Power load prediction result optimization method and device, terminal equipment and storage medium

Abstract

The invention discloses a power load prediction result optimization method and device, terminal equipment and a storage medium, and the method comprises the steps: building a first environment monitoring window at the time when an abnormality occurs when an abnormality standard experience probability is not greater than a probability threshold value, or else, starting to build a second environment monitoring time window at the starting time of a time period; when the number of the second environment monitoring time windows meets a window number threshold value, setting the demarcation point at the ending moment of the current window, and dividing the time period; when the abnormal window proportion is greater than a proportion threshold value in the left time period, setting a new second environment monitoring time window at the ending moment of the current window, otherwise, establishing a new second environment monitoring time window at the moment when the data is abnormal in the right time period; and comparing the predicted environmental parameters and the actual environmental parameters of the window, and compensating the environmental parameters when the predicted environmental parameters and the actual environmental parameters are inconsistent. According to the invention, the optimization of the power load prediction result is realized, and the accuracy of the prediction result is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system load forecasting, and in particular to a method, device, terminal equipment and storage medium for optimizing power load forecasting results. Background Art

[0002] In the power system, power load forecasting is one of the important links to ensure the stability and safety of power supply. Accurate power load forecasting can help grid dispatchers make reasonable power generation plans and effectively avoid power surplus or shortage, thereby ensuring the economy and reliability of the power system. However, there are many factors that affect power load, among which the change of environmental parameters is a key factor that cannot be ignored.

[0003] Existing power load forecasting methods often do not perform other optimization processing after predicting the load, especially lacking consideration of environmental parameters. Considering the importance of environmental parameters for load forecasting, it can be seen that the prediction results of existing schemes are not accurate. Therefore, in order to improve the accuracy of the prediction results, it is necessary to compensate for environmental parameters in the power load forecast results. Summary of the invention

[0004] The present invention provides a method, device, terminal equipment and storage medium for optimizing power load forecasting results, so as to solve the technical problem that the forecasting results in the prior art are not accurate enough.

[0005] In order to solve the above technical problems, an embodiment of the present invention provides a method for optimizing power load forecasting results, comprising:

[0006] Obtaining the standard empirical probability of abnormality for each time period in the dispatching cycle; wherein the dispatching cycle is divided into a number of time periods; the standard empirical probability of abnormality refers to the standard empirical probability of predicting abnormality in power load data;

[0007] For each of the time periods, if the abnormal standard empirical probability of the time period is less than or equal to the preset probability threshold, a first environmental monitoring time window is established at the time when the power load data is predicted to be abnormal within the time period, and the predicted environmental parameters and actual environmental parameters corresponding to each of the first environmental monitoring time windows are obtained;

[0008] For each of the first environmental monitoring time windows, when the predicted environmental parameters corresponding to the first environmental monitoring time window are inconsistent with the actual environmental parameters, performing an environmental parameter compensation operation on the predicted power load data corresponding to the first environmental monitoring time window;

[0009] If the abnormal standard empirical probability is greater than the probability threshold, a second environment monitoring time window is established at the start time of the time period, and the following operations are repeatedly performed:

[0010] Obtaining predicted environmental parameters and actual environmental parameters corresponding to the current second environmental monitoring time window;

[0011] When the predicted environmental parameters corresponding to the second environmental monitoring time window are inconsistent with the actual environmental parameters, an environmental parameter compensation operation is performed on the predicted power load data corresponding to the second environmental monitoring time window, and the second environmental monitoring time window is marked as an abnormal window;

[0012] When the number of all second environment monitoring time windows in the time period is equal to a preset window number threshold, taking the end time of the current second environment monitoring time window as a dividing point, the time period is divided into a left time period before the dividing point and a right time period after the dividing point;

[0013] If the proportion of abnormal windows in the left time period is greater than the preset proportion threshold, a new second environmental monitoring time window is set at the end time of the current second environmental monitoring time window; otherwise, a new second environmental monitoring time window is established at the moment when the power data is predicted to be abnormal in the right time period.

[0014] As a preferred solution, the method for determining abnormality of predicted power load data includes:

[0015] Acquiring actual power load data corresponding to the predicted power load data;

[0016] Calculating the deviation between the predicted power load data and the actual power load data;

[0017] If the deviation value is greater than a preset deviation threshold, the predicted power load data is determined to be abnormal.

[0018] As a preferred solution, before obtaining the abnormal standard empirical probability of each time period in the scheduling cycle, it also includes:

[0019] Obtain forecasted power load data and actual power load data for several historical dispatch cycles;

[0020] For each of the historical dispatching cycles, the abnormal probability of each time period in the historical dispatching cycle is calculated based on the predicted power load data and the actual power load data of the historical dispatching cycle; wherein the abnormal probability refers to the probability of abnormality in the predicted power load data;

[0021] For each time period, the average value of the abnormality probabilities corresponding to the time period in all the historical scheduling cycles is calculated as the abnormality standard empirical probability of the time period.

[0022] As a preferred solution, the predicted environmental parameters are obtained, including:

[0023] Acquire predicted power load data corresponding to the predicted environmental parameter to be acquired as first data;

[0024] Inputting the first data into a trained environmental parameter prediction model so that the environmental parameter prediction model outputs corresponding predicted environmental parameters according to the first data;

[0025] The training process of the environmental parameter prediction model includes:

[0026] Obtaining some historical power load data and its corresponding historical environmental parameters;

[0027] Using the historical power load data as sample data and using the corresponding historical environmental parameters as labels for the sample data;

[0028] Construct an initial model with power load data as input and environmental parameters as output;

[0029] The initial model is trained according to the sample data to obtain a trained environmental parameter prediction model.

[0030] As a preferred solution, the predicted power load data is obtained, including:

[0031] According to the existing power load forecasting model, obtaining input data corresponding to the power load data to be forecasted;

[0032] The input data is input into the power load prediction model so that the power load prediction model outputs corresponding predicted power load data according to the input data.

[0033] As a preferred solution, the environmental parameter compensation operation includes:

[0034] According to the predicted power load data to be compensated and its corresponding actual power load data, the corresponding compensation parameters are obtained from a preset compensation parameter comparison table;

[0035] The predicted electric power load data to be compensated is compensated according to the compensation parameters.

[0036] As a preferred solution, before obtaining the abnormal standard empirical probability of each time period in the scheduling cycle, it also includes:

[0037] Obtain voltage data and frequency data of power equipment;

[0038] Determining the abnormal fluctuation duration of the power equipment according to the voltage data and the frequency data;

[0039] If the abnormal fluctuation duration is longer than the preset duration, no subsequent operation is performed; otherwise, the subsequent operation is continued.

[0040] Based on the above embodiment, another embodiment of the present invention provides a device for optimizing power load forecasting results, including: a data acquisition module and a result optimization module;

[0041] The data acquisition module is used to obtain the abnormal standard empirical probability of each time period in the scheduling cycle; wherein the scheduling cycle is divided into a number of time periods; the abnormal standard empirical probability refers to the standard empirical probability of predicting abnormalities in the power load data;

[0042] The result optimization module is used to establish a first environmental monitoring time window at the moment when the power load data is predicted to be abnormal within the time period, and obtain the predicted environmental parameters and actual environmental parameters corresponding to each of the first environmental monitoring time windows for each of the time periods if the abnormal standard empirical probability of the time period is less than or equal to a preset probability threshold; for each of the first environmental monitoring time windows, when the predicted environmental parameters corresponding to the first environmental monitoring time window are inconsistent with the actual environmental parameters, perform an environmental parameter compensation operation on the predicted power load data corresponding to the first environmental monitoring time window; if the abnormal standard empirical probability is greater than the probability threshold, establish a second environmental monitoring time window at the starting moment of the time period, and repeat the following operations: obtain the predicted environmental parameters and actual environmental parameters corresponding to the current second environmental monitoring time window Parameters; when the predicted environmental parameters corresponding to the second environmental monitoring time window are inconsistent with the actual environmental parameters, the predicted power load data corresponding to the second environmental monitoring time window is compensated for the environmental parameters, and the second environmental monitoring time window is marked as an abnormal window; when the number of all second environmental monitoring time windows in the time period is equal to the preset window number threshold, the end time of the current second environmental monitoring time window is used as the dividing point, and the time period is divided into a left time period before the dividing point and a right time period after the dividing point; if the proportion of abnormal windows in the left time period is greater than the preset proportion threshold, a new second environmental monitoring time window is set at the end time of the current second environmental monitoring time window, otherwise, a new second environmental monitoring time window is established at the moment when the predicted power data in the right time period is abnormal.

[0043] Based on the above embodiments, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, the power load forecasting result optimization method described in the above invention embodiment is implemented.

[0044] Based on the above embodiments, another embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the power load forecasting result optimization method described in the above invention embodiment.

[0045] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0046] The present invention first determines a method for establishing an environmental monitoring time window according to an abnormal standard empirical probability, selects to establish a first environmental monitoring window at the time when the abnormality occurs when the abnormal standard empirical probability is not greater than a probability threshold, compares the predicted environmental parameters and actual environmental parameters of the first environmental monitoring time window, and performs environmental parameter compensation when the two are inconsistent; selects to start establishing a second environmental monitoring time window at the start time of a time period when the abnormal standard empirical probability is greater than the probability threshold; determines the time of a demarcation point according to the number of all current second environmental monitoring time windows, and sets the demarcation point at the end time of the current second environmental monitoring time window when the number of windows meets the window number threshold, and divides the time period; further determines a method for establishing an environmental monitoring time window of a right time period according to the proportion of abnormal windows in a left time period, and sets a new second environmental monitoring time window at the end time of the current second environmental monitoring time window when the proportion of abnormal windows is greater than the proportion threshold, otherwise, establishes a new second environmental monitoring time window at the time when data in the right time period is abnormal; compares the predicted environmental parameters and actual environmental parameters of the second environmental monitoring time window, and performs environmental parameter compensation when the two are inconsistent. The present invention provides a method for compensating environmental parameters for predicted power load data, which can optimize power load prediction results and improve the accuracy of prediction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a flow chart of a method for optimizing power load forecasting results provided by one embodiment of the present invention;

[0048] Figure 2 It is a structural schematic diagram of a power load forecast result optimization device provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 are within the scope of protection of the present invention.

[0050] Embodiment 1

[0051] Please refer to Figure 1 , is a flow chart of a method for optimizing power load forecasting results provided by an embodiment of the present invention, comprising:

[0052] S1. Obtain the standard empirical probability of abnormality in each time period in the dispatching cycle; wherein the dispatching cycle is divided into a number of time periods; the standard empirical probability of abnormality refers to the standard empirical probability of predicting abnormality in power load data.

[0053] In step S1, the scheduling cycle includes several time periods, and the abnormal standard empirical probability of each time period is obtained.

[0054] It should be noted that the dispatch cycle is generally 1 day, which is divided into several time periods of equal length, and the length of the time period is generally 30 minutes. The standard empirical probability of abnormality in a certain time period refers to the standard empirical probability of abnormality in the predicted power load data corresponding to the time period. The abnormality of predicted power load data mainly refers to the large deviation between the predicted data and the actual data.

[0055] In a preferred embodiment, the method for determining abnormality of predicted power load data includes:

[0056] Acquiring actual power load data corresponding to the predicted power load data;

[0057] Calculating the deviation between the predicted power load data and the actual power load data;

[0058] If the deviation value is greater than a preset deviation threshold, the predicted power load data is determined to be abnormal.

[0059] In this embodiment, a method for determining abnormality of predicted power load data is provided, and predicted power load data whose deviation value from actual power data exceeds a deviation threshold is regarded as abnormal.

[0060] In a preferred embodiment, before obtaining the abnormal standard empirical probability of each time period in the scheduling cycle, it also includes:

[0061] Obtain forecasted power load data and actual power load data for several historical dispatch cycles;

[0062] For each of the historical dispatching cycles, the abnormal probability of each time period in the historical dispatching cycle is calculated based on the predicted power load data and the actual power load data of the historical dispatching cycle; wherein the abnormal probability refers to the probability of abnormality in the predicted power load data;

[0063] For each time period, the average value of the abnormality probabilities corresponding to the time period in all the historical scheduling cycles is calculated as the abnormality standard empirical probability of the time period.

[0064] In this embodiment, before obtaining the abnormal standard empirical probability, it is necessary to calculate the abnormal standard empirical probability. The abnormal standard empirical probability is calculated through historical data. First, the predicted power load data and actual power load data of several historical scheduling cycles are obtained, and then for each historical scheduling cycle, the abnormal probability of each time period in the historical scheduling cycle is calculated, and then for the same time period of all historical scheduling cycles, the average value of the abnormal probability of the time period in different historical scheduling cycles is calculated as the abnormal standard empirical probability of the time period.

[0065] It should be noted that the abnormal probability of a certain time period refers to the probability of abnormality in the predicted power load data corresponding to the time period. The time period division of different dispatch cycles is the same. When calculating the abnormal standard empirical probability, the abnormal probability of the same time period in different historical dispatch cycles is counted, and the average value is calculated as the abnormal standard empirical probability of the time period.

[0066] In a preferred embodiment, before obtaining the abnormal standard empirical probability of each time period in the scheduling cycle, it also includes:

[0067] Obtain voltage data and frequency data of power equipment;

[0068] Determining the abnormal fluctuation duration of the power equipment according to the voltage data and the frequency data;

[0069] If the abnormal fluctuation duration is longer than the preset duration, no subsequent operation is performed; otherwise, the subsequent operation is continued.

[0070] In this embodiment, before step S1, the abnormal fluctuation of the power equipment is judged, and the voltage data and frequency data of the power equipment are obtained to determine the abnormal fluctuation duration of the power equipment. If the abnormal fluctuation duration is greater than the preset duration, it means that the power equipment may be abnormal, and the predicted data of the power equipment is not worth optimizing, so there is no need to continue the subsequent optimization operation; if the abnormal fluctuation duration is not greater than the preset market, it is considered that the power equipment is normal, and the subsequent optimization operation is continued.

[0071] It should be noted that the abnormal fluctuation duration of the power equipment refers to the total duration of abnormal fluctuations of the power equipment within a preset operation cycle.

[0072] S2. For each of the time periods, if the standard empirical probability of abnormality in the time period is less than or equal to a preset probability threshold, a first environmental monitoring time window is established at the moment when the power load data is predicted to be abnormal within the time period, and the predicted environmental parameters and actual environmental parameters corresponding to each of the first environmental monitoring time windows are obtained.

[0073] In step S2, for each time period, when the standard empirical probability of abnormality in the time period is less than or equal to the preset probability threshold, a first environmental monitoring time window is established at the moment when the power load data is predicted to be abnormal within the time period, and then the predicted environmental parameters and actual environmental parameters of each first environmental monitoring time window are obtained.

[0074] It should be noted that multiple anomalies may occur in a certain time period, or no anomalies may occur, so multiple first environmental monitoring time windows may be established, or no first environmental monitoring time window may be established. If there is no first environmental monitoring time window, there is no need to further obtain the predicted environmental parameters and actual environmental parameters, and there is no need to perform environmental parameter compensation.

[0075] In a preferred embodiment, obtaining predicted environmental parameters includes:

[0076] Acquire predicted power load data corresponding to the predicted environmental parameter to be acquired as first data;

[0077] Inputting the first data into a trained environmental parameter prediction model so that the environmental parameter prediction model outputs corresponding predicted environmental parameters according to the first data;

[0078] The training process of the environmental parameter prediction model includes:

[0079] Obtaining some historical power load data and its corresponding historical environmental parameters;

[0080] Using the historical power load data as sample data and using the corresponding historical environmental parameters as labels for the sample data;

[0081] Construct an initial model with power load data as input and environmental parameters as output;

[0082] The initial model is trained according to the sample data to obtain a trained environmental parameter prediction model.

[0083] It should be noted that the predicted environmental parameters are obtained by inputting the predicted power load data into the environmental parameter prediction model for prediction.

[0084] In this embodiment, before obtaining the predicted environmental parameters, the predicted environmental parameters to be obtained are known, and the corresponding predicted power load data is first obtained as the first data. The first data is then input into the trained environmental parameter prediction model, so that the environmental parameter prediction model predicts according to the first data and outputs the corresponding predicted environmental parameters.

[0085] The environmental parameter prediction model is trained based on the historical power load data and its corresponding historical environmental parameters as samples and labels. The historical power load data is the actual power load data, not the predicted one. The historical environmental parameters are also the actual environmental parameters, not the predicted ones.

[0086] In a preferred embodiment, obtaining predicted power load data includes:

[0087] According to the existing power load forecasting model, obtaining input data corresponding to the power load data to be forecasted;

[0088] The input data is input into the power load prediction model so that the power load prediction model outputs corresponding predicted power load data according to the input data.

[0089] In this embodiment, a method for obtaining predicted power load data is provided. According to the existing power load prediction model in the prior art, the input data required for prediction can be first obtained, the input data is input into the power load prediction model for prediction, and the corresponding predicted power load data is output.

[0090] S3. For each of the first environmental monitoring time windows, when the predicted environmental parameters corresponding to the first environmental monitoring time window are inconsistent with the actual environmental parameters, an environmental parameter compensation operation is performed on the predicted power load data corresponding to the first environmental monitoring time window.

[0091] If the first environmental monitoring time window is successfully constructed in step S2 and the predicted environmental parameters and actual environmental parameters corresponding to each of the first environmental monitoring time windows are obtained, then step S3 is continued to be executed; otherwise, step S3 is not executed.

[0092] In step S3, for each first environmental monitoring time window, the predicted environmental parameters corresponding to the first environmental monitoring time window are compared with the actual environmental parameters. If the two are inconsistent, it is necessary to perform an environmental parameter compensation operation on the predicted power data corresponding to the first environmental monitoring time window.

[0093] It should be noted that the environmental parameter compensation operation can adopt existing compensation methods.

[0094] In a preferred embodiment, the environmental parameter compensation operation includes:

[0095] According to the predicted power load data to be compensated and its corresponding actual power load data, the corresponding compensation parameters are obtained from a preset compensation parameter comparison table;

[0096] The predicted electric power load data to be compensated is compensated according to the compensation parameters.

[0097] In this embodiment, a specific environmental parameter compensation method is provided. According to the predicted power load data to be compensated and its corresponding actual power load data, corresponding compensation parameters are obtained from a preset compensation parameter comparison table, and the predicted power load data is compensated according to the compensation parameters.

[0098] It should be noted that the compensation parameter comparison table is obtained based on the analysis and summary of historical experience, and can be applied to various environmental parameter compensation scenarios where the deviation between predicted power load data and actual power load data is large.

[0099] S4. If the abnormal standard empirical probability is greater than the probability threshold, a second environment monitoring time window is established at the start time of the time period, and steps S5 to S8 are repeated.

[0100] It should be noted that step S5 to step S8 is a complete iteration in the loop operation, and the iteration is performed according to the consumption of the time window. When a time window is consumed, the next time window is used as the time window of the next iteration.

[0101] S5. Obtain predicted environmental parameters and actual environmental parameters corresponding to the current second environmental monitoring time window.

[0102] In step S5, the predicted environmental parameters and actual environmental parameters corresponding to the current second environmental monitoring time window are obtained.

[0103] S6. When the predicted environmental parameters corresponding to the second environmental monitoring time window are inconsistent with the actual environmental parameters, an environmental parameter compensation operation is performed on the predicted power load data corresponding to the second environmental monitoring time window, and the second environmental monitoring time window is marked as an abnormal window.

[0104] In step S6, for the current second environmental monitoring time window, the predicted environmental parameters corresponding to the second environmental monitoring time window are compared with the actual environmental parameters. If the two are inconsistent, it is necessary to perform environmental parameter compensation operation on the predicted power data corresponding to the current second environmental monitoring time window, and at the same time mark the current second environmental monitoring time window as an abnormal window.

[0105] S7. When the number of all second environment monitoring time windows within the time period is equal to the preset window number threshold, the end time of the current second environment monitoring time window is used as the dividing point to divide the time period into a left time period before the dividing point and a right time period after the dividing point.

[0106] In step S7, the number of all second environment monitoring time windows generated in the time period currently being processed is counted. If it is equal to the preset window number threshold, it means that this iteration triggers the subsequent time period division operation. Taking the end time of the current second environment monitoring time window as the dividing point, the time period currently being processed is divided into a left time period and a right time period. The left time period is the time period before the dividing point, and the right time period is the time period after the dividing point.

[0107] S8. If the proportion of abnormal windows in the left time period is greater than the preset proportion threshold, a new second environmental monitoring time window is set at the end time of the current second environmental monitoring time window; otherwise, a new second environmental monitoring time window is established at the moment when the power data is predicted to be abnormal in the right time period.

[0108] In step S8, a method for creating a second environment monitoring time window for the right time period is provided. The proportion of abnormal windows in the left time period to all windows in the left time period is counted. If the proportion is greater than a preset proportion threshold, a new second environment monitoring time window is set at the end of the current second environment monitoring time window. If the proportion is not greater than the proportion threshold, a second environment monitoring time window is established at the moment when the power data is predicted to be abnormal in the right time period.

[0109] Embodiment 2

[0110] Please refer to Figure 2 , is a schematic diagram of the structure of a power load forecast result optimization device provided by an embodiment of the present invention, comprising: a data acquisition module and a result optimization module;

[0111] The data acquisition module is used to obtain the abnormal standard empirical probability of each time period in the scheduling cycle; wherein the scheduling cycle is divided into a number of time periods; the abnormal standard empirical probability refers to the standard empirical probability of predicting abnormalities in the power load data;

[0112] The result optimization module is used to establish a first environmental monitoring time window at the moment when the power load data is predicted to be abnormal within the time period, and obtain the predicted environmental parameters and actual environmental parameters corresponding to each of the first environmental monitoring time windows for each of the time periods if the abnormal standard empirical probability of the time period is less than or equal to a preset probability threshold; for each of the first environmental monitoring time windows, when the predicted environmental parameters corresponding to the first environmental monitoring time window are inconsistent with the actual environmental parameters, perform an environmental parameter compensation operation on the predicted power load data corresponding to the first environmental monitoring time window; if the abnormal standard empirical probability is greater than the probability threshold, establish a second environmental monitoring time window at the starting moment of the time period, and repeat the following operations: obtain the predicted environmental parameters and actual environmental parameters corresponding to the current second environmental monitoring time window Parameters; when the predicted environmental parameters corresponding to the second environmental monitoring time window are inconsistent with the actual environmental parameters, the predicted power load data corresponding to the second environmental monitoring time window is compensated for the environmental parameters, and the second environmental monitoring time window is marked as an abnormal window; when the number of all second environmental monitoring time windows in the time period is equal to the preset window number threshold, the end time of the current second environmental monitoring time window is used as the dividing point, and the time period is divided into a left time period before the dividing point and a right time period after the dividing point; if the proportion of abnormal windows in the left time period is greater than the preset proportion threshold, a new second environmental monitoring time window is set at the end time of the current second environmental monitoring time window, otherwise, a new second environmental monitoring time window is established at the moment when the predicted power data in the right time period is abnormal.

[0113] Embodiment 3

[0114] Accordingly, an embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements the method for optimizing the power load forecasting results described in the above-mentioned embodiment of the invention.

[0115] Embodiment 4

[0116] Accordingly, an embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the power load forecasting result optimization method described in the above-mentioned embodiment of the invention.

[0117] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0118] Those skilled in the art can clearly understand that for the sake of convenience and brevity, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0119] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0120] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the device, and various interfaces and lines are used to connect various parts of the entire device.

[0121] The memory can be used to store the computer program, and the processor realizes various functions of the device by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0122] The storage medium is a storage medium, and the computer program is stored in the storage medium. When the computer program is executed by the processor, the steps of each method embodiment described above can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer readable media do not include electric carrier signals and telecommunication signals.

[0123] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for optimizing power load forecasting results, characterized in that: include: Obtaining the standard empirical probability of abnormality for each time period in the dispatching cycle; wherein the dispatching cycle is divided into a number of time periods; the standard empirical probability of abnormality refers to the standard empirical probability of predicting abnormality in power load data; For each of the time periods, if the abnormal standard empirical probability of the time period is less than or equal to the preset probability threshold, a first environmental monitoring time window is established at the time when the power load data is predicted to be abnormal within the time period, and the predicted environmental parameters and actual environmental parameters corresponding to each of the first environmental monitoring time windows are obtained; For each of the first environmental monitoring time windows, when the predicted environmental parameters corresponding to the first environmental monitoring time window are inconsistent with the actual environmental parameters, performing an environmental parameter compensation operation on the predicted power load data corresponding to the first environmental monitoring time window; If the abnormal standard empirical probability is greater than the probability threshold, a second environment monitoring time window is established at the start time of the time period, and the following operations are repeatedly performed: Obtaining predicted environmental parameters and actual environmental parameters corresponding to the current second environmental monitoring time window; When the predicted environmental parameters corresponding to the second environmental monitoring time window are inconsistent with the actual environmental parameters, an environmental parameter compensation operation is performed on the predicted power load data corresponding to the second environmental monitoring time window, and the second environmental monitoring time window is marked as an abnormal window; When the number of all second environment monitoring time windows in the time period is equal to a preset window number threshold, taking the end time of the current second environment monitoring time window as a dividing point, the time period is divided into a left time period before the dividing point and a right time period after the dividing point; If the proportion of abnormal windows in the left time period is greater than the preset proportion threshold, a new second environmental monitoring time window is set at the end time of the current second environmental monitoring time window; otherwise, a new second environmental monitoring time window is established at the moment when the power data is predicted to be abnormal in the right time period.

2. The method for optimizing power load forecasting results according to claim 1, characterized in that: The method for determining abnormality of predicted power load data comprises: Acquiring actual power load data corresponding to the predicted power load data; Calculating the deviation between the predicted power load data and the actual power load data; If the deviation value is greater than a preset deviation threshold, the predicted power load data is determined to be abnormal.

3. The method for optimizing power load forecasting results according to claim 1, characterized in that: Before obtaining the abnormal standard empirical probability of each time period in the scheduling cycle, it also includes: Obtain forecasted power load data and actual power load data for several historical dispatch cycles; For each of the historical dispatching cycles, the abnormal probability of each time period in the historical dispatching cycle is calculated based on the predicted power load data and the actual power load data of the historical dispatching cycle; wherein the abnormal probability refers to the probability of abnormality in the predicted power load data; For each time period, the average value of the abnormality probabilities corresponding to the time period in all the historical scheduling cycles is calculated as the abnormality standard empirical probability of the time period.

4. The method for optimizing power load forecasting results according to claim 1, characterized in that: Get predicted environmental parameters, including: Acquire predicted power load data corresponding to the predicted environmental parameter to be acquired as first data; Inputting the first data into a trained environmental parameter prediction model so that the environmental parameter prediction model outputs corresponding predicted environmental parameters according to the first data; The training process of the environmental parameter prediction model includes: Obtaining some historical power load data and its corresponding historical environmental parameters; Using the historical power load data as sample data and using the corresponding historical environmental parameters as labels for the sample data; Construct an initial model with power load data as input and environmental parameters as output; The initial model is trained according to the sample data to obtain a trained environmental parameter prediction model.

5. The method for optimizing power load forecasting results according to claim 4, characterized in that: Get forecasted power load data, including: According to the existing power load forecasting model, obtaining input data corresponding to the power load data to be forecasted; The input data is input into the power load prediction model so that the power load prediction model outputs corresponding predicted power load data according to the input data.

6. The method for optimizing power load forecasting results according to claim 1, characterized in that: The environmental parameter compensation operation includes: According to the predicted power load data to be compensated and its corresponding actual power load data, the corresponding compensation parameters are obtained from a preset compensation parameter comparison table; The predicted electric power load data to be compensated is compensated according to the compensation parameters.

7. The method for optimizing power load forecasting results according to claim 1, characterized in that: Before obtaining the abnormal standard empirical probability of each time period in the scheduling cycle, it also includes: Obtain voltage data and frequency data of power equipment; Determining the abnormal fluctuation duration of the power equipment according to the voltage data and the frequency data; If the abnormal fluctuation duration is longer than the preset duration, no subsequent operation is performed; otherwise, the subsequent operation is continued.

8. A device for optimizing power load forecasting results, characterized in that: include: Data acquisition module and result optimization module; The data acquisition module is used to obtain the abnormal standard empirical probability of each time period in the scheduling cycle; wherein the scheduling cycle is divided into a number of time periods; the abnormal standard empirical probability refers to the standard empirical probability of predicting abnormalities in the power load data; The result optimization module is used to establish a first environmental monitoring time window at the moment when the power load data is predicted to be abnormal within the time period, and obtain the predicted environmental parameters and actual environmental parameters corresponding to each of the first environmental monitoring time windows for each of the time periods, if the abnormal standard empirical probability of the time period is less than or equal to the preset probability threshold; for each of the first environmental monitoring time windows, when the predicted environmental parameters corresponding to the first environmental monitoring time window are inconsistent with the actual environmental parameters, perform environmental parameter compensation operation on the predicted power load data corresponding to the first environmental monitoring time window; if the abnormal standard empirical probability is greater than the probability threshold, establish a second environmental monitoring time window at the starting moment of the time period, and repeat the following operations: obtain the predicted environmental parameters and actual environmental parameters corresponding to the current second environmental monitoring time window Parameters; when the predicted environmental parameters corresponding to the second environmental monitoring time window are inconsistent with the actual environmental parameters, the predicted power load data corresponding to the second environmental monitoring time window is compensated for the environmental parameters, and the second environmental monitoring time window is marked as an abnormal window; when the number of all second environmental monitoring time windows in the time period is equal to the preset window number threshold, the end time of the current second environmental monitoring time window is used as the dividing point, and the time period is divided into a left time period before the dividing point and a right time period after the dividing point; if the proportion of abnormal windows in the left time period is greater than the preset proportion threshold, a new second environmental monitoring time window is set at the end time of the current second environmental monitoring time window, otherwise, a new second environmental monitoring time window is established at the moment when the predicted power data in the right time period is abnormal.

9. A terminal device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the method for optimizing the power load forecasting results as described in any one of claims 1 to 7 is implemented.

10. A storage medium, characterized in that: The storage medium includes a stored computer program, wherein when the computer program is executed, the device where the storage medium is located is controlled to execute the method for optimizing the power load forecasting results according to any one of claims 1 to 7.

Images