Decoupling summer and winter temperature regulation load measuring and calculating method, device and equipment

Abstract

The invention relates to the technical field of power load measurement and calculation, in particular to a decoupling summer and winter temperature regulation load measurement and calculation method. Training a pre-established GAM-based day-ahead load prediction model by using the data sample set; further obtaining an influence function of the temperature on the load, an influence function of the humidity on the load and a characteristic value function of the historical load; decoupling the GAM-based day-ahead load prediction model by using three characteristic value functions to obtain a decoupling load model composed of a temperature characteristic function, a humidity characteristic function and a historical load characteristic function; and carrying out load measurement and calculation by using the decoupling load model. It can be understood that according to the technical scheme shown by the invention, the air temperature item and historical load item characteristic function can directly reflect the relation between the temperature and the load, and an important basis is provided for fine analysis of the power load.

Description

Technical Field

[0001] The present invention relates to the technical field of power load calculation, and in particular to a decoupled summer and winter temperature control load calculation method, device and equipment. Background Art

[0002] In the power industry, accurate short-term load forecasting has always been an important part of ensuring stable operation of the power grid. Especially in summer and winter, temperature changes will cause the temperature control load of the power grid to increase, which brings great challenges to the operation of the power grid. Since the temperature control load cannot be directly obtained through metering equipment, and most existing methods cannot eliminate the influence of characteristic quantities such as humidity on the measurement results, there are certain limitations, resulting in large deviations in the stripping of the temperature control load.

[0003] In the existing technology, the temperature control load calculation method is mainly based on the typical day method of the benchmark curve comparison method. The benchmark curve comparison method can roughly obtain the scale of the cooling load, but there are the following problems: the measurement results cannot eliminate the influence of other characteristics such as humidity and wind speed; the load calculation ignores the cumulative effect of the temperature control load and cannot explain how the cumulative effect drives the load growth under continuous high / low temperatures; the use of the benchmark curve makes the non-temperature control load lose its weekly and monthly variation characteristics.

[0004] Therefore, the current temperature control load calculation method cannot independently obtain the influence of characteristics such as temperature, humidity, and historical load, and the calculation results are inaccurate. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide a decoupled summer and winter temperature control load calculation method, device and equipment to solve the problem that the temperature control load calculation method in the prior art cannot independently obtain the influence of characteristics such as temperature, humidity, historical load, etc., resulting in inaccurate calculation results.

[0006] According to a first aspect of an embodiment of the present invention, a decoupled summer and winter temperature control load calculation method is provided, comprising:

[0007] Obtaining load data and temperature data within a preset time period in the area to be tested, and integrating the load data and the temperature data into a data sample set;

[0008] Using the data sample set to train a pre-established GAM-based day-ahead load forecasting model;

[0009] According to the trained GAM-based day-ahead load forecasting model, the influence function of temperature on load, the influence function of humidity on load and the characteristic value function of historical load are obtained;

[0010] The trained GAM-based day-ahead load forecasting model is decoupled by using the temperature-load influence function, humidity-load influence function and historical load characteristic value function, and a decoupled load model consisting of temperature characteristic function, humidity characteristic function and historical load characteristic function is obtained.

[0011] The decoupled load model is used to perform load calculation.

[0012] Preferably, the trained GAM-based day-ahead load forecasting model is expressed as follows:

[0013] log(x d,t )=T d,t (w)+H d,t (D)+L d,t (x d-,t- )

[0014] Among them, x d,t is the forecast load at time t on day d, t=1,2,...,96; T d,t (w) is the effect of temperature w on the load at time t on day d; H d,t (D) is the effect of humidity D on the load at time t on day d; L d,t (x d-,t- ) is the influence of load characteristic value on the predicted load at time t on day d; x d-,t- is the load value before time t on day d.

[0015] Preferably, the influence function of temperature on load is expressed as the following formula:

[0016]

[0017] Among them, a d,t , b d,t 、c d,t 、e d,t 、f d,t are the influence functions of different temperatures w on the load at time t on day d; n is the lag days of the real-time temperature w, and the maximum lag days N = 7; w d-n,t is the temperature at time t on day dn; m is the lag time of the real-time temperature w, and the maximum lag time M=10; w d,t-m is the temperature at time tm on day d; is the maximum value of temperature w at time t on day d-1; is the minimum value of temperature w at time t on day d-1; is the average value of the temperature w at time t on day d-7.

[0018] Preferably, the influence function of humidity on load is expressed as the following formula:

[0019]

[0020] Among them, g d,t 、j d,t , k d,t , l d,t , o d,t are the influence functions of different humidity D on the load at time t on day d; n is the lag days of the real-time humidity D, and the maximum lag days N = 7; D d-n,t is the temperature at time t on day dn; m is the lag time of the real-time humidity D, and the maximum lag time M=10; D d,t-m is the humidity at tm on day d; is the maximum value of humidity D at time t on day d-1; is the minimum value of humidity D at time t on day d-1; It is the average value of humidity D at time t on day d-7.

[0021] Preferably, the characteristic value function of the historical load is expressed as follows:

[0022]

[0023] Among them, p d,t ,q d,t 、r d,t 、s d,t 、u d,t are the influence functions of different load characteristic values ​​x on the load at time t on day d; n is the number of lag days considered in the prediction model, and the maximum lag day number N = 7; x d-n,t is the load at time t on day dn; m is the maximum number of lag times considered in the forecast model, and the last lag time M = 10; x d,t-m is the load at time tm on day d; is the maximum value of the load at time t on day d-1; is the minimum value of the load at time t on day d-1; It is the average value of the load at time t on day d-7.

[0024] Preferably, in the decoupled load model, the temperature characteristic function includes:

[0025] The temperature elastic part is calculated using the following formula:

[0026] ΔL d,i =(f0(T d,i )-f0(T c2 ))I(T d,i >T c2 )

[0027] Where I(x) is an indicator function, which returns 1 if x is true, otherwise 0; ΔL d,irepresents the temperature elastic part; f0(x) is the temperature calculation function; T d,i is the temperature at the ith hour on the dth day; T c2 is the preset comparison temperature;

[0028] The cumulative portion is calculated using the following formula:

[0029] C d,i =k i A d-1,i

[0030] Among them, C d,i represents the cumulative part; k i represents the cumulative conduction coefficient; A d-1,i represents the cooling load at the i-th hour on day d-1;

[0031] The cooling load is calculated based on the elastic part and the cumulative part of the temperature. The formula is as follows:

[0032] ΔA d,i =ΔL d,i +C d,i =(f0(T d,i )-f0(T c2 ))I(T d,i >T c2 )+k i A d-1,i

[0033] Among them, A d,i For cooling load.

[0034] Preferably, the method further comprises:

[0035] Dividing the data sample set into a training set and a test set;

[0036] When training a pre-established GAM-based day-ahead load forecasting model, a training set in the data sample set is used;

[0037] The test set is used to calculate the model accuracy of the decoupled load model.

[0038] Preferably, using the test set to calculate the model accuracy of the decoupled load model includes:

[0039] Calculate the average absolute percentage using the following formula:

[0040]

[0041] Among them, e MAPE is the mean absolute percentage error; m is the number of observations; y i is the actual value of the i-th observation; is the predicted value or fitted value of the i-th observation;

[0042] It is determined whether the average absolute percentage converges. If not, the parameters of the decoupled load model are adjusted. If so, the parameters of the current decoupled load model are output as the optimal solution.

[0043] According to a second aspect of an embodiment of the present invention, a decoupled summer and winter temperature control load calculation device is provided, comprising:

[0044] A data acquisition and integration module is used to acquire the load data and temperature data within a preset time period of the area to be tested, and integrate the load data and the temperature data into a data sample set; a model training module is used to use the data sample set to train a pre-established GAM-based day-ahead load forecasting model; a model decoupling module is used to obtain the influence function of temperature on load, the influence function of humidity on load, and the characteristic value function of historical load according to the trained GAM-based day-ahead load forecasting model; the trained GAM-based day-ahead load forecasting model is decoupled using the influence function of temperature on load, the influence function of humidity on load, and the characteristic value function of historical load to obtain a decoupled load model composed of a temperature characteristic function, a humidity characteristic function, and a historical load characteristic function; a load measurement module is used to perform load measurement using the decoupled load model.

[0045] According to a third aspect of an embodiment of the present invention, a decoupled summer and winter temperature control load calculation device is provided, comprising:

[0046] A main controller, and a memory connected to the main controller;

[0047] a memory in which program instructions are stored;

[0048] The main controller is used to execute program instructions stored in the memory to perform any of the above methods.

[0049] The technical solution provided by the embodiments of the present invention may have the following beneficial effects:

[0050] It can be understood that the technical solution shown in the present invention can obtain a data sample set; use the data sample set to train a pre-established day-ahead load forecasting model based on GAM; and then obtain the influence function of temperature on load, the influence function of humidity on load, and the characteristic value function of historical load; use the three characteristic value functions to decouple the day-ahead load forecasting model based on GAM, and obtain a decoupled load model composed of a temperature characteristic function, a humidity characteristic function, and a historical load characteristic function; and use the decoupled load model to perform load measurement. It can be understood that in the technical solution shown in the present invention, the characteristic functions of the temperature term and the historical load term can directly reflect the relationship between temperature and load, providing an important basis for refined analysis of power load. At the same time, it provides a more effective load shedding method for power companies and government departments, which helps to ensure the stable operation of the power grid and the efficient supply of electricity.

[0051] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0053] Figure 1 is a schematic diagram of steps of a method for calculating decoupled summer and winter temperature control loads according to an exemplary embodiment;

[0054] Figure 2 It is a schematic block diagram of a decoupled summer and winter temperature control load calculation device according to an exemplary embodiment. DETAILED DESCRIPTION

[0055] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0056] In one embodiment, Figure 1 is a schematic diagram of steps of a method for calculating decoupled summer and winter temperature control loads according to an exemplary embodiment, see Figure 1 , providing a decoupled summer and winter temperature control load calculation method, including:

[0057] Step S11, obtaining load data and temperature data of a region to be tested within a preset time period, and integrating the load data and the temperature data into a data sample set.

[0058] Step S12: using the data sample set to train a pre-established GAM-based day-ahead load forecasting model.

[0059] Step S13: According to the trained GAM-based day-ahead load forecasting model, the influence function of temperature on load, the influence function of humidity on load and the characteristic value function of historical load are obtained.

[0060] Step S14: Decouple the trained GAM-based day-ahead load forecasting model by using the temperature influence function on load, the humidity influence function on load and the characteristic value function of historical load, and obtain a decoupled load model consisting of the temperature characteristic function, the humidity characteristic function and the historical load characteristic function.

[0061] Step S15: Utilize the decoupled load model to perform load calculation.

[0062] It can be understood that the technical solution shown in this embodiment can obtain a data sample set; use the data sample set to train a pre-established day-ahead load forecasting model based on GAM; and then obtain the influence function of temperature on load, the influence function of humidity on load, and the characteristic value function of historical load; use the three characteristic value functions to decouple the day-ahead load forecasting model based on GAM, and obtain a decoupled load model composed of a temperature characteristic function, a humidity characteristic function, and a historical load characteristic function; and use the decoupled load model to perform load measurement. It can be understood that in the technical solution shown in this embodiment, the characteristic functions of the temperature term and the historical load term can directly reflect the relationship between temperature and load, providing an important basis for refined analysis of power load. At the same time, it provides a more effective load shedding method for power companies and government departments, which helps to ensure the stable operation of the power grid and the efficient supply of electricity.

[0063] In specific practice, in step S11, for example, the load and temperature of 96 points of the power grid in the test area from January 2019 to December 2022 can be collected every day, and the data set can be preprocessed to improve the overall data quality, and the preprocessed data can be integrated into a data sample set.

[0064] It should be noted that, considering that the relationship between load and influencing factors has the characteristics of Poisson distribution, the logarithmic function is set as the connection function of the prediction model, and the trained GAM-based day-ahead load prediction model is expressed as follows:

[0065] log(x d,t )=T d,t (w)+H d,t (D)+L d,t (x d-,t- )

[0066] Among them, x d,t is the forecast load at time t on day d, t=1,2,...,96; T d,t (w) is the effect of temperature w on the load at time t on day d; H d,t (D) is the effect of humidity D on the load at time t on day d; L d,t (x d-,t- ) is the influence of load characteristic value on the predicted load at time t on day d; x d-,t- is the load value before time t on day d.

[0067] In specific practice, the mathematical expression of the GAM model is:

[0068] y=f1(x1)+f2(x2)+…+f n (x n )+ε

[0069] Among them, y is the connection function, and the function form can be customized according to the relationship between the driving variable and the predictor variable; x i is the i-th characteristic variable; n is the total number of characteristics; ε is the error term, which follows the normal distribution ε~(0,σ 2 ), which is independent of the features; f i (x i ) is a nonlinear function related to the i-th feature. The most commonly used one is the B-spline function formed by superimposing multiple B-spline bases, which can quantify the feature x i The relationship between the link function.

[0070] It should be noted that, since the temperature and load are in a nonlinear relationship, the influence function of the temperature on the load is estimated by using a cubic spline function and is expressed as follows:

[0071]

[0072] Among them, a d,t , b d,t 、c d,t 、e d,t 、f d,t are the influence functions of different temperatures w on the load at time t on day d; n is the lag days of the real-time temperature w, and the maximum lag days N = 7; w d-n,t is the temperature at time t on day dn; m is the lag time of the real-time temperature w, and the maximum lag time M=10; w d,t-m is the temperature at time tm on day d; is the maximum value of temperature w at time t on day d-1; is the minimum value of temperature w at time t on day d-1; is the average value of the temperature w at time t on day d-7.

[0073] Similarly, the influence function of humidity on load is expressed as follows:

[0074]

[0075] Among them, g d,t 、j d,t , k d,t , l d,t , o d,t are the influence functions of different humidity D on the load at time t on day d; n is the lag days of the real-time humidity D, and the maximum lag days N = 7; D d-n,t is the temperature at time t on day dn; m is the lag time of the real-time humidity D, and the maximum lag time M=10; D d,t-m is the humidity at tm on day d; is the maximum value of humidity D at time t on day d-1; is the minimum value of humidity D at time t on day d-1; It is the average value of humidity D at time t on day d-7.

[0076] The characteristic value function of the historical load is expressed as follows:

[0077]

[0078] Among them, p d,t ,q d,t 、r d,t 、s d,t 、u d,t are the influence functions of different load characteristic values ​​x on the load at time t on day d; n is the number of lag days considered in the prediction model, and the maximum lag day number N = 7; x d-n,t is the load at time t on day dn; m is the maximum number of lag times considered in the forecast model, and the last lag time M = 10; x d,t-m is the load at time tm on day d; is the maximum value of the load at time t on day d-1; is the minimum value of the load at time t on day d-1; It is the average value of the load at time t on day d-7.

[0079] It should be noted that in the decoupled load model, the temperature load is decomposed into the temperature elastic part and the cumulative part. d,i Take this as an example, and explain as follows:

[0080] Calculate the temperature elasticity part: If the temperature T at the i-th hour on the d-th day d,i Meet T d,i >T c2 , then A d,i The temperature elastic part ΔLd,i f0(T d,i )-f0(T c2 ); if T d,i ≤T c2 , then there is no temperature elastic part, that is, ΔL d,i =0.

[0081] The temperature elastic part is calculated using the following formula:

[0082] ΔL d,i =(f0(T d,i )-f0(T c2 ))I(T d,i >T c2 )

[0083] Where I(x) is an indicator function, which returns 1 if x is true, otherwise 0; ΔL d,i represents the temperature elastic part; f0(x) is the temperature calculation function; T d,i is the temperature at the ith hour on the dth day; T c2 is the preset comparison temperature.

[0084] Calculate the cumulative part: The load has a cumulative effect, and the cooling load, as a part of the load, also has the same cumulative effect. d,i The cumulative portion C d,i The cumulative conduction coefficient k i and the cooling load A at the i-th hour on day d-1 d-1,i Calculate, see the following formula, if there is no cooling load in the i-th hour of the d-1 day, then A d-1,i =0.

[0085] The cumulative portion is calculated using the following formula:

[0086] C d,i =k i A d-1,i

[0087] Among them, C d,i represents the cumulative part; k i represents the cumulative conduction coefficient; A d-1,i It represents the cooling load at the i-th hour on the d-1 day.

[0088] The cooling load is calculated based on the elastic part and the cumulative part of the temperature. The formula is as follows:

[0089] ΔA d,i =ΔL d,i +C d,i =(f0(T d,i )-f0(T c2 ))I(T d,i >Tc2 )+k i A d-1,i

[0090] Among them, A d,i For cooling load.

[0091] Referring to the above formula, the temperature load at that moment can be calculated daily, and the humidity load and historical load can be obtained in the same way.

[0092] It should be noted that the method described above further includes:

[0093] The data sample set is divided into a training set and a test set; when training a pre-established GAM-based day-ahead load forecasting model, the training set in the data sample set is used; and the test set is used to calculate the model accuracy of the decoupled load model.

[0094] In specific practice, the data sample set can be divided into a training set and a test set. The training set can be used to train the model, and the test set can be used to test the model. For example, 96 points of data are collected daily from January 2019 to December 2022 of a local power grid. The 96 points of data daily from January 2019 to December 2021 can be used as the training set, and the 96 points of data daily from January to December 2022 can be used as the test set.

[0095] It should be noted that the model accuracy calculation of the decoupled load model using the test set includes:

[0096] Calculate the average absolute percentage using the following formula:

[0097]

[0098] Among them, e MAPE is the mean absolute percentage error; m is the number of observations; y i is the actual value of the i-th observation; is the predicted or fitted value for the ith observation.

[0099] It is determined whether the average absolute percentage converges. If not, the parameters of the decoupled load model are adjusted. If so, the parameters of the current decoupled load model are output as the optimal solution, thereby achieving the purpose of optimizing the load forecasting model.

[0100] According to the second aspect of the embodiment of the present invention, see Figure 2 , a decoupled summer and winter temperature control load calculation device is provided, comprising:

[0101] The data acquisition and integration module 101 is used to acquire the load data and temperature data of the area to be tested within a preset time period, and integrate the load data and the temperature data into a data sample set.

[0102] The model training module 102 is used to train a pre-established GAM-based day-ahead load forecasting model using the data sample set.

[0103] The model decoupling module 103 is used to obtain the influence function of temperature on load, the influence function of humidity on load and the characteristic value function of historical load according to the trained GAM-based day-ahead load forecasting model; and decouple the trained GAM-based day-ahead load forecasting model by using the influence function of temperature on load, the influence function of humidity on load and the characteristic value function of historical load to obtain a decoupled load model composed of the temperature characteristic function, the humidity characteristic function and the historical load characteristic function.

[0104] The load calculation module 104 is used to perform load calculation using the decoupled load model.

[0105] It can be understood that the technical solution shown in this embodiment can obtain the load data and temperature data within the preset time period of the area to be tested through the data acquisition and integration module 101, and integrate the load data and the temperature data into a data sample set; the model training module 102 uses the data sample set to train the pre-established GAM-based day-ahead load forecasting model; the model decoupling module 103 obtains the influence function of temperature on load, the influence function of humidity on load and the characteristic value function of historical load according to the trained GAM-based day-ahead load forecasting model; the influence function of temperature on load, the influence function of humidity on load and the characteristic value function of historical load are used to decouple the trained GAM-based day-ahead load forecasting model to obtain a decoupled load model composed of a temperature characteristic function, a humidity characteristic function and a historical load characteristic function; and the load is measured through the load measurement module 104. It can be understood that in the technical solution shown in this embodiment, the characteristic functions of the temperature term and the historical load term can directly reflect the relationship between temperature and load, and provide an important basis for fine analysis of power load. At the same time, it provides a more effective load stripping method for power companies and government departments, which helps to ensure the stable operation of the power grid and the efficient supply of electricity.

[0106] According to a third aspect of an embodiment of the present invention, a decoupled summer and winter temperature control load calculation device is provided, comprising:

[0107] A main controller, and a memory connected to the main controller;

[0108] a memory in which program instructions are stored;

[0109] The main controller is used to execute program instructions stored in the memory to perform any of the above methods.

[0110] According to a fourth aspect of an embodiment of the present invention, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the method described in any one of the above is implemented.

[0111] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0112] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0113] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0114] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0115] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0116] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0117] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0118] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0119] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of 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. A decoupled summer and winter temperature control load calculation method, characterized in that: include: Obtaining load data and temperature data within a preset time period in the area to be tested, and integrating the load data and the temperature data into a data sample set; Using the data sample set to train a pre-established GAM-based day-ahead load forecasting model; According to the trained GAM-based day-ahead load forecasting model, the influence function of temperature on load, the influence function of humidity on load and the characteristic value function of historical load are obtained; The trained GAM-based day-ahead load forecasting model is decoupled by using the temperature-load influence function, humidity-load influence function and historical load characteristic value function, and a decoupled load model consisting of temperature characteristic function, humidity characteristic function and historical load characteristic function is obtained. The decoupled load model is used to perform load calculation.

2. The method according to claim 1, characterized in that The trained GAM-based day-ahead load forecasting model is expressed as follows: log(x d,t )=T d,t (w)+H d,t (D)+L d,t (x d-,t- ) Among them, x d,t is the forecast load at time t on day d, t=1,2,...,96; T d,t (w) is the effect of temperature w on the load at time t on day d; Hd,t(D) is the effect of humidity D on the load at time t on day d; L d,t (x d-,t- ) is the influence of load characteristic value on the predicted load at time t on day d; x d-,t- is the load value before time t on day d.

3. The method according to claim 2, characterized in that The influence function of temperature on load is expressed as follows: Among them, a d,t , b d,t 、c d,t 、e d,t 、f d,t are the influence functions of different temperatures w on the load at time t on day d; n is the lag days of the real-time temperature w, and the maximum lag days N = 7; w d-n,t is the temperature at time t on day dn; m is the lag time of the real-time temperature w, and the maximum lag time M = 10; w d,t-m is the temperature at time tm on day d; is the maximum value of temperature w at time t on day d-1; is the minimum value of temperature w at time t on day d-1; is the average value of the temperature w at time t on day d-7.

4. The method according to claim 3, characterized in that The influence function of humidity on load is expressed as follows: Among them, g d,t 、j d,t , k d,t , l d,t , o d,t are the influence functions of different humidity D on the load at time t on day d; n is the lag days of the real-time humidity D, and the maximum lag days N = 7; D d-n,t is the temperature at time t on day dn; m is the lag time of the real-time humidity D, and the maximum lag time M=10; D d,t-m is the humidity at tm on day d; is the maximum value of humidity D at time t on day d-1; is the minimum value of humidity D at time t on day d-1; It is the average value of humidity D at time t on day d-7.

5. The method according to claim 4, characterized in that The characteristic value function of the historical load is expressed as follows: Among them, p d,t ,q d,t 、r d,t 、s d,t 、u d,t are the influence functions of different load characteristic values ​​x on the load at time t on day d; n is the number of lag days considered in the prediction model, and the maximum lag day number N = 7; x d-n,t is the load at time t on day dn; m is the maximum number of lag times considered in the forecast model, and the last lag time M = 10; x d,t-m is the load at time tm on day d; is the maximum value of the load at time t on day d-1; is the minimum value of the load at time t on day d-1; It is the average value of the load at time t on day d-7.

6. The method according to claim 5, characterized in that In the decoupled load model, the temperature characteristic function includes: The temperature elastic part is calculated using the following formula: ΔL d,i =(f0(T d,i )-f0(T c2 ))I(T d,i >T c2 ) Where I(x) is an indicator function, which returns 1 if x is true, otherwise 0; ΔL d,i represents the temperature elastic part; f0(x) is the temperature calculation function; T d,i is the temperature at the i-th hour on the d-th day; T c2 is the preset comparison temperature; The cumulative portion is calculated using the following formula: C d,i =k i A d-1,i Among them, C d,i represents the cumulative part; k i represents the cumulative conduction coefficient; A d-1,i represents the cooling load at the i-th hour on day d-1; The cooling load is calculated based on the elastic part and the cumulative part of the temperature. The formula is as follows: ΔA d,i =ΔL d,i +C d,i =(f0(T d,i )-f0(T c2 ))I(T d,i >T c2 )+k i A d-1,i Among them, A d,i For cooling load.

7. The method according to claim 1, characterized in that Also includes: Dividing the data sample set into a training set and a test set; When training a pre-established GAM-based day-ahead load forecasting model, a training set in the data sample set is used; The test set is used to calculate the model accuracy of the decoupled load model.

8. The method according to claim 7, characterized in that Calculating the model accuracy of the decoupled load model using the test set includes: Calculate the average absolute percentage using the following formula: Among them, e MAPE is the mean absolute percentage error; m is the number of observations; y i is the actual value of the i-th observation; is the predicted value or fitted value of the i-th observation; It is determined whether the average absolute percentage converges. If not, the parameters of the decoupled load model are adjusted. If so, the parameters of the current decoupled load model are output as the optimal solution.

9. A decoupled summer and winter temperature control load calculation device, characterized in that: include: A data acquisition and integration module, used to acquire load data and temperature data within a preset time period in the area to be tested, and integrate the load data and the temperature data into a data sample set; A model training module, used for training a pre-established GAM-based day-ahead load forecasting model using the data sample set; The model decoupling module is used to obtain the influence function of temperature on load, the influence function of humidity on load and the characteristic value function of historical load according to the trained GAM-based day-ahead load forecasting model; the influence function of temperature on load, the influence function of humidity on load and the characteristic value function of historical load are used to decouple the trained GAM-based day-ahead load forecasting model to obtain a decoupled load model composed of the temperature characteristic function, the humidity characteristic function and the historical load characteristic function; The load calculation module is used to perform load calculation using the decoupled load model.

10. A decoupled summer and winter temperature control load calculation device, characterized in that: include: A main controller, and a memory connected to the main controller; a memory in which program instructions are stored; The main controller is used to execute program instructions stored in the memory and execute the method according to any one of claims 1 to 8.

Images