Power load prediction method and device, electronic equipment and storage medium

Abstract

The invention provides a power load prediction method and device, electronic equipment and a storage medium. The method comprises the steps of obtaining first target data of a power system, the first target data comprising historical load data and meteorological data, performing data preprocessing on the first target data to obtain second target data, and predicting power load data of the power system according to the second target data based on a target load prediction model, the target load prediction model is constructed based on a Light GBM algorithm. Through the target load prediction model with higher prediction precision provided by the application, in combination with the preprocessed first target data, the accuracy of the predicted power load data can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of power load forecasting, and in particular to a power load forecasting method, device, electronic device and storage medium. Background Art

[0002] With the rapid development of social economy, the demand for electricity in various fields is increasing rapidly, and the stability requirements of the power system are increasing. Accurate load forecasting can help power system operators to conduct efficient power dispatching and resource allocation to meet the electricity needs of users, while reducing energy waste and improving economic benefits. It is of great significance to the optimal dispatching and management of the power grid.

[0003] Different technical solutions have their specific scenarios and limitations. Traditional prediction methods rely too much on historical data and give less consideration to uncertain factors; machine learning prediction methods are easily affected by irrelevant attributes and data imbalance; deep learning prediction methods have complex model structures and slow training speeds. Summary of the invention

[0004] In view of this, the purpose of the present application is to propose a power load forecasting method, device, electronic device and storage medium to solve or partially solve the above-mentioned problems.

[0005] Based on the above objectives, in a first aspect, the present application provides a method for predicting power load, comprising:

[0006] Acquire first target data of the power system, wherein the first target data includes historical load data and meteorological data;

[0007] Performing data preprocessing on the first target data to obtain second target data, wherein the preprocessing includes data cleaning and / or digital conversion;

[0008] Based on the target load prediction model, the power load data of the power system is predicted according to the second target data, and the target load prediction model is constructed based on the LightGBM algorithm.

[0009] In a second aspect of the present application, a power load prediction device is provided, comprising:

[0010] An acquisition module is configured to acquire first target data of the power system, wherein the first target data includes historical load data and meteorological data;

[0011] a preprocessing module, configured to perform data preprocessing on the first target data to obtain second target data;

[0012] The prediction module is configured to predict the power load data of the power system according to the second target data based on a target load prediction model, and the target load prediction model is constructed based on the LightGBM algorithm.

[0013] In a third aspect of the present application, an electronic 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 method described in the first aspect is implemented.

[0014] According to a fourth aspect of the present application, a non-transitory computer-readable storage medium is provided, wherein the non-transitory computer-readable storage medium stores computer instructions, wherein the computer instructions are used to enable a computer to execute the method described in the first aspect.

[0015] From the above, it can be seen that the present application provides a method, device, electronic device and storage medium for power load forecasting. The method includes: obtaining first target data of the power system, the first target data including historical load data and meteorological data, performing data preprocessing on the first target data to obtain second target data, and based on a target load forecasting model, predicting the power load data of the power system according to the second target data, the target load forecasting model is constructed based on the LightGBM algorithm. The target load forecasting model with higher prediction accuracy provided by the present application, combined with the preprocessed first target data, can improve the accuracy of the predicted power load data. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present application or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 A flow chart of an exemplary power load forecasting method according to an embodiment of the present application is shown.

[0018] Figure 2 A schematic flow chart of an exemplary power load forecasting method 200 according to an embodiment of the present application is shown.

[0019] Figure 3 A schematic diagram of an exemplary power load prediction device according to an embodiment of the present application is shown.

[0020] Figure 4 A schematic diagram of an exemplary electronic device according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0021] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0022] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present application should be the usual meanings understood by people with ordinary skills in the field to which the present application belongs. The "first", "second" and similar words used in the embodiments of the present application do not represent any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing in front of the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0023] With the rapid development of social economy, the demand for electricity in various fields is increasing rapidly, and the stability requirements of power systems are increasing day by day. The daily maximum load forecast of power systems is an important process in power industry planning and an important part of the power grid management system. It uses historical power load data and other related influencing factors (such as weather conditions, human activities, time and seasonal characteristics, etc.) to comprehensively forecast and estimate the power load, power demand, and power consumption in future periods. Accurate load forecasting can help power system operators to conduct timely power dispatching and resource allocation to meet users' power demand, while reducing energy waste and improving economic benefits, which is of great significance to the optimal dispatching and management of power grids.

[0024] Existing power system load forecasting schemes mainly include the following methods: (1) Traditional forecasting methods, including time series analysis, trend analysis, and regression analysis, which can take into account periodic factors; (2) Machine learning forecasting methods, including support vector machines, decision trees, and random forests, which can solve nonlinear problems; (3) Deep learning forecasting methods, including convolutional neural networks and long short-term memory networks, which can handle high-dimensional data and sequence dependency problems.

[0025] Different technical solutions have their specific scenarios and limitations. Traditional prediction methods rely too much on historical data and give less consideration to uncertain factors; machine learning prediction methods are easily affected by irrelevant attributes and data imbalance; deep learning prediction methods have complex model structures and slow training speeds.

[0026] In order to at least solve the above-mentioned problems, the present application provides a method, device, electronic device and storage medium for power load forecasting. The method includes: obtaining first target data of the power system, the first target data including historical load data and meteorological data, performing data preprocessing on the first target data to obtain second target data, and predicting the power load data of the power system based on the target load forecasting model according to the second target data, the target load forecasting model being constructed based on the LightGBM algorithm. The target load forecasting model with higher prediction accuracy provided by the present application, combined with the preprocessed first target data, can improve the accuracy of the predicted power load data.

[0027] Figure 1 A flow chart of an exemplary power load forecasting method according to an embodiment of the present application is shown.

[0028] like Figure 1 As shown, the power load forecasting method provided by the present application can first collect data, for example, historical load data and meteorological related data, and then perform data preprocessing, smooth the historical load, encode the meteorological characteristics, etc., and then construct features, screen variables, use the LightGBM algorithm to train the model and perform model evaluation, for example, use MAPE (mean absolute percentage error) to evaluate the model, and finally use trend analysis to adjust the model results, for example, use trend analysis to calculate the cyclical growth coefficient, and adjust the model prediction results as the final predicted daily maximum load.

[0029] A combination of mathematical models and machine learning algorithms is used to predict the daily maximum load, including historical power load, weather (maximum temperature, minimum temperature, weather type, humidity, air pressure, precipitation, etc.), time (year, month, day, whether it is a special holiday, whether it is the day before a holiday, etc.) and season (periodicity coefficient, etc.) characteristics, and the data is processed and analyzed in four dimensions.

[0030] (1) Data Collection

[0031] When collecting data, in some embodiments, historical load data and meteorological data may be collected. The historical load data may include date, province, and daily maximum load. The meteorological data may include meteorological related data collected from a meteorological website, such as meteorological type, maximum temperature, minimum temperature, wind direction, wind speed, air pressure, humidity, visibility, and precipitation.

[0032] (2) Data preprocessing

[0033] In the prediction of daily maximum load of power system, data preprocessing is a crucial step, which directly affects the accuracy and efficiency of the model. The following is the detailed process of data preprocessing:

[0034] In some embodiments, the collected data may be cleaned, and missing values ​​and abnormal values ​​in the power load data may be deleted, and only the first piece of data may be retained for duplicate values.

[0035] In some embodiments, the wind direction feature can be digitally converted, which usually involves converting the azimuth of the wind direction into a numerical representation so that the machine learning model can process it. The wind direction is usually expressed as an angle ranging from 0° to 360°, where 0° represents due north, 45° represents northeast, 90° represents due east, 135° represents southeast, 180° represents due south, 225° represents southwest, and 270° represents due west.

[0036] In some embodiments, the weather type can be digitally converted to roughly divide the weather conditions into four situations: sunny, cloudy, overcast, and rainy, which correspond to four mapping values ​​of 1, 2, 3, and 4. When the weather conditions change during the day, the weather condition with the larger mapping value shall prevail. For example, if the weather on a certain day is "sunny turning to thunderstorm", the mapping value of the weather "sunny" is 1, and the mapping value of the weather "rainy" is 4, so it is considered that the mapping value of the weather on that day is 4.

[0037] (3) Construct features and filter variables

[0038] 1. Feature Engineering:

[0039] For meteorological characteristics, the temperature difference characteristics and average temperature characteristics are calculated based on the maximum temperature and the minimum temperature. The maximum temperature characteristics of the previous day are calculated based on the date and the maximum temperature. Time series characteristics are introduced, such as year, month, day, whether it is a holiday, whether it is the day before a holiday, the number of weeks in the year, the number of days in the week, etc., to improve the accuracy of the prediction.

[0040] 2. Feature selection:

[0041] The main purpose of feature selection is to find features that have a valuable impact on the results. Some of the features constructed in the above steps have a large contribution, while others have a small contribution. The features with a large contribution should be selected. For the meteorological dimension, the Pearson (Person) correlation coefficient is used to select the variables that have the highest impact on the results.

[0042] The Person correlation coefficient is used to detect the degree of linear correlation between two continuous variables. The value range is [-1,1]. Positive values ​​indicate positive correlation, and negative values ​​indicate negative correlation. The larger the absolute value, the higher the degree of linear correlation.

[0043] The Pearson correlation coefficient (ρ) between two variables x,y ) is defined as the quotient of the covariance and standard deviation between two variables (variable x and variable y):

[0044]

[0045] Among them, cov(x,y) represents the covariance between variables x and y, σ x represents the standard deviation of variable x, σ y represents the standard deviation of variable y, μ x represents the expected value of variable x, μ y represents the expected value of the variable y, and E represents the expected value.

[0046] The above formula defines the overall correlation coefficient, commonly used as ρ y,y As a representative symbol. Estimating the sample covariance and standard deviation can obtain the sample Pearson correlation coefficient, expressed as r x,y express:

[0047]

[0048] Among them, x i represents the i-th sample value of variable x, y i represents the i-th sample value of variable y, represents the sample mean of variable x, represents the sample mean of variable y, and n represents the total number of samples.

[0049] Let and represent the zero mean of variables x and y. In fact, the Pearson correlation coefficient of variables x and y is the cosine similarity of and:

[0050]

[0051] Among them, cos(x′,y′) represents the cosine of the angle between two normalized vectors x′ and y′. In geometry, the cosine represents the similarity of the directions of two vectors, x′·y′ represents the dot product of vectors x′ and y′, which is used to measure the similarity of the directions of two vectors, and |x′|·|y′| represents the product of the modulus lengths of vectors x′ and y′, regardless of the direction.

[0052] Usually, the correlation of variables is judged by the following value range:

[0053]

[0054] According to the results of feature selection, among the features of meteorological and time dimensions, the features related to load data are: maximum temperature, temperature difference, humidity, precipitation, whether it is a holiday, whether it is the day before a holiday, the number of weeks in the year, and the number of days in the week.

[0055] (4) Model evaluation based on LightGBM training model

[0056] 1. Divide the training set and test set

[0057] Divide the data set into a training set and a test set with a ratio of 7:3. The training set is used to train the model, and the test set is used to test the model.

[0058] 2. Use LightGBM to train the model

[0059] LightGBM (Light Gradient Boosting Machine) is an algorithm based on gradient boosting decision trees. It gradually reduces the error by building multiple weak learners. Its goal is to minimize a loss function Among them, y is the actual value, is the predicted value of the model. The LightGBM algorithm uses the following loss function:

[0060]

[0061] Among them, i represents the i-th sample, N represents the total number of samples, and y i represents the actual value of the i-th sample, represents the predicted value of the i-th sample, represents the error between the actual value and the predicted value, λ represents the regularization coefficient, and Ω(f) represents the regularization term, which is used to prevent overfitting.

[0062] Histogram algorithm: In some embodiments, in order to speed up the calculation of segmentation points, the LightGBM algorithm divides each eigenvalue into K discrete intervals to form a histogram. This discretization process is expressed as:

[0063]

[0064] Among them, x k represents the discretized eigenvalue, floor represents the mathematical function, x represents the eigenvalue of the second target data, min(x) represents the minimum eigenvalue, max(x) represents the maximum eigenvalue, and K represents the number of intervals in the histogram.

[0065] The boundary of each histogram interval can be used to calculate the information gain of different segmentation points, effectively reducing the computational overhead.

[0066] Leaf-node-based growth strategy: In some embodiments, the LightGBM algorithm grows by splitting the leaf node with the largest error. For the splitting condition of leaf node j, the gain can be calculated by the following formula:

[0067]

[0068] in, Where Gain(j) represents the gain value of the leaf node, g i represents the gradient of the i-th sample, h i represents the Hessian value, which is used to measure the segmentation effect, and λ represents the regularization coefficient. A high gain means a better segmentation point.

[0069] Gradient and loss function optimization: In some embodiments, the LightGBM algorithm uses negative gradient as an approximation of the loss function. Assume that the prediction for round m is Then the negative gradient update formula for the m+1th round is:

[0070]

[0071] Among them, g i represents the negative gradient of the i-th sample in round m+1, Represents the predicted value of the i-th sample With the actual value y i The error between Represents the rate at which the loss function changes with the predicted value.

[0072] Regularization: In some embodiments, the model complexity can be controlled by adding L1 and L2 regularization terms to the loss function, such as:

[0073]

[0074] Among them, α and β are regularization parameters, and by adjusting these parameters, the model can be prevented from overfitting.

[0075] Use the training set data and cross-validation method to train the LightGBM model.

[0076] 3. Use MAPE (mean absolute percentage error) for model evaluation

[0077] MAPE (Mean Absolute Percentage Error) is an indicator to measure the accuracy of forecasting. It calculates the absolute value of the difference between the forecast value and the actual value and converts it into a percentage to express the size of the average error. MAPE can provide an intuitive error rate metric to help evaluate the accuracy of the daily maximum load forecasting model.

[0078] The calculation formula of MAPE is as follows:

[0079]

[0080] Among them, A t represents the actual value at the tth time point, F trepresents the predicted value at the tth time point, and n is the total number of observations.

[0081] After the model training is completed, the test data is input into the model, and prediction is made through the model. The MAPE is calculated using the predicted value and the actual value. The lower the MAPE value, the higher the prediction accuracy.

[0082] (5) Trend analysis method to adjust model results

[0083] Trend analysis, also known as trend curve analysis, is a quantitative forecasting method widely used in power load forecasting. This method analyzes historical load data, analyzes the trend of historical load changes in the same period and the previous period at the observation point, calculates the recent load change rate as the periodic growth coefficient, and multiplies the growth coefficient with the model prediction result to obtain the final predicted daily maximum load.

[0084] Evaluation results after using the method provided in this application:

[0085]

[0086]

[0087] Figure 2 FIG. 2 is a flow chart of an exemplary power load forecasting method 200 according to an embodiment of the present application. Figure 2 As shown, method 200 may include the following steps.

[0088] In step 202, first target data of the power system is acquired, where the first target data includes historical load data and meteorological data.

[0089] In step 204, data preprocessing is performed on the first target data to obtain second target data, wherein the preprocessing includes data cleaning and / or digital conversion.

[0090] In some embodiments, the meteorological data includes wind direction characteristics and meteorological types, and the data preprocessing of the first target data to obtain the second target data further includes: converting the wind direction characteristics into corresponding angles, and converting the meteorological type into corresponding mapping values ​​to obtain the second target data.

[0091] In step 206, based on a target load prediction model, according to the second target data, the power load data of the power system is predicted, and the target load prediction model is constructed based on the LightGBM algorithm.

[0092] In some embodiments, the target load prediction model is trained by the following method: acquiring target training data; constructing feature data based on the target training data; and training a target model constructed based on the LightGBM algorithm based on the feature data to obtain the target load prediction model.

[0093] In some embodiments, the step of predicting the power load data of the power system based on the target load prediction model and the second target data further includes: discretizing the second target data based on a histogram algorithm of the target load prediction model to predict the power load data; wherein the discretization process is expressed as: Among them, x k represents the discretized eigenvalue, floor represents the mathematical function, x represents the eigenvalue of the second target data, min(x) represents the minimum eigenvalue, max(x) represents the maximum eigenvalue, and K represents the number of intervals in the histogram.

[0094] In some embodiments, the loss function of the target load prediction model is expressed as: Among them, i represents the i-th sample, N represents the total number of samples, and y i represents the actual value of the i-th sample, represents the predicted value of the i-th sample, represents the error between the actual value and the predicted value, λ represents the regularization coefficient, and Ω(f) represents the regularization term.

[0095] In some embodiments, the gain of the target load prediction model is expressed as: Where Gain(j) represents the gain value of the leaf node, g i represents the gradient of the i-th sample, h i represents the Hessian value, and λ represents the regularization coefficient.

[0096] In some embodiments, the gradient of the target load prediction model is expressed as: Among them, g i represents the negative gradient of the i-th sample in round m+1, Represents the predicted value of the i-th sample With the actual value y i The error between Represents the rate at which the loss function changes with the predicted value.

[0097] The present application provides a method, device, electronic device and storage medium for predicting electric load. The method includes: obtaining first target data of the electric power system, the first target data including historical load data and meteorological data, performing data preprocessing on the first target data to obtain second target data, and predicting the electric load data of the electric power system based on the target load prediction model according to the second target data, the target load prediction model being constructed based on the LightGBM algorithm. The target load prediction model with higher prediction accuracy provided by the present application, combined with the preprocessed first target data, can improve the accuracy of the predicted electric load data.

[0098] It should be noted that the method of the embodiment of the present application can be performed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present application, and the multiple devices will interact with each other to complete the described method.

[0099] It should be noted that the above describes some embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0100] Based on the same technical concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a power load prediction device.

[0101] refer to Figure 3 , the power load forecasting device comprises:

[0102] The acquisition module 301 is configured to acquire first target data of the power system, where the first target data includes historical load data and meteorological data.

[0103] The preprocessing module 302 is configured to perform data preprocessing on the first target data to obtain second target data, wherein the preprocessing includes data cleaning and / or digital conversion.

[0104] The meteorological data includes wind direction characteristics and meteorological types. The preprocessing module 302 is further configured to convert the wind direction characteristics into corresponding angles and convert the meteorological types into corresponding mapping values ​​to obtain the second target data.

[0105] The prediction module 303 is configured to predict the power load data of the power system according to the second target data based on a target load prediction model, and the target load prediction model is constructed based on the LightGBM algorithm.

[0106] The device also includes a training module, which is configured to obtain target training data; construct feature data based on the target training data; and train a target model constructed based on the LightGBM algorithm based on the feature data to obtain the target load prediction model.

[0107] The prediction module 303 is further configured to discretize the second target data based on the histogram algorithm of the target load prediction model to predict the power load data; wherein the discretization process is expressed as: Among them, x k represents the discretized eigenvalue, floor represents the mathematical function, x represents the eigenvalue of the second target data, min(x) represents the minimum eigenvalue, max(x) represents the maximum eigenvalue, and K represents the number of intervals in the histogram.

[0108] Among them, the loss function of the target load prediction model is expressed as: Among them, i represents the i-th sample, N represents the total number of samples, and y i represents the actual value of the i-th sample, represents the predicted value of the i-th sample, represents the error between the actual value and the predicted value, λ represents the regularization coefficient, and Ω(f) represents the regularization term.

[0109] Wherein, the gain of the target load prediction model is expressed as: Where Gain(j) represents the gain value of the leaf node, g i represents the gradient of the i-th sample, h i represents the Hessian value, and λ represents the regularization coefficient.

[0110] Among them, the gradient of the target load prediction model is expressed as: Among them, g i represents the negative gradient of the i-th sample in round m+1, Represents the predicted value of the i-th sample With the actual value y i The error between Represents the rate at which the loss function changes with the predicted value.

[0111] For the convenience of description, the above device is described in terms of functions divided into various modules. Of course, when implementing the present application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0112] The device of the above embodiment is used to implement the corresponding method 200 in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0113] Based on the same technical concept, corresponding to any of the above-mentioned embodiments, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method 200 described in any of the above embodiments when executing the program.

[0114] Figure 4 A schematic diagram of an exemplary electronic device according to an embodiment of the present application is shown, and the device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are connected to each other in communication within the device through the bus 1050.

[0115] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0116] The memory 1020 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 may store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 1020 and are called and executed by the processor 1010.

[0117] The input / output interface 1030 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0118] The communication interface 1040 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0119] The bus 1050 includes a path that transmits information between the various components of the device (eg, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).

[0120] It should be noted that, although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040 and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present specification, and does not necessarily include all the components shown in the figure.

[0121] The electronic device of the above embodiment is used to implement the corresponding method 200 in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0122] Based on the same technical concept, corresponding to any of the above-mentioned embodiments, the present application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute method 200 described in any of the above embodiments.

[0123] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be achieved by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0124] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the method 200 described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0125] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present application (including the claims) is limited to these examples. In line with the concept of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.

[0126] In addition, to simplify the description and discussion, and in order not to make the embodiments of the present application difficult to understand, the known power supply / ground connection with the integrated circuit (IC) chip and other components may or may not be shown in the provided drawings. In addition, the device can be shown in the form of a block diagram to avoid making the embodiments of the present application difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform to be implemented in the embodiments of the present application (that is, these details should be fully within the scope of understanding of those skilled in the art). In the case of elaborating specific details (e.g., circuits) to describe exemplary embodiments of the present application, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details or when these specific details are changed. Therefore, these descriptions should be considered to be illustrative rather than restrictive.

[0127] Although the present application has been described in conjunction with specific embodiments of the present application, many replacements, modifications and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0128] The embodiments of the present application are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the scope of protection of the present application.

Claims

1. A method for predicting power load, comprising: Acquire first target data of the power system, wherein the first target data includes historical load data and meteorological data; Performing data preprocessing on the first target data to obtain second target data, wherein the preprocessing includes data cleaning and / or digital conversion; Based on the target load prediction model, the power load data of the power system is predicted according to the second target data, and the target load prediction model is constructed based on the LightGBM algorithm.

2. The method of claim 1, wherein: The meteorological data includes wind direction characteristics and meteorological types, and the preprocessing of the first target data to obtain the second target data further includes: The wind direction feature is converted into a corresponding angle, and the meteorological type is converted into a corresponding mapping value to obtain the second target data.

3. The method of claim 1, wherein: The target load forecasting model is trained by the following method: Obtain target training data; Constructing feature data according to the target training data; Based on the feature data, a target model constructed based on the LightGBM algorithm is trained to obtain the target load prediction model.

4. The method of claim 1, wherein: The predicting the power load data of the power system based on the target load prediction model and according to the second target data further comprises: Discretizing the second target data using a histogram algorithm based on a target load prediction model to predict the power load data; The discretization process is expressed as: Among them, x k represents the discretized eigenvalue, floor represents the mathematical function, x represents the eigenvalue of the second target data, min(x) represents the minimum eigenvalue, max(x) represents the maximum eigenvalue, and K represents the number of intervals in the histogram.

5. The method of claim 1, wherein: The loss function of the target load prediction model is expressed as: Among them, i represents the i-th sample, N represents the total number of samples, and y i represents the actual value of the i-th sample, represents the predicted value of the i-th sample, represents the error between the actual value and the predicted value, λ represents the regularization coefficient, and Ω(f) represents the regularization term.

6. The method of claim 1, wherein: The gain of the target load prediction model is expressed as: Where Gain(j) represents the gain value of the leaf node, g i represents the gradient of the i-th sample, h i represents the Hessian value, and λ represents the regularization coefficient.

7. The method of claim 1, wherein: The gradient of the target load prediction model is expressed as: Among them, g i represents the negative gradient of the i-th sample in round m+1, Represents the predicted value of the i-th sample With the actual value y i The error between Represents the rate at which the loss function changes with the predicted value.

8. A power load forecasting device, comprising: An acquisition module is configured to acquire first target data of the power system, wherein the first target data includes historical load data and meteorological data; a preprocessing module, configured to perform data preprocessing on the first target data to obtain second target data; The prediction module is configured to predict the power load data of the power system according to the second target data based on a target load prediction model, and the target load prediction model is constructed based on the LightGBM algorithm.

9. An electronic device 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 method according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 7.

Images