Priority electricity purchase curve prediction and decomposition method considering seasonal difference of supply and demand
By collecting and analyzing seasonal data, combining multiple linear regression analysis and the comprehensive operation index of new energy units, the problem of seasonal changes in power supply and demand in the existing technology has been solved, and a fine prediction of the priority power purchase curve is achieved. and decomposition have improved the utilization efficiency of power resources and the dispatching capacity of new energy power generation.
Patent Information
- Application Number
- CN202411073323.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-06
- Publication Date
- 2024-11-08
AI Technical Summary
The existing prediction and decomposition methods of priority power purchase curve cannot fully consider the seasonal changes in power supply and demand, resulting in the predictions being not refined and dynamic enough, and the inability to effectively utilize new energy resources, affecting the sustainable development of the power system.
By collecting seasonal data of regional power grids, such as monthly average temperature, average monthly precipitation and sunshine time, combined with multiple linear regression analysis, electricity consumption prediction function is constructed, and calculated based on the installed capacity, historical deviation rate and operation time of new energy units. Comprehensive operation index and decompose the priority power purchase curve.
It has achieved more refined and dynamic prediction of the priority power purchase curve, improved the accuracy and practicality of the prediction, helped to effectively utilize power resources, and promoted the rational dispatch and utilization of new energy power generation.
Smart Images

Figure CN118920476A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power forecasting, and in particular to a method for predicting and decomposing a priority power purchase curve taking into account seasonal differences in supply and demand. Background Art
[0002] With the goal of building a new power system with new energy as the main body, my country's renewable energy power generation will reach more than 70% in the future. Historically, there are many challenges in the prediction and decomposition of the priority purchase curve based on the traditional static decomposition method. Considering the seasonality and crop growth needs, it is necessary to seek more sophisticated and dynamic prediction and decomposition methods to achieve effective use of resources and sustainable development of electricity consumption.
[0003] At present, there are few studies on the prediction and decomposition system of priority power purchase curve. The existing methods mainly use relevant algorithms to predict electricity consumption data. Under the requirement of more refined system for decomposition based on predicted data, it is necessary to study the prediction and decomposition method of priority power purchase curve considering seasonal differences based on the theoretical basis of relevant regression algorithms. However, it is unable to fully consider the seasonal changes in electricity supply and demand. Summary of the invention
[0004] Based on the above objectives, the present invention provides a method for predicting and decomposing a priority electricity purchase curve taking into account seasonal differences in supply and demand.
[0005] A method for predicting and decomposing a priority electricity purchase curve considering seasonal differences in supply and demand, comprising the following steps:
[0006] S1: Data collection, collecting seasonal data of the regional power grid throughout the year, including monthly average temperature, monthly average precipitation, sunshine time and historical electricity consumption data for the same period;
[0007] S2: Multiple linear regression analysis, using the collected seasonal related data, conduct multiple linear regression analysis on the priority power purchase curve considering various related data factors, determine the degree of influence factors of each variable, and select the first three influencing factors as the influencing factors of the fitting function;
[0008] S3: variance analysis, based on the selected influencing factors, variance analysis is performed to calculate the coefficients of each variable and construct the electricity consumption prediction function;
[0009] S4: Model verification and adjustment: Based on the analysis results of SPSS software, check the correlation coefficient R in the model summary table, and adjust it according to the preset threshold of the correlation coefficient R. Continue to perform decomposition for the priority power purchase prediction curve that meets the requirements;
[0010] S5: Decomposition of the priority electricity purchase curve, including:
[0011] If the region publishes the forecast output curve of renewable energy, the priority purchase curve will be decomposed to each renewable energy unit based on the forecast output ratio of renewable energy units in the province;
[0012] If the region has not published the day-ahead forecast output curve for new energy sources, the priority power purchase curve will be decomposed according to the comprehensive operating index of each new energy unit.
[0013] Furthermore, the data collection in S1 specifically includes:
[0014] Monthly average temperature collection: Cooperate with the local meteorological bureau to obtain the historical temperature records of the area, including the highest, lowest and average temperatures of each month, or use a public meteorological data service platform or meteorological data API;
[0015] Average monthly precipitation collection: Obtain historical precipitation records from local meteorological bureaus or through meteorological data services;
[0016] Sunshine time collection: obtained through the Meteorological Bureau or the meteorological data service platform, including the total sunshine time per month and the average daily sunshine time;
[0017] Collection of electricity consumption in the same period: Request local power grid companies or relevant power management agencies to provide electricity consumption data for the past few years, including total monthly electricity consumption and peak electricity consumption, or use electricity consumption statistics from publicly released industry reports or government announcements;
[0018] Data integration and standardization: All collected data need to be integrated and standardized to ensure uniform data format, comparability and accuracy of analysis.
[0019] Furthermore, the multivariate linear regression analysis in S2 is calculated as follows:
[0020] Assume that the power load variable y i With the independent variable x i (i=1,2,…,p), respectively replace the sign 1,2,…,p) and express the relationship as follows:
[0021] y i =β0+β1x 1i +β2x 2i +…+β p x pi +ε i ;
[0022] Among them, y i is the priority electricity purchase curve, i=1,2,…,n is the sequence number of the independent variable, β0,β1,…β p is the regression coefficient of the model, ε i is divided by the independent variable x i(i=1,2,…,n) has an impact on the power consumption y i The random variable that produces the influence, that is, the random error, is assumed to have a mean of 0 and a variance of σ 2 , following the same normal distribution N(0,σ 2 ), when the above conditions are met, the priority electricity purchase curve y is obtained i The predicted value of .
[0023] Furthermore, the predicted value of the priority electricity purchase curve yi is expressed as:
[0024] The predicted value is used for parameter estimation based on the least squares method.
[0025] Furthermore, the power consumption prediction function in S3 is expressed as:
[0026] y=C+k1x1+k2x2+k3x3, where C is a constant, and k1, k2, and k3 are the coefficients of the respective variables.
[0027] Furthermore, the preset threshold of the correlation coefficient R in S4 is 0.85;
[0028] If the correlation coefficient R does not reach 0.85, the selected influencing factors are adjusted, and the updated influencing factors are re-introduced into the SPSS software for correlation analysis until the correlation coefficient R reaches 0.85, and the priority power purchase prediction curve that meets the result is further decomposed.
[0029] Furthermore, in S5, if the region publishes the day-ahead predicted output curve of new energy, the distribution is performed according to the corresponding proportions of the predicted quantities A1, A2, A3...An of the units X1, X2, X3...Xn.
[0030] Furthermore, the comprehensive operation index includes the monthly average historical deviation rate of each new energy unit, the new energy installed capacity, and the unit commissioning time factors. The comprehensive operation index of each new energy unit is comprehensively calculated, and based on the comprehensive operation index, the priority power purchase curve is decomposed.
[0031] Furthermore, the decomposition of the priority power purchase curve according to the comprehensive operation index of each new energy unit in S5 specifically includes:
[0032] Obtain the installed capacity C of each new energy unit and the historical deviation rate of each unit in each month in the past two years, which are recorded as A1, A2...An respectively;
[0033] For units that have been put into operation within two years, the installed capacity C×0.8 and the historical average deviation rate The coefficient of new energy installed capacity coefficient is calculated;
[0034] For units that have been put into operation for more than two years, the installed capacity C×0.7 and the historical average deviation rate The coefficient of new energy installed capacity coefficient is calculated;
[0035] For a given historical deviation rate At, if the historical deviation rate At of the unit is greater than 20% for more than 5 times, the unit is deemed to have low reliability, and 0.1 is added to the original reliability coefficient of the unit; if the historical deviation rate At is less than 20%, the unit is deemed to have high reliability, and the original coefficient is kept unchanged, and finally the comprehensive operation index of the new energy unit is obtained.
[0036] Furthermore, the calculation of the comprehensive operation index of the new energy unit includes:
[0037] For a given new energy unit: If the commissioning time is two years or less, and the historical deviation rate is greater than 20% for more than 5 times, the calculation is: If the commissioning time is more than two years and the historical deviation rate is greater than 20% for more than 5 times, the calculation is: If the commissioning time is within two years and the historical deviation rate is greater than 20% for no more than 5 times, the calculation is: If the operation time is more than two years and the historical deviation rate is greater than 20% for no more than 5 times, it is calculated as: Calculation; then add up the comprehensive operating index of each new energy unit and take the corresponding percentage, comprehensively calculate the proportion of the new energy unit in all optional units, multiply the proportion by the total required power generation in the area, and get the required power generation of the generator unit.
[0038] Beneficial effects of the present invention:
[0039] The present invention fully considers the seasonal changes in electricity supply and demand, collects relevant data such as the monthly average temperature, monthly average precipitation, sunshine time, historical electricity consumption in the same period of the year in the regional power grid, and combines it with multivariate linear regression analysis to achieve a more precise and dynamic prediction of the priority electricity purchase curve.
[0040] The present invention also takes into account multiple factors such as the installed capacity of the new energy unit, the historical deviation rate, the unit commissioning time, etc., and proposes a comprehensive operation index of the new energy unit that decomposes the priority power purchase curve, thereby improving the accuracy and practicality of the prediction.
[0041] The present invention can achieve more accurate prediction and decomposition of the priority electricity purchase curve, which helps to improve the effective utilization of power resources and achieve sustainable development of electricity consumption. As the proportion of new energy in the power system continues to increase, this solution provides a more refined priority electricity purchase curve prediction and decomposition method for new energy power generation, which helps to reasonably dispatch and utilize new energy power generation. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0043] Figure 1 The present invention is a flowchart of a provincial priority electricity purchase curve prediction and decomposition method considering seasonal differences according to an embodiment of the present invention. DETAILED DESCRIPTION
[0044] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.
[0045] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.
[0046] like Figure 1 As shown in the figure, a method for predicting and decomposing the priority electricity purchase curve considering the seasonal differences in supply and demand is proposed.
[0047] (1) Collect data related to the regional power grid’s monthly average temperature, monthly average precipitation, sunshine hours, and historical electricity consumption over the same period throughout the year.
[0048] (2) Based on the data, a multivariate linear regression analysis is performed on the priority power purchase curve considering various factors. The calculation method is as follows:
[0049] In power load forecasting, it is assumed that the power load variable y i (Priority electricity purchase curve) and independent variable x i (i=1,2,…,p) (such as temperature, precipitation, sunshine time, historical electricity consumption in the same period, etc., respectively replace the sign 1,2,…,p) can be expressed as:
[0050] y i =β0+β1x1i +β2x 2i +…+β p x pi +ε i
[0051] Where i = 1, 2, ..., n (the sequence number of the independent variable), β0, β1, ... β p is the regression coefficient of the model. i is divided by the independent variable x i (i=1,2,…,n) has an impact on the power consumption y i The random variables that have an impact, that is, random errors, are assumed to have a mean of 0 and a variance of σ 2 , that is, following the same normal distribution N(0,σ 2 ).
[0052] When the above conditions are met, the priority electricity purchase curve y can be obtained according to the above formula: i Predicted value of:
[0053]
[0054] The above formula is based on the least squares method for parameter estimation.
[0055] (3) Based on the selected influencing factors (here are the average monthly precipitation, average monthly temperature, and sunshine time), variance analysis was performed using SPSS software to obtain the coefficients of each variable, and the electricity consumption prediction function y=C+k1x1+k2x2+k3x3 was obtained, where C is a constant, and k1, k2, and k3 are the coefficients of each variable.
[0056] (4) Based on the results of the software analysis, check the correlation coefficient R in the model summary table. If the coefficient R does not reach 0.85, it is necessary to adjust the selected influencing factors and bring the updated influencing factors back into the SPSS software for correlation analysis until the correlation coefficient R reaches 0.85. The priority power purchase prediction curve that meets the result is further decomposed.
[0057] (5) The priority power purchase curve is allocated to each new energy unit according to the proportion of the new energy unit's predicted output in the province, that is, according to the proportion of the predicted output A1, A2, A3...An of the units X1, X2, X3...Xn; if the region has not published the new energy day-ahead output forecast curve, the priority power purchase curve is decomposed according to the comprehensive operation index of each new energy unit, that is, considering multiple factors such as the monthly average historical deviation rate of each new energy unit, the installed capacity of new energy, and the time of unit commissioning, the comprehensive operation index of each new energy unit is comprehensively calculated, and based on this index, the priority power purchase curve is decomposed. The specific steps are as follows:
[0058] First, obtain the installed capacity C of each new energy unit and the historical deviation rate of each unit in each month in the past two years, which are recorded as A1, A2...An respectively. For units that have been put into operation within two years, according to the installed capacity C×0.8, the historical average deviation rate The coefficient of the installed capacity of new energy is calculated; for units that have been put into operation for more than two years, the coefficient is calculated according to the installed capacity C×0.7 and the historical average deviation rate The coefficient of new energy installed capacity is calculated based on the coefficient of .
[0059] At the same time, for a given historical deviation rate At, if the historical deviation rate At of the unit is greater than 20% for more than 5 times, the unit is deemed to have a low reliability, and 0.1 is added to the original reliability coefficient of the unit; if the historical deviation rate At is less than 20%, the unit is deemed to be relatively reliable, and the original coefficient is kept unchanged, and finally the comprehensive operation index of the new energy unit is obtained.
[0060] Specifically, for a given renewable energy unit: if the commissioning time is two years or less, and the historical deviation rate is greater than 20% for more than five times, Calculation; if the operation time is more than two years and the historical deviation rate is greater than 20% for more than 5 times, The commissioning time is two years or less, and the historical deviation rate is greater than 20% for no more than 5 times. The commissioning time is more than two years, and the historical deviation rate is greater than 20% for no more than 5 times. calculate.
[0061] Finally, the comprehensive operation index of each new energy unit is added up and the corresponding percentage is taken to comprehensively obtain the proportion of the new energy unit in all optional units. Finally, multiplying the total required power generation in the area can get the required power generation of the generator unit.
[0062] Specific examples:
[0063] The priority electricity purchase curve is affected by many factors such as urban power supply, total population, seasonal factors, gross national product, etc. It is proposed to use four factors such as the monthly average temperature and precipitation in a certain area as examples to study its impact on the total regional electricity consumption. Table 1 shows the values of the total social electricity consumption in a certain area in 2017 and its influencing factors.
[0064] Table 1 Data values for each month in 2022
[0065]
[0066]
[0067] According to the data of monthly average temperature, monthly average precipitation, sunshine time, regional total electricity consumption, etc., the total electricity consumption of the whole society was predicted by linear regression using SPSS software. According to the multiple linear regression method, the significance level α=0.05 was taken, and the Person correlation analysis was performed on each influencing factor. The results are shown in Table 2.
[0068] Table 2 Correlation analysis
[0069]
[0070] Table 3 Model summary
[0071]
[0072]
[0073] From Table 2, we know that the correlation coefficients between the total social electricity consumption and the five influencing factors are high, and the overall correlation coefficient is also large, so the regression effect is good. SPSS software stepwise regression is used to screen and eliminate the variables that cause multicollinearity. Table 3 is obtained. The overall correlation coefficient R=0.874 between the three independent variables (X1, X2, X3) in model 2 and the dependent variable y indicates that the correlation is significant at the level of α=0.126, indicating that 87.4% of the influencing factors can be explained by this model, indicating that the established regression model is better. That is, the monthly average temperature and monthly average precipitation are selected as independent variables.
[0074] The data were processed using SPSS software, and the predicted value and residual 1 of annual electricity consumption obtained by the multiple linear regression method were obtained using the above formula.
[0075] Table 4 Coefficients
[0076]
[0077] According to the standardized coefficient values of each variable in Table 4, the multivariate linear regression equation of the total social electricity consumption in the region is obtained as follows:
[0078] y1=45.074-1.016x1+0.09x2+0.062x3
[0079] The predicted value of the priority electricity purchase curve in the region is calculated using the derived linear regression equation.
[0080] The average monthly temperature in this area is 13.4℃, the average monthly precipitation is 49.3mm, and the sunshine time is 12h. The multivariate linear regression equation obtained shows that the average monthly electricity consumption of the whole society is 3.664 billion kWh, and the annual electricity consumption is 43.9687 billion kWh.
[0081] Decompose according to the comprehensive operation index of each new energy unit. That is, consider multiple factors such as the monthly average historical deviation rate of each new energy unit, the installed capacity of new energy, and the commissioning time of the unit, and comprehensively calculate the comprehensive operation index of each new energy unit, and based on this index, decompose the priority power purchase curve.
[0082] First, the installed capacity C of each renewable energy unit and the historical deviation rate of each unit in each month in the past two years are obtained, which are recorded as A1, A2...An respectively.
[0083] The monthly historical deviation data of new energy units in this region are as follows:
[0084]
[0085]
[0086] The installed capacity data of new energy generators in the region are as follows:
[0087]
[0088] For units that have been put into operation within two years, the installed capacity C×0.8 and the historical average deviation rate The coefficient of the installed capacity of new energy is calculated; for units that have been put into operation for more than two years, the coefficient is calculated according to the installed capacity C×0.7 and the historical average deviation rate The coefficient of the installed capacity coefficient of new energy is calculated. At the same time, for a given historical deviation rate, if the historical deviation rate of the unit is greater than 20% for more than 5 times, the unit is considered to have low reliability, and 0.1 is added to the original reliability coefficient of the unit; if the historical deviation rate is less than 20%, the unit is considered to be relatively reliable, and the original coefficient remains unchanged. Finally, the comprehensive operation index of the new energy unit is obtained.
[0089] The calculation rules for the comprehensive operation coefficient of new energy units are as follows: That is, for a given new energy unit: if the commissioning time is two years or less, and the historical deviation rate is greater than 20% for more than 5 times, Calculation; if the operation time is more than two years and the historical deviation rate is greater than 20% for more than 5 times, Calculation; if the commissioning time is within two years and the historical deviation rate is greater than 20% for no more than 5 times, Calculation; if the operation time is more than two years and the historical deviation rate is greater than 20% for no more than 5 times, calculate.
[0090]
[0091] Therefore, the number of times that the deviation rate of the second, third, and fourth groups in the above units is greater than 20% exceeds 5 times. The corresponding calculation formula should be used to sum.
[0092]
[0093] Based on the above results, 5.188 billion, 8.046 billion, 7.343 billion, 11.564 billion and 11.872 billion kWh of electricity were allocated to units 1 to 5 respectively.
[0094] The above results are more positive than the original plan of the provincial power grid to predict and decompose the priority power purchase curve that took seasonal differences into consideration.
[0095] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention is limited to these examples. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity.
[0096] The present invention is intended to cover all such substitutions, modifications and variations that fall within the broad scope of the claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for predicting and decomposing a priority electricity purchase curve considering seasonal differences in supply and demand, characterized in that: The following steps are involved: S1: Data collection, collecting seasonal data of the regional power grid throughout the year, including monthly average temperature, monthly average precipitation, sunshine time and historical electricity consumption data for the same period; S2: Multiple linear regression analysis, using the collected seasonal related data, conduct multiple linear regression analysis on the priority power purchase curve considering various related data factors, determine the degree of influence factors of each variable, and select the first three influencing factors as the influencing factors of the fitting function; S3: variance analysis, based on the selected influencing factors, variance analysis is performed to calculate the coefficients of each variable and construct the electricity consumption prediction function; S4: Model verification and adjustment: Based on the analysis results of SPSS software, check the correlation coefficient R in the model summary table, and adjust it according to the preset threshold of the correlation coefficient R. Continue to perform decomposition for the priority power purchase prediction curve that meets the requirements; S5: Decomposition of the priority electricity purchase curve, including: If the region publishes the forecast output curve of renewable energy, the priority purchase curve will be decomposed to each renewable energy unit based on the forecast output ratio of renewable energy units in the province; If the region has not published the day-ahead forecast output curve for new energy sources, the priority power purchase curve will be decomposed according to the comprehensive operating index of each new energy unit.
2. A method for predicting and decomposing a priority electricity purchase curve considering seasonal differences in supply and demand according to claim 1, characterized in that: The data collection in S1 specifically includes: Monthly average temperature collection: Cooperate with the local meteorological bureau to obtain the historical temperature records of the area, including the highest, lowest and average temperatures of each month, or use a public meteorological data service platform or meteorological data API; Average monthly precipitation collection: Obtain historical precipitation records from local meteorological bureaus or through meteorological data services; Sunshine time collection: obtained through the Meteorological Bureau or the meteorological data service platform, including the total sunshine time per month and the average daily sunshine time; Collection of electricity consumption in the same period: Request local power grid companies or relevant power management agencies to provide electricity consumption data for the past few years, including total monthly electricity consumption and peak electricity consumption, or use electricity consumption statistics from publicly released industry reports or government announcements; Data integration and standardization: All collected data need to be integrated and standardized to ensure uniform data format, comparability and accuracy of analysis.
3. A method for predicting and decomposing a priority electricity purchase curve considering seasonal differences in supply and demand according to claim 2, characterized in that: The multivariate linear regression analysis in S2 was calculated as follows: Assume that the power load variable y i With the independent variable x i (i=1,2,…,p), respectively replace the sign 1,2,…,p) and express the relationship as follows: y i =β0+β1x 1i +β2x 2i +…+b p x pi +e i ; Among them, y i is the priority electricity purchase curve, i=1,2,…,n is the sequence number of the independent variable, β0,β1,…β p is the regression coefficient of the model, ε i is divided by the independent variable x i (i=1,2,…,n) has an impact on the power consumption y i The random variable that produces the influence, that is, the random error, is assumed to have a mean of 0 and a variance of σ 2 , following the same normal distribution N(0,σ 2 ), when the above conditions are met, the priority electricity purchase curve y is obtained i The predicted value of .
4. A method for predicting and decomposing a priority electricity purchase curve considering seasonal differences in supply and demand according to claim 3, characterized in that: The priority electricity purchase curve y i The predicted value of is expressed as: The predicted value is used for parameter estimation based on the least squares method.
5. A method for predicting and decomposing a priority electricity purchase curve considering seasonal differences in supply and demand according to claim 4, characterized in that: The power consumption prediction function in S3 is expressed as: y=C+k1x1+k2x2+k3x3, where C is a constant, and k1, k2, and k3 are the coefficients of the respective variables.
6. A method for predicting and decomposing a priority electricity purchase curve considering seasonal differences in supply and demand according to claim 5, characterized in that: The preset threshold of the correlation coefficient R in S4 is 0.85; If the correlation coefficient R does not reach 0.85, the selected influencing factors are adjusted, and the updated influencing factors are re-introduced into the SPSS software for correlation analysis until the correlation coefficient R reaches 0.85, and the priority power purchase prediction curve that meets the result is further decomposed.
7. A method for predicting and decomposing priority electricity purchase curves considering seasonal differences in supply and demand according to claim 6, characterized in that: In S5, if the region publishes the day-ahead predicted output curve of new energy, the distribution is performed according to the corresponding proportions of the predicted quantities A1, A2, A3...An of the units X1, X2, X3...Xn.
8. A method for predicting and decomposing a priority electricity purchase curve considering seasonal differences in supply and demand according to claim 7, characterized in that:The comprehensive operation index includes the monthly average historical deviation rate of each new energy unit, the new energy installed capacity, and the unit commissioning time factors. The comprehensive operation index of each new energy unit is comprehensively calculated, and based on the comprehensive operation index, the priority power purchase curve is decomposed.
9. A method for predicting and decomposing a priority electricity purchase curve considering seasonal differences in supply and demand according to claim 8, characterized in that: The decomposition of the priority power purchase curve according to the comprehensive operation index of each new energy unit in S5 specifically includes: Obtain the installed capacity C of each new energy unit and the historical deviation rate of each unit in each month in the past two years, which are recorded as A1, A2...An respectively; For units that are put into operation within two years, the new energy installed capacity coefficient is calculated according to the coefficient of installed capacity C×0.8 and historical average deviation rate AA×0.2; For units that have been put into operation for more than two years, the installed capacity C×0.7 and the historical average deviation rate The coefficient of new energy installed capacity coefficient is calculated; For a given historical deviation rate At, if the historical deviation rate At of the unit is greater than 20% for more than 5 times, the unit is deemed to have low reliability, and 0.1 is added to the original reliability coefficient of the unit; if the historical deviation rate At is less than 20%, the unit is deemed to have high reliability, and the original coefficient is kept unchanged, and finally the comprehensive operation index of the new energy unit is obtained.
10. A method for predicting and decomposing priority electricity purchase curves considering seasonal differences in supply and demand according to claim 9, characterized in that: The calculation of the comprehensive operation index of the new energy unit includes: For a given new energy unit: If the commissioning time is two years or less, and the historical deviation rate is greater than 20% for more than 5 times, the calculation is: If the commissioning time is more than two years and the historical deviation rate is greater than 20% for more than 5 times, the calculation is: If the commissioning time is within two years and the historical deviation rate is greater than 20% for no more than 5 times, the calculation is: If the operation time is more than two years and the historical deviation rate is greater than 20% for no more than 5 times, it is calculated as: Calculation; then add up the comprehensive operating index of each new energy unit and take the corresponding percentage, comprehensively calculate the proportion of the new energy unit in all optional units, multiply the proportion by the total required power generation in the area, and get the required power generation of the generator unit.
