Disclosure of Invention
In order to solve the problems of low energy management efficiency, poor prediction accuracy and the like in photovoltaic energy storage allocation in the prior art, the invention designs a photovoltaic energy storage allocation method and a photovoltaic energy storage allocation system which can effectively solve the problems.
In order to solve the technical problems, the technical scheme of the invention is as follows:
a photovoltaic energy storage allocation method comprises the following steps:
acquiring environmental data and historical power generation data of a photovoltaic energy storage system;
acquiring weather forecast data of different time periods through a set network weather service platform;
Training a photovoltaic power generation amount prediction model and an electric load demand prediction model based on the acquired data;
Generating energy allocation plans aiming at different time periods and different preset conditions based on the predicted photovoltaic power generation amount and the power load demand;
Selecting an optimal energy configuration scheme from the energy distribution schemes, and performing energy supply control on an energy storage unit of the photovoltaic energy storage system based on the selected scheme;
Collecting operation data of the energy storage unit in real time, evaluating the efficiency level of the energy storage unit, and correcting the energy distribution plan according to an evaluation result;
And respectively carrying out energy supply test on the photovoltaic energy storage system through each energy allocation scheme in the corrected energy allocation scheme, and selecting the energy allocation scheme with the highest priority weight as a final energy supply execution scheme.
Preferably, the training the photovoltaic power generation amount prediction model and the power load demand prediction model based on the acquired data further includes:
acquiring multidimensional data samples of environmental data and historical power generation data of a fusion weather forecast data and a photovoltaic energy storage system;
Deep feature extraction is carried out on the multidimensional data sample, and a first feature vector representing meteorological variation and a second feature vector depicting the performance of the photovoltaic energy storage system are generated;
Respectively carrying out depth coding on the first feature vector and the second feature vector to obtain a first fine-granularity feature vector related to meteorological parameters and a second fine-granularity feature vector related to equipment performance;
dynamically integrating the first fine-grain feature vector and the second fine-grain feature vector to generate a comprehensive target feature vector, and calculating the dynamic alignment loss of the photovoltaic power generation capacity prediction model based on the target feature vector;
And dynamically adjusting and optimizing model parameters of the photovoltaic power generation amount prediction model based on the initial loss and the dynamic alignment loss of the photovoltaic power generation amount prediction model.
Preferably, the generating the energy distribution scheme for different time periods and different preset conditions based on the predicted photovoltaic power generation and the power load demand further comprises:
Dividing the predicted photovoltaic power generation amount and power load demand data according to a time sequence to form a plurality of continuous time period sets, wherein each time period set comprises a photovoltaic power generation amount predicted value and a power load demand predicted value in a corresponding time period;
Setting different energy allocation strategies according to preset conditions for each time period set;
calculating the distribution proportion and specific value of each energy type in each time period by combining a photovoltaic power generation quantity predicted value and an electric power load demand predicted value based on a set energy distribution strategy, and generating a preliminary energy distribution plan;
and verifying the generated preliminary energy distribution plan, ensuring that the plan meets the requirements of safe and stable operation of the power grid, and carrying out necessary adjustment and optimization on the plan by considering the charge and discharge efficiency and service life management of the photovoltaic energy storage system.
Preferably, the selecting an optimal energy configuration scheme from the energy distribution schemes, and controlling energy supply to the energy storage unit of the photovoltaic energy storage system based on the selected scheme further includes:
Comprehensively evaluating the generated multiple energy distribution plans according to preset evaluation indexes;
Selecting an energy configuration scheme with the optimal comprehensive evaluation as an optimal scheme under the current time period and the preset condition;
and carrying out energy supply management on the energy storage unit of the photovoltaic energy storage system based on the selected optimal scheme.
Preferably, the collecting the operation data of the energy storage unit in real time, evaluating the efficiency level of the energy storage unit, and correcting the energy distribution plan according to the evaluation result further includes:
collecting operation data of the photovoltaic energy storage system in real time through a sensor;
Processing and analyzing the collected operation data by utilizing a data analysis technology, and evaluating the efficiency level of the photovoltaic energy storage system;
And correcting the energy distribution plan according to the evaluation result.
Preferably, the energy supply test is performed on the photovoltaic energy storage system by each energy allocation scheme in the modified energy allocation scheme, and selecting the energy allocation scheme with the highest priority weight as the final energy supply execution scheme further includes:
respectively applying each energy allocation scheme in the corrected energy allocation scheme to a photovoltaic energy storage system to perform energy supply test;
In the testing process, monitoring the running state and performance index of the photovoltaic energy storage system in real time;
And selecting the energy configuration scheme with the highest priority weight as a final energy supply execution scheme according to the monitoring result.
A photovoltaic energy storage deployment system, comprising:
The data acquisition module is used for acquiring environmental data and historical power generation data of the photovoltaic energy storage system;
the weather data acquisition module acquires weather forecast data of different time periods in real time or periodically through a set network weather service platform;
The prediction model module is used for training a photovoltaic power generation amount prediction model and an electric load demand prediction model based on the collected environmental data, historical power generation data and meteorological prediction data;
The plan generation module is used for generating an energy distribution plan aiming at energy distribution plans under different time periods and different conditions based on the predicted photovoltaic power generation amount and power load requirements and combining different preset conditions;
The configuration scheme selection module is used for selecting an optimal energy configuration scheme from the energy distribution schemes;
the energy supply control module is used for controlling energy supply of the energy storage unit of the photovoltaic energy storage system based on the selected energy configuration scheme;
The efficiency evaluation and correction module is used for collecting operation data of the photovoltaic energy storage system in real time, evaluating the energy storage efficiency level and correcting the energy distribution plan according to the evaluation result;
And the testing and scheme determining module is used for respectively carrying out energy supply testing on the photovoltaic energy storage system through each energy allocation scheme in the corrected energy allocation scheme, and selecting the energy allocation scheme with the highest priority weight as a final energy supply execution scheme.
The method has the advantages that the environment data and the historical power generation data of the photovoltaic energy storage system are collected, weather forecast data provided by a network weather service platform are combined, the technical scheme can fully utilize multidimensional data resources to construct an accurate photovoltaic power generation capacity forecast model and an electric load demand forecast model, the data fusion and depth feature extraction method remarkably improves the forecast accuracy, the energy distribution plan is more close to reality, the problem of low energy management efficiency caused by forecast errors is solved, secondly, the time sequence division and the setting of preset conditions can be used for generating the energy distribution plan aiming at different time periods and different conditions, the diversity and the practicability of the plan are guaranteed, meanwhile, the generated plan is verified and adjusted, the requirement of safe and stable operation of a power grid is met, the charge and discharge efficiency and the service life management of the photovoltaic energy storage system are further improved, the operation data of the photovoltaic energy storage system are collected in real time through a sensor, the data analysis technology is used for processing and analyzing the data, the energy distribution plan can be optimized according to the level efficiency of the energy storage unit, the energy distribution plan is estimated, the energy distribution plan is more flexibly estimated, and the energy distribution plan is more practical, and the energy distribution and energy distribution situation is more flexible.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the present patent;
for the purpose of better illustrating the embodiments, certain elements of the drawings may be omitted, enlarged or reduced and do not represent the actual product dimensions;
It will be appreciated by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The technical scheme of the invention is further described below with reference to the accompanying drawings and examples.
Examples
A photovoltaic energy storage allocation method, as shown in figure 1, comprises the following steps:
acquiring environmental data and historical power generation data of a photovoltaic energy storage system;
weather forecast data of different time periods are acquired through a set network weather service platform, and the data are critical to the prediction of photovoltaic power generation capacity, including but not limited to illumination intensity, wind speed, precipitation probability and the like.
Training a photovoltaic power generation amount prediction model and an electric load demand prediction model based on the acquired data;
Generating energy allocation plans aiming at different time periods and different preset conditions based on the predicted photovoltaic power generation amount and the power load demand;
Selecting an optimal energy configuration scheme from the energy distribution schemes, and performing energy supply control on an energy storage unit of the photovoltaic energy storage system based on the selected scheme;
Collecting operation data of the energy storage unit in real time, evaluating the efficiency level of the energy storage unit, and correcting the energy distribution plan according to an evaluation result;
And respectively carrying out energy supply test on the photovoltaic energy storage system through each energy allocation scheme in the corrected energy allocation scheme, and selecting the energy allocation scheme with the highest priority weight as a final energy supply execution scheme.
Based on the acquired data, training the photovoltaic power generation amount prediction model and the power load demand prediction model further comprises:
acquiring multidimensional data samples of environmental data and historical power generation data of a fusion weather forecast data and a photovoltaic energy storage system;
Deep feature extraction is carried out on the multidimensional data sample, and a first feature vector representing meteorological variation and a second feature vector depicting the performance of the photovoltaic energy storage system are generated;
Respectively carrying out depth coding on the first feature vector and the second feature vector to obtain a first fine-granularity feature vector related to meteorological parameters and a second fine-granularity feature vector related to equipment performance;
dynamically integrating the first fine-grain feature vector and the second fine-grain feature vector to generate a comprehensive target feature vector, and calculating the dynamic alignment loss of the photovoltaic power generation capacity prediction model based on the target feature vector;
And dynamically adjusting and optimizing model parameters of the photovoltaic power generation amount prediction model based on the initial loss and the dynamic alignment loss of the photovoltaic power generation amount prediction model.
Suitable machine learning algorithms, such as support vector machines, time series analysis, random forests, neural networks, etc., are selected as the basis for the predictive model.
The generating of the energy distribution scheme for different time periods and different preset conditions based on the predicted photovoltaic power generation and the power load demand further comprises:
Dividing the predicted photovoltaic power generation amount and power load demand data according to a time sequence to form a plurality of continuous time period sets, wherein each time period set comprises a photovoltaic power generation amount predicted value and a power load demand predicted value in a corresponding time period;
Different energy allocation strategies are set according to preset conditions, such as energy cost, environmental constraint, system stability requirement and the like, for each time period set, and include, but are not limited to, preferentially using photovoltaic power generation, enabling a photovoltaic energy storage system to supplement power, calling a standby power supply and the like.
Calculating the distribution proportion and specific value of each energy type in each time period by combining a photovoltaic power generation quantity predicted value and an electric power load demand predicted value based on a set energy distribution strategy, and generating a preliminary energy distribution plan;
and verifying the generated preliminary energy distribution plan, ensuring that the plan meets the requirements of safe and stable operation of the power grid, and carrying out necessary adjustment and optimization on the plan by considering the charge and discharge efficiency and service life management of the photovoltaic energy storage system.
The selecting an optimal energy configuration scheme from the energy distribution schemes, and controlling energy supply to the energy storage unit of the photovoltaic energy storage system based on the selected scheme further comprises:
comprehensively evaluating the generated multiple energy allocation plans according to preset evaluation indexes, such as the lowest energy cost, the lowest environmental impact, the highest system efficiency and the like;
Selecting an energy configuration scheme with the optimal comprehensive evaluation as an optimal scheme under the current time period and the preset condition;
Based on the selected optimal scheme, energy supply management is carried out on the energy storage unit of the photovoltaic energy storage system, and the energy supply management comprises the steps of setting a charging and discharging strategy of the photovoltaic energy storage system, controlling input and output power of the photovoltaic energy storage system and the like.
The real-time operation data of the energy storage unit is collected, the efficiency level of the energy storage unit is estimated, and the energy distribution plan is corrected according to the estimation result, and the method further comprises the following steps:
The operation data of the photovoltaic energy storage system are collected in real time through the sensor, and the operation data comprise key parameters such as electric quantity, temperature, voltage and current of the photovoltaic energy storage system.
And processing and analyzing the collected operation data by utilizing a data analysis technology, and evaluating the efficiency level of the photovoltaic energy storage system, wherein the efficiency level comprises energy storage efficiency, cycle life and the like.
And correcting the energy distribution plan according to the evaluation result, such as adjusting the charging and discharging strategy of the photovoltaic energy storage system, optimizing the energy distribution proportion and the like, so as to improve the efficiency of the photovoltaic energy storage system and prolong the service life of the photovoltaic energy storage system.
The energy supply test is carried out on the photovoltaic energy storage system through each energy allocation scheme in the corrected energy allocation scheme, and the selection of the energy allocation scheme with the highest priority weight as the final energy supply execution scheme further comprises the following steps:
respectively applying each energy allocation scheme in the corrected energy allocation scheme to a photovoltaic energy storage system to perform energy supply test;
In the testing process, the running state and performance index of the photovoltaic energy storage system are monitored in real time, such as system stability, energy utilization efficiency and the like.
According to the monitoring result, the energy configuration scheme with the highest priority weight is selected as the final energy supply execution scheme, and the determination of the priority weight can be based on a plurality of factors, such as energy cost, environmental impact, system efficiency, and the like.
In the implementation, firstly, environmental data such as temperature, humidity, illumination intensity and the like and historical power generation data such as power generation capacity, power generation efficiency and the like in a past period are collected through sensors and a network of the photovoltaic energy storage system, and meanwhile, weather forecast data including but not limited to sunlight time, cloud cover, wind speed and the like in a future period are obtained through a set network weather service platform.
The method comprises the steps of obtaining a multi-dimensional data sample which is fused with weather forecast data and photovoltaic energy storage system data, carrying out deep feature extraction to generate feature vectors which represent weather variation and photovoltaic energy storage system performance, then carrying out depth coding to obtain fine-grained feature vectors, dynamically integrating the fine-grained feature vectors to generate comprehensive target feature vectors, and calculating dynamic alignment loss of the photovoltaic energy generation prediction model so as to dynamically adjust and optimize model parameters.
Based on a trained prediction model, predicting photovoltaic power generation capacity and power load demand in a future period, dividing prediction data into a plurality of continuous time period sets according to a time sequence, wherein each set comprises a photovoltaic power generation capacity predicted value and a power load demand predicted value in a corresponding time period, setting different energy distribution strategies according to preset conditions, such as power grid safe and stable operation requirements, charging and discharging efficiency and service life management of a photovoltaic energy storage system, energy cost, environmental constraint, system stability requirements and the like, calculating distribution proportion and specific values of each energy type in each time period, generating a preliminary energy distribution plan, and checking and adjusting the generated plan to ensure that the generated plan meets all preset conditions.
Then, comprehensively evaluating the generated multiple energy distribution plans according to preset evaluation indexes such as lowest energy cost, lowest environmental impact, highest system efficiency and the like, selecting an energy configuration scheme with the optimal comprehensive evaluation as an optimal scheme under the current time period and preset conditions, and performing energy supply management on an energy storage unit of the photovoltaic energy storage system based on the selected optimal scheme.
In the energy supply process, operation data of the photovoltaic energy storage system, such as electric quantity, temperature, voltage and the like of an energy storage unit, are collected in real time through a sensor, the data are processed and analyzed by utilizing a data analysis technology, the efficiency level of the photovoltaic energy storage system is estimated, and an energy distribution plan is dynamically adjusted according to an estimation result so as to optimize the operation efficiency and stability of the system.
And finally, respectively applying each energy configuration scheme in the corrected energy distribution scheme to the photovoltaic energy storage system to carry out energy supply test, and in the test process, monitoring the running state and performance indexes of the system, such as power generation efficiency, energy storage efficiency, system stability and the like in real time, and selecting the energy configuration scheme with the highest priority weight as a final energy supply execution scheme according to the monitoring result so as to realize optimal allocation of the photovoltaic energy storage system.
Also provided is a computer readable storage medium, the computer readable storage medium including a photovoltaic energy storage deployment method program, which when executed by a processor, implements the steps of a photovoltaic energy storage deployment method as described above
A photovoltaic energy storage deployment system, as shown in fig. 2, comprising:
The data acquisition module is used for acquiring environmental data and historical power generation data of the photovoltaic energy storage system;
the weather data acquisition module acquires weather forecast data of different time periods in real time or periodically through a set network weather service platform;
The prediction model module is used for training a photovoltaic power generation amount prediction model and an electric load demand prediction model based on the collected environmental data, historical power generation data and meteorological prediction data;
The plan generation module is used for generating an energy distribution plan aiming at energy distribution plans under different time periods and different conditions based on the predicted photovoltaic power generation amount and power load requirements and combining different preset conditions;
The configuration scheme selection module is used for selecting an optimal energy configuration scheme from the energy distribution schemes;
the energy supply control module is used for controlling energy supply of the energy storage unit of the photovoltaic energy storage system based on the selected energy configuration scheme;
The efficiency evaluation and correction module is used for collecting operation data of the photovoltaic energy storage system in real time, evaluating the energy storage efficiency level and correcting the energy distribution plan according to the evaluation result;
And the testing and scheme determining module is used for respectively carrying out energy supply testing on the photovoltaic energy storage system through each energy allocation scheme in the corrected energy allocation scheme, and selecting the energy allocation scheme with the highest priority weight as a final energy supply execution scheme.
In specific implementation, a photovoltaic energy storage allocation system is constructed, and the system consists of a data acquisition module, a meteorological data acquisition module, a prediction model module, a plan generation module, a configuration plan selection module, an energy supply control module, a performance evaluation and correction module and a test and plan determination module, wherein the system is arranged in a control center of the photovoltaic energy storage system, so that data interaction and command transmission smoothness among all modules are ensured.
After the system is built, initialization setting is carried out, wherein the initialization setting comprises setting the frequency of data acquisition, the timing task of a meteorological data acquisition module, the training parameters of a prediction model and the like, and meanwhile, energy allocation plan templates under different conditions are defined and preset, so that a foundation is provided for subsequent automatic generation of plans.
The data acquisition module periodically collects environmental data such as temperature, humidity, illumination intensity and the like and historical power generation data such as generated energy, power generation efficiency and the like from a sensor network of the photovoltaic energy storage system, and the data are stored in a database of the system and are used for subsequent analysis and prediction.
The weather data acquisition module acquires weather forecast data of the future week including illumination intensity, temperature, wind speed and the like at regular time every day through a set network weather service platform, and the data are also stored in a database and used for training a prediction model together with environment data and historical power generation data.
The prediction model module utilizes collected environmental data, historical power generation data and weather forecast data, adopts a machine learning algorithm, such as time sequence analysis, a neural network and the like to train a photovoltaic power generation amount prediction model and a power load demand prediction model, can predict photovoltaic power generation amount and power load demand in a future period, and in order to improve the accuracy of prediction, the model needs to be updated and retrained regularly, and a system can automatically trigger the updating process of the model whenever new data is collected and stored.
The plan generation module automatically generates an energy distribution plan aiming at different time periods and under different conditions according to the predicted photovoltaic power generation amount and power load requirements and by combining different preset conditions such as energy price, power grid stability, environmental protection policy and the like.
The configuration scheme selection module selects an optimal energy configuration scheme from the generated energy allocation schemes, and the process involves a trade-off of a plurality of factors, such as cost effectiveness, environmental impact, energy utilization rate and the like, and the system can sort the schemes according to preset priority weights to select the optimal scheme.
The energy supply control module is used for controlling energy supply of the energy storage unit of the photovoltaic energy storage system according to the selected energy configuration scheme, and the energy supply control module comprises the steps of adjusting the power generation of the photovoltaic panel, the charging and discharging strategies of the energy storage battery and the like.
The efficiency evaluation and correction module collects operation data of the photovoltaic energy storage system in real time, evaluates energy storage efficiency levels, such as energy storage efficiency and energy loss, and if the energy storage efficiency levels are found to be lower than expected, the system corrects the energy distribution scheme so as to improve the energy storage efficiency.
The testing and scheme determining module respectively performs energy supply tests on the photovoltaic energy storage system through each energy allocation scheme in the modified energy allocation scheme, and the tests aim at verifying feasibility and effect of the scheme.
After the test is completed, the system selects the energy configuration scheme with the highest priority weight as the final energy supply execution scheme according to the test result, and the scheme is transmitted to the energy supply control module for guiding the actual operation of the photovoltaic energy storage system.
The same or similar reference numerals correspond to the same or similar components;
The terms describing the positional relationship in the drawings are merely illustrative, and are not to be construed as limiting the present patent;
It is to be understood that the above-described embodiments of the present invention are provided by way of illustration only and not as limitations of the embodiments of the present invention, and that various other changes and modifications may be made by one skilled in the art based on the above description, without the necessity of or without intending to be exhaustive of all embodiments, and any modifications, equivalents, improvements and modifications etc. within the spirit and principles of the present invention are intended to be included within the scope of the appended claims.