Isolation switch intelligent monitoring system

Abstract

The invention relates to the technical field of disconnecting switch monitoring, in particular to a disconnecting switch intelligent monitoring system which comprises a data acquisition unit, a data transceiving unit, a data collection unit, a data transmission unit and a data processing unit. The data acquisition unit is used for acquiring multi-source sensing data reflecting the operating state of the isolating switch; the data transceiving unit is connected with the data acquisition unit and is used for receiving the multi-source sensing data and carrying out primary processing on the multi-source sensing data; the data collection unit is connected with the data receiving and transmitting unit and is used for receiving and collecting the data from the data receiving and transmitting unit and storing the data; the data transmission unit is connected with the data collection unit and is used for remotely transmitting the collected data to a rear end; and the data processing unit is connected with the data transmission unit and is used for analyzing the received data so as to evaluate the operation state of the disconnecting switch and carry out fault prediction. The system can timely and accurately find the abnormity, and effectively improves the reliability and safety of a railway power supply system.

Description

Technical Field

[0001] This invention relates to the field of disconnector switch monitoring technology, and more specifically to an intelligent disconnector switch monitoring system. Background Technology

[0002] In railway power supply systems, contact wire disconnect switches are indispensable key equipment, undertaking core responsibilities such as distributing power and isolating faulty areas. They play a vital role in ensuring the stable operation of railway power supply systems and the safe operation of trains. Their normal operation is directly related to the reliability and safety of the entire railway power supply network. Once a disconnect switch fails, it may not only cause power outages and affect the normal operation of trains, but may also lead to serious safety accidents and cause incalculable losses.

[0003] As a key component of the overhead contact line disconnector, the performance of the contactors directly affects the working efficiency and safety of the switch. Contact temperature rise is mainly caused by poor contact, overload current, or poor heat dissipation. During long-term operation, the contact surface may experience poor contact due to oxidation, wear, and other factors, leading to increased contact resistance. When current flows through, more heat is generated, further increasing the contact temperature. When an overload current occurs in the circuit, the increased current also causes a sharp increase in the heat generated by the contacts. High temperatures may damage the contact material, leading to changes in its performance, accelerated wear, and further increased contact resistance, thereby reducing the reliability of the switch.

[0004] In addition, disconnect switches operate outdoors for extended periods in complex and variable environments, exposed to various factors such as wind, sun, rain, and dust. This can cause transmission components to malfunction, which may result in inaccurate opening and closing operations due to rust, wear, or jamming. Consequently, these malfunctions can affect the distribution of electrical energy and the isolation function of faulty areas, thereby reducing the reliability and safety of the power supply system.

[0005] Meanwhile, in railway power supply systems, the current and voltage parameters of disconnecting switches are important indicators reflecting their operating status. The magnitude of the current directly affects the load condition of the disconnecting switch. Overload current may lead to problems such as contact temperature rise and equipment damage. Voltage stability also affects the normal operation of the disconnecting switch. Excessively high or low voltage may damage the insulation performance and contact performance of the disconnecting switch. Therefore, real-time monitoring of the current and voltage of the disconnecting switch can promptly detect abnormal changes in current and voltage, so as to take corresponding measures for adjustment and protection, and ensure the safe and stable operation of the disconnecting switch.

[0006] In conclusion, developing an intelligent monitoring system for disconnecting switches to monitor their operational status in real time and provide early warnings of faults is of great significance for improving the reliability and safety of railway power supply systems. Summary of the Invention

[0007] To address the aforementioned issues, this invention provides an intelligent monitoring system for disconnecting switches. This system, through multi-source data acquisition and efficient transmission and processing, can promptly and accurately detect anomalies, effectively improving the reliability and safety of railway power supply systems.

[0008] The technical solution of the present invention is as follows:

[0009] The intelligent monitoring system for disconnecting switches includes a data acquisition unit, a data transceiver unit, a data aggregation unit, a data transmission unit, and a data processing unit.

[0010] The data acquisition unit is used to acquire multi-source sensor data reflecting the operating status of the disconnector switch. The multi-source sensor data includes at least angle data representing the mechanical position of the disconnector switch, temperature data representing the contact state of the contacts, and current and voltage data representing the electrical load.

[0011] The data transceiver unit, connected to the data acquisition unit, is used to receive multi-source sensor data and perform preliminary processing.

[0012] The data aggregation unit, connected to the data transceiver unit, is used to receive and aggregate data from the data transceiver unit and store it.

[0013] The data transmission unit, connected to the data aggregation unit, is used to remotely transmit the aggregated data to the backend.

[0014] The data processing unit, connected to the data transmission unit, is used to analyze the received data to assess the operating status of the disconnecting switch and predict faults, and to generate alarm information when an anomaly is detected.

[0015] Furthermore, the data acquisition unit includes:

[0016] Angle sensors are used to monitor angle changes during the opening and closing of disconnecting switches;

[0017] Temperature rise sensor is used to monitor temperature changes in the contacts of disconnecting switches;

[0018] A current sensor is installed in the current loop of the disconnecting switch to measure the current value in real time.

[0019] A voltage sensor is installed in the voltage circuit of the disconnecting switch to measure the voltage value in real time.

[0020] Furthermore, the data processing unit is configured as follows:

[0021] Calculate the deviation between the current angle and the standard angle, and issue an emergency alarm signal when the deviation exceeds the first-level set threshold and an early warning signal when it exceeds the second-level set threshold;

[0022] Calculate the contact temperature change rate and issue an emergency alarm signal when the temperature change rate exceeds the first-level set threshold and an early warning signal when it exceeds the second-level set threshold.

[0023] Furthermore, the temperature rise sensor is deployed in the current-carrying transition zone of the stationary contact, which is determined through finite element temperature field simulation.

[0024] Furthermore, the data transceiver unit is a passive short-range transceiver device, which is deployed on the sensor body, cable, or switch jumper adapter board.

[0025] The intelligent monitoring method for disconnecting switches includes the following steps:

[0026] S1: Multi-source data acquisition, which collects the angle, contact temperature, current and voltage data of the disconnector switch through angle sensor, temperature rise sensor, current sensor and voltage sensor respectively;

[0027] S2: Data preprocessing and transmission: Data from each sensor is received through the data transceiver unit, and after verification, error correction and format conversion, it is transmitted to the data aggregation unit.

[0028] S3: Data collection and storage. Data is received through the data collection unit, and timestamps are aligned, data is cleaned and integrated, and then stored in the local database.

[0029] S4: Data transmission: The data processed by the data collection unit is sent to the data processing unit through the data transmission unit.

[0030] S5: Operation status analysis and fault prediction. The data processing unit analyzes and processes the received data to assess the operation status of the disconnector switch and predict faults, and issues an alarm when an anomaly occurs.

[0031] Furthermore, S5 includes the following steps:

[0032] S51: Data preprocessing, verifying and correcting the collected raw angle data, and using wavelet transform for noise reduction to extract low-frequency trend components for analysis;

[0033] S52: Pattern construction, based on a large amount of angle change data collected and labeled in the actual operating environment of disconnecting switches under normal and fault conditions, a hidden Markov model is trained to obtain state patterns representing different operating states.

[0034] S53: Real-time pattern matching. After preprocessing the angle data collected in real time, the obtained sequence is matched with the trained state pattern to calculate the matching degree. The operating status of the disconnector is determined based on the matching degree result.

[0035] Furthermore, S51 includes the following steps:

[0036] Data integrity verification: Traverse the original angle data and check whether there are any missing data at each sampling time; if there are missing data, use linear interpolation to fill the gaps based on data from adjacent time points to ensure the continuity of the time series;

[0037] Outlier removal and correction: Outliers are detected statistically based on the Grubbs criterion. The mean and standard deviation of the data are calculated, data points that deviate from the mean by more than a critical value are removed, and the neighborhood mean is used as a replacement.

[0038] Time-frequency domain noise suppression: Wavelet transform is used to decompose the angle signal into multiple scales. The Daubechies wavelet is selected to extract the low-frequency trend component. High-frequency noise is removed by thresholding, while retaining the characteristic signal that reflects the mechanical motion state.

[0039] Furthermore, S52 includes the following steps:

[0040] Hidden state definition: Define multiple hidden states that represent the health status of the device;

[0041] Initial state probability allocation: Based on prior knowledge or historical data, set the initial state probability distribution;

[0042] State transition probability configuration: Construct the state transition probability matrix;

[0043] Observation probability matrix generation: Based on the preprocessed angle data features, the probability distribution of observations corresponding to each hidden state is generated through statistical modeling or parameter estimation methods.

[0044] Furthermore, S53 includes the following steps:

[0045] Real-time data acquisition and preprocessing: synchronously acquire angle data and perform verification, interpolation and wavelet denoising to generate feature sequences of the same dimension as the training data;

[0046] Optimal state path decoding: Using the trained HMM model, the matching probability between the real-time feature sequence and each state pattern is calculated by the Viterbi algorithm, and the optimal state transition path is generated by decoding.

[0047] Multi-level fault threshold determination: Calculate the similarity between the optimal path and the normal mode, and trigger alarms in stages based on preset thresholds.

[0048] The beneficial effects of this invention are as follows:

[0049] 1. The intelligent monitoring system for disconnecting switches disclosed in this invention comprehensively collects multi-source sensor data reflecting the operating status of the disconnecting switch through a data acquisition unit. This data includes angle data characterizing the mechanical position of the disconnector, temperature data representing the contact state of the contacts, and current and voltage data of the electrical load. This multi-source data acquisition method can comprehensively reflect the operating status of the disconnecting switch from multiple dimensions, providing rich and reliable data support for accurately assessing its operating status. Compared with single data monitoring, it can more accurately capture subtle changes in equipment operation.

[0050] 2. The intelligent monitoring system for disconnecting switches disclosed in this invention includes a data transceiver unit that performs preliminary processing on the collected multi-source sensor data, a data aggregation unit that receives and aggregates the data and stores it, and a data transmission unit that remotely transmits the aggregated data to the backend. Each unit has a clear division of labor and works in concert to ensure that the data can be transmitted and processed in a timely and accurate manner, avoiding data loss or delay, and providing a guarantee for subsequent data analysis.

[0051] 3. The intelligent monitoring system for disconnecting switches disclosed in this invention analyzes the received data, evaluates the operating status of the disconnecting switch, and predicts faults. By calculating key indicators such as the deviation between the current angle and the standard angle, and the contact temperature change rate, and comparing them with set thresholds, an alarm message is generated when an abnormality is detected. This intelligent analysis method can detect potential problems in the equipment in advance, issue timely warnings, buy valuable processing time for maintenance personnel, effectively prevent the further expansion of the fault, and reduce the probability of accidents.

[0052] 4. The intelligent monitoring system for disconnecting switches disclosed in this invention has a temperature rise sensor deployed in the current-carrying transition zone of the stationary contact, which is determined by finite element temperature field simulation. This simulation-based precise deployment method can more accurately monitor the temperature change of the contact, improve the sensitivity and accuracy of temperature monitoring, and promptly detect abnormal temperature rise caused by poor contact, overload current, or poor heat dissipation, thus providing strong support for ensuring contact performance and switch reliability.

[0053] 5. The intelligent monitoring system for disconnecting switches disclosed in this invention uses a passive short-range transceiver unit for its data transceiver unit, which is deployed on the sensor body, cable, or switch jumper adapter board. This design makes the installation of the data transceiver unit more flexible and convenient, adaptable to different equipment layouts and environmental requirements. At the same time, the passive design also reduces the system's energy consumption and complexity.

[0054] 6. The intelligent monitoring method for disconnecting switches disclosed in this invention employs a series of advanced data analysis techniques. For example, preprocessing operations such as integrity verification, outlier removal and correction, and time-frequency domain noise suppression are performed on the raw angle data to improve data quality. Based on Hidden Markov Model training, state patterns representing different operating states are obtained, and the operating state of the disconnecting switch is determined through real-time pattern matching. This model-based analysis method can more accurately identify the operating mode and fault characteristics of the equipment, improving the accuracy and reliability of fault diagnosis.

[0055] 7. The intelligent monitoring method for disconnecting switches disclosed in this invention adopts a hierarchical fault early warning mechanism in the configuration of the data processing unit and the fault determination of the intelligent monitoring method. According to the situation where the deviation value, temperature change rate and other indicators exceed different set thresholds, emergency alarm signals and early warning signals are issued respectively. This hierarchical early warning method can take different countermeasures according to the severity of the fault, so that maintenance personnel can arrange maintenance work more rationally, improve maintenance efficiency and reduce maintenance costs. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of the composition of the intelligent monitoring system for disconnecting switches according to an embodiment of the present invention;

[0057] Figure 2 This is a flowchart of the intelligent monitoring method for disconnecting switches according to an embodiment of the present invention;

[0058] Figure 3 This is a flowchart of the operation status analysis and fault prediction in the intelligent monitoring method for disconnecting switches according to an embodiment of the present invention. Detailed Implementation

[0059] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of this application. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to this application are not shown or described in the specification. This is to avoid obscuring the core parts of this application with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0060] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0061] like Figure 1 As shown, the intelligent monitoring system for disconnecting switches includes: a data acquisition unit (angle sensor, temperature rise sensor, current sensor, voltage sensor), a data transceiver unit, a data aggregation unit, a data transmission unit, and a data processing unit.

[0062] The data acquisition unit includes an angle sensor, a temperature rise sensor, a current sensor, and a voltage sensor.

[0063] Among them: the angle sensor is used to monitor the angle change of the disconnecting switch during the opening and closing process, so as to determine whether the disconnecting switch is in the correct position;

[0064] The temperature rise sensor monitors the temperature change of the contact, collects the temperature change of the contact, and transmits the temperature information to the data transceiver unit;

[0065] The current sensor is installed in the current loop of the disconnector switch, which can measure the current value in real time and accurately, and transmit the current data to the data transceiver unit.

[0066] A voltage sensor is used to monitor the voltage across the disconnector switch. Installed in the voltage circuit of the disconnector switch, it accurately measures the voltage value and transmits the voltage data to the data transceiver unit.

[0067] The data transceiver unit adopts a passive short-range transceiver device, which is deployed on the sensor body, cable, or switch jumper adapter board to receive data from angle sensor, temperature rise sensor, current sensor, and voltage sensor. After preliminary processing and integration of these data, it is transmitted to the data aggregation unit.

[0068] The data collection unit is housed within the operating mechanism housing or mounted on a support column. When the operating mechanism housing is electrically operated, the data collection unit shares the same power supply with it. When the operating mechanism housing is manually operated, the data collection unit has a separate power supply module to receive data from the data transceiver unit and further collect and process the data. The data collection unit also has data storage capabilities, allowing it to save historical data for subsequent analysis and retrieval.

[0069] The data transmission unit transmits the data processed by the data aggregation unit to the data processing unit, and then transmits it to the back-end platform via optical fiber or short-distance transmission such as LoRa.

[0070] The data processing unit receives data from the data transmission unit, analyzes and processes the data, evaluates and predicts the operating status of the disconnecting switch, determines whether there are potential faults in the disconnecting switch, and when an abnormality is detected, the data processing unit can promptly issue an alarm signal and transmit the fault information to the relevant monitoring center or the terminal equipment of maintenance personnel so that timely measures can be taken to handle the situation.

[0071] When the temperature rise sensor detects that the rate of temperature change at the contact exceeds the threshold, the data processing unit evaluates and predicts the operating status of the disconnecting switch and determines that there is a potential fault in the disconnecting switch. The temperature rise rate is calculated as (current temperature - initial temperature) / time interval.

[0072] An emergency alarm signal is issued when the rate of temperature change exceeds the first-level set threshold; a warning signal is issued when the rate of temperature change exceeds the second-level set threshold, wherein the first-level set threshold must be greater than the second-level set threshold.

[0073] When the angle sensor detects that the disconnector switch is not fully closed or fully open, the data processing unit evaluates and predicts the operating status of the disconnector switch, determining that there is a potential fault. The standard closing angle and standard opening angle are set separately, and the deviation between the current angle and the standard angle is:

[0074] An emergency alarm signal is issued when the deviation between the current angle and the standard angle exceeds the first-level set threshold; a warning signal is issued when the deviation between the current angle and the standard angle exceeds the second-level set threshold.

[0075] like Figure 2 The specific steps of the intelligent monitoring method for disconnecting switches are shown below:

[0076] S1. Multi-source data acquisition: The angle change data of the disconnector switch during opening and closing is acquired in real time through the angle sensor; the temperature change data of the stationary contact of the disconnector switch is acquired through the temperature rise sensor; and the real-time data of the current circuit and voltage circuit of the disconnector switch are acquired through the current sensor and voltage sensor, respectively.

[0077] S2. Data preprocessing and transmission: The data transceiver unit receives data from various sensors, performs verification, error correction and format conversion, and then transmits it to the data aggregation unit.

[0078] S3. Data aggregation and storage: The data aggregation unit receives data transmitted by the data transceiver unit, performs timestamp alignment, data cleaning and correlation integration, and stores it in the local database;

[0079] S4. Data transmission to the backend platform: The data transmission unit sends the data processed by the data aggregation unit to the data processing unit through short-distance transmission methods such as optical fiber or LoRa.

[0080] S5. Operation Status Analysis and Fault Prediction: After receiving the data, the data processing unit performs the following analyses: Based on the angle data, it determines whether the disconnecting switch is in the correct position; based on the temperature rise data, it monitors the risk of contact overheating and predicts the temperature trend by combining historical data; based on the current and voltage data, it assesses load anomalies and insulation performance degradation.

[0081] The nonlinear angle change pattern recognition is used in operation status analysis and fault prediction. The specific implementation process is as follows:

[0082] S51 Data Preprocessing

[0083] 1.1 Data Validation and Error Correction

[0084] Data integrity check: The collected raw angle data is checked to ensure that each sampling time has a corresponding angle value. If missing data exists, linear interpolation is used to fill it in. If at time point t... i and t i+2 There is data, and t i+1 If data is missing, then t i+1 Angle value A i and A i+2 t i and t i+2 The angle value at any given time.

[0085] Outlier Detection and Handling: Outlier detection is performed using Grubbs' criterion. For a set of angle data... Calculate its mean and standard deviation If a certain data A j satisfy Where g(n,α) is the Grubbs' critical value related to the sample size n and the significance level α, then the judgment is A. j Outliers are identified and removed, then replaced with the average of adjacent data.

[0086] 1.2 Wavelet Transform Processing

[0087] Wavelet basis function selection: The Daubechies wavelet (dbN, where N is the order of the vanishing moments of the wavelet) is selected as the basis function. For isolating switch angle data, db4 or db6 wavelets are usually chosen because they have good performance in signal denoising and feature extraction.

[0088] Signal decomposition: Wavelet transform is used to decompose the angle signal into different frequency bands. Mallat's algorithm is employed for multi-scale decomposition, decomposing the original signal A(t) into approximate parts A0. j (t) and details D j (t), where j is the decomposition scale. Perform 3-level wavelet decomposition to decompose the original signal into A3(t), D3(t), D2(t) and D1(t).

[0089] Noise Removal and Feature Preservation: High-frequency noise is removed by setting a threshold. For detailed features (D...) j (t) Thresholding is performed, and common thresholding methods include hard thresholding and soft thresholding. The formula for hard thresholding is: The soft thresholding formula is: Where T is the set threshold. After removing high-frequency noise, the low-frequency trend component A3(t) is retained as data for subsequent analysis.

[0090] S52 pattern construction

[0091] 2.1 Data Collection and Labeling

[0092] Data Collection: In the actual operating environment of the disconnector switch, a large amount of angle change data under normal and fault conditions is collected. Angle sensors are installed on the disconnector switch, and angle data is recorded during different operations (opening and closing). The operating status of the disconnector switch (normal, mechanical jamming, loose transmission components, and other fault types) is also recorded.

[0093] Data labeling: The collected data is labeled, and each angle change sequence is associated with the corresponding operating state. For angle sequences of normal closing operations, they are labeled as "normal closing"; for angle sequences of closing operations with mechanical jamming faults, they are labeled as "mechanical jamming fault closing".

[0094] 2.2 Hidden Markov Model (HMM) Parameter Initialization

[0095] State Definition: Define the hidden state of the HMM based on the operating state of the disconnector switch. Three hidden states can be defined: S1 (normal operation state), S2 (mechanical jamming state), and S3 (loose transmission components state).

[0096] Initial probability distribution π: Initialize the initial state probability distribution π = (π1, π2, π3), where π i This indicates that the system is initially in state S. i The probability is determined based on prior knowledge or experience. π = (0.7, 0.15, 0.15), indicating that the system is initially in a normal motion state with a relatively high probability.

[0097] State transition probability matrix A: Initialize the state transition probability matrix A = (a ij ) 3×3 , where a ij Indicates from state S i Transition to state S j The probability of . Let a. 11 =0.8, a 12 =0.1, a 13 =0.1 indicates that under normal motion conditions, there is a high probability of continuing to maintain normal motion, and there is also a certain probability of transitioning to a state of mechanical jamming or loose transmission components.

[0098] Observation probability matrix B: Initialize the observation probability matrix B = (b j (k)) 3×M , where b j (k) represents the state S j The probability of observing the k-th observation (angle change feature). Based on the characteristics of the preprocessed angle data, determine the range of observation values ​​and the discretization method, and initialize the observation probability matrix using statistical methods or parameter estimation methods.

[0099] 2.3 Model Training

[0100] Data preparation: The labeled angle change data sequence is used as training data. For each training sequence... , where o t It is the angular change characteristic observation value at time t.

[0101] Baum-Welch Algorithm: The Baum-Welch algorithm is used to train the Hidden Markov Model (HMM) parameters. This algorithm is an expectation-maximization (EM) algorithm that maximizes the likelihood probability of the training data by iteratively updating the HMM parameters λ=(π,A,B). The specific steps are as follows:

[0102] E-step (expected step): Based on the current model parameters λ (i) Calculate the forward probability α t (i) and backward probability β t (i), and the state occupancy probability γ t (i) and state transition probability ξ t (i,j).

[0103] M-step (maximization step): Update the model parameters λ based on the probability values ​​calculated in the E-step. (i+1) The updated formula is:

[0104]

[0105]

[0106]

[0107] Iteration Termination Condition: Set an iteration termination condition, such as the maximum number of iterations or a threshold for the change in likelihood probability. When the termination condition is met, the iteration stops, and the trained HMM model is obtained.

[0108] S53 Real-time Mode Matching

[0109] 3.1 Real-time data acquisition and preprocessing

[0110] Data acquisition: Real-time acquisition of angle data from disconnect switches, and conversion of this data into an angle change sequence.

[0111] Data preprocessing: The real-time acquired angle data undergoes the same preprocessing steps as the training data, including data verification, error correction, and wavelet transform, to obtain the angle variation feature sequence of the low-frequency trend component. .

[0112] 3.2 Viterbi Algorithm for Calculating Matching Degree

[0113] Initialization: Based on the trained HMM model parameters λ=(π,A,B), initialize the Viterbi variable δ1(i)=π. i b i (o real1 ), where i=1,2,3 correspond to 3 hidden states respectively. At the same time, the backtracking path ψ1(i)=0 is initialized.

[0114] Recursion: For ,calculate and

[0115] , where j=1,2,3.

[0116] Termination: Calculation and

[0117] Backtracking path: From Beginning, according to ψ t (j) Backtracking yields the optimal state sequence .

[0118] 3.3 Fault Diagnosis

[0119] Matching degree calculation: Calculate the matching degree between real-time data and each pattern. The matching degree can be measured by calculating the similarity between the optimal state sequence and the normal pattern state sequence. Calculate the Hamming distance d between two state sequences, and the matching degree. , where T is the length of the state sequence.

[0120] Threshold comparison: Set the matching threshold Mth If the real-time data matches the normal pattern with a degree of M... normal <M th The matching degree M with a certain failure mode (such as mechanical jamming mode) fault ≥M th If so, it can be determined that the disconnecting switch may have this type of mechanical fault.

[0121] Anomaly Alarm and Response: When an anomaly is detected, the data processing unit generates alarm information, including device ID, alarm type, anomaly parameters and timestamp, and sends it to the monitoring center or maintenance personnel's terminal device.

[0122] Selection of temperature rise sensor installation location: COMSOL was used to perform finite element temperature field simulation on the contact part of the contact network disconnector of the electrified railway. Based on the heat conduction principle of the contact system, a heat conduction model of the contact disconnector was established. The steady-state temperature field distribution under different rated current gradients was distinguished, and the heat-sensitive area was determined. From the heat temperature distribution, it was found that, apart from the contact area of ​​the moving and stationary contacts and the disconnector, the current-carrying transition area of ​​the stationary contact is the most temperature-sensitive area. Moreover, this location has space for the installation of sensitive components. Therefore, the current-carrying transition area of ​​the stationary contact of the disconnector was selected as the temperature rise monitoring area.

[0123] The mathematical model for heat conduction in the contact wire disconnector is as follows:

[0124]

[0125] in, Indicates the thermal conductivity in the x and y directions; α represents the material density; c represents the material specific heat; q represents the heat generation rate of the conductor; α represents the convective heat dissipation coefficient; T represents the unknown boundary temperature; T0 represents the known air temperature; K represents the thermal conductivity; and t represents time.

[0126] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the embodiments of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples. The above use of specific examples to illustrate the present invention is only for the purpose of helping to understand the present invention and is not intended to limit the present invention. For those skilled in the art, based on the concept of the present invention, several simple deductions, modifications, or substitutions can be made.

Claims

1. An intelligent monitoring system for disconnecting switches, characterized in that, It includes a data acquisition unit, a data transceiver unit, a data aggregation unit, a data transmission unit, and a data processing unit; The data acquisition unit is used to acquire multi-source sensor data reflecting the operating status of the disconnector switch. The multi-source sensor data includes at least angle data representing the mechanical position of the disconnector switch, temperature data representing the contact state of the contacts, and current and voltage data representing the electrical load. The data transceiver unit, connected to the data acquisition unit, is used to receive multi-source sensor data and perform preliminary processing. The data aggregation unit, connected to the data transceiver unit, is used to receive and aggregate data from the data transceiver unit and store it. The data transmission unit, connected to the data aggregation unit, is used to remotely transmit the aggregated data to the backend. The data processing unit, connected to the data transmission unit, is used to analyze the received data to assess the operating status of the disconnecting switch and predict faults, and to generate alarm information when an anomaly is detected.

2. The intelligent monitoring system for disconnecting switches as described in claim 1, characterized in that, The data acquisition unit includes: Angle sensors are used to monitor angle changes during the opening and closing of disconnecting switches; Temperature rise sensor is used to monitor temperature changes in the contacts of disconnecting switches; A current sensor is installed in the current loop of the disconnecting switch to measure the current value in real time. A voltage sensor is installed in the voltage circuit of the disconnecting switch to measure the voltage value in real time.

3. The intelligent monitoring system for disconnecting switches as described in claim 2, characterized in that, The data processing unit is configured as follows: Calculate the deviation between the current angle and the standard angle, and issue an emergency alarm signal when the deviation exceeds the first-level set threshold and an early warning signal when it exceeds the second-level set threshold; Calculate the contact temperature change rate and issue an emergency alarm signal when the temperature change rate exceeds the first-level set threshold and an early warning signal when it exceeds the second-level set threshold.

4. The intelligent monitoring system for disconnecting switches as described in claim 1, characterized in that, The temperature rise sensor is deployed in the current-carrying transition zone of the stationary contact, which was determined through finite element temperature field simulation.

5. The intelligent monitoring system for disconnecting switches as described in claim 1, characterized in that, The data transceiver unit is a passive short-range transceiver device, which is installed on the sensor body, cable, or switch jumper adapter board.

6. A method for intelligent monitoring of disconnecting switches, based on the intelligent monitoring system for disconnecting switches according to any one of claims 1 to 4, characterized in that, Includes the following steps: S1: Multi-source data acquisition, which collects the angle, contact temperature, current and voltage data of the disconnector switch through angle sensor, temperature rise sensor, current sensor and voltage sensor respectively; S2: Data preprocessing and transmission: Data from each sensor is received through the data transceiver unit, and after verification, error correction and format conversion, it is transmitted to the data aggregation unit. S3: Data collection and storage. Data is received through the data collection unit, and timestamps are aligned, data is cleaned and integrated, and then stored in the local database. S4: Data transmission: The data processed by the data collection unit is sent to the data processing unit through the data transmission unit. S5: Operation status analysis and fault prediction. The data processing unit analyzes and processes the received data to assess the operation status of the disconnector switch and predict faults, and issues an alarm when an anomaly occurs.

7. The intelligent monitoring method for disconnecting switches as described in claim 6, characterized in that, S5 includes the following steps: S51: Data preprocessing, verifying and correcting the collected raw angle data, and using wavelet transform for noise reduction to extract low-frequency trend components for analysis; S52: Pattern construction, based on a large amount of angle change data collected and labeled in the actual operating environment of disconnecting switches under normal and fault conditions, a hidden Markov model is trained to obtain state patterns representing different operating states. S53: Real-time pattern matching. After preprocessing the angle data collected in real time, the obtained sequence is matched with the trained state pattern to calculate the matching degree. The operating status of the disconnector is determined based on the matching degree result.

8. The intelligent monitoring method for disconnecting switches as described in claim 7, characterized in that, S51 includes the following steps: Data integrity verification: Traverse the original angle data and check whether there are any missing data at each sampling time; if there are missing data, use linear interpolation to fill the gaps based on data from adjacent time points to ensure the continuity of the time series; Outlier removal and correction: Outliers are detected statistically based on the Grubbs criterion. The mean and standard deviation of the data are calculated, data points that deviate from the mean by more than a critical value are removed, and the neighborhood mean is used as a replacement. Time-frequency domain noise suppression: Wavelet transform is used to decompose the angle signal into multiple scales. The Daubechies wavelet is selected to extract the low-frequency trend component. High-frequency noise is removed by thresholding, while retaining the characteristic signal that reflects the mechanical motion state.

9. The intelligent monitoring method for disconnecting switches as described in claim 7, characterized in that, S52 includes the following steps: Hidden state definition: Define multiple hidden states that represent the health status of the device; Initial state probability allocation: Based on prior knowledge or historical data, set the initial state probability distribution; State transition probability configuration: Construct the state transition probability matrix; Observation probability matrix generation: Based on the preprocessed angle data features, the probability distribution of observations corresponding to each hidden state is generated through statistical modeling or parameter estimation methods.

10. The intelligent monitoring method for disconnecting switches as described in claim 7, characterized in that, S53 includes the following steps: Real-time data acquisition and preprocessing: synchronously acquire angle data and perform verification, interpolation and wavelet denoising to generate feature sequences of the same dimension as the training data; Optimal state path decoding: Using the trained HMM model, the matching probability between the real-time feature sequence and each state pattern is calculated by the Viterbi algorithm, and the optimal state transition path is generated by decoding. Multi-level fault threshold determination: Calculate the similarity between the optimal path and the normal mode, and trigger alarms in stages based on preset thresholds.

Images