Meteorological sensitive load prediction method based on reference load stripping

Abstract

The invention provides a meteorological sensitive load prediction method based on reference load stripping, and the method comprises the steps: firstly obtaining historical load data and historical temperature data, the historical temperature data being historical temperature data; analyzing the historical load data and the historical temperature data, and fitting the relationship between the load data and the meteorological data by using a polynomial function to obtain a relational expression between the load and the meteorological data; calculating according to the characteristics of the polynomial function to obtain a temperature interval in which the load is the least sensitive, and calculating a reference load value based on the load data in the interval; and finally, predicting future load data according to the historical load data and the historical temperature data, and obtaining a predicted value of the meteorological sensitive load based on the reference load value and the predicted load value. The method does not depend on the extraction of historical meteorological sensitive loads, and can guarantee the prediction accuracy.

Description

Technical Field

[0001] The present invention belongs to the technical field of load forecasting, and in particular relates to a meteorological sensitive load forecasting method based on benchmark load stripping. Background Art

[0002] With the intensification of global climate change and the continuous expansion of the scale of power systems, power load forecasting, especially meteorological sensitive load forecasting considering meteorological factors, has become an important part of power system operation and management. Meteorological sensitive load forecasting aims to achieve accurate forecasting of future power loads by comprehensively considering historical meteorological data, power load data and socio-economic factors, so as to optimize the allocation of power resources and improve the safety and economy of power grid operation. Most existing methods directly use the historical data of meteorological sensitive loads to predict future meteorological sensitive loads, but the extraction of historical data of meteorological sensitive loads is difficult and difficult to obtain.

[0003] Therefore, it is necessary to study other methods to forecast meteorological sensitive loads. Summary of the invention

[0004] The purpose of the present invention is to provide a meteorological sensitive load prediction method based on benchmark load stripping. By extracting the benchmark load value and the predicted load value, the predicted value of the meteorological sensitive load is obtained, which can reduce the impact of the difficulty in extracting historical meteorological sensitive load data while ensuring the accuracy of the prediction.

[0005] In a first aspect, the present application provides a meteorological sensitive load prediction method based on benchmark load stripping, comprising:

[0006] Obtain historical load data and historical meteorological data, where the historical meteorological data is historical temperature data. Obtain the optimum temperature range based on the historical load data and historical temperature data. Calculate the average value of the historical load data corresponding to the optimum temperature range as the benchmark load value. Predict future load data based on the historical load data and historical temperature data to obtain a predicted load value, and subtract the benchmark load value from the predicted load value to obtain a predicted value for the meteorological sensitive load.

[0007] Temperature is the most important factor affecting power load. This application predicts the meteorological sensitive load in the future period based on historical temperature data and historical load data, which can effectively predict the future meteorological sensitive load and ensure the accuracy of the prediction.

[0008] In a possible implementation, the acquiring of historical load data and historical meteorological data, wherein the historical meteorological data is historical temperature data, includes:

[0009] Obtain historical load data L and historical temperature data C, where L = {L i,t,i=1,2,…,N,t=1,2,…,M}, where L i,t represents the load data at the tth time point on the i-th day, C i,t Represents the temperature data at the t-th time point on the i-th day, N is the number of days selected, and M represents the number of time points in a day.

[0010] Since the electricity consumption characteristics of working days and weekends are quite different, the load data is classified according to the two date types of working days and weekends, and L is divided into L w and L r , where L w and L r Represents the load data for working days and rest days respectively.

[0011] In a possible implementation, a time point is taken every 15 minutes, and M=96.

[0012] In a possible implementation, analyzing the historical load data and the historical temperature data to obtain the optimum temperature range includes:

[0013] Convert the temperature at each moment into the cumulative temperature. The conversion formula is as follows:

[0014]

[0015] In the formula, is the accumulated temperature data at the tth time point on the i-th day, J is the set time point number, P j+1 To calculate C i,t-j The weight coefficient of the temperature accumulation effect, which changes with the passage of j. The closer j is to the current moment, i.e., the tth moment, the greater the weight; C i,t-j is the temperature data at the jth time point before the tth time point on the i-th day;.

[0016] Taking into account the influence of continuous high temperature, through this step, the temperature at each moment is corrected and converted into cumulative temperature, which can improve the accuracy of the prediction.

[0017] Since load is closely related to weather, a polynomial function can be used to describe the relationship between them. Therefore, for the load data L on working days and rest days, i,t ∈L w and L i,t ∈L r , use the polynomial function to fit the relationship between the accumulated temperature data and the historical load data, and get two relationship equations for working days and rest days respectively. The expression of the polynomial function is as follows:

[0018] Q=f(C * )=w0+w1C* +w2(C * ) 2 +...+w n (C * ) n

[0019] Where Q is the fitted load value, w0~w n is the weight parameter of the polynomial function, n is the degree of the polynomial, which can be set according to the specific scenario, C * is the accumulated temperature.

[0020] The fitted polynomial function is analyzed. First, the derivative Q' of the polynomial function is calculated to obtain the accumulated temperature C when Q' = 0. * The value of is expressed as {c1,c2,...,c m}, where m≥1. The reason for having multiple temperature values ​​is that there may be multiple points where Q'=0. Then calculate when C * ={c1,c2,...,c m}, and take the value of Q when Q is minimum, and the accumulated temperature c min , this temperature is the point where the load is least sensitive to temperature, and the temperature range where the temperature value fluctuates up and down ε is set as the insensitive zone of the load to temperature, that is, the optimal temperature range, expressed as C b =[c min -ε,c min +ε], where the value of the interval parameter ε can be adjusted according to the regional load characteristics;

[0021] In a possible implementation, calculating the average value of the historical load data corresponding to the optimum temperature range as the reference load value includes:

[0022] For all Assume that the cumulative temperature of working days and rest days is within the optimal temperature range C b The numbers in the interval are W and R respectively, and the subscripts of the accumulated temperature data of working days and rest days in this interval are represented as sets and Calculate the corresponding average load value as the benchmark load value for working days and rest days. and

[0023]

[0024] In a possible implementation, predicting future load data based on historical load data and historical temperature data to obtain a predicted load value, and subtracting a reference load value from the predicted load value to obtain a predicted value for a weather-sensitive load includes:

[0025] Construct a load forecasting model L = F (C, s, t), where L is the load, C is the temperature, s is the day type, and t is the time;

[0026] Using historical load data and historical temperature data to train the load forecasting model to obtain a trained load forecasting model;

[0027] Get forecast values ​​of weather data The meteorological data, day type and time are input into the load forecasting model to obtain the corresponding forecast load value. When the day type is a working day, the predicted load value is When the day type is a rest day, get the predicted value

[0028] The predicted value of the meteorological sensitive load is calculated based on the predicted load value and the benchmark value. The difference between the predicted load value and the benchmark load on working days and rest days is calculated respectively to obtain the predicted value of the meteorological sensitive load:

[0029]

[0030] In the formula, and They are the predicted values ​​of meteorological sensitive loads on working days and weekends, respectively.

[0031] In a second aspect, the present application provides a meteorological sensitive load prediction system based on benchmark load stripping, comprising:

[0032] A data acquisition module, used to acquire historical load data and historical meteorological data, wherein the historical meteorological data is historical temperature data;

[0033] An optimum temperature range acquisition module, used for acquiring an optimum temperature range based on the historical load data and the historical temperature data;

[0034] A reference load value calculation module, used to calculate the average value of the historical load data corresponding to the optimum temperature range as the reference load value;

[0035] The prediction module is used to predict future load data based on the historical load data and the historical temperature data to obtain a predicted load value, and to subtract the reference load value from the predicted load value to obtain a predicted value for the meteorological sensitive load.

[0036] In a possible implementation, the acquiring of historical load data and historical meteorological data, wherein the historical meteorological data is historical temperature data, includes:

[0037] Obtain historical load data L and historical temperature data C, where L = {L i,t,i=1,2,…,N,t=1,2,…,M},L i,t represents the load data at the tth time point on the i-th day, C i,t represents the temperature data at the t-th time point on the i-th day, N is the number of days selected, and M represents the number of time points in a day;

[0038] The load data is classified according to the two date types of working days and rest days, that is, L is divided into L w and L r , where L w and L r Represents the load data for working days and rest days respectively.

[0039] In a possible implementation, the acquiring the optimum temperature range based on the historical load data and the historical temperature data includes:

[0040] Convert the temperature at each moment into the cumulative temperature. The conversion formula is as follows:

[0041]

[0042] In the formula, is the accumulated temperature data at the tth time point on the i-th day, J is the set time point number, P j+1 To calculate C i,t-j The weight coefficient of the temperature accumulation effect, which changes with the passage of j. The closer j is to the current moment, the greater the weight; C i,t-j is the temperature data at the jth time point before the tth time point on the i-th day;

[0043] The load data L for working days and rest days are i,t ∈L w and L i,t ∈L r , use the polynomial function to fit the relationship between the accumulated temperature data and the historical load data, and get two relationship equations for working days and rest days respectively. The expression of the polynomial function is as follows:

[0044] Q=f(C * )=w0+w1C * +w2(C * ) 2 +...+w n (C * ) n

[0045] Where Q is the fitted load value, w0~w n is the weight parameter of the polynomial function, n is the degree of the polynomial, C * is the accumulated temperature;

[0046] The fitted polynomial function is analyzed. First, the derivative Q' of the polynomial function is calculated to obtain the accumulated temperature C when Q' = 0. * The value of is expressed as {c1,c2,...,c m}, where m≥1;

[0047] Calculate when C * ={c1,c2,...,c m}, and take the value of Q when Q is minimum, and the accumulated temperature c min , cumulative temperature c min The point where the load is least sensitive to temperature; according to c min The optimum temperature range is determined by the set interval parameters, expressed as C b =[c min -ε,c min +ε].

[0048] In a possible implementation, calculating the average value of the historical load data corresponding to the optimum temperature range as the reference load value includes:

[0049] For all Assume that the accumulated temperature data on working days and rest days is within the optimum temperature range C b The number of days in the optimal temperature range is W and R respectively, and the working days and rest days are placed in the optimal temperature range C b The subscripts of the accumulated temperature data in the set are and Calculate the corresponding average load value as the benchmark load value for working days and rest days. and

[0050]

[0051] In a possible implementation, predicting future load data according to the historical load data and the historical temperature data to obtain a predicted load value, and subtracting the reference load value from the predicted load value to obtain a predicted value of the meteorological sensitive load includes:

[0052] Construct a load forecasting model L = F (C, s, t), where L is the load, C is the temperature, s is the day type, and t is the time;

[0053] Using historical load data and historical temperature data to train the load forecasting model to obtain a trained load forecasting model;

[0054] Get forecast values ​​of weather data The meteorological data, day type and time are input into the load forecasting model to obtain the corresponding forecast load value. When the day type is a working day, the predicted load value is When the day type is a rest day, get the predicted value

[0055] The predicted value of the meteorological sensitive load is calculated based on the predicted load value and the benchmark value. The difference between the predicted load value and the benchmark load on working days and rest days is calculated respectively to obtain the predicted value of the meteorological sensitive load:

[0056]

[0057] In the formula, and They are the predicted values ​​of meteorological sensitive loads on working days and weekends, respectively.

[0058] In a third aspect, the present application provides an electronic device, including: a memory and a processor;

[0059] The memory is used to store computer programs;

[0060] The processor is used to call the computer program to execute the method as described above.

[0061] In a fourth aspect, the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed on an electronic device, the electronic device implements the method described above.

[0062] In a fifth aspect, the present application provides a computer program product, including a computer program, and when the computer program is run on an electronic device, the electronic device implements the method as described above.

[0063] The specific implementation methods of the third to fifth aspects of the present application can refer to the implementation methods of the first aspect, which will not be repeated here.

[0064] Beneficial effects:

[0065] The present invention extracts the benchmark load based on historical load data and meteorological data, predicts the future load, and finally calculates the predicted value of the meteorological sensitive load based on the predicted load value and the benchmark load value. The existing technical solutions for predicting meteorological sensitive loads mostly rely on historical meteorological sensitive load data. However, there are great difficulties in extracting such data, and it is difficult to ensure accuracy. The use of data with no guaranteed accuracy for meteorological sensitive load prediction has a great impact on the accuracy of the prediction results. The present invention extracts the benchmark load and obtains the predicted value of the meteorological sensitive load based on the predicted load value and the benchmark load. This method is more accurate and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 The present invention is a flow chart of a meteorological sensitive load prediction method based on benchmark load stripping.

[0067] Figure 2 It is a principle diagram of a meteorological sensitive load prediction method based on benchmark load stripping of the present invention. DETAILED DESCRIPTION

[0068] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0069] The present invention provides a meteorological sensitive load prediction method based on benchmark load stripping, such as Figure 1 As shown, including:

[0070] S10: Acquire historical load data and historical meteorological data, wherein the historical meteorological data is historical temperature data.

[0071] Obtain historical load data L and historical temperature data C, where L = {L i,t ,i=1,2,…,N,t=1,2,…,M}, where L i,t represents the load data at the tth time point on the i-th day, C i,t Represents the temperature data at the t-th time point on the i-th day, N is the number of days selected, and M represents the number of time points in a day.

[0072] Since the electricity consumption characteristics of working days and weekends are quite different, the load data is classified according to the two date types of working days and weekends, and L is divided into L w and L r , where L w and L r Represents the load data for working days and rest days respectively.

[0073] In some embodiments, a time point is taken every 15 minutes, and M=96.

[0074] S20: Obtaining an optimum temperature range based on historical load data and historical temperature data.

[0075] Convert the temperature at each moment into the cumulative temperature. The conversion formula is as follows:

[0076]

[0077] In the formula, is the cumulative temperature data at the t-th time point on the i-th day; J is the set time point number, which can be selected according to specific needs. For example, 4 time points or 16 time points can be taken to calculate the cumulative temperature. In this case, J = 3 or J = 15; P j+1 To calculate C i,t-j The weight coefficient of the temperature accumulation effect changes with the passage of j. The closer j is to the current moment, i.e., the tth moment, the greater the weight; C i,t-j is the temperature data at the jth time point before the tth time point on the i-th day.

[0078] Taking into account the influence of continuous high temperature, through this step, the temperature at each moment is corrected and converted into cumulative temperature, which can improve the accuracy of the prediction.

[0079] Since load is closely related to weather, a polynomial function can be used to describe the relationship between them. Therefore, for the load data L on working days and rest days, i,t ∈L w and L i,t ∈L r , use the polynomial function to fit the relationship between the accumulated temperature data and the historical load data, and get two relationship equations for working days and rest days respectively. The expression of the polynomial function is as follows:

[0080] Q=f(C * )=w0+w1C * +w2(C * ) 2 +...+w n (C * ) n

[0081] Where Q is the fitted load value, w0~w n is the weight parameter of the polynomial function, n is the degree of the polynomial, which can be set according to the specific scenario. In this scenario, the value range of n is generally 3 to 5. * is the accumulated temperature.

[0082] The fitted polynomial function is analyzed. First, the derivative Q' of the polynomial function is calculated to obtain the accumulated temperature C when Q' = 0. * The value of is expressed as {c1,c2,...,c m}, where m≥1. The reason for having multiple temperature values ​​is that there may be multiple points where L'=0. Then calculate when C * ={c1,c2,...,c m}, and take the value of Q when Q is minimum, and the accumulated temperature c min, this temperature is the point where the load is least sensitive to temperature, and the temperature range where the temperature value fluctuates up and down ε is set as the insensitive zone of the load to temperature, that is, the optimal temperature range, expressed as C b =[c min -ε,c min +ε], where the value of the interval parameter ε can be adjusted according to the regional load characteristics;

[0083] S30: Calculate the average value of the historical load data corresponding to the optimum temperature range as the reference load value.

[0084] For all Assume that the number of accumulated temperature data on working days and rest days in this interval is W and R respectively, and the subscripts of the temperature data on working days and rest days in this interval are represented as the set and Calculate the corresponding average load value as the benchmark load value for working days and rest days. and

[0085]

[0086] In the formula, and are the baseline loads for working days and rest days respectively.

[0087] This step is based on the average value of the historical load data corresponding to the optimum temperature range as the benchmark load value, that is, the load value after removing the meteorological sensitive load.

[0088] S40: Predict future load data based on historical load data and historical temperature data to obtain a predicted load value, and subtract the reference load value from the predicted load value to obtain a predicted value for the meteorological sensitive load.

[0089] Construct a load forecasting model L = F (C, s, t), where L is the load, C is the temperature, s is the day type, and t is the time;

[0090] Using historical load data and historical temperature data to train the load forecasting model to obtain a trained load forecasting model;

[0091] Get forecast values ​​of weather data The meteorological data, day type and time are input into the load forecasting model to obtain the corresponding forecast load value. When the day type is a working day, the predicted load value is When the day type is a rest day, get the predicted value

[0092] The predicted value of the meteorological sensitive load is calculated based on the predicted load value and the benchmark value. The difference between the predicted load value and the benchmark load on working days and rest days is calculated respectively to obtain the predicted value of the meteorological sensitive load:

[0093]

[0094] In the formula, and They are the predicted values ​​of meteorological sensitive loads on working days and weekends, respectively.

[0095] The embodiment of the present application also provides a meteorological sensitive load prediction system based on benchmark load stripping, including:

[0096] A data acquisition module, used to acquire historical load data and historical meteorological data, wherein the historical meteorological data is historical temperature data;

[0097] An optimum temperature range acquisition module, used for acquiring an optimum temperature range based on the historical load data and the historical temperature data;

[0098] A reference load value calculation module, used to calculate the average value of the historical load data corresponding to the optimum temperature range as the reference load value;

[0099] The prediction module is used to predict future load data based on the historical load data and the historical temperature data to obtain a predicted load value, and to subtract the reference load value from the predicted load value to obtain a predicted value for the meteorological sensitive load.

[0100] In some embodiments, the acquiring of historical load data and historical meteorological data, wherein the historical meteorological data is historical temperature data, includes:

[0101] Obtain historical load data L and historical temperature data C, where L = {L i,t ,i=1,2,…,N,t=1,2,…,M},L i,t represents the load data at the tth time point on the i-th day, C i,t represents the temperature data at the t-th time point on the i-th day, N is the number of days selected, and M represents the number of time points in a day;

[0102] The load data is classified according to the two date types of working days and rest days, that is, L is divided into L w and L r , where L w and L r Represents the load data for working days and rest days respectively.

[0103] In some embodiments, the obtaining the optimum temperature range based on the historical load data and the historical temperature data includes:

[0104] Convert the temperature at each moment into the cumulative temperature. The conversion formula is as follows:

[0105]

[0106] In the formula, is the accumulated temperature data at the tth time point on the i-th day, J is the set time point number, P j+1 To calculate C i,t-j The weight coefficient of the temperature accumulation effect, which changes with the passage of j. The closer j is to the current moment, the greater the weight; C i,t-j is the temperature data at the jth time point before the tth time point on the i-th day;

[0107] The load data L for working days and rest days are i,t ∈L w and L i,t ∈L r , use the polynomial function to fit the relationship between the accumulated temperature data and the historical load data, and get two relationship equations for working days and rest days respectively. The expression of the polynomial function is as follows:

[0108] Q=f(C * )=w0+w1C * +w2(C * ) 2 +...+w n (C * ) n

[0109] Where Q is the fitted load value, w0~w n is the weight parameter of the polynomial function, n is the degree of the polynomial, C * is the accumulated temperature;

[0110] The fitted polynomial function is analyzed. First, the derivative Q' of the polynomial function is calculated to obtain the accumulated temperature C when Q' = 0. * The value of is expressed as {c1,c2,...,c m}, where m≥1;

[0111] Calculate when C * ={c1,c2,...,c m}, and take the value of Q when Q is minimum, and the accumulated temperature c min , cumulative temperature c min The point where the load is least sensitive to temperature; according to c min The optimum temperature range is determined by the set interval parameters, expressed as C b =[c min -ε,c min +ε].

[0112] In some embodiments, calculating the average value of the historical load data corresponding to the optimum temperature range as the reference load value includes:

[0113] For all Assume that the accumulated temperature data on working days and rest days is within the optimum temperature range C b The number of days in the optimal temperature range is W and R respectively, and the working days and rest days are placed in the optimal temperature range C b The subscripts of the accumulated temperature data in the set are and Calculate the corresponding average load value as the benchmark load value for working days and rest days. and

[0114]

[0115] In some embodiments, predicting future load data according to the historical load data and the historical temperature data to obtain a predicted load value, and subtracting the benchmark load value from the predicted load value to obtain a predicted value of the meteorological sensitive load includes:

[0116] Construct a load forecasting model L = F (C, s, t), where L is the load, C is the temperature, s is the day type, and t is the time;

[0117] Using historical load data and historical temperature data to train the load forecasting model to obtain a trained load forecasting model;

[0118] Get forecast values ​​of weather data The meteorological data, day type and time are input into the load forecasting model to obtain the corresponding forecast load value. When the day type is a working day, the predicted load value is When the day type is a rest day, get the predicted value

[0119] The predicted value of the meteorological sensitive load is calculated based on the predicted load value and the benchmark value. The difference between the predicted load value and the benchmark load on working days and rest days is calculated respectively to obtain the predicted value of the meteorological sensitive load:

[0120]

[0121] In the formula, and They are the predicted values ​​of meteorological sensitive loads on working days and weekends, respectively.

[0122] The embodiment of the present application also provides an electronic device, including: a memory and a processor;

[0123] The memory is used to store computer programs;

[0124] The processor is used to call the computer program to execute the method as described above.

[0125] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed on an electronic device, the electronic device implements the method as described above.

[0126] An embodiment of the present application further provides a computer program product, including a computer program. When the computer program is executed on an electronic device, the electronic device implements the method as described above.

[0127] The embodiments of the present application also provide a system, an electronic device, a computer-readable storage medium, and a computer program product. The specific implementation methods can refer to the specific embodiments of the above methods and will not be repeated here.

[0128] Please refer to the template and make sure to use the same words before and after. (For those who do not have a description of the corresponding computer storage medium, you can send this template. If you have a similar description, you do not need to give it.)

[0129] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, electronic devices, computer-readable storage media or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application may be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0130] The present application is described with reference to the flowcharts and / or block diagrams of the methods, systems, electronic devices, computer-readable storage media, and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1A device that provides the functions specified in a block or multiple blocks.

[0131] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0132] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0133] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0134] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A meteorological sensitive load forecasting method based on benchmark load stripping, characterized in that: include: Acquire historical load data and historical meteorological data, wherein the historical meteorological data is historical temperature data; Based on the historical load data and the historical temperature data, obtaining an optimum temperature range; Calculate the average value of the historical load data corresponding to the optimum temperature range as the benchmark load value; Future load data is predicted based on the historical load data and the historical temperature data to obtain a predicted load value, and the benchmark load value is subtracted from the predicted load value to obtain a predicted value for the meteorological sensitive load.

2. The method according to claim 1, characterized in that The acquiring of historical load data and historical meteorological data, wherein the historical meteorological data is historical temperature data, includes: Obtain historical load data L and historical temperature data C, where L = {L i,t ,i=1,2,…,N,t=1,2,…,M},L i,t represents the load data at the tth time point on the i-th day, C i,t represents the temperature data at the t-th time point on the i-th day, N is the number of days selected, and M represents the number of time points in a day; The load data is classified according to the two date types of working days and rest days, that is, L is divided into L w and L r , where L w and L r Represents the load data for working days and rest days respectively.

3. The method according to claim 2, characterized in that The obtaining of the optimum temperature range based on the historical load data and the historical temperature data includes: Convert the temperature at each moment into the cumulative temperature. The conversion formula is as follows: In the formula, is the accumulated temperature data at the tth time point on the i-th day, J is the set time point number, P j+1 To calculate C i,t-j The weight coefficient of the temperature accumulation effect, which changes with the passage of j. The closer j is to the current moment, the greater the weight; C i,t-j is the temperature data at the jth time point before the tth time point on the i-th day; The load data L for working days and rest days are i,t ∈L w and L i,t ∈L r , use the polynomial function to fit the relationship between the accumulated temperature data and the historical load data, and get two relationship equations for working days and rest days respectively. The expression of the polynomial function is as follows: Q=f(C * )=w0+w1C * +w2(C * ) 2 +...+w n (C * ) n Where Q is the fitted load value, w0~w n is the weight parameter of the polynomial function, n is the degree of the polynomial, C * is the accumulated temperature; The fitted polynomial function is analyzed. First, the derivative Q' of the polynomial function is calculated to obtain the accumulated temperature C when Q' = 0. * The value of is expressed as {c1,c2,...,c m }, where m≥1; Calculate when C * ={c1,c2,...,c m }, and take the value of Q when Q is minimum, and the accumulated temperature c min , cumulative temperature c min The point where the load is least sensitive to temperature; according to c min The optimum temperature range is determined by the set interval parameters, expressed as C b =[c min -ε,c min +ε].

4. The method according to claim 3, characterized in that Calculating the average value of the historical load data corresponding to the optimum temperature range as the benchmark load value includes: For all Assume that the accumulated temperature data on working days and rest days is within the optimum temperature range C b The number of days in the optimal temperature range is W and R respectively, and the working days and rest days are placed in the optimal temperature range C b The subscripts of the accumulated temperature data in the set are and Calculate the corresponding average load value as the benchmark load value for working days and rest days. and 5. The method according to claim 4, characterized in that Predicting future load data according to the historical load data and the historical temperature data to obtain a predicted load value, and subtracting the reference load value from the predicted load value to obtain a predicted value for the meteorological sensitive load, including: Construct a load forecasting model L = F (C, s, t), where L is the load, C is the temperature, s is the day type, and t is the time; Using historical load data and historical temperature data to train the load forecasting model to obtain a trained load forecasting model; Get forecast values ​​of weather data The meteorological data, day type and time are input into the load forecasting model to obtain the corresponding forecast load value. When the day type is a working day, the predicted load value is When the day type is a rest day, get the predicted value The predicted value of the meteorological sensitive load is calculated based on the predicted load value and the benchmark value. The difference between the predicted load value and the benchmark load on working days and rest days is calculated respectively to obtain the predicted value of the meteorological sensitive load: In the formula, and They are the predicted values ​​of meteorological sensitive loads on working days and weekends, respectively.

6. A meteorological sensitive load forecasting system based on benchmark load stripping, characterized in that: include: A data acquisition module, used to acquire historical load data and historical meteorological data, wherein the historical meteorological data is historical temperature data; An optimum temperature range acquisition module, used for acquiring an optimum temperature range based on the historical load data and the historical temperature data; A reference load value calculation module, used to calculate the average value of the historical load data corresponding to the optimum temperature range as the reference load value; The prediction module is used to predict future load data based on the historical load data and the historical temperature data to obtain a predicted load value, and to subtract the reference load value from the predicted load value to obtain a predicted value for the meteorological sensitive load.

7. The system according to claim 6, characterized in that The acquiring of historical load data and historical meteorological data, wherein the historical meteorological data is historical temperature data, includes: Obtain historical load data L and historical temperature data C, where L = {L i,t ,i=1,2,…,N,t=1,2,…,M},L i,t represents the load data at the tth time point on the i-th day, C i,t represents the temperature data at the t-th time point on the i-th day, N is the number of days selected, and M represents the number of time points in a day; The load data is classified according to the two date types of working days and rest days, that is, L is divided into L w and L r , where L w and L r Represents the load data for working days and rest days respectively.

8. The system according to claim 7, characterized in that The obtaining of the optimum temperature range based on the historical load data and the historical temperature data includes: Convert the temperature at each moment into the cumulative temperature. The conversion formula is as follows: In the formula, is the accumulated temperature data at the tth time point on the i-th day, J is the set time point number, P j+1 To calculate C i,t-j The weight coefficient of the temperature accumulation effect, which changes with the passage of j. The closer j is to the current moment, the greater the weight; C i,t-j is the temperature data at the jth time point before the tth time point on the i-th day; The load data L for working days and rest days are i,t ∈L w and L i,t ∈L r , use the polynomial function to fit the relationship between the accumulated temperature data and the historical load data, and get two relationship equations for working days and rest days respectively. The expression of the polynomial function is as follows: Q=f(C * )=w0+w1C * +w2(C * ) 2 +...+w n (C * ) n Where Q is the fitted load value, w0~w n is the weight parameter of the polynomial function, n is the degree of the polynomial, C * is the accumulated temperature; The fitted polynomial function is analyzed. First, the derivative Q' of the polynomial function is calculated to obtain the accumulated temperature C when Q' = 0. * The value of is expressed as {c1,c2,...,c m }, where m≥1; Calculate when C * ={c1,c2,...,c m }, and take the value of Q when Q is minimum, and the accumulated temperature c min , cumulative temperature c min The point where the load is least sensitive to temperature; according to c min The optimum temperature range is determined by the set interval parameters, expressed as C b =[c min -ε,c min +ε].

9. The system according to claim 8, characterized in that Calculating the average value of the historical load data corresponding to the optimum temperature range as the benchmark load value includes: For all Assume that the accumulated temperature data on working days and rest days is within the optimum temperature range C b The number of days in the optimal temperature range is W and R respectively, and the working days and rest days are placed in the optimal temperature range C b The subscripts of the accumulated temperature data in the set are and Calculate the corresponding average load value as the benchmark load value for working days and rest days. and 10. The system according to claim 9, characterized in that Predicting future load data according to the historical load data and the historical temperature data to obtain a predicted load value, and subtracting the reference load value from the predicted load value to obtain a predicted value for the meteorological sensitive load, including: Construct a load forecasting model L = F (C, s, t), where L is the load, C is the temperature, s is the day type, and t is the time; Using historical load data and historical temperature data to train the load forecasting model to obtain a trained load forecasting model; Get forecast values ​​of weather data The meteorological data, day type and time are input into the load forecasting model to obtain the corresponding forecast load value. When the day type is a working day, the predicted load value is When the day type is a rest day, get the predicted value The predicted value of the meteorological sensitive load is calculated based on the predicted load value and the benchmark value. The difference between the predicted load value and the benchmark load on working days and rest days is calculated respectively to obtain the predicted value of the meteorological sensitive load: In the formula, and They are the predicted values ​​of meteorological sensitive loads on working days and weekends, respectively.

11. An electronic device, characterized in that: include: Memory and processor; The memory is used to store computer programs; The processor is configured to call the computer program to execute the method according to any one of claims 1 to 5.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed on an electronic device, the electronic device implements the method according to any one of claims 1 to 5.

13. A computer program product, comprising a computer program, characterized in that When the computer program is executed on an electronic device, the electronic device implements the method according to any one of claims 1 to 5.

Images