New energy multi-dimensional uncertainty intelligent evaluation method considering prediction error

Abstract

The invention relates to the technical field of novel power system dispatching, in particular to a new energy multi-dimensional uncertainty intelligent evaluation method considering prediction errors, and the method comprises the steps: carrying out the kernel density estimation method probability distribution fitting of the prediction errors of a new energy station, so as to calculate the average prediction error of the output of the station, then establishing a new energy station output uncertainty evaluation system under an intermittent level, a randomness level and a volatility level, dividing a sample set into a plurality of scene sets, extracting typical scenes under different scene sets, and then establishing a four-layer structure model for new energy station output uncertainty evaluation; and new energy uncertainty indexes in different typical scenes are calculated, so that a new energy output uncertainty intelligent evaluation model based on the neural network is constructed. Therefore, the problems that the output uncertainty of the new energy station is difficult to comprehensively and quantitatively describe and the short-term scheduling uncertainty of the power system is difficult to accurately evaluate in the prior art are solved.

Description

Technical Field

[0001] This application relates to the field of novel power system dispatching technology, and in particular to an intelligent assessment method for multidimensional uncertainty of new energy sources that takes into account prediction errors. Background Technology

[0002] Against the backdrop of advancing the "dual carbon" goals, the "Action Plan for Peak Carbon Reduction by 2030" clearly proposes to build a new power system with a gradually increasing proportion of new energy sources and promote the large-scale optimization and allocation of clean power resources. Driven by this policy guidance and the needs of energy transition, the new energy industry, represented by wind power and photovoltaics, has developed rapidly, achieving a significant increase in installed capacity and becoming a core force in the construction of the new power system.

[0003] In related technologies, the characteristics and influencing factors of new energy power output have been studied and understood to a certain extent. At the technical level, it is clear that the output of wind power and photovoltaic power is directly affected by meteorological parameters such as wind speed, light intensity, and temperature, resulting in significant uncertainty in their output. This uncertainty can lead to risks such as wind and solar power curtailment and load shedding in the power system, posing a challenge to the safe and stable operation of the system. At the same time, there are already evaluation methods for the uncertainty of the output of new energy power plants in related technologies, which can be used for preliminary assessment of the uncertainty.

[0004] However, in related technologies, the evaluation of the uncertainty of the output of existing new energy power plants is mostly based on a single perspective and does not take into account multiple operating scenarios. This makes it difficult to comprehensively assess the degree and impact of uncertainty, which in turn limits the reference significance of the evaluation results for the optimized operation of the power system. It cannot provide sufficient support for the power system to cope with the challenges of uncertainty in the output of new energy, and also restricts the promotion of the optimized allocation of clean power resources in the new power system. This issue urgently needs to be addressed. Summary of the Invention

[0005] This application provides an intelligent assessment method for multidimensional uncertainty of new energy sources that considers prediction errors, in order to solve the problem that related technologies mostly assess the output uncertainty of new energy power plants from a single perspective, without taking into account multiple operating scenarios, which makes it difficult to comprehensively quantify and describe the output uncertainty of new energy power plants and to accurately assess the short-term dispatch uncertainty of the power system.

[0006] The first aspect of this application provides a method for intelligent assessment of multidimensional uncertainty of new energy considering prediction errors, comprising the following steps: fitting the probability distribution of prediction errors of new energy power plants using kernel density estimation to calculate the average prediction error of power plant output; establishing an assessment system for the uncertainty of new energy power plant output under intermittent, random, and fluctuating levels based on the prediction error distribution corresponding to the average prediction error of power plant output and the temporal characteristics of new energy output; dividing the sample set into multiple scenario sets based on the characteristics of new energy output, and extracting at least one typical scenario under different scenario sets; establishing a four-layer structure model for assessing the uncertainty of new energy power plant output based on the assessment system and the at least one typical scenario to calculate the new energy uncertainty index under different typical scenarios; and using the new energy uncertainty index to establish a model training dataset to construct a neural network-based intelligent assessment model for the uncertainty of new energy output.

[0007] Through the above technical means, the embodiments of this application can accurately quantify the power plant output prediction error through kernel density estimation, laying a reliable data foundation for assessment. Then, a three-level uncertainty assessment system is constructed to achieve comprehensive coverage of the uncertainty of new energy power output, avoiding the limitations of a single dimension. The system also divides the scenario set to extract typical scenarios, allowing the assessment to focus on different operating conditions and improve the assessment's relevance. Subsequently, a four-layer structural model is established to systematically calculate scenario-based uncertainty indicators, making the assessment more structured and accurate. Based on the indicators, a neural network intelligent assessment model is constructed to achieve efficient and intelligent assessment of the uncertainty of new energy power output. Overall, this provides scientific and accurate decision support for power system dispatch optimization, risk prevention and control, and power plant operation planning.

[0008] Optionally, in one embodiment of this application, the step of fitting the probability distribution of the prediction error of the new energy power station using the kernel density estimation method to calculate the average error of the power station output prediction includes: fitting the probability distribution of the new energy power output prediction error using the kernel density estimation strategy in the preset nonparametric fitting strategy; and based on the probability distribution, using the probability of the new energy power output in each error interval as a weighting coefficient to perform a weighted summation of the new energy daily power output error as the average error of the power station output prediction.

[0009] Through the above technical means, the embodiments of this application can use the kernel density estimation method to fit the probability distribution of the prediction error of new energy power plants, avoiding the deviation that may occur due to the need to pre-set the distribution type in parameter fitting, thereby making the error distribution description more in line with the data characteristics. At the same time, the probability of the error interval can be used as a weighting coefficient to calculate the average error by weighted summation of the daily power output error of new energy power plants, effectively reflecting the statistical characteristics of different errors, accurately obtaining the average error of power plant output prediction, and improving the reliability of the assessment.

[0010] Optionally, in one embodiment of this application, the establishment of the uncertainty assessment system for the output of new energy power plants under the intermittent, stochastic, and volatile levels includes: collecting and preprocessing data on the prediction error distribution and the time series characteristics of the new energy output to obtain processed data; proposing new energy uncertainty assessment indicators under the intermittent, stochastic, and volatile levels based on the processed data; and establishing the uncertainty assessment system for the output of new energy power plants based on the new energy uncertainty assessment indicators.

[0011] Through the above technical means, the embodiments of this application can ensure the accuracy of the evaluation data and avoid the interference of noise in the original data by collecting and preprocessing data. Then, evaluation indicators are proposed from three levels, breaking through the limitations of single-angle evaluation in related technologies, realizing multi-level coverage of the uncertainty of new energy power output, thereby establishing an evaluation system for the uncertainty of new energy power plant output, providing a structured framework for comprehensively and systematically quantifying the uncertainty of new energy power plant output, and improving the reliability of the evaluation.

[0012] Optionally, in one embodiment of this application, the establishment of a four-layer structure model for assessing the uncertainty of new energy power plant output, in order to calculate the uncertainty index of new energy under different typical scenarios, includes: establishing a four-level hierarchical structure model containing a target layer, a first-level indicator layer, a second-level indicator layer, and a scheme layer, to propose multiple evaluation indicators for the uncertainty of new energy power plant output; based on multiple evaluation indicators, using a preset scaling strategy to compare the indicators pairwise, completing the hierarchical single ranking and multi-level ranking, determining the final calculation method of the comprehensive index, so as to obtain the new energy uncertainty index.

[0013] Through the above technical means, the embodiments of this application can establish a four-level hierarchical structure model, avoiding the one-sidedness of evaluation caused by single or scattered indicators. By pre-setting scaling strategies and hierarchical ordering, the importance of indicators and comprehensive calculation methods can be determined, ensuring the rationality of the calculation of uncertainty indicators, thereby obtaining uncertainty indicators under different typical scenarios and improving the accuracy and reliability of the evaluation.

[0014] Optionally, in one embodiment of this application, the step of using the new energy uncertainty index to establish a model training dataset to construct a neural network-based intelligent assessment model for new energy output uncertainty includes: using the secondary index as the input layer and uncertainty as the output layer to build a neural network model; calculating the new energy uncertainty based on historical output data to form a model training dataset, and training the neural network model to construct the neural network-based intelligent assessment model for new energy output uncertainty.

[0015] Through the above technical means, the embodiments of this application can construct a dataset and train a model based on historical data, so that the model can adapt to the output characteristics of different new energy power plants and has a stronger generalization ability. At the same time, the trained intelligent model can quickly output uncertain results by inputting secondary indicators, which greatly improves the evaluation efficiency and provides efficient intelligent support for real-time scheduling and risk prediction of the power system.

[0016] Optionally, in one embodiment of this application, the expression for calculating the new energy uncertainty index may be, but is not limited to, the following: , in, i Index symbols representing new energy power stations This indicates the uncertainty index of the new energy source. n Indicates the total number of indicators. k Indicates the index symbol of the indicator. Indicates the first k The weighting coefficients of each indicator Indicates the first i Feature vectors of the scheme layer of a new energy power station m This indicates the total number of new energy power stations.

[0017] Through the above technical means, the embodiments of this application can reflect the differences in importance of indicators at different levels through the weight coefficients of the indicators. By combining the calculation with the feature vector of the scheme layer, the evaluation deviation caused by simple superposition of indicators or ignoring the scene characteristics is avoided, making the uncertain indicators more in line with the actual working conditions.

[0018] The second aspect of this application provides an intelligent assessment device for multidimensional uncertainty of new energy considering prediction errors, comprising: a fitting module for fitting the probability distribution of prediction errors of new energy power plants using kernel density estimation to calculate the average prediction error of the power plant output; an establishment module for establishing an assessment system for the uncertainty of new energy power plant output under intermittent, random, and fluctuating levels based on the prediction error distribution corresponding to the average prediction error of the power plant output and the time series characteristics of new energy output; an extraction module for dividing the sample set into multiple scenario sets based on the characteristics of new energy output and extracting at least one typical scenario under different scenario sets; a calculation module for establishing a four-layer structure model for assessing the uncertainty of new energy power plant output based on the assessment system and the at least one typical scenario to calculate the new energy uncertainty index under different typical scenarios; and an assessment module for establishing a model training dataset using the new energy uncertainty index to construct an intelligent assessment model for the uncertainty of new energy output based on a neural network.

[0019] Through the above technical means, the embodiments of this application can accurately quantify the power plant output prediction error through kernel density estimation, laying a reliable data foundation for assessment. Then, a three-level uncertainty assessment system is constructed to achieve comprehensive coverage of the uncertainty of new energy power output, avoiding the limitations of a single dimension. The system also divides the scenario set to extract typical scenarios, allowing the assessment to focus on different operating conditions and improve the assessment's relevance. Subsequently, a four-layer structural model is established to systematically calculate scenario-based uncertainty indicators, making the assessment more structured and accurate. Based on the indicators, a neural network intelligent assessment model is constructed to achieve efficient and intelligent assessment of the uncertainty of new energy power output. Overall, this provides scientific and accurate decision support for power system dispatch optimization, risk prevention and control, and power plant operation planning.

[0020] Optionally, in one embodiment of this application, the fitting module includes: fitting the probability distribution of the new energy power output prediction error using a kernel density estimation strategy in a preset nonparametric fitting strategy; and based on the probability distribution, using the probability of new energy power output in each error interval as a weighting coefficient to perform a weighted summation of the new energy power output error as the average error of the power station power output prediction.

[0021] Through the above technical means, the embodiments of this application can use the kernel density estimation method to fit the probability distribution of the prediction error of new energy power plants, avoiding the deviation that may occur due to the need to pre-set the distribution type in parameter fitting, thereby making the error distribution description more in line with the data characteristics. At the same time, the probability of the error interval can be used as a weighting coefficient to calculate the average error by weighted summation of the daily power output error of new energy power plants, effectively reflecting the statistical characteristics of different errors, accurately obtaining the average error of power plant output prediction, and improving the reliability of the assessment.

[0022] Optionally, in one embodiment of this application, the establishment module includes: collecting and preprocessing data on the prediction error distribution and the time-series characteristics of the new energy output to obtain processed data; proposing new energy uncertainty assessment indicators at the intermittent, random, and volatile levels based on the processed data; and establishing the new energy power station output uncertainty assessment system based on the new energy uncertainty assessment indicators.

[0023] Through the above technical means, the embodiments of this application can ensure the accuracy of the evaluation data and avoid the interference of noise in the original data by collecting and preprocessing data. Then, evaluation indicators are proposed from three levels, breaking through the limitations of single-angle evaluation in related technologies, realizing multi-level coverage of the uncertainty of new energy power output, thereby establishing an evaluation system for the uncertainty of new energy power plant output, providing a structured framework for comprehensively and systematically quantifying the uncertainty of new energy power plant output, and improving the reliability of the evaluation.

[0024] Optionally, in one embodiment of this application, the calculation module includes: establishing a four-level hierarchical structure model comprising a target layer, a first-level indicator layer, a second-level indicator layer, and a scheme layer to propose multiple evaluation indicators for the uncertainty of power output of new energy power plants; based on multiple evaluation indicators, using a preset scaling strategy to compare the indicators pairwise, completing hierarchical single-ranking and multi-level ranking, determining the final calculation method of the comprehensive indicator, so as to obtain the new energy uncertainty indicator.

[0025] Through the above technical means, the embodiments of this application can establish a four-level hierarchical structure model, avoiding the one-sidedness of evaluation caused by single or scattered indicators. By pre-setting scaling strategies and hierarchical ordering, the importance of indicators and comprehensive calculation methods can be determined, ensuring the rationality of the calculation of uncertainty indicators, thereby obtaining uncertainty indicators under different typical scenarios and improving the accuracy and reliability of the evaluation.

[0026] Optionally, in one embodiment of this application, the evaluation module includes: building a neural network model by using the secondary index as the input layer and uncertainty as the output layer; calculating the uncertainty of new energy sources based on historical power output data to form a model training dataset; and training the neural network model to construct the intelligent evaluation model for uncertainty of new energy output based on the neural network.

[0027] Through the above technical means, the embodiments of this application can construct a dataset and train a model based on historical data, so that the model can adapt to the output characteristics of different new energy power plants and has a stronger generalization ability. At the same time, the trained intelligent model can quickly output uncertain results by inputting secondary indicators, which greatly improves the evaluation efficiency and provides efficient intelligent support for real-time scheduling and risk prediction of the power system.

[0028] Optionally, in one embodiment of this application, the expression for calculating the new energy uncertainty index may be, but is not limited to, the following: , in, i Index symbols representing new energy power stations This indicates the uncertainty index of the new energy source. n Indicates the total number of indicators. k Indicates the index symbol of the indicator. Indicates the first k The weighting coefficients of each indicator Indicates the first i Feature vectors of the scheme layer of a new energy power station m This indicates the total number of new energy power stations.

[0029] Through the above technical means, the embodiments of this application can reflect the differences in importance of indicators at different levels through the weight coefficients of the indicators. By combining the calculation with the feature vector of the scheme layer, the evaluation deviation caused by simple superposition of indicators or ignoring the scene characteristics is avoided, making the uncertain indicators more in line with the actual working conditions.

[0030] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the program to implement the intelligent assessment method for multidimensional uncertainty of new energy considering prediction errors as described in the above embodiments.

[0031] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described intelligent assessment method for multidimensional uncertainty of new energy sources, taking into account prediction errors.

[0032] The fifth aspect of this application provides a computer program product, including a computer program that, when executed, implements the above-described intelligent assessment method for multidimensional uncertainty of new energy sources that takes into account prediction errors.

[0033] This application's embodiments can use kernel density estimation to fit the probability distribution of new energy power plant prediction errors to calculate the average error of power plant output prediction. Then, it establishes an assessment system for the uncertainty of new energy power plant output under intermittent, stochastic, and fluctuating levels. The sample set is divided into multiple scenario sets, and typical scenarios under different scenario sets are extracted. A four-layer structural model for assessing the uncertainty of new energy power plant output is then established to calculate new energy uncertainty indicators under different typical scenarios. This constructs a neural network-based intelligent assessment model for new energy output uncertainty, achieving comprehensive coverage of new energy output uncertainty, avoiding the limitations of a single dimension. The system calculates scenario-based uncertainty indicators, making the assessment more structured and accurate, achieving efficient and intelligent assessment of new energy output uncertainty. Overall, it provides scientific and accurate decision support for power system dispatch optimization, risk prevention and control, and power plant operation planning. This solves the problem that related technologies often assess the uncertainty of new energy power plant output from a single perspective, failing to consider multiple operating scenarios, resulting in difficulties in comprehensively quantifying the uncertainty of new energy power plant output and accurately assessing the short-term dispatch uncertainty of the power system.

[0034] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0035] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of a smart evaluation method for multidimensional uncertainty of new energy considering prediction error, provided according to an embodiment of this application. Figure 2 This is a schematic diagram of an assessment system for the uncertainty of power output of new energy power plants under intermittent, random, and fluctuating conditions according to an embodiment of this application; Figure 3 This is a schematic diagram of a four-layer structural model for assessing the output uncertainty of a new energy power station according to an embodiment of this application; Figure 4 This is a block diagram of a new energy multidimensional uncertainty intelligent evaluation device that takes into account prediction errors, according to an embodiment of this application. Figure 5 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0036] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0037] The following describes an intelligent assessment method for multidimensional uncertainty of new energy sources that takes into account prediction errors, based on an embodiment of this application, with reference to the accompanying drawings. To address the issues raised in the background that many related technologies assess the uncertainty of renewable energy power plant output from a single perspective, failing to consider multiple operating scenarios and thus hindering the comprehensive quantification of this uncertainty and accurate assessment of short-term power system dispatch uncertainty, this application provides a multi-dimensional intelligent assessment method for renewable energy uncertainty that considers prediction errors. This method uses kernel density estimation to fit the probability distribution of renewable energy power plant prediction errors to calculate the average error of power plant output prediction. It then establishes an assessment system for renewable energy power plant output uncertainty at the intermittent, stochastic, and fluctuating levels. Furthermore, it divides the sample set into multiple scenario sets and extracts typical scenarios from each set, establishing a four-layer structural model for assessing renewable energy power plant output uncertainty. This model calculates renewable energy uncertainty indicators under different typical scenarios, thereby constructing a neural network-based intelligent assessment model for renewable energy output uncertainty. This achieves comprehensive coverage of renewable energy output uncertainty, avoids the limitations of a single dimension, and systematically calculates scenario-based uncertainty indicators, making the assessment more structured and accurate. This results in efficient and intelligent assessment of renewable energy output uncertainty, providing scientific and precise decision support for power system dispatch optimization, risk prevention and control, and power plant operation planning. This solves the problem that many related technologies assess the uncertainty of power output from renewable energy power plants from a single perspective, without taking into account multiple operating scenarios, which makes it difficult to fully quantify the uncertainty of power output from renewable energy power plants and to accurately assess the uncertainty of short-term power system dispatch.

[0038] Specifically, Figure 1 This is a flowchart of a new energy multidimensional uncertainty intelligent assessment method that considers prediction error, according to an embodiment of this application.

[0039] like Figure 1 As shown, the intelligent assessment method for multidimensional uncertainty of new energy considering prediction errors includes the following steps: In step S101, the probability distribution of the prediction error of the new energy power station is fitted using the kernel density estimation method to calculate the average error of the power station output prediction.

[0040] In the embodiments of this application, the prediction error of a new energy power station can be understood as the deviation between the output-related parameters (such as power generation and power output) pre-calculated by the new energy power station (such as wind farm and photovoltaic power station) when carrying out output prediction work and the actual output parameters obtained by monitoring during the actual operation of the station.

[0041] Furthermore, kernel density estimation can be understood as a method that does not rely on prior knowledge of the data distribution, nor does it impose any pre-defined assumptions on the distribution pattern of the data. It estimates and fits the actual distribution pattern of the data by utilizing the characteristics of the data sample itself and constructing a probability density function based on the kernel function. Therefore, kernel density estimation can effectively adapt to the characteristic that the error in the prediction of new energy output has no fixed distribution pattern due to the influence of factors such as meteorology.

[0042] In addition, the average error of power plant output prediction can be understood as an indicator to measure the accuracy of the power plant output prediction results. The smaller the value of the average error of power plant output prediction, the smaller the overall deviation between the prediction results and the actual output.

[0043] In actual implementation, the embodiments of this application can use the kernel density estimation method to fit the probability distribution of the prediction error of new energy power plants in order to calculate the average error of the power plant output prediction. The calculation process of the average error of the power plant output prediction of new energy power plants is described in detail below.

[0044] Specifically, in one embodiment of this application, the probability distribution of the prediction error of the new energy power station is fitted using the kernel density estimation method to calculate the average error of the power station output prediction. This includes: fitting the probability distribution of the new energy power output prediction error using the kernel density estimation strategy in the preset nonparametric fitting strategy; and based on the probability distribution, using the probability of the new energy power output in each error interval as a weighting coefficient to perform a weighted summation of the new energy power output daily error as the average error of the power station output prediction.

[0045] In actual implementation, the embodiments of this application can adopt the kernel density estimation method in the non-parametric fitting strategy. There is no need to preset a fixed probability distribution type. Based on the data characteristics of the prediction error of the new energy power station, a probability distribution that can truly reflect the error distribution law can be fitted. Then, based on the fitted probability distribution, the occurrence probability corresponding to each error interval can be extracted and used as a weighting coefficient to perform a weighted summation operation on the daily power output error of the new energy power station to obtain the average error of the power station power output prediction.

[0046] In some cases, embodiments of this application can collect historical actual power generation from renewable energy power plants. PW and historical predictions PW 0, to calculate the prediction error. The expression for the prediction error calculation formula can be, but is not limited to, as follows: (1) Furthermore, the embodiments of this application address the prediction error. e probability density function For kernel density estimation fitting, the expression of the fitting formula can be, but is not limited to, as follows: (2) in, h For bandwidth, e i For error sample points, K (·) represents the kernel density function. n It represents the total number of sample points.

[0047] For the fitted probability density function described above, the probability distribution function can be obtained by integrating it in this embodiment. Furthermore, this embodiment can arrange the prediction errors from smallest to largest, and then select the minimum value... a Maximum value b Divide into equal parts N There are intervals, each denoted as ( ). a i , b i ),error e Falling in i The probability of each interval P i The expression can be, but is not limited to, as: (3) Furthermore, in embodiments of this application, the average value of the error within the interval can be used to represent the error of the entire interval, and the errors are weighted and summed using probability values ​​to calculate the average error. The expression for the average error calculation formula can be, but is not limited to, as follows: (4) In summary, the embodiments of this application can use the kernel density estimation method to fit the probability distribution of the prediction error of new energy power plants, avoiding the deviation that may occur due to the need to pre-set the distribution type in parameter fitting. This makes the error distribution description more closely match the data characteristics. At the same time, the probability of the error interval can be used as a weighting coefficient to calculate the average error by weighted summation of the daily power output error of new energy power plants. This effectively reflects the statistical characteristics of different errors, accurately obtains the average error of power plant output prediction, and improves the reliability of the assessment.

[0048] In step S102, based on the prediction error distribution corresponding to the average prediction error of the power plant output and the time series characteristics of new energy power output, an assessment system for the uncertainty of new energy power plant output under the intermittent, random, and volatile levels is established.

[0049] In the embodiments of this application, the power output time sequence characteristics of new energy can be understood as the changing patterns and characteristics of the power output data of new energy power stations in a continuous time dimension (such as hours, days, months), which can reflect the dynamic changing trend of power over time.

[0050] In addition, the intermittency dimension describes the discontinuous and periodically interrupted power output of new energy power plants in the time dimension; the randomness dimension describes the uncertain changes in the power output of new energy power plants caused by unpredictable random factors; and the volatility dimension measures the magnitude and frequency of changes in the power output of new energy power plants over a continuous period of time.

[0051] In addition, the power output uncertainty assessment system for new energy power plants is a systematic assessment framework constructed to quantify and analyze the degree of power output fluctuation caused by factors such as natural condition fluctuations and prediction errors.

[0052] In actual implementation, the embodiments of this application can establish an assessment system for the uncertainty of new energy power plant output under the intermittent, random, and volatile levels based on the prediction error distribution corresponding to the average error of power plant output prediction and the time series characteristics of new energy power output. The establishment of the assessment system for the uncertainty of new energy power plant output under the intermittent, random, and volatile levels will be described in detail below.

[0053] Specifically, in one embodiment of this application, a new energy power station output uncertainty assessment system is established under the intermittent, random, and volatile levels, including: collecting and preprocessing data on the prediction error distribution and the time series characteristics of new energy output to obtain processed data; proposing new energy uncertainty assessment indicators under the intermittent, random, and volatile levels based on the processed data; and establishing a new energy power station output uncertainty assessment system based on the new energy uncertainty assessment indicators.

[0054] In actual implementation, this application embodiment collects prediction error distribution data and output time series characteristic data of new energy power stations, and preprocesses the data to obtain standardized processed data. Then, based on the processed data, targeted new energy uncertainty assessment indicators can be designed from three levels: intermittency, randomness, and volatility, to reflect the uncertainty characteristics at each level. In this way, multi-level assessment indicators are integrated to establish a new energy power station output uncertainty assessment system to cover the key uncertainty levels of new energy power station output.

[0055] For example, such as Figure 2 As shown in the embodiments of this application, new energy uncertainty assessment indicators are proposed under the intermittent, stochastic, and volatile levels, including: (1) Intermittent level assessment indicators Because the output of new energy sources is limited by meteorological factors, resulting in intermittent power generation, there may be insufficient output and difficulty in supporting grid demand during periods of no wind or no sunlight. This application's embodiments set the probability of insufficient output as the evaluation index at the intermittency level. P Insufficient effort and expectations EThe expressions for calculating the probability of insufficient output and the expected output can be, but are not limited to, as follows: (5) (6) in, X t Indicates time t The active power of wind or solar power plants T The total number of time periods. X ref This indicates the reference power used to determine insufficient output from renewable energy power plants.

[0056] (2) Randomness level assessment indicators Since there is a deviation between the actual output of new energy and the predicted value, this application uses the prediction error probability distribution and the average error as evaluation indicators under the randomness level.

[0057] (3) Volatility assessment indicators This application uses a fluctuation coefficient to characterize the fluctuation of current renewable energy output, and calculates the standard deviation and average value of renewable energy daily output. The fluctuation coefficient is then calculated by dividing the standard deviation by the average value. Simultaneously, the fluctuation rate of renewable energy daily output is selected. flu Peak-valley difference pv As a volatility indicator, the formula for calculating volatility and peak-to-trough difference can be expressed, but is not limited to, as follows: (7) (8) In summary, the embodiments of this application can ensure the accuracy of the evaluation data and avoid the interference of noise in the raw data through data collection and preprocessing. Then, evaluation indicators are proposed from three levels, breaking through the limitations of single-angle evaluation in related technologies, realizing multi-level coverage of the uncertainty of new energy power output, thereby establishing an evaluation system for the uncertainty of new energy power plant output, providing a structured framework for comprehensively and systematically quantifying the uncertainty of new energy power plant output, and improving the reliability of the evaluation.

[0058] In step S103, based on the characteristics of new energy output, the sample set is divided into multiple scenario sets, and at least one typical scenario under each scenario set is extracted.

[0059] In the embodiments of this application, the power output characteristics of new energy sources can be understood as the power output change attributes of new energy power stations affected by natural conditions, which may include, but are not limited to, attributes such as fluctuation amplitude, duration, peak or trough periods.

[0060] In addition, the sample set is a dataset formed by organizing specific data containing power output characteristics according to a unified dimension; the scenario set is a combination of multiple specific power output scenarios that can cover different power output states (such as normal power output, large fluctuations, and extreme power output) after the sample set is processed by data cleaning, clustering, or random simulation; the typical scenario is a scenario that is further screened or refined from the scenario set, which can represent the main modes of new energy power output (such as peak power output, low-end power output, and stable power output) and has statistical significance, which can simplify the analysis of the power output pattern of new energy.

[0061] In actual implementation, this application embodiment uses the key characteristics of new energy output as the classification basis, divides the sample set into multiple scene sets according to feature similarity, and then extracts typical scenes that can represent the output pattern of the set from the scene set, so as to realize the classification and representative extraction of complex samples.

[0062] In some cases, embodiments of this application may select one year of renewable energy output data, unify its time granularity to 1 hour, perform data preprocessing to obtain data for 365 output scenarios, and use this as the total scenario set with dimensions (365, 24). Then, the maximum daily power fluctuation of renewable energy output under each scenario can be calculated. R The expression for the maximum power fluctuation calculation formula can be, but is not limited to, as follows: (9) In addition, the embodiments of this application can be based on the daily maximum power fluctuation. R The total scene set is divided into typical scene set, fluctuating scene set, and extreme scene set based on the value. The specific division rule is as follows: calculate the value for each of the 365 data points. R ,exist R When the installed capacity exceeds 20% of the local renewable energy power station capacity, it is classified as an extreme scenario set; R When the installed capacity of new energy power plants is between 10% and 20%, it is classified as a fluctuating scenario set; otherwise, it is classified as a typical scenario set.

[0063] Furthermore, embodiments of this application can use the Load curve-ISODATA clustering algorithm (Load Curve-Iterative Self-Organizing Data Analysis Techniques Algorithm Clustering Algorithm) to extract typical scenarios from different scenario sets. It should be noted that in related technologies, the k-means clustering algorithm requires the number of clusters to be determined in advance. K Value, and optimal K The selection of values ​​often involves multiple trials and empirical judgments to arrive at the optimal value. KValue calculation is complex and difficult to determine accurately in advance. The ISODATA algorithm (Iterative Self-organizing Data Analysis Techniques Algorithm) addresses this by setting variable values ​​during the clustering process. K This addresses the aforementioned issues. The Load curve-ISODATA clustering algorithm is an improvement upon the ISODATA algorithm. It achieves efficient convergence by optimizing the selection of initial cluster centers and introducing kernel functions to learn the high-dimensional features of the data.

[0064] This application's embodiments reduce the analytical complexity of uncertainties in new energy output by classifying samples and extracting typical scenarios, and align with actual output characteristics, simplifying data processing costs, thereby reducing the complexity of the assessment while ensuring its reliability.

[0065] In step S104, based on the uncertainty assessment system for the output of new energy power plants and at least one typical scenario, a four-layer structural model for assessing the uncertainty of the output of new energy power plants is established to calculate the uncertainty index of new energy under different typical scenarios.

[0066] In the embodiments of this application, the new energy uncertainty index under different typical scenarios can be understood as a quantitative index designed to match the characteristics of each scenario for different typical modes of new energy output, and used to accurately measure the degree of uncertainty of new energy output under typical scenarios.

[0067] In actual implementation, this application embodiment can establish a four-layer structure model for assessing the uncertainty of new energy power plant output based on the new energy power plant output uncertainty assessment system and at least one typical scenario. Then, the analytic hierarchy process can be used to calculate the new energy uncertainty index under different typical scenarios. The following is a detailed introduction to the process of establishing the four-layer structure model for assessing the uncertainty of new energy power plant output and calculating the new energy uncertainty index under different typical scenarios.

[0068] Specifically, in one embodiment of this application, a four-layer structure model for assessing the uncertainty of power output of new energy power plants is established to calculate the uncertainty index of new energy under different typical scenarios. This includes: establishing a four-level hierarchical structure model containing a target layer, a first-level indicator layer, a second-level indicator layer, and a scheme layer to propose multiple evaluation indicators for the uncertainty of power output of new energy power plants; based on multiple evaluation indicators, using a preset scaling strategy to compare the indicators pairwise, completing the hierarchical single ranking and multi-level ranking, determining the final calculation method of the comprehensive indicator, and obtaining the uncertainty index of new energy.

[0069] Understandably, the Analytic Hierarchy Process (AHP) treats a complex multi-objective decision problem as a system, decomposing it into multiple indicators at multiple levels. It calculates the ranking of each level through qualitative indicator fuzzy quantification, reducing the problem to solving for the weight or superiority / inferiority of the lowest level relative to the highest level. It is suitable for decision problems with hierarchical and intersecting evaluation indicators, where the indicator values ​​are difficult to describe quantitatively.

[0070] In some cases, embodiments of this application can establish a hierarchical structure for indicator evaluation, with the lowest level representing alternative solutions, the highest level representing the overall objective, and intermediate levels representing indicator layers. This allows for flexible selection based on different application scenarios. Subsequently, indicators within each level can be compared pairwise using a 1-9 scale. Indicators i relative to indicators j To determine the importance of a matrix, construct a judgment matrix. This allows us to calculate the largest eigenvalue and corresponding eigenvector of the judgment matrix, and then perform a consistency check on it. If the consistency check passes, the constructed matrix can be considered valid. H If the error is small, the normalized feature vector becomes the weight coefficient of all indicators in this level of indicator layer; otherwise, the judgment matrix is ​​modified until the consistency test is passed. In the embodiments of this application, the expression of the consistency index used for consistency testing may be, but is not limited to, as follows: (10) in, To determine the largest eigenvalue of a matrix, N To determine the order of a matrix, This represents the average consistency index, which is a definite constant. It should be noted that... When the matrix passes the consistency check, it is considered to have passed the consistency check.

[0071] Furthermore, after sorting the indicators at this level, the next level of indicators is sorted, and so on down to the lowest level of solutions. The weights of each level are multiplied together to obtain the score and ranking of each solution in the lowest level relative to the overall goal of the highest level, which can be used as the basis for solution decision-making.

[0072] It is understandable that the uncertainty of new energy output is affected by a variety of factors. Since there are many static and dynamic parameters to consider, the embodiments of this application can establish a four-layer structure model and apply the analytic hierarchy process to calculate a comprehensive index of new energy uncertainty for each index.

[0073] like Figure 3As shown, the uncertainty of new energy output is the ultimate goal of the evaluation; therefore, in this embodiment, the uncertainty of new energy output is set as target layer A. Intermittency, randomness, and volatility are the main factors considered in the evaluation; therefore, in this embodiment, intermittency, randomness, and volatility are set as the first-level indicator layer C1. For intermittency, the second-level indicators can consider the probability of insufficient output and the expected insufficient output; for randomness, the second-level indicators can consider the mean of prediction errors and the probability distribution of prediction errors; for volatility, the second-level indicators can consider the fluctuation coefficient, power fluctuation rate, and power peak-to-valley difference. These second-level indicators are uniformly listed as the second-level indicator layer C2. Assuming the total number of indicators in layer C2 is... n , No. k Individual markers are The scheme layer represents all renewable energy power stations whose uncertainties need to be evaluated, denoted as layer P. Assume the total number of renewable energy power stations is [number missing]. m , No. i Each station is recorded as .

[0074] In some cases, after dividing the scenario set, embodiments of this application can determine the importance of the three C1 indicators in each typical scenario, and perform single-level ranking accordingly to obtain the eigenvector corresponding to the largest eigenvalue of the judgment matrix, denoted as . Furthermore, in this embodiment, the process can be moved down to the secondary index layer and repeated to obtain the feature vector of the secondary index layer C2 relative to C1. Subsequently, in this embodiment of the application, the weight coefficients of layer C1 and layer C2 can be multiplied accordingly to obtain a comprehensive weight vector of all indicators. w 0, where each indicator The weighting coefficient is w 0k ( k =1,2,..., n Furthermore, in this embodiment, the process is moved down to the solution layer and a single-layer sorting of the solutions is performed, targeting the indicators. The feature vector of the scheme layer is Then, a multi-level sort is performed to obtain the first... i A new energy power station New energy uncertainty indicators.

[0075] Optionally, in one embodiment of this application, the expression for calculating the new energy uncertainty index may be, but is not limited to, the following: , in, i Index symbols representing new energy power stations Indicators representing uncertainty in new energy sources n Indicates the total number of indicators.k Indicates the index symbol of the indicator. Indicates the first k The weighting coefficients of each indicator Indicates the first i Feature vectors of the scheme layer of a new energy power station m This indicates the total number of new energy power stations.

[0076] In actual implementation, the embodiments of this application can calculate the new energy uncertainty index through the weight coefficient of the index and the feature vector of the scheme layer.

[0077] The embodiments of this application can reflect the differences in importance of indicators at different levels through the weight coefficients of the indicators. By combining the calculation with the feature vector of the scheme layer, the evaluation deviation caused by simple superposition of indicators or ignoring the scene characteristics is avoided, making the uncertain indicators more in line with the actual working conditions.

[0078] In summary, the embodiments of this application can establish a four-level hierarchical structure model, avoiding the one-sidedness of evaluation caused by single or scattered indicators. By pre-setting scaling strategies and hierarchical ordering, the importance of indicators and comprehensive calculation methods can be determined, ensuring the rationality of the calculation of uncertain indicators. Thus, the uncertainty indicators obtained under different typical scenarios can improve the accuracy and reliability of the evaluation.

[0079] In step S105, a model training dataset is established using new energy uncertainty indicators to construct a neural network-based intelligent assessment model for new energy output uncertainty.

[0080] In the embodiments of this application, the model training dataset can be understood as a structured dataset that allows the model to learn the inherent laws of the data. It is used to provide learning samples for the model, allowing the model to continuously adjust its internal parameters (such as weights and thresholds) through iterative calculations, and gradually master the correlation between input and output.

[0081] In addition, the intelligent assessment model for uncertainty of renewable energy output based on neural networks can be understood as an intelligent assessment tool that integrates neural network technology to quantify the uncertainty of renewable energy power plant output.

[0082] In actual implementation, the embodiments of this application can use new energy uncertainty indicators to establish a model training dataset in order to construct a neural network-based intelligent assessment model for new energy output uncertainty. The process of constructing a neural network-based intelligent assessment model for new energy output uncertainty is described in detail below.

[0083] Specifically, in one embodiment of this application, a model training dataset is established using new energy uncertainty indicators to construct a neural network-based intelligent assessment model for new energy output uncertainty. This includes: using secondary indicators as the input layer and uncertainty as the output layer to build a neural network model; calculating new energy uncertainty based on historical output data to form a model training dataset; and training the neural network model to construct a neural network-based intelligent assessment model for new energy output uncertainty.

[0084] In actual implementation, this application embodiment sets the secondary indicators that reflect the intermittency, randomness, and volatility of new energy output as the model input layer, and sets the new energy uncertainty indicators to be evaluated as the output layer, thus building a neural network model that directly links the input and output. Then, based on historical output data, the secondary indicators and uncertainty indicators for the corresponding historical periods are calculated to form a model training dataset with paired input and output. This dataset is then used to iteratively train the built neural network model, allowing the model to autonomously learn the mapping relationship between the secondary indicators and the uncertainty of new energy, so as to construct a neural network-based intelligent assessment model for the uncertainty of new energy output.

[0085] In summary, the embodiments of this application can construct datasets and train models based on historical data, enabling the models to adapt to the output characteristics of different new energy power plants and have stronger generalization capabilities. At the same time, the trained intelligent models can quickly output uncertain results by inputting secondary indicators, greatly improving evaluation efficiency and providing efficient intelligent support for real-time scheduling and risk prediction of power systems.

[0086] The following section describes the intelligent assessment method for multidimensional uncertainty of new energy that considers prediction error, as proposed in this application, using a specific embodiment.

[0087] This application embodiment can select the output data and meteorological measurement results of a certain new energy power station on a certain day, calculate the average error of the power station output prediction, and then establish an uncertainty assessment system for the power station output under the intermittent, random, and fluctuating levels based on the prediction error distribution corresponding to the average error of the power station output prediction and the time series characteristics of new energy output. At the same time, the sample set is divided into multiple scenario sets according to the power station output characteristics, and the Load curve-ISODATA clustering algorithm is used to extract typical scenarios under different scenario sets. Then, the analytic hierarchy process is used to calculate the uncertainty index of the power station on that day, which is used as a training data.

[0088] Subsequently, in this embodiment, multiple stations and multi-day data can be selected to repeat the above calculations, constructing a training dataset for the BP (BackPropagation) neural network. Further, in this embodiment, secondary indicator values ​​can be used as the input layer and uncertainty as the output layer. Based on this, the structural design and parameter selection of the BP neural network are completed, a BP neural network model is built, and 10% of the dataset is selected as test samples to train the BP neural network, thereby obtaining a neural network-based intelligent assessment model for the uncertainty of new energy output.

[0089] The intelligent assessment method for multidimensional uncertainty of new energy sources, which considers prediction errors, proposed in this application, can fit the probability distribution of prediction errors of new energy power plants using kernel density estimation to calculate the average error of power plant output prediction. This allows for the establishment of an assessment system for the uncertainty of new energy power plant output at the intermittent, stochastic, and fluctuating levels. The sample set is divided into multiple scenario sets, and typical scenarios under different scenario sets are extracted. A four-layer structural model for assessing the uncertainty of new energy power plant output is then established to calculate new energy uncertainty indicators under different typical scenarios. This constructs a neural network-based intelligent assessment model for the uncertainty of new energy power plant output, achieving comprehensive coverage of new energy output uncertainty, avoiding the limitations of a single dimension, and systematically calculating scenario-based uncertainty indicators. This makes the assessment more structured and accurate, achieving efficient and intelligent assessment of new energy output uncertainty. Overall, this provides scientific and accurate decision support for power system dispatch optimization, risk prevention and control, and power plant operation planning. Therefore, this solves the problem that related technologies often assess the uncertainty of new energy power plant output from a single perspective, failing to consider multiple operating scenarios, resulting in difficulties in comprehensively quantifying the uncertainty of new energy power plant output and accurately assessing the short-term dispatch uncertainty of the power system.

[0090] Next, referring to the accompanying drawings, we describe the intelligent assessment device for multidimensional uncertainty of new energy sources that takes into account prediction errors, according to an embodiment of this application.

[0091] Figure 4 This is a block diagram of a new energy multidimensional uncertainty intelligent evaluation device that takes into account prediction errors, according to an embodiment of this application.

[0092] like Figure 4 As shown, the intelligent assessment device 40 for multidimensional uncertainty of new energy considering prediction error includes: a fitting module 100, an establishment module 200, an extraction module 300, a calculation module 400, and an assessment module 500.

[0093] Among them, the fitting module 100 is used to perform kernel density estimation probability distribution fitting on the prediction error of new energy power stations in order to calculate the average error of power station output prediction.

[0094] Module 200 is established to build an assessment system for the uncertainty of new energy power plant output at the intermittent, stochastic, and volatile levels, based on the prediction error distribution corresponding to the average error of power plant output prediction and the time series characteristics of new energy power output.

[0095] The extraction module 300 is used to divide the sample set into multiple scenario sets based on the characteristics of new energy output, and extract at least one typical scenario under each scenario set.

[0096] The calculation module 400 is used to establish a four-layer structure model for assessing the uncertainty of new energy power plant output based on the uncertainty assessment system of new energy power plant output and at least one typical scenario, so as to calculate the new energy uncertainty index under different typical scenarios.

[0097] The evaluation module 500 is used to establish a model training dataset using new energy uncertainty indicators in order to construct an intelligent evaluation model for new energy output uncertainty based on neural networks.

[0098] Optionally, in one embodiment of this application, the fitting module 100 includes a fitting unit and a calculation unit.

[0099] The fitting unit is used to fit the probability distribution of the new energy output prediction error using the kernel density estimation strategy in the preset nonparametric fitting strategy.

[0100] The calculation unit is used to use the probability distribution of new energy power output in each error interval as a weighting coefficient to perform a weighted summation of the daily power output error of new energy sources, which is then used as the average error of the power station power output prediction.

[0101] Optionally, in one embodiment of this application, the establishment module 200 includes: a processing unit, a first generation unit, and an establishment unit.

[0102] The processing unit is used to collect and preprocess data on the prediction error distribution and the time-series characteristics of new energy output to obtain processed data.

[0103] The first generation unit is used to propose new energy uncertainty assessment indicators at the intermittent, stochastic, and volatile levels based on the processed data.

[0104] Establish a unit to build an assessment system for the output uncertainty of new energy power plants based on new energy uncertainty assessment indicators.

[0105] Optionally, in one embodiment of this application, the calculation module 400 includes a second generation unit and a third generation unit.

[0106] The second generation unit is used to establish a four-level hierarchical model that includes a target layer, a first-level indicator layer, a second-level indicator layer, and a scheme layer, in order to propose multiple evaluation indicators for the uncertainty of power output of new energy power plants.

[0107] The third generation unit is used to compare the indicators pairwise based on multiple evaluation indicators and using a preset scaling strategy to complete the hierarchical single ranking and multi-level ranking, and determine the final calculation method of the comprehensive indicator in order to obtain the uncertainty indicator of new energy.

[0108] Optionally, in one embodiment of this application, the evaluation module 500 includes: a construction unit and an evaluation unit.

[0109] The building unit is used to build a neural network model by taking secondary indicators as input and uncertainty as output.

[0110] The evaluation unit is used to calculate the uncertainty of new energy sources based on historical power output data, to form a model training dataset, and to train the neural network model to build an intelligent evaluation model for the uncertainty of new energy power output based on neural networks.

[0111] Optionally, in one embodiment of this application, the expression for calculating the new energy uncertainty index may be, but is not limited to, the following: , in, i Index symbols representing new energy power stations Indicators representing uncertainty in new energy sources n Indicates the total number of indicators. k Indicates the index symbol of the indicator. Indicates the first k The weighting coefficients of each indicator Indicates the first i Feature vectors of the scheme layer of a new energy power station m This indicates the total number of new energy power stations.

[0112] It should be noted that the foregoing explanation of the intelligent assessment method for multidimensional uncertainty of new energy considering prediction error also applies to the intelligent assessment device for multidimensional uncertainty of new energy considering prediction error in this embodiment, and will not be repeated here.

[0113] The intelligent assessment device for multidimensional uncertainty of new energy power plants, which considers prediction errors according to the embodiments of this application, can perform kernel density estimation probability distribution fitting on the prediction errors of new energy power plants to calculate the average error of power plant output prediction. Then, it establishes an assessment system for the uncertainty of new energy power plant output under intermittent, stochastic, and fluctuating levels. The sample set is divided into multiple scenario sets, and typical scenarios under different scenario sets are extracted. A four-layer structural model for assessing the uncertainty of new energy power plant output is then established to calculate the uncertainty index of new energy under different typical scenarios. This constructs a neural network-based intelligent assessment model for the uncertainty of new energy power plant output, achieving comprehensive coverage of the uncertainty of new energy power plant output, avoiding the limitations of a single dimension. The system calculates scenario-based uncertainty indicators, making the assessment more structured and accurate, achieving efficient and intelligent assessment of the uncertainty of new energy power plant output. Overall, it provides scientific and accurate decision support for power system dispatch optimization, risk prevention and control, and power plant operation planning. This solves the problem that related technologies often assess the uncertainty of new energy power plant output from a single perspective, failing to consider multiple operating scenarios, resulting in the difficulty in comprehensively quantifying the uncertainty of new energy power plant output and accurately assessing the short-term dispatch uncertainty of the power system.

[0114] Figure 5 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. The electronic device may include: The memory 501, the processor 502, and the computer program stored on the memory 501 and capable of running on the processor 502.

[0115] When the processor 502 executes the program, it implements the intelligent evaluation method for multidimensional uncertainty of new energy that takes into account prediction errors, as provided in the above embodiments.

[0116] Furthermore, electronic devices also include: Communication interface 503 is used for communication between memory 501 and processor 502.

[0117] The memory 501 is used to store computer programs that can run on the processor 502.

[0118] Memory 501 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0119] If the memory 501, processor 502, and communication interface 503 are implemented independently, then the communication interface 503, memory 501, and processor 502 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 5 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0120] Optionally, in a specific implementation, if the memory 501, processor 502, and communication interface 503 are integrated on a single chip, then the memory 501, processor 502, and communication interface 503 can communicate with each other through an internal interface.

[0121] Processor 502 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0122] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-mentioned intelligent assessment method for multidimensional uncertainty of new energy considering prediction errors.

[0123] This application also provides a computer program product, including a computer program that, when executed, implements the above-mentioned intelligent assessment method for multidimensional uncertainty of new energy considering prediction errors.

[0124] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0125] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0126] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0127] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0128] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or more of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0129] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it includes one or a combination of the steps of the method embodiments.

[0130] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0131] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A smart assessment method for multidimensional uncertainty of new energy sources considering prediction errors, characterized in that, Includes the following steps: The kernel density estimation method is used to fit the probability distribution of the prediction error of the new energy power station in order to calculate the average error of the power station output prediction. Based on the prediction error distribution corresponding to the average prediction error of the power plant output and the time series characteristics of new energy power output, an assessment system for the uncertainty of new energy power plant output under the intermittent, random, and fluctuating levels is established. Based on the characteristics of new energy output, the sample set is divided into multiple scenario sets, and at least one typical scenario under each scenario set is extracted. Based on the aforementioned uncertainty assessment system for new energy power plant output and the aforementioned at least one typical scenario, a four-layer structural model for uncertainty assessment of new energy power plant output is established to calculate new energy uncertainty indicators under different typical scenarios. The aforementioned new energy uncertainty index is used to establish a model training dataset in order to construct a neural network-based intelligent assessment model for new energy output uncertainty.

2. The method according to claim 1, characterized in that, The step of performing kernel density estimation probability distribution fitting on the prediction error of the new energy power station to calculate the average error of the power station output prediction includes: The kernel density estimation strategy in the preset nonparametric fitting strategy is used to fit the probability distribution of the prediction error of new energy power output; Based on the probability distribution, the probability of new energy output falling within each error interval is used as a weighting coefficient to perform a weighted summation of the new energy daily output error, which is then used as the average error of the power station output prediction.

3. The method according to claim 1, characterized in that, The establishment of an assessment system for the uncertainty of power output from renewable energy power plants at the intermittent, stochastic, and volatile levels includes: Data collection and preprocessing are performed on the prediction error distribution and the time series characteristics of the new energy output to obtain processed data; Based on the processed data, new energy uncertainty assessment indicators are proposed at the intermittent, stochastic, and volatile levels. Based on the aforementioned new energy uncertainty assessment indicators, an assessment system for the output uncertainty of new energy power plants is established.

4. The method according to claim 1, characterized in that, The four-layer structural model for assessing the uncertainty of power output from renewable energy power plants is established to calculate renewable energy uncertainty indicators under different typical scenarios, including: A four-level hierarchical model comprising a target layer, a primary indicator layer, a secondary indicator layer, and a scheme layer is established to propose multiple evaluation indicators for the uncertainty of power output of new energy power plants. Based on multiple evaluation indicators, a pre-defined scaling strategy is used to compare the indicators pairwise, complete hierarchical single ranking and multi-level ranking, determine the final calculation method of the comprehensive indicator, and obtain the uncertainty indicator of new energy.

5. The method according to claim 4, characterized in that, The step of establishing a model training dataset using the aforementioned new energy uncertainty indicators to construct a neural network-based intelligent assessment model for new energy output uncertainty includes: A neural network model is built by using the secondary indicators as the input layer and uncertainty as the output layer. The uncertainty of new energy sources is calculated based on historical power output data to form a model training dataset, and the neural network model is trained to construct the intelligent assessment model for uncertainty of new energy output based on neural networks.

6. The method according to claim 4 or the method thereof, characterized in that, The formula for calculating the uncertainty index of the new energy source is as follows: ， in, i Index symbols representing new energy power stations This indicates the uncertainty index of the new energy source. n Indicates the total number of indicators. k Indicates the index symbol of the indicator. Indicates the first k The weighting coefficients of each indicator Indicates the first i Feature vectors of the scheme layer of a new energy power station m This indicates the total number of new energy power stations.

7. A smart assessment device for multidimensional uncertainty of new energy considering prediction error, characterized in that, include: The fitting module is used to fit the probability distribution of the prediction error of the new energy power station using the kernel density estimation method, so as to calculate the average error of the power station output prediction. A module is established to build an uncertainty assessment system for new energy power plant output under intermittent, random, and fluctuating levels, based on the prediction error distribution corresponding to the average prediction error of the power plant output and the time series characteristics of new energy power output. The extraction module is used to divide the sample set into multiple scenario sets based on the characteristics of new energy output, and extract at least one typical scenario under each scenario set. The calculation module is used to establish a four-layer structure model for assessing the uncertainty of new energy power plant output based on the new energy power plant output uncertainty assessment system and the at least one typical scenario, so as to calculate the new energy uncertainty index under different typical scenarios. The evaluation module is used to establish a model training dataset using the new energy uncertainty indicators, so as to construct an intelligent evaluation model for new energy output uncertainty based on neural networks.

8. An electronic device, characterized in that, include: The memory, the processor, and the computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the intelligent assessment method for multidimensional uncertainty of new energy considering prediction errors as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the intelligent assessment method for multidimensional uncertainty of new energy sources that takes into account prediction errors, as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, The computer program is executed to implement the intelligent assessment method for multidimensional uncertainty of new energy considering prediction errors as described in any one of claims 1-6.

Images