A power market deduction multi-scenario generation method and related device

Abstract

The application belongs to a scene generation method based on predicted data, and aims at the problem that the generated deduction scene in the current power market simulation deduction lacks rigorous theoretical basis and covering ability for small probability extreme events in actual production, and provides a power market deduction multi-scene generation method and related device, obtains historical operation data, pre-processes the historical operation data, visualizes the pre-processed operation data through a histogram, superimposes a Gaussian distribution fitting curve on the error histogram, obtains the corresponding error probability density function according to the fitting result, and converts the new energy uncertainty problem in the power market into a prediction error uncertainty problem. Then, a Latin hypercube probability sampling method is used to sample the cumulative distribution function to obtain the corresponding combination sample, and finally the combination sample is superimposed on the historical operation data to obtain prediction data for power market deduction, and the power market deduction multi-scene is generated according to the prediction data.

Description

Technical Field

[0001] This application pertains to a scenario generation method based on predictive data, specifically involving a method and related apparatus for generating multiple scenarios for power market simulation. Background Technology

[0002] In new power systems, the high proportion of renewable energy integration poses unprecedented challenges to the operation of the electricity market. In particular, the randomness and intermittency of renewable energy significantly increase the uncertainty of electricity supply and demand balance. Against this backdrop, analyzing and verifying policies and rules through electricity market simulation has become a key means of exploring the operating mechanism of new power systems, and an important tool to support policy formulation, market optimization, and risk assessment.

[0003] The core of electricity market simulation lies in accurately reproducing real-world market behavior and evaluating the adaptability and robustness of market mechanisms to different operating conditions under various scenarios. However, this process places extremely high demands on the rationality, comprehensiveness, and diversity of the input data. Typically, simulation systems are based on historical operating data, fine-tuned using certain rules or experience to generate simulation scenarios. However, this method is limited to simulating known conditions and conventional operating scenarios, lacking a rigorous theoretical foundation and the ability to cover low-probability extreme events in actual production. Summary of the Invention

[0004] This application addresses the technical problem that current power market simulation scenarios lack a rigorous theoretical foundation and the ability to cover low-probability extreme events in actual production. It provides a method and related apparatus for generating multiple scenarios for power market simulation.

[0005] To achieve the above objectives, this application adopts the following technical solution:

[0006] Firstly, this application proposes a method for generating multiple scenarios for electricity market simulation, including:

[0007] Obtain historical operational data;

[0008] The historical running data is preprocessed to obtain preprocessed running data, and the preprocessed running data is visualized using a histogram to obtain an error histogram.

[0009] A Gaussian distribution fitting curve is superimposed on the error histogram, and the corresponding error probability density function is obtained based on the fitting result;

[0010] Calculate the cumulative distribution function corresponding to the Gaussian distribution based on the error probability density function;

[0011] The cumulative distribution function is sampled using the Latin hypercube probability sampling method to obtain the corresponding combined sample;

[0012] The combined samples are superimposed on historical operating data to obtain predictive data for power market projection.

[0013] Multiple scenarios for power market simulation are generated based on the predicted data.

[0014] Furthermore, the error preprocessing of the historical operating data includes abnormal data correction, missing value supplementation, and data normalization.

[0015] Furthermore, the abnormal data correction includes:

[0016] The Laida criterion was used to screen outliers;

[0017] Outlier correction can be performed using at least one of the following methods: imputing outliers with the mean, correcting outliers as missing data, or correcting missing values ​​for outliers.

[0018] Perform similar date replacement and delete outliers.

[0019] Furthermore, the missing value supplementation includes: supplementing missing values ​​using Lagrange interpolation.

[0020] Furthermore, the electricity market includes wind power and solar power;

[0021] The formula for obtaining the corresponding error probability density function based on the fitting result is as follows:

[0022]

[0023] in, Let be the error probability density function for wind power. for t The error value of photovoltaic prediction at any given time. for t The standard deviation of photovoltaic prediction error at any given time. for t The expected value of the photovoltaic prediction error at time point. Let be the error probability density function for photovoltaics. for t The error value of wind power prediction at any given time. for t Standard deviation of wind power forecast error at any given time for t The expected value of wind power prediction error at any given time.

[0024] Furthermore, the cumulative distribution function corresponding to the Gaussian distribution is calculated using the following formula:

[0025]

[0026] in, Let be the cumulative distribution function corresponding to the Gaussian distribution followed by the wind power prediction error. Let be the cumulative distribution function corresponding to the Gaussian distribution that the photovoltaic prediction error follows.

[0027] Furthermore, the method of sampling the cumulative distribution function using the Latin hypercube probability sampling method to obtain the corresponding combined sample includes:

[0028] Divide the cumulative distribution function into n equal intervals to obtain the set T of all intervals; where n is the number of pre-generated scenes.

[0029] Randomly select a uniformly distributed value from each of the n intervals;

[0030] By mapping all selected uniformly distributed values ​​to a Gaussian distribution using the inverse transformation sampling method, sample points of the Gaussian distribution are obtained.

[0031] Based on the Gaussian distribution of sample points, the prediction errors for wind power and photovoltaic power are arranged according to time periods to form a combined sample.

[0032] Furthermore, the generation of multiple scenarios for electricity market simulation based on the predicted data includes:

[0033] A distributed computing module is used to perform parallel calculations on each set of forecast data to generate corresponding power market simulation scenarios.

[0034] Furthermore, before employing a distributed computing module to perform parallel computations on each set of prediction data, the method further includes: packaging the prediction data and the computation task into a mirror image.

[0035] Secondly, this application proposes a multi-scenario generation system for electricity market simulation, including:

[0036] The data module is used to acquire historical operational data;

[0037] An error preprocessing module is used to preprocess the historical running data to obtain preprocessed running data, and to visualize the preprocessed running data using a histogram to obtain an error histogram.

[0038] The Gaussian fitting module is used to overlay a Gaussian distribution fitting curve onto the error histogram and obtain the corresponding error probability density function based on the fitting result.

[0039] The statistics module is used to calculate the cumulative distribution function corresponding to the Gaussian distribution based on the error probability density function;

[0040] The sampling module is used to sample the cumulative distribution function using the Latin hypercube probability sampling method to obtain the corresponding combined samples;

[0041] The overlay module is used to overlay the combined samples onto historical operating data to obtain predictive data for power market projection.

[0042] The simulation module is used to generate multiple scenarios for power market simulation based on the predicted data.

[0043] Thirdly, this application proposes an electronic device, including: a memory and one or more processors; the memory is coupled to the processors; wherein the memory stores computer program code, the computer program code including computer instructions, and when the computer instructions are executed by the processor, the electronic device performs the steps of the above-described method for generating multiple scenarios for power market simulation.

[0044] Fourthly, this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method for generating multiple scenarios for power market simulation.

[0045] Compared with the prior art, this application has the following beneficial effects:

[0046] This application proposes a method for generating multiple scenarios for power market simulation. After acquiring historical operating data, to ensure accuracy in subsequent data processing, the historical operating data is first preprocessed for errors. Then, the preprocessed operating data is visualized using histograms to understand the distribution trend of errors. Next, a Gaussian distribution fitting curve is superimposed on the error histogram, and the corresponding error probability density function is obtained based on the fitting result, transforming the uncertainty problem of new energy in the power market into a prediction error uncertainty problem. Then, the Latin hypercube probability sampling method is used to sample the cumulative distribution function, obtaining corresponding combined samples to ensure uniform coverage of the sample space within the uncertainty range, suitable for multi-scenario analysis. Finally, the combined samples are superimposed on the historical operating data to obtain prediction data for power market simulation, and multiple power market simulation scenarios are generated based on the prediction data. This application improves the adaptability and comprehensiveness of power market simulation to complex operating conditions.

[0047] This application also proposes a multi-scenario generation system, electronic device, and computer storage medium for power market simulation, which possess all the advantages of the aforementioned multi-scenario generation method for power market simulation. Attached Figure Description

[0048] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a schematic diagram of the first method for generating multiple scenarios in the power market simulation of this application.

[0050] Figure 2 This is a schematic diagram of the second process for generating multiple scenarios in the power market simulation method of this application.

[0051] Figure 3 This is a schematic diagram illustrating the principle of the DistRun distributed computing module.

[0052] Figure 4 This is a schematic diagram illustrating the process principle of the multi-scenario generation method for electricity market simulation in this application.

[0053] Figure 5 This is a schematic diagram of a multi-scenario generation system for power market simulation in this application. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0055] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0056] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0057] In the description of the embodiments of this application, it should be noted that if terms such as "upper," "lower," "horizontal," or "inner" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of the invention is in use, they are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this application. In addition, terms such as "first" and "second" are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0058] Furthermore, the use of the term "horizontal" does not imply that the component must be absolutely horizontal, but rather that it can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0059] In the description of the embodiments of this application, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0060] In new power systems, with the significant increase in the integration of new energy sources (such as wind and solar power), the operation of the electricity market faces unprecedented challenges. These new energy sources exhibit significant randomness and intermittency; their power generation output is affected by various uncontrollable factors such as weather conditions and seasonal variations, leading to a significant increase in the uncertainty of electricity supply and demand balance. This uncertainty also puts enormous pressure on the stable operation of the power system and the effective dispatch of the electricity market. To address these challenges, electricity market simulation has become an important analytical tool. By constructing simulation models of the electricity market and conducting simulations based on actual or hypothetical market conditions, the actual effects of different policies and rules in the electricity market can be analyzed and verified in depth. This simulation not only helps explore the operating mechanisms of new power systems but also provides policymakers with a scientific basis, helping to formulate more rational and effective electricity market policies. Simultaneously, electricity market simulation is also an important tool for market optimization and risk assessment. By simulating electricity supply and demand under different market conditions, the adaptability and robustness of market mechanisms can be evaluated, and potential market problems and risks can be identified. This helps market operators adjust market rules in a timely manner, optimize market structure, and improve the efficiency and stability of the electricity market.

[0061] The core of electricity market simulation lies in accurately reproducing real-world market behavior. This requires the simulation system to comprehensively and accurately reflect the complex relationships and dynamic changes within the electricity market. To achieve this, the simulation system needs to be built and run based on a large amount of input data. This input data includes historical operating data, market rules, the cost structure of power generation companies, and user demand behavior. Their rationality and comprehensiveness are crucial to the accuracy of the simulation results. However, relying solely on historical operating data for simulation is far from sufficient. Historical data only reflects past market conditions and cannot cover all possible future scenarios. Especially for low-probability but significant extreme events, historical data often lacks sufficient samples for simulation. Therefore, the simulation system also needs fine-tuning through certain rules or experience to generate more diverse and realistic simulation scenarios. However, this rule- and experience-based fine-tuning method also has limitations. It mainly relies on manually set rules and parameters and lacks a rigorous theoretical foundation. Furthermore, due to the limitations of human cognition and experience, this method may not be able to fully cover all possible operating conditions and extreme events. Therefore, in order to improve the accuracy and reliability of electricity market simulation, it is necessary to continuously explore and introduce new theories and methods to better simulate and predict the dynamic changes in the electricity market.

[0062] In actual operation, low-probability scenarios such as extreme weather events and large-scale grid connection of renewable energy may cause the market to deviate from its normal trajectory, triggering sharp price fluctuations, market failures, and even power supply crises. Although these scenarios are rare, their potential impact can be catastrophic. Therefore, how to construct input datasets that can reflect low-probability extreme events in power market simulations has become a key focus and challenge in current research. Thus, for new power systems with a high proportion of renewable energy integration, rationally constructing input data and conducting power market simulations is a crucial step in improving market resilience and efficiency. By constructing comprehensive and scientific simulation scenarios, the power market will be better able to adapt to the challenges of future complex operating environments, while providing key decision support for policymakers and ensuring the achievement of system stability and economic goals.

[0063] Based on the above, this application proposes a method and related apparatus for generating multiple scenarios for power market simulation. The following is a detailed description of this application in conjunction with embodiments and accompanying drawings.

[0064] like Figure 1 The diagram shown is a first illustration of the multi-scenario generation method for electricity market simulation in this application, which may include:

[0065] S101, retrieve historical operation data.

[0066] Historical operating data forms the basis for electricity market projections, helping to understand market operating patterns and potential risks. Collected historical operating data can include power generation, electricity consumption, electricity prices, records of market participant behavior, weather conditions, and equipment failure records. Data sources may include power company operating reports, publicly available data from market regulators, and weather records from meteorological departments. The time span of the data can be set according to forecasting requirements; for example, it can cover different market cycles and extreme events to ensure the comprehensiveness and representativeness of the data.

[0067] S102, perform error preprocessing on the historical running data to obtain preprocessed running data, and visualize the preprocessed running data using a histogram to obtain an error histogram.

[0068] Error preprocessing helps eliminate noise and bias in data, improving its reliability and usability. Histogram visualization provides an intuitive view of data distribution, helping to identify data characteristics and potential problems.

[0069] When formulating day-ahead power generation plans, power grid dispatching departments need to know the next day's active power forecasts for renewable energy and system load demand across the entire grid. Due to the vast scale of the power system and its complex and variable operating conditions, including orderly power consumption, channel obstruction, extreme weather, and equipment failures, historical operating databases contain some missing and anomalies. To avoid adverse data interfering with the accuracy of renewable energy and load forecasts, historical data should be preprocessed. After preprocessing, histograms are used to visualize the preprocessed operating data, showing the data distribution characteristics and error patterns, forming an error histogram.

[0070] S103, Gaussian distribution fitting curve is superimposed on the error histogram, and the corresponding error probability density function is obtained based on the fitting result.

[0071] Gaussian distribution fitting helps in understanding the distribution patterns of data errors, providing a theoretical basis for subsequent sampling and extrapolation. The probability density function is an important tool for describing the probability of a random variable's values, providing detailed statistical information about data errors. A Gaussian distribution is used to fit the error histogram to describe the probability distribution characteristics of the data. By adjusting the parameters of the Gaussian distribution, the fitted curve is made to closely resemble the contour of the histogram. After fitting, the probability density function describing the data error can be obtained from the fitting results.

[0072] S104. Calculate the cumulative distribution function corresponding to the Gaussian distribution based on the error probability density function.

[0073] The cumulative distribution function (CDF) is fundamental for probability sampling, providing a method for drawing samples from a probability density function. In electricity market simulations, the CDF helps generate forecast data that conforms to the actual error distribution. Using the error probability density function, the corresponding Gaussian cumulative distribution function can be calculated. The cumulative distribution function describes the probability that a random variable takes a value less than or equal to a specific value.

[0074] S105, The cumulative distribution function is sampled using the Latin hypercube probability sampling method to obtain the corresponding combined sample.

[0075] Latin hypercube sampling is an efficient probability sampling method that reduces the sample size while ensuring representativeness. By applying Latin hypercube sampling to the cumulative distribution function, a set of samples with a specific probability distribution can be obtained, which constitute the composite samples required for inference. Latin hypercube sampling improves sampling efficiency and accuracy, ensuring the diversity and representativeness of the inference scenarios. The composite samples are the foundation for generating the inference scenarios, and their quality and quantity directly affect the reliability and comprehensiveness of the inference results.

[0076] S106, The combined samples are superimposed on historical operating data to obtain forecast data for power market projection.

[0077] By overlaying the combined samples obtained from Latin hypercube sampling onto historical operating data, predictive data containing error information is formed. This predictive data retains the overall characteristics of the historical data while incorporating the error distribution information obtained through sampling. By overlaying and combining samples, the predictive data can more accurately reflect the potential risks and uncertainties of the market.

[0078] S107, Generate multiple scenarios for power market simulation based on the predicted data.

[0079] Based on forecast data, multiple electricity market simulation scenarios can be generated. Each scenario represents a possible future operating state of the market and can include forecasts of key variables such as power generation, electricity consumption, electricity prices, and market participant behavior. These scenarios can cover different market conditions and extreme events to ensure the comprehensiveness and robustness of the simulation.

[0080] This application addresses the limitations of existing technologies in multi-scenario power market simulation and proposes a multi-scenario simulation data generation method based on probability distribution. This method can generate output curves of new energy units according to the needs of different power market simulation scenarios, and combine these data with known quantities to form the boundary conditions for power market simulation. This method provides data support for the analysis and verification of continuous simulation of multi-scenario power markets. Regarding the prediction and uncertainty modeling of new energy output, this application uses a Gaussian distribution to statistically model the prediction error of new energy. To generate diverse scenario data, this application combines Latin hypercube sampling technology to sample from the fitted Gaussian distribution, generating new energy output data and system load prediction data covering multiple scenarios. The Latin hypercube sampling technology ensures uniform coverage of the sample space, enabling the simulation data to include both conventional scenarios and low-probability extreme scenarios, thereby improving the adaptability and comprehensiveness of power market simulation to complex operating conditions.

[0081] like Figure 2 The diagram shown is a second flowchart illustrating the multi-scenario generation method for electricity market simulation in this application, which may include:

[0082] S201, Historical Data Preprocessing.

[0083] To avoid adverse data interfering with the accuracy of new energy forecasting and load forecasting, the historical data is preprocessed as follows:

[0084] (1) Abnormal data identification and correction.

[0085] Values ​​that deviate significantly from the sample distribution range are considered outliers. In the power system operation reserve capacity demand assessment model, the sample data consists of the day-ahead forecast deviation of new energy sources and the day-ahead forecast deviation of load. When the sample size is large enough, the Laida criterion can usually be used to screen outliers. Generally, values ​​that deviate from the total sample mean by more than three times the standard deviation are considered outliers.

[0086]

[0087] in, This refers to errors in new energy forecasting or load forecasting. The sample mean. This represents the sample standard deviation.

[0088] When a value satisfies the Laida criterion, it is considered an outlier, and the following outlier handling methods can be applied:

[0089] 1) Mean filling: Take the average of the same number of samples before and after the outlier as the filling value.

[0090] 2) Missing value correction: Treat outliers as missing data in the sample.

[0091] 3) Similar day replacement: For a small portion of the data anomalies, the anomaly data are replaced with similar days that have a high correlation with the predicted power.

[0092] 4) Direct deletion: For data anomalies that occur for several consecutive days in the historical database, the outliers can be deleted directly if the sample size is large enough.

[0093] (2) Missing value supplementation.

[0094] Missing values ​​in the prediction error were filled using Lagrange interpolation.

[0095]

[0096] Where k is the total number of sampling periods, For interpolation basis functions, for The prediction error corresponding to the time period This is the sampling time period number. This is the sampling time period number. For the sampling period, For sampling Time period For sampling Time period. The missing values ​​can be calculated by substituting the time period of the missing sample into the Lagrange interpolation formula.

[0097] (3) Sample data normalization.

[0098] After completing outlier replacement and missing value imputation, the sample population needs to be normalized for easier data processing.

[0099]

[0100] in, This represents the actual value of the new energy prediction error. This represents the per-unit value of the new energy prediction error. This represents the installed capacity of new energy sources for the corresponding time period.

[0101] S202, Draw an error distribution histogram.

[0102] Visualize the distribution of error data using histograms. Histograms show the sample frequency within each error range, helping us understand the distribution trend of errors.

[0103] S203, New Energy Prediction Error Fitting Based on Gaussian Distribution.

[0104] The probabilistic model-based scene generation method fits the probability density distribution of historical data and samples the probability distribution to obtain the output scene.

[0105] This application considers the output of new energy sources as the sum of predicted power and uncertain predicted power errors. To reduce model complexity and facilitate quantitative analysis of uncertainties in wind and solar power generation, the uncertainty problem of wind and solar power output is transformed into a prediction error uncertainty problem. Therefore, the actual power generation of wind and solar power can be expressed as follows:

[0106]

[0107] in, For photovoltaic at time t, Let t be the actual wind power output at time t. Let t be the predicted photovoltaic power output. Let t be the predicted wind power output. Let be the photovoltaic prediction error value at time t. Let t be the wind power prediction error value at time t.

[0108] The prediction errors for wind power and photovoltaic power follow a Gaussian distribution, and the corresponding probability density function can be expressed as:

[0109]

[0110] in, Let be the error probability density function for wind power. for t The error value of photovoltaic prediction at any given time. for t The standard deviation of photovoltaic prediction error at any given time. for t The expected value of the photovoltaic prediction error at time point. Let be the error probability density function for photovoltaics. for t The error value of wind power prediction at any given time. for t Standard deviation of wind power forecast error at any given time for t The expected value of wind power prediction error at any given time.

[0111] S204, Latin hypercube probability sampling.

[0112] After fitting a Gaussian distribution to the prediction error of new energy sources, Latin hypercube sampling (LHS) can be used to generate forecast data for electricity market simulation. This method ensures uniform coverage of the sample space within the uncertainty range, making it well-suited for multi-scenario analysis.

[0113] Latin hypercube sampling is a stratified sampling method that ensures that a sample point is selected from each distribution interval, uniformly distributed throughout the space. For applications using the Gaussian distribution, the key steps are as follows:

[0114] 1) Set the number of samples.

[0115] Define the number of samples n to be generated, which is the number of scenarios we want to generate.

[0116] 2) Hierarchical interval division.

[0117] The cumulative distribution function (CDF) describes the probability that a random variable takes a value less than or equal to a specific value. When sampling, the inverse transformation of the CDF is typically used to generate samples that conform to a specific distribution. The CDF corresponding to the Gaussian distribution followed by the aforementioned wind power and photovoltaic prediction errors is shown below:

[0118]

[0119] Divide the cumulative distribution function (CDF) of the Gaussian distribution into n equal intervals. The range of the cumulative distribution function F(x) is [0, 1]. Divide it into n segments, with each interval having a width of 1 / n. Then, the set T of all intervals can be represented as:

[0120] .

[0121] 3) Samples are drawn from the interval.

[0122] Randomly select a sample point within each interval to ensure coverage of the entire distribution.

[0123] For the i Given a range of values, randomly select a uniformly distributed value. :

[0124] .

[0125] 4) Map the sample points to a Gaussian distribution.

[0126] These uniformly distributed points are mapped onto a Gaussian distribution using the inverse transform sampling method.

[0127] Inverse transform of the cumulative distribution function of the Gaussian distribution ,Will Convert sample points to Gaussian distribution :

[0128]

[0129] in, It is the inverse function of the Gaussian distribution and can be calculated using numerical methods.

[0130] above That is, the first i The prediction error values ​​corresponding to each stratified interval can be used to obtain samples of wind power and solar power output prediction errors using the above method. These error samples are independent and can be arranged according to time periods to form combined samples:

[0131]

[0132] S205 generates multi-scene data based on Latin hypercube.

[0133] The generated combined samples are superimposed on the original new energy forecast data to obtain new energy forecast data that can be used for extrapolation and analysis. Specifically, the predicted value at each time point is added to its corresponding error sample value, so that the predicted value at each time point becomes closer to the actual situation that may occur in operation due to the inclusion of error samples.

[0134] S206, parallel computing of multiple simulation scenarios.

[0135] Assume there are n sets of new energy prediction data in the simulation scenario. These data are extracted and generated from a specified sample set according to the simulation requirements. Each set of data reflects different new energy output trends, covering various types of energy changes such as wind power and solar power.

[0136] As an example, to quickly process these simulation scenarios and obtain calculation results, MindOptStudio's DistRun distributed computing module can be used, such as... Figure 3 The diagram shown illustrates the principle of the DistRun distributed computing module. It's important to note that MindOpt Studio is a cloud platform that provides an easy-to-use interface and powerful backend computing capabilities. The DistRun distributed computing module is a key component of MindOpt Studio, leveraging distributed computing technology to improve the efficiency of solving optimization problems. By breaking down large computational tasks into multiple smaller tasks and processing them in parallel across multiple computers, the DistRun module significantly reduces solution time and improves the utilization of computing resources. Each computing node independently handles the corresponding scenario computation based on task allocation, avoiding bottlenecks caused by serial computation and accelerating the overall computation process.

[0137] Based on the functions and usage of the DistRun module, in order to utilize distributed cluster scheduling tasks, multi-scenario simulation tasks containing new energy prediction data can be packaged into images and created as batch tasks. This allows for parallel processing of multiple scenarios' clearing preparation, file generation, optimization calculations, and post-event statistical clearing tasks, improving simulation efficiency and obtaining simulation results faster. In practical applications, containerization technology (such as Docker) can be used when packaging into images.

[0138] To summarize the methods of the above embodiments of this application, as follows: Figure 4 The diagram shown illustrates the process principle of the multi-scenario generation method for power market projection proposed in this application. Based on the probability distribution model of new energy and load forecasting errors, the method uses Latin hypercube sampling to generate simulated forecasting error data. The validity is then verified based on the probability density functions of the simulated and historical data. Finally, the entire process, from parallel clearing of multiple scenarios based on the DistRun module, completes the generation of multiple projection scenarios from load forecasting data and new energy forecasting data.

[0139] As an example, subsequent analysis is conducted based on the multi-scenario power market simulation generated using the aforementioned method, to further illustrate the role of this application:

[0140] After completing the parallel clearing of simulation examples such as "extreme heat and no wind," we conducted an in-depth analysis of key statistical indicators, including market electricity price information, market participant revenue, unit output, and market supply and demand. Through meticulous analysis of this data, we calculated the expected value of each indicator, revealing the average performance of these indicators in the simulation examples. Simultaneously, calculating variance allows for a quantitative measure of market volatility, helping to identify key factors that may affect market stability. Further analysis revealed that the stability of market electricity prices is closely related to the actual output of renewable energy units, while the volatility of thermal power unit revenue may be affected by the supply-demand ratio and market bidding space. Sensitivity analysis and risk assessment provided us with profound insights into potential risks in market operations, guiding us to develop more robust market strategies. In practical applications, a Gaussian distribution is fitted based on the characteristics of new energy prediction error samples in different forms such as extreme heat and no wind. Multiple sets of prediction error data under this scenario are generated through Latin hypercube sampling. Then, multiple sets of case data are generated by combining other invariants in the case data. This constitutes a complete set of multiple sets of inference case data based on the characteristics of the specified scenario. Mindopt is used to call multiple resources to realize the time-series parallel computation of multiple sets of inference scenarios, thereby improving the computational efficiency. The final statistical analysis data of the calculation cases is obtained based on expectation and variance analysis.

[0141] In the electricity market, variance statistics and expected value statistics are key tools for evaluating market performance and supporting decision-making. Analysis of variance (ANOVA) is used to compare significant differences in electricity price fluctuations and generating capacity output across different time periods or market conditions, while expected value statistics focus on average market performance, such as average market electricity price and average generating capacity output. Through in-depth analysis of the clearing results of simulation groups such as "extremely hot and windless," we collected key statistical indicators including market electricity price, generating capacity revenue, output, and supply-demand ratio. Calculating the expected value of these indicators provides the central trend of market performance, while variance measures the volatility of the projected data. Smaller price variance indicates relatively stable market prices, while larger generating capacity variance may indicate greater fluctuations in the performance of specific generating units across different simulations. The application of these statistical results not only helps quantify market performance but also assesses the electricity market performance of simulation groups such as "extremely hot and windless," providing quantitative support for research on the high proportion of renewable energy participation in electricity market operations. Based on the characteristics of new energy prediction error samples in different forms such as extreme heat and no wind, a Gaussian distribution is fitted, and multiple sets of prediction error data under this scenario are generated by sampling. Then, multiple sets of case data are generated by combining other invariants in the case data, thus forming a complete set of multiple inference case data based on the characteristics of the specified scenario.

[0142] To address the limitations of existing technologies in multi-scenario power market simulation, this application proposes a method and related apparatus for generating multiple power market simulation scenarios. This method can generate output curves of new energy generating units based on the requirements of different target simulation scenarios, and combine these data with known quantities as boundary conditions for power market simulation. This technology supports continuous simulation of power markets across multiple scenarios, providing comprehensive quantitative evaluation support for the analysis and verification of policies, rules, and operational mechanisms.

[0143] In the prediction and uncertainty modeling of new energy output, prediction errors typically concentrate around the predicted value, exhibiting a certain peak characteristic, and may also show significant deviations due to randomness and volatility. Therefore, this application employs a Gaussian distribution to statistically model new energy prediction errors, accurately characterizing the probability distribution of the error through mean and standard deviation parameters. The Gaussian distribution has a clear theoretical foundation and flexible applicability, effectively describing the main characteristics of new energy prediction errors, while also facilitating mathematical processing and sampling generation.

[0144] To generate diverse scenario data, this application combines Latin hypercube sampling technology to sample from a fitted Gaussian distribution, generating renewable energy output data and system load forecast data covering multiple scenarios. Latin hypercube sampling can uniformly cover the sample space within the uncertainty range, ensuring that the simulation data includes both conventional scenarios and low-probability extreme scenarios, thereby improving the adaptability and comprehensiveness of power market simulation under complex operating conditions. Through the modeling and data generation method combining Gaussian distribution and Latin hypercube sampling, this application can accurately characterize the uncertainty of renewable energy output, providing a scientific and comprehensive input foundation for power market simulation, and effectively supporting continuous simulation and analysis verification of multiple scenarios under different simulation objectives and needs.

[0145] like Figure 5 The diagram shown is a schematic representation of a multi-scenario generation system for electricity market simulation according to this application, which may include:

[0146] The data module is used to acquire historical operational data;

[0147] An error preprocessing module is used to preprocess the historical running data to obtain preprocessed running data, and to visualize the preprocessed running data using a histogram to obtain an error histogram.

[0148] The Gaussian fitting module is used to overlay a Gaussian distribution fitting curve onto the error histogram and obtain the corresponding error probability density function based on the fitting result.

[0149] The statistics module is used to calculate the cumulative distribution function corresponding to the Gaussian distribution based on the error probability density function;

[0150] The sampling module is used to sample the cumulative distribution function using the Latin hypercube probability sampling method to obtain the corresponding combined samples;

[0151] The overlay module is used to overlay the combined samples onto historical operating data to obtain predictive data for power market projection.

[0152] The simulation module is used to generate multiple scenarios for power market simulation based on the predicted data.

[0153] In some embodiments of the power market simulation multi-scenario generation system of this application, the error preprocessing module includes an abnormal data correction unit, a missing value supplementation unit, a data normalization processing unit, and a histogram visualization unit; the abnormal data correction unit is used to correct abnormal data in the historical operating data; the missing value supplementation unit is used to supplement missing values ​​in the data after abnormal data correction; the data normalization processing unit is used to normalize the data after missing value supplementation; and the histogram visualization unit is used to visualize the preprocessed operating data using a histogram to obtain an error histogram.

[0154] In some embodiments of the power market simulation multi-scenario generation system of this application, the abnormal data correction unit is used to screen outliers using the Laida criterion; the abnormal data correction unit is also used to correct outliers using at least one of the following methods: filling out outliers with the mean, correcting outliers as missing data, replacing outliers with similar days, and deleting outliers.

[0155] In some embodiments of the power market simulation multi-scenario generation system of this application, the missing value supplementation unit is used to supplement missing values ​​using the Lagrange interpolation method.

[0156] In some embodiments of the power market simulation multi-scenario generation system of this application, the power market includes wind power and photovoltaic power;

[0157] The Gaussian fitting module uses the following formula to calculate the corresponding error probability density function based on the fitting results:

[0158]

[0159] in, Let be the error probability density function for wind power. for t The error value of photovoltaic prediction at any given time. for t The standard deviation of photovoltaic prediction error at any given time. for t The expected value of the photovoltaic prediction error at time point. Let be the error probability density function for photovoltaics. for t The error value of wind power prediction at any given time. for t Standard deviation of wind power forecast error at any given time for t The expected value of wind power prediction error at any given time.

[0160] In some embodiments of the multi-scenario generation system for electricity market simulation in this application, the statistical module uses the following formula to calculate the cumulative distribution function corresponding to the Gaussian distribution:

[0161] .

[0162] In some embodiments of the power market simulation multi-scenario generation system of this application, the sampling module includes an equal division unit, a value selection unit, an inverse transformation sampling unit, and an arrangement unit;

[0163] The equal division unit is used to divide the cumulative distribution function into n equal intervals to obtain the set T of all intervals; where n is the preset number of generated scenes;

[0164] The value selection unit is used to randomly select a uniformly distributed value from each of the n intervals.

[0165] The inverse transformation sampling unit is used to map all selected uniformly distributed values ​​to a Gaussian distribution using the inverse transformation sampling method, thereby obtaining Gaussian distributed sample points.

[0166] The arrangement unit is used to arrange the prediction errors of wind power and photovoltaic power according to the time period based on the Gaussian distributed sample points to form a combined sample.

[0167] It should be noted that, in the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For instance, the division of each module is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another device, or some features may be ignored or not executed. The modules described as separate components may or may not be physically separated. The components shown as modules may be one or more physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the modules can be selected to achieve the purpose of the solution in this embodiment according to actual needs.

[0168] Furthermore, in the various embodiments of the present invention, the modules can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The integrated unit described above can be implemented in hardware or as a software functional unit.

[0169] This application also provides an electronic device, which may include one or more processors, memory and communication interfaces.

[0170] The memory, communication interface, and processor are coupled together. For example, the memory, communication interface, and processor can be coupled together via a bus.

[0171] The communication interface is used for data transmission with other devices. The memory stores computer program code. This computer program code includes computer instructions, which, when executed by the processor, cause the electronic device to perform the steps of the aforementioned multi-scenario generation method for electricity market simulation.

[0172] The processor can be a processor or controller, such as a Central Processing Unit (CPU), a general-purpose processor, a Digital Signal Processor (DSP), an Application-Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with this disclosure. The processor can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc. The processor can be used to support an electronic device in performing the method steps provided in the above embodiments.

[0173] The bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. These buses can be categorized as address buses, data buses, control buses, etc.

[0174] This application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the steps of the above-described method for generating multiple scenarios for power market simulation.

[0175] The computer-readable storage media involved in this application include random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or any other form of storage media known in the art.

[0176] For descriptions of relevant parts of the power market simulation multi-scenario generation system, electronic equipment, and computer-readable storage medium provided in this application's embodiments, please refer to the detailed descriptions of the corresponding parts in the power market simulation multi-scenario generation method provided in this application's embodiments; they will not be repeated here. Furthermore, parts of the technical solutions provided in this application that are consistent with the implementation principles of corresponding technical solutions in the prior art have not been described in detail to avoid excessive elaboration.

[0177] The above are merely preferred embodiments of this application and are not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for generating multiple scenarios for electricity market simulation, including new energy output scenarios for wind power and photovoltaic power; characterized in that, include: Obtain historical operational data; The historical running data is preprocessed to obtain preprocessed running data, and the preprocessed running data is visualized using a histogram to obtain an error histogram. A Gaussian distribution fitting curve is superimposed on the error histogram, and the corresponding error probability density function is obtained based on the fitting result: in, Let be the error probability density function for wind power. for t The error value of photovoltaic prediction at any given time. for t The standard deviation of photovoltaic prediction error at any given time. for t The expected value of the photovoltaic prediction error at time point. Let be the error probability density function for photovoltaics. for t The error value of wind power prediction at any given time. for t Standard deviation of wind power forecast error at any given time for t The expected value of the wind power prediction error at any given time; Based on the error probability density function, calculate the cumulative distribution function corresponding to the Gaussian distribution: in, Let be the cumulative distribution function corresponding to the Gaussian distribution followed by the wind power prediction error. Let be the cumulative distribution function corresponding to the Gaussian distribution followed by the photovoltaic prediction error; The cumulative distribution function is sampled using the Latin hypercube probability sampling method to obtain the corresponding combined sample: Divide the cumulative distribution function into n equal intervals to obtain the set T of all intervals; where n is the number of pre-generated scenes. Randomly select a uniformly distributed value from each of the n intervals; By using the inverse transformation sampling method, all selected uniformly distributed values ​​are mapped onto a Gaussian distribution to obtain Gaussian distributed sample points. Based on the Gaussian distribution of sample points, the prediction errors for wind power and photovoltaic power are arranged according to time periods to form a combined sample. The combined samples are superimposed on historical operating data to obtain predictive data for power market projection. Multiple scenarios for power market simulation are generated based on the predicted data.

2. The method for generating multiple scenarios for electricity market simulation according to claim 1, characterized in that, The error preprocessing of the historical operating data includes abnormal data correction, missing value supplementation, and data normalization.

3. The method for generating multiple scenarios for electricity market simulation according to claim 2, characterized in that, The correction of the abnormal data includes: The Laida criterion was used to screen outliers; Outlier correction can be performed using at least one of the following methods: imputing outliers with the mean, correcting outliers as missing data, or correcting missing values ​​for outliers. Perform similar date replacement and delete outliers.

4. The method for generating multiple scenarios for electricity market simulation according to claim 2, characterized in that, The missing value supplementation includes: supplementing missing values ​​using Lagrange interpolation.

5. The method for generating multiple scenarios for electricity market simulation according to claim 1, characterized in that, The generation of multiple scenarios for electricity market simulation based on the predicted data includes: A distributed computing module is used to perform parallel calculations on each set of forecast data to generate corresponding power market simulation scenarios.

6. The method for generating multiple scenarios for electricity market simulation according to claim 5, characterized in that, Before using a distributed computing module to perform parallel calculations on each set of prediction data, the method further includes: packaging the prediction data and the task of performing the calculations into an image.

7. A multi-scenario generation system for electricity market simulation, wherein the electricity market includes wind power and photovoltaic power; characterized in that, include: The data module is used to acquire historical operational data; An error preprocessing module is used to preprocess the historical running data to obtain preprocessed running data, and to visualize the preprocessed running data using a histogram to obtain an error histogram. The Gaussian fitting module is used to overlay a Gaussian distribution fitting curve onto the error histogram and obtain the corresponding error probability density function based on the fitting result. in, Let be the error probability density function for wind power. for t The error value of photovoltaic prediction at any given time. for t The standard deviation of photovoltaic prediction error at any given time. for t The expected value of the photovoltaic prediction error at time point. Let be the error probability density function for photovoltaics. for t The error value of wind power prediction at any given time. for t Standard deviation of wind power forecast error at any given time for t The expected value of the wind power prediction error at any given time; The statistics module is used to calculate the cumulative distribution function corresponding to the Gaussian distribution based on the error probability density function. in, Let be the cumulative distribution function corresponding to the Gaussian distribution followed by the wind power prediction error. Let be the cumulative distribution function corresponding to the Gaussian distribution followed by the photovoltaic prediction error; The sampling module is used to sample the cumulative distribution function using the Latin hypercube probability sampling method to obtain the corresponding combined samples; The overlay module is used to overlay the combined samples onto historical operating data to obtain predictive data for power market projection. The simulation module is used to generate multiple scenarios for power market simulation based on the predicted data.

8. The multi-scenario generation system for power market simulation according to claim 7, characterized in that, The error preprocessing module includes an abnormal data correction unit, a missing value supplementation unit, a data normalization processing unit, and a histogram visualization unit. The abnormal data correction unit is used to correct abnormal data in the historical operation data; The missing value supplementation unit is used to supplement missing values ​​in the data after the abnormal data has been corrected. The data normalization processing unit is used to perform data normalization processing on the data after missing values ​​are filled in; The histogram visualization unit is used to visualize the preprocessed running data using histograms to obtain an error histogram.

9. The multi-scenario generation system for power market simulation according to claim 8, characterized in that, The abnormal data correction unit is used to filter out outliers using the Laida criterion; The abnormal data correction unit is also used for Use at least one of the following methods to correct outliers: The functions include: imputing outliers with the mean, correcting outliers as missing data, replacing outliers with similar days, and deleting outliers.

10. The multi-scenario generation system for power market simulation according to claim 8, characterized in that, The missing value supplementation unit is used to supplement missing values ​​using Lagrange interpolation.

11. The multi-scenario generation system for power market simulation according to claim 10, characterized in that, The sampling module includes an equal division unit, a value selection unit, an inverse transformation sampling unit, and an arrangement unit; The equal division unit is used to divide the cumulative distribution function into n equal intervals to obtain the set T of all intervals; where n is the preset number of generated scenes; The value selection unit is used to randomly select a uniformly distributed value from each of the n intervals. The inverse transformation sampling unit is used to map all selected uniformly distributed values ​​to a Gaussian distribution using the inverse transformation sampling method, thereby obtaining Gaussian distributed sample points. The arrangement unit is used to arrange the prediction errors of wind power and photovoltaic power according to the time period based on the Gaussian distributed sample points to form a combined sample.

12. An electronic device, characterized in that, include: A memory, one or more processors; the memory is coupled to the processors; wherein the memory stores computer program code, the computer program code including computer instructions, and when the computer instructions are executed by the processor, the electronic device performs the steps of the power market simulation multi-scenario generation method as described in any one of claims 1-6.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the multi-scenario generation method for power market simulation as described in any one of claims 1-6.

Images