Detailed Description
The present application 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 application 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 application.
The power grid operation risk assessment method provided by the embodiment of the application can be applied to an application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The data storage system may store data that the server 104 needs to process. The data storage system may be integrated on the server 104 or may be located on a cloud or other network server.
The server 104 acquires power grid operation state data and meteorological monitoring data; generating a power grid section data set according to the power grid running state data and the meteorological monitoring data; in an off-line stage of the power grid, constructing and generating an countermeasure network model according to the power grid section data set; acquiring at least one type of power grid operation risk data by generating an countermeasure network model; generating a power grid operation risk scene set according to at least one type of power grid operation risk data, wherein the power grid operation risk scene set comprises at least one power grid operation risk scene; performing deduction calculation on at least one power grid operation risk scene by adopting a time sequence operation simulation method to generate at least one power grid look-ahead scene sequence; acquiring a preset risk assessment index and current power grid operation scene information in an online operation stage of the power grid; according to the preset risk assessment index and the current power grid operation scene information, a power grid operation risk assessment result is obtained, and a corresponding power grid prospective operation scene sequence is obtained through matching.
The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices, and portable wearable devices, where the internet of things devices may be smart speakers, smart televisions, smart air conditioners, smart vehicle devices, and the like. The portable wearable device may be a smart watch, smart bracelet, headset, or the like. The server 104 may be implemented as a stand-alone server or as a server cluster of multiple servers.
In an exemplary embodiment, as shown in fig. 2, a power grid operation risk assessment method is provided, and an example of application of the method to the server in fig. 1 is described, which includes the following steps S202 to S216.
Wherein:
Step S202, power grid running state data and meteorological monitoring data are obtained.
Specifically, the power grid operation state data obtained from the electric measurement system specifically includes system power flow data, topology connection information, connection load information and the like, and the obtained meteorological monitoring data specifically includes grid meteorological data such as grid wind speed, wind direction, temperature, humidity and the like.
Step S204, a power grid section data set is generated according to the power grid running state data and the meteorological monitoring data.
Specifically, the collected power grid operation state data and meteorological monitoring data are preprocessed and integrated, and the preprocessing comprises data cleaning, denoising, correction and the like. And ensuring the accuracy and consistency of the data. And selecting specific section ranges and parameters according to the specific purpose of grid analysis and evaluation. These parameters may be grid components within a particular area or particular grid characteristics, such as transmission line length, substation capacity, etc. And generating a power grid section data set by utilizing the preprocessed and integrated data, and marking and classifying the power grid section data set. The marking can be performed according to indexes such as the running risk level, the fault probability and the like of the power grid so as to facilitate subsequent risk assessment and analysis. And then, verifying and validating the generated power grid section data set to ensure the integrity and accuracy of the data. This may include comparison and comparative analysis with actual observed data.
The grid section data set expressed in step S204 is specifically expressed as:
N＝[n1,n2,...ni,...,nm]；
Wherein ni is a characteristic information vector of the ith section, smi is the mth electrical measurement information of the ith section, and wmi is the mth weather measurement information of the ith section.
Step S206, constructing and generating an countermeasure network model according to the power grid section data set in an off-line stage of the power grid.
Specifically, in the offline stage of the power grid, building a generation countermeasure network (GAN) model according to the power grid section dataset refers to utilizing the generation capability of the GAN model, and training the GAN model using the power grid section dataset as an input to learn and generate virtual power grid section data with similar characteristics.
Generating an countermeasure network is a neural network model that consists of a generator and a arbiter. Training by countermeasure allows the generator to gradually generate virtual data that is similar to real data, while the arbiter attempts to accurately distinguish between real data and generated data. This countermeasure training results in a gradual increase in the performance of the generator, and the generated data also more and more closely approximates the distribution of real data.
On the basis of constructing the grid section dataset, more virtual grid section data can be generated by using the GAN model. The GAN model can simulate a new composite dataset by learning correlations and distributions between grid profile data. To train this GAN model, it is necessary to define a suitable loss function for evaluating the difference between the data generated by the generator and the real data and updating the parameters of the model by means of a back propagation algorithm.
The constructed GAN model may be used to generate virtual grid section data similar to the grid section data set. Such virtual data can be used to supplement the deficiencies of existing data sets, expand the training set scale, and improve the generalization ability of the model. Meanwhile, the generated virtual data can be used for simulating various risk scenes in the running process of the power grid and used for risk assessment and decision support.
By constructing and generating the countermeasure network model and training with the grid section dataset, a model can be obtained that can generate virtual grid section data that is similar to real grid data. The model can generate data with similar characteristics and statistical properties by learning the distribution and characteristics of the power grid data, and provides more data resources for power grid operation risk assessment and related applications.
Step S208, obtaining at least one type of grid operation risk data by generating an countermeasure network model.
Specifically, a trained GAN model is used to generate a new batch of grid operational risk data. The generated data will have similar characteristics and distribution as the training data. And extracting the characteristics of the generated data and the original data to obtain a high-dimensional risk characteristic vector. Various feature extraction methods, such as Principal Component Analysis (PCA) or other dimension reduction techniques, may be used.
Step S210, generating a power grid operation risk scene set according to at least one type of power grid operation risk data, wherein the power grid operation risk scene set comprises at least one power grid operation risk scene, and calculating and screening high-dimensional risk characteristics through a K-Means clustering method.
In particular, one common scenario in which the risk of grid operation is known is power load fluctuations. During grid operation, load fluctuations may lead to power system overloads, voltage fluctuations, and other potential risks. The following is an example scenario set, which includes a grid operation risk scenario caused by load fluctuations:
scene 1: voltage drop during high load
Description of: during high loads, the power system is faced with a huge load demand, exceeding the supply capacity. This results in a drop in supply voltage, which may lead to a malfunction or power outage of the user equipment.
The characteristics are as follows: high load, supply voltage drop, potential user equipment failure or power outage risk.
The solution is as follows: increasing power generation capacity, improving power system scheduling and management strategies, and optimizing power supply networks.
Scene 2: sudden load increase
Description of: due to certain events (e.g., sudden weather, equipment failure, etc.), the power system is subject to sudden load increases, exceeding the capacity of the system, which may lead to overloads and voltage drops.
The characteristics are as follows: sudden load increases, power system capacity overloads, potential overloads, and voltage drop risks.
The solution is as follows: implementing load management policies, responding to load changes in time, improving schedulability of power supply and distribution equipment.
Scene 3: unbalanced load distribution
Description of: due to uneven load distribution or power supply system failure, some areas or lines carry excessive load, which may cause overload of a partial area of the grid, while other partial loads are too light.
The characteristics are as follows: unbalanced load distribution, overload of the area or line, and lighter load in other areas.
The solution is as follows: optimizing load distribution, improving power supply network planning, expanding the power transmission and distribution capacity of partial areas.
These risk scenarios provide examples of the risks that may be posed by load fluctuations in grid operation and their potential solutions. Note that the actual grid risk scenario may be more complex, requiring determination based on specific grid data and conditions.
The extracted high-dimensional risk feature vector is input into a K-Means clustering algorithm, and risk data are divided into different clusters. K-Means is a commonly used unsupervised learning algorithm that automatically clusters similar data points together. The high-dimensional risk characteristics can be calculated and screened through the K-Means clustering method to help identify and analyze the risk modes and abnormal conditions in the power grid operation scene. This may provide valuable information to the power system manager to take corresponding measures to address potential risk issues.
Step S212, performing deduction calculation on at least one power grid operation risk scene by adopting a time sequence operation simulation method, and generating at least one power grid look-ahead scene sequence.
Specifically, the time sequence operation simulation method is a commonly used power grid risk assessment tool, and a power grid look-ahead scene sequence is deduced and calculated by simulating the power grid operation condition. Firstly, one or more power grid operation risk scenes such as load sudden increase, power transmission line faults and the like are selected, and parameters required by simulation such as occurrence probability, repair time and the like are determined. And then, performing power grid operation simulation calculation according to the determined risk scene and parameters by a time sequence operation simulation method. The simulation can consider the conditions of load change, start-stop of the generator set, fault occurrence, repair and the like. According to the simulation calculation result, the state of the power grid at different time points is obtained, and information on reliability, stability, vulnerability and the like of the power grid can be analyzed. Possible problems can be identified and future grid operation can be predicted. And generating a power grid look-ahead scene sequence according to the analysis result. These sequences may help the grid operator to formulate a reasonable operating strategy to cope with future risk scenarios that may occur.
Step S214, in the online operation stage of the power grid, acquiring a preset risk assessment index and current power grid operation scene information.
Specifically, in an online operation stage of the power grid, real-time power grid operation data including parameters such as power load, generator set output, voltage and current are obtained through monitoring equipment such as sensors and monitoring devices. These data can be used to evaluate the operating state and performance of the grid. Relevant environmental data such as weather data, air temperature, humidity, etc. are acquired. These data may affect the operation of the grid, such as the intensity of wind in a wind power generation scenario. And (3) formulating a series of risk assessment indexes according to the safety and reliability requirements of the power grid. These indicators can be used to assess the risk level of the grid in different scenarios, such as voltage stability, frequency stability, capacity utilization, etc. These indices may be set and modified according to actual conditions. And analyzing and reviewing the running condition of the power grid by utilizing the historical data and the fault records to acquire the past running scene information of the power grid. Such information may be used to improve the accuracy of risk assessment and identify possible problems and risks. And calculating and evaluating the risk of the power grid by using a corresponding risk evaluation model and algorithm according to the preset risk evaluation index and the collected power grid operation scene information. The models and algorithms can be selected and adjusted according to actual requirements and scenes to achieve accurate risk assessment.
Step S216, according to preset risk assessment indexes and current power grid operation scene information, power grid operation risk assessment results are obtained, and corresponding power grid prospective operation scene sequences are obtained in a matching mode.
Specifically, the collected power grid operation scene information and a preset risk assessment index are utilized, and the corresponding risk assessment model and algorithm are utilized to assess the operation risk of the power grid. And judging the current risk level of the power grid according to the preset index and the evaluation result. And matching corresponding power grid prospective operation scene sequences according to the risk assessment result. These sequences may be pre-planned strategies and measures for coping with grid operation at different risk levels. For example, in the event of a sudden increase in electrical load, there may be a corresponding genset start-stop strategy; under the condition of power transmission line faults, corresponding line switching and repairing strategies can be provided. And further analyzing and optimizing the operation strategy of the power grid according to the power grid look-ahead operation scene sequence. This may include adjusting the output of the genset, optimizing power load scheduling, formulating backup power strategies, etc. to improve the safety and reliability of the grid.
In the power grid operation risk assessment method, the power grid section data set is generated by acquiring the power grid operation state data and the meteorological monitoring data, and then the power grid operation risk data is generated by utilizing the generation countermeasure network model. The method utilizes the advantages of the GAN model, can more accurately generate the power grid operation risk data, and further improves the accuracy of risk assessment. By using the power grid operation risk data generated by the generation countermeasure network model, various power grid operation risk scenes can be generated. And then, carrying out deduction calculation on the risk scenes by using a time sequence operation simulation method to generate a power grid look-ahead scene sequence. Therefore, the running conditions of the power grid under different risk scenes can be better evaluated, and possible problems can be prejudged in advance. And in the online operation stage of the power grid, acquiring a preset risk assessment index and current power grid operation scene information, acquiring a power grid operation risk assessment result according to the information, and matching a corresponding power grid prospective operation scene sequence. The process can be performed in real time, provides real-time risk assessment results, and helps a decision maker to quickly make adjustments in the running process. In conclusion, the novel power grid operation risk assessment method utilizes the GAN model and the time sequence operation simulation method, so that the accuracy and timeliness of power grid risk assessment can be improved, and the power grid operation is more stable and reliable.
In an exemplary embodiment, as shown in fig. 3, the step of performing a time sequence operation simulation method to perform deduction calculation on at least one power grid operation risk scene to generate at least one power grid look-ahead scene sequence includes steps S302 to S306. Wherein: the time sequence operation simulation method comprises a Monte Carlo decision tree;
Step S302, taking the cross entropy function with the weight as a loss function of a Monte Carlo decision tree, and acquiring an evaluation value of at least one power grid operation risk scene through the Monte Carlo decision tree.
And step S304, dividing and processing the evaluation value of at least one power grid operation risk scene to obtain a processing result.
Step S306, adopting a parallel deduction architecture, and performing multithread parallel deduction calculation on the processing result through a Monte Carlo decision tree to generate at least one power grid look-ahead scene sequence. And (3) providing branching Cheng Tuiyan calculation for the multi-branch scene, and outputting a power grid operation deduction scene sequence under different targets including new energy consumption, economic optimization, safety limitation and the like.
Specifically, the weighted cross entropy function can be used as a loss function of the Monte Carlo decision tree to evaluate the grid operation risk scenario. The Monte Carlo decision tree is a probability-based decision tree model, and uncertainty factors and weights can be fully considered. When the Monte Carlo decision tree is constructed, the nodes and boundary conditions of the decision tree can be set according to the characteristics of the power grid and risk scenes to be evaluated. These conditions may include power load level, genset status, transmission line parameters, and the like. Ensuring the rationality and operability of the decision tree. The decision tree is constructed and optimized using the monte carlo method. And randomly sampling and simulating and evaluating by a Monte Carlo method to obtain a sampled power grid operation risk scene set and operation information of each scene, and calculating risk evaluation values under different decision paths.
Specifically, a weighted cross entropy function is used to calculate a risk assessment value, and the result is adjusted according to the weights. And finally, obtaining one or more power grid operation risk scenes with higher evaluation values. And for the power grid operation risk scene with higher evaluation value, the classification processing can be carried out according to the actual requirements and the priorities. For example, the evaluation values may be divided into different priorities for better handling and control.
Specifically, a parallel deduction architecture is adopted, and multithread parallel deduction calculation is carried out on the processing result through a Monte Carlo decision tree, so that at least one power grid look-ahead scene sequence is generated. These prospective scenario sequences may be used to formulate power grid emergency handling policies, adjust power load scheduling, etc. to handle different power grid operational risks.
In this embodiment, through evaluation and multi-thread parallel deduction calculation of the monte carlo decision tree, a power grid look-ahead scene sequence based on probability and weight can be provided to assist decision making and planning measures, so that the safety and reliability of power grid operation are improved.
In an exemplary embodiment, dividing the evaluation value of at least one power grid operation risk scenario to obtain a processing result includes:
Obtaining the maximum deduction expected evaluation value of each power grid operation risk scene through a preset evaluation value calculation formula;
Dividing at least one power grid operation risk scene according to the maximum deduction expected evaluation value to obtain a processing result with a multi-section power grid operation risk scene sequence;
Wherein, the evaluation value calculation formula is:
Wherein, E is a desired operator; t is the access step length; q (vt) is the prize value accessed to scene vt; n (vp) is the number of times the parent scene is accessed; n (vc) is the number of accesses to the sub-scene; vpen,i is an evaluation penalty term; gamma is a constant coefficient and gamma >0.
In this embodiment, according to a given evaluation value calculation formula, deduction calculation is performed on each power grid operation risk scene. And calculating the maximum deduction expected evaluation value of each scene according to the expected operator and other related parameters. And dividing each power grid operation risk scene according to the maximum deduction expected evaluation value. The grid operation risk scenario may be divided into a plurality of paragraphs according to different ranges of evaluation values or set thresholds, each paragraph representing a different risk level or processing priority. And combining the power grid operation risk scene sequences of all the paragraphs according to the dividing processing result to form a processing result with a plurality of segments of power grid operation risk scene sequences. Each scene sequence may represent a different risk level such that the corresponding countermeasure may be adjusted for the different risk levels.
In the dividing process, a proper evaluation value range and a proper threshold value and a corresponding punishment item can be set according to specific requirements and related constraint conditions. Therefore, the power grid operation risk scene can be divided into a plurality of paragraphs according to different evaluation values and constraint conditions, so that risk management and corresponding control strategies planning can be better carried out.
In an exemplary embodiment, the obtaining manner of the loss function of the Monte Carlo decision tree includes:
Acquiring the scene number in a power grid operation risk scene set;
Sampling from a power grid operation risk scene set to obtain a current scene;
Analyzing whether the current scene is positioned in a preset target scene set, analyzing whether the current scene is positioned in a corresponding probability value in the preset target scene set, and analyzing whether the number of the scenes positioned in the target scene set accounts for the duty ratio of the number of all scenes in the power grid operation risk scene set to obtain an analysis result;
and obtaining a loss function of the Monte Carlo decision tree according to the analysis result.
The loss function of the Monte Carlo decision tree is:
Wherein m is the number of scenes; when the current scene is located in the target scene set, li is 1, and when the current scene is not located in the target scene set, li is 0; p1i、p2i is the probability value of whether the current scene is in the target scene set or not; rkey is the duty weight of the number of scenes located in the target scene set; the duty weight is the number of scenes that are not in the target scene set.
In this embodiment, the first part of the loss function is used to measure the ability of the model to correctly classify a scene as being in the target scene set, and the second part is used to measure the ability of the model to correctly classify a scene as not being in the target scene set. By weighing the two parts, the penalty of the correct classification and the incorrect classification is weighted by the loss function. In a Monte Carlo decision tree, the optimization algorithm would attempt to minimize the loss function to find the best decision tree model. According to specific weight and scene probability values, the algorithm can determine optimal decision points and partitioning strategies so as to reduce misclassification and missing the situation of a target scene as much as possible, and accordingly prediction accuracy and decision effect of the model are improved.
In an exemplary embodiment, dividing the evaluation value of at least one power grid operation risk scenario to obtain a processing result includes:
Setting constraint conditions, restraining the Monte Carlo decision tree, and dividing and processing the evaluation value of at least one power grid operation risk scene through the constrained Monte Carlo decision tree to obtain a processing result, wherein the constraint conditions comprise evaluation penalty items, and the expression of the evaluation penalty items is as follows:
∑Vpen,i＝Vpen,bl+Vpen,op+Vpen,pf；
Wherein Vpen,bl、Vpen,op、Vpen,pf is an operation constraint penalty term, a safety stability penalty term and a power balance penalty term respectively.
In this embodiment, the optimization algorithm may use this objective function, in combination with other constraints, to adjust the partitioning strategy and decision points of the Monte Carlo decision tree to minimize the value of the objective function. By the constraint and optimization method, a processing result can be obtained, and the situation of the running risk scene of the target power grid can be better divided and evaluated while the constraint condition is met.
In an exemplary embodiment, obtaining a preset risk value evaluation index includes:
Acquiring a loss-of-load risk value, a voltage out-of-limit risk value and a line out-of-limit risk value of a power grid;
Acquiring a preset risk value evaluation index according to a preset weight ratio according to a load loss risk value, a voltage out-of-limit risk value and a line out-of-limit risk value; wherein, the expression of the load loss risk value is:
Wherein K is the analog state number; probability that P (Ek) is system state Ek; sC(Ek) is the severity of the load shedding consequences of system state Ek; pci is the cut load quantity at the load node i, ND is the number of load nodes;
The expression of the voltage threshold risk value is:
Wherein K is the analog state number; probability that P (Ek) is system state Ek; sv(Ek) is the severity of the voltage out-of-limit consequence in the system state Ek, NI is the number of system nodes; vi is the voltage of the different nodes i;
the expression of the line out-of-limit risk value is:
Wherein K is the analog state number; probability that P (Ek) is system state Ek; ss(Ek) is the severity of the voltage out-of-limit consequence in system state Ek; NL is the number of system nodes, PLi is the active power flowing through leg i, and PLimax is the maximum active power allowed to flow through leg i.
In this embodiment, the technical effects of obtaining the load loss risk value, the voltage out-of-limit risk value and the line out-of-limit risk value of the power grid, and calculating the preset risk value evaluation index according to the risk values are to improve the stability and reliability of the power grid.
The loss of load risk value, the voltage out-of-limit risk value and the line out-of-limit risk value of the power grid are important indicators in the operation of the power grid, and reflect the stability and reliability of the power grid under specific conditions. Through calculation and analysis of the risk values, possible problems in the power grid can be found in time, and corresponding measures are taken to solve the problems, so that the probability of power grid faults is reduced, and the stability and reliability of the power grid are improved.
The preset risk value evaluation index is calculated according to the actual running condition and the historical data of the power grid, and can comprehensively reflect the risk levels of the power grid in different aspects. By comparing the preset risk value evaluation index with the actual risk value, the running state and reliability of the power grid can be evaluated, and a reference basis is provided for running management and maintenance of the power grid, so that safe and stable running of the power grid is better ensured.
In summary, the loss-of-load risk value, the voltage out-of-limit risk value and the line out-of-limit risk value of the power grid are obtained, and the preset risk value evaluation index is calculated according to the preset weight ratio, so that the stability and the reliability of the power grid can be improved, and an important reference basis is provided for operation management and maintenance of the power grid.
In the most specific embodiment, referring to fig. 4, fig. 4 is a schematic flow chart of the most specific embodiment.
Acquiring power grid running state data and meteorological monitoring data, generating a power grid section data set, constructing a generated countermeasure network model, generating a power grid running risk scene set, wherein the current iteration number is 1, clustering and discriminant analysis of the running risk scenes, sampling by a Monte Carlo method, and screening by using a decision tree with a weighted cross entropy function through an evaluation value. And setting constraint conditions of the Monte Carlo method, performing constraint when performing deduction calculation, then verifying the accuracy of the Monte Carlo method, performing parallel deduction calculation when the accuracy meets the conditions, obtaining a power grid operation risk assessment result and a power grid prospective operation scene sequence, and re-adding the constraint conditions or reducing the constraint conditions when the accuracy does not meet the conditions. After the parallel deduction calculation is completed, judging whether all scenes are traversed, if not, iterating the steps again, if so, generating a power grid comprehensive risk assessment model according to the result of the parallel deduction calculation, and then carrying out offline processing on the model. And then in the online operation stage of the power grid, acquiring preset risk assessment indexes and current power grid operation scene information, carrying out real-time risk assessment through the model to obtain a more accurate risk assessment result, and then rolling back the risk assessment result to the model for deep learning.
According to the embodiment, the power grid operation risk assessment test is performed based on IEEE39 nodes, the number of nodes is 39, the number of lines is 38, the number of power grid operation simulation scenes is 10000, the time resolution of time sequence deduction calculation is 15 minutes, and the deduction time window is 24 hours.
As shown in FIG. 5, the evaluation results are shown in a grading manner based on the power system risk evaluation test of the generated countermeasure network and time sequence deduction method, the evaluation results are judged to include 1-4, 11-15 and 25-36 risk alarm lines, and the risk alarm nodes are 15, 17 and 23. The risk comprehensive evaluation risk value is 0.731, belongs to the whole-network risk alarm level, and the single-section risk evaluation calculation time is 1.39 seconds.
The embodiment test result shows that the risk assessment method provided by the invention can accurately and rapidly assess the risk lines and nodes of the power grid, solve the comprehensive operation risk level of the power grid, and meet the accuracy and timeliness requirements of scheduling business risk identification.
It should be understood that, although the steps in the flowcharts related to the above embodiments are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.
Based on the same inventive concept, the embodiment of the application also provides a power grid operation risk assessment device for realizing the power grid operation risk assessment method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation in the embodiments of the power grid operation risk assessment device or devices provided below may be referred to the limitation of the power grid operation risk assessment method hereinabove, and will not be described herein.
In an exemplary embodiment, referring to fig. 7, there is provided a power grid operation risk assessment apparatus, including: the data acquisition module 702 is configured to acquire power grid operation state data and meteorological monitoring data;
the data integration module 704 is configured to generate a grid section dataset according to the grid operation state data and the meteorological monitoring data;
The modeling module 706 is configured to construct and generate an countermeasure network model according to the grid section dataset in an offline stage of the grid; acquiring at least one type of power grid operation risk data by generating an countermeasure network model;
a processing module 708, configured to generate a grid operation risk scenario set according to at least one type of grid operation risk data, where the grid operation risk scenario set includes at least one grid operation risk scenario;
The deduction evaluation module 710 is configured to perform deduction calculation on at least one power grid operation risk scene by using a time sequence operation simulation method, so as to generate at least one power grid look-ahead scene sequence;
The data acquisition module 702 is further configured to acquire a preset risk assessment indicator and current power grid operation scene information in an online operation stage of the power grid;
the online application module 712 is configured to obtain a power grid operation risk assessment result according to a preset risk assessment index and current power grid operation scene information, and obtain a corresponding power grid look-ahead operation scene sequence in a matching manner.
In one exemplary embodiment, a timing operation simulation method includes a Monte Carlo decision tree; deduction evaluation module 710 is further configured to:
taking the cross entropy function with the weight as a loss function of a Monte Carlo decision tree, and acquiring an evaluation value of at least one power grid operation risk scene through the Monte Carlo decision tree;
Dividing the evaluation value of at least one power grid operation risk scene to obtain a processing result;
And (3) adopting a parallel deduction architecture, and performing multithread parallel deduction calculation on the processing result through a Monte Carlo decision tree to generate at least one power grid look-ahead scene sequence.
In one exemplary embodiment, deduction evaluation module 710 is further configured to:
Obtaining the maximum deduction expected evaluation value of each power grid operation risk scene through a preset evaluation value calculation formula;
Dividing at least one power grid operation risk scene according to the maximum deduction expected evaluation value to obtain a processing result with a multi-section power grid operation risk scene sequence;
Wherein, the evaluation value calculation formula is:
Wherein, E is a desired operator; t is the access step length; q (vt) is the prize value accessed to scene vt; n (vp) is the number of times the parent scene is accessed; n (vc) is the number of accesses to the sub-scene; vpen,i is an evaluation penalty term; gamma is a constant coefficient and gamma >0.
In one exemplary embodiment, the loss function of the Monte Carlo decision tree is:
Wherein m is the number of scenes; when the current scene is located in the target scene set, li is 1, and when the current scene is not located in the target scene set, li is 0; p1i、p2i is the probability value of whether the current scene is in the target scene set or not; rkey is the duty weight of the number of scenes located in the target scene set; the duty weight is the number of scenes that are not in the target scene set.
In one exemplary embodiment, deduction evaluation module 710 is further configured to: setting constraint conditions, restraining the Monte Carlo decision tree, and dividing and processing the evaluation value of at least one power grid operation risk scene through the constrained Monte Carlo decision tree to obtain a processing result, wherein the constraint conditions comprise evaluation penalty items, and the expression of the evaluation penalty items is as follows:
∑Vpen,i＝Vpen,bl+Vpen,op+Vpen,pf；
Wherein Vpen,bl、Vpen,op、Vpen,pf is an operation constraint penalty term, a safety stability penalty term and a power balance penalty term respectively.
In an exemplary embodiment, the data acquisition module 702 is further configured to:
Acquiring a loss-of-load risk value, a voltage out-of-limit risk value and a line out-of-limit risk value of a power grid;
Acquiring a preset risk value evaluation index according to a preset weight ratio according to a load loss risk value, a voltage out-of-limit risk value and a line out-of-limit risk value; wherein, the expression of the load loss risk value is:
Wherein K is the analog state number; probability that P (Ek) is system state Ek; sC(Ek) is the severity of the load shedding consequences of system state Ek; pci is the cut load quantity at the load node i, ND is the number of load nodes;
The expression of the voltage threshold risk value is:
Wherein K is the analog state number; probability that P (Ek) is system state Ek; sv(Ek) is the severity of the voltage out-of-limit consequence in the system state Ek, NI is the number of system nodes; vi is the voltage of the different nodes i;
the expression of the line out-of-limit risk value is:
Wherein K is the analog state number; probability that P (Ek) is system state Ek; ss(Ek) is the severity of the voltage out-of-limit consequence in system state Ek; NL is the number of system nodes, PLi is the active power flowing through leg i, and PLimax is the maximum active power allowed to flow through leg i.
The modules in the power grid operation risk assessment device can be all or partially implemented by software, hardware and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one exemplary embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, an Input/Output interface (I/O) and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device 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 for storing the power grid operation state data and the meteorological monitoring data. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a grid operation risk assessment method.
It will be appreciated by those skilled in the art that the structure shown in FIG. 7 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In one exemplary embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:
acquiring power grid running state data and meteorological monitoring data;
Generating a power grid section data set according to the power grid running state data and the meteorological monitoring data;
In an off-line stage of a power grid, constructing and generating an countermeasure network model according to the power grid section data set;
Acquiring at least one type of power grid operation risk data through the generation of the countermeasure network model;
generating a power grid operation risk scene set according to the at least one type of power grid operation risk data, wherein the power grid operation risk scene set comprises at least one power grid operation risk scene;
Performing deduction calculation on at least one power grid operation risk scene by adopting a time sequence operation simulation method to generate at least one power grid look-ahead scene sequence;
acquiring a preset risk assessment index and current power grid operation scene information in an online operation stage of the power grid;
and acquiring a power grid operation risk assessment result according to the preset risk assessment index and the current power grid operation scene information, and matching to acquire a corresponding power grid prospective operation scene sequence.
In one embodiment, the processor when executing the computer program further performs the steps of:
The time sequence operation simulation method comprises a Monte Carlo decision tree; the method for simulating the time sequence operation is adopted to carry out deduction calculation on at least one power grid operation risk scene to generate at least one power grid look-ahead scene sequence, and comprises the following steps:
taking the cross entropy function with the weight as a loss function of a Monte Carlo decision tree, and acquiring an evaluation value of at least one power grid operation risk scene through the Monte Carlo decision tree;
Dividing the evaluation value of at least one power grid operation risk scene to obtain a processing result;
And (3) adopting a parallel deduction architecture, and performing multithread parallel deduction calculation on the processing result through a Monte Carlo decision tree to generate at least one power grid look-ahead scene sequence.
In one embodiment, the processor when executing the computer program further performs the steps of:
Dividing the evaluation value of at least one power grid operation risk scene to obtain a processing result, wherein the processing result comprises the following steps:
Obtaining the maximum deduction expected evaluation value of each power grid operation risk scene through a preset evaluation value calculation formula;
Dividing at least one power grid operation risk scene according to the maximum deduction expected evaluation value to obtain a processing result with a multi-section power grid operation risk scene sequence;
Wherein, the evaluation value calculation formula is:
Wherein, E is a desired operator; t is the access step length; q (vt) is the prize value accessed to scene vt; n (vp) is the number of times the parent scene is accessed; n (vc) is the number of accesses to the sub-scene; vpen,i is an evaluation penalty term; gamma is a constant coefficient and gamma >0.
In one embodiment, the processor when executing the computer program further performs the steps of:
The obtaining mode of the loss function of the Monte Carlo decision tree comprises the following steps:
Acquiring the scene number in a power grid operation risk scene set;
Sampling from a power grid operation risk scene set to obtain a current scene;
Analyzing whether the current scene is positioned in a preset target scene set, analyzing whether the current scene is positioned in a corresponding probability value in the preset target scene set, and analyzing whether the number of the scenes positioned in the target scene set accounts for the duty ratio of the number of all scenes in the power grid operation risk scene set to obtain an analysis result;
and obtaining a loss function of the Monte Carlo decision tree according to the analysis result.
The loss function of the Monte Carlo decision tree is:
Wherein m is the number of scenes; when the current scene is located in the target scene set, li is 1, and when the current scene is not located in the target scene set, li is 0; p1i、p2i is the probability value of whether the current scene is in the target scene set or not; rkey is the duty weight of the number of scenes located in the target scene set; the duty weight is the number of scenes that are not in the target scene set.
In one embodiment, the processor when executing the computer program further performs the steps of:
Dividing the evaluation value of at least one power grid operation risk scene to obtain a processing result, wherein the processing result comprises the following steps:
Setting constraint conditions, restraining the Monte Carlo decision tree, and dividing and processing the evaluation value of at least one power grid operation risk scene through the constrained Monte Carlo decision tree to obtain a processing result, wherein the constraint conditions comprise evaluation penalty items, and the expression of the evaluation penalty items is as follows:
∑Vpen,i＝Vpen,bl+Vpen,op+Vpen,pf；
Wherein Vpen,bl、Vpen,op、Vpen,pf is an operation constraint penalty term, a safety stability penalty term and a power balance penalty term respectively.
In one embodiment, the processor when executing the computer program further performs the steps of:
acquiring a preset risk value evaluation index, including:
Acquiring a loss-of-load risk value, a voltage out-of-limit risk value and a line out-of-limit risk value of a power grid;
Acquiring a preset risk value evaluation index according to a preset weight ratio according to a load loss risk value, a voltage out-of-limit risk value and a line out-of-limit risk value; wherein, the expression of the load loss risk value is:
Wherein K is the analog state number; probability that P (Ek) is system state Ek; sC(Ek) is the severity of the load shedding consequences of system state Ek; pci is the cut load quantity at the load node i, ND is the number of load nodes;
The expression of the voltage threshold risk value is:
Wherein K is the analog state number; probability that P (Ek) is system state Ek; sv(Ek) is the severity of the voltage out-of-limit consequence in the system state Ek, NI is the number of system nodes; vi is the voltage of the different nodes i;
the expression of the line out-of-limit risk value is:
Wherein K is the analog state number; probability that P (Ek) is system state Ek; ss(Ek) is the severity of the voltage out-of-limit consequence in system state Ek; NL is the number of system nodes, PLi is the active power flowing through leg i, and PLimax is the maximum active power allowed to flow through leg i.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring power grid running state data and meteorological monitoring data;
Generating a power grid section data set according to the power grid running state data and the meteorological monitoring data;
In an off-line stage of a power grid, constructing and generating an countermeasure network model according to the power grid section data set;
Acquiring at least one type of power grid operation risk data through the generation of the countermeasure network model;
generating a power grid operation risk scene set according to the at least one type of power grid operation risk data, wherein the power grid operation risk scene set comprises at least one power grid operation risk scene;
Performing deduction calculation on at least one power grid operation risk scene by adopting a time sequence operation simulation method to generate at least one power grid look-ahead scene sequence;
acquiring a preset risk assessment index and current power grid operation scene information in an online operation stage of the power grid;
and acquiring a power grid operation risk assessment result according to the preset risk assessment index and the current power grid operation scene information, and matching to acquire a corresponding power grid prospective operation scene sequence.
In one embodiment, the computer program when executed by the processor further performs the steps of:
The time sequence operation simulation method comprises a Monte Carlo decision tree; the method for simulating the time sequence operation is adopted to carry out deduction calculation on at least one power grid operation risk scene to generate at least one power grid look-ahead scene sequence, and comprises the following steps:
taking the cross entropy function with the weight as a loss function of a Monte Carlo decision tree, and acquiring an evaluation value of at least one power grid operation risk scene through the Monte Carlo decision tree;
Dividing the evaluation value of at least one power grid operation risk scene to obtain a processing result;
And (3) adopting a parallel deduction architecture, and performing multithread parallel deduction calculation on the processing result through a Monte Carlo decision tree to generate at least one power grid look-ahead scene sequence.
In one embodiment, the computer program when executed by the processor further performs the steps of:
Dividing the evaluation value of at least one power grid operation risk scene to obtain a processing result, wherein the processing result comprises the following steps:
Obtaining the maximum deduction expected evaluation value of each power grid operation risk scene through a preset evaluation value calculation formula;
Dividing at least one power grid operation risk scene according to the maximum deduction expected evaluation value to obtain a processing result with a multi-section power grid operation risk scene sequence;
Wherein, the evaluation value calculation formula is:
Wherein, E is a desired operator; t is the access step length; q (vt) is the prize value accessed to scene vt; n (vp) is the number of times the parent scene is accessed; n (vc) is the number of accesses to the sub-scene; vpen,i is an evaluation penalty term; gamma is a constant coefficient and gamma >0.
In one embodiment, the computer program when executed by the processor further performs the steps of:
The obtaining mode of the loss function of the Monte Carlo decision tree comprises the following steps:
Acquiring the scene number in a power grid operation risk scene set;
Sampling from a power grid operation risk scene set to obtain a current scene;
Analyzing whether the current scene is positioned in a preset target scene set, analyzing whether the current scene is positioned in a corresponding probability value in the preset target scene set, and analyzing whether the number of the scenes positioned in the target scene set accounts for the duty ratio of the number of all scenes in the power grid operation risk scene set to obtain an analysis result;
and obtaining a loss function of the Monte Carlo decision tree according to the analysis result.
The loss function of the Monte Carlo decision tree is:
Wherein m is the number of scenes; when the current scene is located in the target scene set, li is 1, and when the current scene is not located in the target scene set, li is 0; p1i、p2i is the probability value of whether the current scene is in the target scene set or not; rkey is the duty weight of the number of scenes located in the target scene set; the duty weight is the number of scenes that are not in the target scene set.
In one embodiment, the computer program when executed by the processor further performs the steps of:
Dividing the evaluation value of at least one power grid operation risk scene to obtain a processing result, wherein the processing result comprises the following steps:
Setting constraint conditions, restraining the Monte Carlo decision tree, and dividing and processing the evaluation value of at least one power grid operation risk scene through the constrained Monte Carlo decision tree to obtain a processing result, wherein the constraint conditions comprise evaluation penalty items, and the expression of the evaluation penalty items is as follows:
∑Vpen,i＝Vpen,bl+Vpen,op+Vpen,pf；
Wherein Vpen,bl、Vpen,op、Vpen,pf is an operation constraint penalty term, a safety stability penalty term and a power balance penalty term respectively.
In one embodiment, the computer program when executed by the processor further performs the steps of:
acquiring a preset risk value evaluation index, including:
Acquiring a loss-of-load risk value, a voltage out-of-limit risk value and a line out-of-limit risk value of a power grid;
Acquiring a preset risk value evaluation index according to a preset weight ratio according to a load loss risk value, a voltage out-of-limit risk value and a line out-of-limit risk value; wherein, the expression of the load loss risk value is:
Wherein K is the analog state number; probability that P (Ek) is system state Ek; sC(Ek) is the severity of the load shedding consequences of system state Ek; pci is the cut load quantity at the load node i, ND is the number of load nodes;
The expression of the voltage threshold risk value is:
Wherein K is the analog state number; probability that P (Ek) is system state Ek; sv(Ek) is the severity of the voltage out-of-limit consequence in the system state Ek, NI is the number of system nodes; vi is the voltage of the different nodes i;
the expression of the line out-of-limit risk value is:
Wherein K is the analog state number; probability that P (Ek) is system state Ek; ss(Ek) is the severity of the voltage out-of-limit consequence in system state Ek; NL is the number of system nodes, PLi is the active power flowing through leg i, and PLimax is the maximum active power allowed to flow through leg i.
In one embodiment, a computer program product is provided comprising a computer program which, when executed by a processor, performs the steps of:
acquiring power grid running state data and meteorological monitoring data;
Generating a power grid section data set according to the power grid running state data and the meteorological monitoring data;
In an off-line stage of a power grid, constructing and generating an countermeasure network model according to the power grid section data set;
Acquiring at least one type of power grid operation risk data through the generation of the countermeasure network model;
generating a power grid operation risk scene set according to the at least one type of power grid operation risk data, wherein the power grid operation risk scene set comprises at least one power grid operation risk scene;
Performing deduction calculation on at least one power grid operation risk scene by adopting a time sequence operation simulation method to generate at least one power grid look-ahead scene sequence;
acquiring a preset risk assessment index and current power grid operation scene information in an online operation stage of the power grid;
and acquiring a power grid operation risk assessment result according to the preset risk assessment index and the current power grid operation scene information, and matching to acquire a corresponding power grid prospective operation scene sequence.
In one embodiment, the computer program when executed by the processor further performs the steps of:
The time sequence operation simulation method comprises a Monte Carlo decision tree; the method for simulating the time sequence operation is adopted to carry out deduction calculation on at least one power grid operation risk scene to generate at least one power grid look-ahead scene sequence, and comprises the following steps:
taking the cross entropy function with the weight as a loss function of a Monte Carlo decision tree, and acquiring an evaluation value of at least one power grid operation risk scene through the Monte Carlo decision tree;
Dividing the evaluation value of at least one power grid operation risk scene to obtain a processing result;
And (3) adopting a parallel deduction architecture, and performing multithread parallel deduction calculation on the processing result through a Monte Carlo decision tree to generate at least one power grid look-ahead scene sequence.
In one embodiment, the computer program when executed by the processor further performs the steps of:
Dividing the evaluation value of at least one power grid operation risk scene to obtain a processing result, wherein the processing result comprises the following steps:
Obtaining the maximum deduction expected evaluation value of each power grid operation risk scene through a preset evaluation value calculation formula;
Dividing at least one power grid operation risk scene according to the maximum deduction expected evaluation value to obtain a processing result with a multi-section power grid operation risk scene sequence;
Wherein, the evaluation value calculation formula is:
Wherein, E is a desired operator; t is the access step length; q (vt) is the prize value accessed to scene vt; n (vp) is the number of times the parent scene is accessed; n (vc) is the number of accesses to the sub-scene; vpen,i is an evaluation penalty term; gamma is a constant coefficient and gamma >0.
In one embodiment, the computer program when executed by the processor further performs the steps of:
The obtaining mode of the loss function of the Monte Carlo decision tree comprises the following steps:
Acquiring the scene number in a power grid operation risk scene set;
Sampling from a power grid operation risk scene set to obtain a current scene;
Analyzing whether the current scene is positioned in a preset target scene set, analyzing whether the current scene is positioned in a corresponding probability value in the preset target scene set, and analyzing whether the number of the scenes positioned in the target scene set accounts for the duty ratio of the number of all scenes in the power grid operation risk scene set to obtain an analysis result;
and obtaining a loss function of the Monte Carlo decision tree according to the analysis result.
The loss function of the Monte Carlo decision tree is:
Wherein m is the number of scenes; when the current scene is located in the target scene set, li is 1, and when the current scene is not located in the target scene set, li is 0; p1i、p2i is the probability value of whether the current scene is in the target scene set or not; rkey is the duty weight of the number of scenes located in the target scene set; the duty weight is the number of scenes that are not in the target scene set.
In one embodiment, the computer program when executed by the processor further performs the steps of:
Dividing the evaluation value of at least one power grid operation risk scene to obtain a processing result, wherein the processing result comprises the following steps:
Setting constraint conditions, restraining the Monte Carlo decision tree, and dividing and processing the evaluation value of at least one power grid operation risk scene through the constrained Monte Carlo decision tree to obtain a processing result, wherein the constraint conditions comprise evaluation penalty items, and the expression of the evaluation penalty items is as follows:
∑Vpen,i＝Vpen,bl+Vpen,op+Vpen,pf；
Wherein Vpen,bl、Vpen,op、Vpen,pf is an operation constraint penalty term, a safety stability penalty term and a power balance penalty term respectively.
In one embodiment, the computer program when executed by the processor further performs the steps of:
acquiring a preset risk value evaluation index, including:
Acquiring a loss-of-load risk value, a voltage out-of-limit risk value and a line out-of-limit risk value of a power grid;
Acquiring a preset risk value evaluation index according to a preset weight ratio according to a load loss risk value, a voltage out-of-limit risk value and a line out-of-limit risk value; wherein, the expression of the load loss risk value is:
Wherein K is the analog state number; probability that P (Ek) is system state Ek; sC(Ek) is the severity of the load shedding consequences of system state Ek; pci is the cut load quantity at the load node i, ND is the number of load nodes;
The expression of the voltage threshold risk value is:
Wherein K is the analog state number; probability that P (Ek) is system state Ek; sv(Ek) is the severity of the voltage out-of-limit consequence in the system state Ek, NI is the number of system nodes, Vi is the voltage of different nodes i;
the expression of the line out-of-limit risk value is:
Wherein K is the analog state number; probability that P (Ek) is system state Ek; ss(Ek) is the severity of the voltage out-of-limit consequence in system state Ek; NL is the number of system nodes, PLi is the active power flowing through leg i, and PLimax is the maximum active power allowed to flow through leg i.
It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the present application are both information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data are required to meet the related regulations.
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, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magneto-resistive random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (PHASE CHANGE Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in various forms such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor referred to in the embodiments provided in the present application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, or the like, but is not limited thereto.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of the application should be assessed as that of the appended claims.