Electricity consumption prediction optimization method, system and equipment based on significant event influence

Abstract

The power consumption prediction optimization method based on the influence of the major events comprises the steps of selecting a first time period influenced by the major events, obtaining daily power consumption of each date in the first time period, selecting a second time period not influenced by the major events, selecting typical days of normal periods corresponding to each date in the first time period in the second time period, acquiring daily electricity consumption of typical days in each normal period; constructing a major event influence index model, and calculating a major event influence index according to the major event influence index model and the daily electricity consumption; calculating an influence factor according to the important event influence index and the daily electricity consumption; and correcting the electricity consumption predicted value and the electricity consumption acceleration predicted value according to the influence factors. According to the design, the predicted values of the electricity consumption and the speed increase in a period of time in the future can be dynamically corrected based on the influence of major events on daily electricity consumption, and data support is provided for formulating a future electricity consumption plan target.

Description

Technical Field

[0001] The present invention relates to the technical field of electricity consumption prediction, and in particular to an electricity consumption prediction optimization method, system and device based on the impact of major events, which are mainly used to improve the accuracy of electricity consumption prediction and provide a reference for the prediction of electricity consumption growth rate under the influence of major events in the future. Background Art

[0002] At present, the total electricity consumption of the whole society in the region is composed of the electricity consumption of the whole industry / sector and the electricity consumption of residents. Among them, the electricity consumption of the whole industry / sector can be divided into the electricity consumption of the primary industry, the secondary industry, and the tertiary industry according to the industry; according to the industry / sector, it can be divided into 20 categories and 97 major categories such as agriculture, forestry, animal husbandry and fishery, industry, and construction. When a major event occurs in the region (such as an epidemic, a game, or a major meeting), the daily electricity consumption of the relevant industries, industries, and sectors in the region will change significantly compared with the typical daily electricity consumption in normal times of the year. Therefore, it is necessary to construct an electricity consumption forecasting optimization method based on the impact of major events, analyze the trend of electricity consumption changes in different industries or industries / sectors in the region under the influence of major events, and quantitatively analyze the impact of major events on the decrease and increase of electricity consumption in the region and sub-industries, industries / sectors, and then revise the forecast value and planned target value of annual (quarterly, monthly) electricity consumption. In addition, decoupling the electricity consumption affected by major events from the regional electricity consumption is conducive to providing a reference for the prediction of electricity consumption growth rate under the influence of major events in the future.

[0003] At present, in the research of related fields, the analysis and evaluation or forecasting research work is mainly carried out based on the monthly electricity consumption data, and the daily electricity consumption data of different industries and industrial / sectoral regions are not used to carry out analysis, evaluation and forecasting work, and the degree of research refinement needs to be further improved; in related studies, the model is mainly constructed based on the daily electricity consumption data of customers, and the influence of factors such as temperature changes and holidays on electricity consumption changes is not eliminated, which will lead to certain deviations in the forecasting analysis conclusions; in related studies, more emphasis is placed on the research of forecasting methods, while this method mainly focuses on the correction of forecasting targets, which makes up for the shortcomings of forecasting methods to a certain extent. Summary of the invention

[0004] The purpose of the present invention is to overcome the problem that the existing electricity consumption prediction method in the prior art does not take into account the impact of major events on the electricity consumption trend of various industries / sectors, and to provide an electricity consumption prediction optimization method, system and device based on the impact of major events.

[0005] To achieve the above objectives, the technical solution of the present invention is:

[0006] The power consumption prediction optimization method based on the impact of major events includes the following steps:

[0007] S1. Select a time period affected by a major event as a first time period, obtain the daily power consumption of each date in the first time period, select a time period not affected by the major event as a second time period, select a normal period typical day corresponding to each date in the first time period in the second time period, and obtain the daily power consumption of each normal period typical day;

[0008] S2. construct a major event impact index model, and calculate the major event impact index based on the major event impact index model, the daily power consumption on each date in the first time period, and the daily power consumption on typical days in each normal period;

[0009] S3. Calculate the impact factor based on the major event impact index, the daily power consumption on each date in the first time period, and the daily power consumption on a typical day in each normal period;

[0010] S4. Modify the electricity consumption forecast value and electricity consumption growth rate forecast value according to the influencing factors.

[0011] The first time period is the entire time since the occurrence of the major event in the current year / quarter / month, and the length of the first time period is not less than 3 days;

[0012] The second time period is one year before the major event occurs.

[0013] In step S1, selecting the typical day of the normal period corresponding to the m-th day i in the first time period in the second time period specifically includes:

[0014] S101, to eliminate the impact of holidays, filter out dates with the same holiday attributes as i in the second time period;

[0015] S102. To eliminate the influence of temperature, select a day j with the closest temperature to i from among the dates with the same holiday attribute as i, where j is a typical day in the normal period corresponding to the m-th day i in the first time period.

[0016] In step S102, selecting a day j with the temperature closest to i from the dates with the same holiday attribute as i specifically includes:

[0017] Calculate the temperature difference between each date with the same holiday attribute as i and i:

[0018]

[0019] In formula (1), is the temperature difference between the nth day and the ith day on the date with the same holiday attribute as i, is the highest temperature on the nth day of the date with the same holiday attribute as i, is the maximum temperature of i, is the lowest temperature on the nth day of the date with the same holiday attribute as i, is the lowest temperature of i;

[0020] Select the day j with the smallest temperature difference from i among all the dates with the same holiday attribute as i;

[0021]

[0022] In formula (2), is the nth day among the dates with the same holiday attribute as i, and d is the number of dates with the same holiday attribute as i.

[0023] In step S2, a major event impact index model is constructed, and the major event impact index is calculated according to the major event impact index model, the daily power consumption of each date in the first time period, and the daily power consumption of typical days in each normal period, specifically including:

[0024] Constructing a major event impact index model:

[0025]

[0026] In formula (3), EF m is the major event impact index corresponding to the mth day in the first time period, is the daily electricity consumption on the mth day in the first time period, is the daily electricity consumption of a typical day in a normal period corresponding to the mth day in the first time period;

[0027] The major event impact index corresponding to each date in the first time period is calculated according to formula (3).

[0028] In step S3, calculating the impact factor according to the major event impact index, the daily power consumption on each date in the first time period, and the daily power consumption on each typical day in the normal period specifically includes:

[0029] The impact factor of major events on each industry / sector is calculated according to the following formula:

[0030]

[0031] In formula (4), ECY x is the impact factor of major events on industry / sector x, p is the number of days in the first time period, is the daily electricity consumption of industry x on the mth day in the first time period, STAT is the daily electricity consumption of a typical day in a normal period corresponding to the mth day in the first time period of the x industry / sector. m is the influence coefficient corresponding to the mth day in the first time period;

[0032] STAT mThe value of is determined according to the following conditions:

[0033] If the following three conditions A, B, and C are met at the same time, STAT m = -1;

[0034] A, m∈[p1,p2];

[0035] B. p2-p1≥k;

[0036] C.

[0037] If the following three conditions D, E, and F are met at the same time, STAT m =1;

[0038] D, m∈[p3,p4];

[0039] E, p3-p4 ≥ k;

[0040] F.

[0041] If conditions A, B, and C or conditions D, E, and F cannot be met at the same time, STAT m =0;

[0042] The k is a set time threshold, k≥2.

[0043] In step S4, the correction of the power consumption forecast value and the power consumption growth rate forecast value according to the influencing factor specifically includes:

[0044] S401. Obtain the electricity consumption forecast value TYD of industry / sector x in this year / quarter / month x , the predicted growth rate of electricity consumption of x industry / sector in this year / quarter / month TYDSP x ;

[0045] ECY factor of impact on industry / sector x by major events x The electricity consumption forecast value TYD for x industry / sector in this year / quarter / month x , the predicted growth rate of electricity consumption of x industry / sector in this year / quarter / month TYDSP x To make corrections:

[0046] TYDX x =TYD x +ECY x (5);

[0047]

[0048] In formulas (5)-(6), TYDX xTYDSPX is the revised value of the electricity consumption forecast for the x industry / sector in this year / quarter / month. x is the revised value of the forecast value of electricity consumption growth rate of industry / sector x in this year / quarter / month, The electricity consumption of x industry / sector in the previous year / quarter / month;

[0049] S402. Calculate the total electricity consumption TYDX of this year / quarter / month and the total electricity consumption growth rate TYDSPX of this year / quarter / month based on the revised value of the electricity consumption forecast value of each industry / sector in this year / quarter / month and the revised value of the electricity consumption growth rate forecast value of each industry / sector in this year / quarter / month:

[0050]

[0051] TYDSPX=(TYDX-LYYD) / LYYD (8);

[0052] In formulas (7)-(8), l is the total number of industries / sectors, TYDX x LYYD is the revised value of the forecast electricity consumption of industry / sector x in this year / quarter / month, and LYYD is the total electricity consumption of each industry / sector in the previous year / quarter / month.

[0053] A power consumption forecast value correction system based on the impact of major events, the system is used to execute the above-mentioned power consumption forecast optimization method based on the impact of major events, specifically comprising:

[0054] A data collection module is used to select a time period affected by a major event as a first time period, wherein the first time period is the entire time since the occurrence of the major event in the current year / quarter / month, and the length of the first time period is not less than 3 days, and the daily electricity consumption of each date in the first time period is obtained. A time period not affected by the major event is selected as a second time period, wherein the second time period is one year before the occurrence of the major event, and within the second time period, typical days of a normal period corresponding to each date in the first time period are selected, and the daily electricity consumption of each typical day of the normal period is obtained.

[0055] The daily power consumption of a typical day in a normal period corresponding to the mth day i in the first time period is calculated according to the following method:

[0056] S101, to eliminate the impact of holidays, filter out dates with the same holiday attributes as i in the second time period;

[0057] S102: To eliminate the influence of temperature, select a day j with the closest temperature to i from among the dates with the same holiday attribute as i:

[0058] Calculate the temperature difference between each date with the same holiday attribute as i and i:

[0059]

[0060] In formula (1), is the temperature difference between day n and day i on the date with the same holiday attribute as day i; is the highest temperature on the nth day of the date with the same holiday attribute as i, is the maximum temperature of i, is the lowest temperature on the nth day of the date with the same holiday attribute as i, is the lowest temperature of i;

[0061] Select the day j with the smallest temperature difference from i among all the dates with the same holiday attribute as i;

[0062]

[0063] In formula (2), is the nth day among the dates with the same holiday attribute as i, and d is the number of dates with the same holiday attribute as i;

[0064] j is the typical day of the normal period corresponding to the m-th day i in the first time period.

[0065] Daily electricity consumption of j The calculation is performed according to the following formula:

[0066]

[0067] In formula (9), is the daily electricity consumption on the nth day of the date with the same holiday attribute as i, d is the number of dates with the same holiday attribute as i, and the daily electricity consumption on the nth day of the date with the same holiday attribute as i is directly obtained from the power system operation data.

[0068] In the data acquired by the data collection module, the daily electricity consumption includes the daily electricity consumption of each industry / sector and the daily electricity consumption of the entire industry / sector.

[0069] Model building module, used to build a major event impact index model;

[0070] The major event impact index model is as follows:

[0071]

[0072] In formula (3), EF m is the major event impact index corresponding to the mth day in the first time period, is the daily electricity consumption on the mth day in the first time period, It is the daily electricity consumption of a typical day in the normal period corresponding to the mth day in the first time period.

[0073] The major event impact index calculation module is used to calculate the major event impact index based on the major event impact index model, the daily power consumption of each date in the first time period, and the daily power consumption of typical days in each normal period. Among them, the major event impact index can be calculated by the daily power consumption of the corresponding industry / sector, or by the total daily power consumption.

[0074] The impact factor calculation module is used to calculate the impact factor according to the impact index of the major event, the daily power consumption of each date in the first time period, and the daily power consumption of each typical day in the normal period. The impact factor is calculated according to the impact index of the major event, the daily power consumption of each date in the first time period, and the daily power consumption of each typical day in the normal period. Specifically, the impact factor is calculated according to the impact index of the major event, the daily power consumption of each date in the first time period, and the daily power consumption of each typical day in the normal period.

[0075]

[0076] In formula (4), ECY x is the impact factor of major events on industry / sector x, p is the number of days in the first time period, is the daily electricity consumption of industry / sector x on the mth day in the first time period, STAT is the daily electricity consumption of a typical day in a normal period corresponding to the mth day in the first time period of the x industry / sector. m is the influence coefficient corresponding to the mth day in the first time period;

[0077] STAT m The value of is determined according to the following conditions:

[0078] If the following three conditions A, B, and C are met at the same time, STAT m = -1;

[0079] A, m∈[p1,p2];

[0080] B. p2-p1≥k;

[0081] C.

[0082] If the following three conditions D, E, and F are met at the same time, STAT m =1;

[0083] D, m∈[p3,p4];

[0084] E, p3-p4 ≥ k;

[0085] F.

[0086] If conditions A, B, and C or conditions D, E, and F cannot be met at the same time, STAT m =0;

[0087] The k is a set time threshold, k≥2.

[0088] The correction module is used to correct the power consumption forecast value and the power consumption growth rate forecast value according to the influencing factors. Correcting the power consumption forecast value and the power consumption growth rate forecast value according to the influencing factors specifically includes:

[0089] S401, ECY, the impact factor of major events on industry / sector x x The electricity consumption forecast value TYD for x industry / sector in this year / quarter / month x , the predicted growth rate of electricity consumption of x industry / sector in this year / quarter / month TYDSP x To make corrections:

[0090] TYDX x =TYD x +ECY x (5);

[0091]

[0092] In formulas (5)-(6), TYDX x TYDSPX is the revised value of the electricity consumption forecast for the x industry / sector in this year / quarter / month. x is the revised value of the forecast value of electricity consumption growth rate of industry / sector x in this year / quarter / month, The electricity consumption of x industry / sector in the previous year / quarter / month;

[0093] S402. Calculate the total electricity consumption TYDX of this year / quarter / month and the total electricity consumption growth rate TYDSPX of this year / quarter / month based on the revised value of the electricity consumption forecast value of each industry / sector in this year / quarter / month and the revised value of the electricity consumption growth rate forecast value of each industry / sector in this year / quarter / month:

[0094]

[0095] TYDSPX=(TYDX-LYYD) / LYYD (8);

[0096] In formulas (7)-(8), l is the total number of industries / sectors, TYDX x LYYD is the revised value of the forecast electricity consumption of industry / sector x in this year / quarter / month, and LYYD is the total electricity consumption of each industry / sector in the previous year / quarter / month.

[0097] A device for predicting and optimizing power consumption based on the impact of major events, comprising a memory and a processor, wherein the memory is used to store computer program codes and transmit the computer program codes to the processor;

[0098] The processor is used to execute the above-mentioned electricity consumption prediction optimization method based on the impact of major events according to the instructions in the computer program code.

[0099] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the above-mentioned method for optimizing electricity consumption prediction based on the impact of major events.

[0100] Compared with the prior art, the present invention has the following beneficial effects:

[0101] 1. The present invention constructs a major event impact index model in the power consumption forecast optimization method based on the impact of major events, and calculates the major event impact index through the major event impact index model, the daily power consumption of each date in the time period affected by the major event, and the daily power consumption of each date in the normal period corresponding to each date in the time period affected by the major event. The impact factor is calculated according to the major event impact index, the daily power consumption of each date in the time period affected by the major event, and the daily power consumption of each date in the normal period corresponding to each date in the time period affected by the major event. The power consumption forecast value and the power consumption growth rate forecast value are corrected according to the impact factor, which can make up for the deficiency of the existing power consumption forecast method that does not consider the sudden major event impact factor. Therefore, in this design, it is judged whether the power consumption is affected by the major event based on the daily power consumption, and the current power consumption forecast value and the power consumption growth rate forecast value are corrected. The corrected power consumption forecast value can more accurately reflect the trend of power consumption changes, which makes up for the deficiency of the existing forecast method to a certain extent, and can also provide data support for formulating future power consumption plan targets.

[0102] 2. In the electricity consumption prediction optimization method based on the impact of major events of the present invention, since major events may have different impacts on different industries / sectors, the method of calculating by industry / sector is adopted when calculating the impact factor, and the impact factor of major events on each industry / sector is calculated, and then the electricity consumption forecast value of the corresponding industry / sector is corrected by the impact factor. Finally, the total industry / sector electricity consumption and the growth rate of the total industry / sector electricity consumption in the region are calculated according to the corrected value of the electricity consumption forecast value of different industries / sectors. Calculation and correction by industry / sector can obtain more accurate prediction data, and the change trend of electricity consumption of different industries or industries / sectors under the influence of major events can be obtained. Therefore, in this design, the impact factor is calculated by distinguishing industries / sectors, and the change trend of electricity consumption of different industries or industries / sectors under the influence of major events can be obtained, and the forecast data of the growth rate of electricity consumption of the whole industry / sector and the whole industry / sector electricity consumption can also be accurately obtained.

[0103] 3. In the power consumption prediction optimization method based on the impact of major events of the present invention, when selecting a typical day in the normal period corresponding to any date in the time period affected by the major event, the holiday attributes and temperature difference are considered, and the impact of holidays and temperature is eliminated in the calculation process, and the calculation result is more accurate. Therefore, the impact of holidays and temperature on daily power consumption is eliminated in this design, and the calculation result is accurate.

[0104] 4. The power consumption forecast optimization method, system and device based on the influence of major events of the present invention provide a major event impact index model for correcting the power consumption forecast value, which can reflect the degree of influence of major events on regional power consumption from the perspective of daily power consumption changes. The model is simple, highly operable, and has higher precision in data analysis. At the same time, the power consumption forecast value correction system and device provided in this design can automatically and dynamically correct the power consumption forecast results during the duration of the major event, making the forecast results more accurate and reliable. Therefore, in this design, the power consumption forecast value correction system and device can automatically and dynamically correct the power consumption forecast results. BRIEF DESCRIPTION OF THE DRAWINGS

[0105] Figure 1 It is a flow chart of the power consumption prediction optimization method based on the impact of major events of the present invention.

[0106] Figure 2 It is a flow chart of annual (quarterly, monthly) forecast electricity consumption and its growth rate value correction.

[0107] Figure 3 It is a structural diagram of the electricity consumption forecast value correction system based on the impact of major events.

[0108] Figure 4 It is a structural diagram of the power consumption prediction and optimization equipment based on the impact of major events. DETAILED DESCRIPTION

[0109] The present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0110] Embodiment 1:

[0111] The present invention is a method, system and device for optimizing power consumption prediction for a certain region. The changes in the daily power consumption of related industries, industries / sectors in a certain region during a major event compared with the typical daily power consumption in normal periods of normal years are used to construct a major event impact index model based on the change in daily power consumption, calculate the impact factor of the major event on the change in power consumption of different industries / sectors in the region, and calculate the impact of the major event on the annual / quarterly / monthly power consumption and its growth rate of different industries, industries / sectors in the region according to the impact factor, and then correct the annual / quarterly / monthly total power consumption forecast value and the total power consumption growth rate forecast value of all industries / sectors in the region, which can more accurately predict the annual / quarterly / monthly power consumption of the region in that year, and evaluate the impact of major events on power consumption, so as to provide a reference for the future power demand forecasting work under the influence of major events, and further improve the accuracy of analysis and prediction. In the power consumption prediction optimization method based on the impact of major events, the daily power consumption is the daily power consumption of a selected region, and this selected region can be one or several areas / cities / provinces.

[0112] See also Figure 1 ,The power consumption prediction optimization method based on the impact of major events includes the following steps:

[0113] S1. Based on the selected region, a time period affected by a major event is selected as a first time period, and the daily power consumption of each date in the first time period is obtained. A time period not affected by a major event is selected as a second time period, and a typical day of a normal period corresponding to each date in the first time period is selected in the second time period, and the daily power consumption of each typical day of a normal period is obtained. The daily power consumption is the daily power consumption of the selected region, and the daily power consumption includes the daily power consumption of each industry / sector, and the total daily power consumption of all industries / sectors.

[0114] In step S1, the first time period is the entire time since the occurrence of the major event in the current year / quarter / month, that is, the entire time after the major event occurs in the current year / quarter / month and before the current date. The length of the first time period is not less than 3 days; the second time period is the one year before the major event occurs.

[0115] When selecting the first time period, if it is necessary to revise the forecast electricity consumption and the forecast electricity consumption growth rate for this month, the first time period shall be the entire time since the major event occurred in this month; if it is necessary to revise the forecast electricity consumption and the forecast electricity consumption growth rate for this quarter, the first time period shall be the entire time since the major event occurred in this quarter; if it is necessary to revise the forecast electricity consumption and the forecast electricity consumption growth rate for this year, the first time period shall be the entire time since the major event occurred in this year.

[0116] In step S1, in order to eliminate the impact of holidays and temperature and facilitate the subsequent calculation of the major event impact index EF i The typical day of the normal period corresponding to any day in the first time period is the day with the same holiday attribute as the day in the second time period, and the day with the temperature closest to the day. The typical day of the normal period corresponding to the mth day i in the first time period is selected in the second time period, specifically including:

[0117] S101, to eliminate the impact of holidays, filter out dates with the same holiday attributes as i in the second time period;

[0118] S102: To eliminate the influence of temperature, based on the temperature of the selected area, select a day j with the temperature closest to i from the dates with the same holiday attribute as i, specifically including:

[0119] Calculate the temperature difference between each date with the same holiday attribute as i and i:

[0120]

[0121] In formula (1), is the temperature difference between day n and day i on the date with the same holiday attribute as day i; is the highest temperature on the nth day of the date with the same holiday attribute as i, is the maximum temperature of i, is the lowest temperature on the nth day of the date with the same holiday attribute as i, is the lowest temperature of i. The highest temperature and the lowest temperature are the temperatures of the selected area.

[0122] Select the day j with the smallest temperature difference from i among all the dates with the same holiday attribute as i;

[0123]

[0124] In formula (2), is the nth day among the dates with the same holiday attribute as i, d is the number of dates with the same holiday attribute as i, and j is the typical day in the normal period corresponding to the mth day i in the first time period.

[0125] In order to find the day with the temperature closest to the i-th day during the period of major events, we calculate it by formula (1): The value of By taking the smallest value as the target for screening, we can find the day j with the temperature closest to the i-th day during the period affected by the major event.

[0126] The typical day j in the normal period corresponding to the mth day i in the first time period meets the following conditions;

[0127]

[0128] In formula (10), is the holiday attribute of the typical day j in the normal period corresponding to the m-th day i in the first time period, is the holiday attribute of the mth day i in the first time period. From formula (1), it can be seen that the holiday attribute of the typical day j in the normal period corresponding to the mth day i in the first time period is the same as the holiday attribute of the mth day i in the first time period. For example, when a certain day is a holiday, the holiday attribute of the day is 0, when a certain day is not a holiday, the holiday attribute of the day is 1, and when the holiday attributes of two dates are equal, the holiday attributes of the two dates are the same.

[0129] The daily electricity consumption of a typical day j in the normal period corresponding to the mth day i in the first time period The calculation is performed according to the following formula:

[0130]

[0131] In formula (9), is the daily electricity consumption on the nth day of the date with the same holiday attribute as i, d is the number of dates with the same holiday attribute as i, and the daily electricity consumption on the nth day of the date with the same holiday attribute as i is obtained from the power system operation data. From formula (9), we can see that It is the daily electricity consumption of the day with the smallest temperature difference from i among the days with the same holiday attribute as i.

[0132] S2. Construct a major event impact index model, and calculate the major event impact index based on the major event impact index model, the daily electricity consumption on each date in the first time period, and the daily electricity consumption on typical days in each normal period.

[0133] In step S2, the major event impact index model is as follows:

[0134]

[0135] In formula (3), EF m is the major event impact index corresponding to the mth day in the first time period, is the daily electricity consumption on the mth day in the first time period, is the daily electricity consumption of a typical day in a normal period corresponding to the mth day in the first time period;

[0136] In formula (3), It can be the total daily electricity consumption of the entire industry / sector, that is, the sum of the daily electricity consumption of each industry / sector. It can also be based on the daily electricity consumption of a certain industry / sector, such as the daily industrial electricity consumption. and The calculation is based on the total daily electricity consumption of the entire industry / sector.

[0137] According to formula (3) and the acquired daily electricity consumption data, the major event impact index corresponding to each date in the first time period is calculated.

[0138] In step S3, calculating the impact factor according to the major event impact index, the daily power consumption on each date in the first time period, and the daily power consumption on each typical day in the normal period specifically includes:

[0139] According to formula (4), calculate the impact factor of major events on each industry / sector:

[0140]

[0141] In formula (4), ECY x is the impact factor of major events on industry / sector x, p is the number of days in the first time period, is the daily electricity consumption of industry / sector x on the mth day in the first time period, STAT is the daily electricity consumption of a typical day in a normal period corresponding to the mth day in the first time period of the x industry / sector. m is the influence coefficient corresponding to the mth day in the first time period;

[0142] STAT m The value of is determined according to the following conditions:

[0143] If the following three conditions A, B, and C are met at the same time, STAT m = -1;

[0144] A, m∈[p1,p2], p1, p2 are both positive integers greater than 0;

[0145] B. p2-p1≥k;

[0146] C.

[0147] If the following three conditions D, E, and F are met at the same time, STAT m =1;

[0148] D, m∈[p3,p4], p3, p4 are both positive integers greater than 0;

[0149] E, p3-p4 ≥ k;

[0150] F.

[0151] If conditions A, B, and C or conditions D, E, and F cannot be met at the same time, STAT m =0;

[0152] The k is a set time threshold, k≥2.

[0153] By STAT m From the value determination conditions, it can be seen that if the major event impact index corresponding to each day from the p1th day to the p2th day in the first time period is greater than 100, it means that the power consumption from the p1th day to the p2th day in the first time period is negatively affected by the major event, which will lower the power consumption forecast value and the power consumption growth rate. It is necessary to correct the power consumption forecast value and the power consumption growth rate forecast value. At this time, the value of the impact coefficient corresponding to each day from the p1th day to the p2th day in the first time period is -1. If the major event impact index corresponding to each day from the p3th day to the p4th day in the first time period is less than 100, it means that the power consumption from the p3th day to the p4th day in the first time period is positively affected by the major event, which will increase the power consumption forecast value and the power consumption growth rate. It is necessary to correct the power consumption forecast value and the power consumption growth rate forecast value. At this time, the value of the impact coefficient corresponding to each day from the p3th day to the p4th day in the first time period is 1. Other situations other than the above two situations are regarded as fluctuations. There is no need to correct the electricity consumption forecast value and the electricity consumption growth rate forecast value, and the corresponding impact coefficient is 0.

[0154] Since major events have different impacts on different industries / sectors, if ECY x If ECY is significantly greater than zero, it means that the major event has a positive effect on the electricity consumption of industry / sector x; x If it is significantly less than zero, it means that the major event has a negative impact on the electricity consumption of x industry / sector. Among them, x industry / sector can be industries such as automobiles, building materials, steel, etc., or industries such as industry and service.

[0155] S4. Based on the selected region, the electricity consumption forecast value and the electricity consumption growth forecast value are revised according to the influencing factors, including:

[0156] S401, ECY, the impact factor of major events on industry / sector x x The electricity consumption forecast value TYD for x industry / sector in this year / quarter / month x, the predicted growth rate of electricity consumption of x industry / sector in this year / quarter / month TYDSP x To make corrections:

[0157] TYDX x =TYD x +ECY x (5);

[0158]

[0159] In formulas (5)-(6), TYDX x TYDSPX is the revised value of the electricity consumption forecast for the x industry / sector in this year / quarter / month. x is the revised value of the forecast value of electricity consumption growth rate of industry / sector x in this year / quarter / month, is the electricity consumption of x industry / sector in the previous year / quarter / month, TYD x TYDX x , The unit is billion kilowatt-hours.

[0160] In step S401, the monthly / quarterly / annual electricity consumption forecast value TYD of industry / sector x is x According to the historical annual growth trend of electricity consumption in the industry, the forecast is obtained by combining the time series method and the multivariate linear regression method. This forecast method is a common forecast method and will not be described here one by one. Monthly / quarterly / annual electricity consumption growth rate forecast value of x industry / sector TYDSP x Monthly / quarterly / annual electricity consumption forecast value TYD based on x industry / sector x The electricity consumption of the previous month / quarter / year is calculated and obtained by querying the power system operation data or through prediction.

[0161] In step S401, if it is necessary to revise the electricity consumption forecast value and the electricity consumption growth rate forecast value for this month, then TYD x TYDSP is the electricity consumption forecast for industry / sector x this month. x TYDX is the predicted growth rate of electricity consumption of industry / sector x in this month. x is the revised value of the forecast value of electricity consumption of industry / sector x this month, TYDSPXx is the revised value of the forecast value of electricity consumption growth rate of industry / sector x this month, is the electricity consumption of industry / sector x in the previous month; if the electricity consumption forecast value and electricity consumption growth rate forecast value of this quarter need to be revised, then TYD x TYDSP is the electricity consumption forecast for industry / sector x in this quarter. xTYDX is the predicted growth rate of electricity consumption in the x industry / sector in this quarter. x TYDSPX is the revised value of the electricity consumption forecast for industry / sector x in this quarter. x is the revised value of the forecast value of electricity consumption growth rate of industry / sector x in this quarter, is the electricity consumption of industry / sector x in the previous quarter; if the electricity consumption forecast value and electricity consumption growth rate forecast value for this year need to be revised, then TYD x TYDSP is the electricity consumption forecast for industry / sector x in this year. x TYDX is the predicted growth rate of electricity consumption in the x industry / sector in this year. x TYDSPX is the revised value of the electricity consumption forecast for the x industry / sector in this year. x is the revised value of the forecast value of the electricity consumption growth rate of the x industry / sector in this year, The electricity consumption of industry / sector x in the previous month.

[0162] S402. Calculate the total electricity consumption TYDX of this year / quarter / month and the total electricity consumption growth rate TYDSPX of this year / quarter / month based on the revised value of the electricity consumption forecast value of each industry / sector in this year / quarter / month and the revised value of the electricity consumption growth rate forecast value of each industry / sector in this year / quarter / month:

[0163]

[0164] TYDSPX=(TYDX-LYYD) / LYYD (8);

[0165] In formulas (7)-(8), l is the total number of industries / sectors, TYDX x is the revised value of the electricity consumption forecast for the x industry / sector in this year / quarter / month, It represents the sum of the revised values ​​of the electricity consumption forecast values ​​of l industries / sectors in this year / quarter / month. LYYD is the total electricity consumption of each industry / sector in the previous year / quarter / month. The units of TYDX and LYYD are billion kWh. The value of LYYD is obtained from the power system operation data.

[0166] In step S402, if the electricity consumption forecast value and the electricity consumption growth rate forecast value for this month need to be revised, LYYD is the total electricity consumption of each industry / sector in the previous month; if the electricity consumption forecast value and the electricity consumption growth rate forecast value for this quarter need to be revised, LYYD is the total electricity consumption of each industry / sector in the previous quarter; if the electricity consumption forecast value and the electricity consumption growth rate forecast value for this year need to be revised, LYYD is the total electricity consumption of each industry / sector in the previous year.

[0167] Major events refer to events that have an impact on industries within the electricity consumption forecast area, including the occurrence of epidemics, the holding of large-scale events, the convening of major conferences, the holding of international expositions, major emergencies, etc.

[0168] Taking the epidemic as an example, the impact of the epidemic on electricity consumption in the industrial and service sectors of a selected region is analyzed, as shown in Table 1 below:

[0169] Table 1: Analysis of the impact of the epidemic on industrial / industry electricity consumption, taking industrial and service electricity consumption as an example

[0170]

[0171]

[0172] According to the data in Table 1, if k ≥ 30, the epidemic has a negative impact on industrial electricity consumption for 30 consecutive days and a negative impact on service electricity consumption for 30 consecutive days during the period from October 1 to January 30 of the year. i The index has been greater than 100 for 30 consecutive days, indicating that electricity consumption has been affected by major events). The epidemic has had a certain impact on the region's industry and service industries.

[0173] The impact of the epidemic on the growth of electricity consumption in industry and service industries in a selected region is shown in Table 2 below:

[0174] Table 2 Impact of the epidemic on regional industrial and service electricity consumption

[0175] Unit: 10,000 kWh

[0176]

[0177]

[0178] According to the data in Table 2, the impact of the epidemic on industrial electricity consumption in the selected region from October 1 to October 31 of the year ECY 工业 The impact on electricity consumption in the service industry in the selected region is -510 million kWh ECY 服务业 It reached -0.72 billion kWh, and the total electricity consumption in the affected regions was reduced by 582 million kWh.

[0179] Taking the annual electricity consumption plan target as an example, the electricity consumption forecast value TYD for that year is 267 billion kWh, and the growth rate forecast value TYDSP is 8.0%. Among them: The electricity consumption forecast value TYD for the service industry is 服务业 It is 53 billion kWh, and the growth rate is predicted to be TYDSP 服务业 The forecast value of industrial electricity consumption is TYD 工业 It is 152 billion kWh, and the growth rate is predicted to be TYDSP工业 It is 2.7%.

[0180] After revising the planned target value and growth rate of electricity consumption in each industry by calculation at the end of October, the revised value of the electricity consumption forecast value in the selected region for the year was 266.418 billion kWh, and the revised value of the growth rate forecast value was 7.78%, including: the revised value of the electricity consumption forecast value of the service industry was TYDX 服务业 It is 52.928 billion kWh, and the revised value of the growth rate forecast is TYDSPX 服务业 12.4%: revised value of industrial electricity consumption forecast TYDX 工业 It is 151.49 billion kWh, and the revised value of the growth rate forecast is TYDSPX 工业 It is 2.4%.

[0181] According to this method, taking into account the impact of the current year's epidemic, it is expected that the electricity consumption growth rate will be reduced by 0.4 percentage points. Therefore, this method can correct the electricity consumption forecast for a period of time in the future according to the impact of major events, further improve the accuracy of electricity consumption forecasts, and then dynamically correct the electricity consumption and growth rate targets for selected regions.

[0182] In summary, taking the epidemic in a certain region in a certain year as a major event, the analysis and application process of an electricity consumption forecast optimization method based on daily electricity consumption changes is introduced. First, by constructing an epidemic impact index model based on daily electricity consumption changes, the impact of major events on industry and service industries and the number of days affected can be analyzed; then, by measuring the contribution of major events to the changes in electricity consumption in industry and service industries, the impact of major events on electricity consumption and its growth rate in industry and service industries is calculated respectively. Finally, the impact of major events on regional electricity consumption and electricity consumption growth rate is summarized and analyzed, and then the regional annual (quarterly, monthly) electricity consumption forecast value and its growth rate forecast value are revised. The quantitative analysis conclusions of this method can provide decision-making references for macro decision makers to evaluate the impact of major events from the perspective of electricity consumption, and can also provide analysis and calculation methods for the forecast of electricity demand under the influence of major events in the future. In addition, this method decouples the impact of major events on electricity consumption from regional electricity consumption, which is conducive to providing reference for the forecast of electricity consumption growth rate under the influence of major events in the future. In the next step, we can summarize the mechanisms and processes of different major events on electricity consumption changes, form a major event impact analysis database, provide a reference for future electricity demand forecasting work under the influence of major events, and further improve the accuracy of analysis and prediction.

[0183] Embodiment 2:

[0184] A power consumption forecast value correction system based on the impact of major events, the system is used to execute the power consumption forecast optimization method based on the impact of major events as described in Example 1, specifically comprising:

[0185] The data collection module is used to select a time period affected by a major event as a first time period, obtain the daily power consumption of each date in the first time period, select a time period not affected by a major event as a second time period, select a normal period typical day corresponding to each date in the first time period in the second time period, and obtain the daily power consumption of each normal period typical day. The data collection module is used to execute step S1 in embodiment 1;

[0186] The model building module is used to build a major event impact index model. The model building module is used to execute the step of building a major event impact index model in Example 1.

[0187] The major event impact index calculation module is used to calculate the major event impact index according to the major event impact index model, the daily power consumption of each date in the first time period, and the daily power consumption of each typical day in the normal period. The major event impact index calculation module is used to execute the step of calculating the major event impact index according to the major event impact index model, the daily power consumption of each date in the first time period, and the daily power consumption of each typical day in the normal period in Example 1.

[0188] The impact factor calculation module is used to calculate the impact factor according to the major event impact index, the daily power consumption of each date in the first time period, and the daily power consumption of each typical day in the normal period. The impact factor calculation module is used to execute step S3 in embodiment 1.

[0189] The value correction module is used to correct the power consumption prediction value and the power consumption growth rate prediction value according to the influencing factor. The value correction module is used to execute step S4 in embodiment 1.

[0190] Embodiment 3:

[0191] A device for predicting and optimizing power consumption based on the impact of major events, comprising a memory and a processor, wherein the memory is used to store computer program codes and transmit the computer program codes to the processor;

[0192] The processor is used to execute the electricity consumption prediction optimization method based on the impact of major events as described in Example 1 according to the instructions in the computer program code.

[0193] Embodiment 4:

[0194] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method for optimizing electricity consumption prediction based on the impact of major events as described in Example 1.

[0195] Generally speaking, the computer instructions for implementing the method of the present invention may be carried in any combination of one or more computer-readable storage media. Non-transitory computer-readable storage media may include any computer-readable media, except for the signal itself that is temporarily propagating.

[0196] Computer-readable storage media, for example, may be, but are not limited to, electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or devices, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection with one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with a system, device or device for executing instructions.

[0197] One or more programming languages ​​or their combinations can be used to write computer program codes for performing the operation of the present invention. These programming languages ​​include object-oriented programming languages, such as Java, Smalltalk, C++, and also include conventional procedural programming languages, such as C language or similar programming languages. In particular, Python language suitable for neural network calculations and platform frameworks based on TensorFlow, PyTorch, etc. can be used. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through various types of networks, including a local area network (LAN) or a wide area network (WAN), or an Internet connection is performed through an Internet service provider.

[0198] Although the embodiments of the present invention have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. The power consumption forecast optimization method based on the impact of major events is characterized by: The optimization method comprises the following steps: S1. Select a time period affected by a major event as a first time period, obtain the daily power consumption of each date in the first time period, select a time period not affected by the major event as a second time period, select a typical day of a normal period corresponding to each date in the first time period in the second time period, and obtain the daily power consumption of each typical day of the normal period; S2. construct a major event impact index model, and calculate the major event impact index based on the major event impact index model, the daily power consumption on each date in the first time period, and the daily power consumption on typical days in each normal period; S3. Calculate the impact factor based on the major event impact index, the daily power consumption on each date in the first time period, and the daily power consumption on a typical day in each normal period; S4. Modify the electricity consumption forecast value and electricity consumption growth rate forecast value according to the influencing factors.

2. The power consumption prediction optimization method based on the impact of major events according to claim 1 is characterized by: In step S1, the first time period is the entire time since the occurrence of the major event in the current year / quarter / month, and the length of the first time period is not less than 3 days; The second time period is one year before the major event occurs.

3. The power consumption prediction optimization method based on the impact of major events according to claim 1 or 2 is characterized in that: In step S1, selecting the typical day of the normal period corresponding to the m-th day i in the first time period in the second time period specifically includes: S101, to eliminate the impact of holidays, filter out dates with the same holiday attributes as i in the second time period; S102. To eliminate the influence of temperature, select a day j with the closest temperature to i from among the dates with the same holiday attribute as i, where j is a typical day in the normal period corresponding to the m-th day i in the first time period.

4. The method for optimizing power consumption prediction based on the impact of major events according to claim 3 is characterized in that: In step S102, selecting a day j with the temperature closest to i from the dates with the same holiday attribute as i specifically includes: Calculate the temperature difference between each date with the same holiday attribute as i and i: In formula (1), is the temperature difference between day n and day i on the date with the same holiday attribute as day i; is the highest temperature on the nth day of the date with the same holiday attribute as i, is the maximum temperature of i, is the lowest temperature on the nth day of the date with the same holiday attribute as i, is the lowest temperature of i; Select the day j with the smallest temperature difference from i among all the dates with the same holiday attribute as i; In formula (2), is the nth day among the dates with the same holiday attribute as i, and d is the number of dates with the same holiday attribute as i.

5. The power consumption prediction optimization method based on the impact of major events according to claim 1 or 2 is characterized in that: In step S2, a major event impact index model is constructed, and the major event impact index is calculated according to the major event impact index model, the daily power consumption of each date in the first time period, and the daily power consumption of typical days in each normal period, specifically including: Constructing a major event impact index model: In formula (3), EF m is the major event impact index corresponding to the mth day in the first time period, is the daily electricity consumption on the mth day in the first time period, is the daily electricity consumption of a typical day in a normal period corresponding to the mth day in the first time period; The major event impact index corresponding to each date in the first time period is calculated according to formula (3).

6. The method for optimizing power consumption prediction based on the impact of major events according to claim 5 is characterized in that: In step S3, calculating the impact factor according to the major event impact index, the daily power consumption on each date in the first time period, and the daily power consumption on each typical day in the normal period specifically includes: The impact factor of major events on each industry / sector is calculated according to the following formula: In formula (4), ECY x is the impact factor of major events on industry / sector x, p is the number of days in the first time period, is the daily electricity consumption of industry x on the mth day in the first time period, STAT is the daily electricity consumption of a typical day in a normal period corresponding to the mth day in the first time period of the x industry / sector. m is the influence coefficient corresponding to the mth day in the first time period; STAT m The value of is determined according to the following conditions: If the following three conditions A, B, and C are met at the same time, STAT m = -1; A, m∈[p1,p2]; B. p2-p1≥k; C、 If the following three conditions D, E, and F are met at the same time, STAT m =1; D, m∈[p3,p4]; E, p3-p4 ≥ k; F、 If conditions A, B, and C or conditions D, E, and F cannot be met at the same time, STAT m =0; The k is a set time threshold, k≥2.

7. The method for optimizing power consumption prediction based on the impact of major events according to claim 6 is characterized in that: In step S4, the correction of the power consumption forecast value and the power consumption growth rate forecast value according to the influencing factor specifically includes: S401, ECY, the impact factor of major events on industry / sector x x The electricity consumption forecast value TYD for x industry / sector in this year / quarter / month x , the predicted growth rate of electricity consumption of x industry / sector in this year / quarter / month TYDSP x To make corrections: TYDX x =TIME x +ECY x (5); TYDSPX x =TYDSP x +ECY x / LYYD t x (6); In formulas (5)-(6), TYDX x TYDSPX is the revised value of the electricity consumption forecast for the x industry / sector in this year / quarter / month. x is the revised value of the forecast value of electricity consumption growth rate of industry / sector x in this year / quarter / month, The electricity consumption of x industry / sector in the previous year / quarter / month; S402. Calculate the total electricity consumption TYDX of this year / quarter / month and the total electricity consumption growth rate TYDSPX of this year / quarter / month based on the revised value of the electricity consumption forecast value of each industry / sector in this year / quarter / month and the revised value of the electricity consumption growth rate forecast value of each industry / sector in this year / quarter / month: TYDSPX=(TYDX-LYYD) / LYYD (8); In formulas (7)-(8), l is the total number of industries / sectors, TYDX x LYYD is the revised value of the forecast electricity consumption of industry / sector x in this year / quarter / month, and LYYD is the total electricity consumption of each industry / sector in the previous year / quarter / month.

8. A power consumption forecast value correction system based on the impact of major events, characterized by: The system is used to execute the power consumption prediction optimization method based on the impact of major events as described in any one of claims 1 to 7, specifically comprising: A data collection module is used to select a time period affected by a major event as a first time period, obtain the daily power consumption of each date in the first time period, select a time period not affected by the major event as a second time period, select a normal period typical day corresponding to each date in the first time period in the second time period, and obtain the daily power consumption of each normal period typical day; Model building module, used to build a major event impact index model; A major event impact index calculation module, used to calculate the major event impact index based on the major event impact index model, the daily power consumption of each date in the first time period, and the daily power consumption of typical days in each normal period; An impact factor calculation module, used to calculate the impact factor according to the impact index of the major event, the daily power consumption of each date in the first time period, and the daily power consumption of typical days in each normal period; The correction module is used to correct the power consumption forecast value and the power consumption growth rate forecast value according to the influencing factors.

9. A device for predicting and optimizing power consumption based on the impact of major events, characterized in that: The method comprises a memory and a processor, wherein the memory is used to store computer program code and transmit the computer program code to the processor; The processor is configured to execute the method according to any one of claims 1 to 7 according to instructions in the computer program code.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any one of claims 1 to 7 is implemented.

Images