Power distribution network supply and insurance demand prediction method combined with dynamic extreme environment simulation

Abstract

The invention discloses a power distribution network supply guarantee demand prediction method combined with dynamic extreme environment simulation, and relates to the technical field of power systems, and the method comprises the steps: building an extreme environment power database, carrying out the classification calibration, obtaining a calibration database, and extracting power consumption behaviors based on the calibration database to construct a power consumption demand response model. Parameters are extracted according to extreme environment simulation information, and a simulation model is generated by combining data modeling such as distribution characteristics of a power distribution network. And constructing a strategy library according to the calibration database, optimizing the simulation parameters to determine optimization parameters, optimizing the simulation model based on the optimization parameters, and establishing a prediction model and performing prediction. The technical problems of inaccurate, incomplete and inflexible power distribution network supply and guarantee demand prediction and lack of a continuous optimization mechanism in an extreme environment in the prior art are solved, and the technical effect of guaranteeing stable operation and continuous power supply of the power distribution network under the extreme condition is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and in particular to a method for predicting power supply demand of distribution networks combined with dynamic extreme environment simulation. Background Art

[0002] In the distribution network operation guarantee scenario, the supply demand problem in extreme environments is particularly prominent, and the contradiction between supply resource demand is relatively more prominent. Accurately predicting supply demand and rationally allocating resources have become crucial links in ensuring the stable operation of distribution networks. Traditional distribution network supply forecasting methods are often extensive and one-sided, relying only on limited environmental data or simple models, and insufficient consideration of the dynamic changes in extreme environments. There is a lack of comprehensive analysis and utilization of user electricity consumption behavior and distribution network distribution characteristics, making it difficult to accurately and effectively predict the distribution network supply demand in extreme environments. There are unreasonable situations in resource allocation, resulting in insufficient supply capacity in some key areas and waste of resources in some areas. The formulation of supply strategies is relatively simple and fixed, and cannot cope well with the actual operation of distribution networks that are constantly changing in extreme environments.

[0003] At the current stage, relevant technologies have the technical problems of inaccurate, incomplete, and inflexible forecasting of distribution network supply demand in extreme environments and lack of continuous optimization mechanism. Summary of the invention

[0004] This application provides a distribution network supply demand prediction method combined with dynamic extreme environment simulation, establishes an extreme environment power database and classifies and calibrates it, obtains a calibration database, and extracts power consumption behavior based on it to build a power demand response model. Parameters are extracted according to extreme environment simulation information, and a simulation model is generated by combining data modeling such as distribution characteristics of the distribution network. A strategy library is constructed based on the calibration database, and the optimization parameters are determined by optimizing the simulation parameters. The simulation model is optimized based on the optimization parameters, and a prediction model is established and predicted, thereby achieving the technical effect of ensuring the stable operation and continuous power supply of the distribution network under extreme conditions.

[0005] This application provides a distribution network supply demand prediction method combined with dynamic extreme environment simulation, including: Establish an extreme environment power database, classify and calibrate the extreme environment power database, and obtain an extreme environment calibrated power database, wherein the extreme environment calibrated power database includes extreme environment power data of each calibrated category parameter; extract power consumption behavior based on the extreme environment calibrated power database, obtain a power demand response behavior data set, perform identification training on the extreme environment calibrated power database and the power demand response behavior data set, and construct an extreme environment power demand response model; extract parameters from the extreme environment calibrated power database according to extreme environment simulation information, determine extreme environment dynamic simulation parameters, perform simulation modeling based on the distribution characteristic data of the target distribution network, the power demand response model and the extreme environment dynamic simulation parameters, and generate a distribution network supply demand simulation model; construct a distribution network power supply reconstruction strategy library based on the extreme environment calibrated power database, perform strategy matching optimization on the extreme environment dynamic simulation parameters based on the distribution network power supply reconstruction strategy library, and determine the power supply reconstruction strategy optimization parameters; optimize the distribution network supply demand simulation model based on the power supply reconstruction strategy optimization parameters, establish a distribution network supply demand prediction model, and predict the supply demand through the distribution network supply demand prediction model.

[0006] The distribution network supply demand prediction method combined with dynamic extreme environment simulation proposed in this application is to first establish an extreme environment power database and classify and calibrate it, obtain a calibration database, and extract power consumption behavior based on it to build a power demand response model. Extract parameters according to extreme environment simulation information, and generate a simulation model by combining data modeling such as distribution characteristics of the distribution network. Build a strategy library based on the calibration database, optimize the simulation parameters to determine the optimization parameters, optimize the simulation model based on the optimization parameters, establish a prediction model and make predictions, so as to achieve the technical effect of ensuring the stable operation and continuous power supply of the distribution network under extreme conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0008] Figure 1 A flow chart of a method for predicting distribution network supply demand combined with dynamic extreme environment simulation provided in an embodiment of the present application; Figure 2 A schematic diagram of a flow chart for obtaining a fault discriminator in a distribution network supply demand prediction method combined with dynamic extreme environment simulation provided in an embodiment of the present application. DETAILED DESCRIPTION

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0010] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0011] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0012] The present application embodiment provides a distribution network supply demand prediction method combined with dynamic extreme environment simulation, such as Figure 1 As shown, the method includes: Step S100, establish an extreme environment power database, classify and calibrate the extreme environment power database, and obtain an extreme environment calibration power database, which includes extreme environment power data of each calibration category parameter. Specifically, the extreme environment power database refers to a collection of power-related data collected under extreme environmental conditions. Determine the data source channel, such as power company operation records, meteorological department monitoring data, field monitoring site collection, etc., prepare data collection equipment, clarify data collection indicators, such as power consumption fluctuations, voltage change amplitude, equipment failure frequency, etc., carry out extensive data collection work, and obtain initial extreme environment power data. The initial extreme environment power data reflects the initial state of the power system when the extreme environment just appears, and its characteristic is the initial nature of the data. Continue to expand the scope and depth of data collection, obtain comprehensive power data during the complete extreme environment, and the power data during the complete extreme environment covers the overall status of the power system during the entire extreme environment continuous stage, which is characterized by the integrity of the data. Sort and summarize the collected data to establish an extreme environment power database.

[0013] In a possible implementation, an extreme environment power database is established, the extreme environment power database is classified and calibrated, and an extreme environment calibrated power database is obtained, the extreme environment calibrated power database includes extreme environment power data of each calibration category parameter, and step S100 further includes step S110, obtaining extreme environment calibration factor information, the extreme environment calibration factor information includes extreme temperature and humidity, climate conditions, geographical terrain, and altitude weather. Specifically, the extreme environment calibration factor information obtained is clearly obtained, and the information covers aspects such as extreme temperature and humidity, climate conditions, geographical terrain, and altitude weather.

[0014] Step S120, fill the extreme environment calibration factor information with content through the extreme environment event library to obtain an extreme environment factor entity content set. Specifically, the calibration factor information is filled with content with the help of the existing extreme environment event library, which contains a large number of detailed records and related data of various extreme environment events that have occurred in history. By matching and integrating with these data, the originally more abstract calibration factor information is transformed into a specific and rich extreme environment factor entity content set, so that it contains more detailed and practical information.

[0015] Step S130, designing a physical factor content coding system based on the extreme environment calibration factor information. Specifically, after obtaining the extreme environment calibration factor information (such as extreme temperature and humidity, climate conditions, geographical terrain, altitude weather, etc.), designing a physical factor content coding system, conducting an in-depth analysis of each extreme environment calibration factor, understanding its possible value range and variation characteristics, for example, extreme temperatures may range from dozens of degrees below zero to dozens of degrees above zero, and climate conditions may include a variety of different types, determining the basic rules and structure of the coding, choosing to use digital coding, letter coding, or a combination of digital and letter coding, considering the length and readability of the coding, both to cover all possible situations and to be easy to understand and use, for extreme temperature and humidity, according to The specific value ranges of temperature and humidity are divided into different intervals, and a unique code is assigned to each interval. For climate conditions, they can be classified and coded according to different climate types. For geographical terrain, they can be coded according to the main categories of terrain. For altitude weather, they can be coded according to different stages of altitude and corresponding weather characteristics. When designing the coding system, a certain amount of expansion space should be reserved to cope with new extreme environmental calibration factors or changes in value ranges that may appear in the future. Ultimately, a complete, reasonable and applicable entity factor content coding system will be formed, which will provide clear and definite standards and basis for the subsequent coding work of extreme environmental factor entity content sets.

[0016] Step S140, encode the extreme environment factor entity content set through the entity factor content encoding system to obtain an extreme environment entity content encoding set. Specifically, the designed encoding system encodes the obtained extreme environment factor entity content set. During the encoding process, the rules and requirements of the encoding system are strictly followed to ensure the accuracy and consistency of the encoding, thereby obtaining an extreme environment entity content encoding set.

[0017] Step S150, using the extreme environment entity content code set as a calibration category parameter, dividing and calibrating the extreme environment power database, and obtaining the extreme environment calibrated power database. Specifically, using the obtained extreme environment entity content code set as a calibration category parameter, the previously established extreme environment power database is carefully divided and calibrated, and the data in the database is classified and marked according to different extreme environment characteristics, so as to obtain a complete and accurate extreme environment calibrated power database, providing strong support for subsequent analysis and application.

[0018] Step S200, extracting electricity consumption behavior based on the extreme environment calibrated power database, obtaining an electricity demand response behavior data set, performing identification training on the extreme environment calibrated power database and the electricity demand response behavior data set, and constructing an extreme environment electricity demand response model. Specifically, the extreme environment electricity demand response model refers to a model constructed based on the extreme environment calibrated power database to reflect the law of electricity demand response. Obtain an extreme environment calibrated power database, prepare data processing and analysis tools, set relevant parameters for extraction and training, such as data screening conditions, number of model training rounds, etc., perform electricity consumption behavior extraction operations, and obtain an electricity demand response behavior data set. The electricity demand response behavior data set reflects the patterns and characteristics of user electricity consumption behaviors in extreme environments, and its characteristic is focusing on behavioral responses. Continue to perform identification training on the data to construct an extreme environment electricity demand response model, which can predict the user's electricity demand under different extreme environmental conditions, and its characteristic is reflecting the law of demand.

[0019] In one possible implementation, electricity consumption behavior is extracted based on the extreme environment calibrated power database to obtain a power demand response behavior data set, identification training is performed on the extreme environment calibrated power database and the power demand response behavior data set, and an extreme environment power demand response model is constructed. Step S200 further includes step S210, performing necessary demand identification and necessary data extraction on the power demand response behavior data set to obtain a necessary power demand behavior data set. Specifically, work is carried out on the electricity demand response behavior dataset, and each data in the dataset is carefully analyzed to identify the necessary needs. For example, the dataset includes various electricity consumption behaviors such as lighting, cooling / heating, and communications. Through analysis, it is assumed that lighting electricity consumption behavior is more important in a specific scenario, while cooling / heating electricity consumption behavior is relatively less important in the scenario. In this way, lighting-related data is taken as the necessary demand that is focused on, and demand-related data is extracted from the dataset. Lighting electricity consumption data is recorded according to time and electricity consumption, such as lighting electricity consumption in different time periods of a day. Through screening, only lighting electricity consumption data with large electricity consumption or in a specific time period (such as night) are retained, thereby obtaining the necessary electricity demand behavior dataset.

[0020] Step S220, mapping and identifying the extreme environment calibrated power database with the necessary power demand behavior data set according to the calibration category parameters, and obtaining the extreme environment power demand sample data set. Specifically, the extreme environment calibrated power database is classified and sorted according to its calibration category parameters, for example, it is divided according to different extreme temperature ranges (assuming that it is divided into low temperature, medium temperature and high temperature), climate types (assuming that it is divided into heavy rain, strong wind and drought) and other calibration categories, and each calibration category is mapped and identified with the necessary power demand behavior data set, and the power data under the same calibration category is associated with the corresponding necessary power demand data. For example, for the low temperature calibration category, the degree of matching between it and the lighting power demand data in terms of time and power consumption is calculated. Assuming that the lighting power consumption generally increases by a certain proportion under low temperature conditions, these changes are compared and counted to obtain the extreme environment power demand sample data set.

[0021] Step S230, performing neural network learning and training on the extreme environment electricity demand sample data set to generate an initial electricity demand analysis model. Specifically, the acquired extreme environment electricity demand sample data set is used for neural network learning and training, and the data set is input into the neural network model. By adjusting the parameters of the model, the model is allowed to learn the patterns and rules in the data. During the training process, the model continuously compares the difference between the predicted electricity demand and the actual electricity demand, gradually reduces the difference, so that the model's prediction becomes more and more accurate, and generates an initial electricity demand analysis model.

[0022] Step S240, the performance of the initial power demand analysis model is verified and iteratively optimized to construct the extreme environment power demand response model. Specifically, the performance of the initial power demand analysis model is verified, and a series of evaluation indicators and test data are used to test the accuracy, reliability and generalization ability of the model. By comparing with the actual power consumption data, the average deviation between the power consumption predicted by the model and the actual power consumption is calculated. If the average deviation is large, it means that the model is not accurate enough. It is necessary to find the problems and deficiencies of the model based on the verification results, and then iteratively optimize by adjusting the structure of the model (such as increasing or decreasing the connection of certain neurons), parameters (such as changing the learning rate) or increasing training data, and finally construct an extreme environment power demand response model that can accurately reflect the power demand response in extreme environments.

[0023] Step S300, extract parameters of the extreme environment calibrated power database according to the extreme environment simulation information, determine the extreme environment dynamic simulation parameters, perform simulation modeling based on the distribution characteristic data of the target distribution network, the power demand response model and the extreme environment dynamic simulation parameters, and generate a distribution network supply demand simulation model. Specifically, the distribution network supply demand simulation model refers to a model generated under a specific process for predicting the supply status of the distribution network under extreme environments. Obtain the preset extreme environment information to be simulated, perform parameter extraction operations on the extreme environment calibrated power database, and determine the extreme environment dynamic simulation parameters. Obtain the distribution characteristic data of the target distribution network, integrate the power demand response model and the extreme environment dynamic simulation parameters, carry out simulation modeling, and generate a distribution network supply demand simulation model. This model reflects the supply and demand situation of the distribution network under extreme environments. It is characterized by the ability to accurately predict supply and demand, obtain preset specific information about the extreme environment to be simulated, such as a detailed description of the extreme environment that may occur in a specific area during a specific period of time, covering specific extremely high or extremely low temperature values, the intensity and duration of severe weather types, etc. According to the extreme environment simulation information, an in-depth analysis of the extreme environment calibration power database is carried out. Parameters associated with the current simulation information are selected from the database. For example, if the simulation is a high temperature extreme environment, the performance change parameters of the power equipment under past high temperature conditions are extracted, such as the overload limit value of the transformer, the heat dissipation capacity of the line, etc.; at the same time, the user's power demand change parameters at high temperatures are also extracted, such as the growth rate of air conditioning use. Through comprehensive measurement and calculation of parameters, the dynamic simulation parameters that can accurately reflect the current extreme environmental characteristics are identified, and the distribution characteristic data of the target distribution network is obtained, including the specific location, model, capacity and connection method of various types of power equipment (such as transformers, poles, wires, etc.) in the distribution network, as well as detailed information such as the distribution location of users in the distribution network and the distribution status of power loads. The power demand response model, the already determined extreme environment dynamic simulation parameters and the distribution characteristic data of the target distribution network are integrated for simulation modeling. In the modeling process, the basic physical model is first constructed according to the equipment distribution and user distribution of the distribution network, and the power demand response model is integrated into it, so that it can predict the changes in users' power consumption behavior in extreme environments based on the extreme environment dynamic simulation parameters. Then, the extreme environment dynamic simulation parameters are applied to the model to simulate the performance changes of power equipment in such extreme environments, such as increased line resistance and increased transformer losses.By continuously adjusting and optimizing the parameters and relationships of each part of the model, it can accurately reflect the actual operating conditions of the distribution network under extreme environments, and ultimately generate a simulation model that can effectively predict the power supply needs of the distribution network under extreme environments. The model can output key information such as voltage changes at each node in the distribution network under specific extreme environments, line load conditions, fault locations, and required power supply capacity, providing a solid decision-making basis for the planning, operation, and maintenance of the distribution network.

[0024] In a possible implementation, according to the extreme environment simulation information, the extreme environment calibration power database is parameter extracted, the extreme environment dynamic simulation parameters are determined, and simulation modeling is performed based on the distribution characteristic data of the target distribution network, the power demand response model and the extreme environment dynamic simulation parameters to generate a distribution network supply demand simulation model. Step S300 further includes step S310, using the three-dimensional modeling technology to perform spatial modeling on the distribution characteristic data of the target distribution network to generate a distribution network distribution architecture spatial model. Specifically, the three-dimensional modeling technology is used to process the distribution characteristic data of the target distribution network. The distribution characteristic data contains detailed information such as the precise position, height, orientation and mutual connection relationship of each power equipment (such as transformers, poles, wires, etc.) in the distribution network. Professional three-dimensional modeling software and tools are used to convert the data into an intuitive three-dimensional model. In the modeling process, the three-dimensional shape and position of each power equipment are accurately constructed according to the actual physical size and spatial layout, thereby generating a distribution network distribution architecture spatial model. The model can clearly display the architecture and layout of the distribution network in space.

[0025] Step S320, the power supply simulation modeling is performed on the distribution network distribution architecture space model through the power demand response model and the extreme environment dynamic simulation parameters to obtain a basic supply guarantee demand simulation model. Specifically, the power demand response model and the extreme environment dynamic simulation parameters are introduced into the distribution network distribution architecture space model. The power demand response model can predict the changes in user demand under different power consumption conditions, and the extreme environment dynamic simulation parameters reflect the performance changes of power equipment under extreme environments. The power supply simulation modeling is performed on the distribution network distribution architecture space model. Through complex calculations and simulations, the power flow, voltage changes, power distribution and other factors are considered to simulate the power supply of the distribution network under different power demands and extreme environments, thereby obtaining a basic supply guarantee demand simulation model.

[0026] Step S330, compare and verify the basic supply and demand simulation model to obtain the model reliability coefficient, and optimize the parameters of the basic supply and demand simulation model based on the model reliability coefficient to obtain the distribution network supply and demand simulation model. Specifically, compare and verify the obtained basic supply and demand simulation model, compare the output results of the model with the actual monitoring data, historical experience data or other reliable reference data, analyze the degree of difference between the model prediction results and the actual situation through a series of evaluation indicators and calculation methods, and calculate the model reliability coefficient based on the comparison results. The coefficient can be a value between 0 and 1. The closer the value is to 1, the more reliable the model is. Based on the calculated model reliability coefficient, adjust and optimize the parameters of the basic supply and demand simulation model. It is necessary to adjust the parameters of the power equipment, the coefficient of power demand, the influencing factors of the extreme environment, etc. Through continuous experiments and improvements, the prediction results of the model are closer to the actual situation, and finally obtain an accurate and reliable distribution network supply and demand simulation model.

[0027] Step S400, construct a distribution network power supply reconstruction strategy library according to the extreme environment calibrated power database, perform strategy matching optimization on the extreme environment dynamic simulation parameters based on the distribution network power supply reconstruction strategy library, and determine the power supply reconstruction strategy optimization parameters. Specifically, the distribution network power supply reconstruction strategy optimization parameters refer to key indicators for optimizing the distribution network power supply reconstruction strategy determined according to specific processes and conditions. Construct a distribution network power supply reconstruction strategy library based on the extreme environment calibrated power database. Based on this strategy library, perform strategy matching optimization on the extreme environment dynamic simulation parameters, and finally determine the power supply reconstruction strategy optimization parameters based on the extreme environment calibrated power database. Carry out in-depth analysis and research on the database to extract the characteristics and laws of the impact of different extreme environments on the operation of the distribution network. Construct a distribution network power supply reconstruction strategy library based on this. During the construction, for each possible extreme weather scenario, such as typhoon, rainstorm, snowstorm, high temperature, etc., carefully formulate specific and detailed reconstruction strategies. Taking typhoon weather as an example, the carefully formulated reconstruction strategy may include: identifying key loads, such as hospitals, emergency command centers, etc., and giving priority to ensuring the stable power supply of these loads; optimizing the structure of the distribution network, which may include adjusting the connection method of the line, adding backup power supplies, or enhancing the carrying capacity of certain lines; it may also involve adjusting the configuration of transformers to adapt to changes in electricity demand in different regions. Through specific reconstruction strategies, the wind resistance and power supply capacity of the distribution network in typhoon weather are improved, and the strategy matching optimization of the extreme environment dynamic simulation parameters is carried out based on the constructed distribution network power supply reconstruction strategy library. The extreme environment dynamic simulation parameters are input into the strategy library, and through a series of calculations and analyses, the reconstruction strategy that best matches these parameters is sought. In the matching process, a variety of factors are comprehensively considered, such as current power demand, equipment operating status, possible failure risks, etc., and the preliminary matching strategy is further optimized and adjusted to ensure that it can adapt to specific extreme environment dynamic simulation parameters to the greatest extent. Finally, the power supply reconstruction strategy optimization parameters are finalized. The parameters clarify the specific details and key indicators of the optimal reconstruction strategy that the distribution network should implement under specific extreme environments, providing precise guidance for the actual operation and management of the distribution network.

[0028] In a possible implementation, a distribution network power supply reconstruction strategy library is constructed according to the extreme environment calibration power database, and the extreme environment dynamic simulation parameters are optimized by strategy matching based on the distribution network power supply reconstruction strategy library to determine the power supply reconstruction strategy optimization parameters, such as Figure 2Said step S400 further includes step S410, performing strategy matching on the extreme environment dynamic simulation parameters based on the distribution network power supply reconstruction strategy library, and determining the target power supply reconstruction strategy. Specifically, based on the constructed distribution network power supply reconstruction strategy library, the extreme environment dynamic simulation parameters are input therein, and the strategy library contains a variety of power supply reconstruction strategies formulated for various extreme environmental conditions. For example, the extreme environment dynamic simulation parameters may show that under certain extreme weather conditions, the power load will rise sharply, and some areas may experience large voltage fluctuations. Carefully compare the various strategies in the strategy library, different line switching arrangements, transformer adjustment strategies, etc., to determine which strategy can more effectively cope with the current situation of such a sharp increase in load and voltage fluctuations, thereby determining the target power supply reconstruction strategy.

[0029] Step S420, according to the target power supply reconstruction strategy, the target distribution network is subjected to parameter optimization analysis to obtain the reconstruction strategy parameter optimization threshold. Specifically, according to the determined target power supply reconstruction strategy, the relevant parameters of the target distribution network are deeply optimized and analyzed, and the equipment performance, load distribution, line capacity and other factors of the distribution network are fully considered. For example, assuming that the target power supply reconstruction strategy is to adjust the output voltage of a certain transformer, it is necessary to calculate whether the power supply capacity of the area after adjustment can meet the load demand. Through such analysis and calculation, the parameter range that can achieve the optimal reconstruction effect is determined, and then the reconstruction strategy parameter optimization threshold is obtained.

[0030] Step S430, obtain the reconstruction optimization target, evaluate and fit the reconstruction optimization target, and obtain the fitness function of the distribution network reconstruction effect. Specifically, obtain clear reconstruction optimization targets, such as improving power supply capacity, ensuring voltage stability, improving power quality, etc. For improving power supply capacity, it is necessary to calculate whether the maximum power that can be supplied under different reconstruction schemes can meet the expected load growth; for voltage stability, it is necessary to calculate whether the voltage fluctuation range of each node is within the allowable range; for power quality, it is necessary to evaluate indicators such as harmonic content and voltage deviation, and convert specific optimization targets into quantifiable and calculable forms, so as to obtain a fitness function that can measure the effect of distribution network reconstruction.

[0031] Step S440, based on the fitness function of the distribution network reconstruction effect, the reconstruction strategy parameter optimization threshold is optimized for parameters, and the power supply reconstruction strategy optimization parameters are determined. Specifically, based on the obtained fitness function of the distribution network reconstruction effect, the parameters within the reconstruction strategy parameter optimization threshold are finely optimized, and different parameter combinations are tried, such as changing the resistance of certain lines, adjusting the gear of the transformer, etc., and the corresponding fitness values ​​are calculated, just like selecting the one that best meets the optimization goal from many schemes. After repeated exploration and comparison, the most ideal power supply reconstruction strategy optimization parameters are finally determined, which provides a strong guarantee for the stable operation and efficient power supply of the distribution network in extreme environments.

[0032] In a possible implementation, the reconstruction strategy parameter optimization threshold is optimized based on the distribution network reconstruction effect fitness function to determine the power supply reconstruction strategy optimization parameter, and step S440 further includes step S441, randomly obtaining multiple reconstruction strategy parameters within the reconstruction strategy parameter optimization threshold. Specifically, within the range defined by the reconstruction strategy parameter optimization threshold, multiple reconstruction strategy parameters are obtained by random generation, for example, the reconstruction strategy parameters include the switching mode of the line, the adjustment range of the transformer, etc., and multiple different schemes are randomly combined from various possibilities allowed by the threshold, and each scheme is a set of reconstruction strategy parameters.

[0033] Step S442, based on the distribution network supply demand simulation model, the multiple reconstruction strategy parameters are simulated to obtain multiple distribution network reconstruction simulation parameters. Specifically, the multiple reconstruction strategy parameters obtained randomly are respectively input into the distribution network supply demand simulation model for simulation. During the simulation, the model will calculate various possible extreme environments and power demand conditions according to the parameters, such as calculating the amount of current passing through a certain line under the input parameters; or calculating whether the voltage of a certain node is increased or decreased under this parameter setting, and how much it changes, thereby obtaining multiple distribution network reconstruction simulation parameters.

[0034] Step S443, the multiple distribution network reconstruction simulation parameters are evaluated for their effects through the distribution network reconstruction effect fitness function to obtain multiple reconstruction effect fitness information. Specifically, the distribution network reconstruction effect fitness function is used to evaluate the effects of multiple distribution network reconstruction simulation parameters. The fitness function measures the effect of each simulation parameter according to a pre-set reconstruction optimization goal. For example, the maximum power that can be supplied under the simulation parameter is calculated to be increased by how much compared to before; to ensure voltage stability, check whether the voltage fluctuation of each node has become smaller and how much smaller; to improve power quality, check whether the harmonic content has been reduced and to what extent. Through comparison and measurement, a score or evaluation is given to each simulation parameter, and finally multiple reconstruction effect fitness information is obtained.

[0035] Step S444, based on the multiple reconstruction effect fitness information, the multiple reconstruction strategy parameters are compared and optimized to determine the power supply reconstruction strategy optimization parameters. Specifically, based on the obtained multiple reconstruction effect fitness information, the multiple reconstruction strategy parameters are carefully compared and optimized, each reconstruction strategy parameter and its corresponding fitness information are taken out and compared with each other, and the reconstruction strategy parameter with the highest fitness and the best effect is found, and determined as the final power supply reconstruction strategy optimization parameters. The power supply reconstruction strategy optimization parameters can provide guidance for the reconstruction and optimization of the distribution network in extreme environments, and achieve more stable and efficient power supply.

[0036] In a possible implementation, the multiple reconstruction strategy parameters are compared and optimized based on the multiple reconstruction effect fitness information to determine the power supply reconstruction strategy optimization parameters, and step S444 further includes step S4441, based on the multiple reconstruction effect fitness information, the multiple distribution network reconstruction simulation parameters are proportionally screened to obtain a set of superior reconstruction strategy parameters. Specifically, according to the multiple reconstruction effect fitness information obtained, the multiple distribution network reconstruction simulation parameters are proportionally screened, the reconstruction effect fitness is arranged in order from high to low, the screening ratio is set, and from the sorted parameters, the parameters that are arranged in front and correspond to the set screening ratio are selected, and the selected parameters constitute the set of superior reconstruction strategy parameters.

[0037] Step S4442, set the parameter variation step length, and mutate and expand the optimal reconstruction strategy parameter set based on the parameter variation step length to obtain a reconstruction strategy parameter variation population. Specifically, set the parameter variation step length, for example, for a certain parameter, set an initial value, and the possible value of the parameter after mutation is between the initial value minus the step length and the initial value plus the step length. Based on the set step length, mutate and expand each parameter in the optimal reconstruction strategy parameter set, thereby obtaining a series of new parameter combinations to form a reconstruction strategy parameter variation population.

[0038] Step S4443, iterative optimization is performed within the reconstruction strategy parameter variation population until the preset number of iterations is reached, and the power supply reconstruction strategy optimization parameters are determined. Specifically, the iterative optimization process is started within the reconstruction strategy parameter variation population. In each iteration, the parameters in the population are evaluated and compared. For example, if the fitness of a group of parameters is low, then in the next round, a larger variation is made to it; if the fitness is high, a smaller adjustment is made, and the above process is repeated until the preset number of iterations is reached. When the preset number of iterations is reached, the group with the highest fitness is found from the parameter combination obtained in the last round of iterations, and it is determined as the final power supply reconstruction strategy optimization parameter. The parameters are the optimal solutions obtained after multiple screening, adjustment and optimization, which can provide the best strategic guidance for the reconstruction of the distribution network and ensure the stable and efficient operation of the distribution network in extreme environments.

[0039] Step S500, optimize the supply demand of the distribution network supply demand simulation model based on the power supply reconstruction strategy optimization parameters, establish a distribution network supply demand prediction model, and predict the supply demand through the distribution network supply demand prediction model. Specifically, the distribution network supply demand prediction scheme refers to a process for predicting the distribution network supply demand based on specific parameters and models, obtaining the power supply reconstruction strategy optimization parameters obtained through rigorous calculation and screening, introducing them into the distribution network supply demand simulation model, optimizing the model for supply demand, and then constructing a distribution network supply demand prediction model, and implementing supply demand prediction through the model, obtaining the obtained power supply reconstruction strategy optimization parameters, and integrating the parameters into the distribution network supply demand simulation model, as if injecting new vitality and accuracy into the model. For example, if the optimization parameters involve the re-planning of certain key lines, then the connection of these lines will be adjusted accordingly in the model. If it is related to the optimization of specific equipment, such as the upgrade or replacement of transformers, the performance parameters of the equipment in the model will be updated. By integrating the optimization parameters into the simulation model, a full range of supply demand optimization is implemented to enable the model to more realistically and accurately reflect the operating conditions and supply capacity of the distribution network under various extreme environments. On this basis, a distribution network supply demand prediction model is further built. The model focuses on reducing the adverse effects of extreme environments on the power supply of the distribution network. In the construction process, the advantages and characteristics of the optimized simulation model are fully integrated, and a large amount of historical data and actual operation experience are combined. For example, the past The problems and countermeasures presented by the distribution network under extreme weather conditions, converting experience into part of the model, and performing meticulous preprocessing on the collected data, including removing outliers and standardizing the data, to ensure the quality and consistency of the data. In terms of the model architecture design, carefully select algorithms and structures that can effectively capture time series characteristics and complex relationships in order to better predict future supply and demand. When the distribution network supply and demand forecasting model is successfully constructed, it can be put into actual supply and demand forecasting operations, and the real-time operating data of the current distribution network, such as the current, voltage, power, etc. of each node, as well as the latest extreme weather forecast information and changes in power load, can be used to predict the supply and demand of the power grid. The model will use its complex internal calculation and reasoning logic to comprehensively analyze and process the information, and finally output a detailed forecast of the distribution network's supply demand in the future, covering important contents such as emerging power shortage areas and time periods, the stress conditions of key equipment under extreme weather conditions, and the reliability assessment of the overall power supply. The forecast results can provide the power department with valuable decision-making basis, helping it to formulate reasonable power allocation plans, arrange equipment maintenance plans, prepare emergency power supplies in advance, thereby greatly improving the distribution network's supply capacity and power supply reliability under extreme weather conditions, and ensuring that users' normal electricity needs are met.

[0040] In one possible implementation, the distribution network supply demand simulation model is optimized based on the power supply reconstruction strategy optimization parameters, a distribution network supply demand prediction model is established, and supply demand prediction is performed through the distribution network supply demand prediction model. Step S500 further includes step S510, performing supply demand prediction and actual comparison feedback based on the distribution network supply demand prediction model to obtain prediction accuracy feedback parameters. Specifically, the distribution network supply demand prediction model is used to predict the supply demand. According to the input data such as the current distribution network operation status, weather conditions, power load, etc., the prediction results of the distribution network supply demand in the future are given, and the prediction results are compared with the actual distribution network supply demand. The actual operation data, including the actual current, voltage, power of each node, the actual power load changes, and the actual supply situation, are collected and sorted out, and compared with the prediction results one by one. In the process of comparison, the difference between the prediction result and the actual situation is calculated. For example, the predicted power gap is compared with the actual power gap, the predicted voltage fluctuation range is compared with the actual voltage fluctuation range, etc. Through comparison and calculation, the prediction accuracy feedback parameter is obtained, which can quantitatively reflect the accuracy of the prediction model.

[0041] Step S520, the prediction accuracy feedback parameters are subjected to model optimization analysis to obtain model parameter optimization rules. Specifically, the obtained prediction accuracy feedback parameters are subjected to in-depth model optimization analysis to analyze the deviations and deficiencies of the model reflected by the parameters, such as insufficient estimation of the impact of extreme weather or inaccurate grasp of the changing trend of power load. Through analysis, the direction in which the model needs to be improved and optimized is found, thereby obtaining model parameter optimization rules, which include adjusting the weights of certain parameters, increasing or decreasing the consideration of certain factors, or improving the algorithm and structure of the model.

[0042] Step S530, iteratively update the parameters of the distribution network supply demand prediction model based on the model parameter optimization rules to obtain the distribution network supply demand optimization prediction model. Specifically, based on the obtained model parameter optimization rules, iteratively update the parameters of the distribution network supply demand prediction model, modify and adjust the various parameters in the model according to the rules, perform prediction and comparison, and repeat the above process until the prediction results of the model can achieve the expected accuracy and reliability, and finally obtain the distribution network supply demand optimization prediction model.

[0043] The embodiment of the present application adopts the method of establishing an extreme environment power database and classifying and calibrating it, obtaining a calibration database, and constructing a power demand response model based on the power consumption behavior extracted from it. Parameters are extracted according to the extreme environment simulation information, and a simulation model is generated by combining data modeling such as the distribution characteristics of the distribution network. A strategy library is constructed based on the calibration database, and the optimization parameters are determined by optimizing the simulation parameters. The simulation model is optimized based on the optimization parameters, and a prediction model is established and predicted, thereby achieving the technical effect of ensuring the stable operation and continuous power supply of the distribution network under extreme conditions.

[0044] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A distribution network supply demand prediction method combined with dynamic extreme environment simulation is characterized by: The method comprises: Establishing an extreme environment power database, classifying and calibrating the extreme environment power database, and obtaining an extreme environment calibrated power database, wherein the extreme environment calibrated power database includes extreme environment power data of each calibration category parameter; Extracting electricity consumption behavior based on the extreme environment calibrated power database to obtain an electricity demand response behavior data set, performing identification training on the extreme environment calibrated power database and the electricity demand response behavior data set, and constructing an extreme environment electricity demand response model; Extract parameters of the extreme environment calibration power database according to the extreme environment simulation information, determine the extreme environment dynamic simulation parameters, perform simulation modeling based on the distribution characteristic data of the target distribution network, the power demand response model and the extreme environment dynamic simulation parameters, and generate a distribution network supply demand simulation model; According to the extreme environment calibration power database, a distribution network power supply reconstruction strategy library is constructed, and based on the distribution network power supply reconstruction strategy library, strategy matching optimization is performed on the extreme environment dynamic simulation parameters to determine the power supply reconstruction strategy optimization parameters; The distribution network supply demand simulation model is optimized for supply demand based on the power supply reconstruction strategy optimization parameters, a distribution network supply demand prediction model is established, and supply demand prediction is performed through the distribution network supply demand prediction model.

2. The method for predicting the supply demand of a distribution network combined with dynamic extreme environment simulation according to claim 1, characterized in that: The step of obtaining the extreme environment calibration power database includes: Acquire extreme environment calibration factor information, wherein the extreme environment calibration factor information includes extreme temperature and humidity, climate conditions, geographical terrain, and altitude weather; Filling the extreme environment calibration factor information with content through the extreme environment event library to obtain an extreme environment factor entity content set; Designing a physical factor content coding system based on the extreme environment calibration factor information; Encoding the extreme environment factor entity content set by the entity factor content encoding system to obtain an extreme environment entity content encoding set; The extreme environment entity content code set is used as a calibration category parameter to divide and calibrate the extreme environment power database to obtain the extreme environment calibrated power database.

3. The method for predicting the demand for distribution network supply combined with dynamic extreme environment simulation according to claim 1, characterized in that: The construction of the extreme environment electricity demand response model includes: Performing necessary demand identification and necessary data extraction on the electricity demand response behavior data set to obtain a necessary electricity demand behavior data set; Mapping the extreme environment calibration power database with the necessary power demand behavior data set according to calibration category parameters to obtain an extreme environment power demand sample data set; Performing neural network learning training on the extreme environment electricity demand sample data set to generate an initial electricity demand analysis model; The performance of the initial electricity demand analysis model is verified and iteratively optimized to construct the extreme environment electricity demand response model.

4. The method for predicting the demand for distribution network supply combined with dynamic extreme environment simulation according to claim 1, characterized in that: The generating of the distribution network supply guarantee demand simulation model comprises: Using the three-dimensional modeling technology to perform spatial modeling on the distribution characteristic data of the target distribution network, a distribution network distribution architecture spatial model is generated; Through the power demand response model and the extreme environment dynamic simulation parameters, the power supply simulation model is performed on the distribution network distribution architecture space model to obtain a basic power supply demand simulation model; The basic supply guarantee demand simulation model is compared and verified to obtain a model reliability coefficient, and parameters of the basic supply guarantee demand simulation model are optimized based on the model reliability coefficient to obtain the distribution network supply guarantee demand simulation model.

5. The method for predicting the demand for distribution network supply combined with dynamic extreme environment simulation according to claim 1, characterized in that: The step of determining the optimization parameters of the power supply reconstruction strategy includes: Performing strategy matching on the extreme environment dynamic simulation parameters based on the distribution network power supply reconstruction strategy library to determine a target power supply reconstruction strategy; Performing parameter optimization analysis on the target distribution network according to the target power supply reconstruction strategy to obtain a reconstruction strategy parameter optimization threshold; Obtaining a reconstruction optimization target, evaluating and fitting the reconstruction optimization target, and obtaining a distribution network reconstruction effect fitness function; Based on the distribution network reconstruction effect fitness function, the reconstruction strategy parameter optimization threshold is optimized to determine the power supply reconstruction strategy optimization parameter.

6. The method for predicting the demand for distribution network supply combined with dynamic extreme environment simulation according to claim 5, characterized in that: The step of determining the power supply reconstruction strategy optimization parameter includes: Randomly obtaining a plurality of reconstruction strategy parameters within the reconstruction strategy parameter optimization threshold; Simulating the multiple reconstruction strategy parameters based on the distribution network supply guarantee demand simulation model to obtain multiple distribution network reconstruction simulation parameters; Evaluating the effects of the plurality of distribution network reconstruction simulation parameters by using the distribution network reconstruction effect fitness function to obtain a plurality of reconstruction effect fitness information; The multiple reconstruction strategy parameters are compared and optimized based on the multiple reconstruction effect fitness information to determine the power supply reconstruction strategy optimization parameters.

7. The method for predicting the demand for distribution network supply combined with dynamic extreme environment simulation according to claim 6, characterized in that: The step of determining the power supply reconstruction strategy optimization parameter includes: Based on the plurality of reconstruction effect fitness information, the plurality of distribution network reconstruction simulation parameters are proportionally screened to obtain an optimal reconstruction strategy parameter set; Setting a parameter variation step length, mutating and expanding the optimal reconstruction strategy parameter set based on the parameter variation step length, and obtaining a reconstruction strategy parameter variation population; An iterative optimization is performed within the reconstruction strategy parameter variation population until a preset number of iterations is reached to determine the optimization parameters of the power supply reconstruction strategy.

8. The method for predicting the demand for distribution network supply combined with dynamic extreme environment simulation according to claim 1, characterized in that: The method comprises: Based on the distribution network supply demand forecasting model, supply demand forecasting and actual comparison feedback are performed to obtain forecast accuracy feedback parameters; Performing model optimization analysis on the prediction accuracy feedback parameters to obtain model parameter optimization rules; The parameters of the distribution network supply demand prediction model are iteratively updated based on the model parameter optimization rules to obtain the distribution network supply demand optimization prediction model.

Images