Intelligent fault diagnosis method for high-throughput data network for power production
By building a high-throughput data network model and combining multimodal convolutional neural network and gradient improvement decision tree model, the problem of inaccurate fault diagnosis of optical transmission networks in power production is solved, and the stability and reliability of the power system are improved.
Patent Information
- Application Number
- CN202510206338.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-25
- Publication Date
- 2025-07-01
AI Technical Summary
The prior art is difficult to efficiently process high-throughput data of the optical transmission network during power production, resulting in inaccurate fault diagnosis and high maintenance costs, which affects the stability and reliability of the power system.
Build a high-throughput data network model, combine multimodal convolutional neural network and gradient improvement decision tree model, perform data preprocessing, fault feature extraction and real-time diagnosis of optical transmission networks, optimize model parameters, and formulate maintenance strategies.
It realizes rapid and accurate diagnosis of optical transmission network failures during power production, improves the stability and reliability of the power system, and reduces maintenance costs.
Smart Images
Figure CN120238780A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to power communication technology, and in particular to an intelligent fault diagnosis method for high-throughput data networks for power production. Background Art
[0002] During the power production process, with the rapid development of smart grids, the amount of data that power communication networks need to process has increased sharply, including status monitoring data of power equipment, control instructions, and user electricity consumption information, etc. The high-throughput transmission and real-time processing of these data are crucial for ensuring the safe, stable, and efficient operation of the power system.
[0003] To cope with the large amount of data generated in power production, high-throughput data processing technology has emerged. This technology can efficiently process and analyze massive data, extract valuable information, and provide strong support for the operation and management of the power system. High-throughput data processing technology usually includes multiple links such as data acquisition, storage, processing, analysis, and visualization, and requires the use of advanced computing resources and algorithms to achieve.
[0004] Intelligent diagnosis technology is a new method that uses advanced technologies such as artificial intelligence, machine learning, and data mining to predict, diagnose, and maintain power equipment. In power production, intelligent diagnosis technology can monitor the operating status of equipment in real time, discover potential fault hazards, and take measures in advance for repair, thereby avoiding the adverse effects caused by equipment failures on the power system. At the same time, intelligent diagnosis technology can also improve the accuracy and efficiency of fault diagnosis, reduce maintenance costs, and improve the overall reliability of the power system. Summary of the Invention
[0005] The technical problem to be solved by the present invention is to provide an intelligent fault diagnosis method for high-throughput data networks for power production in view of the defects in the prior art.
[0006] The technical solution adopted by the present invention to solve its technical problems is: an intelligent fault diagnosis method for high-throughput data networks for power production, including the following steps:
[0007] Step 1) During the power production process, collect the operating status data of the optical transport network; the operating data of the optical transport network includes the bandwidth utilization rate, data transmission rate, bit error rate, delay, and jitter data of the optical transport network; the status data of the optical transport network includes the fault type data corresponding to the operating data;
[0008] Step 2) Preprocess the collected operating data of the optical transport network, including removing outliers and noise;
[0009] Step 3) Based on the preprocessed data, construct a high-throughput data network model; the high-throughput data network model is used to reflect the operating status and performance of the optical transport network during the power production process;
[0010] Step 4) Continuously monitor the operating status of the optical transport network and update the model data in real time;
[0011] Step 5) Conduct network fault diagnosis for the high-throughput data network model of power production;
[0012] Step 6) According to the network fault diagnosis results during the power production process, formulate corresponding maintenance strategies for maintenance.
[0013] According to the above solution, in step 4), it further includes: analyzing the constructed model, evaluating the accuracy and reliability of the high-throughput data network model, and optimizing the high-throughput data network model according to the analysis results.
[0014] According to the above solution, in step 5), the network fault diagnosis is carried out as follows:
[0015] 5.1) Design a multi-modal convolutional neural network model, and the neural network model structure includes a convolutional layer, a pooling layer, and a fully connected layer;
[0016] 5.2) Input the operation data of the high-throughput data network model into the multi-modal convolutional neural network model to extract the local features of the data;
[0017] 5.3) Use the local features extracted by the multi-modal convolutional neural network as the input of the gradient boosting decision tree model;
[0018] 5.4) Use the gradient boosting algorithm to iteratively train multiple gradient boosting decision tree models, and each decision tree model is trained based on the residual of the previous model;
[0019] 5.5) Compare the actual fault situation with the model diagnosis results, and continuously optimize and update the model parameters according to the deviation of the diagnosis results;
[0020] 5.6) Use the trained fault diagnosis model to conduct fault diagnosis on the real-time operation data of the optical transport network; the fault diagnosis model includes a multi-modal convolutional neural network model and a gradient boosting decision tree model.
[0021] According to the above solution, in step 5.5), continuously optimizing and updating the model parameters includes adjusting the convolution kernel size, stride, pooling method in the multi-modal convolutional neural network, and the learning rate, number of iterations, and tree depth in the gradient boosting decision tree.
[0022] According to the above solution, step 6) includes:
[0023] Output the diagnostic result of the high-throughput data network system status according to step 5), including whether the system is operating normally, whether there are faults, and the type information of the faults;
[0024] According to the diagnostic result, obtain the specific location and specific maintenance strategy of the fault according to the preset fault analysis file, such as regular maintenance, equipment replacement, preventive maintenance, etc.
[0025] The beneficial effects produced by the present invention are:
[0026] The present invention provides a data network fault diagnosis method for power production requirements. First, a high-throughput data network model for power production is constructed, then network fault diagnosis is carried out, the fault diagnosis is intelligently optimized, and finally the prediction and maintenance of faults are carried out to achieve the stability and reliability of power production.
[0027] By combining the multi-modal convolutional neural network model and the gradient boosting decision tree model, the present invention can quickly and accurately diagnose various network faults that may occur in the power production process, effectively improve the stability and reliability of power production, and has wide application value. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] The present invention will be further described below in conjunction with the drawings and embodiments. In the drawings:
[0029] Figure 1 is the flowchart of the method of the embodiment of the present invention;
[0030] Figure 2 is the flowchart of the fault diagnosis method of the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0031] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0032] As Figure 1 shown, a high-throughput data network fault intelligent diagnosis method for power production includes the following steps:
[0033] Step 1) Install high-precision sensors and monitoring devices in the optical transport network of power production to collect the operation status data of the optical transport network; the operation data of the optical transport network includes the bandwidth utilization rate, data transmission rate, bit error rate, delay and jitter data of the optical transport network; the status data of the optical transport network includes the fault type data corresponding to the operation data;
[0034] Step 2) Preprocess the collected operation data of the optical transport network, including removing outliers and noise;
[0035] Step 3) Based on the preprocessed data, construct a high-throughput data network model; the high-throughput data network model is used to reflect the operation status and performance of the optical transport network during the power production process;
[0036] Step 4) Continuously monitor the operation status of the optical transport network and update the model data in real time;
[0037] Then analyze the constructed model, evaluate the accuracy and reliability of the high-throughput data network model, and optimize the high-throughput data network model according to the analysis results;
[0038] Step 5) Conduct network fault diagnosis for the high-throughput data network model of power production;
[0039] Such as Figure 2 , conduct network fault diagnosis, specifically as follows:
[0040] 5.1) Design a multi-modal convolutional neural network model, and the neural network model structure includes a convolutional layer, a pooling layer, and a fully connected layer;
[0041] 5.2) Input the operation data of the high-throughput data network model into the multi-modal convolutional neural network model to extract the local features of the data;
[0042] 5.3) Use the local features extracted by the multi-modal convolutional neural network as the input of the gradient boosting decision tree model;
[0043] 5.4) Use the gradient boosting algorithm to iteratively train multiple gradient boosting decision tree models, and each decision tree model is trained based on the residual of the previous model;
[0044] 5.5) Compare the actual fault situation with the model diagnosis result, and continuously optimize and update the model parameters according to the deviation of the diagnosis result;
[0045] Continuously optimize and update the model parameters, including adjusting the convolution kernel size, stride, pooling method in the multi-modal convolutional neural network, and the learning rate, number of iterations, and tree depth in the gradient boosting decision tree;
[0046] 5.6) Use the trained fault diagnosis model to conduct fault diagnosis on the real-time operation data of the optical transport network; the fault diagnosis model includes a multi-modal convolutional neural network model and a gradient boosting decision tree model;
[0047] Step 6) According to the network fault diagnosis result during the power production process, formulate corresponding maintenance strategies for maintenance;
[0048] Output the diagnostic results of the high-throughput data network system status according to step 5), including whether the system is operating normally, whether there are faults, and the type information of the faults;
[0049] According to the diagnostic results, obtain the specific location and specific maintenance strategies of the faults according to the preset fault analysis file, such as regular maintenance, equipment replacement, preventive maintenance, etc.
[0050] Among them, the fault analysis file is established based on historical fault data and operation status data, and according to the actual operation situation and maintenance effect, continuously optimize the fault maintenance strategy, collect feedback and new data, and update the fault analysis file.
[0051] The intelligent fault diagnosis method for high-throughput data networks for power production provided by the present invention can quickly and accurately diagnose various network faults that may occur in the power production process, effectively improve the stability and reliability of power production, and has wide application value.
[0052] It should be understood that those of ordinary skill in the art can make improvements or transformations according to the above description, and all such improvements and transformations should fall within the protection scope of the appended claims of the present invention.
Claims
1. A high-throughput data network fault intelligent diagnosis method for power production, characterized in that: The following steps are involved: Step 1) During the power production process, operating status data of the optical transmission network is collected; the operating data of the optical transmission network includes bandwidth utilization, data transmission rate, bit error rate, delay and jitter data of the optical transmission network; the status data of the optical transmission network includes fault type data corresponding to the operating data; Step 2) preprocessing the collected operation data of the optical transmission network, including removing outliers and noise; Step 3) constructing a high-throughput data network model based on the preprocessed data; the high-throughput data network model is used to reflect the operating status and performance of the optical transmission network during the power production process; Step 4) continuously monitoring the operating status of the optical transmission network and updating the model data in real time; Step 5) Conducting network fault diagnosis for the high-throughput data network model of power production; Step 6) According to the network fault diagnosis results in the power production process, formulate corresponding maintenance strategies for maintenance.
2. The high-throughput data network fault intelligent diagnosis method for power production according to claim 1 is characterized in that: The step 4) also includes: analyzing the constructed model, evaluating the accuracy and reliability of the high-throughput data network model, and optimizing the high-throughput data network model according to the analysis results.
3. The high-throughput data network fault intelligent diagnosis method for power production according to claim 1 is characterized in that In step 5), network fault diagnosis is performed as follows: 5.1) Design a multimodal convolutional neural network model. The neural network model structure includes convolutional layers, pooling layers and fully connected layers; 5.2) Input the high-throughput data network model running data into the multimodal convolutional neural network model to extract local features of the data; 5.3) Using the local features extracted by the multimodal convolutional neural network as the input of the gradient boosting decision tree model; 5.4) Using the gradient boosting algorithm, iteratively train multiple gradient boosting decision tree models, where each decision tree model is trained based on the residual of the previous model; 5.5) Compare the actual fault situation with the model diagnosis results, and continuously optimize and update the model parameters according to the deviation of the diagnosis results; 5.6) Use the trained model to diagnose real-time optical transport network operation data.
4. The high-throughput data network fault intelligent diagnosis method for power production according to claim 3 is characterized in that In the step 5.5), the model parameters are continuously optimized and updated, including adjusting the convolution kernel size, step size, pooling method in the multimodal convolutional neural network, and the learning rate, number of iterations and tree depth in the gradient boosting decision tree.
5. The high-throughput data network fault intelligent diagnosis method for power production according to claim 1 is characterized in that: The step 6) comprises: According to step 5), the diagnosis result of the high-throughput data network system status is output, including whether the system is running normally, whether there is a fault, and the type of fault information; According to the diagnosis results, combined with the preset fault analysis file, the specific location of the fault and the specific maintenance strategy are obtained.
6. An electronic device, characterized in that: include: one or more processors; as well as a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors are caused to execute the method according to any one of claims 1 to 5.
7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.