Power data cleaning method, device, equipment, medium and program product

Abstract

The invention relates to an electric power data cleaning method and device, equipment, a medium and a program product. The method comprises the steps that multiple pieces of electric power data of an electric power system are acquired, and the multiple pieces of electric power data comprise at least one of voltage, temperature, current, power and electricity consumption; respectively determining information gain values of the plurality of electric power data for electric power load prediction; screening target data from the multiple pieces of power data according to the multiple pieces of information gain values; and performing data cleaning optimization on the target data according to a data cleaning strategy to obtain optimized data. By adopting the method, the accuracy of power data cleaning can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of big data, in particular to a power data cleaning method and device, computer equipment, computer readable storage medium and computer program product. BACKGROUND

[0002] With the transformation of the power system to digitalization and intelligentization, power load forecasting becomes the core support for power grid dispatching and resource allocation, and high-quality power data is a key prerequisite for improving forecasting accuracy.

[0003] At present, multiple power data are directly cleaned, and the sources of multiple power data are relatively rich. Directly cleaning multiple power data may lead to some power data that is still not high-quality power data for power load forecasting even after data cleaning, resulting in low accuracy of data cleaning. SUMMARY

[0004] Therefore, it is necessary to provide a power data cleaning method, device, computer equipment, computer readable storage medium and computer program product capable of reducing the operation complexity of power data cleaning to solve the above technical problems.

[0005] In a first aspect, the present application provides a power data cleaning method, which comprises:

[0006] obtaining multiple power data of a power system, wherein the multiple power data comprise at least one of voltage, temperature, current, power and power consumption;

[0007] determining information gain values of the multiple power data for power load forecasting, respectively;

[0008] selecting target data from the multiple power data according to the multiple information gain values;

[0009] performing data cleaning optimization on the target data according to a data cleaning strategy to obtain optimized data.

[0010] As one of the embodiments, the determining of the information gain values of the multiple power data for power load forecasting, respectively, comprises:

[0011] determining confusion degree information of each power data, respectively;

[0012] extracting feature information of each confusion degree information;

[0013] for each power data, determining an information gain value of the power data for power load forecasting according to the difference between the confusion degree information and the feature information of the power data.

[0014] As one of the embodiments, the filtering target data from the plurality of power data according to the plurality of information gain values comprises:

[0015] Obtaining a preset information gain threshold corresponding to the plurality of information gain values;

[0016] Filtering target data with an information gain value greater than the preset information gain threshold from the plurality of power data.

[0017] As one of the embodiments, the data cleaning optimization of the target data according to the data cleaning strategy comprises:

[0018] According to the noise removal strategy in the data cleaning strategy, the target data is removed to obtain denoising data;

[0019] According to the abnormal identification strategy in the data cleaning strategy, the denoising data is identified to obtain abnormal data and normal data;

[0020] According to the data repair strategy in the data cleaning strategy, the abnormal data is optimized to obtain repair data;

[0021] According to the format conversion strategy in the data cleaning strategy, the normal data and the repair data are format converted to obtain optimized data.

[0022] As one of the embodiments, after the data cleaning optimization of the target data according to the data cleaning strategy to obtain the optimized data, the method further comprises:

[0023] Determining the data quality index value of the optimized data;

[0024] According to the data quality index value, the data cleaning strategy is optimized.

[0025] As one of the embodiments, the determination of the data quality index value of the optimized data comprises at least one of the following:

[0026] Determining the data integrity of the optimized data, and determining the data integrity as the data quality index value of the optimized data;

[0027] Determining the measured data corresponding to the optimized data, and determining the difference between the optimized data and the measured data as the data quality index value of the optimized data;

[0028] Determining the data consistency index value of the optimized data, and determining the data consistency index value as the data quality index value of the optimized data.

[0029] In a second aspect, the present application further provides an electric power data cleaning device, comprising:

[0030] a data acquisition module configured to acquire a plurality of electric power data of an electric power system, wherein the plurality of electric power data comprises at least one of voltage, temperature, current, power and power consumption;

[0031] a gain determination module configured to determine information gain values of the plurality of electric power data for electric power load forecasting respectively;

[0032] a data screening module configured to screen target data from the plurality of electric power data according to the information gain values;

[0033] a data cleaning module configured to perform data cleaning optimization on the target data according to a data cleaning strategy to obtain optimized data.

[0034] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the processor implements the following steps when executing a computer program:

[0035] acquiring a plurality of electric power data of an electric power system, wherein the plurality of electric power data comprises at least one of voltage, temperature, current, power and power consumption;

[0036] determining information gain values of the plurality of electric power data for electric power load forecasting respectively;

[0037] screening target data from the plurality of electric power data according to the information gain values;

[0038] performing data cleaning optimization on the target data according to a data cleaning strategy to obtain optimized data.

[0039] In a fourth aspect, the present application further provides a computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the following steps:

[0040] acquiring a plurality of electric power data of an electric power system, wherein the plurality of electric power data comprises at least one of voltage, temperature, current, power and power consumption;

[0041] determining information gain values of the plurality of electric power data for electric power load forecasting respectively;

[0042] screening target data from the plurality of electric power data according to the information gain values;

[0043] performing data cleaning optimization on the target data according to a data cleaning strategy to obtain optimized data.

[0044] In a fifth aspect, the present application also provides a computer program product comprising a computer program which, when executed by a processor, implements the following steps:

[0045] obtaining a plurality of power data of a power system, wherein the plurality of power data comprises at least one of voltage, temperature, current, power and power consumption;

[0046] respectively determining an information gain value of the plurality of power data for power load forecasting;

[0047] screening target data from the plurality of power data according to a plurality of the information gain values;

[0048] performing data cleaning optimization on the target data according to a data cleaning strategy to obtain optimized data.

[0049] The power data cleaning method, device, computer equipment, computer readable storage medium and computer program product described above, by obtaining a plurality of power data of a power system, wherein the plurality of power data comprises at least one of voltage, temperature, current, power and power consumption; respectively determining an information gain value of the plurality of power data for power load forecasting; screening target data from the plurality of power data according to a plurality of the information gain values; performing data cleaning optimization on the target data according to a data cleaning strategy to obtain optimized data.

[0050] In this way, when obtaining a plurality of power data, the target data is screened from the plurality of power data according to the information gain value of the plurality of power data for power load forecasting, and then the target data is cleaned, so that the object of data cleaning, i.e. the target data, is the data helpful for power load forecasting, and the accuracy of the target data is further ensured through the data cleaning process, thus improving the accuracy of power data cleaning. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed to be used in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0052] Figure 1 An application environment diagram of the power data cleaning method in an embodiment;

[0053] Figure 2 A flowchart of the power data cleaning method in an embodiment;

[0054] Figure 3This is a flowchart illustrating the steps of cleaning and optimizing the target data according to a data cleaning strategy in one embodiment to obtain optimized data.

[0055] Figure 4 This is a structural block diagram of a power data cleaning device in one embodiment;

[0056] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0058] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0059] The power data cleaning method provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or located in the cloud or on other network servers. Server 104 acquires multiple power data from the power system, including at least one of voltage, temperature, current, power, and electricity consumption; determines the information gain value of each of the multiple power data for power load prediction; filters target data from the multiple power data based on the multiple information gain values; and performs data cleaning and optimization on the target data according to a data cleaning strategy to obtain optimized data. Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. Terminal 102 can be a device with the target application deployed on it.

[0060] In one exemplary embodiment, such as Figure 2 As shown, a power data cleaning method is provided, which can be applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 202 to 208. Wherein:

[0061] In step 202, a plurality of power data of a power system is acquired, wherein the plurality of power data comprises at least one of voltage, temperature, current, power and power consumption.

[0062] In step 202, the power data comprises at least one of voltage, temperature, current, power and power consumption collected from various data sources, wherein the data sources comprise but are not limited to substations, power transmission lines, data acquisition components deployed on user terminal nodes, and the data acquisition components comprise but are not limited to smart meters, current sensors, voltage sensors, weather sensors and power monitors; and the power data comprises at least one of inventory data (i.e. historical data) and real-time data.

[0063] In step 204, information gain values of the plurality of power data for power load forecasting are determined respectively.

[0064] In step 204, the information gain values of the plurality of power data for power load forecasting are determined respectively, which comprises: determining a degree of disorder information of each of the power data; extracting feature information of each of the degree of disorder information; and determining, for each of the power data, an information gain value of the power data for power load forecasting according to a difference between the degree of disorder information and the feature information of the power data.

[0065] In step 204, the feature information comprises a weighted average of the degree of disorder information.

[0066] Optionally, step 204 can be implemented by a random forest algorithm or a gradient boosting tree algorithm.

[0067] In this way, the influence / contribution of the power data for power load forecasting is quantified by the information gain value, avoiding the subjectivity of “empirical screening”, and ensuring that the target data screened can improve the accuracy of power load forecasting.

[0068] In step 206, target data is screened from the plurality of power data according to the plurality of information gain values.

[0069] In step 206, the target data with an information gain value greater than a preset information gain threshold value is screened from the plurality of power data.

[0070] The preset information gain threshold value can be set by a user as needed, and can also be an empirical value, which is not limited herein.

[0071] In step 208, the target data is optimized by data cleaning according to a data cleaning strategy, to obtain optimized data.

[0072] In step 208, the data cleaning strategy comprises at least one of a noise removal strategy, an abnormality identification strategy, a data repair strategy and a format conversion strategy.

[0073] Optionally, step 208 is followed by the method further comprising determining the optimized data as input data of the power prediction model.

[0074] Optionally, step 208 is followed by the method further comprising determining the optimized data as input data of the power prediction model.

[0075] Optionally, step 208 is followed by the method further comprising determining a data quality indicator value of the optimized data, and optimizing the data cleaning strategy according to the data quality indicator value.

[0076] Further, determining the data quality indicator value of the optimized data comprises at least one of the following: determining a data completeness of the optimized data, and determining the data completeness as the data quality indicator value of the optimized data; determining a difference between the optimized data and corresponding measured data, and determining the difference as the data quality indicator value of the optimized data; determining a data consistency indicator value of the optimized data, and determining the data consistency indicator value as the data quality indicator value of the optimized data; determining a model indicator value corresponding to the optimized data, and determining the model indicator value as the data quality indicator value of the optimized data.

[0077] As an embodiment, optimizing the data cleaning strategy according to the data quality indicator value comprises at least one of the following: optimizing the data cleaning strategy in a case where a data missing rate represented by the data quality indicator value is greater than a preset missing rate threshold; optimizing the data cleaning strategy in a case where a difference represented by the data quality indicator value is greater than a preset difference threshold; optimizing the data cleaning strategy in a case where a data format unification rate represented by the data quality indicator value is less than a preset format unification rate threshold; optimizing the data cleaning strategy in a case where a sampling frequency unification rate represented by the data quality indicator value is less than a preset sampling frequency unification rate threshold; optimizing the data cleaning strategy in a case where a model prediction accuracy represented by the data quality indicator value is lower than a preset accuracy threshold.

[0078] The preset missing rate threshold, the preset difference threshold, the preset format unification rate threshold, the preset sampling frequency unification rate threshold, and the preset accuracy threshold can be set by a user as needed, or can be empirical values, which are not limited herein.

[0079] In this way, a closed-loop logic of intelligent cleaning optimization and dynamic iteration is realized, and the accuracy of data cleaning is ensured.

[0080] Optionally, the optimization of the data cleaning strategy comprises at least one of the following: adjusting a wavelet basis function included in a noise removal strategy in the data cleaning strategy; adjusting a field radius in a DBSCAN algorithm included in an anomaly identification strategy in the data cleaning strategy; adjusting a minimum point quantity in the DBSCAN algorithm included in the anomaly identification strategy in the data cleaning strategy; adjusting a K value in a K-Nearest Neighbor algorithm included in a data repair strategy in the data cleaning strategy; adjusting a weight in a weighted average method of the K-Nearest Neighbor algorithm included in the data repair strategy in the data cleaning strategy.

[0081] In this way, an optimization manner of the data cleaning strategy is provided, and the optimization of the data cleaning strategy can be achieved from the above-mentioned multiple dimensions, so as to seek an optimal data cleaning strategy.

[0082] Optionally, after step 208, the method further comprises: determining a data quality indicator value of the optimized data; and optimizing a collection frequency of the power data according to the data quality indicator value.

[0083] Further, the determination of the data quality indicator value of the optimized data comprises at least one of the following: determining a data completeness of the optimized data, and determining the data completeness as the data quality indicator value of the optimized data; determining a measured data corresponding to the optimized data, and determining a data quality indicator value of the optimized data according to a difference between the optimized data and the measured data; determining a data consistency indicator value of the optimized data, and determining the data consistency indicator value as the data quality indicator value of the optimized data; and determining a model indicator value corresponding to the optimized data, and determining the model indicator value as the data quality indicator value of the optimized data.

[0084] As an embodiment, the optimization of the collection frequency of the power data according to the data quality indicator value comprises at least one of the following: in a case where a data missing rate represented by the data quality indicator value is greater than a preset missing rate threshold, optimizing the collection frequency of the power data; in a case where a difference represented by the data quality indicator value is greater than a preset difference threshold, optimizing the collection frequency of the power data; in a case where a data format unification rate represented by the data quality indicator value is less than a preset format unification rate threshold, optimizing the collection frequency of the power data; in a case where a sampling frequency unification rate represented by the data quality indicator value is less than a preset sampling frequency unification rate threshold, optimizing the collection frequency of the power data; and in a case where a model prediction accuracy represented by the data quality indicator value is lower than a preset accuracy threshold, optimizing the collection frequency of the power data.

[0085] Optionally, the optimization of the collection frequency of the power data comprises at least one of the following: adjusting the collection frequency of the power data to be lower; and adjusting the collection frequency of the power data to be higher.

[0086] The increasing the collection frequency of the power data can include increasing the collection frequency of the power data to 1.2 to 1.5 times of the collection frequency before the adjustment.

[0087] In this way, the collection frequency of the power data is adjusted, and iterative optimization of the power data cleaning is achieved from the dimension of the collection frequency of the power data.

[0088] In the power data cleaning method, a plurality of power data of a power system is obtained, information gain values of the plurality of power data for power load prediction are determined respectively, target data is selected from the plurality of power data according to the information gain values, and the target data is cleaned and optimized according to a data cleaning strategy to obtain optimized data. When the plurality of power data is obtained, the target data is selected from the plurality of power data according to the information gain values of the plurality of power data for power load prediction, and the target data is cleaned. The object of data cleaning, that is, the target data, is data that is helpful for power load prediction. The accuracy of the target data is further ensured through the data cleaning process, and thus the accuracy of the power data cleaning is improved.

[0089] In one exemplary embodiment, as shown in FIG. 3, step 208 includes steps 302 to 308. Wherein: Figure 3 Step 302: removing noise from the target data according to a noise removal strategy in the data cleaning strategy to obtain denoised data.

[0090] Exemplarily, step 302 includes: removing noise from the target data by using a wavelet analysis algorithm included in the noise removal strategy in the data cleaning strategy to obtain denoised data.

[0091] Further, removing noise from the target data by using the wavelet analysis algorithm to obtain denoised data includes: decomposing the target data into high-frequency subband data and low-frequency subband data by using a wavelet basis function in the wavelet analysis algorithm, reconstructing the high-frequency subband data by using a soft threshold to obtain reconstructed data, and determining the reconstructed data and the low-frequency subband data as the denoised data.

[0092]

[0093] ​As an embodiment, the high-frequency sub-band data divides the frequency band frequency from low to high to include first sub-band data, second sub-band data, third sub-band data, fourth sub-band data and fifth sub-band data; and the soft threshold reconstruction of the high-frequency sub-band data to obtain the reconstructed data includes: for each sub-band data in the first sub-band data, the second sub-band data and the third sub-band data, determining the sub-band coefficient corresponding to the sub-band data, determining the sub-band threshold based on the sub-band coefficient, in the case that the sub-band data is less than the sub-band threshold, modifying the sub-band data to 0, in the case that the sub-band data is not less than the sub-band threshold, modifying the sub-band data to the difference between the sub-band data and the sub-band threshold, and determining the modified sub-band data as the reconstructed data.

[0094] Step 304, according to the abnormality identification strategy in the data cleaning strategy, identifying the denoising data to obtain abnormal data and normal data.

[0095] Exemplarily, step 304 includes: identifying the denoising data by the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm included in the abnormality identification strategy in the data cleaning strategy to obtain abnormal data and normal data.

[0096] Among them, the field radius in the DBSCAN algorithm is set to 0.8-1.2, the number of minimum points is set to 5-8, and the denoising data that does not meet the density requirement is determined as abnormal data.

[0097] Step 306, according to the data repair strategy in the data cleaning strategy, optimizing the abnormal data to obtain repair data.

[0098] Exemplarily, step 306 includes: according to the K-nearest neighbor algorithm included in the data repair strategy in the data cleaning strategy, optimizing the abnormal data to obtain repair data.

[0099] Further, according to the K-nearest neighbor algorithm, the abnormal data is optimized to obtain repair data, including: by calculating the Euclidean distance between the abnormal data and other data, selecting the K nearest data, and using the weighted average method to calculate the repair data of the abnormal data.

[0100] Among them, the K value is set to 3-5, and the weight in the weighted average method is the reciprocal of the distance.

[0101] Thus, based on the characteristics of power noise high frequency and effective signal low frequency, through the frequency separation ability of wavelet analysis, the dual goals of "precise denoising + retaining core information" are achieved. This choice not only conforms to the basic principles of signal processing, but also meets the business needs of power load forecasting, and overcomes the defects of data distortion caused by excessive smoothing in traditional denoising methods, ensuring the accuracy of data cleaning.

[0102] Step 308, according to the format conversion strategy in the data cleaning strategy, the normal data and the repair data are converted to obtain the optimized data.

[0103] Exemplarily, step 308 includes: according to the format conversion strategy in the data cleaning strategy, the normal data and the repair data are converted into JSON (JavaScript Object Notation, JavaScript Object Notation) format to obtain the optimized data.

[0104] Thus, through data format unification, multi-source data collection, intelligent cleaning, dynamic optimization, and prediction support the key link of the whole process. By eliminating the structural differences of multi-source data, it not only guarantees the integrity and accuracy of the time series data set, but also provides a "non-interference" basis for model training, effect verification, and dynamic iteration.

[0105] In this embodiment, the target data is removed according to the noise removal strategy in the data cleaning strategy to obtain denoising data; the denoising data is identified according to the abnormal identification strategy in the data cleaning strategy to obtain abnormal data and normal data; the abnormal data is optimized according to the data repair strategy in the data cleaning strategy to obtain repair data; the normal data and the repair data are converted according to the format conversion strategy in the data cleaning strategy to obtain the optimized data, realizing the whole process of denoising-elimination-repair-conversion, and ensuring the accuracy of the optimized data.

[0106] It can be understood that during the power data cleaning process, various problems may occur to cause abnormal power data cleaning. If not checked, it will affect the subsequent data report generation.

[0107] As a detailed embodiment, a plurality of power data of a power system is acquired, wherein the plurality of power data comprises at least one of voltage, temperature, current, power and power consumption; the confusion degree information of each power data is determined respectively; the feature information of each confusion degree information is extracted; for each power data, the information gain value of the power data for power load prediction is determined according to the difference between the confusion degree information and the feature information of the power data; a preset information gain threshold value corresponding to a plurality of information gain values is acquired; target data with an information gain value greater than the preset information gain threshold value is screened from the plurality of power data; noise removal is performed on the target data according to a noise removal strategy in a data cleaning strategy, to obtain denoised data; the denoised data is identified according to an abnormality identification strategy in the data cleaning strategy, to obtain abnormal data and normal data; the abnormal data is optimized according to a data repair strategy in the data cleaning strategy, to obtain repaired data; the normal data and the repaired data are format-converted according to a format conversion strategy in the data cleaning strategy, to obtain optimized data; a data quality index value of the optimized data is determined; and the data cleaning strategy is optimized according to the data quality index value.

[0108] In this way, by acquiring a plurality of power data of a power system, determining information gain values of the plurality of power data for power load prediction respectively, screening target data from the plurality of power data according to a plurality of information gain values, and performing data cleaning optimization on the target data according to a data cleaning strategy to obtain optimized data, the information gain values of the plurality of power data for power load prediction are used as a basis for screening target data from the plurality of power data, and the target data is cleaned in turn, so that the object of data cleaning, i.e. the target data, is data that is helpful for power load prediction, and the accuracy of the target data is further ensured through the data cleaning process, thereby improving the accuracy of power data cleaning.

[0109] It should be understood that although the steps in the flowcharts involved in the above embodiments are shown in sequence according to the arrows, the steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, the execution of the steps is not strictly limited in sequence, and the steps can be executed in other orders. Moreover, at least some of the steps in the flowcharts involved in the above embodiments can include multiple steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of the steps or stages is not necessarily sequential, but can be alternately or alternately executed with at least part of other steps or steps or stages in other steps. It can be understood that the steps in different embodiments can be freely combined as needed, and various non-contradictory schemes formed by the combination are within the scope of protection of the present application.

[0110] Based on the same inventive concept, the embodiments of the present application also provide a power data cleaning device for implementing the above-mentioned power data cleaning method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more power data cleaning device embodiments provided below can refer to the limitations of the power data cleaning method described above, which will not be repeated here.

[0111] In one exemplary embodiment, as shown in Figure 4 A power data cleaning device is provided, comprising: a data acquisition module, a gain determination module, a data screening module, and a data cleaning module, wherein:

[0112] The data acquisition module is configured to acquire a plurality of power data of a power system, wherein the plurality of power data comprises at least one of voltage, temperature, current, power, and power consumption.

[0113] The gain determination module is configured to determine an information gain value of each of the plurality of power data for power load prediction.

[0114] The data screening module is configured to screen target data from the plurality of power data according to a plurality of information gain values.

[0115] The data cleaning module is configured to perform data cleaning optimization on the target data according to a data cleaning strategy to obtain optimized data.

[0116] In one embodiment, the gain determination module is further configured to determine a degree of confusion information of each of the power data, extract feature information of each of the degree of confusion information, and determine the information gain value of each of the power data for power load prediction according to the difference between the degree of confusion information and the feature information of the power data.

[0117] In one of the embodiments, the data screening module is further configured to: obtain a preset information gain threshold corresponding to the plurality of information gain values; and screen target data with an information gain value greater than the preset information gain threshold from the plurality of power data.

[0118] In one of the embodiments, the data cleaning module is further configured to: perform noise removal on the target data according to a noise removal strategy in the data cleaning strategy to obtain denoised data; perform identification on the denoised data according to an abnormality identification strategy in the data cleaning strategy to obtain abnormal data and normal data; perform optimization on the abnormal data according to a data repair strategy in the data cleaning strategy to obtain repaired data; and perform format conversion on the normal data and the repaired data according to a format conversion strategy in the data cleaning strategy to obtain optimized data.

[0119] In one of the embodiments, after the data cleaning optimization device obtains the optimized data by performing data cleaning optimization on the target data according to the data cleaning strategy, the data cleaning optimization device further includes an iterative optimization module configured to: determine a data quality index value of the optimized data; and optimize the data cleaning strategy according to the data quality index value.

[0120] In one of the embodiments, the iterative optimization module is further configured to perform at least one of the following: determine a data completeness of the optimized data, and determine the data completeness as the data quality index value of the optimized data; determine real measurement data corresponding to the optimized data, and determine a difference between the optimized data and the real measurement data as the data quality index value of the optimized data; and determine a data consistency index value of the optimized data, and determine the data consistency index value as the data quality index value of the optimized data.

[0121] The modules in the power data cleaning device can be all or partially implemented by software, hardware, or a combination thereof. The modules can be embedded in or independent of a processor in a computer device in a hardware form, or stored in a memory in the computer device in a software form, so as to be called and executed by a processor to perform operations corresponding to the modules.

[0122] In one of the embodiments, a computer device is provided, which can be a terminal. An internal structure diagram of the computer device can be as shown in FIG. 1. Figure 5As shown in the figure. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through the system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capability. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be realized through WIFI, mobile cellular network, near field communication (NFC) or other technologies. The computer program is executed by the processor to realize a power data cleaning method. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0123] Those skilled in the art can understand that, Figure 5 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0124] In one exemplary embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps in each of the method embodiments described above.

[0125] In one exemplary embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the steps in each of the method embodiments described above.

[0126] In one exemplary embodiment, a computer program product is provided, including a computer program, and the computer program is executed by a processor to implement the steps in each of the method embodiments described above.

[0127] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0128] It can be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments of each method. In the embodiments provided in the present application, any reference to memory, database or other medium can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0129] Any technical features in the above embodiments can be combined, and for the sake of brevity, not all possible combinations are described above, however, any combination of these technical features is deemed to be within the scope of the present application.

[0130] The above embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it should not be understood as a limitation on the patent scope of the present application. It should be pointed out that, for ordinary skilled persons in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A power data cleaning method, characterized in that, The method comprises: acquiring a plurality of power data of a power system, wherein the plurality of power data comprises at least one of voltage, temperature, current, power and power consumption; respectively determining information gain values of the plurality of power data for power load prediction; screening target data from the plurality of power data according to the plurality of information gain values; optimizing the target data according to a data cleaning strategy to obtain optimized data.

2. The method of claim 1, wherein, The respectively determining information gain values of the plurality of power data for power load prediction comprises: respectively determining chaos degree information of each power data; extracting feature information of each chaos degree information; for each power data, determining an information gain value of the power data for power load prediction according to a difference between the chaos degree information and the feature information of the power data.

3. The method of claim 1, wherein, The screening target data from the plurality of power data according to the plurality of information gain values comprises: acquiring a preset information gain threshold corresponding to the plurality of information gain values; screening target data with an information gain value greater than the preset information gain threshold from the plurality of power data.

4. The method of claim 1, wherein, The optimizing the target data according to a data cleaning strategy to obtain optimized data comprises: removing noise from the target data according to a noise removal strategy in the data cleaning strategy to obtain denoised data; identifying the denoised data according to an abnormality identification strategy in the data cleaning strategy to obtain abnormal data and normal data; optimizing the abnormal data according to a data repair strategy in the data cleaning strategy to obtain repaired data; converting the format of the normal data and the repaired data according to a format conversion strategy in the data cleaning strategy to obtain optimized data.

5. The method of claim 1, wherein, After the optimizing the target data according to a data cleaning strategy to obtain optimized data, the method further comprises: determining a data quality index value of the optimized data; optimizing the data cleaning strategy according to the data quality index value.

6. The method of claim 5, wherein, The determining a data quality index value of the optimized data comprises at least one of: determining a data integrity of the optimized data and determining the data integrity as the data quality index value of the optimized data; determining measured data corresponding to the optimized data, and determining a data quality index value of the optimized data according to a difference between the optimized data and the measured data; determining a data consistency index value of the optimized data, and determining the data consistency index value as the data quality index value of the optimized data.

7. An electric power data cleaning device, characterized by, The device comprises: a data acquisition module configured to acquire a plurality of power data of a power system; a gain determination module configured to respectively determine information gain values of the plurality of power data for power load prediction; a data screening module configured to screen target data from the plurality of power data according to the plurality of information gain values; a data cleaning module configured to optimize the target data according to a data cleaning strategy to obtain optimized data.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 6.

Images