New energy consumption capability assessment method and system fusing time sequence production simulation and random production simulation
By integrating timing and random production simulation methods, a new energy consumption capacity evaluation model is constructed and dynamically adjusted, which solves the problem of inaccurate evaluation in the existing technology, and achieves more efficient new energy utilization and power system stability.
Patent Information
- Application Number
- CN202411219420.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-02
- Publication Date
- 2025-02-07
AI Technical Summary
When evaluating the consumption capacity of new energy, existing technologies often conduct production simulations in a single dimension, and fail to fully understand the actual operation of the system, resulting in underestimating or overestimating the true consumption capacity of the evaluation results, affecting the accuracy of the evaluation.
Using a method of fusion timing production simulation and random production simulation, the fusion model is constructed based on historical and simulated data, and the results output from the fusion model are analyzed and uncertainty analysis are performed, and the model parameters are dynamically adjusted to improve the accuracy of evaluation. sex.
It improves the accuracy of the assessment of new energy consumption capacity, ensures the efficient utilization of new energy, and enhances the safe and stable operation ability of the power system.
Smart Images
Figure CN119398484A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of power evaluation, and in particular relates to a method and system for evaluating new energy consumption capacity by integrating time-series production simulation and random production simulation. Background Art
[0002] At present, the proportion of new energy access in the power grid is constantly increasing. Due to the uncertainty of new energy power generation, the phenomenon of wind and solar power abandonment is serious, and the absorption of new energy is difficult. The uncertainty of new energy power generation can easily lead to transient voltage instability, broadband oscillation and other prominent problems. Therefore, accurately evaluating the absorption capacity of new energy is of great significance to improving the absorption level of new energy in the power system and ensuring the safe and stable operation of the power system.
[0003] Existing assessments of new energy absorption capacity mostly use single-dimensional production simulations, which only considers time-series production or randomness, which may lead to an incomplete understanding of the actual operation of the system. The assessment results are likely to underestimate or overestimate the actual absorption capacity, affecting the accuracy of the assessment.
[0004] At the same time, existing assessment methods often ignore the fact that the output of new energy sources (such as wind power and photovoltaic power) changes dramatically with meteorological conditions. If the changing patterns and fluctuation characteristics of different periods are not fully considered, it will be difficult to accurately predict the response capacity of the power system. Forecast errors may lead to investment decision-making errors, such as excessive or insufficient infrastructure construction, improper allocation of peak-shaving resources, etc. System reliability and economic evaluation may be inaccurate, affecting long-term development and strategy adjustments.
[0005] Therefore, how to propose a new energy consumption capacity assessment method and system that integrates time-series production simulation and random production simulation to accurately assess the new energy consumption capacity is an urgent problem to be solved by technical personnel in this field. Summary of the invention
[0006] In order to solve the problems existing in the prior art, the present invention proposes a new energy consumption capacity evaluation method and system integrating time-series production simulation and random production simulation.
[0007] The technical solution of the present invention is as follows:
[0008] A new energy consumption capacity evaluation method integrating sequential production simulation and random production simulation includes:
[0009] Based on historical renewable energy output data and simulated renewable energy output data, build and train time series production models and stochastic production models;
[0010] The time series production results output by the time series production model and the random production results output by the time series production model are integrated to obtain a comprehensive prediction value of new energy power generation;
[0011] Performing analysis on the comprehensive forecast value of the output of the new energy power generation at different time scales to obtain forecast values of the comprehensive forecast value at multiple time scales, thereby obtaining a multi-scale analysis result, and performing uncertainty analysis on the comprehensive forecast value of the output of the new energy power generation to obtain a confidence interval and an error range of the time series production result and the random production result, thereby obtaining a forecast credibility and potential risk assessment result;
[0012] Adopting a reinforcement learning algorithm to dynamically adjust the parameters of the time series production model and the random production model, and iterating the time series production model and the random production model after the dynamic parameter adjustment to obtain new multi-scale analysis results and new prediction credibility and potential risk assessment results;
[0013] Based on the time-series production results output by the time-series production model after dynamic parameter adjustment, the random production results output by the random production model after dynamic parameter adjustment, the new multi-scale analysis results obtained through iteration, and the new prediction credibility and potential risk assessment results, a new energy absorption capacity assessment system is constructed to conduct a comprehensive assessment of the new energy absorption capacity.
[0014] Furthermore, the specific method of constructing and training the time series production model and the random production model based on the historical new energy output data and the simulated new energy output data includes:
[0015] Collect and pre-process historical renewable energy output data;
[0016] Generate a variety of renewable energy output scenarios using random simulation methods to obtain simulated renewable energy output data under different renewable energy output scenarios;
[0017] The historical new energy output data and the simulated new energy output data are integrated to form a new training data set;
[0018] Based on the new training dataset, a time series production model and a random production model are constructed and trained.
[0019] Furthermore, the specific method of generating a plurality of renewable energy output scenarios by using the random simulation method and obtaining simulated renewable energy output data under different renewable energy output scenarios includes:
[0020] Monte Carlo simulation is used to generate a variety of renewable energy output scenarios. In the Monte Carlo simulation, random samples are generated using the following formula:
[0021] X i =μ+δZ i (i=1,2,...,n)
[0022] Among them, Xi represents the result of the ith simulation; μ is the mean, δ is the standard deviation, and Zi is a standard normally distributed random variable.
[0023] Furthermore, the random production model is based on Monte Carlo simulation and output prediction is performed through the following formula:
[0024]
[0025] Among them, P sim is the output of the random production model, X i is the result of the ith simulation, and n is the number of simulations.
[0026] Furthermore, the time series production model adopts a long short-term memory network model, whose input is historical new energy output data and output is the new energy output forecast value, which is expressed as:
[0027] y t =LSTM(x t-n ,x t-n+1 ,...,x t )
[0028] Among them, y t is the predicted output at time t; x t-n , x t-n+1 , ..., x t They are the historical renewable energy output data input at the tnth moment, the t-n+1th moment, ..., and the tnth moment respectively; LSTM is a long short-term memory network model.
[0029] Furthermore, the specific step of fusing the time series production result output by the time series production model and the random production result output by the time series production model includes:
[0030] The time series production results output by the time series production model and the random production results output by the time series production model are weightedly fused, and the weighted fusion formula is:
[0031] P combined =αP LSTM +βP sim
[0032] Among them, P combined is the comprehensive forecast value of the integrated new energy power generation; P LSTM It is the timing production result output by the timing production model; sim is the random production result output by the random production model; α and β are the weight coefficients of the optimized sequential production model and the random production model respectively.
[0033] Furthermore, the specific method of analyzing the comprehensive prediction value of the new energy power generation output at different time scales to obtain the prediction value of the comprehensive prediction value at multiple time scales includes:
[0034] The comprehensive forecast value of the new energy power generation output is analyzed for trend and seasonality on long-term data, and for randomness and volatility on short-term data, and the formula is:
[0035] X long-term =T(t)+S(t)
[0036] X short-term =R(t)+W(t)
[0037] Among them, X long-term is the predicted value in the long time scale; T(t) is the trend component; S(t) is the seasonal component; X short-term is the seasonal component; R(t) is the predicted value under short time scale; R(t) is the random component; W(t) is the volatility component.
[0038] Furthermore, the specific method of performing uncertainty analysis on the comprehensive forecast value of the new energy power generation output and obtaining the confidence interval and error range of the time series production result and the random production result includes:
[0039] The confidence intervals for the results of sequential production simulation and random production simulation are calculated using the formula:
[0040]
[0041] Among them, CI is the confidence interval; is the sample mean; Z is the critical value of the standard normal distribution corresponding to the confidence level; δ is the sample standard deviation; n is the sample size.
[0042] Furthermore, the specific method of dynamically adjusting the parameters of the time series production model and the random production model based on the reinforcement learning algorithm, and iterating the time series production model and the random production model after the dynamic parameter adjustment to obtain new multi-scale analysis results and new prediction credibility and potential risk assessment results includes:
[0043] Adopting a reinforcement learning algorithm to dynamically adjust parameters of the sequential production model and the random production model;
[0044] Obtaining new sequential production results and new random production results according to the sequential production model and the random production model after dynamic parameter adjustment;
[0045] The new time-series production results and the new random production results are integrated to obtain the new comprehensive forecast value of renewable energy power generation;
[0046] Analyze the comprehensive forecast value of new renewable energy power generation at different time scales to obtain new forecast values at multiple time scales, thereby obtaining new multi-scale analysis results; and perform uncertainty analysis on the comprehensive forecast value of new renewable energy power generation to obtain new confidence intervals and error ranges, thereby obtaining forecast credibility and potential risk assessment results;
[0047] The above process is continuously repeated, thereby iterating the time series production model and the random production model adjusted by dynamic parameters to obtain new multi-scale analysis results and new prediction credibility and potential risk assessment results.
[0048] A new energy consumption capacity evaluation system integrating time-series production simulation and random production simulation, including a model building module, a comprehensive prediction module, an analysis module, a parameter adjustment module and an evaluation module;
[0049] The model building module is used to build and train a time series production model and a random production model based on historical new energy output data and simulated new energy output data;
[0050] The comprehensive prediction module is used to merge the time series production results output by the time series production model and the random production results output by the time series production model to obtain a comprehensive prediction value of new energy power generation;
[0051] The analysis module is used to analyze the comprehensive prediction value of the new energy power generation output at different time scales, obtain the prediction values of the comprehensive prediction value at multiple time scales, thereby obtaining multi-scale analysis results, and perform uncertainty analysis on the comprehensive prediction value of the new energy power generation output, thereby obtaining the confidence interval and error range of the time series production result and the random production result, and obtaining the prediction credibility and potential risk assessment results;
[0052] The parameter adjustment module is used to dynamically adjust the parameters of the time series production model and the random production model by using a reinforcement learning algorithm, and iterate the time series production model and the random production model after the dynamic parameter adjustment to obtain new multi-scale analysis results and new prediction credibility and potential risk assessment results;
[0053] The evaluation module is used to build a new energy absorption capacity evaluation system and conduct a comprehensive evaluation of the new energy absorption capacity based on the time series production results output by the time series production model after dynamic parameter adjustment, the random production results output by the random production model after dynamic parameter adjustment, the iterative new multi-scale analysis results and the new prediction credibility and potential risk evaluation results.
[0054] An electronic device comprises a memory and a processor, wherein the memory stores a computer program, and the processor is used to call and run the computer program stored in the memory to execute any of the methods described above.
[0055] A computer-readable storage medium stores a computer program, wherein the computer program implements the steps of any of the above methods when executed by a processor.
[0056] Compared with the prior art, the present invention has the following beneficial effects:
[0057] The present invention proposes a new energy consumption capacity assessment method and system that integrates time-series production simulation and random production simulation. The method effectively combines the advantages of time-series production simulation and random production simulation, improves the assessment accuracy of new energy consumption capacity, and ensures efficient utilization of new energy.
[0058] The evaluation method of the present invention dynamically adjusts the parameters of the time series production model and the random production model through a reinforcement learning algorithm, and iterates the time series production model and the random production model after the dynamic parameter adjustment to obtain new multi-scale analysis results and new prediction credibility and potential risk assessment results. This process can continuously update and optimize the model parameters to make the multi-scale analysis results, prediction credibility and potential risk assessment results more accurate.
[0059] The evaluation method of the present invention is based on the time-series production results output by the time-series production model after dynamic parameter adjustment, the random production results output by the random production model after dynamic parameter adjustment, the iterative new multi-scale analysis results and the new prediction credibility and potential risk assessment results, to construct a new energy absorption capacity evaluation system, thereby enabling a more comprehensive comprehensive evaluation. BRIEF DESCRIPTION OF THE DRAWINGS
[0060] Figure 1 It is a flow chart of the new energy consumption capacity evaluation method integrating time-series production simulation and random production simulation in the embodiment. DETAILED DESCRIPTION
[0061] The present invention is further explained below in conjunction with the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, various equivalent modifications of the present invention by those skilled in the art all fall within the scope defined by the claims attached to this application.
[0062] Embodiment 1:
[0063] A new energy consumption capacity evaluation method integrating sequential production simulation and random production simulation of the present invention comprises:
[0064] S1. Based on historical renewable energy output data and simulated renewable energy output data, build and train a time series production model and a random production model;
[0065] S2, integrating the time series production results output by the time series production model and the random production results output by the time series production model to obtain a comprehensive prediction value of new energy power generation;
[0066] S3. Analyze the comprehensive forecast value of renewable energy power generation output at different time scales, obtain the forecast value of the comprehensive forecast value at multiple time scales, and thereby obtain multi-scale analysis results (such as in-depth analysis of simulation data at long time scales (such as seasonal changes, annual fluctuations, etc.) and short time scales (such as daily changes, hourly fluctuations), which helps to understand the changing pattern and absorption potential of renewable energy output within different time frames), and perform uncertainty analysis on the comprehensive forecast value of renewable energy power generation output, thereby obtaining the confidence interval and error range of time series production results and random production results, and obtaining the prediction credibility and potential risk assessment results (such as evaluating the credibility and potential risks of the prediction by calculating the confidence interval and probability distribution of the results. This step directly depends on the accuracy of the output results of the previous model);
[0067] S4. Use reinforcement learning algorithm to dynamically adjust the parameters of the time series production model and the random production model, and iterate the time series production model and the random production model after dynamic parameter adjustment to obtain new multi-scale analysis results and new prediction credibility and potential risk assessment results;
[0068] S5. Based on the time series production results output by the time series production model after dynamic parameter adjustment, the random production results output by the random production model after dynamic parameter adjustment, the new multi-scale analysis results obtained through iteration, and the new prediction credibility and potential risk assessment results, a new energy absorption capacity assessment system is constructed to conduct a comprehensive assessment of the new energy absorption capacity.
[0069] Embodiment 2:
[0070] This embodiment is further designed on the basis of the first embodiment in that, in this embodiment, based on the historical new energy output data and the simulated new energy output data, the specific method of constructing and training the time series production model and the random production model includes:
[0071] Collect and preprocess historical renewable energy output data; (Preprocessing includes removing outliers, filling missing data, and normalizing data.)
[0072] Generate a variety of renewable energy output scenarios using random simulation methods to obtain simulated renewable energy output data under different renewable energy output scenarios;
[0073] The historical new energy output data and the simulated new energy output data are integrated to form a new training data set;
[0074] Based on the new training dataset, a time series production model and a random production model are constructed and trained.
[0075] Embodiment three:
[0076] This embodiment is further designed on the basis of the second embodiment in that a variety of new energy output scenarios are generated by using a random simulation method in this embodiment, and the specific method for obtaining simulated new energy output data under different new energy output scenarios includes:
[0077] Monte Carlo simulation is used to generate a variety of renewable energy output scenarios. In the Monte Carlo simulation, random samples are generated using the following formula:
[0078] X i =μ+δZ i (i=1,2,...,n)
[0079] Among them, Xi represents the result of the ith simulation; μ is the mean, δ is the standard deviation, and Z i is a standard normally distributed random variable.
[0080] Embodiment 4:
[0081] This embodiment is further designed on the basis of the third embodiment in that the random production model in this embodiment is based on Monte Carlo simulation and output prediction is performed through the following formula:
[0082]
[0083] Among them, P sim is the output of the random production model, X i is the result of the ith simulation, and n is the number of simulations.
[0084] Embodiment five:
[0085] This embodiment is further designed on the basis of the second embodiment in that the time series production model in this embodiment adopts a long short-term memory network model (LSTM), whose input is historical new energy output data, and output is the new energy output forecast value, which is expressed as:
[0086] y t =LSTM(x t-n ,x t-n+1 ,...,x t )
[0087] Among them, y t is the predicted output at time t; x t-n , x t-n+1 , ..., x tThey are the historical renewable energy output data input at the tnth moment, the t-n+1th moment, ..., and the tnth moment (the historical renewable energy output data is generally a time series); LSTM is a long short-term memory network model.
[0088] Embodiment six:
[0089] This embodiment is further designed on the basis of the first embodiment in that the specific steps of fusing the time series production results output by the time series production model and the random production results output by the time series production model in this embodiment include:
[0090] The time series production results output by the time series production model and the random production results output by the time series production model are weighted fused. The weighted fusion formula is:
[0091] P combined =αP LSTM +βP sim
[0092] Among them, P combined is the comprehensive forecast value of the integrated new energy power generation; P LSTM It is the timing production result output by the timing production model; sim is the random production result output by the random production model; α and β are the weight coefficients of the optimized sequential production model and the random production model respectively.
[0093] Embodiment seven:
[0094] This embodiment is further designed on the basis of the first embodiment in that in this embodiment, the comprehensive prediction value of the output of renewable energy power generation is analyzed at different time scales, and the specific method of obtaining the prediction value of the comprehensive prediction value at multiple time scales includes:
[0095] The comprehensive forecast value of renewable energy power generation output is analyzed for trend and seasonality of long-term data, and for randomness and volatility of short-term data. The formula is:
[0096] X long-term =T(t)+S(t)
[0097] X short-term =R(t)+W(t)
[0098] Among them, X long-term is the predicted value in the long time scale; T(t) is the trend component; S(t) is the seasonal component; X short-term is the seasonal component; R(t) is the predicted value under short time scale; R(t) is the random component; W(t) is the volatility component.
[0099] Embodiment eight:
[0100] This embodiment is further designed on the basis of the first embodiment in that in this embodiment, uncertainty analysis is performed on the comprehensive forecast value of the output of renewable energy power generation, and the specific method of obtaining the confidence interval and error range of the time series production results and the random production results includes:
[0101] The confidence intervals for the results of sequential production simulation and random production simulation are calculated using the formula:
[0102]
[0103] Among them, CI is the confidence interval; is the sample mean; Z is the critical value of the standard normal distribution corresponding to the confidence level; δ is the sample standard deviation; n is the sample size.
[0104] Embodiment nine:
[0105] This embodiment is further designed on the basis of the first embodiment in that, in this embodiment, the specific method of dynamically adjusting the parameters of the time series production model and the random production model based on the reinforcement learning algorithm, and iterating the time series production model and the random production model after the dynamic parameter adjustment to obtain new multi-scale analysis results and new prediction credibility and potential risk assessment results includes:
[0106] Reinforcement learning algorithm is used to dynamically adjust parameters of sequential production model and random production model;
[0107] Obtain new sequential production results and new random production results according to the sequential production model and the random production model adjusted by dynamic parameters;
[0108] The new time-series production results and the new random production results are integrated to obtain the new comprehensive forecast value of renewable energy power generation;
[0109] Analyze the comprehensive forecast value of new renewable energy power generation at different time scales to obtain new forecast values at multiple time scales, thereby obtaining new multi-scale analysis results; and perform uncertainty analysis on the comprehensive forecast value of new renewable energy power generation to obtain new confidence intervals and error ranges, thereby obtaining forecast credibility and potential risk assessment results;
[0110] The above process is repeated continuously, thereby iterating the time series production model and the random production model adjusted by dynamic parameters to obtain new multi-scale analysis results and new prediction credibility and potential risk assessment results.
[0111] Embodiment ten:
[0112] The present invention provides a new energy consumption capacity evaluation system integrating sequential production simulation and random production simulation, characterized in that it includes a model building module, a comprehensive prediction module, an analysis module, a parameter adjustment module and an evaluation module;
[0113] Model building module, used to build and train time series production models and stochastic production models based on historical renewable energy output data and simulated renewable energy output data;
[0114] The comprehensive prediction module is used to integrate the time series production results output by the time series production model and the random production results output by the time series production model to obtain a comprehensive prediction value of renewable energy power generation;
[0115] An analysis module is used to analyze the comprehensive prediction value of the output of renewable energy power generation at different time scales, obtain the prediction values of the comprehensive prediction value at multiple time scales, thereby obtaining multi-scale analysis results, and perform uncertainty analysis on the comprehensive prediction value of the output of renewable energy power generation, thereby obtaining the confidence interval and error range of the time series production results and the random production results, and obtaining the prediction credibility and potential risk assessment results;
[0116] The parameter adjustment module is used to dynamically adjust the parameters of the time series production model and the random production model using a reinforcement learning algorithm, and to iterate the time series production model and the random production model after the dynamic parameter adjustment to obtain new multi-scale analysis results and new prediction credibility and potential risk assessment results;
[0117] The evaluation module is used to build a new energy absorption capacity evaluation system and conduct a comprehensive evaluation of the new energy absorption capacity based on the time series production results output by the time series production model after dynamic parameter adjustment, the random production results output by the random production model after dynamic parameter adjustment, the new multi-scale analysis results obtained through iteration, and the new prediction credibility and potential risk assessment results.
[0118] Embodiment eleven:
[0119] An electronic device comprises a memory and a processor, wherein the memory stores a computer program, and the processor is used to call and run the computer program stored in the memory to execute any of the methods described above.
[0120] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any of the above methods are implemented.
[0121] The above description is only a specific implementation of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a technician familiar with the technical field within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention.
Claims
1. A new energy consumption capacity evaluation method integrating time-series production simulation and random production simulation, characterized in that: include: Based on historical renewable energy output data and simulated renewable energy output data, build and train time series production models and stochastic production models; The time series production results output by the time series production model and the random production results output by the time series production model are integrated to obtain a comprehensive prediction value of new energy power generation; Performing analysis on the comprehensive forecast value of the output of the new energy power generation at different time scales to obtain forecast values of the comprehensive forecast value at multiple time scales, thereby obtaining a multi-scale analysis result, and performing uncertainty analysis on the comprehensive forecast value of the output of the new energy power generation to obtain a confidence interval and an error range of the time series production result and the random production result, thereby obtaining a forecast credibility and potential risk assessment result; Adopting a reinforcement learning algorithm to dynamically adjust the parameters of the time series production model and the random production model, and iterating the time series production model and the random production model after the dynamic parameter adjustment to obtain new multi-scale analysis results and new prediction credibility and potential risk assessment results; Based on the time-series production results output by the time-series production model after dynamic parameter adjustment, the random production results output by the random production model after dynamic parameter adjustment, the new multi-scale analysis results obtained through iteration, and the new prediction credibility and potential risk assessment results, a new energy absorption capacity assessment system is constructed to conduct a comprehensive assessment of the new energy absorption capacity.
2. The new energy consumption capacity evaluation method integrating sequential production simulation and random production simulation according to claim 1 is characterized in that: The specific method of constructing and training the time series production model and the random production model based on the historical new energy output data and the simulated new energy output data includes: Collect and pre-process historical renewable energy output data; Generate a variety of renewable energy output scenarios using random simulation methods to obtain simulated renewable energy output data under different renewable energy output scenarios; The historical new energy output data and the simulated new energy output data are integrated to form a new training data set; Based on the new training dataset, a time series production model and a random production model are constructed and trained.
3. The new energy consumption capacity evaluation method integrating sequential production simulation and random production simulation according to claim 2 is characterized in that: The specific method of using the random simulation method to generate a variety of new energy output scenarios and obtain simulated new energy output data under different new energy output scenarios includes: Monte Carlo simulation is used to generate a variety of renewable energy output scenarios. In the Monte Carlo simulation, random samples are generated using the following formula: X i =μ+δZ i (i=1,2,...,n) Among them, Xi represents the result of the ith simulation; μ is the mean, δ is the standard deviation, and Z i is a standard normally distributed random variable.
4. The new energy consumption capacity evaluation method integrating sequential production simulation and random production simulation according to claim 3 is characterized in that: The stochastic production model is based on Monte Carlo simulation and output prediction is performed using the following formula: Among them, P sim is the output of the random production model, X i is the result of the ith simulation, and n is the number of simulations.
5. The new energy consumption capacity evaluation method integrating sequential production simulation and random production simulation according to claim 2 is characterized in that: The time series production model adopts a long short-term memory network model, whose input is historical new energy output data and output is the new energy output forecast value, which is expressed as: y t =LSTM(x t-n ,x t-n+1 ,...,x t ) Among them, y t is the predicted output at time t; x t-n , x t-n+1 , ..., x t They are the historical renewable energy output data input at the tnth moment, the t-n+1th moment, ..., and the tnth moment respectively; LSTM is a long short-term memory network model.
6. The new energy consumption capacity evaluation method integrating sequential production simulation and random production simulation according to claim 1 is characterized in that: The specific steps of fusing the sequential production results output by the sequential production model and the random production results output by the sequential production model include: The time series production results output by the time series production model and the random production results output by the time series production model are weightedly fused, and the weighted fusion formula is: P combined =αP LSTM +βP sim Among them, P combined is the comprehensive forecast value of the integrated new energy power generation; P LSTM It is the timing production result output by the timing production model; sim is the random production result output by the random production model; α and β are the weight coefficients of the optimized sequential production model and the random production model respectively.
7. The new energy consumption capacity evaluation method integrating sequential production simulation and random production simulation according to claim 1 is characterized in that: The specific method of analyzing the comprehensive prediction value of the new energy power generation output at different time scales to obtain the prediction value of the comprehensive prediction value at multiple time scales includes: The comprehensive forecast value of the new energy power generation output is analyzed for trend and seasonality on long-term data, and for randomness and volatility on short-term data, and the formula is: X long-term =T(t)+S(t) X short-term =R(t)+W(t) Among them, X long-term is the predicted value in the long time scale; T(t) is the trend component; S(t) is the seasonal component; X short-term is the seasonal component; R(t) is the predicted value under short time scale; R(t) is the random component; W(t) is the volatility component.
8. The new energy consumption capacity evaluation method integrating sequential production simulation and random production simulation according to claim 1 is characterized in that: The specific method of performing uncertainty analysis on the comprehensive forecast value of the new energy power generation output and obtaining the confidence interval and error range of the time series production result and the random production result includes: The confidence intervals for the results of sequential production simulation and random production simulation are calculated using the formula: Among them, CI is the confidence interval; is the sample mean; Z is the critical value of the standard normal distribution corresponding to the confidence level; δ is the sample standard deviation; n is the sample size.
9. The new energy consumption capacity evaluation method integrating sequential production simulation and random production simulation according to claim 1 is characterized in that: The specific method of dynamically adjusting the parameters of the time series production model and the random production model based on the reinforcement learning algorithm, and iterating the time series production model and the random production model after the dynamic parameter adjustment to obtain new multi-scale analysis results and new prediction credibility and potential risk assessment results includes: Adopting a reinforcement learning algorithm to dynamically adjust parameters of the sequential production model and the random production model; Obtaining new sequential production results and new random production results according to the sequential production model and the random production model after dynamic parameter adjustment; The new time-series production results and the new random production results are integrated to obtain the new comprehensive forecast value of renewable energy power generation; Analyze the comprehensive forecast value of new renewable energy power generation at different time scales to obtain new forecast values at multiple time scales, thereby obtaining new multi-scale analysis results; and perform uncertainty analysis on the comprehensive forecast value of new renewable energy power generation to obtain new confidence intervals and error ranges, thereby obtaining forecast credibility and potential risk assessment results; The above process is continuously repeated, thereby iterating the time series production model and the random production model adjusted by dynamic parameters to obtain new multi-scale analysis results and new prediction credibility and potential risk assessment results.
10. A new energy consumption capacity evaluation system integrating sequential production simulation and random production simulation, characterized in that: It includes model building module, comprehensive prediction module, analysis module, parameter adjustment module and evaluation module; The model building module is used to build and train a time series production model and a random production model based on historical new energy output data and simulated new energy output data; The comprehensive prediction module is used to merge the time series production results output by the time series production model and the random production results output by the time series production model to obtain a comprehensive prediction value of new energy power generation; The analysis module is used to analyze the comprehensive prediction value of the new energy power generation output at different time scales, obtain the prediction values of the comprehensive prediction value at multiple time scales, thereby obtaining multi-scale analysis results, and perform uncertainty analysis on the comprehensive prediction value of the new energy power generation output, thereby obtaining the confidence interval and error range of the time series production result and the random production result, and obtaining the prediction credibility and potential risk assessment results; The parameter adjustment module is used to dynamically adjust the parameters of the time series production model and the random production model by using a reinforcement learning algorithm, and iterate the time series production model and the random production model after the dynamic parameter adjustment to obtain new multi-scale analysis results and new prediction credibility and potential risk assessment results; The evaluation module is used to build a new energy absorption capacity evaluation system and conduct a comprehensive evaluation of the new energy absorption capacity based on the time series production results output by the time series production model after dynamic parameter adjustment, the random production results output by the random production model after dynamic parameter adjustment, the iterative new multi-scale analysis results and the new prediction credibility and potential risk evaluation results.