Data processing method and device, electronic equipment and storage medium
By classifying power data to determine processing needs and allocating tasks accordingly, the method optimizes data processing efficiency and resource utilization in edge and cloud systems, addressing the inefficiencies of existing systems.
Patent Information
- Application Number
- CN202411122438.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-15
- Publication Date
- 2024-12-03
AI Technical Summary
Existing data processing systems face challenges in efficiently allocating tasks between cloud and edge computing resources, leading to waste of computing resources and low computing efficiency due to unreasonable task distribution.
A data processing method that classifies power data based on classification results to determine required processing capacity, transmitting certain data to the cloud server and processing others locally on the edge computing device, utilizing a pre-trained joint classification model for enhanced accuracy and efficiency.
This approach optimizes task allocation, improves computing efficiency, and enhances resource utilization by aligning data processing tasks with the characteristics of power data, thereby improving collaborative control between edge and cloud systems.
Smart Images

Figure CN119066575A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing technology, and in particular to a data processing method, device, electronic equipment and storage medium. Background Art
[0002] With the widespread application of modern information technology and advanced communication technology in power systems, complex power network operation and control, a large amount of operation data is generated during the operation of power systems. Massive data and diversified demands have brought greater challenges to data processing.
[0003] Cloud computing can be used to process large-scale data processing tasks corresponding to massive data, but it has limitations in low-latency applications. The computing resources on the edge are limited and cannot handle large-scale or complex data processing tasks. For the above situation, the existing technology adopts a cloud-edge collaborative control method to process data processing tasks, but it is unable to reasonably allocate data processing tasks and cannot meet the needs of data computing and processing. There are problems of waste of computing resources and low computing efficiency. Summary of the invention
[0004] The present invention provides a data processing method, device, electronic device and storage medium to solve the problems of unreasonable data processing task allocation and low computing efficiency existing in the prior art.
[0005] According to one aspect of the present invention, there is provided a data processing method, which is applied to an edge computing device, comprising:
[0006] Acquiring power data to be processed, wherein the power data to be processed includes operation data of the power system at multiple sampling moments;
[0007] Classify the power data to be processed and determine the classification results corresponding to the power data to be processed; wherein the classification results are used to characterize the equipment processing capacity required for the processing of the power data to be processed;
[0008] Transmitting the unprocessed power data whose classification results meet the first condition to the cloud server, and receiving the processing result of the unprocessed power data by the cloud server;
[0009] The power data to be processed whose classification results meet the second condition is processed locally on the edge computing device to obtain the processing result of the power data to be processed.
[0010] Optionally, obtaining the power data to be processed includes: collecting original power data according to a preset data collection method; performing data preprocessing on the original power data to obtain initial power data to be processed, wherein the preprocessing includes one or more of data alignment processing, data type conversion processing, data encoding format conversion processing, data standardization processing and data normalization processing; performing data quality check on the initial power data to be processed to obtain power data to be processed that meets preset data qualification requirements.
[0011] Optionally, the power data to be processed is classified and processed to determine the classification results corresponding to the power data to be processed, including: classifying the power data to be processed based on a pre-trained joint classification model to determine the classification results corresponding to the power data to be processed, wherein the pre-trained joint classification model includes an independent component analysis sub-model, anomaly detection sub-model and classification sub-model.
[0012] Optionally, classification processing is performed on the power data to be processed based on a pre-trained joint classification model, including: extracting features of the power data to be processed through an independent component analysis sub-model to obtain target feature data; processing the target feature data through an anomaly detection sub-model to determine the abnormal score data corresponding to each feature data in the target feature data; fusing the abnormal score data with the target feature data to obtain fused feature data, and classifying the fused feature data through a classification sub-model to obtain a classification result corresponding to the power data to be processed.
[0013] Optionally, before transmitting the pending power data whose classification results meet the first condition to the cloud server, the steps include: compressing and encrypting the pending power data whose classification results meet the first condition; and determining the priority of the pending power data based on the classification result of the pending power data, wherein the priority is used to control the upload order of the pending power data.
[0014] Optionally, the power data to be processed whose classification results meet the second condition is locally processed on the edge computing device to obtain the processing result of the power data to be processed, including: locally processing the power data to be processed whose classification results meet the second condition based on a preset algorithm to obtain the processing result of the power data to be processed, wherein the preset algorithm includes one or more of a dynamic programming algorithm and a greedy algorithm.
[0015] Optionally, the processing results of the cloud server on the processed power data include one or more of load forecasting results, adjusted power generation plan, load transfer strategy and maintenance time information; the method also includes: updating the locally stored target control strategy based on the processing results of the cloud server on the processed power data, wherein the target control strategy is used to control the operation process of the equipment in the power system.
[0016] According to another aspect of the present invention, there is provided a data processing device, comprising:
[0017] A data acquisition module, used to acquire power data to be processed, wherein the power data to be processed includes operation data of the power system at multiple sampling moments;
[0018] A classification result determination module is used to classify the power data to be processed and determine the classification result corresponding to the power data to be processed; wherein the classification result is used to characterize the equipment processing capacity required for the processing process of the power data to be processed;
[0019] A data transmission module, used to transmit the unprocessed power data whose classification results meet the first condition to the cloud server, and receive the processing result of the unprocessed power data from the cloud server;
[0020] The data processing module is used to locally process the power data to be processed whose classification results meet the second condition on the edge computing device to obtain the processing results of the power data to be processed.
[0021] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:
[0022] at least one processor; and
[0023] a memory communicatively connected to at least one processor; wherein,
[0024] The memory stores a computer program that can be executed by at least one processor. The computer program is executed by at least one processor so that the at least one processor can execute the data processing method of any embodiment of the present invention.
[0025] According to another aspect of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium stores computer instructions, and the computer instructions are used to implement the data processing method of any embodiment of the present invention when executed by a processor.
[0026] The technical solution of the embodiment of the present invention is to obtain the power data to be processed, wherein the power data to be processed includes the operation data of the power system at multiple sampling moments; classify the power data to be processed, and determine the classification results corresponding to the power data to be processed; wherein the classification results are used to characterize the equipment processing capacity required for the processing process of the power data to be processed; transmit the power data to be processed whose classification results meet the first condition to the cloud server, and receive the processing results of the power data to be processed by the cloud server; and locally process the power data to be processed whose classification results meet the second condition on the edge computing device to obtain the processing results of the power data to be processed. This solution classifies the power data to be processed, determines the classification results, determines the data uploaded to the cloud server and determines the data to be processed on the edge computing device according to the classification results, realizes the reasonable allocation of data processing tasks of the edge computing device and the cloud server, makes full use of the data processing capabilities and resources of the cloud server and the edge computing device, solves the problems of unreasonable allocation of data processing tasks and low computing efficiency, makes the allocation of data processing tasks more in line with the characteristics of power data, helps to improve the collaborative control process between the edge computing device and the cloud server, and improves data processing efficiency and resource utilization.
[0027] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0029] Figure 1 is a flow chart of a data processing method provided by Embodiment 1 of the present invention;
[0030] Figure 2 is a flow chart of a data processing method provided by Embodiment 2 of the present invention;
[0031] Figure 3 is a flow chart of a data processing method applicable to an embodiment of the present invention;
[0032] Figure 4 is a structural schematic diagram of a data processing device provided in Embodiment 3 of the present invention;
[0033] Figure 5It is a structural schematic diagram of an electronic device for implementing the data processing method of an embodiment of the present invention. DETAILED DESCRIPTION
[0034] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0035] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0036] Embodiment 1
[0037] Figure 1 is a flowchart of a data processing method provided in the first embodiment of the present invention. This embodiment is applicable to the case of performing data processing. The method can be executed by a data processing device. The data processing device can be implemented in the form of hardware and / or software. The data processing device can be configured in an edge computing device, a computer or other electronic device. The method can be applied to edge computing devices, such as Figure 1 As shown, the method includes:
[0038] S110: Acquire power data to be processed, wherein the power data to be processed includes operation data of the power system at multiple sampling moments.
[0039] Among them, the power data to be processed specifically represents the data generated by the power system during operation, including but not limited to voltage data, current data, power factor data, temperature data, harmonic data, vibration data and image data. The various power data can be collected by setting corresponding data acquisition equipment, wherein the acquisition equipment includes but is not limited to voltage sensors, current sensors, temperature sensors, wind turbine vibration monitoring equipment and image acquisition equipment. It should be noted that different types of data have different sampling frequencies. The sampling frequency of each data type can be pre-set according to actual needs, and then data can be collected according to the sampling frequency of the device. The collected data can be stored in a preset storage space for calling at any time, or the power data to be processed can be obtained by real-time acquisition. The power data to be processed can be obtained by real-time acquisition or directly read from a preset target storage space.
[0040] Specifically, during the operation of the power system, data collection can be performed through a preset data collection plan, and the data collection equipment can be controlled to continuously collect real-time power data to obtain the operation data of the power system at multiple sampling moments for subsequent data processing. The information in the data collection plan may include data collection items and the collection frequency corresponding to each data collection item. The data collection equipment converts the collected data into a standardized data format and transmits it to the edge computing device. Exemplarily, the standardized data format may be IEC61850 or DNP3.
[0041] Optionally, obtaining the power data to be processed includes: collecting original power data according to a preset data collection method; performing data preprocessing on the original power data to obtain initial power data to be processed, wherein the preprocessing includes one or more of data alignment processing, data type conversion processing, data encoding format conversion processing, data standardization processing and data normalization processing; performing data quality check on the initial power data to be processed to obtain power data to be processed that meets preset data qualification requirements.
[0042] Among them, the preset data collection method can be a real-time collection method and a timed collection method. The original power data specifically refers to the data collected by the collection equipment and has not been processed in any way. Data alignment processing is used to unify the data of time series with different sampling frequencies to the same time base, and data alignment processing can be performed by resampling. Data type conversion processing can be a process of unifying the precision and unit of data, which can be processed by calling a preset data type conversion algorithm. Data encoding format conversion processing can be a process of unifying the encoding format of data to a preset encoding format, which can be processed by calling a preset encoding format conversion algorithm. Data standardization and normalization are used to map different data on the same scale for easy comparison and analysis. Standardization and normalization methods include but are not limited to Min-Max normalization, Z-score normalization and Robust Scaler normalization. Different types of data have different standardization and normalization methods. For example, voltage, current and other data are mapped to the [0,1] interval using Min-Max normalization, harmonic data with a lot of noise or outliers are normalized using Robust Scaler, power factor is generally normalized using Min-Max, temperature data is normalized using Z-score, and vibration data is normalized using Robust Scaler. Data quality inspection is used to detect whether there is unreasonable data in power data. The unreasonable data can be determined by the threshold method. The corresponding threshold range is set in advance for each type of data in the power data, and each data in the power data is determined to meet the corresponding threshold range. If there is unsatisfactory power data, the unsatisfactory power data is corrected to ensure that the power data used for subsequent processing is qualified data. The preset data qualification requirement is specifically used to determine whether the power data is qualified, and can be set according to the normal fluctuation range of the data. For example, for each data in the power data to be processed, it can be set that when the data meets the corresponding preset threshold, it indicates that the data meets the preset data qualification requirement and is qualified data. Otherwise, it indicates that the data does not meet the preset data qualification requirement and is unqualified data. When it is determined that there is unqualified data, the unqualified data is corrected and preprocessed and data quality checked again.
[0043] Specifically, the original power data is collected in real-time or timed manner; the original power data is preprocessed according to one or more preprocessing algorithms of data alignment processing, data type conversion processing, data encoding format conversion processing, data standardization processing and data normalization processing to obtain initial power data to be processed; the preset threshold values corresponding to each type of data in the initial power data to be processed are obtained, the initial power data to be processed are traversed, and each data in the initial power data to be processed is compared with the corresponding preset threshold values respectively. If the data meets the corresponding preset threshold values, it indicates that the data is qualified data, otherwise, the data is unqualified data.
[0044] In this embodiment, the collected raw power data is preprocessed to eliminate dimensional differences between different data to facilitate comparison and analysis. The quality of the acquired power data to be processed is also guaranteed through data quality checks, which helps to improve the accuracy of data processing.
[0045] S120, classify the power data to be processed and determine the classification result corresponding to the power data to be processed; wherein the classification result is used to characterize the equipment processing capacity required for the processing process of the power data to be processed.
[0046] It should be noted that the power data to be processed includes multiple data types, such as voltage data, current data, temperature data, harmonic data and vibration data. The computational complexity and real-time requirements of the data processing process of different data types are different. The processing process of abnormal information generated by data of different data types is also different, resulting in different requirements for the processing capacity of executing the data processing process. It is necessary to classify the power data to be processed to determine the equipment required for the processing process of the power data to be processed according to the classification results. The processing equipment includes but is not limited to edge computing equipment and cloud servers. In this embodiment, the power data to be processed is specifically classified and processed to determine the data to be processed on the edge computing device and the data to be processed on the cloud server. The classification processing can be classified by a preset classification model, or it can be classified according to a preset data classification mapping relationship table. The classification processing method can be set according to actual needs.
[0047] Specifically, when the power data to be processed is obtained, the power data to be processed is classified by calling a pre-trained classification model to obtain the classification result of the power data to be processed, wherein the classification result is used to characterize the equipment processing capacity required for the processing process of the power data to be processed. Exemplarily, the classification result includes multiple classification levels, and the classification result can be divided into five levels. The first level has extremely low requirements for the equipment processing capacity required for the processing process, and can process data that fluctuates within a normal range, that is, it can complete daily monitoring tasks. The second level has a higher requirement for the equipment processing capacity required for the processing process than the first level, and can process data with slight fluctuations, that is, process slightly abnormal data. The third level has a higher requirement for the equipment processing capacity required for the processing process than the second level, and can process abnormal data with a first degree of computational complexity. The fourth level has a higher requirement for the equipment processing capacity required for the processing process than the third level, and can process abnormal data with a second degree of computational complexity. The fifth level has a higher requirement for the equipment processing capacity required for the processing process than the fourth level, and can process abnormal data with a third degree of computational complexity. The first degree of computational complexity is lower than the second degree of computational complexity, and the second degree of computational complexity is lower than the third degree of computational complexity. It is understandable that the higher the computational complexity, the higher the demand for equipment processing capabilities, and the higher the abnormality of abnormal data, the higher the demand for equipment processing capabilities. In addition, the processing of some data also has certain requirements for low latency. Therefore, the latency factor is also taken into account in the classification process to ensure that the classification results can accurately characterize the demand for processing capabilities of the data processing process of the power data to be processed, so as to accurately determine the data processing equipment corresponding to the power data to be processed. Preferably, for devices with low latency requirements, edge computing devices are used for local calculations, which avoids wasted time during data transmission.
[0048] In this embodiment, the power data to be processed is classified to determine the processing capability requirements of the power data to be processed for the data processing process, and the corresponding processing equipment is determined according to the classification results of the power data to be processed. This helps to reasonably allocate computing resources for the power data to be processed, avoid resource waste and untimely response, and help improve data processing efficiency.
[0049] S130, transmitting the unprocessed power data whose classification results meet the first condition to the cloud server, and receiving the processing result of the unprocessed power data by the cloud server.
[0050] The first condition can be understood as a condition set for determining the data to be uploaded to the cloud server, and is specifically set according to the classification result. For example, if the classification result is characterized by multiple classification levels, at least one classification level in the classification result can be used as the first condition, that is, the data of the fifth level can be used as the data uploaded to the cloud server, or the data of the fourth level and the fifth level can be used as the data uploaded to the cloud server. The first condition can be set in advance and stored in a preset storage space for easy access at any time.
[0051] Specifically, the classification results are traversed, the classification results are compared with the first condition, and the power data to be processed that meets the first condition is screened out as the data to be transmitted, and the data to be transmitted is transmitted to the cloud server, and the cloud reader processes the received power data to be processed to obtain the processing result corresponding to the power data to be processed. The cloud server feeds back the processing result of the determined power data to be processed to the corresponding edge processing device. The cloud server can be responsible for processing large-scale and complex computing tasks.
[0052] In this embodiment, by setting a first condition and performing screening based on the first condition and the classification result, the power data to be processed that meets the first condition is screened out, and the power data to be processed is uploaded to the cloud server, and the cloud server performs computing tasks on the power data to be processed to realize complex processing tasks on the power data to be processed, thereby solving the problem of being limited by the computing power and resources of the edge computing device, so that the needs for processing complex computing tasks and data can be met.
[0053] Optionally, before transmitting the pending power data whose classification results meet the first condition to the cloud server, the steps include: compressing and encrypting the pending power data whose classification results meet the first condition; and determining the priority of the pending power data based on the classification result of the pending power data, wherein the priority is used to control the upload order of the pending power data.
[0054] Among them, the priority of the power data to be processed specifically represents the upload order of the data, and can be set for the power data to be processed according to the classification results. For example, the higher the classification level in the classification result, the higher the corresponding priority. A mapping relationship table of the priorities corresponding to each classification result can also be set in advance, and the priority corresponding to the power data to be processed can be determined according to the mapping relationship table.
[0055] Specifically, when it is determined that the power data to be processed meets the first condition, the power data to be processed is compressed and encrypted to improve the data transmission speed and data security, wherein the encryption algorithm can adopt the AES-256 encryption algorithm, and the data compression can adopt the GZIP algorithm for compression, or the data encryption algorithm and data compression method can be selected according to actual needs. In addition, the priority of the power data to be processed transmitted to the cloud server can be set according to the classification result of the power data to be processed, and the upload order of the power data to be processed can be controlled according to the priority. Specifically, the data with the highest priority is uploaded first, and the corresponding power data to be processed is uploaded in descending order of priority. Among them, the transmission protocol used to upload the power data to be processed to the cloud server can be the MQTT protocol or the CoAP protocol, which is suitable for the Internet of Things scenario.
[0056] In a specific embodiment, the cloud server receives the power data to be processed, decrypts the power data to be processed, and uses it for in-depth analysis and global optimization processing. The big data processing platform can be used for processing, wherein the big data processing platform can be Apache Hadoop or Spark; the data is processed through a deep learning model, which can be an LSTM or Transformer model, and the power data to be processed is used as the input data of the model. At the same time, historical load data, weather information, and holiday information are also used as input parameters. The deep learning model processes the data and outputs the model processing results, wherein the model processing results include the load forecast results within a preset time period in the future, and the optimization suggestions and maintenance plans are determined based on the forecast results. The optimization suggestions include adjusting the power generation plan and the load transfer plan, and the maintenance plan includes the equipment preventive maintenance schedule. The cloud server feeds back the optimization suggestions and maintenance plans to the edge computing device.
[0057] S140. Process the power data to be processed whose classification results meet the second condition locally on the edge computing device to obtain a processing result of the power data to be processed.
[0058] The second condition can be specifically understood as a condition set for determining the data to be processed locally, and is specifically set according to the classification result. For example, if the classification result is characterized by multiple classification levels, then at least one classification level in the classification result can be used as the second condition, that is, the data of the first level can be used as the data to be processed locally, or the data of the first level and the second level can be used as the data to be processed locally. The second condition can be set in advance and stored in a preset storage space for easy access at any time. Preferably, data that does not meet the first condition can be directly used as data to be processed locally.
[0059] Specifically, the classification results are traversed, the classification results are compared with the second condition, and the power data to be processed that meets the second condition is screened out as data to be processed locally, that is, the screened power data to be processed is processed on the edge computing device to obtain the processing result of the power data to be processed.
[0060] Optionally, the power data to be processed whose classification results meet the second condition is locally processed on the edge computing device to obtain the processing result of the power data to be processed, including: locally processing the power data to be processed whose classification results meet the second condition based on a preset algorithm to obtain the processing result of the power data to be processed, wherein the preset algorithm includes one or more of a dynamic programming algorithm and a greedy algorithm.
[0061] Specifically, the power data to be processed whose classification results meet the second condition is processed locally by a preset algorithm to obtain the processing result of the power data to be processed, that is, the edge computing device performs real-time analysis and response load balancing processing on the power data to be processed, so as to minimize the peak-to-valley difference, improve power supply stability and simple fault diagnosis. The preset algorithm includes one or more of a dynamic programming algorithm and a greedy algorithm.
[0062] Based on the above embodiments, the processing results of the cloud server on the processed power data include one or more of load forecasting results, adjusted power generation plan, load transfer strategy and maintenance time information; the method also includes: updating the locally stored target control strategy based on the processing results of the cloud server on the processed power data, wherein the target control strategy is used to control the operation process of the equipment in the power system.
[0063] Specifically, one or more of the load forecast results, adjusted power generation plan, load transfer strategy and maintenance time information fed back by the cloud server are received, and the various data fed back by the cloud server are analyzed. The analysis results can be compared with the target control strategy, and the corresponding parameter information in the target control strategy can be updated according to the analysis results to determine the updated target control strategy, which is used to control the equipment in the power system to operate according to the updated target control strategy.
[0064] Preferably, after controlling the equipment in the power system to operate according to the updated target control strategy, the execution effect of the updated target control strategy is continuously monitored, the operating data of the power system is collected, and the operating data is fed back to the cloud server in real time for further optimization, thereby realizing closed-loop control between the edge computing device and the cloud server, improving the processing efficiency of the power data, and thereby improving the control effect of the power system.
[0065] The technical solution of this embodiment is to obtain the power data to be processed, wherein the power data to be processed includes the operation data of the power system at multiple sampling moments; classify the power data to be processed and determine the classification results corresponding to the power data to be processed; wherein the classification results are used to characterize the equipment processing capacity required for the processing process of the power data to be processed; transmit the power data to be processed whose classification results meet the first condition to the cloud server, and receive the processing results of the power data to be processed by the cloud server; and locally process the power data to be processed whose classification results meet the second condition on the edge computing device to obtain the processing results of the power data to be processed. This solution classifies the power data to be processed, determines the classification results, determines the data to be uploaded to the cloud server and determines the data to be processed on the edge computing device according to the classification results, realizes the reasonable allocation of data processing tasks of the edge computing device and the cloud server, makes full use of the data processing capabilities and resources of the cloud server and the edge computing device, solves the problems of unreasonable allocation of data processing tasks and low computing efficiency, makes the allocation of data processing tasks more in line with the characteristics of power data, helps to improve the collaborative control process between the edge computing device and the cloud server, and improves data processing efficiency and resource utilization.
[0066] Embodiment 2
[0067] Figure 2 : is a flow chart of a data processing method provided by the second embodiment of the present invention. This embodiment is a further optimization of the method of the above embodiment. Optionally, the power data to be processed is classified based on a pre-trained joint classification model to determine the classification result corresponding to the power data to be processed, wherein the pre-trained joint classification model includes an independent component analysis sub-model, anomaly detection sub-model and classification sub-model. Figure 2 As shown, the method includes:
[0068] S210: Obtain power data to be processed, wherein the power data to be processed includes operation data of the power system at multiple sampling moments.
[0069] S220, classify the power data to be processed based on the pre-trained joint classification model to determine the classification result corresponding to the power data to be processed, wherein the pre-trained joint classification model includes an independent component analysis sub-model, an anomaly detection sub-model and a classification sub-model.
[0070] Among them, the pre-trained joint classification model specifically refers to the integration of the independent component analysis sub-model, the anomaly detection sub-model and the classification sub-model. The independent component analysis sub-model is used to extract features from the power data to be processed. For example, the independent component analysis sub-model adopts the standard ICA model; the anomaly detection sub-model is used to perform anomaly detection and determine the anomaly score data corresponding to the feature data of the power data to be processed. For example, the anomaly detection sub-model can adopt the IF model (isolation forest model); the classification sub-model is used to classify the data output by the independent component analysis sub-model and the anomaly detection sub-model to obtain the classification result. For example, the classification sub-model can adopt the SVM (Support Vector Machine) classifier. Then, the pre-trained joint classification model can be characterized by the ICA-IF-SVM joint model.
[0071] Specifically, a pre-trained joint classification model is integrated in the edge computing device, and the acquired power data to be processed is classified and processed through the pre-trained joint classification model to obtain a classification result corresponding to the power data to be processed.
[0072] Optionally, classification processing is performed on the power data to be processed based on a pre-trained joint classification model, including: extracting features of the power data to be processed through an independent component analysis sub-model to obtain target feature data; processing the target feature data through an anomaly detection sub-model to determine the abnormal score data corresponding to each feature data in the target feature data; fusing the abnormal score data with the target feature data to obtain fused feature data, and classifying the fused feature data through a classification sub-model to obtain a classification result corresponding to the power data to be processed.
[0073] The independent component analysis sub-model in the pre-trained joint classification model is used to extract features from the power data to be processed to obtain the target feature data corresponding to the power data to be processed. The target feature data is used as the input parameter of the anomaly detection sub-model, which is processed by the anomaly detection sub-model to determine the anomaly score data corresponding to each feature data in the target feature data. The anomaly score data is fused into the target feature data of the power data to be processed as new feature information of the power data to be processed to obtain the fused feature data. The fused feature data is used as the input data of the classification sub-model, which is processed by the classification sub-model to output the classification result corresponding to the power data to be processed.
[0074] In this embodiment, the abnormal score data of the power data to be processed is determined by the abnormal detection sub-model, and the abnormal score data is added as a new data feature to the target feature data of the power data to be processed to determine the classification result, which helps to improve the accuracy of data classification.
[0075] Preferably, before extracting features from the power data to be processed by the independent component analysis sub-model in the pre-trained joint classification model, the power data to be processed is standardized so that its mean is zero and its variance is unified to 1, thereby eliminating dimensional differences between different data sets.
[0076] For the construction of the independent component analysis sub-model, a sample data set is obtained, and the sample data set is standardized to obtain a new data matrix p. Taking p as input, it is decomposed into:
[0077] p=Bx+e
[0078] Among them, B is the mixing matrix, x is the independent element vector, and e is the error vector. The goal of the ICA sub-model is to estimate the matrix B and x through the observed sample data set, so that the separated signal can be closest to the source signal. Therefore, it is necessary to select a suitable objective function and optimization algorithm to train B for different observed signals to maximize or minimize it.
[0079] Furthermore, the separation matrix C and the reconstructed separation matrix x′ are calculated, namely:
[0080] x'=PC
[0081] The data needs to be centered and whitened so that its covariance matrix P is a unit matrix.
[0082] e{p T p}=HΛH T
[0083] Among them, the H matrix is composed of eigenvectors corresponding to the eigenvalues, and the Λ matrix is a diagonal matrix composed of the eigenvalues of the covariance matrix P.
[0084] The covariance matrix P is whitened by singular value decomposition (SVD) to obtain the whitened vector n, and then the independent element s is estimated and calculated:
[0085] n=WP
[0086]
[0087] Where W is the whitening matrix and n is the whitening vector, we can get:
[0088] n=WP=WBx
[0089]
[0090] Finally, the FastICA algorithm is used to obtain the independent element matrix s. The DW value is used to measure whether there are important features in ICA through the DW criterion. The DW value is obtained according to the following formula:
[0091]
[0092] Among them, δS i ,δS i-1 is the residual in the consecutive points in the vector, and n is the variable.
[0093] The IF (Isolation Forest) sub-model is applied to the independent components extracted by the ICA sub-model to identify and isolate potential anomalies or fault points. The IF sub-model evaluates the abnormality of data points by randomly constructing decision trees. Anomalies are often more easily isolated due to their characteristics of deviating from the normal pattern. λ pieces of data are extracted from the original data set to form a sub-dataset M, which will be used to build the isolation forest. For each sub-dataset, an isolation forest is constructed based on the independent components extracted by the ICA sub-model. During the construction of each tree, a dimension q∈M is randomly selected and a cutting point a is determined. The data set is divided into a left subtree and a right subtree using the cutting point a, where data with q≤a is placed in the left subtree and data with q>a is placed in the right subtree. Construction and cutting are recursively applied to each subtree until the stopping condition is met, such as reaching the maximum tree depth d or only one data point is left in the child node. Repeat the above steps to generate k isolated trees to form an isolation forest.
[0094] The constructed isolation forest sub-model is used to calculate the anomaly score data for each data point in the original data set. The shorter the path length (i.e., the number of edges from the root to the leaf) of the data point in the isolation tree, the more abnormal the data point is. The anomaly of the data point is evaluated based on the anomaly score. A low score (i.e., a short path length) indicates that the data point may be abnormal, while a high score (i.e., a long path length) indicates that the data point is normal. The features extracted by the ICA sub-model are combined with the anomaly score determined by the isolation forest sub-model to provide a more comprehensive data analysis. For example, data points that are considered important components in the output results of the ICA sub-model and receive low anomaly scores in the isolation forest sub-model can be preferentially examined.
[0095] The SVM sub-model is used as a classifier to take the data processed by the ICA sub-model and the IF sub-model as input, and the SVM sub-model is used to make the final classification decision. The SVM sub-model effectively separates and classifies different fault types by finding the optimal segmentation hyperplane in the high-dimensional space: Assume that the sample training data set is H = {(x1, y1), (x2, y2), …, (x m ,y m )},y i ∈{-1,+1}; the basic form of the SVM sub-model can be expressed as:
[0096]
[0097] The constraints are: ξ≥0,i=1,2,3,…,l. Where: φ is a nonlinear mapping; ω is the normal vector of the SVM hyperplane; ξ i is the slack variable; parameter C is the penalty factor; x i is the i-th fault sample; b is the offset of the SVM hyperplane; y i is a category label. During the training process, the corresponding category label can be calibrated for the training data for model training. Then the Lagrangian function is constructed to transform it into a dual function for solution:
[0098]
[0099] The constraints are: In the formula, α i is the Lagrange multiplier, so the decision function is:
[0100]
[0101] Where: K(x,x i ) is the kernel function.
[0102] Therefore, when SVM classifies data, if the data is linearly distributed, SVM will perform linear classification on it; if the data is nonlinearly distributed, SVM will map it to a high-dimensional space and introduce kernel functions to solve high-dimensional calculation problems, thereby efficiently and accurately classifying the data.
[0103] Through multi-classification support SVM technology, the acquired power data to be processed is classified, and the classification results can be set according to actual needs, so as to achieve the optimal allocation of computing resources based on the classification results. Exemplarily, the power data to be processed is intelligently divided into five levels through a joint classification model, and level 1-2 lightweight data processing is recommended to edge computing devices for processing to take advantage of its fast response and low latency. Level 3-5 data processing tasks, including medium, heavy and extremely heavy tasks, are more suitable for execution in a cloud server environment to give full play to its powerful computing power and storage resources. This grading strategy not only improves the efficiency of resource utilization, but also ensures the flexibility and adaptability of task processing.
[0104] S230, transmitting the unprocessed power data whose classification results meet the first condition to the cloud server, and receiving the processing result of the unprocessed power data by the cloud server.
[0105] S240. Process the power data to be processed whose classification results meet the second condition locally on the edge computing device to obtain a processing result of the power data to be processed.
[0106] In a specific embodiment, Figure 3The flowchart of the data processing method shown in the figure first performs data collection to obtain data collected by sensors and data collected by cameras, preprocesses the obtained data, performs data quality check on the data preprocessing results, stores qualified data in a preset data warehouse, and performs data correction processing on unqualified data, returns to the data preprocessing stage to continue preprocessing and data quality check, obtains the power data to be processed from the data warehouse, uses the power data to be processed as input data of the ICA-IF-SVM joint classification model, obtains the corresponding classification result, determines the data to be processed in the edge computing device and the data to be transmitted to the cloud server according to the classification result, and the edge computing device processes the data calculated locally to obtain the processing result; performs data encryption processing and data compression processing on the data uploaded to the cloud server, transmits the encrypted and compressed data to the cloud server via network transmission, and the cloud server receives the power data to be processed, decrypts and verifies the received power data to be processed, and transmits the decrypted and verified data to the cloud server. The data is stored for deep analysis and can be processed by a convolutional neural network and comprehensively optimized to generate a scheduling plan. The scheduling plan is returned to the corresponding edge computing device as a processing result. Accordingly, before returning the scheduling plan as a processing result to the corresponding edge computing device, the processing result is encrypted. When the edge computing device receives the encrypted processing result, it decrypts and verifies the processing result, parses the decrypted data, obtains the parsing result, and updates the locally stored local strategy, i.e., the target control strategy, according to the parsing result. The equipment controlling the power system operates according to the updated target control strategy, executes the instructions corresponding to the target control strategy and monitors the effect of the instructions, i.e., monitors the execution result of the updated target control strategy, collects the operation data in real time, and re-determines the data in the operation data that needs to be uploaded to the cloud server. In this way, closed-loop collaborative control between the edge computing device and the cloud server is realized, which improves the efficiency of data processing and the stability of the power system while realizing the reasonable allocation and utilization of computing resources.
[0107] The technical solution of this embodiment is to obtain the power data to be processed, wherein the power data to be processed includes the operation data of the power system at multiple sampling moments; classify the power data to be processed based on the pre-trained joint classification model, and determine the classification results corresponding to the power data to be processed, wherein the pre-trained joint classification model includes an independent component analysis sub-model, anomaly detection sub-model and classification sub-model; transmit the power data to be processed whose classification results meet the first condition to the cloud server, and receive the processing results of the power data to be processed by the cloud server; and locally process the power data to be processed whose classification results meet the second condition on the edge computing device to obtain the processing results of the power data to be processed. This solution classifies the power data to be processed, determines the classification results, determines the data uploaded to the cloud server and determines the data to be processed on the edge computing device according to the classification results, realizes the reasonable allocation of data processing tasks of the edge computing device and the cloud server, makes full use of the data processing capabilities and resources of the cloud server and the edge computing device, solves the problems of unreasonable allocation of data processing tasks and low computing efficiency, makes the allocation of data processing tasks more in line with the characteristics of power data, helps to improve the collaborative control process between the edge computing device and the cloud server, and improves data processing efficiency and resource utilization.
[0108] Embodiment 3
[0109] Figure 4 Schematic diagram of the structure of a data processing device provided by Embodiment 3 of the present invention. Figure 4 As shown, the device comprises:
[0110] The data acquisition module 410 is used to acquire the power data to be processed, wherein the power data to be processed includes the operation data of the power system at multiple sampling moments;
[0111] The classification result determination module 420 is used to classify the power data to be processed and determine the classification result corresponding to the power data to be processed; wherein the classification result is used to characterize the equipment processing capacity required for the processing process of the power data to be processed;
[0112] The data transmission module 430 is used to transmit the power data to be processed whose classification results meet the first condition to the cloud server, and receive the processing result of the power data to be processed by the cloud server;
[0113] The data processing module 440 is used to locally process the power data to be processed whose classification results meet the second condition on the edge computing device to obtain the processing results of the power data to be processed.
[0114] The technical solution of this embodiment is to obtain the power data to be processed through the data acquisition module, wherein the power data to be processed includes the operation data of the power system at multiple sampling moments; the classification result determination module classifies the power data to be processed and determines the classification result corresponding to the power data to be processed; wherein the classification result is used to characterize the equipment processing capacity required for the processing process of the power data to be processed; the data transmission module transmits the power data to be processed whose classification result meets the first condition to the cloud server, and receives the processing result of the power data to be processed by the cloud server; the data processing module processes the power data to be processed whose classification result meets the second condition locally on the edge computing device, and obtains the processing result of the power data to be processed. This solution classifies the power data to be processed, determines the classification result, determines the data to be uploaded to the cloud server and determines the data to be processed on the edge computing device according to the classification result, realizes the reasonable allocation of data processing tasks of the edge computing device and the cloud server, makes full use of the data processing capacity and resources of the cloud server and the edge computing device, solves the problem of unreasonable allocation of data processing tasks and low computing efficiency, makes the allocation of data processing tasks more in line with the characteristics of power data, and improves data processing efficiency.
[0115] On the basis of the above embodiment, optionally, the data acquisition module 410 is specifically used to collect original power data according to a preset data collection method; perform data preprocessing on the original power data to obtain initial power data to be processed, wherein the preprocessing includes one or more of data alignment processing, data type conversion processing, data encoding format conversion processing, data standardization processing and data normalization processing; perform data quality check on the initial power data to be processed to obtain power data to be processed that meets the preset data qualification requirements.
[0116] Optionally, the classification result determination module 420 includes a classification result determination unit, which is used to classify the power data to be processed based on a pre-trained joint classification model to determine the classification result corresponding to the power data to be processed, wherein the pre-trained joint classification model includes an independent component analysis sub-model, an anomaly detection sub-model and a classification sub-model. The classification result determination unit is specifically used to extract features of the power data to be processed through the independent component analysis sub-model to obtain target feature data; process the target feature data through the anomaly detection sub-model to determine the abnormal score data corresponding to each feature data in the target feature data; fuse the abnormal score data with the target feature data to obtain the fused feature data, and classify the fused feature data through the classification sub-model to obtain the classification result corresponding to the power data to be processed.
[0117] Optionally, before transmitting the pending power data whose classification results meet the first condition to the cloud server, the device is also used to compress and encrypt the pending power data whose classification results meet the first condition; and determine the priority of the pending power data based on the classification result of the pending power data, wherein the priority is used to control the upload order of the pending power data.
[0118] Optionally, the data processing module 440 is specifically used to locally process the power data to be processed whose classification results meet the second condition based on a preset algorithm to obtain a processing result of the power data to be processed, wherein the preset algorithm includes one or more of a dynamic programming algorithm and a greedy algorithm.
[0119] Optionally, the processing results of the cloud server on the processed power data include one or more of load forecasting results, adjusted power generation plan, load transfer strategy and maintenance time information; the device is also specifically used to update the locally stored target control strategy based on the processing results of the cloud server on the processed power data, wherein the target control strategy is used to control the operation process of the equipment in the power system.
[0120] The data processing device provided in the embodiment of the present invention can execute the data processing method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.
[0121] Embodiment 4
[0122] Figure 5 1 is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.
[0123] like Figure 5As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.
[0124] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.
[0125] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as a data processing method.
[0126] In some embodiments, the data processing method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the data processing method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the data processing method in any other appropriate manner (e.g., by means of firmware).
[0127] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.
[0128] The computer program for implementing the data processing method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.
[0129] Embodiment 5
[0130] Embodiment 5 of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to cause a processor to execute a data processing method, the method comprising:
[0131] Acquiring power data to be processed, wherein the power data to be processed includes operation data of the power system at multiple sampling moments;
[0132] Classify the power data to be processed and determine the classification results corresponding to the power data to be processed; wherein the classification results are used to characterize the equipment processing capacity required for the processing of the power data to be processed;
[0133] Transmitting the unprocessed power data whose classification results meet the first condition to the cloud server, and receiving the processing result of the unprocessed power data by the cloud server;
[0134] The power data to be processed whose classification results meet the second condition is processed locally on the edge computing device to obtain the processing result of the power data to be processed.
[0135] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
[0136] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).
[0137] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.
[0138] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.
[0139] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.
[0140] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A data processing method, characterized in that: Applied to edge computing devices, including: Acquiring power data to be processed, wherein the power data to be processed includes operation data of the power system at multiple sampling moments; Classify the power data to be processed and determine the classification result corresponding to the power data to be processed; wherein the classification result is used to characterize the equipment processing capacity required for the processing process of the power data to be processed; Transmitting the power data to be processed whose classification result meets the first condition to a cloud server, and receiving a processing result of the power data to be processed by the cloud server; The power data to be processed whose classification result meets the second condition is processed locally on the edge computing device to obtain a processing result of the power data to be processed.
2. The method according to claim 1, characterized in that Obtain power data to be processed, including: Collect raw power data according to the preset data collection method; Performing data preprocessing on the raw power data to obtain initial power data to be processed, wherein the preprocessing includes one or more of data alignment processing, data type conversion processing, data encoding format conversion processing, data standardization processing, and data normalization processing; A data quality check is performed on the initial power data to be processed to obtain the power data to be processed that meets preset data qualification requirements.
3. The method according to claim 1, characterized in that The classifying the power data to be processed and determining the classification result corresponding to the power data to be processed includes: The power data to be processed is classified based on a pre-trained joint classification model to determine a classification result corresponding to the power data to be processed, wherein the pre-trained joint classification model includes an independent component analysis sub-model, an anomaly detection sub-model and a classification sub-model.
4. The method according to claim 3, characterized in that The classifying process of the to-be-processed power data based on the pre-trained joint classification model includes: Extracting features of the power data to be processed by using the independent component analysis sub-model to obtain target feature data; Processing the target feature data through the anomaly detection sub-model to determine anomaly score data corresponding to each feature data in the target feature data; The abnormal score data is fused with the target feature data to obtain fused feature data, and the fused feature data is classified by the classification sub-model to obtain a classification result corresponding to the power data to be processed.
5. The method according to claim 1, characterized in that Before transmitting the to-be-processed power data whose classification result satisfies the first condition to a cloud server, the method includes: compressing and encrypting the power data to be processed whose classification results meet the first condition; and The priority of the power data to be processed is determined based on the classification result of the power data to be processed, wherein the priority is used to control the upload order of the power data to be processed.
6. The method according to claim 1, characterized in that The processing of the power data to be processed whose classification result satisfies the second condition locally on the edge computing device to obtain a processing result of the power data to be processed includes: Based on a preset algorithm, the power data to be processed whose classification result meets the second condition is locally processed to obtain a processing result of the power data to be processed, wherein the preset algorithm includes one or more of a dynamic programming algorithm and a greedy algorithm.
7. The method according to claim 1, characterized in that The processing result of the power data to be processed by the cloud server includes one or more of load forecasting result, adjustment of power generation plan, load transfer strategy and maintenance time information; The method further comprises: The locally stored target control strategy is updated based on the processing result of the power data to be processed by the cloud server, wherein the target control strategy is used to control the operation process of the equipment in the power system.
8. A data processing device, characterized in that: include: A data acquisition module, used to acquire power data to be processed, wherein the power data to be processed includes operation data of the power system at multiple sampling moments; A classification result determination module, used to classify the power data to be processed and determine the classification result corresponding to the power data to be processed; wherein the classification result is used to characterize the equipment processing capacity required for the processing process of the power data to be processed; A data transmission module, configured to transmit the power data to be processed whose classification result satisfies a first condition to a cloud server, and receive a processing result of the power data to be processed by the cloud server; The data processing module is used to locally process the power data to be processed whose classification results meet the second condition on the edge computing device to obtain a processing result of the power data to be processed.
9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the data processing method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the data processing method according to any one of claims 1 to 7 when executed.
Citation Information
Patent Citations
Method for classifying appearances of vehicles based on multi-feature fusion of surveillance video
CN103886331A
A structure damage identification method based on independent components and a support vector machine
CN109948673A
Computing unloading method and system based on cloud edge cooperation
CN112004239A
Electric power data edge calculation system and calculation method
CN112925646A
Cited By
Data cleaning method, device and system for edge end and storage medium
CN120216884A
Information acquisition monitoring system and method for semiconductor accessory equipment
CN120993839A
Data processing method and device based on edge computing and medium
CN121833279A
A data processing method, device and medium based on edge computing
CN121833279B
