Power grid load forecasting method, device, modeling method, computer equipment and medium

Abstract

The present invention discloses a method, device, modeling method, computer equipment and medium for predicting power grid load. The prediction method comprises: obtaining the net load original data of the power grid; performing data preprocessing on the net load original data to obtain a gray generation sequence; determining a target gray prediction model according to the gray generation sequence, wherein the target gray prediction model is a discrete gray prediction model established based on a target fluctuation coefficient; and determining a net load prediction value according to the target gray prediction model. The present invention constructs an improved gray prediction model by introducing parameters that characterize load fluctuation, and uses the improved gray prediction model to predict power grid load, with high prediction accuracy, which is conducive to improving the economic dispatch and safe operation performance of the power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of power system load forecasting, and in particular to a power grid load forecasting method, device, modeling method, computer equipment and medium. Background Art

[0002] With the access of renewable energy sources such as hydropower and photovoltaics to the power grid, the volatility and randomness of the power grid have increased, and the economic dispatch and stable operation of the power grid are facing greater challenges. Net load is the difference between the microgrid load and the output of renewable energy. Its accurate prediction is the key to ensure the economic dispatch and safe operation of the microgrid.

[0003] In the prior art, commonly used load forecasting methods include: traditional gray forecast model (Gray Forecast Model, GM) forecasting method and discrete gray forecast model (Discrete Gray Forecast Model, GM) forecasting method. The gray forecast model can predict a small amount of net load data series with low data integrity and reliability.

[0004] However, the existing prediction methods have the following problems: the traditional grey prediction model has certain errors in the process of converting differential solutions to differential solutions; the discrete grey prediction model does not fully consider the volatility and randomness of load data after the access of renewable energy. When the above two models are applied to the renewable energy microgrid system, the prediction results have certain deviations, and the prediction accuracy of load data is low, which will affect the safe operation and economic dispatch of the power grid. Summary of the invention

[0005] The present invention provides a power grid load forecasting method, device, modeling method, computer equipment and medium, so as to improve the grey forecasting model by introducing parameters representing load volatility, and achieve high forecasting accuracy.

[0006] According to one aspect of the present invention, a method for predicting power grid load is provided, comprising:

[0007] Obtain the original data of net load of the power grid;

[0008] Performing data preprocessing on the raw data of the net load to obtain a grey generation sequence;

[0009] Determining a target grey prediction model according to the grey generation sequence, wherein the target grey prediction model is a discrete grey prediction model established based on a target fluctuation coefficient;

[0010] The net load forecast value is determined according to the target grey forecast model.

[0011] Optionally, the function expression of the target grey prediction model is:

[0012] x (1) (k+1)=μ1×x (1) (k) α +μ2

[0013] Among them, α is the volatility coefficient, x (1) (k+1) is the (k+1)th data in the gray generation sequence, x (1) (k) is the kth data in the gray generation sequence, μ1 is the development coefficient, and μ2 is the gray action amount.

[0014] According to another aspect of the present invention, a method for modeling a power grid load prediction model is provided, which is used in the above load prediction method, and the modeling method comprises:

[0015] Obtain the original data of net load of the power grid;

[0016] Performing data preprocessing on the raw data of the net load to obtain a grey generation sequence;

[0017] Create a discrete grey prediction model with fluctuation coefficient;

[0018] Determine the target fluctuation coefficient, development coefficient and grey action amount of the discrete grey prediction model according to the grey generation sequence;

[0019] A target grey prediction model is created according to the target fluctuation coefficient, the development coefficient and the grey action.

[0020] According to another aspect of the present invention, there is provided a power grid load forecasting device for executing the above-mentioned power grid load forecasting method, the device comprising: an original data acquisition unit for acquiring the original net load data of the power grid; a data processing unit for performing data preprocessing on the original net load data to obtain a grey generation sequence; a model correction unit for determining a target grey prediction model according to the grey generation sequence, the target grey prediction model being a discrete grey prediction model established based on a target fluctuation coefficient; and a prediction execution unit for determining a net load prediction value according to the target grey prediction model.

[0021] According to another aspect of the present invention, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the above-mentioned power grid load forecasting method is implemented; or, the above-mentioned modeling method is implemented.

[0022] According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the program implements the above-mentioned power grid load forecasting method; or, implements the above-mentioned modeling method.

[0023] The technical solution of the embodiment of the present invention obtains a gray generation sequence by performing data preprocessing on the net load original data, and determines a target gray prediction model based on the gray generation sequence, wherein the target gray prediction model is a discrete gray prediction model established based on a target fluctuation coefficient. The net load prediction value is determined according to the target gray prediction model, which solves the problem of low load prediction accuracy of the existing renewable energy microgrid. The gray prediction model is improved by introducing a fluctuation coefficient that characterizes the degree of change of load data, which is beneficial to improving the prediction accuracy and improving the economic dispatch and safe operation performance of the renewable energy microgrid. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 is a flow chart of a power grid load prediction method provided by Embodiment 1 of the present invention;

[0026] Figure 2 is a flow chart of another power grid load forecasting method provided by Embodiment 1 of the present invention;

[0027] Figure 3 is a flow chart of another power grid load prediction method provided by the first embodiment of the present invention;

[0028] Figure 4 It is a flow chart of a method for modeling a power grid load prediction model provided by the second embodiment of the present invention;

[0029] Figure 5 It is a structural schematic diagram of a power grid load prediction device provided by Embodiment 3 of the present invention;

[0030] Figure 6 It is a structural diagram of a computer device provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0032] It should be noted that the terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or apparatus.

[0033] Embodiment 1

[0034] Figure 1 This is a flow chart of a power grid load forecasting method provided in Embodiment 1 of the present invention. This embodiment can be applied to the application scenario of load forecasting of a renewable energy microgrid, wherein the renewable energy microgrid can be a hydropower microgrid, a photovoltaic microgrid, or a wind power generation microgrid. The method can be executed by a power grid load forecasting device, which can be implemented in the form of hardware and / or software, and can be configured in a power grid dispatching system.

[0035] like Figure 1 As shown, the power grid load prediction method specifically includes the following steps:

[0036] Step S1: Obtain the original data of the net load of the power grid.

[0037] Among them, the net load raw data is the data calculated by deducting the renewable energy output raw data from the grid load raw data; the grid load raw data is the sum of the power consumption of various types of electrical equipment borne by the grid at any time; the renewable energy output raw data is the power output power of the renewable energy power generation system. Typically, the renewable energy power generation system can be a hydropower system.

[0038] In this step, the original data of the net load of the power grid is obtained, which specifically includes the following steps: obtaining a sampling period and a sampling step, recording the original data of the power grid load and the original data of renewable energy output of multiple consecutive sampling nodes based on the sampling step and the sampling period, and calculating the original data of the net load of each sampling node according to the original data of the power grid load and the original data of renewable energy output.

[0039] The sampling period is a period of time before the target period of net load forecasting.

[0040] For example, if the number of data in the payload original data is defined as N, the payload original data sequence can be expressed as: (0) ={x (0) (1), x (0) (2), ..., x (0) (N)}, where X (0) Represents the original data sequence of the payload, x(0) (1) represents the net load raw data of the first sampling node, x (0) (2) represents the net load raw data of the second sampling node, ..., x (0) (N) represents the net load raw data of the Nth sampling node.

[0041] Step S2: preprocess the original data of the net load to obtain a grey generation sequence.

[0042] Among them, the gray generated sequence is a data sequence with load change regularity established based on the original data of net load. In a microgrid equipped with renewable energy, the original data of net load has large volatility and stability. Using the gray generated sequence as model training data can reduce the impact of the randomness of the original data of net load on model training.

[0043] Optionally, the data preprocessing may include an accumulation generation operation, and the gray generation sequence includes: a net load accumulation generation sequence.

[0044] For example, if the number of data in the payload original data is defined as N, the gray generated sequence can be expressed as X (1) ={x (1) (1), x (1) (2), ..., x (1) (N)}, where X (1) represents the net load accumulation generation sequence, x (1) (1) represents the accumulated net load data of the first sampling node, x (1) (2) represents the net load accumulation data of the second sampling node, ..., c (1) (N) represents the net load accumulation data of the Nth sampling node. The net load accumulation data and the net load original data satisfy: x (1) (k) represents the accumulated net load data of the kth sampling node, x (0) (i) represents the net load raw data of the i-th sampling node, k and i are positive integers greater than or equal to 1. Based on the above formula, x (1) (1) is equal to the net load raw data of the first sampling node, x (1) (2) is equal to the cumulative sum of the net load raw data from the first sampling node to the net load raw data from the second sampling node, ..., x (1) (N) is equal to the cumulative sum of the net load raw data from the 1st sampling node to the net load raw data of the Nth sampling node.

[0045] Step S3: determining a target grey prediction model according to the grey generation sequence, wherein the target grey prediction model is a discrete grey prediction model established based on a target fluctuation coefficient.

[0046] Among them, the target grey prediction model can be a first-order univariate discrete grey model containing three parameters. The three parameters of the model include fluctuation coefficient, development coefficient and grey action. Among them, the fluctuation coefficient is a parameter that characterizes the degree of change of the original data of the net load, and the target fluctuation coefficient is the target value of the fluctuation coefficient calculated based on the specific data in the grey generation sequence; the development coefficient is a parameter that characterizes the development trend of the predicted value of the grey generation sequence; and the grey action is a parameter that reflects the relationship between data changes.

[0047] Optionally, the function expression of the target grey prediction model can be:

[0048] x (1) (k+1)=μ1×x (1) (k) α +μ2 (Formula 1)

[0049] Among them, α is the volatility coefficient, x (1) (k+1) is the (k+1)th data in the gray generation sequence, x (1) (k) is the kth data in the grey generation sequence, μ1 is the development coefficient, and μ2 is the grey action amount.

[0050] Combined with the reference formula 1, substituting k=1, 2, 3, ..., N-1 into the above formula 1, the following simultaneous equations are obtained:

[0051]

[0052] Introducing matrix-vector notation:

[0053]

[0054]

[0055]

[0056] In this step, the assignment method and error analysis method can be used to calculate the fluctuation coefficient α, and the least squares method can be used to calculate the development coefficient μ1 and the gray action μ2. The calculated target fluctuation coefficient α1, development coefficient μ1 and gray action μ2 are substituted into the above formula 1 to obtain the final target gray prediction model.

[0057] Step S4: Determine the net load forecast value according to the target grey forecast model.

[0058] In this step, the data in the gray generation sequence can be substituted into the target gray prediction model to calculate the predicted value of the net load accumulation data in the time period after the sampling period. The net load forecast value is calculated based on the inverse process of the accumulation generation operation

[0059] Specifically, before predicting the net load for the target period, the period before the target period is first determined as the sampling period, and the sampling step is set. The original data of the grid load and the original data of renewable energy output of N sampling nodes are recorded. The original data of the grid load of the same sampling node is subtracted from the original data of renewable energy output to obtain the original data of the net load corresponding to the N sampling nodes. The N original data of net load are accumulated and generated to obtain the gray generation sequence X (1) (i.e., the net load accumulation generation sequence). (1) Substitute the data in the pre-established three-parameter discrete grey prediction model. The three parameters include: fluctuation coefficient α, development coefficient μ1 and grey action μ2. Calculate the target fluctuation coefficient α1 by the assignment method and error analysis method, and calculate the development coefficient μ1 and grey action μ2 by the least squares method. Substitute the calculated target fluctuation coefficient α1, development coefficient μ1 and grey action μ2 into the above formula 1 to obtain the final target grey prediction model. Then, substitute the data in the grey generation sequence into the target grey prediction model to calculate the net load cumulative data prediction value of the time period after the sampling period. The net load forecast value is calculated based on the inverse process of the accumulation generation operation

[0060] Therefore, by preprocessing the original data of the net load, a gray generation sequence is obtained, and the target gray prediction model is determined according to the gray generation sequence, wherein the target gray prediction model is a discrete gray prediction model established based on the target fluctuation coefficient. The net load prediction value is determined according to the target gray prediction model, which solves the problem that the prediction results of the existing prediction methods have large deviations after the renewable energy is connected to the power grid. By introducing the fluctuation coefficient that characterizes the degree of change of the load data, the gray prediction model is improved, which is conducive to improving the prediction accuracy.

[0061] Optionally, Figure 2 is a flow chart of another power grid load forecasting method provided by the first embodiment of the present invention. Figure 1 Based on this, a specific implementation method for determining the target fluctuation coefficient is exemplarily shown.

[0062] like Figure 2 As shown, the above step S3 specifically includes the following steps:

[0063] Step S301: Obtain a discrete grey prediction model containing a fluctuation coefficient.

[0064] Among them, the function structure of the discrete grey prediction model is the same as the function structure of the target grey prediction model, and the specific function expression is shown in Formula 1.

[0065] Step S302: assigning values ​​to the fluctuation coefficient of the discrete grey prediction model based on a preset assignment interval and a preset step size to obtain an assignment model.

[0066] Among them, the fluctuation coefficient is a parameter that characterizes the degree of change of the original data of the net load. According to the difference in the data in the sequence, the fluctuation coefficient may fluctuate around the value of 1.

[0067] Optionally, the preset value assignment interval may be: [0.5, 1.5]; the preset step size may be 0.01.

[0068] For example, in combination with reference formula 1, if the volatility coefficient is defined as 1.01, the corresponding assignment model is: (1) (k+1)=μ1×x (1) (k) 1.01 +μ2.

[0069] Step S303: Perform error analysis on the predicted value and the true value of the assignment model, and determine the target fluctuation coefficient according to the error analysis objective function.

[0070] The predicted value of the assignment model is the net load prediction value calculated according to the assignment model, and the true value is the net load original data calculated according to the sampling data.

[0071] Optionally, the error analysis objective function may be a minimum average absolute percentage error between the predicted value and the true value as the objective function, wherein the average absolute percentage error is the average percentage of the deviation between the absolute value of the deviation between all individual net load predicted values ​​and the net load original data and the actual value; or, a minimum average absolute error between the predicted value and the true value as the objective function, wherein the average absolute error is the average of the absolute values ​​of the deviation between all individual net load predicted values ​​and the net load original data.

[0072] For example, the minimum average absolute percentage error between the predicted value and the true value is used as the objective function. If the net load original data sequence X is defined (0) The number of data in is N, then the expression of the error analysis objective function is shown in Formula 2:

[0073]

[0074] Where, k = 1, 2, ..., N; Represents the predicted value of the assignment model.

[0075] The minimum average absolute error between the predicted value and the true value is the objective function. If the net load original data sequence X is defined as (0) The number of data in is N, then the expression of the error analysis objective function is shown in Formula 3:

[0076]

[0077] Where, k = 1, 2, ..., N; Represents the predicted value of the assignment model.

[0078] Optionally, determining the target fluctuation coefficient according to the error analysis objective function includes: determining the target fluctuation coefficient according to the minimum average absolute percentage error between the predicted value of the assignment model and the true value; or determining the target fluctuation coefficient according to the minimum average absolute error between the predicted value of the assignment model and the true value.

[0079] Specifically, in combination with Formula 2 and Formula 3, the volatility coefficient of the assignment model corresponding to the minimum mean absolute percentage error is determined as the target volatility coefficient; or, the volatility coefficient of the assignment model corresponding to the minimum mean absolute error is determined as the target volatility coefficient.

[0080] Step S304: Determine the target grey prediction model according to the target fluctuation coefficient.

[0081] Specifically, the target fluctuation coefficient is defined as α1, and the corresponding target grey prediction model is DGM(1,1,α1). The least square method is used to calculate the grey parameters (development coefficient μ1 and grey action μ2) of the target grey prediction model DGM(1,1,α1) to ensure that the true value x (1) (k+1) and net load forecast value The minimum simulation error S satisfies:

[0082]

[0083] Matlab software is used to perform iterative operations, calculate the development coefficient μ1 and grey action μ2 of DGM (1,1, α1), and obtain the final target grey prediction model.

[0084] Therefore, the technical solution of the embodiment of the present invention constructs an improved grey prediction model by introducing parameters that characterize load volatility, and uses the improved grey prediction model to predict power grid load. The model is simple and has high prediction accuracy.

[0085] Optionally, Figure 3 is a flowchart of another power grid load prediction method provided by the first embodiment of the present invention. Figure 1 On this basis, before executing the prediction, a specific implementation method of model verification is added to ensure the accuracy of model prediction.

[0086] like Figure 3 As shown, the load forecasting method specifically includes the following steps:

[0087] Step S1: Obtain the original data of the net load of the power grid.

[0088] Step S2: preprocess the original data of the net load to obtain a grey generation sequence.

[0089] Step S3: Determine the target grey prediction model according to the grey generation sequence.

[0090] Step S501: performing error analysis on the predicted value and the true value of the target grey prediction model.

[0091] Step S502: Correct the fluctuation coefficient of the target grey prediction model according to the error analysis result.

[0092] Step S503: Establish a new target grey prediction model according to the corrected fluctuation coefficient.

[0093] Optionally, the error analysis method includes at least one of the following: mean absolute error analysis method, mean absolute percentage error analysis method and accuracy grade evaluation method.

[0094] The mean absolute error analysis method is the average of the absolute values ​​of the deviations between all individual net load prediction values ​​and the net load original data. The expression of the mean absolute error refers to the above formula three.

[0095] The mean absolute percentage error analysis method is the average percentage of the absolute value of the deviation between all individual net load forecast values ​​and the net load original data and the actual value. The expression of the mean absolute percentage error can refer to the above formula 2.

[0096] The standards for accuracy grade evaluation can be found in Table 1 below.

[0097] Accuracy level Relative error Level 1 (Excellent) 0.01 Level 2 (good) 0.05 Level 3 (poor) 0.10 Level 4 (failed) 0.20

[0098] As shown in Table 1, the relative error can be the average absolute percentage error, that is, if the average percentage of the absolute value of the deviation between the single net load prediction value of the target grey prediction model and the net load original data and the actual value is less than 0.01, the accuracy level of the target grey prediction model is level one (excellent).

[0099] Specifically, after performing error analysis on the predicted value and the true value of the target grey prediction model, if the error of the target grey prediction model is less than the preset error threshold, or the accuracy level of the target grey prediction model reaches the preset level (e.g., level 1), the target grey prediction model is determined to meet the requirements; if the error of the target grey prediction model is greater than the preset error threshold, or the accuracy level of the target grey prediction model does not reach the preset level (e.g., level 1), the fluctuation coefficient of the target grey prediction model is corrected until the error or accuracy level of the target grey prediction model reaches the set standard. In the subsequent load forecasting process, the corrected target grey prediction model is used to calculate the net load forecast value.

[0100] The prediction method of the present invention is described in detail below with reference to specific examples.

[0101] Exemplarily, the sampling period can be defined as 7:00-9:00, the sampling step is 15 minutes, the original data of the grid load and the original data of the renewable energy output of each sampling node are recorded, and a set of sampling data as shown in Table 2 is established. The sampling period is 2 hours, the sampling step is 15 minutes, and the data of 8 sampling nodes are recorded in total.

[0102]

[0103]

[0104] Combined with reference to Table 2, the net load raw data corresponding to the 8 sampling nodes are calculated using the method described in step S1 as follows:

[0105] X (0) ={448.11, 618.92, 783.93, 947, 1100.95, 1273.3, 1427.89, 1579.83}

[0106] Among them, 448.11 is the original data of the net load of the first sampling node, 618.92 is the original data of the net load of the second sampling node, ..., 1579.83 is the original data of the net load of the eighth sampling node.

[0107] After obtaining the original data of the net load, the method described in step S2 is used to perform the accumulation generation operation to obtain the gray generation sequence:

[0108] X (1) ={448.11, 1067.03, 1850.62, 2797.62, 3898.57, 5171.87, 6599.76, 8179.59}

[0109] After getting the payload original data sequence X (0) and gray generated sequence X (1) Afterwards, the gray parameters of the target gray prediction model are determined by the method described in the above steps S301 to S305. The minimum mean absolute percentage error between the predicted value and the true value is taken as the target function. The target fluctuation coefficient α1 is calculated to be 0.94. At this time, the target gray prediction model is DGM (1, 1, 0.94). Matlab software is used to perform iterative operations and the development coefficient μ1 = 1.9846 and the gray action μ2 = 455.0049 of DGM (1, 1, 0.94) are calculated. Based on this, the function expression of the target gray prediction model is: (1)(k+1)=1.9846×x (1) (k) 0.94 +455.0049.

[0110] Using the method described in step S4 above, the gray generation sequence X (1) Substitute the data in into the target grey prediction model, and calculate the net load prediction values ​​of each sampling node as follows:

[0111] {448.11, 623.45, 782.67, 943.64, 1106.12, 1268.29, 1427.90, 1582.62}.

[0112] By adopting the method recorded in the above step S501, an error analysis is performed on the net load original data and the net load forecast value of each sampling node, and the relative error of the target grey prediction model is obtained to be 0.0028, and the relative error is less than 0.01. It is determined that the accuracy level of the target grey prediction model is level one, and the net load forecast for the target period can be carried out according to the target grey prediction model that meets the requirements.

[0113] Exemplarily, combined with the data in Table 2, taking the target period for load forecasting as 9:00-10:00 as an example, four forecast results can be obtained, namely: the net load forecast value at 9:15 is 1730.24; the net load forecast value at 9:30 is 1868.77; the net load forecast value at 9:45 is 1996.51; and the net load forecast value at 10:00 is 2112.07.

[0114] Therefore, the technical solution of the embodiment of the present invention solves the problem of low load prediction accuracy of the existing renewable energy microgrid by introducing parameters that characterize load volatility and determining the net load prediction value based on the improved discrete gray prediction model. The gray prediction model is improved by introducing a fluctuation coefficient that characterizes the degree of change of load data, which is beneficial to improving the prediction accuracy and improving the economic dispatch and safe operation performance of the renewable energy microgrid.

[0115] Embodiment 2

[0116] Based on the above embodiments, a second embodiment of the present invention provides a method for modeling a power grid load prediction model, which is used in the above load prediction method. The prediction model in this embodiment introduces parameters that characterize load volatility to improve the discrete grey prediction model.

[0117] Figure 4 It is a flow chart of a method for modeling a power grid load prediction model provided in the second embodiment of the present invention.

[0118] like Figure 4 As shown, the modeling method specifically includes the following steps:

[0119] Step S10: Obtaining the original data of the net load of the power grid.

[0120] Step S20: preprocess the original payload data to obtain a grey generation sequence.

[0121] Step S30: creating a discrete grey prediction model containing a fluctuation coefficient, wherein the discrete grey prediction model is a first-order univariate discrete grey prediction model.

[0122] Among them, the fluctuation coefficient is a parameter that characterizes the degree of change of the original data of the net load.

[0123] Step S40: Determine the target fluctuation coefficient, development coefficient and grey action amount of the discrete grey prediction model according to the grey generation sequence.

[0124] Step S50: Determine the target grey prediction model according to the target fluctuation coefficient, development coefficient and grey action amount.

[0125] Optionally, the function expression of the target grey prediction model is: (1) (k+1)=μ1×x (1) (k) α +μ2; where α is the volatility coefficient, x (1) (k+1) is the (k+1)th data in the gray generation sequence, x (1) (k) is the kth data in the grey generation sequence, μ1 is the development coefficient, and μ2 is the grey action amount.

[0126] Optionally, the target fluctuation coefficient of the discrete grey prediction model is determined according to the grey generation sequence, including: assigning a value to the fluctuation coefficient of the discrete grey prediction model based on a preset assignment interval and a preset step size to obtain an assignment model; performing an error analysis on the predicted value and the true value of the assignment model, and determining the target fluctuation coefficient according to the error analysis objective function.

[0127] Optionally, determining the target fluctuation coefficient according to the error analysis objective function includes: determining the target fluctuation coefficient according to the minimum average absolute percentage error between the predicted value of the assignment model and the true value; or determining the target fluctuation coefficient according to the minimum average absolute error between the predicted value of the assignment model and the true value.

[0128] Optionally, the preset value assignment interval is: [0.5, 1.5]; the preset step size is 0.01.

[0129] Therefore, the technical solution of the embodiment of the present invention, by introducing parameters that characterize load volatility to construct an improved grey prediction model, is conducive to improving the prediction accuracy of the model and enhancing the economic dispatch and safe operation performance of the renewable energy microgrid.

[0130] Embodiment 3

[0131] Based on any of the above embodiments, embodiment three of the present invention provides a power grid load forecasting device, which can execute the power grid load forecasting method provided by any of the above embodiments, or the modeling method provided by any of the above embodiments, and has functional modules and beneficial effects corresponding to the execution method.

[0132] Figure 5 It is a structural schematic diagram of a power grid load prediction device provided in Embodiment 3 of the present invention.

[0133] like Figure 5 As shown, the device 00 includes: an original data acquisition unit 101, a data processing unit 102, a model correction unit 103 and a prediction execution unit 104. The original data acquisition unit 101 is used to obtain the original net load data of the power grid; the data processing unit 102 is used to perform data preprocessing on the original net load data to obtain a gray generation sequence; the model correction unit 103 is used to determine a target gray prediction model according to the gray generation sequence, and the target gray prediction model is a discrete gray prediction model established based on a target fluctuation coefficient; the prediction execution unit 104 is used to determine a net load prediction value according to the target gray prediction model.

[0134] Optionally, the function expression of the target grey prediction model is: (1) (k+1)=μ1×x (1) (k) α +μ2; where α is the volatility coefficient, x (1) (k+1) is the (k+1)th data in the gray generation sequence, x (1) (k) is the kth data in the grey generation sequence, μ1 is the development coefficient, and μ2 is the grey action amount.

[0135] Optionally, the model correction unit 103 is used to execute: obtaining a discrete gray prediction model containing a fluctuation coefficient; assigning a value to the fluctuation coefficient of the discrete gray prediction model based on a preset assignment interval and a preset step size to obtain an assignment model; performing an error analysis on the predicted value and the true value of the assignment model, and determining a target fluctuation coefficient based on an error analysis objective function; and determining a target gray prediction model based on the target fluctuation coefficient.

[0136] Optionally, determining the target fluctuation coefficient according to the error analysis objective function includes: determining the target fluctuation coefficient according to the minimum average absolute percentage error between the predicted value of the assignment model and the true value; or determining the target fluctuation coefficient according to the minimum average absolute error between the predicted value of the assignment model and the true value.

[0137] Optionally, the preset value assignment interval is: [0.5, 1.5]; the preset step size is 0.01.

[0138] Embodiment 4

[0139] Based on any of the above embodiments, embodiment 4 of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the above-mentioned power grid load forecasting method is implemented; or, the above-mentioned power grid load forecasting method is implemented.

[0140] Figure 6 : is a schematic diagram of the structure of a computer device provided in Embodiment 4 of the present invention. The computer device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The computer device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0141] like Figure 6 As shown, the computer device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the computer device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0142] A number of components in the computer device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the computer device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0143] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 11 executes the various methods and processes described above, such as a power grid load forecasting method or a modeling method.

[0144] In some embodiments, the above-mentioned power grid load forecasting method or modeling method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 18 .

[0145] In some embodiments, part or all of the computer program may be loaded and / or installed on the computer device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the power grid load forecasting method or modeling method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the power grid load forecasting method or modeling method in any other appropriate manner (e.g., by means of firmware).

[0146] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0147] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0148] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, 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 of the foregoing.

[0149] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0150] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0151] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0152] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0153] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting power grid load, characterized in that: include: Obtain the original data of net load of the power grid; Performing data preprocessing on the raw data of the net load to obtain a grey generation sequence; Determining a target grey prediction model according to the grey generation sequence, wherein the target grey prediction model is a discrete grey prediction model established based on a target fluctuation coefficient; Determine a net load forecast value according to the target grey forecast model; Determining a target grey prediction model according to the grey generation sequence includes: Obtain discrete grey prediction model with fluctuation coefficient; Assigning a value to the fluctuation coefficient of the discrete grey prediction model based on a preset assignment interval and a preset step size to obtain an assignment model; Performing error analysis on the predicted value and the true value of the assignment model, and determining the target fluctuation coefficient according to the error analysis objective function; Determining the target grey prediction model according to the target fluctuation coefficient; Determining the target fluctuation coefficient according to the error analysis objective function includes: Determine the target fluctuation coefficient according to the minimum mean absolute percentage error between the predicted value and the true value of the assignment model; or The target fluctuation coefficient is determined according to the minimum mean absolute error between the predicted value and the true value of the assignment model.

2. The method according to claim 1, characterized in that: The functional expression of the target grey prediction model is: x (1) (k+1)=μ1×x (1) (k) α +μ2 Among them, α is the volatility coefficient, x (1) (k+1) is the (k+1)th data in the gray generation sequence, x (1) (k) is the kth data in the gray generation sequence, μ1 is the development coefficient, and μ2 is the gray action amount.

3. The method according to claim 1, characterized in that The preset value assignment interval is: [0.5, 1.5]; the preset step size is 0.

01.

4. The method according to claim 1, characterized in that: Also includes: Performing error analysis on the predicted value and the true value of the target grey prediction model; The fluctuation coefficient of the target grey prediction model is corrected according to the error analysis result.

5. A method for modeling a power grid load prediction model, characterized in that: For the load forecasting method described in any one of claims 1 to 4, the modeling method comprises: Obtain the original data of net load of the power grid; Performing data preprocessing on the raw data of the net load to obtain a grey generation sequence; Create a discrete grey prediction model with fluctuation coefficient; Determine the target fluctuation coefficient, development coefficient and grey action amount of the discrete grey prediction model according to the grey generation sequence; A target grey prediction model is created according to the target fluctuation coefficient, the development coefficient and the grey action.

6. A power grid load prediction device, characterized in that: The device is used to execute the power grid load forecasting method according to any one of claims 1 to 4, comprising: A raw data acquisition unit, used to acquire raw data of net load of the power grid; A data processing unit, used for performing data preprocessing on the raw data of the net load to obtain a grey generation sequence; A model correction unit, used for determining a target grey prediction model according to the grey generation sequence, wherein the target grey prediction model is a discrete grey prediction model established based on a target fluctuation coefficient; The prediction execution unit is used to determine the net load prediction value according to the target grey prediction model.

7. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the power grid load forecasting method as described in any one of claims 1 to 4 is implemented; Alternatively, implement the power grid load prediction modeling method as described in claim 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the power grid load forecasting method as described in any one of claims 1 to 4 is implemented; Alternatively, implement the power grid load prediction modeling method as described in claim 5.

Images