Disclosure of Invention
The invention mainly aims to provide a sponge city design optimization method and system based on rainfall measurement and calculation, which solve the problems of how to scientifically and reasonably predict rainfall modes and influences and effectively alleviate urban waterlogging.
In order to achieve the above purpose, the invention provides a sponge city design optimization method based on rainfall measurement, which comprises the following steps:
acquiring historical rainfall data and climate data of a target sponge city, and carrying out frequency distribution analysis on the historical rainfall data to obtain rainfall analysis data;
Predicting the occurrence probability of rainfall events in the target sponge city based on the rainfall analysis data and the climate data to obtain probability distribution of future rainfall and future measured rainfall;
carrying out space analysis on the target sponge city through a preset geographic information system to obtain a distribution map of easy water accumulation points;
Evaluating the waterlogging risk level of each region in the target sponge city based on the distribution map of the easy-to-accumulate water points, the future measured rainfall and the probability distribution of the future rainfall by adopting a Bayesian network algorithm to obtain a waterlogging risk map, wherein the waterlogging risk map comprises a high risk region, a medium risk region and a low risk region;
And carrying out layout optimization design on the target sponge city based on the waterlogging risk map by using a preset multi-target optimization algorithm to obtain an optimal configuration scheme.
Further, performing frequency distribution analysis on the historical rainfall data to obtain rainfall analysis data, including:
Acquiring original historical rainfall data of a plurality of meteorological sites of the target sponge city, and cleaning the original historical rainfall data to remove abnormal values and missing values and obtain historical rainfall data;
performing rainfall trend analysis on the historical rainfall data to obtain rainfall trend data;
performing multi-scale wavelet transformation on the rainfall trend data to extract rainfall characteristics under different time scales so as to obtain multi-scale rainfall characteristic data;
Carrying out extreme rainfall event probability distribution calculation based on the rainfall trend data and the multi-scale rainfall characteristic data by adopting an extreme value theory to obtain probability distribution of the extreme rainfall event, wherein the probability distribution of the extreme rainfall event is the probability distribution of the maximum rainfall and the longest duration;
and obtaining rainfall analysis data based on the probability distribution of the extreme rainfall event, the multi-scale rainfall characteristic data and the rainfall trend data, wherein the rainfall analysis data comprises rainfall intensity, rainfall duration and rainfall interval period.
Further, the predicting the occurrence probability of the rainfall event in the target sponge city based on the rainfall analysis data and the climate data, to obtain probability distribution of future rainfall and future measured rainfall, includes:
Randomly sampling the rainfall intensity based on a Markov chain Monte Carlo algorithm to determine the transition path probability of rainfall of different intensities;
carrying out state hiding analysis and mining on the rainfall duration to obtain state probability distribution of the hiding duration;
Carrying out association feature extraction on the rainfall interval period by using a conditional random field algorithm to construct a conditional probability relation of the rainfall interval period, rainfall intensity and duration, and obtaining an interval period conditional probability relation;
Carrying out data combination on the transition path probability, the hidden duration state probability distribution and the interval period conditional probability relation to obtain rainfall combination data;
constructing rainfall probability density data based on the rainfall combination data;
probability distribution adjustment is carried out on the rainfall probability density data through the maximum entropy principle, and adjustment probability density data is obtained;
and predicting the occurrence probability of rainfall events in the target sponge city based on the adjustment probability density data and the climate data to obtain probability distribution of future rainfall and future measured rainfall.
Further, the spatial analysis is performed on the target sponge city through a preset geographic information system to obtain a distribution map of easy water accumulation points, including:
performing DEM analysis on the topographic and topographic data of the target sponge city through a preset geographic information system to obtain a topographic feature map, wherein the obtained topographic feature map comprises gradient, slope direction and elevation;
simulating a surface runoff path in the topographic feature map by using a preset hydrological model to obtain a runoff simulation path map;
referring to underground pipeline data in the target sponge city in a database, and performing network analysis on the underground pipeline data to identify distribution and connection relations of the underground pipelines, so as to obtain underground pipeline distribution network data;
Estimating the interaction of the surface runoff and the underground pipeline based on a runoff simulation path diagram and the underground pipeline distribution network data by using a network superposition analysis method to obtain a runoff influence diagram;
Classifying and analyzing land utilization data of the target sponge city, and identifying influences of different land utilization types on the surface runoff to obtain a land utilization influence diagram;
And obtaining a distribution diagram of easy water accumulation points based on the runoff influence diagram and the land utilization influence diagram by using a multi-criterion decision analysis method.
Further, the estimating, by using a bayesian network algorithm, the waterlogging risk level of each region in the target sponge city based on the distribution map of the easy-to-accumulate-water point, the future measured rainfall and the probability distribution of the future rainfall, to obtain a waterlogging risk map, includes:
Based on a Bayesian network structure learning algorithm, performing feature analysis on the easy-to-accumulate-point distribution map to determine the dependency relationship among nodes, and obtaining an easy-to-accumulate-point network structure;
Generating a plurality of rainfall situations and ponding situations corresponding to the rainfall situations based on the easily ponding point network structure, the future measured rainfall and the probability distribution of the future rainfall through a preset Monte Carlo simulation technology;
Carrying out waterlogging risk assessment on each rainfall scene and the corresponding ponding situation to obtain risk index data, wherein the risk index data comprise submerging depth and ponding time;
Dividing the region of the target sponge city into different risk grade categories based on the risk index data to obtain a risk grade classification result, wherein the risk grade classification result comprises a high risk category, a medium risk category and a low risk category;
And marking a risk level classification result on the distribution map of the easy-to-accumulate water points to obtain a waterlogging risk map, wherein the waterlogging risk map is a specific position and range of a high-risk area, a medium-risk area and a low-risk area.
Further, the performing layout optimization design on the sponge city based on the waterlogging risk map by using a preset multi-objective optimization algorithm to obtain an optimal configuration scheme includes:
Extracting and analyzing hydrogeologic features of the waterlogging risk area of the sponge city based on the waterlogging risk map to obtain waterlogging hydrogeologic parameters, wherein the waterlogging hydrogeologic parameters comprise groundwater level changes corresponding to permeability coefficients;
Constructing a permeability coefficient-groundwater level change curve based on groundwater level change corresponding to the permeability coefficient;
performing multi-objective optimizing calculation on the sponge city based on the permeability coefficient-groundwater level change curve through a preset multi-objective optimizing algorithm to obtain a sponge facility layout candidate scheme set;
performing spatial clustering analysis on the sponge facility layout candidate scheme set to obtain a layout scheme clustering result;
performing strategy game analysis on each cluster representing scheme in the layout scheme clustering result through a preset game theory algorithm to obtain a game balanced layout scheme set;
and under a plurality of rainfall situations, carrying out layout optimization design on the sponge city based on the game balanced layout scheme set by using a neural network algorithm to obtain an optimal configuration scheme.
Further, the step of performing multi-objective optimization calculation on the target sponge city based on the permeability coefficient-groundwater level change curve by a preset multi-objective optimization algorithm to obtain a sponge facility layout candidate scheme set includes:
determining a change rule of the permeability coefficient and the groundwater level based on the permeability coefficient-groundwater level change curve through a preset multi-objective optimization algorithm;
Performing sensitivity analysis on the permeability coefficient based on the change rule to obtain the influence degree of the permeability coefficient on the groundwater level change under different rainfall situations;
Carrying out hydrological characteristic analysis on the sponge city based on the influence degree and the change rule to obtain a hydrological response mode and hydrological characteristic data in the waterlogging risk area;
And performing multi-objective optimizing calculation on the sponge city based on the hydrological response mode and the hydrological characteristic data to obtain a sponge facility layout candidate scheme set.
The invention also provides a sponge city design optimization system based on rainfall measurement, which comprises:
The first analysis module is used for acquiring historical rainfall data and climate data of a target sponge city, and carrying out frequency distribution analysis on the historical rainfall data to obtain rainfall analysis data;
the prediction module is used for predicting the occurrence probability of rainfall events in the target sponge city based on the rainfall analysis data and the climate data to obtain probability distribution of future rainfall and future measured rainfall;
the second analysis module is used for carrying out space analysis on the target sponge city through a preset geographic information system to obtain a distribution diagram of easy water accumulation points;
the evaluation module is used for evaluating the waterlogging risk level of each area in the target sponge city based on the easy-to-accumulate-point distribution map, the future measured rainfall and the probability distribution of the future rainfall by adopting a Bayesian network algorithm to obtain a waterlogging risk map, wherein the waterlogging risk map comprises a high risk area, a medium risk area and a low risk area;
And the design module is used for carrying out layout optimization design on the target sponge city based on the waterlogging risk map by utilizing a preset multi-target optimization algorithm to obtain an optimal configuration scheme.
The invention also provides a computer device comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of any of the methods described above when the computer program is executed.
The invention also provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method of any of the preceding claims.
The sponge city design optimization method based on rainfall measurement comprises the steps of obtaining historical rainfall data and climate data of a target sponge city, carrying out frequency distribution analysis on the historical rainfall data to obtain rainfall analysis data, predicting occurrence probability of rainfall events in the target sponge city based on the rainfall analysis data and the climate data to obtain probability distribution of future rainfall and future measured rainfall, carrying out space analysis on the target sponge city through a preset geographic information system to obtain an easy-to-accumulate-point distribution map, evaluating waterlogging risk levels of all areas in the target sponge city based on the easy-to-accumulate-point distribution map, the future measured rainfall and the probability distribution of the future rainfall by adopting a Bayesian network algorithm to obtain a waterlogging risk map, carrying out layout optimization design on the target sponge city by utilizing a preset multi-objective optimization algorithm based on the waterlogging risk map to obtain an optimal configuration scheme, solving the problems of how to scientifically predict the mode and influence and effectively relieving the urban waterlogging problem, realizing that the easy-to-accumulate-point distribution map and the probability distribution map are comprehensively considered, and the risk level of future rainfall is different, and providing a detailed risk management and a risk management effect for a user is provided for the urban waterlogging risk level.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, fig. 1 is a schematic diagram illustrating steps of a sponge city design optimization method based on rainfall measurement according to an embodiment of the present invention;
the embodiment of the invention provides a sponge city design optimization method based on rainfall measurement, which comprises the following steps:
step S1, historical rainfall data and climate data of a target sponge city are obtained, and frequency distribution analysis is carried out on the historical rainfall data to obtain rainfall analysis data.
Specifically, in order to realize the step of acquiring the historical rainfall data and the climate data of the target sponge city and performing frequency distribution analysis on the historical rainfall data to obtain rainfall analysis data, the rainfall records and the related climate data of the target sponge city for years are collected. The data can be obtained from the national weather bureau or local weather station, or can be obtained by satellite remote sensing technology. For example, in a particular sponge city construction project, a project team may download daily rainfall data over the past decade by accessing an official website of a local meteorological department, while collecting contemporaneous climate information such as air temperature, humidity, wind speed, etc. And then, processing the collected historical rainfall data by using professional data analysis software, such as Pandas libraries in SPSS or Python programming languages, removing invalid values and abnormal values, and ensuring the accuracy and reliability of the data. After data cleaning is completed, the next step is to perform frequency distribution analysis on the rainfall data after cleaning, namely counting the occurrence times or frequency of different rainfall in a specific time period. For example, the daily rainfall can be calculated as the number of days in each of the three intervals of less than 10 mm, between 10 and 50 mm, and greater than 50 mm. By drawing a histogram or a cumulative frequency curve, the distribution situation of different rainfall intensities is intuitively displayed, so that researchers are helped to understand the rainfall pattern and seasonal variation rule of the target sponge city. The analysis result not only provides an important reference basis for the subsequent rainfall event prediction, but also enables the urban waterlogging risk assessment to be more scientific and reasonable.
And S2, predicting the occurrence probability of rainfall events in the target sponge city based on the rainfall analysis data and the climate data, and obtaining probability distribution of future rainfall and future measured rainfall.
Specifically, after the collection and frequency distribution analysis of the historical rainfall data and the climate data of the target sponge city are completed, the next step is to predict the occurrence probability of the future rainfall event based on the analysis results, so as to obtain the probability distribution of the future rainfall and the future measured rainfall. This process relies primarily on statistical models and machine learning algorithms to capture the relationship between rainfall patterns and climate factors by training the models. For example, in a specific application scenario, a project team may employ a time series analysis method, such as an ARIMA model, or a machine learning method, such as a random forest and support vector machine, using existing rainfall analysis data and climate data as input variables, training the model to learn association rules between these variables and rainfall events. On the basis, the model can predict rainfall conditions in a certain future time period, including the occurrence probability of rainfall events with different intensities and the expected total rainfall amount. In order to ensure the accuracy of prediction, model parameters are continuously adjusted and optimized, such as selecting an optimal model super-parameter combination through a cross-validation method, or introducing more influencing factors, such as soil humidity, vegetation coverage and the like, so as to improve the generalization capability of the model. Meanwhile, the influence of climate change on a rainfall mode is considered, a training data set can be updated regularly, and the latest observation data is added, so that the model can adapt to the change of environmental conditions, and the timeliness and the accuracy of prediction are maintained. For example, in the sponge city construction project, if the model predicts that continuous medium to heavy rain weather will exist in the future for several weeks, maintenance work of the drainage system can be done in advance according to the prediction, monitoring of easy water accumulation points is enhanced, and measures are taken in time to prevent possible waterlogging disasters. Therefore, through scientific and reasonable rainfall prediction, powerful data support and technical support are provided for the construction and management of sponge cities.
And S3, carrying out space analysis on the target sponge city through a preset geographic information system to obtain a distribution map of easy water accumulation points.
Specifically, after analysis of historical rainfall data and climate data of a target sponge city and prediction of occurrence probability of a future rainfall event are realized, the next important step is to perform spatial analysis on the target sponge city through a preset Geographic Information System (GIS) so as to obtain a distribution diagram of easy-to-accumulate water points. The process involves utilizing the powerful spatial data analysis function of GIS, combining multiple factors such as topography, underground pipeline layout, building density, etc., and accurately identifying the position where water accumulation possibly occurs in the urban area. For example, in a specific sponge city construction project, a project team may first import a Digital Elevation Model (DEM) of a city on a GIS platform, and based on this, superimpose vector layers of urban roads, rivers and lakes, green parks, etc., to construct a complete urban spatial database. Subsequently, by analyzing the information of the slope of the terrain, the direction and the speed of the water flow, potential low-lying areas where the water flow is converged are identified, and the areas are often high-lying areas of urban inland inundation. Meanwhile, the bearing capacity and potential bottleneck of the existing drainage system can be evaluated by combining the distribution condition of the urban drainage pipe network, particularly the position information of key nodes such as inspection wells, pump stations and the like, and the sections with unsmooth drainage can be further screened out. After the space data analysis is completed, the GIS system can automatically generate a detailed easy-to-accumulate-point distribution map, and the specific positions and the surrounding environment characteristics of each easy-to-accumulate-point are clearly marked. The distribution map not only provides an important basis for subsequent waterlogging risk assessment, but also provides visual guidance for urban planners and managers in the process of carrying out drainage facility reconstruction and sponge urban construction. For example, according to the distribution diagram of the easy water accumulation points, project teams can prioritize the addition of green infrastructure such as permeable pavement, rainwater garden and the like in a high risk area, improve the rainwater permeability of the ground surface, and relieve the pressure of a city drainage system, thereby effectively reducing the risk of waterlogging.
And S4, evaluating the waterlogging risk level of each area in the target sponge city based on the distribution map of the easy-to-accumulate-point, the future measured rainfall and the probability distribution of the future rainfall by adopting a Bayesian network algorithm to obtain a waterlogging risk map, wherein the waterlogging risk map comprises a high risk area, a medium risk area and a low risk area.
Specifically, after the preparation of a distribution diagram of easy water accumulation points of a target sponge city is completed, a Bayesian network algorithm is adopted in the next step, the waterlogging risk level of each region in the target sponge city is estimated based on the distribution diagram, the future measured rainfall and the probability distribution of the future rainfall, and finally a waterlogging risk map is formed, wherein the map divides the region into a high risk region, a medium risk region and a low risk region. In particular, bayesian network algorithms are probabilistic inference models that can handle uncertainty problems and are well suited for assessing risk in complex environments. In this process, we first need to collect various factors related to waterlogging, such as the position of the easy-to-accumulate water point, the historical rainfall record, the topography and topography characteristics, the efficiency of the drainage system, etc., and then take these factors as nodes of the bayesian network. Each node represents a variable, and the connections between nodes represent causal relationships or correlations between the variables. For example, in the application scenario of sponge cities, if the topography of a region is low and there is a large amount of hard ground around, even if the drainage system is perfect, the region may still become a water spot in the event of heavy rainfall. Next, we incorporate future measured rainfall and probability distribution of future rainfall into bayesian network model, which is to predict the possibility of waterlogging in each region in different rainfall situations more accurately. For example, for a city expected to experience continuous heavy rain in the next few days, the weather forecast data can be used to adjust node weights related to rainfall intensity and duration in the Bayesian network, so as to calculate the probability of each easy-to-accumulate point being converted into an actual accumulated point under different rainfall level conditions. By the method, the Bayesian network can comprehensively consider the influence of various factors, and a waterlogging risk level is allocated to each area so as to generate a complete waterlogging risk map. Finally, based on the generated waterlogging risk map, city managers can take more targeted measures to reduce the influence caused by waterlogging. For example, the construction of drainage facilities is enhanced in a high risk area, emergency response capability is improved, greening design is optimized in a medium risk area, rainwater absorption capability is improved, and resource investment can be properly reduced in a low risk area, so that more attention is paid to daily maintenance and management. Therefore, the occurrence probability of waterlogging disasters can be effectively reduced, limited city construction funds can be reasonably distributed, and the maximization of resource utilization is achieved.
And S5, carrying out layout optimization design on the target sponge city based on the waterlogging risk map by using a preset multi-target optimization algorithm to obtain an optimal configuration scheme.
Specifically, after the waterlogging risk map is obtained, the next key step is to utilize a preset multi-objective optimization algorithm to perform layout optimization design on the target sponge city based on the waterlogging risk map so as to obtain an optimal configuration scheme. This process involves balancing and optimizing a number of objectives including, but not limited to, reducing risk of flooding, improving rainwater utilization, controlling construction costs, and the like. The multi-objective optimization algorithm is able to find one or more pareto optimal solutions among many possible designs, i.e. solutions that cannot further improve other objectives without sacrificing one objective. For example, in the construction of sponge cities, the minimization of risk of waterlogging, the maximization of rainwater recovery efficiency and the minimization of construction costs can be set as three main optimization objectives. In the implementation, firstly, quantitative indexes of the targets need to be defined, for example, the risk of waterlogging can be measured by multiplying the occurrence probability of waterlogging by the influence range, the rainwater recovery efficiency can be determined according to the ratio of the rainwater collection amount to the total precipitation amount, and the construction cost is directly calculated according to the actual expenses such as the material cost, the construction cost and the like. Next, these quantization indexes are input into a multi-objective optimization algorithm, such as a genetic algorithm, a particle swarm optimization algorithm, etc., which can search for an optimal solution by simulating the evolutionary process or swarm behavior of the nature. In the optimization process, the algorithm can continuously generate new design schemes, evaluate the advantages and disadvantages of each scheme according to a preset objective function, and gradually converge on an optimal solution. Taking a specific sponge city construction project as an example, assuming that the project is located in an old urban area where waterlogging is likely to occur, the project team has obtained a detailed waterlogging risk map through the previous steps, showing which areas belong to the high risk area, the medium risk area and the low risk area. Based on the map, a team can explore different design schemes by utilizing a multi-objective optimization algorithm, such as increasing the proportion of permeable pavement and a rainwater garden in a high-risk area so as to enhance the rainwater permeability of the ground surface, arranging more rainwater collecting pools and artificial wetlands in a medium-risk area so as to improve the storage and purification capacities of rainwater, and mainly carrying out greening transformation in a low-risk area so as to improve the ecological environment quality of a city. After multiple rounds of iterative optimization, an optimal allocation scheme which can effectively reduce waterlogging risk and also give consideration to economic benefit and social benefit is finally selected, so that powerful technical support is provided for scientific planning and construction of sponge cities.
In a specific embodiment, performing frequency distribution analysis on the historical rainfall data to obtain rainfall analysis data, including:
Acquiring original historical rainfall data of a plurality of meteorological sites of the target sponge city, and cleaning the original historical rainfall data to remove abnormal values and missing values and obtain historical rainfall data;
performing rainfall trend analysis on the historical rainfall data to obtain rainfall trend data;
performing multi-scale wavelet transformation on the rainfall trend data to extract rainfall characteristics under different time scales so as to obtain multi-scale rainfall characteristic data;
Carrying out extreme rainfall event probability distribution calculation based on the rainfall trend data and the multi-scale rainfall characteristic data by adopting an extreme value theory to obtain probability distribution of the extreme rainfall event, wherein the probability distribution of the extreme rainfall event is the probability distribution of the maximum rainfall and the longest duration;
and obtaining rainfall analysis data based on the probability distribution of the extreme rainfall event, the multi-scale rainfall characteristic data and the rainfall trend data, wherein the rainfall analysis data comprises rainfall intensity, rainfall duration and rainfall interval period.
Specifically, the process of obtaining rainfall analysis data is a complex and fine task by carrying out frequency distribution analysis on the historical rainfall data, and the process relates to a plurality of links such as data acquisition, cleaning, trend analysis, feature extraction, probability distribution calculation and the like. The process aims to comprehensively understand the rainfall characteristics of the target sponge city, and provides a solid foundation for subsequent rainfall event prediction and waterlogging risk management. Specifically, raw historical rainfall data is first acquired from multiple weather sites in a target sponge city. These data typically comprise daily rainfall records for many years, either stored in the form of a spreadsheet or in the form of a database. To ensure the quality of the data, these raw data must be cleaned to remove outliers and missing values, resulting in historical rainfall data that can be used for analysis. For example, in a particular sponge city construction project, a project team may obtain daily rainfall data for the past twenty years from a local weather office, then script using Python or R language, automatically detect and delete those data points that are significantly erroneous, such as negative rainfall or extremely high values outside of reasonable limits, while filling in missing date data, ensuring the integrity of the data sequence. After the data are cleaned, the rainfall trend analysis is carried out on the historical rainfall data so as to reveal the change rule of the rainfall mode along with time. This stage of work typically employs statistical methods, such as linear regression analysis or moving average, to identify long-term rainfall trends. For example, a project team may use SPSS software to perform linear regression analysis on the cleaned rainfall data to find a trend line of rainfall over time, which may help us determine if there is a trend in the area of increasing or decreasing rainfall year by year. This trend analysis is critical to understanding the impact of climate change on the rainfall pattern, and also provides important background information for subsequent rainfall prediction. On the basis of obtaining rainfall trend data, the data are further subjected to multi-scale wavelet transformation to extract rainfall features at different time scales. wavelet transform is a time-frequency analysis tool, particularly suitable for processing non-stationary signals, which can decompose rainfall data into components on multiple time scales, revealing the rainfall characteristics at different time scales. For example, in the application scenario of sponge cities, a project team may utilize wavelet transform functions in MATLAB to convert daily rainfall data into coefficients on different time scales reflecting various rainfall characteristics from short term fluctuations to long term trends. By comparing coefficient changes at different time scales, it can be found that rainfall activity is more active at certain specific time scales, which features are very useful for understanding the occurrence mechanism of rainfall events. After the multi-scale rainfall characteristic data are provided, the extreme value theory is adopted, and the extreme rainfall event probability distribution calculation is carried out based on the rainfall trend data and the multi-scale rainfall characteristic data. Extremum theory is a branch of statistics, dedicated to analyzing the probability distribution of extreme events, which can help us estimate the likelihood of extreme rainfall events occurring within a certain time. For example, in a sponge city construction project, a project team may calculate the probability distribution of maximum rainfall and longest duration using gummel distribution or Weibull distribution models in extremum theory, in combination with rainfall trend data and multi-scale rainfall signature data. This means that a team can predict what the probability of the maximum daily rainfall for a particular location exceeding a certain threshold, or what the probability of the longest consecutive days of rainfall exceeding a certain threshold, in the next year. The probability distribution of the extreme rainfall events provides scientific basis for planning and designing the urban drainage system, helps decision makers to reasonably allocate resources, and improves the flood fighting capacity of cities. Finally, comprehensively obtaining rainfall analysis data based on probability distribution of extreme rainfall events, multi-scale rainfall characteristic data and rainfall trend data, wherein the rainfall analysis data comprises key parameters such as rainfall intensity, rainfall duration, rainfall interval period and the like. These parameters not only describe the basic characteristics of the rainfall event, but also reflect the distribution law of the rainfall event in time and space. For example, in sponge city applications, a project team can identify from rainfall analysis data which months are more prone to high intensity short duration rainfall and which areas are more prone to be hot spots of long duration continuous rainfall due to terrain reasons. The information is important to the establishment of an effective waterlogging prevention and control strategy, so that urban managers can be helped to make full preparation before coming in rainy seasons, and the influence of waterlogging disasters on urban life is reduced. In summary, by performing frequency distribution analysis on the historical rainfall data, the process of obtaining rainfall analysis data is a multi-step, multi-technology integrated process, which requires interdisciplinary knowledge and skills. From data acquisition to cleaning, trend analysis, feature extraction and probability distribution calculation, each step is crucial, and a complete rainfall characteristic analysis framework is formed together. the framework not only provides scientific data support for the construction and management of sponge cities, but also provides valuable references for the hydrological research in other fields.
In a specific embodiment, predicting the occurrence probability of the rainfall event in the target sponge city based on the rainfall analysis data and the climate data to obtain probability distribution of future rainfall and future measured rainfall comprises:
Randomly sampling the rainfall intensity based on a Markov chain Monte Carlo algorithm to determine the transition path probability of rainfall of different intensities;
carrying out state hiding analysis and mining on the rainfall duration to obtain state probability distribution of the hiding duration;
Carrying out association feature extraction on the rainfall interval period by using a conditional random field algorithm to construct a conditional probability relation of the rainfall interval period, rainfall intensity and duration, and obtaining an interval period conditional probability relation;
Carrying out data combination on the transition path probability, the hidden duration state probability distribution and the interval period conditional probability relation to obtain rainfall combination data;
constructing rainfall probability density data based on the rainfall combination data;
probability distribution adjustment is carried out on the rainfall probability density data through the maximum entropy principle, and adjustment probability density data is obtained;
and predicting the occurrence probability of rainfall events in the target sponge city based on the adjustment probability density data and the climate data to obtain probability distribution of future rainfall and future measured rainfall.
Specifically, predicting the occurrence probability of rainfall events in the target sponge city based on the rainfall analysis data and the climate data, and obtaining probability distribution of future rainfall and a process of measuring rainfall in the future are highly comprehensive tasks, and relate to knowledge in multiple fields such as statistics, machine learning, probability theory and the like. The main purpose of the process is to accurately predict rainfall events through advanced algorithms and models, and provide scientific basis for urban waterlogging risk management. Specifically, the rainfall intensities first need to be randomly sampled using a Markov Chain Monte Carlo (MCMC) algorithm to determine the transition path probabilities of the different intensity rainfall. the MCMC algorithm is an iterative method for extracting samples from complex probability distributions, particularly suited to deal with problems in high-dimensional parameter spaces. For example, in a sponge city construction project, a project team may generate a series of random samples using the MCMC algorithm based on a sequence of rainfall intensities in historical rainfall data, the samples representing transition paths between different rainfall intensities. By analyzing these samples, the probability of shifting from one rainfall intensity to another can be calculated, thereby constructing a shift matrix of rainfall intensities. After the transition path probability of the rainfall intensity is obtained, the state hiding analysis and mining are carried out on the rainfall duration so as to obtain the state probability distribution of the hiding duration. A method similar to the Hidden Markov Model (HMM) is used here, which assumes that there are some unobserved states behind the observed rainfall duration, which affect the distribution of the rainfall duration. For example, a project team may divide the rainfall duration into several states, such as "short time", "medium time", and "long time", and then use the EM algorithm (expectation maximization algorithm) to estimate the probability distribution of the rainfall duration in each state. In this way, not only can the internal structure of the rainfall duration be revealed, it can be identified which factors may lead to an extension or shortening of the rainfall duration. And then, carrying out associated feature extraction on the rainfall interval period by using a Conditional Random Field (CRF) algorithm to construct a conditional probability relation of the rainfall interval period, the rainfall intensity and the duration, and obtaining the conditional probability relation of the interval period. CRF is a statistical modeling method for sequence labeling and structured prediction, and is particularly suitable for processing dependency relations in time sequence data. In the sponge city application scenario, a project team may consider the rainfall interval period, rainfall intensity, and duration as multiple features over a time sequence, and then learn the interactions between these features using the CRF model. By training the CRF model, the conditional probability of each feature at different time points can be obtained, so that the relationship between the rainfall interval period and the other two features is established. After the steps are completed, data combination is required to be carried out on the transition path probability, the hidden duration time state probability distribution and the interval period conditional probability relation, so that rainfall combination data are obtained. This process effectively integrates information acquired from different angles to form a complete rainfall event description. For example, a project team may incorporate the transition probabilities of rainfall intensities, the state probability distribution of rainfall durations, and the conditional probability relationship of rainfall interval periods into a unified data structure for subsequent processing. The data combination not only improves the information integrity, but also lays a foundation for subsequent rainfall prediction. Constructing rainfall probability density data based on the rainfall combination data is a key step of the next step. This step aims at converting the combined data into the form of a probability density function, facilitating the calculation and adjustment of the probability distribution. For example, a project team may map each feature in the rainfall portfolio data onto a continuous probability density function using a Kernel Density Estimation (KDE) method, resulting in probability density distributions for rainfall events under different conditions. By constructing rainfall probability density data, the possibility of different rainfall characteristic combinations can be more intuitively seen, and richer information is provided for prediction. And carrying out probability distribution adjustment on the rainfall probability density data through a maximum entropy principle to obtain adjustment probability density data. The principle of maximum entropy is a method of finding the least deterministic (i.e. entropy maximum) probability distribution under given constraints, which ensures that the predicted outcome is as close to real as possible, while avoiding overfitting. In sponge city applications, the project team may set some prior knowledge as constraints, such as average rainfall intensity and duration in historical rainfall data, and then readjust the rainfall probability density data using a maximum entropy model. This has the advantage that the prediction result can be made more robust, reducing errors due to data noise. And finally, predicting the occurrence probability of rainfall events in the target sponge city based on the adjustment probability density data and the climate data, and obtaining probability distribution of future rainfall and future measured rainfall. The work at this stage is to combine all pre-processed data with current and future climate data to estimate the probability of occurrence of a future rainfall event by constructing a suitable predictive model. For example, a project team may use a Bayesian network or deep learning model with the adjusted rainfall probability density data and the latest climate forecast data as inputs to output a joint probability distribution of different rainfall intensities, durations, and interval periods over a period of time in the future. By the method, not only the occurrence probability of the future rainfall event can be predicted, but also the corresponding rainfall can be estimated, and scientific decision support is provided for the construction and management of the sponge city. In summary, predicting the occurrence probability of the rainfall event in the target sponge city based on the rainfall analysis data and the climate data, and obtaining the probability distribution of the future rainfall and the process of measuring and calculating the rainfall in the future are a multi-step and multi-level complex task. From the application of the Markov chain Monte Carlo algorithm to the adjustment of the maximum entropy principle, each step is continuously optimizing and refining the prediction model, and the final purpose is to provide more accurate and reliable rainfall prediction results, help cities to better cope with waterlogging risks and realize sustainable development.
In a specific embodiment, the performing spatial analysis on the target sponge city through a preset geographic information system to obtain a distribution map of easy water accumulation points includes:
performing DEM analysis on the topographic and topographic data of the target sponge city through a preset geographic information system to obtain a topographic feature map, wherein the obtained topographic feature map comprises gradient, slope direction and elevation;
simulating a surface runoff path in the topographic feature map by using a preset hydrological model to obtain a runoff simulation path map;
referring to underground pipeline data in the target sponge city in a database, and performing network analysis on the underground pipeline data to identify distribution and connection relations of the underground pipelines, so as to obtain underground pipeline distribution network data;
Estimating the interaction of the surface runoff and the underground pipeline based on a runoff simulation path diagram and the underground pipeline distribution network data by using a network superposition analysis method to obtain a runoff influence diagram;
Classifying and analyzing land utilization data of the target sponge city, and identifying influences of different land utilization types on the surface runoff to obtain a land utilization influence diagram;
And obtaining a distribution diagram of easy water accumulation points based on the runoff influence diagram and the land utilization influence diagram by using a multi-criterion decision analysis method.
Specifically, the process of obtaining the distribution map of the easily accumulated water points by carrying out space analysis on the target sponge city through a preset geographic information system is a complex and fine work, and relates to a plurality of steps such as topography analysis, hydrologic simulation, underground pipeline network analysis, land utilization classification analysis, multi-criterion decision analysis and the like. The process aims to comprehensively evaluate the hydrologic characteristics and the infrastructure conditions in the city, so that easy water accumulation points are identified, and scientific basis is provided for urban waterlogging risk management and sponge city construction. Specifically, firstly, the topographic and topographic data of the target sponge city needs to be subjected to DEM (digital elevation model) analysis through a preset geographic information system so as to obtain a topographic feature map. The topography map includes information such as gradient, slope direction and elevation, which are the basis for evaluating the path of surface runoff and the direction of water flow. For example, in a particular sponge city construction project, a project team may use ArcGIS software to load DEM data for a target city and then run a terrain analysis tool to generate a terrain profile that includes grade, slope and elevation. The map can clearly show the height fluctuation and the topography trend of the city, and provides an important geographical background for the subsequent hydrologic simulation. After the topographic feature map is obtained, a surface runoff path in the topographic feature map is simulated by using a preset hydrological model, and a runoff simulation path map is obtained. The choice of hydrologic model depends on the characteristics of the investigation region and the type of data available, common models include SWAT（Soil and Water Assessment Tool）、HEC-HMS（Hydrologic Engineering Center's Hydrologic Modeling System）, etc. The model can simulate the surface runoff path and flow change after rainfall according to the factors such as the terrain features, the soil types, the vegetation coverage and the like. For example, a project team can use a HEC-HMS model to input gradient and slope data in a topographic feature map and soil and vegetation information, set different rainfall scenarios, and run the model to obtain a surface runoff path map under different rainfall conditions. These path diagrams not only show the direction of the water flow, but also reflect the speed and flow of the water flow, helping to identify potential water accumulation areas. Meanwhile, the underground pipeline data in the target sponge city is required to be consulted in a database, and network analysis is carried out on the underground pipeline data so as to identify the distribution and connection relation of the underground pipelines and obtain the underground pipeline distribution network data. Underground pipelines include various types of rainwater pipelines, sewage pipelines, water supply pipelines and the like, and the distribution and connection conditions of the pipelines directly influence the discharge of surface runoff and the change of underground water level. For example, a project team may access a municipal engineering database of a city, derive a GIS map of an underground pipeline, and then build a network model of the underground pipeline using network analysis tools in ArcGIS. By analyzing the attributes of the length, diameter, material and the like of the pipelines and the connection relation among the pipelines, key nodes and bottleneck sections in the pipe network can be identified, and the information is important for evaluating the drainage capacity of the underground pipelines. And evaluating the interaction of the surface runoff and the underground pipeline based on the runoff simulation path diagram and the underground pipeline distribution network data by using a network superposition analysis method to obtain a runoff influence diagram. Network overlay analysis refers to overlaying layers of different types of geographic information together for spatial analysis to reveal interactions between different elements. For example, a project team may overlay the runoff simulation path graph with the pipeline distribution network data to analyze intersections and adjacent areas of the surface runoff path with the pipeline. From this analysis, it can be identified which of the underground lines may be damaged by the impact of surface runoff and which areas of the underground lines have insufficient drainage capacity, resulting in surface water. Such information will help to preferentially repair and reform the underground pipeline, improving the drainage efficiency of the city. And carrying out classification analysis on the land utilization data of the target sponge city, and identifying the influence of different land utilization types on the surface runoff to obtain a land utilization influence diagram. Land use types include residential, commercial, industrial, greenfield, water, etc., and different land use types have different effects on the generation and flow of surface runoff. For example, a project team may use remote sensing images and urban planning maps to classify land use types and then calculate area proportions and distribution characteristics for each land use type using GIS software. Then, the influence of different land utilization types on the surface runoff is evaluated by combining parameters such as the water permeability, the surface roughness and the like of the surface covering. For example, hardened floors (e.g., concrete, asphalt) can result in rapid increases in surface runoff, while greenbelts and wetlands can slow down surface runoff, increasing penetration and evaporation of rain. By this analysis it can be identified which areas are prone to water accumulation due to unreasonable land utilization. And finally, comprehensively considering the mutual influence of the surface runoff and the underground pipeline and the influence of different land utilization types on the surface runoff based on the runoff influence diagram and the land utilization influence diagram by utilizing a multi-criterion decision analysis method to obtain a distribution diagram of easy water accumulation points. Multi-criterion decision analysis is a comprehensive evaluation method that can trade off between multiple evaluation criteria to select an optimal solution. For example, a project team may set a series of evaluation metrics such as surface runoff speed, underground pipeline drainage capacity, land utilization type, etc., and then assign weights to each metric, calculate a composite score for each region using Analytical Hierarchy Process (AHP) or fuzzy comprehensive evaluation. Dividing the city into a high risk area, a medium risk area and a low risk area according to the grading, and finally generating a distribution map of easy water accumulation points. The map can intuitively show which areas in the city are most prone to water accumulation, and provides clear guidance for city planning and management. In summary, the process of obtaining the distribution map of the easily accumulated water point by spatially analyzing the target sponge city through the preset geographic information system is a multi-step and multi-technology integrated process. From the analysis of topography and topography to hydrologic simulation, to the analysis of underground pipeline network and land utilization classification analysis, each step is of great importance, and a comprehensive urban waterlogging risk assessment framework is formed together. The framework not only provides scientific data support for the construction and management of sponge cities, but also provides valuable references for urban planning and environmental protection in other fields.
In a specific embodiment, the estimating, by using a bayesian network algorithm, the waterlogging risk level of each region in the target sponge city based on the easy-to-accumulate-point distribution map, the future measured rainfall and the probability distribution of the future rainfall to obtain a waterlogging risk map includes:
Based on a Bayesian network structure learning algorithm, performing feature analysis on the easy-to-accumulate-point distribution map to determine the dependency relationship among nodes, and obtaining an easy-to-accumulate-point network structure;
Generating a plurality of rainfall situations and ponding situations corresponding to the rainfall situations based on the easily ponding point network structure, the future measured rainfall and the probability distribution of the future rainfall through a preset Monte Carlo simulation technology;
Carrying out waterlogging risk assessment on each rainfall scene and the corresponding ponding situation to obtain risk index data, wherein the risk index data comprise submerging depth and ponding time;
Dividing the region of the target sponge city into different risk grade categories based on the risk index data to obtain a risk grade classification result, wherein the risk grade classification result comprises a high risk category, a medium risk category and a low risk category;
And marking a risk level classification result on the distribution map of the easy-to-accumulate water points to obtain a waterlogging risk map, wherein the waterlogging risk map is a specific position and range of a high-risk area, a medium-risk area and a low-risk area.
Specifically, the process of evaluating the waterlogging risk level of each region in the target sponge city based on the easy-to-accumulate-point distribution map, the future measured rainfall and the future rainfall probability distribution by adopting the Bayesian network algorithm to obtain the waterlogging risk map is a highly comprehensive and technical task, and relates to a plurality of steps of Bayesian network structure learning, monte Carlo simulation, waterlogging risk evaluation, risk level division and the like. The main purpose of the process is to comprehensively evaluate urban waterlogging risk through advanced algorithms and models, and provide scientific basis for urban waterlogging management and sponge urban construction. Specifically, firstly, a Bayesian network structure learning algorithm is required to perform feature analysis on the easy-to-accumulate-point distribution map so as to determine the dependency relationship among nodes and obtain the easy-to-accumulate-point network structure. The Bayesian network is a probability graph model, can express causal relation and dependency relation among variables, and is particularly suitable for processing uncertainty problems. For example, in a sponge city construction project, a project team may use a structure learning algorithm, such as the K2 algorithm or TAN (Tree Augmented Naive Bayes) algorithm, to perform a feature analysis on each point in the easy-to-accumulate-point distribution map to identify which points have significant dependencies between themselves. By the method, a network structure reflecting the mutual influence among the easily accumulated water points can be constructed, and a foundation is provided for subsequent risk assessment. After the easy-to-accumulate-point network structure is obtained, a plurality of rainfall situations and accumulated water situations corresponding to the rainfall situations are generated based on the easy-to-accumulate-point network structure, the future measured rainfall and the probability distribution of the future rainfall through a preset Monte Carlo simulation technology. monte Carlo simulation is a method of solving problems by random sampling, and is particularly suitable for processing complex probability distribution and uncertainty problems. For example, a project team may generate a series of random rainfall scenarios based on a pool-prone spot network structure, combined with future measured rainfall and probability distribution of future rainfall. Each rainfall scenario includes parameters such as time, intensity, and duration of the rainfall. Then, simulating each rainfall scene by using a hydrologic model, and calculating the water accumulation depth and water accumulation time of each easy water accumulation point under different rainfall scenes. In this way, a large number of possible ponding situations can be generated, providing a rich data support for waterlogging risk assessment. And carrying out waterlogging risk assessment on each rainfall scene and the corresponding ponding situation to obtain risk index data, which is one of the core steps for assessing urban waterlogging risk. The risk index data comprise submerging depth and water accumulation time, and the two indexes can comprehensively reflect the influence degree of waterlogging on cities. For example, a project team may set different risk thresholds, such as a water depth exceeding 30 cm or a water time exceeding 6 hours, as high risk. And then, calculating the risk value of each easy water accumulation point according to the water accumulation depth and the water accumulation time under each rainfall scene. By the method, risk index data of each easily-accumulated water point under different rainfall situations can be obtained, and a basis is provided for subsequent risk classification. And dividing the region of the target sponge city into different risk grade categories based on the risk index data to obtain a risk grade classification result. The process needs to set reasonable risk classification standards, and scientificity and practicability of the evaluation result are ensured. For example, a project team may categorize risk categories into high risk categories, medium risk categories, and low risk categories. The specific division criteria may be based on a statistical distribution of risk indicator data, such as dividing the first 20% of the regions with highest risk values into high risk categories, the middle 60% into medium risk categories, and the remaining 20% into low risk categories. In this way, the urban area can be divided into different risk classes, and explicit risk partition information is provided for the urban manager. And finally, marking a risk level classification result on the distribution diagram of the easy-to-accumulate water points to obtain a waterlogging risk map. The waterlogging risk map not only shows the specific positions and the specific ranges of the high risk area, the medium risk area and the low risk area in the city, but also can intuitively reflect the overall distribution condition of the urban waterlogging risks. For example, a project team may label areas of each risk level on the water spot prone profile using different colors or symbols, such as red for high risk areas, yellow for medium risk areas, and green for low risk areas. By the method, a detailed waterlogging risk map can be generated, and visual data support is provided for urban waterlogging management and sponge urban construction. In summary, the process of estimating the waterlogging risk level of each region in the target sponge city based on the distribution map of the easy-to-accumulate-water points, the future measured rainfall and the probability distribution of the future rainfall by adopting the bayesian network algorithm and obtaining the waterlogging risk map is a multi-step and multi-level complex task. Learning from a Bayesian network structure, monte Carlo simulation, waterlogging risk assessment and risk classification, and continuously optimizing and refining an assessment model in each step, wherein the final purpose is to provide a more accurate and reliable waterlogging risk assessment result, help cities to better cope with waterlogging risks, and realize sustainable development.
In a specific embodiment, the performing layout optimization design on the sponge city based on the waterlogging risk map by using a preset multi-objective optimization algorithm to obtain an optimal configuration scheme includes:
Extracting and analyzing hydrogeologic features of the waterlogging risk area of the sponge city based on the waterlogging risk map to obtain waterlogging hydrogeologic parameters, wherein the waterlogging hydrogeologic parameters comprise groundwater level changes corresponding to permeability coefficients;
Constructing a permeability coefficient-groundwater level change curve based on groundwater level change corresponding to the permeability coefficient;
performing multi-objective optimizing calculation on the sponge city based on the permeability coefficient-groundwater level change curve through a preset multi-objective optimizing algorithm to obtain a sponge facility layout candidate scheme set;
performing spatial clustering analysis on the sponge facility layout candidate scheme set to obtain a layout scheme clustering result;
performing strategy game analysis on each cluster representing scheme in the layout scheme clustering result through a preset game theory algorithm to obtain a game balanced layout scheme set;
and under a plurality of rainfall situations, carrying out layout optimization design on the sponge city based on the game balanced layout scheme set by using a neural network algorithm to obtain an optimal configuration scheme.
Specifically, the process of performing layout optimization design on the sponge city based on the waterlogging risk map by using a preset multi-objective optimization algorithm to obtain an optimal configuration scheme is a highly comprehensive and technical task, and involves multiple steps of hydrogeologic feature extraction and analysis, multi-objective optimization calculation, spatial cluster analysis, strategy game analysis, neural network optimization and the like. The main purpose of the process is to optimally design the layout of the sponge city by a scientific method and an advanced algorithm so as to reduce the risk of waterlogging to the maximum extent and improve the rainwater management capability of the city. Specifically, firstly, the waterlogging risk area of the sponge city needs to be extracted and analyzed based on a waterlogging risk map to obtain waterlogging hydrogeological parameters. These parameters include the change in groundwater level corresponding to the permeability coefficient, which are important indicators for assessing surface water and groundwater interactions. For example, in a sponge city construction project, a project team may use GIS software to extract a geological profile and a soil type distribution map of each waterlogging risk region in combination with a waterlogging risk map and geological survey data. By analyzing the data, the permeability coefficients of different areas can be calculated, and the change condition of the groundwater level can be monitored, so that the detailed waterlogging hydrogeological parameters can be obtained. And after obtaining the waterlogging hydrogeological parameters, constructing a permeability coefficient-groundwater level change curve based on groundwater level change corresponding to the permeability coefficient. The curve can intuitively show the relation between the permeability coefficient and the groundwater level, and provides important reference information for subsequent optimal design. For example, a project team may use statistical software, such as R language or Python, to perform regression analysis on the extracted permeability coefficient and groundwater level data, and draw a permeability coefficient-groundwater level change curve. By analyzing the curve, the influence of groundwater level change in areas with high permeability coefficient can be identified, and the areas with high risk of waterlogging are possibly places with high risk and need important optimization. And carrying out multi-target optimizing calculation on the sponge city based on the permeability coefficient-groundwater level change curve by a preset multi-target optimizing algorithm to obtain a sponge facility layout candidate scheme set. The multi-objective optimization algorithm can trade-off among multiple objectives, finding the optimal solution or pareto frontier. For example, a project team may use a multi-objective optimization algorithm such as NSGA-II (Non-dominated Sorting Genetic Algorithm II) or MOEA/D (Multiobjective Evolutionary Algorithm based on Decomposition) to take as input a permeability coefficient-groundwater level change curve while taking into account other factors such as construction costs, environmental impact, etc., to perform multi-objective optimization calculations. Through multiple iterations, the algorithm can generate a set of candidate solutions for the sponge facility layout that strike a good balance among multiple targets. And performing spatial clustering analysis on the sponge facility layout candidate scheme set to obtain a layout scheme clustering result. Spatial cluster analysis can classify similar layout schemes into one category, thereby reducing the complexity of subsequent analysis. For example, a project team may use a K-means clustering algorithm or a DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm to perform cluster analysis on candidate solutions generated by the multi-objective optimization. By setting reasonable clustering parameters, the candidate schemes can be divided into a plurality of clusters, and schemes in each cluster have similar layout characteristics. These clustering results provide the basis for subsequent strategy game analysis. And carrying out strategy game analysis on each cluster representing scheme in the layout scheme clustering result through a preset game theory algorithm to obtain a game balanced layout scheme set. Game theory is a mathematical method for analyzing interactions and strategy selections among decision makers, and is particularly suitable for processing multi-subject optimization problems. For example, a project team may use a game theory approach such as Nash equalization (Nash Equilibrium) or Multi-agent reinforcement learning (Multi-Agent Reinforcement Learning) to conduct a strategy game analysis on the representative solutions in each cluster. By simulating interactions between different principals (e.g., city authorities, residents, businesses, etc.), one or more balanced layout schemes can be found that achieve an optimal balance between the benefits of the various principals. And finally, under a plurality of rainfall situations, carrying out layout optimization design on the sponge city based on the game balanced layout scheme set by using a neural network algorithm to obtain an optimal configuration scheme. The neural network algorithm can process complex nonlinear relations and is particularly suitable for prediction and optimization tasks. For example, a project team may construct a neural network model using a deep learning framework such as TensorFlow or PyTorch, input data including rainfall scenarios, a set of game balancing layout schemes, and a waterlogging risk map, etc. By training the model, the effect of each layout scheme under different rainfall situations can be predicted, so that one layout scheme with optimal performance under various rainfall situations is selected. The finally generated optimal configuration scheme not only can effectively reduce waterlogging risk, but also can give consideration to economy and environmental friendliness, and provides scientific basis for construction and management of sponge cities. In summary, the process of performing layout optimization design on the sponge city based on the waterlogging risk map by using the preset multi-objective optimization algorithm to obtain the optimal configuration scheme is a multi-step and multi-level complex task. The method comprises the steps of extracting hydrogeologic features, carrying out multi-objective optimizing calculation, carrying out spatial clustering analysis and strategy game analysis, and continuously optimizing and refining a layout scheme in each step, wherein the final purpose is to provide a more scientific and practical sponge city layout optimizing scheme, help cities to better cope with waterlogging risks, and realize sustainable development.
In a specific embodiment, the performing, by a preset multi-objective optimization algorithm, multi-objective optimization calculation on the target sponge city based on the permeability coefficient-groundwater level change curve to obtain a sponge facility layout candidate scheme set includes:
determining a change rule of the permeability coefficient and the groundwater level based on the permeability coefficient-groundwater level change curve through a preset multi-objective optimization algorithm;
Performing sensitivity analysis on the permeability coefficient based on the change rule to obtain the influence degree of the permeability coefficient on the groundwater level change under different rainfall situations;
Carrying out hydrological characteristic analysis on the sponge city based on the influence degree and the change rule to obtain a hydrological response mode and hydrological characteristic data in the waterlogging risk area;
And performing multi-objective optimizing calculation on the sponge city based on the hydrological response mode and the hydrological characteristic data to obtain a sponge facility layout candidate scheme set.
Specifically, the process of performing multi-objective optimizing calculation on the target sponge city based on the permeability coefficient-groundwater level change curve through a preset multi-objective optimizing algorithm to obtain a sponge facility layout candidate scheme set is a highly comprehensive and technical task, and involves multiple steps of determining permeability coefficient and groundwater level change rule, performing sensitivity analysis, performing hydrologic characteristic analysis, performing multi-objective optimizing calculation and the like. The main purpose of the process is to optimally design the layout of the sponge city by a scientific method and an advanced algorithm so as to reduce the risk of waterlogging to the maximum extent and improve the rainwater management capability of the city. Specifically, firstly, a change rule of the permeability coefficient and the groundwater level is determined based on the permeability coefficient-groundwater level change curve through a preset multi-objective optimization algorithm. The purpose of this step is to establish a mathematical relationship between the permeability coefficient and the groundwater level, providing a basis for subsequent analysis. For example, in sponge city construction projects, a project team may use statistical software, such as R language or Python, to perform regression analysis on historical data of permeability coefficients and groundwater level, fitting a curve reflecting the relationship between the two. The curve can intuitively show how the permeability coefficient influences the change of the groundwater level, and provides a basis for the subsequent sensitivity analysis. After the change rule of the permeability coefficient and the groundwater level is determined, sensitivity analysis is carried out on the permeability coefficient based on the change rule, and the influence degree of the permeability coefficient on the groundwater level change under different rainfall situations is obtained. Sensitivity analysis is a method for evaluating the influence of different parameter changes on the output of a system, and is particularly suitable for processing multivariable problems. For example, a project team may set different rainfall scenarios, such as light rain, medium rain, and heavy rain, and then observe the change in groundwater level by adjusting the value of the permeability coefficient in each rainfall scenario. In this way, the sensitivity of the permeability coefficient under different rainfall situations can be quantitatively evaluated, and the influence of the permeability coefficient change of which areas on the groundwater level is most obvious can be identified. These sensitivity analysis results provide important reference information for subsequent hydrologic characteristics analysis. And carrying out hydrologic characteristic analysis on the sponge city based on the influence degree and the change rule to obtain a hydrologic response mode and hydrologic characteristic data in the waterlogging risk area. The hydrologic characteristic analysis refers to identifying hydrologic response modes and characteristic parameters of different areas through comprehensive analysis of hydrologic data. For example, a project team may use a hydrological model, such as SWAT (Soil AND WATER ASSESSMENT Tool) or MIKE SHE (INTEGRATED HYDROLOGICAL MODELLING SYSTEM), in combination with the results of the sensitivity analysis to conduct detailed hydrological characterization of the waterlogging risk region. By simulating surface runoffs, groundwater flows and soil moisture dynamics under different rainfall situations, hydrological response modes of each area, such as a fast response type, a slow response type and the like, can be identified. Meanwhile, key hydrologic characteristic data such as runoff coefficient, groundwater supply rate, soil saturation and the like can be extracted. These hydrological response patterns and feature data provide important input information for subsequent multi-objective optimization calculations. And performing multi-objective optimizing calculation on the sponge city based on the hydrological response mode and the hydrological characteristic data to obtain a sponge facility layout candidate scheme set. The multi-objective optimization algorithm can trade-off among multiple objectives, finding the optimal solution or pareto frontier. For example, a project team may use a multi-objective optimization algorithm such as NSGA-II (Non-dominated Sorting Genetic Algorithm II) or MOEA/D (Multiobjective Evolutionary Algorithm based on Decomposition) to take the hydrologic response pattern and the characteristic data as input, while taking other factors into account, such as construction cost, environmental impact, community acceptance, etc., to perform multi-objective optimization calculations. Through multiple iterations, the algorithm can generate a set of candidate solutions for the sponge facility layout that strike a good balance among multiple targets. Each candidate scheme not only can effectively reduce waterlogging risk, but also can give consideration to economy and environmental friendliness, and provides various choices for city managers. In summary, the process of performing multi-objective optimization calculation on the target sponge city based on the permeability coefficient-groundwater level change curve through the preset multi-objective optimization algorithm to obtain the sponge facility layout candidate scheme set is a multi-step and multi-level complex task. From determining the change rule of the permeability coefficient and the groundwater level to sensitivity analysis, hydrologic characteristic analysis and multi-objective optimizing calculation, each step is continuously optimized and refined in layout scheme, and the final purpose is to provide a more scientific and practical sponge city layout optimization scheme, help cities to better cope with waterlogging risks and realize sustainable development. through the process, project teams can provide powerful technical support and decision basis for the construction and management of sponge cities.
The sponge city design optimization method based on rainfall measurement in the embodiment of the present invention is described above, and the sponge city design optimization system based on rainfall measurement in the embodiment of the present invention is described below, referring to fig. 2, and one embodiment of the sponge city design optimization system based on rainfall measurement in the embodiment of the present invention includes:
The first analysis module 21 is configured to obtain historical rainfall data and climate data of a target sponge city, and perform frequency distribution analysis on the historical rainfall data to obtain rainfall analysis data;
The prediction module 22 is configured to predict occurrence probability of a rainfall event in the target sponge city based on the rainfall analysis data and the climate data, so as to obtain probability distribution of future rainfall and future measured rainfall;
The second analysis module 23 is configured to perform spatial analysis on the target sponge city through a preset geographic information system, so as to obtain a distribution diagram of easy-to-accumulate water points;
The evaluation module 24 is configured to evaluate a waterlogging risk level of each region in the target sponge city based on the easy-to-accumulate-point distribution map, the future measured rainfall and the probability distribution of the future rainfall by using a bayesian network algorithm, so as to obtain a waterlogging risk map, where the waterlogging risk map includes a high risk region, a medium risk region and a low risk region;
The design module 25 is configured to perform layout optimization design on the target sponge city based on the waterlogging risk map by using a preset multi-objective optimization algorithm, so as to obtain an optimal configuration scheme.
In this embodiment, for specific implementation of each unit in the above system embodiment, please refer to the description in the above method embodiment, and no further description is given here.
Referring to fig. 3, a computer device is further provided in an embodiment of the present invention, and the internal structure of the computer device may be as shown in fig. 3. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected by a system bus. Wherein the computer is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used for communicating with an external terminal through a network connection. Which computer program, when being executed by a processor, carries out the above-mentioned method.
It will be appreciated by those skilled in the art that the architecture shown in fig. 3 is merely a block diagram of a portion of the architecture in connection with the present inventive arrangements and is not intended to limit the computer devices to which the present inventive arrangements are applicable.
An embodiment of the present invention also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the above method. It is understood that the computer readable storage medium in this embodiment may be a volatile readable storage medium or a nonvolatile readable storage medium.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium provided by the present invention and used in embodiments may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (SYNCHLINK) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising one does not exclude the presence of other like elements in a process, apparatus, article, or method that comprises the element.
The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the invention, and all equivalent structures or equivalent processes using the descriptions and drawings of the present invention or direct or indirect application in other related technical fields are included in the scope of the present invention.