Power failure intelligent prediction platform based on power data

By applying wavelet transformation, deep learning and distributed database technologies in the power failure intelligent prediction platform, combined with support vector machine and random forest algorithm, the problems of low prediction accuracy and slow response speed in the existing power failure prediction technology are solved, and the problem of low prediction accuracy and slow response speed is achieved. Efficient and intelligent prediction and early warning of power failures ensure the stable operation of the power system.

CN119026786AInactive Publication Date: 2024-11-26SHANDONG ENERGY SHENGLUNENG CHEM ORDOS NEW ENERGY CO LTD +1

Patent Information

Application Number
CN202410864165.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-30
Publication Date
2024-11-26
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing power fault prediction technology is difficult to process massive and multi-dimensional power data, has low prediction accuracy, lacks self-learning and optimization capabilities, and is unable to adapt to the dynamic changes of the power system, resulting in untimely and inaccurate fault prediction.

Method used

A smart prediction platform for power failure based on power data is designed, using data acquisition, preprocessing, storage, analysis, model training and early warning modules, combining wavelet transformation, deep learning, distributed database, support vector machine and random forest algorithm. Achieve intelligent prediction and early warning of power failures.

Benefits of technology

It significantly improves the accuracy and timeliness of prediction of power failures, reduces the frequency and impact range of faults, and ensures the safe and stable operation of the power system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119026786A_ABST
    Figure CN119026786A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of power monitoring and prediction, in particular to a power fault intelligent prediction platform based on power data, which comprises a data acquisition module, a data preprocessing module, a data storage module, a data analysis module, a model training module, a fault prediction module and an early warning module. Wherein the data acquisition module is used for acquiring power data from a power system; the data preprocessing module is used for processing the collected power data; the data storage module is used for storing the preprocessed power data; the data analysis module is used for identifying potential power failure risks; the model training module is used for training a power failure prediction model; and the fault prediction module is used for performing fault prediction on the real-time power data. According to the invention, through multi-sensor data acquisition, advanced algorithm analysis and distributed database storage, accurate prediction and timely early warning of power faults are realized, and the stability and reliability of a power system are significantly improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of power monitoring and prediction, and in particular to a power failure intelligent prediction platform based on power data. Background Art

[0002] As the scale of power systems continues to expand and the complexity increases, the prediction and prevention of power failures have become increasingly important. The power system is a complex network structure consisting of multiple links such as power generation, transmission, substation and distribution. Each link may fail due to various factors, such as equipment aging, external environmental interference, operational errors, etc. Traditional power failure prediction methods mainly rely on manual experience and simple statistical analysis. These methods often make it difficult to identify fault precursors in a timely and accurate manner, resulting in difficulty in taking effective measures in a timely manner when a fault occurs, which seriously affects the stable operation of the power system and the reliability of power supply. Therefore, there is an urgent need for an efficient and intelligent power failure prediction platform to achieve early warning and rapid response to power failures.

[0003] There are many shortcomings in the existing power fault prediction technology. First, traditional methods are difficult to process massive and multi-dimensional power data and cannot fully explore the potential information in the data, resulting in low prediction accuracy. Secondly, most of the existing systems are independently operated modules, lacking effective data fusion and comprehensive analysis capabilities, and it is difficult to provide comprehensive and accurate fault prediction. In addition, the existing fault prediction models lack self-learning and optimization capabilities and cannot adapt to the dynamic changes of the power system. Based on the above problems, the present invention proposes an intelligent power fault prediction platform based on power data, which aims to realize intelligent prediction of power faults through the comprehensive application of big data analysis, machine learning and distributed computing technology, improve the accuracy and timeliness of prediction, thereby effectively reducing the frequency and impact range of power faults, and ensuring the safe and stable operation of the power system. Summary of the invention

[0004] Based on the above objectives, the present invention provides an intelligent power failure prediction platform based on power data.

[0005] An intelligent power fault prediction platform based on power data includes a data acquisition module, a data preprocessing module, a data storage module, a data analysis module, a model training module, a fault prediction module and an early warning module; wherein:

[0006] Data acquisition module: used to collect power data from the power system;

[0007] Data preprocessing module: connected to the data acquisition module, used for cleaning and normalizing the collected power data;

[0008] Data storage module: connected to the data preprocessing module, used to store preprocessed power data;

[0009] Data analysis module: connected to the data storage module, used to analyze the stored power data and identify potential power failure risks;

[0010] Model training module: connected to the data analysis module, used to train the power failure prediction model based on historical power failure data;

[0011] Fault prediction module: connected to the model training module, used to apply the trained power fault prediction model to perform fault prediction on real-time power data and output prediction results;

[0012] Early warning module: connected to the fault prediction module, used to generate fault early warning information according to the prediction results and notify relevant personnel;

[0013] The data acquisition module collects data through sensors installed at various nodes of the power system, the data preprocessing module processes data using wavelet transform, the data storage module uses a distributed database to store data, the data analysis module uses a deep learning algorithm to analyze data, the model training module uses supervised learning to train models, the fault prediction module uses support vector machines and random forest algorithms to predict faults, and the early warning module uses a multi-level early warning mechanism to ensure the timeliness and accuracy of early warning information.

[0014] Furthermore, the data acquisition module includes voltage sensors and current sensors installed on transformers and transmission lines. By monitoring the changes in voltage and current, the operating status data of the power system is collected in real time to provide basic data for subsequent data analysis. Specifically, the voltage sensor and current sensor convert the voltage and current signals into digital signals that can be processed and send them to the data preprocessing module.

[0015] Furthermore, the data preprocessing module includes a data denoising unit for denoising the collected power data. Specifically, the data denoising unit adopts a wavelet transform algorithm and utilizes the multi-resolution characteristics of the wavelet transform to decompose and reconstruct the signal at different scales, thereby effectively removing noise in the data and improving the quality of the data.

[0016] Furthermore, the data storage module adopts distributed database technology to improve the reliability and access speed of data storage. Specifically, the distributed database realizes parallel storage and access of data by distributing data on multiple storage nodes.

[0017] Furthermore, the data analysis module includes a pattern recognition unit and an anomaly detection unit. Specifically, the pattern recognition unit uses a deep learning algorithm to perform feature extraction and pattern recognition on power data to identify the normal operation mode and fault mode of the power system; the anomaly detection unit detects potential abnormal situations by comparing real-time data and historical data based on the identified pattern.

[0018] Furthermore, the model training module includes a historical data storage unit and a model optimization unit, which are used to train and optimize the power fault prediction model. Specifically, the historical data storage unit stores a large amount of historical power fault data, and the model optimization unit adopts a combination of supervised learning and unsupervised learning to optimize the performance of the power fault prediction model by continuously adjusting model parameters.

[0019] Furthermore, the fault prediction module includes a real-time data acquisition unit and a prediction result output unit. Specifically, the real-time data acquisition unit acquires real-time power data from the data storage module, and the prediction result output unit applies a trained power fault prediction model to perform fault prediction on the real-time power data and outputs the prediction result.

[0020] Furthermore, the fault prediction module uses a support vector machine and a random forest algorithm to perform fault prediction. Specifically, the support vector machine classifies the data by constructing a hyperplane to perform fault prediction in a manner that maximizes the classification boundary. The specific formula is: Among them, x is the input feature vector, x i is the support vector, α i is the Lagrange multiplier, y i is the label of the support vector, K(x i ,x) is the kernel function, b is the bias term. Through this formula, the power data can be classified to identify the fault risk. The random forest algorithm constructs multiple decision trees, votes on the results of each tree, and integrates the prediction results of each tree to improve the accuracy and robustness of the prediction. The formula of the random forest algorithm is: in, is the final prediction result of the random forest, N is the number of decision trees, and f i (x) is the prediction result of the i-th decision tree.

[0021] Furthermore, the early warning module includes an early warning information generation unit and an early warning information sending unit, which send early warning information via text messages, emails and alarms. Specifically, the early warning information generation unit generates corresponding early warning information based on the prediction results of the fault prediction module, and the early warning information sending unit sends the early warning information to relevant personnel in a variety of ways to ensure that response measures are taken in a timely manner.

[0022] Furthermore, the data acquisition module also includes a temperature sensor and a power sensor for collecting temperature and power data. Specifically, the temperature sensor monitors the temperature changes of the power equipment in real time, and the power sensor monitors the power flow of the power system, providing more dimensional data support for power fault prediction.

[0023] Beneficial effects of the present invention:

[0024] The present invention realizes accurate prediction and early warning of power failures by comprehensively applying various advanced technologies such as data collection, preprocessing, analysis, storage and prediction, and significantly improves the reliability and stability of the power system. First, the present invention collects data through various sensors installed at various nodes of the power system, and uses wavelet transform and principal component analysis algorithms to preprocess and extract features of the data, thereby ensuring the high quality and accuracy of the data. Secondly, the preprocessed data is analyzed by using a deep learning algorithm, and fault prediction is performed in combination with support vector machine and random forest algorithm, thereby effectively improving the accuracy and robustness of fault prediction.

[0025] The present invention adopts distributed database technology for data storage, which improves the reliability and access speed of data storage, continuously optimizes the power fault prediction model through the model training module, ensures that the model can adapt to the dynamic changes of the power system, and the early warning module timely generates and sends fault warning information through a multi-level early warning mechanism, ensuring that relevant personnel can quickly take countermeasures. The overall platform adopts a modular design, the modules are closely related, the data flow is smooth, and it has high scalability and adaptability. In summary, the present invention effectively solves the problems of low prediction accuracy, slow response speed and poor system stability in the prior art, and provides strong technical support for the operation and maintenance of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0027] Figure 1 Schematic diagram of a smart power failure prediction platform according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0029] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0030] like Figure 1 As shown, a power fault intelligent prediction platform based on power data includes a data acquisition module, a data preprocessing module, a data storage module, a data analysis module, a model training module, a fault prediction module and an early warning module; wherein:

[0031] Data acquisition module: used to collect power data from the power system;

[0032] Data preprocessing module: connected to the data acquisition module, used to clean and normalize the collected power data;

[0033] Data storage module: connected to the data preprocessing module, used to store preprocessed power data;

[0034] Data analysis module: connected to the data storage module, used to analyze the stored power data and identify potential power failure risks;

[0035] Model training module: connected with the data analysis module, used to train the power failure prediction model based on historical power failure data;

[0036] Fault prediction module: connected to the model training module, used to apply the trained power fault prediction model to predict faults on real-time power data and output the prediction results;

[0037] Early warning module: connected with the fault prediction module, used to generate fault early warning information according to the prediction results and notify relevant personnel;

[0038] The data acquisition module collects data through sensors installed at various nodes of the power system. The data preprocessing module uses wavelet transform to process data. The data storage module uses a distributed database to store data. The data analysis module uses a deep learning algorithm to analyze data. The model training module uses supervised learning to train models. The fault prediction module uses support vector machines and random forest algorithms to predict faults. The early warning module uses a multi-level early warning mechanism to ensure the timeliness and accuracy of early warning information.

[0039] The data acquisition module includes voltage sensors and current sensors installed on transformers and transmission lines. By monitoring the changes in voltage and current, it collects the operating status data of the power system in real time and provides basic data for subsequent data analysis. Specifically, the voltage sensor and current sensor convert the voltage and current signals into digital signals that can be processed and send them to the data preprocessing module.

[0040] The data preprocessing module includes a data denoising unit, which is used to denoise the collected power data. Specifically, the data denoising unit adopts a wavelet transform algorithm and utilizes the multi-resolution characteristics of the wavelet transform to decompose and reconstruct the signal at different scales, thereby effectively removing noise from the data and improving the quality of the data.

[0041] The data storage module adopts distributed database technology to improve the reliability and access speed of data storage. Specifically, the distributed database realizes parallel storage and access of data by distributing data on multiple storage nodes, thereby improving the system's fault tolerance and data access efficiency.

[0042] The data analysis module includes a pattern recognition unit and an anomaly detection unit. Specifically, the pattern recognition unit uses a deep learning algorithm to perform feature extraction and pattern recognition on power data to identify the normal operation mode and fault mode of the power system; the anomaly detection unit detects potential abnormal situations by comparing real-time data and historical data based on the identified patterns.

[0043] The model training module includes a historical data storage unit and a model optimization unit, which are used to train and optimize the power fault prediction model. Specifically, the historical data storage unit stores a large amount of historical power fault data, and the model optimization unit adopts a combination of supervised learning and unsupervised learning to optimize the performance of the power fault prediction model by continuously adjusting the model parameters.

[0044] The fault prediction module includes a real-time data acquisition unit and a prediction result output unit. Specifically, the real-time data acquisition unit acquires real-time power data from the data storage module, and the prediction result output unit applies the trained power fault prediction model to perform fault prediction on the real-time power data and outputs the prediction result.

[0045] The fault prediction module uses support vector machine (SVM) and random forest algorithm for fault prediction. Specifically, the support vector machine classifies the data by constructing a hyperplane to maximize the classification boundary for fault prediction. The specific formula is: Among them, x is the input feature vector, x i is the support vector, α i is the Lagrange multiplier, y i is the label of the support vector, K(x i ,x) is the kernel function, b is the bias term. Through this formula, the power data can be classified to identify the fault risk. The random forest algorithm builds multiple decision trees and votes on the results of each tree to improve the accuracy and robustness of the prediction by combining the prediction results of each tree. The formula of the random forest algorithm is: in, is the final prediction result of the random forest, N is the number of decision trees, and f i (x) is the prediction result of the i-th decision tree. Through the above combination, the fault prediction module can efficiently and accurately predict the faults of the power system.

[0046] The early warning module includes an early warning information generation unit and an early warning information sending unit, which send early warning information via SMS, email and alarm. Specifically, the early warning information generation unit generates corresponding early warning information according to the prediction results of the fault prediction module, and the early warning information sending unit sends the early warning information to relevant personnel in a variety of ways to ensure that timely response measures are taken.

[0047] The data acquisition module also includes temperature sensors and power sensors for collecting temperature and power data. Specifically, the temperature sensor monitors the temperature changes of power equipment in real time, and the power sensor monitors the power flow of the power system, providing more dimensional data support for power fault prediction.

[0048] The present invention is intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A power failure intelligent prediction platform based on power data, characterized in that: It includes data acquisition module, data preprocessing module, data storage module, data analysis module, model training module, fault prediction module and early warning module; among which: Data acquisition module: used to collect power data from the power system; Data preprocessing module: connected to the data acquisition module, used for cleaning and normalizing the collected power data; Data storage module: connected to the data preprocessing module, used to store preprocessed power data; Data analysis module: connected to the data storage module, used to analyze the stored power data and identify potential power failure risks; Model training module: connected to the data analysis module, used to train the power failure prediction model based on historical power failure data; Fault prediction module: connected to the model training module, used to apply the trained power fault prediction model to perform fault prediction on real-time power data and output prediction results; Early warning module: connected to the fault prediction module, used to generate fault early warning information according to the prediction results and notify relevant personnel; The data acquisition module collects data through sensors installed at various nodes of the power system, the data preprocessing module processes data using wavelet transform, the data storage module uses a distributed database to store data, the data analysis module uses a deep learning algorithm to analyze data, the model training module uses supervised learning to train models, the fault prediction module uses support vector machines and random forest algorithms to predict faults, and the early warning module uses a multi-level early warning mechanism to ensure the timeliness and accuracy of early warning information.

2. The power failure intelligent prediction platform based on power data according to claim 1 is characterized in that: The data acquisition module includes voltage sensors and current sensors installed on transformers and transmission lines. By monitoring the changes in voltage and current, the data on the operating status of the power system is collected in real time to provide basic data for subsequent data analysis. Specifically, the voltage sensor and current sensor convert the voltage and current signals into digital signals that can be processed and send them to the data preprocessing module.

3. The power failure intelligent prediction platform based on power data according to claim 1 is characterized in that: The data preprocessing module includes a data denoising unit, which is used to denoise the collected power data. Specifically, the data denoising unit adopts a wavelet transform algorithm and utilizes the multi-resolution characteristics of the wavelet transform to decompose and reconstruct the signal at different scales, thereby effectively removing noise in the data and improving the quality of the data.

4. The power failure intelligent prediction platform based on power data according to claim 1 is characterized in that: The data storage module adopts distributed database technology to improve the reliability and access speed of data storage. Specifically, the distributed database realizes parallel storage and access of data by distributing data on multiple storage nodes.

5. The power failure intelligent prediction platform based on power data according to claim 1 is characterized in that: The data analysis module includes a pattern recognition unit and an anomaly detection unit. Specifically, the pattern recognition unit uses a deep learning algorithm to perform feature extraction and pattern recognition on power data to identify the normal operating mode and fault mode of the power system; the anomaly detection unit detects potential abnormal situations based on the identified pattern by comparing real-time data and historical data.

6. The power failure intelligent prediction platform based on power data according to claim 1 is characterized in that: The model training module includes a historical data storage unit and a model optimization unit, which are used to train and optimize the power failure prediction model. Specifically, the historical data storage unit stores a large amount of historical power failure data, and the model optimization unit adopts a combination of supervised learning and unsupervised learning to optimize the performance of the power failure prediction model by continuously adjusting model parameters.

7. The power failure intelligent prediction platform based on power data according to claim 1 is characterized in that: The fault prediction module includes a real-time data acquisition unit and a prediction result output unit. Specifically, the real-time data acquisition unit acquires real-time power data from the data storage module, and the prediction result output unit uses a trained power fault prediction model to perform fault prediction on the real-time power data and outputs the prediction result.

8. The power failure intelligent prediction platform based on power data according to claim 7 is characterized in that: The fault prediction module uses a support vector machine and a random forest algorithm to perform fault prediction. Specifically, the support vector machine classifies the data by constructing a hyperplane to perform fault prediction in a manner that maximizes the classification boundary. The specific formula is: Among them, x is the input feature vector, x i is the support vector, α i is the Lagrange multiplier, y i is the label of the support vector, K(x i , x) is the kernel function, b is the bias term. Through this formula, the power data can be classified to identify the fault risk. The random forest algorithm constructs multiple decision trees, votes on the results of each tree, and integrates the prediction results of each tree to improve the accuracy and robustness of the prediction. The formula of the random forest algorithm is: in, is the final prediction result of the random forest, N is the number of decision trees, and f i (x) is the prediction result of the i-th decision tree.

9. The power failure intelligent prediction platform based on power data according to claim 1 is characterized in that: The early warning module includes an early warning information generation unit and an early warning information sending unit, which send early warning information via text messages, emails and alarms. Specifically, the early warning information generation unit generates corresponding early warning information based on the prediction results of the fault prediction module, and the early warning information sending unit sends the early warning information to relevant personnel in a variety of ways to ensure that response measures are taken in a timely manner.

10. The power failure intelligent prediction platform based on power data according to claim 1 is characterized in that: The data acquisition module also includes a temperature sensor and a power sensor for collecting temperature and power data. Specifically, the temperature sensor monitors the temperature changes of the power equipment in real time, and the power sensor monitors the power flow of the power system, providing more dimensional data support for power fault prediction.

Citation Information

Patent Citations

  • Charging pile fault identification method and device based on self-coding and random forest

    CN117633662A

  • Power transformer fault early warning method

    CN117741288A

  • Power transmission line fault detection method and system based on artificial intelligence

    CN117972582A

  • Distribution network 10KV overhead line fault early warning and positioning system based on AI

    CN118050591A

  • Intelligent operation and maintenance cloud platform for industrial equipment

    CN118154174A

Cited By

  • Intelligent detection system of power transmission on-line monitoring device

    CN116629627A

  • New energy power transmission state monitoring method and system based on artificial intelligence algorithm

    CN119944951A