Substation Disconnector Fault Monitoring System

Abstract

The present invention discloses a fault monitoring system for a transformer substation isolating switch, including a data acquisition module, an environmental interference detection module, a data fusion module, a fault prediction module, and a dynamic threshold adjustment module; the data acquisition module collects the operating parameters of the transformer substation isolating switch through multiple redundant sensors, including current, voltage, temperature, and vibration signals, and transmits the collected data to a data processing unit in real time. The present invention uses advanced noise filtering and wavelet transform technology to accurately filter out interference signals and ensure data accuracy through the collaborative work of the data acquisition module, the environmental interference detection module, and the data fusion module. The Kalman filter is used to fuse multi-sensor data to reduce the risk of misjudgment of faults, and a fault prediction module and a dynamic threshold adjustment module are introduced to predict faults and dynamically adjust thresholds through machine learning algorithms to ensure that the system operates stably in complex environments and reduce the risk of power outages.

Description

Technical Field

[0001] The invention relates to the technical field of transformer substation isolating switch fault monitoring, and in particular to a transformer substation isolating switch fault monitoring system. Background Art

[0002] The substation disconnector fault monitoring system is a system dedicated to monitoring and detecting the operating status of the substation disconnector. The disconnector is a vital device in the power system. Its main function is to cut off the power supply to facilitate equipment maintenance or emergency handling. Since the disconnector directly affects the safe and stable operation of the power system, it is crucial to detect its failure in a timely manner. The monitoring system can collect multiple operating parameters such as current, voltage, temperature, vibration, etc. in real time through sensors installed on the disconnector, and transmit these data to the monitoring platform. The monitoring platform analyzes the data, and once an abnormality is detected, the system will issue an alarm signal to prompt the operator to perform maintenance. In addition, the system can also record long-term operating data to help predict the occurrence of faults and improve maintenance efficiency. Through this system, power companies can significantly reduce the risk of power outages caused by disconnector failures and improve the reliability and safety of the power grid.

[0003] The prior art has the following deficiencies:

[0004] In existing substation disconnector fault monitoring systems, sensor false alarms can lead to serious consequences. Specifically, when the sensors used to monitor the disconnector status are disturbed by the environment (such as electromagnetic interference or temperature changes), erroneous data may be generated, causing the system to misjudge the disconnector failure and trigger unnecessary alarms. Such false alarms not only increase the burden on maintenance personnel, but may also lead to unnecessary power outages, affect the normal operation of the power grid, and reduce the stability and continuity of power supply, thereby causing economic losses to users and causing service interruptions.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The purpose of the present invention is to provide a fault monitoring system for a transformer substation isolating switch, which effectively solves the problem of false alarms caused by electromagnetic interference and temperature fluctuations of sensors through the synergistic effect of a data acquisition module, an environmental interference detection module and a data fusion module. By adopting an advanced noise filtering algorithm and combining spectrum analysis and wavelet transform technology, interference signals can be accurately filtered out to ensure data accuracy. At the same time, a Kalman filter is used to fuse multiple sensor data to reduce fault misjudgments and missed judgments caused by single sensor errors. A fault prediction module and a dynamic threshold adjustment module are introduced to perform trend analysis through a machine learning algorithm, predict faults and dynamically adjust thresholds to ensure that the system operates stably in a complex environment, reduce the risk of power outages, and improve the continuity of power supply to solve the problems in the above-mentioned background technology.

[0007] In order to achieve the above-mentioned object, the present invention provides the following technical solutions: a substation disconnector fault monitoring system, comprising a data acquisition module, an environmental interference detection module, a data fusion module, a fault prediction module and a dynamic threshold adjustment module;

[0008] The data acquisition module collects the operating parameters of the substation disconnector through multiple redundant sensors, including current, voltage, temperature and vibration signals, and transmits the collected data to the data processing unit in real time;

[0009] The environmental interference detection module uses advanced noise filtering algorithms, combined with spectrum analysis and wavelet transform technology, to detect environmental interference and denoise sensor data, identify and filter out abnormal data;

[0010] The data fusion module uses Kalman filter to fuse the output data of multiple sensors based on multi-sensor data fusion technology to generate accurate disconnector status evaluation values ​​to improve the reliability and anti-interference ability of monitoring data;

[0011] The fault prediction module introduces a prediction algorithm based on machine learning to perform trend analysis on data collected over a long period of time, establish a health status model for disconnectors, predict faults and issue early warnings to reduce the risk of misjudgment;

[0012] The dynamic threshold adjustment module uses an adaptive algorithm to dynamically adjust the fault judgment threshold of the disconnector according to the grid operating conditions and environmental changes, ensuring accurate judgment of the operating status of the disconnector in complex environments to avoid false alarms or missed alarms.

[0013] Preferably, an advanced noise filtering algorithm is used, combined with spectrum analysis and wavelet transform technology, to perform environmental interference detection and denoising on sensor data. The specific steps are as follows:

[0014] Preprocess the raw data collected by the sensor, including sampling, quantization and signal denoising;

[0015] After completing data acquisition and preliminary filtering, the signal is further analyzed through spectrum analysis technology;

[0016] The signal is decomposed into multiple scales by wavelet transform technology, effectively separating the noise and signal components existing in different time scales;

[0017] After completing the wavelet decomposition, the noise at different scales is suppressed based on the advanced noise filtering algorithm and the pure signal is reconstructed.

[0018] Preferably, based on multi-sensor data fusion technology, a Kalman filter is used to fuse the output data of multiple sensors, and the specific steps are as follows:

[0019] Acquire data from multiple different types of sensors and establish the initial state of the Kalman filter, including setting the state vector, state transfer matrix, and noise covariance matrix;

[0020] In the prediction stage of the Kalman filter, the state vector at the current moment and the corresponding error covariance matrix are predicted through the state transfer equation according to the dynamic model of the disconnector;

[0021] In the update phase, the actual observation data collected by multiple sensors are compared with the predicted values, and weighted adjustment is performed through the Kalman gain to generate the final state estimate;

[0022] The disconnector status is evaluated based on the fused state estimation results.

[0023] Preferably, in order to further improve system performance, an adaptive optimization mechanism is introduced to dynamically adjust the noise covariance matrix of the Kalman filter;

[0024] By analyzing the long-term data of the sensor, the performance changes of the sensor under different working conditions are identified, and the Kalman filter parameters are adaptively optimized to improve the ability to suppress noise in different environments and the sensitivity to abnormal data.

[0025] Preferably, a prediction algorithm based on machine learning is introduced to perform trend analysis and fault prediction. The specific steps are as follows:

[0026] Long-term collection of multi-dimensional data related to disconnector operation and feature engineering of the data;

[0027] After feature extraction is completed, a machine learning model suitable for fault prediction is selected and trained;

[0028] After the model training is completed and deployed to the system, real-time data will be continuously input into the model through sensors for predictive analysis. The machine learning model will evaluate the health status of the disconnector based on the input real-time data and predict future failures through trend analysis.

[0029] When the model detects a potential failure risk, the system issues an early warning message, prompting maintenance personnel to conduct inspections or repairs.

[0030] Preferably, an adaptive algorithm is used to dynamically adjust the disconnector fault determination threshold, and the specific steps are as follows:

[0031] Real-time monitoring of the operating conditions of the power grid and external environmental factors. At this stage, all relevant environmental and operating condition data will be recorded and input into the system's adaptive algorithm model to ensure that subsequent threshold adjustments can be made based on the latest actual conditions;

[0032] After collecting real-time environmental and operating data, the system analyzes these data through pattern recognition technology to identify the changing patterns of the power grid under different operating conditions. The system uses a combination of statistical analysis and machine learning to analyze the correlation between current environmental conditions and the fault history of the disconnector.

[0033] After completing data analysis and pattern recognition, the fault determination threshold of the disconnector is dynamically adjusted through an adaptive algorithm;

[0034] In the process of adjusting the threshold by the adaptive algorithm, the system establishes a feedback mechanism to continuously optimize the threshold adjustment strategy by evaluating the monitoring effect after each threshold adjustment.

[0035] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0036] The present invention effectively solves the problem of false alarms caused by external factors such as electromagnetic interference and temperature fluctuations in the prior art through the synergy of the data acquisition module, the environmental interference detection module and the data fusion module. It adopts an advanced noise filtering algorithm and combines spectrum analysis and wavelet transform technology to accurately detect and filter out interference signals to ensure the accuracy of sensor data. In addition, the output data of multiple sensors are fused through the Kalman filter to generate a more accurate isolation switch status evaluation value, which greatly reduces the fault misjudgment and missed judgment caused by the error of a single sensor.

[0037] The present invention further improves the early warning capability of fault detection and the ability to adapt to complex environments by introducing a fault prediction module and a dynamic threshold adjustment module. Through a trend analysis algorithm based on machine learning, it is possible to conduct an in-depth analysis of long-term operating data, build a health status model of the disconnector, and predict possible fault states in advance, thereby reducing the risk of power outages caused by sudden faults. At the same time, through an adaptive algorithm, the fault judgment threshold is dynamically adjusted according to the actual operating conditions and environmental changes of the power grid to ensure that the system can still accurately judge the operating status of the disconnector in various complex environments, effectively avoiding the occurrence of false alarms and missed alarms, thereby improving the continuity of power supply and the stability of system operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0039] Figure 1 It is a module schematic diagram of the substation disconnector fault monitoring system of the present invention. DETAILED DESCRIPTION

[0040] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0041] The present invention provides Figure 1 The substation disconnector fault monitoring system shown includes a data acquisition module, an environmental interference detection module, a data fusion module, a fault prediction module and a dynamic threshold adjustment module;

[0042] The data acquisition module collects the operating parameters of the substation disconnector through multiple redundant sensors, including current, voltage, temperature and vibration signals, and transmits the collected data to the data processing unit in real time;

[0043] The environmental interference detection module uses advanced noise filtering algorithms, combined with spectrum analysis and wavelet transform technology, to detect and denoise sensor data, identify and filter out data anomalies such as electromagnetic interference and temperature fluctuations that may cause false alarms;

[0044] The data fusion module uses Kalman filter to fuse the output data of multiple sensors based on multi-sensor data fusion technology to generate accurate disconnector status evaluation values ​​to improve the reliability and anti-interference ability of monitoring data;

[0045] The fault prediction module introduces a prediction algorithm based on machine learning to perform trend analysis on data collected over a long period of time, establish a health status model for the disconnector, and provide early warning by predicting possible fault conditions to reduce the risk of misjudgment.

[0046] The dynamic threshold adjustment module uses an adaptive algorithm to dynamically adjust the fault judgment threshold of the disconnector according to the grid operating conditions and environmental changes, ensuring that the operating status of the disconnector can still be accurately judged in complex environments to avoid false alarms or missed alarms.

[0047] Combining spectrum analysis and wavelet transform technology, the specific steps of environmental interference detection and denoising of sensor data can be divided into the following four main stages:

[0048] Step 1: Data preprocessing and signal sampling

[0049] First, the raw data collected from the sensor must be preprocessed, including sampling, quantization, and signal denoising. The data captured by the sensor is usually accompanied by a variety of interference signals in the environment, such as electromagnetic interference, temperature fluctuations, etc. To ensure the integrity and consistency of the data, the system needs to set an appropriate sampling rate and sampling accuracy so that the key signals of the sensor can be collected without losing details. At this point, a simple filter (such as a low-pass filter) can be used to perform preliminary cleaning of the data to remove obvious high-frequency interference signals, laying the foundation for subsequent spectrum analysis. The goal of this stage is to convert the data from its original complex state into a relatively pure preliminary signal.

[0050] Step 2: Spectrum Analysis

[0051] After completing data collection and preliminary filtering, the next step is to further analyze the signal through spectrum analysis technology. Spectral analysis can help identify the frequency components in sensor data, especially in disconnector monitoring, where some abnormal frequency components (such as high-frequency interference or harmonic signals) are often caused by external interference. By using spectrum analysis methods such as Fourier transform (FFT), the time domain signal is converted into a frequency domain signal, and the distribution of the signal at different frequencies can be clearly seen. Based on this spectrum information, the system can set a bandwidth filtering range to remove interference signals that are significantly beyond the normal operating frequency range, thereby improving the purity and reliability of the signal.

[0052] Step 3: Wavelet transform for multi-scale signal decomposition

[0053] Wavelet transform is a technology suitable for non-stationary signal processing. It can decompose the signal at multiple scales and effectively separate the noise and signal components existing at different time scales. In this step, the sensor data is decomposed into different components to better capture abnormal events and short-term interference. Compared with the traditional Fourier transform, the wavelet transform can process the time domain and frequency domain information of the signal at the same time, so it is suitable for dynamic monitoring scenarios such as disconnectors. By selecting an appropriate wavelet basis (such as Daubechies wavelet or Coiflets wavelet), the system can separate the high-frequency noise and low-frequency trends in the signal, effectively remove the noise, and retain the key signal features for subsequent analysis.

[0054] Step 4: Advanced Noise Filtering and Signal Reconstruction

[0055] After completing the wavelet decomposition, the last step is to suppress the noise at different scales based on the advanced noise filtering algorithm and reconstruct the pure signal. By setting the thresholds for the high-frequency and low-frequency parts of the wavelet components, the system can automatically identify and remove the components containing interference, while retaining the signal data with practical significance. In this step, the system uses adaptive threshold technology to dynamically adjust the denoising parameters according to different environmental conditions (such as temperature, humidity, and electromagnetic field changes) to ensure the accuracy of signal reconstruction. Finally, the decomposed signal is restored to the time domain through the inverse wavelet transform to generate purified data after noise filtering, providing a reliable data basis for fault detection and status assessment.

[0056] Through the combination of these four steps, the system can effectively filter environmental noise, improve the accuracy of sensor data, reduce the false alarm rate, and thus improve the overall performance of the disconnector fault monitoring system. This method not only relies on existing spectrum analysis and wavelet transform technology, but also combines adaptive filtering and dynamic threshold algorithms to ensure that high-quality data processing results can be provided in complex environments.

[0057] Based on multi-sensor data fusion technology, the specific steps of using Kalman filter to fuse the output data of multiple sensors can be divided into the following four stages:

[0058] Step 1: Sensor data collection and initialization

[0059] First, the system acquires data from multiple different types of sensors (such as current sensors, voltage sensors, temperature sensors, etc.). These sensors measure multiple physical quantities related to the state of the disconnector, forming raw data inputs in multiple dimensions. Since the physical quantities measured by each sensor have different units, ranges, and errors, each sensor must be initialized before data fusion, including normalizing the data of each sensor to ensure that the data of different sensors can be processed on the same scale. In addition, it is necessary to establish the initial state of the Kalman filter, including setting the state vector, state transfer matrix, and noise covariance matrix. The key to this step is to lay an accurate initial state for the subsequent Kalman filter and ensure that the data of all sensors are processed within a consistent framework during the data fusion process.

[0060] Step 2: Kalman filter prediction stage

[0061] In the prediction stage of the Kalman filter, according to the dynamic model of the disconnector, the system first predicts the state vector at the current moment and the corresponding error covariance matrix through the state transfer equation. The dynamic model can be constructed based on the physical characteristics of the disconnector and the historical operation data. For example, assuming that the changes in current and voltage are linear and stable when the disconnector is working normally, the theoretical values ​​of each physical quantity at the next moment can be predicted through the state transfer matrix. The prediction stage mainly uses the state estimate of the previous moment to infer the state at the current moment, while taking into account the influence of system noise and measurement noise. Through this prediction, the system can know the state change trend of the disconnector in advance, thus laying the foundation for subsequent data updates.

[0062] Step 3: Kalman filter update phase

[0063] In the update phase, the system compares the actual observation data collected by multiple sensors with the predicted values, and performs weighted adjustments through the Kalman gain to generate the final state estimate. In this step, the Kalman gain plays an important role in balancing the predicted value and the observed value. By calculating the Kalman gain, the system can judge the credibility of the sensor data, and the data with higher weight will have a greater impact on the final state estimate. For sensor data with low reliability or high noise, the Kalman gain will automatically reduce its weight, thereby reducing its impact on the fusion result. Through this dynamic adjustment, the system can fuse heterogeneous data from multiple sensors to generate an accurate and robust disconnector state assessment value.

[0064] Step 4: State evaluation and adaptive optimization

[0065] Finally, the system evaluates the state of the disconnector based on the fused state estimation results. At this point, the fused data has higher anti-interference ability and stability, so the state evaluation can more accurately reflect the actual operating status of the disconnector.

[0066] To further improve system performance, an adaptive optimization mechanism can be introduced to dynamically adjust the noise covariance matrix of the Kalman filter. By analyzing the long-term data of the sensor, the system can identify the performance changes of the sensor under different working conditions and adaptively optimize the Kalman filter parameters, thereby improving the ability to suppress noise in different environments and the sensitivity to abnormal data. This optimization not only improves the accuracy of monitoring data, but also enhances the robustness of the system, enabling it to maintain efficient fault detection capabilities under various complex working conditions.

[0067] Multi-sensor data fusion technology based on Kalman filtering can significantly improve the accuracy and reliability of disconnector status monitoring. Kalman filtering effectively reduces the impact of noise and error of a single sensor by fusing data from multiple sensors, and continuously adjusts the filter parameters through adaptive optimization, so that the system can still operate stably in a complex power grid operating environment.

[0068] The specific steps of introducing prediction algorithms based on machine learning for trend analysis and fault prediction can be divided into the following four stages:

[0069] Step 1: Data collection and feature engineering

[0070] First, the system needs to collect multi-dimensional data related to the operation of the disconnector for a long time. These data may include current, voltage, temperature, vibration and other operating parameters. In order to ensure the effectiveness of the machine learning model, the amount of data collected should cover different operating conditions, including normal operating conditions and abnormal conditions before the failure. After the data collection is completed, the data needs to be processed by feature engineering. The core of feature engineering is to extract key features related to the health status of the disconnector, including statistical characteristics of time series data (such as mean, standard deviation, maximum value, minimum value, etc.), and frequency characteristics obtained based on signal processing technology (such as Fourier transform, wavelet transform). In addition, the anomaly detection algorithm can mark the fault moment in the historical data and provide labeled data for the model. The key to this step is to convert the raw data into useful information that the model can understand and process through data preprocessing and feature extraction.

[0071] Step 2: Model selection and training

[0072] After feature extraction, the next step is to select a machine learning model suitable for fault prediction and train it. Common prediction models include supervised learning and unsupervised learning methods, such as support vector machine (SVM), random forest (RandomForest), long short-term memory network (LSTM), etc. For time series data, deep learning models such as LSTM are particularly suitable because they can capture long-term dependencies, especially the disconnector failure may show a trend of slow accumulation. During the training process, the system will use historical operation data to supervise the model and adjust the model parameters to improve the prediction accuracy. In order to improve the generalization ability of the model, the cross-validation technology is used to evaluate the model to ensure that it can work stably on different data sets. Through model training, the system will obtain a prediction model that can accurately capture fault precursors and operation trends.

[0073] Step 3: Real-time data input and trend analysis

[0074] After the model training is completed and deployed to the system, real-time data will be continuously input into the model through sensors for predictive analysis. The machine learning model evaluates the health status of the disconnector based on the input real-time data and predicts the possibility of future failures through trend analysis. Specifically, the model analyzes the operating data of multiple dimensions in real time and identifies potential failure modes in combination with historical data. For example, time series analysis based on models such as LSTM can capture signs of gradual deterioration in the performance of the disconnector, thereby issuing an early warning before a formal failure occurs. In addition, the system can also use sliding window technology to segment real-time data so that the model can continuously update the prediction results and reflect the health status of the disconnector in a timely manner. The goal of this step is to timely identify the precursors of failures and provide early warnings through the trend analysis function of the model to reduce the risk of misjudgment.

[0075] Step 4: Fault prediction and adaptive optimization

[0076] When the model detects a potential risk of failure, the system will issue an early warning message to prompt maintenance personnel to check or repair it. At the same time, in order to further improve the prediction accuracy of the model, the system will also introduce an adaptive optimization mechanism. This mechanism continuously optimizes the model through online learning methods to ensure that the model can adapt to new data patterns and environmental changes. For example, when the disconnector operates in different seasons and under different load conditions, its operating characteristics may change. The system gradually updates the model parameters through adaptive learning to avoid false positives or false negatives. In addition, after a fault occurs, the system can use the newly collected fault data to retrain the model to further enhance its fault prediction ability. Through this adaptive optimization mechanism, the system can not only accurately predict the faults of the disconnector, but also dynamically adjust the model to adapt to the complex and changing power grid operation environment.

[0077] The fault prediction algorithm based on machine learning can effectively analyze the operating trend of the disconnector and provide accurate warning before the fault occurs. Through long-term data collection, in-depth feature extraction and efficient model training, the system can capture potential fault modes and continuously improve the prediction accuracy through adaptive optimization, making the power grid operation more stable and reliable.

[0078] The specific steps of using the adaptive algorithm to dynamically adjust the disconnector fault determination threshold can be divided into the following four stages:

[0079] Step 1: Real-time monitoring of environmental and operating conditions

[0080] First, the system needs to monitor the operating conditions of the power grid and external environmental factors in real time, such as temperature, humidity, load fluctuations, electromagnetic interference, etc. These environmental conditions have a significant impact on the working state of the disconnector, especially in extreme environments, where sensors may deviate, causing the static setting of the judgment threshold to no longer apply. The system continuously collects external environmental data through a variety of sensors (such as ambient temperature sensors, electromagnetic field sensors, etc.), and combines the operating data of the power grid (such as load changes, power factor, etc.) to provide the necessary basic data support for subsequent adaptive adjustments. At this stage, all relevant environmental and operating condition data will be recorded and input into the system's adaptive algorithm model to ensure that subsequent threshold adjustments can be made based on the latest actual conditions.

[0081] Step 2: Data Analysis and Pattern Recognition

[0082] After collecting real-time environmental and operating data, the system will analyze this data through pattern recognition technology to identify the changing patterns of the power grid under different operating conditions. The system uses a combination of statistical analysis and machine learning to analyze the correlation between current environmental conditions and the fault history of disconnectors. For example, the system will use cluster analysis or association rule algorithms to find patterns that are prone to sensor false alarms or failures under specific environments (such as high temperature and high load). Through this analysis, the system can identify which sensor readings are prone to fluctuations under specific conditions and which thresholds need to be adjusted. The goal of this step is to provide a basis for subsequent adaptive threshold adjustments to ensure that adjustments made in response to environmental changes are targeted and scientifically based.

[0083] Step 3: Adaptive algorithm threshold adjustment

[0084] After completing data analysis and pattern recognition, the system dynamically adjusts the fault judgment threshold of the disconnector through an adaptive algorithm. The adaptive algorithm can use a feedback-based control system, combined with environmental conditions and historical operating data, to adjust the threshold of the sensor to ensure its stability in complex environments. For example, the recursive least squares method (RLS) or adaptive Kalman filter algorithm can be used to dynamically calculate new thresholds based on real-time changing environmental data and sensor errors. The adjustment process of the adaptive algorithm is continuous. The system will continuously correct the threshold determined by the sensor according to different time periods or environmental conditions to ensure that the operating status of the disconnector can be accurately judged under extreme conditions. This process ensures the flexibility and accuracy of threshold adjustment and effectively reduces the risk of false alarms or missed alarms.

[0085] Step 4: Threshold optimization and feedback mechanism

[0086] In the process of adjusting the threshold by the adaptive algorithm, the system will also establish a feedback mechanism to continuously optimize the threshold adjustment strategy by evaluating the monitoring effect after each threshold adjustment. Specifically, the system will record the accuracy of each fault judgment and warning, and provide feedback and correction to the adaptive algorithm based on the actual results. If the threshold adjustment under certain environmental conditions is too frequent or the misjudgment rate increases, the system will automatically optimize the adaptive algorithm to adjust its sensitivity to environmental changes. At the same time, the system can gradually accumulate more environmental data and fault experience through the online learning mechanism, and continuously improve the accuracy and robustness of the adaptive algorithm. This feedback mechanism can not only improve the accuracy of fault detection, but also enable the system to better adapt to the complex and dynamically changing power grid operating environment, ensuring the effectiveness and reliability of the system in long-term use.

[0087] The fault judgment threshold adjustment based on the adaptive algorithm can realize the accurate monitoring and dynamic adjustment of the disconnector status. This method not only takes into account the impact of complex environmental conditions on sensor data, but also continuously optimizes the judgment criteria of the system through adaptive adjustment and feedback mechanism, avoiding false alarms or missed alarms, and improving the stability and operation efficiency of the power grid.

[0088] The present invention effectively solves the problem of false alarms caused by external factors such as electromagnetic interference and temperature fluctuations in the prior art through the synergy of the data acquisition module, the environmental interference detection module and the data fusion module. It adopts an advanced noise filtering algorithm and combines spectrum analysis and wavelet transform technology to accurately detect and filter out interference signals to ensure the accuracy of sensor data. In addition, the output data of multiple sensors are fused through the Kalman filter to generate a more accurate isolation switch status evaluation value, which greatly reduces the fault misjudgment and missed judgment caused by the error of a single sensor.

[0089] The present invention further improves the early warning capability of fault detection and the ability to adapt to complex environments by introducing a fault prediction module and a dynamic threshold adjustment module. Through a trend analysis algorithm based on machine learning, it is possible to conduct an in-depth analysis of long-term operating data, build a health status model of the disconnector, and predict possible fault states in advance, thereby reducing the risk of power outages caused by sudden faults. At the same time, through an adaptive algorithm, the fault judgment threshold is dynamically adjusted according to the actual operating conditions and environmental changes of the power grid to ensure that the system can still accurately judge the operating status of the disconnector in various complex environments, effectively avoiding the occurrence of false alarms and missed alarms, thereby improving the continuity of power supply and the stability of system operation.

[0090] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0091] It should be noted that, in this article, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or still includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including a..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0092] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0093] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0094] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0095] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0096] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0097] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0098] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0099] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. The substation disconnector fault monitoring system is characterized by: It includes data acquisition module, environmental interference monitoring module, data fusion module, fault prediction module and dynamic threshold adjustment module; The data acquisition module collects the operating parameters of the substation disconnector through multiple redundant sensors, including current, voltage, temperature and vibration signals, and transmits the collected data to the data processing unit in real time; The environmental interference detection module uses advanced noise filtering algorithms, combined with spectrum analysis and wavelet transform technology, to detect environmental interference and denoise sensor data, identify and filter out abnormal data; The data fusion module uses Kalman filter to fuse the output data of multiple sensors based on multi-sensor data fusion technology to generate accurate disconnector status evaluation values ​​to improve the reliability and anti-interference ability of monitoring data; The fault prediction module introduces a prediction algorithm based on machine learning to perform trend analysis on data collected over a long period of time, establish a health status model for disconnectors, predict faults and issue early warnings to reduce the risk of misjudgment; The dynamic threshold adjustment module uses an adaptive algorithm to dynamically adjust the fault judgment threshold of the disconnector according to the grid operating conditions and environmental changes, ensuring accurate judgment of the operating status of the disconnector in complex environments to avoid false alarms or missed alarms; Advanced noise filtering algorithms are used, combined with spectrum analysis and wavelet transform technology, to detect environmental interference and denoise sensor data. The specific steps are as follows: Preprocess the raw data collected by the sensor, including sampling, quantization and signal denoising; The data captured by the sensor are accompanied by a variety of interference signals in the environment, including electromagnetic interference and temperature fluctuations; After completing data acquisition and preliminary filtering, the signal is further analyzed through spectrum analysis technology; The signal is decomposed into multiple scales by wavelet transform technology, effectively separating the noise and signal components existing in different time scales; After completing the wavelet decomposition, the noise at different scales is suppressed based on the advanced noise filtering algorithm, and the pure signal is reconstructed; Based on multi-sensor data fusion technology, the Kalman filter is used to fuse the output data of multiple sensors. The specific steps are as follows: Acquire data from multiple different types of sensors and establish the initial state of the Kalman filter, including setting the state vector, state transfer matrix, and noise covariance matrix; In the prediction stage of the Kalman filter, the state vector at the current moment and the corresponding error covariance matrix are predicted through the state transfer equation according to the dynamic model of the disconnector; In the update phase, the actual observation data collected by multiple sensors are compared with the predicted values, and weighted adjustment is performed through the Kalman gain to generate the final state estimate; Evaluate the disconnector status based on the fused state estimation results; To further improve system performance, an adaptive optimization mechanism is introduced to dynamically adjust the noise covariance matrix of the Kalman filter; By analyzing the long-term data of the sensor, the performance changes of the sensor under different working conditions are identified, and the Kalman filter parameters are adaptively optimized to improve the ability to suppress noise in different environments and the sensitivity to abnormal data; Through the synergy of the data acquisition module, environmental interference detection module and data fusion module, the false alarms caused by electromagnetic interference and temperature fluctuations of the sensor are solved; by adopting advanced noise filtering algorithms and combining spectrum analysis and wavelet transform technology, interference signals are filtered out to ensure data accuracy; at the same time, the Kalman filter is used to fuse multiple sensor data to reduce fault misjudgments and missed judgments caused by single sensor errors.

2. The substation disconnector fault monitoring system according to claim 1, characterized in that: Introduce a prediction algorithm based on machine learning to perform trend analysis and fault prediction. The specific steps are as follows: Long-term collection of multi-dimensional data related to disconnector operation and feature engineering of the data; After feature extraction is completed, a machine learning model suitable for fault prediction is selected and trained; After the model training is completed and deployed to the system, real-time data will be continuously input into the model through sensors for predictive analysis. The machine learning model will evaluate the health status of the disconnector based on the input real-time data and predict future failures through trend analysis. When the model detects a potential failure risk, the system issues an early warning message, prompting maintenance personnel to conduct inspections or repairs.

3. The substation disconnector fault monitoring system according to claim 1, characterized in that: Use the adaptive algorithm to dynamically adjust the disconnector fault determination threshold. The specific steps are as follows: Real-time monitoring of the operating conditions of the power grid and external environmental factors. At this stage, all relevant environmental and operating condition data will be recorded and input into the system's adaptive algorithm model to ensure that subsequent threshold adjustments can be made based on the latest actual conditions; After collecting real-time environmental and operating data, the system analyzes these data through pattern recognition technology to identify the changing patterns of the power grid under different operating conditions. The system uses a combination of statistical analysis and machine learning to analyze the correlation between current environmental conditions and the fault history of the disconnector. After completing data analysis and pattern recognition, the fault determination threshold of the disconnector is dynamically adjusted through an adaptive algorithm; In the process of adjusting the threshold by the adaptive algorithm, the system establishes a feedback mechanism to continuously optimize the threshold adjustment strategy by evaluating the monitoring effect after each threshold adjustment.

Images