A multi-dimensional medium- and long-term power generation forecasting method

Abstract

The present invention provides a multi-dimensional medium- and long-term power generation forecasting method, which relates to the technical field of multi-dimensional power generation forecasting. Based on the analysis of historical power generation data acquired by a power information analysis platform of a target city, a power generation forecast deviation adjustment coefficient is obtained, and the power generation forecast deviation adjustment coefficient is used to numerically represent the deviation between the historical actual power generation of the target city and the predicted power generation; relevant information about power generation influencing factors of the target city is acquired, and the relevant information about power generation influencing factors is combined with the power generation forecast deviation adjustment coefficient analysis to obtain a multi-dimensional power generation forecast index, and the multi-dimensional power generation forecast index is used to represent the comprehensive power generation forecast status of the target city; the multi-dimensional power generation forecast index is used to perform power generation forecasting on the medium- and long-term power generation of the target city, the power generation forecast result is analyzed, and whether the multi-dimensional power generation forecast index is suitable for the medium- and long-term power generation forecast of the target city is evaluated.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-dimensional power generation prediction, and in particular to a multi-dimensional medium- and long-term power generation prediction method. Background Art

[0002] With the growth of energy demand and the adjustment of energy structure, accurate prediction and planning of power generation are crucial for energy planning and management. Medium- and long-term power generation forecasts can provide important reference for energy decision makers, helping them make appropriate energy investment and adjustment decisions to meet future needs.

[0003] Power generation is affected by many factors, such as weather changes, economic development, policy adjustments, etc. Traditional power generation forecasting methods usually only consider a few factors and ignore other important influencing factors. Multi-dimensional medium- and long-term power generation forecasting methods can comprehensively consider multiple factors, improve forecast accuracy and provide comprehensive forecast results.

[0004] With the continuous advancement of data collection and processing technology, more and more relevant data can be used for power generation forecasting. For example, it has become easier to obtain and process weather data, economic data, energy market data, etc., making multi-dimensional medium- and long-term power generation forecasting possible. Based on decades of meteorological resource simulation data of the target city, historical meteorological change characteristics analysis, medium- and long-term numerical meteorological forecasts, historical power generation data, etc., through big data mining and machine learning technology, research and construction of multi-dimensional medium- and long-term power forecasting models for the entire network, region, single station, etc., is expected to output medium- and long-term time-sharing rolling power forecasting data for the entire network, region, and single station, providing more accurate power forecasting data for medium- and long-term trading position decisions, which can effectively reduce the risks caused by too many or too few medium- and long-term trading positions.

[0005] Therefore, in response to the above problems, a multi-dimensional medium- and long-term power generation forecasting method is urgently needed. Summary of the invention

[0006] The present invention provides a multi-dimensional medium- and long-term power generation forecasting method, which solves the problem of unbalanced and inaccurate prediction of discharge capacity due to a single factor, and the instability and large error points of single meteorological forecast source and single economic and quarterly forecasts causing single-point deviation in power generation forecasting of new energy sites.

[0007] In order to solve the above-mentioned invention object, the technical solution provided by the present invention is as follows:

[0008] A multi-dimensional medium- and long-term power generation forecasting method comprises the following steps: based on the analysis of historical power generation data obtained from a power information analysis platform of a target city, a power generation forecast deviation adjustment coefficient is obtained, and the power generation forecast deviation adjustment coefficient is used to numerically represent the deviation between the historical actual power generation and the predicted power generation of the target city; based on the market economy factors, renewable energy factors, seasonal factors and power generation technology factors of the power generation influencing factors of the target city, relevant information on the power generation influencing factors of the target city is obtained, and the relevant information on the power generation influencing factors is combined with the power generation forecast deviation adjustment coefficient analysis to obtain a multi-dimensional power generation forecast index, and the multi-dimensional power generation forecast index is used to represent the comprehensive power generation forecast status of the target city; the multi-dimensional power generation forecast index is used to forecast the medium- and long-term power generation of the target city, and the power generation forecast result is analyzed to evaluate whether the multi-dimensional power generation forecast index is suitable for the medium- and long-term power generation forecast of the target city.

[0009] Optionally, the specific analysis process for obtaining the power generation forecast deviation adjustment coefficient is: based on the target city power information analysis platform, obtaining the historical power generation data of the target city, the historical power generation data specifically including: the power generation forecast value of the target city in the historical quarter, and the actual power generation value of the target city in the historical quarter; by analyzing the power generation forecast value of the target city in the historical quarter and the actual power generation value of the target city in the historical quarter, the power generation forecast deviation adjustment coefficient of the target city is obtained.

[0010] Optionally, the specific calculation formula of the power generation forecast deviation adjustment coefficient of the target city is:

[0011]

[0012] In the formula represents the target city's power generation forecast deviation adjustment coefficient, f represents the historical sampling period number of the power generation forecast deviation adjustment coefficient, f=1,2,3,...,f', f' represents the total number of historical sampling periods of the power generation forecast deviation adjustment coefficient, j represents the quarter number, j=1,2,3,4, α Yfj represents the power generation forecast value of the jth quarter in the historical sampling period of the fth power generation forecast deviation adjustment coefficient, α Sfj It represents the actual value of power generation in the jth quarter of the historical sampling period of the fth power generation forecast deviation adjustment coefficient.

[0013] Optionally, the specific analysis process for obtaining the multidimensional power generation forecast index is: based on the target city's multidimensional medium- and long-term power generation forecast assessment needs, obtaining information related to the target city's power generation influencing factors, the target city's power generation influencing factors specifically include market economic factor information, renewable energy factor information, seasonal factor information, and power generation technology factor information; by analyzing the market economic factor information, renewable energy factor information, seasonal factor information, and power generation technology factor information, respectively, to obtain the market economic impact coefficient, renewable energy impact coefficient, seasonal impact coefficient, and power generation technology impact coefficient; combining the market economic impact coefficient, renewable energy impact coefficient, seasonal impact coefficient and the target city's power generation forecast deviation adjustment coefficient analysis to obtain the multidimensional power generation forecast index.

[0014] Optionally, the specific calculation formula of the multi-dimensional power generation prediction index is:

[0015]

[0016] In the formula represents the multi-dimensional power generation prediction index, β SC represents the market economic impact coefficient, β NY represents the renewable energy impact coefficient, β JJ represents the seasonal influence coefficient, β JS represents the influence coefficient of power generation technology, q1 represents the weight factor of market economy influence coefficient, q2 represents the weight factor of renewable energy influence coefficient, q3 represents the weight factor of seasonal influence coefficient, q4 represents the weight factor of power generation technology influence coefficient, It represents the power generation forecast deviation adjustment coefficient of the target city.

[0017] Optionally, the information related to the market economic factors specifically includes: the fuel price growth rate, the electricity price growth rate, and the GDP growth rate; the information related to the renewable energy factors specifically includes: the wind speed change rate in the next quarter, the failure rate of wind power generation equipment, the rainfall change rate in the next quarter, the failure rate of hydropower generation equipment, the light intensity change rate in the next quarter, and the failure rate of photovoltaic power generation equipment; the information related to the seasonal factors specifically includes: the increase rate of electricity demand in the next quarter; the information related to the power generation technology factors specifically includes: the power generation technology cost reduction rate and the power generation efficiency growth rate.

[0018] Optionally, the market economy impact coefficient is obtained by analyzing the fuel price growth rate, the electricity price growth rate, and the GDP growth rate to obtain the market economy impact coefficient, and the market economy impact coefficient is used to numerically represent the impact of the predicted economic changes and energy market in the target city on power generation; the renewable energy impact coefficient is obtained by analyzing the wind speed change rate in the next quarter, the failure rate of wind power generation equipment, the rainfall change rate in the next quarter, the failure rate of hydropower generation equipment, the light intensity change rate in the next quarter, and the failure rate of photovoltaic power generation equipment to obtain the renewable energy impact coefficient, and the renewable energy impact coefficient is used to numerically represent the impact of the predicted changes in renewable energy in the target city on power generation; the seasonal impact coefficient is obtained by analyzing the increase rate of electricity demand in the next quarter to obtain the seasonal impact coefficient, and the seasonal impact coefficient is used to numerically represent the impact of the predicted changes in electricity demand in the next quarter in the target city on power generation; the power generation technology impact coefficient is obtained by weighted analysis of the power generation technology cost reduction rate and the power generation efficiency growth rate to obtain the power generation technology impact coefficient, and the power generation technology impact coefficient is used to numerically represent the impact of the predicted target city power generation technology on power generation.

[0019] Optionally, the specific analysis process for evaluating whether the multidimensional power generation prediction index is applicable to the medium- and long-term power generation forecast of the target city is: constructing a multidimensional medium- and long-term electricity forecast model for the target city; performing a multidimensional medium- and long-term power generation simulation forecast on the multidimensional medium- and long-term electricity forecast model based on the multidimensional power generation prediction index to obtain a power generation prediction evaluation value; obtaining a power generation prediction evaluation threshold based on the target city's multidimensional medium- and long-term power generation prediction assessment requirements; performing a comparative analysis between the power generation prediction evaluation value and the power generation prediction evaluation threshold; when the power generation prediction evaluation value is greater than or equal to the power generation prediction evaluation threshold, judging that the multidimensional power generation prediction index is applicable to the medium- and long-term power generation forecast of the target city, and using the multidimensional power generation prediction index to predict the multidimensional medium- and long-term power generation of the target city; when the power generation prediction evaluation value is less than the power generation prediction evaluation threshold, judging that the multidimensional power generation prediction index is not applicable to the medium- and long-term power generation forecast of the target city, obtaining the latest data on information related to the target city's power generation influencing factors to optimize and adjust the multidimensional power generation prediction index.

[0020] Optionally, the multi-dimensional medium- and long-term electricity forecasting model for the target city is specifically constructed as follows: obtaining historical meteorological resource simulation data, historical meteorological change characteristics analysis, medium- and long-term numerical meteorological forecasts, and historical power generation data of the target city, and preprocessing the target city's historical meteorological resource simulation data for decades, historical meteorological change characteristics analysis, medium- and long-term numerical meteorological forecasts, and historical power generation data, the preprocessing including processing missing values ​​and outliers; based on the target city's multi-dimensional medium- and long-term power generation forecast assessment needs, using big data mining and machine learning technologies to build a multi-dimensional medium- and long-term electricity forecasting model for the target city.

[0021] Optionally, the specific analysis process for obtaining the power generation forecast evaluation value is: based on the multi-dimensional power generation forecast index, a power generation forecast simulation is performed on the medium- and long-term power generation of the target city; based on the power generation forecast simulation result analysis, the power generation forecast accuracy of each forecast simulation is obtained; by analyzing the power generation forecast accuracy of each forecast simulation, the power generation forecast evaluation value is obtained, and the calculation formula of the power generation forecast evaluation value is:

[0022]

[0023] Where D PG represents the power generation forecast evaluation value, c represents the forecast simulation number of each power generation forecast simulation, c=1,2,3,...,c', c' represents the total number of forecast simulations, η JZc Represents the power generation prediction accuracy of the cth prediction simulation.

[0024] Compared with the prior art, the above technical solution has at least the following beneficial effects:

[0025] (1) The present invention can numerically represent the deviation between historical power generation and predicted power generation by obtaining the power generation forecast deviation adjustment coefficient, thereby helping to improve the accuracy of power generation forecast and making the forecast result closer to the actual situation; by obtaining information related to the power generation influencing factors of the target city and analyzing it in combination with the power generation forecast deviation adjustment coefficient, a multi-dimensional power generation forecast index can be obtained, which means that the forecast is no longer dependent on a single factor, but takes into account the effects of multiple influencing factors, and can provide a more comprehensive and accurate power generation forecast; by using the multi-dimensional power generation forecast index to predict the medium- and long-term power generation of the target city and analyzing the forecast results, the applicability of the index can be evaluated, which helps to verify the reliability and effectiveness of the index and provide a more reliable basis for the medium- and long-term power generation forecast of the target city.

[0026] (2) The present invention adopts data processing technology to predict the power generation of the target city by using big data technology to perform deviation distribution, error probability statistics, and error probability interval statistics of power generation prediction results in different historical periods, fully explore the temporal and spatial variation laws of wind, solar and hydropower generation, and use model parameter optimization, error processing, weight coefficient, error statistics correction, dynamic optimization deep learning combination algorithm and other methods to perform model combination prediction in order to reduce the inherent limitations of the single prediction algorithm. The instability and frequency of large error points in the power generation prediction of a single meteorological forecast source and a single economic energy source are reduced, thereby reducing the single-point deviation of the power generation prediction of new energy sites, improving the accuracy of power generation prediction, considering multi-dimensional influencing factors and comprehensively evaluating the applicability of power generation prediction indexes, thereby providing more reliable prediction results and decision support for the medium- and long-term power generation prediction of the target city. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0028] Figure 1 This is a flow chart of a multi-dimensional medium- and long-term power generation prediction method of the present invention. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] 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 words "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. Similarly, words such as "one", "one" or "the" do not indicate a quantitative limitation, but indicate the existence of at least one. Words such as "include" or "comprise" 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. Words such as "connect" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect.

[0031] It should be noted that the terms "up", "down", "left", "right", "front", "back", etc. used in the present invention are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0032] The present invention provides a multi-dimensional medium- and long-term power generation prediction method to address the problems of unbalanced and inaccurate prediction of discharge capacity due to a single factor, and the instability and large error points in the prediction of power generation of new energy stations caused by a single meteorological forecast source and a single economic and quarterly forecast.

[0033] like Figure 1 As shown, an embodiment of the present invention provides a multi-dimensional medium- and long-term power generation forecasting method, comprising the following steps: based on the historical power generation data obtained by the power information analysis platform of the target city, a power generation forecast deviation adjustment coefficient is obtained, and the power generation forecast deviation adjustment coefficient numerically represents the deviation between the historical actual power generation of the target city and the predicted power generation; based on the market economic factors, renewable energy factors, seasonal factors and power generation technology factors of the target city's power generation influencing factors, relevant information on the power generation influencing factors of the target city is obtained, and the relevant information on the power generation influencing factors is combined with the power generation forecast deviation adjustment coefficient analysis to obtain a multi-dimensional power generation forecast index, and the multi-dimensional power generation forecast index represents the comprehensive power generation forecast status of the target city; the multi-dimensional power generation forecast index is used to perform power generation forecast on the medium- and long-term power generation of the target city, the power generation forecast result is analyzed, and whether the multi-dimensional power generation forecast index is suitable for the medium- and long-term power generation forecast of the target city.

[0034] Specifically, the power generation forecast deviation adjustment coefficient can be obtained through the target city power information analysis platform analysis and can also be obtained through a more accurate calculation method. The specific calculation method is as follows: based on the target city power information analysis platform, the historical power generation data of the target city is obtained. The historical power generation data includes the target city's historical quarter power generation forecast value and the target city's historical quarter power generation actual value; by analyzing the target city's historical quarter power generation forecast value and the target city's historical quarter power generation actual value, the target city's power generation forecast deviation adjustment coefficient is obtained. The specific calculation formula of the target city's power generation forecast deviation adjustment coefficient is:

[0035]

[0036] In the formula It represents the target city's power generation forecast deviation adjustment coefficient, which is used to numerically represent the deviation between the target city's historical actual power generation and the forecast power generation. f represents the historical sampling period number of the power generation forecast deviation adjustment coefficient, f=1,2,3,...,f', f' represents the total number of historical sampling periods of the power generation forecast deviation adjustment coefficient, j represents the quarter number, j=1,2,3,4, α Yfj represents the power generation forecast value of the jth quarter in the historical sampling period of the fth power generation forecast deviation adjustment coefficient, α Sfj It represents the actual value of power generation in the jth quarter of the historical sampling period of the fth power generation forecast deviation adjustment coefficient.

[0037] In this implementation plan, the target city's historical quarter power generation forecast value and the target city's historical quarter power generation actual value are obtained in the following manner: the target city's energy department, power grid management department or relevant energy research institution records and monitors the city's power generation data, including the historical quarter power generation forecast value and actual value; uses statistical models or data analysis methods to process and analyze the historical quarter power generation data, including comparing the forecast value and actual value, calculating deviations, analyzing trends, and calculating the target city's power generation forecast deviation adjustment coefficient, which can identify forecast deviations and trends, and based on the historical forecast deviation adjustment coefficient, can correct the forecast result to make it closer to the actual power generation, which helps to improve the accuracy of power generation forecasts and can be applied in future power generation forecasts to more accurately correct the forecast results and improve the accuracy and reliability of the forecast.

[0038] Specifically, the multidimensional power generation prediction index can not only be obtained by further analysis through the machine learning integration model, using integration methods such as K-means clustering model or support vector machine model, combining the prediction results of multiple basic models to obtain a more accurate multidimensional power generation prediction index, but also can be calculated in the following way. The specific analysis process is: based on the target city's multi-dimensional medium- and long-term power generation forecast assessment needs, the target city's power generation influencing factors related information is obtained, and the target city's power generation influencing factors related information specifically includes market economic factors related information, renewable energy factors related information, seasonal factors related information, and power generation technology factors related information; by analyzing the market economic factors related information, renewable energy factors related information, seasonal factors related information, and power generation technology factors related information respectively, the market economic impact coefficient, renewable energy impact coefficient, seasonal impact coefficient, and power generation technology impact coefficient are obtained; the market economic impact coefficient, renewable energy impact coefficient, seasonal impact coefficient, and power generation technology impact coefficient are combined with the target city's power generation forecast deviation adjustment coefficient analysis to obtain the multidimensional power generation prediction index. The specific calculation formula of the multidimensional power generation prediction index is:

[0039]

[0040] In the formula represents the multi-dimensional power generation prediction index, β SC represents the market economic impact coefficient, β NY represents the renewable energy impact coefficient, β JJ represents the seasonal influence coefficient, β JS represents the influence coefficient of power generation technology, q1 represents the weight factor of market economy influence coefficient, q2 represents the weight factor of renewable energy influence coefficient, q3 represents the weight factor of seasonal influence coefficient, q4 represents the weight factor of power generation technology influence coefficient, It represents the power generation forecast deviation adjustment coefficient of the target city.

[0041] In this implementation plan, taking multiple influencing factors into account in the power generation forecast can obtain more comprehensive and comprehensive forecast results. The market economy impact coefficient can reflect the impact of factors such as power market supply and demand and fuel prices on power generation; the renewable energy impact coefficient can take into account the scale and use of renewable energy; the seasonal impact coefficient can reflect the impact of seasonal changes on power generation, and the power generation technology impact coefficient can reflect the impact of power generation technology changes on power generation. By combining these factors with the power generation forecast deviation adjustment coefficient of the target city, a more comprehensive and accurate power generation forecast index can be obtained; the multi-dimensional power generation forecast index constructed by integrating these factors can more accurately predict future power generation and help power generation departments, power market operators and decision makers make more accurate decisions; the market economy The setting of weight factors corresponding to the impact coefficient, renewable energy impact coefficient, seasonal impact coefficient and power generation technology impact coefficient depends on the specific forecasting needs and data analysis results. First, historical data analysis, regression analysis and time series analysis are used to analyze the market economic impact coefficient, renewable energy impact coefficient and seasonal impact coefficient to understand the actual impact of each influencing factor on power generation. According to the analysis results, the importance of each influencing factor to the power generation forecast is evaluated. Larger influencing factors are assigned higher weights, while smaller influencing factors are assigned lower weights. Decisions are made based on actual data and the experience and judgment of experts in related fields. Experts in related fields evaluate the importance and impact of each influencing factor and provide knowledge and insights on related factors.

[0042] Specifically, information related to market economic factors includes the fuel price growth rate, electricity price growth rate, and GDP growth rate; information related to renewable energy factors includes the wind speed change rate in the next quarter, the failure rate of wind power generation equipment, the rainfall change rate in the next quarter, the failure rate of hydropower generation equipment, the light intensity change rate in the next quarter, and the failure rate of photovoltaic power generation equipment; information related to seasonal factors includes: the increase rate of electricity demand in the next quarter; information related to power generation technology factors includes: the reduction rate of power generation technology costs and the growth rate of power generation efficiency.

[0043] In this implementation plan, the fuel price growth rate refers to the growth rate of fuel prices within a specific period of time; the electricity price growth rate represents the changing trend of electricity prices within a specific period of time; the GDP growth rate represents the growth rate of the gross domestic product of the target city within a specific period of time; the wind speed change rate in the next quarter represents the changing trend of wind energy resources in the next quarter; the wind power generation equipment failure rate: the wind power generation equipment failure rate represents the frequency or probability of wind power generation equipment failure; the rainfall change rate in the next quarter represents the change of rainfall in the next quarter relative to the historical average; the hydropower equipment failure rate represents the frequency or probability of hydropower equipment failure; the light intensity change rate in the next quarter represents the sunlight intensity in the next quarter. Changing trend; The failure rate of photovoltaic power generation equipment indicates the frequency or probability of failure of photovoltaic power generation equipment; The increase rate of electricity demand in the next quarter indicates the growth rate of urban electricity demand in the next quarter; The reduction rate of power generation technology cost indicates the rate of reduction of the cost of unit power generation over time. The cost reduction is due to technological innovation, economies of scale, and accumulation of production experience. A higher cost reduction rate means that the technology cost is reduced faster; The growth rate of power generation efficiency indicates the growth rate of the amount of electrical energy output that can be generated by unit energy input over time. The efficiency growth is due to technological improvement and process optimization. A higher efficiency growth rate means that the same energy input can generate more electricity, thereby improving power generation efficiency.

[0044] Specifically, the market economic impact coefficient can be obtained through the target city power information analysis platform analysis and can also be obtained through a more accurate calculation method. The specific calculation method is as follows: by analyzing the fuel price growth rate, electricity price growth rate, and GDP growth rate, the market economic impact coefficient is obtained. The market economic impact coefficient is used to numerically represent the impact of the predicted target city economic changes and energy market on power generation. The calculation formula of the market economic impact coefficient is:

[0045]

[0046] Where β SC represents the market economic impact coefficient, g represents the sampling period number of the market economic impact coefficient, g=1,2,3,...,g', g' represents the total number of sampling periods of the market economic impact coefficient, χ 1g represents the fuel price growth rate in the sampling period of the market economic impact coefficient of the ggth period, χ 2g represents the growth rate of electricity price in the sampling period of the g-th market economic impact coefficient, χ 3grepresents the GDP growth rate of the g-th market economic impact coefficient sampling period, r1 represents the weight factor of the fuel price growth rate, r2 represents the weight factor of the electricity price growth rate, and r3 represents the weight factor of the GDP growth rate; the renewable energy impact coefficient can be obtained through the target city power information analysis platform analysis and analysis, and can also be obtained through a more accurate calculation method. The specific calculation method is as follows: by analyzing the wind speed change rate in the next quarter, the failure rate of wind power generation equipment, the rainfall change rate in the next quarter, the failure rate of hydropower generation equipment, the light intensity change rate in the next quarter, and the failure rate of photovoltaic power generation equipment, the renewable energy impact coefficient is obtained. The renewable energy impact coefficient is used to numerically represent the impact of the predicted target city's renewable energy changes on power generation. The calculation formula of the renewable energy impact coefficient is:

[0047]

[0048] Where β NY represents the renewable energy impact coefficient, h represents the sampling period number of the renewable energy impact coefficient, h=1,2,3,...,h', h' represents the total number of renewable energy impact coefficient sampling periods, δ F1h represents the wind speed change rate in the next quarter of the hth renewable energy impact coefficient sampling period, δ F2h represents the failure rate of wind power equipment in the hth renewable energy impact coefficient sampling period, δ S1h represents the rainfall change rate in the next quarter of the sampling period of the hth renewable energy impact coefficient, δ S2h represents the failure rate of hydropower equipment in the hth renewable energy impact coefficient sampling period, δ G1h represents the rate of change of light intensity in the next quarter of the sampling period of the hth renewable energy impact coefficient, δ G2h represents the failure rate of photovoltaic power generation equipment in the hth renewable energy impact coefficient sampling period, t1 represents the weight factor of renewable energy meteorological changes, t2 represents the weight factor of renewable energy power generation equipment, k1 represents the weight factor of wind power generation, k2 represents the weight factor of hydropower generation, and k3 represents the weight factor of photovoltaic power generation; the seasonal impact coefficient can be obtained through the target city power information analysis platform analysis and analysis, and can also be obtained through a more accurate calculation method. The specific calculation method is as follows: by analyzing the increase rate of electricity demand in the next quarter, the seasonal impact coefficient is obtained. The seasonal impact coefficient is used to numerically represent the impact of the predicted target city's electricity demand change in the next quarter on power generation. The calculation formula of the seasonal impact coefficient is:

[0049]

[0050] Where β JJrepresents the seasonal influence coefficient, b represents the sampling period number of the seasonal influence coefficient, b=1,2,3,...,b', b' represents the total number of seasonal influence coefficient sampling periods, γ Db Indicates the increase rate of electricity demand in the next quarter of the bth sampling period of the next seasonal impact coefficient; the power generation technology impact coefficient can be obtained through the target city power information analysis platform analysis and analysis, and can also be obtained through a more accurate calculation method. The specific calculation method is as follows: The power generation technology impact coefficient is obtained by weighted analysis of the power generation technology cost reduction rate and the power generation efficiency growth rate. The power generation technology impact coefficient is used to numerically represent the impact of the predicted target city power generation technology on the power generation. The calculation formula of the power generation technology impact coefficient is:

[0051]

[0052] Where β JS represents the power generation technology impact coefficient, p represents the sampling period number of the power generation technology impact coefficient, p=1,2,3,...,p', p' represents the total number of sampling periods of the power generation technology impact coefficient, ω 1p represents the reduction rate of power generation technology cost in the sampling period of the pth power generation technology impact coefficient, ω 2p It represents the growth rate of power generation efficiency in the sampling period of the pth power generation technology influence coefficient, m1 represents the weight factor of the power generation technology cost reduction rate, and m2 represents the weight factor of the power generation efficiency growth rate.

[0053] In this implementation plan, the fuel price growth rate can be obtained through energy market monitoring agencies, energy company reports, or government department data statistics, and the weight factor is determined based on factors such as the proportion of fuel in power generation costs and changes in market supply and demand; the electricity price growth rate is obtained from the energy market, power company reports or government statistics bureaus, and the weight factor can take into account the impact of electricity prices on power generation costs and consumer electricity costs; the GDP growth rate is obtained from official economic statistics or economic research institutions, and the weight factor is set based on the degree of correlation between GDP and urban power generation demand and economic activities; the wind speed change rate in the next quarter is obtained through meteorological data, wind energy resource assessment reports, historical wind speed data, etc., and the weight factor is set based on the proportion of wind power generation in energy production and the impact of wind speed on power generation; the wind power equipment failure rate is obtained through data reports from equipment manufacturers and operators, equipment maintenance records, etc., and the weight factor takes into account the impact of equipment failure on power generation losses; the rainfall change rate in the next quarter is obtained through meteorological data, reports from meteorological agencies, and data from water conservancy departments, and the weight factor is set based on the impact of hydropower generation in The weight factor is set based on the proportion of photovoltaic power generation in energy production and the degree of influence of rainfall on power generation; the failure rate of hydropower equipment is obtained through data reports and equipment maintenance records of equipment manufacturers and operators, and the weight factor takes into account the influence of equipment failure on power generation loss; the rate of change of light intensity in the next quarter is obtained through meteorological data, solar energy resource assessment report and historical light data, and the weight factor is set according to the proportion of photovoltaic power generation in energy production and the degree of influence of light intensity on power generation; the failure rate of photovoltaic power generation equipment is obtained through data reports and equipment maintenance records of equipment manufacturers and operators, and the weight factor takes into account the degree of influence of equipment failure on power generation loss; the increase rate of electricity demand in the next quarter is predicted or estimated through historical electricity consumption data, urban development planning, electricity consumption surveys, etc.; the way to obtain the cost reduction rate of power generation technology and the growth rate of power generation efficiency depends on investigating technology and the market, collecting relevant data and analyzing them. The setting of weight factors depends on the specific situation and research purpose, and the opinions and judgments of experts in related fields are comprehensively considered, and the weight factors are set in combination with their experience to ensure the credibility and accuracy of the research results.

[0054] Specifically, the specific analysis process for evaluating whether the multidimensional power generation forecast index is applicable to the medium- and long-term power generation forecast of the target city is as follows: construct a multidimensional medium- and long-term power forecast model for the target city; perform a multidimensional medium- and long-term power generation simulation forecast on the multidimensional medium- and long-term power forecast model based on the multidimensional power generation forecast index to obtain a power generation forecast evaluation value; obtain a power generation forecast evaluation threshold based on the target city's multidimensional medium- and long-term power generation forecast assessment needs; compare and analyze the power generation forecast evaluation value with the power generation forecast evaluation threshold; when the power generation forecast evaluation value is greater than or equal to the power generation forecast evaluation threshold, the multidimensional power generation forecast index is applicable to the medium- and long-term power generation forecast of the target city, and the multidimensional power generation forecast index is used to predict the multidimensional medium- and long-term power generation of the target city; when the power generation forecast evaluation value is less than the power generation forecast evaluation threshold, the multidimensional power generation forecast index is not applicable to the medium- and long-term power generation forecast of the target city, and the latest data on information related to the target city's power generation influencing factors is obtained to optimize and adjust the multidimensional power generation forecast index.

[0055] In this implementation plan, by evaluating the multi-dimensional medium- and long-term power generation simulation prediction results, a power generation prediction evaluation value is obtained to measure the accuracy of the prediction, which is helpful to evaluate the effect of the current prediction model and provide a basis for subsequent optimization and adjustment. Based on the multi-dimensional medium- and long-term power generation prediction assessment needs of the target city, a power generation prediction evaluation threshold is set. The threshold is used to determine whether the multi-dimensional power generation prediction index is applicable to the medium- and long-term power generation prediction of the target city. By comparing and analyzing the relationship between the power generation prediction evaluation value and the threshold, the feasibility of the prediction model can be determined. When the multi-dimensional power generation prediction index is not applicable to the target city, the power generation prediction index can be optimized and adjusted by obtaining the latest information related to power generation influencing factors. The accuracy and adaptability of the prediction model can be improved, thereby better meeting the medium- and long-term power generation prediction needs of the target city.

[0056] Specifically, the specific construction method of the multi-dimensional medium- and long-term electricity forecasting model for the target city is: obtain the historical meteorological resource simulation data, historical meteorological change characteristics analysis, medium- and long-term numerical meteorological forecasts, and historical power generation data of the target city; pre-process the target city’s historical meteorological resource simulation data for decades, historical meteorological change characteristics analysis, medium- and long-term numerical meteorological forecasts, and historical power generation data. The pre-processing includes processing missing values ​​and outliers; based on the target city’s multi-dimensional medium- and long-term power generation forecast assessment needs, use big data mining and machine learning technology to obtain the target city’s multi-dimensional medium- and long-term electricity forecasting model.

[0057] In this implementation plan, preprocessing helps to improve the accuracy and completeness of the data, thereby improving the reliability and accuracy of the prediction model; using big data mining and machine learning technology, based on the multi-dimensional medium- and long-term power generation forecast assessment needs of the target city, a multi-dimensional medium- and long-term electricity forecast model is constructed, which helps to provide accurate medium- and long-term electricity supply and demand forecasts, and provide a reference for power planning and scheduling. Using the prediction model for multi-dimensional medium- and long-term electricity forecasts can provide more accurate prediction results for the energy planning and scheduling of the target city, help optimize the allocation and utilization of energy resources, improve the reliability and economy of energy supply, and promote the sustainable development of energy.

[0058] Specifically, the power generation forecast evaluation value can not only be obtained by further analysis through the machine learning integrated model, using integrated methods such as K-means clustering model or support vector machine model to combine the evaluation results of multiple basic models to obtain a more accurate power generation forecast evaluation value, but also can be calculated in the following way. The specific analysis process is: based on the multi-dimensional power generation forecast index, the medium- and long-term power generation of the target city is simulated for power generation forecast; based on the analysis of the power generation forecast simulation results, the power generation forecast accuracy of each forecast simulation is obtained; by analyzing the power generation forecast accuracy of each forecast simulation, the power generation forecast evaluation value is obtained. The power generation forecast evaluation value is used to evaluate the prediction effect of the multi-dimensional power generation forecast index on the multi-dimensional medium- and long-term power generation of the target city. The calculation formula of the power generation forecast evaluation value is:

[0059]

[0060] Where D PG represents the power generation forecast evaluation value, c represents the forecast simulation number of each power generation forecast simulation, c=1,2,3,...,c', c' represents the total number of forecast simulations, η JZc Represents the power generation prediction accuracy of the cth prediction simulation.

[0061] In this implementation scheme, the accuracy of power generation forecast can be improved by simulating the medium- and long-term power generation of the target city based on the multi-dimensional power generation forecast index. The multi-dimensional power generation forecast index takes into account historical meteorological data, numerical meteorological forecasts and other relevant indicators, and comprehensively considers the impact of multiple factors on power generation, thereby improving the accuracy of the forecast; analyzing the power generation forecast simulation results and obtaining the power generation forecast accuracy of each forecast simulation can help evaluate the quality of the forecast model and understand the forecast accuracy of the model for different time periods; analyzing the power generation forecast accuracy of each forecast simulation can obtain a power generation forecast evaluation value, and comprehensively evaluating the prediction effect of the multi-dimensional power generation forecast index on the multi-dimensional medium- and long-term power generation of the target city will help understand the stability and reliability of the forecast model and provide a reference for decision making.

[0062] The accuracy of power generation prediction for each prediction simulation can be obtained by comparing the difference between the simulation results and the actual observation data. Various prediction accuracy indicators, such as root mean square error, mean absolute error, and correlation coefficient are used to evaluate the accuracy of the prediction and measure the closeness between the prediction results and the actual situation, so as to obtain the accuracy of power generation prediction for each prediction simulation.

[0063] There are a few points to note:

[0064] (1) The drawings of the embodiments of the present invention only relate to the structures related to the embodiments of the present invention, and other structures may refer to the general design.

[0065] (2) For the sake of clarity, in the drawings used to describe the embodiments of the present invention, the thickness of the layers or regions is exaggerated or reduced, that is, these drawings are not drawn according to the actual scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being "on" or "under" another element, the element may be "directly" "on" or "under" the other element or there may be intermediate elements.

[0066] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to obtain new embodiments.

[0067] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A multi-dimensional medium- and long-term power generation forecasting method, characterized in that: The following steps are involved: Based on the historical power generation data obtained by the power information analysis platform of the target city, the power generation forecast deviation adjustment coefficient is obtained. The power generation forecast deviation adjustment coefficient is used to numerically represent the deviation between the historical actual power generation and the predicted power generation of the target city; Based on the market economic factors, renewable energy factors, seasonal factors and power generation technology factors of the target city's power generation influencing factors, relevant information on the target city's power generation influencing factors is obtained, and the multi-dimensional power generation prediction index is obtained by combining the relevant information on the power generation influencing factors with the power generation forecast deviation adjustment coefficient analysis. The multi-dimensional power generation prediction index is used to represent the comprehensive power generation forecast status of the target city; Use the multi-dimensional power generation forecast index to forecast the medium- and long-term power generation of the target city, analyze the power generation forecast results, and evaluate whether the multi-dimensional power generation forecast index is suitable for the medium- and long-term power generation forecast of the target city; Among them, the power generation forecast deviation adjustment coefficient can be obtained through the target city power information analysis platform analysis and can also be obtained through a more accurate calculation method. The specific acquisition method is as follows: Acquire historical power generation data of the target city based on the target city power information analysis platform, wherein the historical power generation data includes a predicted power generation value of the target city in a historical quarter and an actual power generation value of the target city in a historical quarter; By analyzing the target city's historical quarterly power generation forecast value and the target city's historical quarterly power generation actual value, the target city's power generation forecast deviation adjustment coefficient is obtained; The specific calculation formula of the power generation forecast deviation adjustment coefficient of the target city is: In the formula represents the target city's power generation forecast deviation adjustment coefficient, f represents the historical sampling period number of the power generation forecast deviation adjustment coefficient, f=1,2,3,...,f', f' represents the total number of historical sampling periods of the power generation forecast deviation adjustment coefficient, j represents the quarter number, j=1,2,3,4, α Yfj represents the power generation forecast value of the jth quarter in the historical sampling period of the fth power generation forecast deviation adjustment coefficient, α Sfj represents the actual value of power generation in the jth quarter of the historical sampling period of the fth power generation forecast deviation adjustment coefficient; The specific analysis process of obtaining the multi-dimensional power generation prediction index is as follows: Based on the target city's multi-dimensional medium- and long-term power generation forecast assessment needs, obtain information related to the target city's power generation influencing factors, wherein the target city's power generation influencing factors specifically include information related to market economic factors, information related to renewable energy factors, information related to seasonal factors, and information related to power generation technology factors; By analyzing the information related to market economic factors, renewable energy factors, seasonal factors, and power generation technology factors, the market economic impact coefficient, renewable energy impact coefficient, seasonal impact coefficient, and power generation technology impact coefficient are obtained; Combining the market economy impact coefficient, renewable energy impact coefficient, seasonal impact coefficient and target city's power generation forecast deviation adjustment coefficient analysis, a multi-dimensional power generation forecast index is obtained; The specific calculation formula of the multi-dimensional power generation prediction index is: In the formula represents the multi-dimensional power generation prediction index, β SC represents the market economic impact coefficient, β NY represents the renewable energy impact coefficient, β JJ represents the seasonal influence coefficient, β JS represents the influence coefficient of power generation technology, q1 represents the weight factor of market economy influence coefficient, q2 represents the weight factor of renewable energy influence coefficient, q3 represents the weight factor of seasonal influence coefficient, q4 represents the weight factor of power generation technology influence coefficient, represents the power generation forecast deviation adjustment coefficient of the target city; The market economic factors-related information specifically includes: fuel price growth rate, electricity price growth rate, GDP growth rate; The relevant information of the renewable energy factors specifically includes: the wind speed change rate in the next quarter, the failure rate of wind power generation equipment, the rainfall change rate in the next quarter, the failure rate of hydropower generation equipment, the light intensity change rate in the next quarter, and the failure rate of photovoltaic power generation equipment; The seasonal factor related information specifically includes: the rate of increase in electricity demand in the next quarter; The relevant information of the power generation technology factors specifically includes: power generation technology cost reduction rate, power generation efficiency growth rate; The specific method for obtaining the market economic impact coefficient is as follows: By analyzing the fuel price growth rate, electricity price growth rate, and GDP growth rate, the market economic impact coefficient is obtained, which is used to numerically represent the impact of the predicted target city's economic changes and the energy market on power generation; The specific method for obtaining the renewable energy impact coefficient is as follows: By analyzing the wind speed change rate in the next quarter, the failure rate of wind power generation equipment, the rainfall change rate in the next quarter, the failure rate of hydropower generation equipment, the light intensity change rate in the next quarter, and the failure rate of photovoltaic power generation equipment, the renewable energy impact coefficient is obtained. The renewable energy impact coefficient is used to numerically represent the impact of the predicted renewable energy changes on the power generation in the target city; The specific method for obtaining the seasonal influence coefficient is as follows: By analyzing the increase rate of electricity demand in the next quarter, a seasonal impact coefficient is obtained, and the seasonal impact coefficient is used to numerically represent the impact of the change in electricity demand in the next quarter on the power generation of the predicted target city; The specific method for obtaining the power generation technology influence coefficient is as follows: By performing a weighted analysis on the reduction rate of power generation technology cost and the growth rate of power generation efficiency, the power generation technology impact coefficient is obtained, and the power generation technology impact coefficient is used to numerically represent the impact of the power generation technology on the power generation of the predicted target city; The specific analysis process for evaluating whether the multi-dimensional power generation forecast index is applicable to the medium- and long-term power generation forecast of the target city is as follows: Construct a multi-dimensional medium- and long-term electricity forecasting model for the target city; Based on the multi-dimensional power generation prediction index, a multi-dimensional medium- and long-term power generation simulation prediction is performed on the multi-dimensional medium- and long-term power generation prediction model to obtain a power generation prediction evaluation value; Based on the multi-dimensional medium- and long-term power generation forecast assessment needs of the target city, the power generation forecast assessment threshold is obtained; Compare and analyze the power generation forecast assessment value with the power generation forecast assessment threshold. When the power generation forecast assessment value is greater than or equal to the power generation forecast assessment threshold, determine that the multi-dimensional power generation forecast index is suitable for the medium- and long-term power generation forecast of the target city, and use the multi-dimensional power generation forecast index to predict the multi-dimensional medium- and long-term power generation of the target city. When the power generation forecast assessment value is less than the power generation forecast assessment threshold, it is determined that the multi-dimensional power generation forecast index is not suitable for the medium- and long-term power generation forecast of the target city, and the latest data on the power generation influencing factors of the target city are obtained to optimize and adjust the multi-dimensional power generation forecast index; The specific construction method of the multi-dimensional medium- and long-term electricity forecasting model for the target city is as follows: Obtain the historical meteorological resource simulation data, historical meteorological change characteristics analysis, medium- and long-term numerical meteorological forecasts, and historical power generation data of the target city, and preprocess the historical meteorological resource simulation data, historical meteorological change characteristics analysis, medium- and long-term numerical meteorological forecasts, and historical power generation data of the target city for decades, including processing missing values ​​and outliers; Based on the target city's multi-dimensional medium- and long-term power generation forecast assessment needs, a multi-dimensional medium- and long-term power forecast model for the target city is built using big data mining and machine learning technology; The specific analysis process for obtaining the power generation prediction evaluation value is as follows: Conduct power generation forecast simulation for the target city’s medium- and long-term power generation based on a multi-dimensional power generation forecast index; The power generation prediction accuracy of each prediction simulation is obtained based on the analysis of power generation prediction simulation results; By analyzing the power generation prediction accuracy of each prediction simulation, the power generation prediction evaluation value is obtained. The calculation formula of the power generation prediction evaluation value is: Where D PG represents the power generation forecast evaluation value, c represents the forecast simulation number of each power generation forecast simulation, c=1,2,3,...,c', c' represents the total number of forecast simulations, η JZc Represents the power generation prediction accuracy of the cth prediction simulation.

Images