An intelligent anti-leakage monitoring and protection system

Abstract

The present invention relates to the field of leakage monitoring technology, and in particular to an intelligent anti-leakage monitoring and protection system. It includes a leakage detection module, an edge computing module, an analysis and control module, a hierarchical warning module and a system management module that are connected in communication in sequence. The leakage detection module monitors the leakage status of switchgear and ground wires and collects current data. The edge computing module uses real-time current detection algorithm and Fourier transform to analyze data, extract current characteristics, and perform instantaneous current limiting. The analysis and control module uses a deep learning algorithm to evaluate the current recovery situation and predict the risk of leakage. The hierarchical warning module implements low, medium and high-level warning strategies. The system management module monitors the equipment through the Internet of Things technology and pushes notifications through an APP. The present invention combines a variety of technical means, identifies leakage hazards through big data analysis, and provides preventive maintenance suggestions, thereby achieving comprehensive monitoring and protection of the system and improving the safety and intelligence level of the system.

Description

Technical Field

[0001] The invention relates to the technical field of leakage monitoring, and in particular to an intelligent leakage prevention monitoring and protection system. Background Art

[0002] With the rapid development of social economy and the popularization of power application, power safety issues have received increasing attention. Leakage not only causes waste of electric energy, but more importantly, it may bring serious safety hazards to people and equipment. Traditional leakage protection devices mostly use simple mechanical circuit breakers or leakage protectors, which have certain limitations in protection effect, sensitivity and intelligence. Especially in complex power environments, these traditional devices are difficult to achieve real-time monitoring and precise protection, which easily leads to leakage accidents. At present, there are still the following problems: data processing in existing systems often relies on central servers, there are delays in data transmission and processing, and real-time data analysis and processing cannot be achieved, affecting the timeliness and accuracy of leakage monitoring; the existing system lacks intelligent data analysis capabilities and cannot perform real-time evaluation and prediction of leakage risks. This deficiency limits the system's preventive maintenance capabilities and cannot effectively prevent potential leakage faults; the early warning mechanism of the existing system is usually relatively simple, and it is difficult to take corresponding countermeasures according to different leakage risk levels. Summary of the invention

[0003] To solve the above problems, the present invention provides an intelligent anti-leakage monitoring and protection system, which combines sensing technology, information processing technology and communication technology, accurately identifies leakage hazards through big data analysis, and provides preventive maintenance suggestions and a multi-level alarm mechanism, thereby realizing comprehensive monitoring and protection of the system.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] An intelligent anti-leakage monitoring and protection system, comprising a leakage detection module, an edge computing module, an analysis and control module, a hierarchical warning module and a system management module which are sequentially connected in communication;

[0006] The leakage detection module is used to monitor the leakage status of the switch device and the ground wire using a sensor network, and collect current data at the same time; the current data includes the current and voltage of the switch device, and the resistance and frequency of the ground wire;

[0007] The edge computing module is used to locally process the current data, use real-time current detection algorithm and Fourier transform fast data analysis to extract the current phase angle and current waveform information, and realize instantaneous current limiting operation through parallel high-speed relays and reactors;

[0008] The analysis and control module is used to receive and analyze the data processed by the edge computing module, use the deep learning algorithm to evaluate the current recovery in real time, and perform predictive analysis on the leakage risk based on the real-time monitoring data and historical records;

[0009] The hierarchical warning module is used to implement a multi-level warning strategy based on the analysis results of the analysis and control module, including: low-level warning, medium-level warning and high-level warning;

[0010] The system management module is used to manage and monitor the entire system. It uses the Internet of Things technology to monitor in real time the ground wire contact with the ground, the aging degree of the ground wire and the signal reception status of the system equipment and switch boxes at all levels, and pushes real-time notifications to users through APP or mini-programs.

[0011] Furthermore, the leakage detection module includes an intelligent sensing unit and a ground monitoring unit;

[0012] The intelligent sensing unit is used to monitor the current and voltage of the switch device, collect electrical parameter data in real time, and transmit the data to the edge computing module through wireless communication;

[0013] The ground wire monitoring unit is used to monitor the resistance and frequency of the ground wire using high-frequency impedance measurement technology, and transmit the monitoring data to the edge computing module.

[0014] Furthermore, the formula for the high frequency impedance measurement technique is as follows:

[0015]

[0016] Among them, Z(f) represents the impedance of the ground wire at frequency f; R represents the resistance of the ground wire; f represents the signal frequency; L represents the inductance of the ground wire, and the inductance represents the degree of resistance of the ground wire to the change of current; C represents the capacitance of the ground wire.

[0017] Furthermore, the operation process of the edge computing module includes the following steps:

[0018] receiving real-time current and voltage data from the leakage detection module, including current and voltage values ​​of the switchgear, and resistance and frequency data of the ground wire;

[0019] Preprocess the collected data, including denoising, filtering and data smoothing, and use wavelet transform or Kalman filtering to remove noise;

[0020] Performing frequency domain analysis on the preprocessed data by fast Fourier transform and extracting current characteristics, wherein the current characteristics include current phase angle and waveform information, current peak value, root mean square value, harmonic components and phase difference;

[0021] Based on the extracted current characteristics, instantaneous current limiting operation is achieved through high-speed relays and reactors;

[0022] All processed data is stored locally and backed up to the cloud regularly.

[0023] Furthermore, the formula of the fast Fourier transform is as follows:

[0024]

[0025] Among them, X(k) represents the spectrum value at frequency k, which is used to analyze the frequency domain characteristics of current and voltage; X(n) represents the sampled current data or voltage data at the nth time point; N represents the total number of sampling points, that is, the number of data points collected within a certain time window; k represents the frequency sequence number, from 0 to N-1, which is used to determine each frequency component; j represents the imaginary unit, which is used to calculate the complex spectrum value.

[0026] Furthermore, the operation process of the analysis control module includes the following steps:

[0027] Receive and analyze the current and voltage data of the edge computing module, including the current phase angle, waveform information, peak value, root mean square value, harmonic component and phase difference;

[0028] The parsed data is input into a deep learning algorithm model that combines the self-attention mechanism with a temporal convolutional network, and the data at different time points are weighted through the self-attention mechanism;

[0029] The temporal convolutional network is used to perform convolution operations on the data processed by the self-attention mechanism, extract the current recovery features in the time series, and generate feature vectors;

[0030] Based on the generated feature vectors, conduct leakage risk assessment and prediction analysis, combine real-time monitoring data and historical records, identify potential leakage hazards through model training and parameter optimization, and calculate the leakage risk probability;

[0031] Generate a detailed analysis report, including the severity of leakage risk, potential hidden danger locations and potential causes.

[0032] Furthermore, the formula of the deep learning algorithm model is as follows:

[0033]

[0034] Where y is the leakage risk probability, X represents the input current data feature vector, including phase angle, waveform information, peak value, RMS value, harmonic component and phase difference; W represents the learnable weight matrix in the self-attention mechanism; d k Represents the dimension of the current data feature vector; W cRepresents the convolution kernel of the temporal convolutional network; b c Represents the bias of the temporal convolutional network; W o and b o represents the weight and bias of the output layer; σ and softmax represent the activation function.

[0035] Furthermore, when the low-level warning specifically detects a slight leakage, the system sends a prompt to the user through the APP or mini-program, suggesting that the device and line be checked;

[0036] When the intermediate warning specifically detects a moderate leakage, the system performs local current limiting through the edge computing module and sends an emergency alert to the user, suggesting immediate inspection and repair;

[0037] When the advanced warning specifically detects a serious leakage, the system automatically cuts off the power supply, implements an emergency power-off through the edge computing module, and sends an emergency notification to the user and relevant safety management personnel, requiring immediate processing.

[0038] Further, the system management module includes a device management unit and a visualization unit;

[0039] The device management unit is used to manage and maintain the hardware devices of the system, including status monitoring, fault diagnosis, maintenance records and remote control of the equipment;

[0040] The visualization unit is used to push real-time notifications to users via an APP or mini-program.

[0041] The beneficial effects of the present invention are:

[0042] Through the sensor network of the leakage detection module of the present invention, the system can monitor the leakage status of the switchgear and the ground wire in real time and collect detailed current data. This comprehensive monitoring ensures that the system can detect leakage in time and ensure the safety of electricity use. The edge computing module uses real-time current detection algorithm and Fourier transform to perform rapid data analysis to extract the phase angle and current waveform information of the current. This not only improves the efficiency of data processing, but also realizes instantaneous current limiting operation through parallel high-speed relays and reactors, further improving the response speed and accuracy of the system. The analysis and control module evaluates the current recovery situation in real time through a deep learning algorithm, and performs leakage risk prediction and analysis in combination with real-time monitoring data and historical records. This intelligent analysis capability enables the system to not only respond to the current leakage situation, but also predict potential risks and prevent them in advance. The hierarchical warning module implements a multi-level warning strategy based on the results of the analysis and control module, including low-level, intermediate and high-level warnings. The multi-level warning mechanism can take corresponding countermeasures according to different risk levels, improving the flexibility and reliability of the system. The system management module uses the Internet of Things technology to monitor the ground contact, ground aging and signal reception status of system equipment and switch boxes at all levels in real time, and notifies users in real time through APP or mini-program push. This comprehensive management method not only ensures the stable operation of the system, but also allows users to understand the system status at any time and take timely response measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a module schematic diagram of an intelligent anti-leakage monitoring and protection system of the present invention.

[0044] Figure 2 It is a flowchart of the operation process of the edge computing module in the present invention.

[0045] Figure 3 It is a flow chart of the operation process of the analysis control module in the present invention. DETAILED DESCRIPTION

[0046] See also Figure 1-3 As shown, the present invention relates to an intelligent anti-leakage monitoring and protection system.

[0047] Example

[0048] An intelligent anti-leakage monitoring and protection system, comprising a leakage detection module, an edge computing module, an analysis and control module, a hierarchical warning module and a system management module which are sequentially connected in communication;

[0049] The leakage detection module is used to monitor the leakage status of the switch device and the ground wire using a sensor network, and collect current data at the same time; the current data includes the current and voltage of the switch device, and the resistance and frequency of the ground wire;

[0050] Wherein, the leakage detection module includes an intelligent sensing unit and a ground monitoring unit;

[0051] The intelligent sensing unit is used to monitor the current and voltage of the switch device, collect electrical parameter data in real time, and transmit the data to the edge computing module through wireless communication;

[0052] The ground wire monitoring unit is used to monitor the resistance and frequency of the ground wire using high-frequency impedance measurement technology, and transmit the monitoring data to the edge computing module;

[0053] It should be noted that the intelligent sensing unit is used to monitor the current and voltage of the switchgear, and is mainly composed of the following parts:

[0054] Hall effect current sensors are used to measure current. They detect the current flowing through a conductor and convert it into a measurable voltage output with high sensitivity and fast response.

[0055] Voltage sensors are used to measure voltage, usually using a voltage divider or capacitive coupling, and can accurately measure voltage changes in electrical equipment.

[0056] The data collector is used to collect the electrical parameter data output by the sensor. It has a multi-channel input interface, which can simultaneously receive data from multiple sensors and perform preliminary processing such as filtering and denoising.

[0057] Wireless communication module: Zigbee module is used for short-distance data transmission, with the characteristics of low power consumption and high reliability, suitable for sensor networks in a small range. LoRa module is used for long-distance data transmission, with the characteristics of wide coverage and strong anti-interference ability, suitable for large-scale monitoring areas.

[0058] The embedded processor is responsible for processing and formatting the raw data collected by the data collector, and transmitting the data to the edge computing module through the wireless communication module.

[0059] The ground wire monitoring unit is used to monitor the resistance and frequency of the ground wire using high-frequency impedance measurement technology. It mainly consists of the following parts:

[0060] The high-frequency signal generator is used to generate a high-frequency signal and inject it into the ground line. The high-frequency signal can more effectively detect the impedance change of the ground line and provide more accurate ground line status information.

[0061] The impedance measuring device calculates the impedance of the ground wire by detecting the signal reflected from the ground wire. The device can respond quickly and provide high-precision impedance measurement results.

[0062] The data processor analyzes and processes the measured impedance data to extract the resistance and frequency information of the ground wire. The processor usually uses a high-performance microcontroller or FPGA to achieve fast data processing.

[0063] In addition, the system also includes an anti-electric shock monitoring module for real-time detection of the electrical safety status of the switchgear and the ground wire to prevent people from electric shock;

[0064] The anti-electric shock monitoring module detects whether a human body touches the electrical equipment through a highly sensitive electric shock sensor, and immediately triggers an alarm when an electric shock risk is detected;

[0065] The system is equipped with a dual protection mechanism. Once the anti-electric shock monitoring module detects the risk of electric shock, it immediately sends a signal to the edge computing module to initiate emergency power-off measures, and at the same time transmits the alarm information to the system management module and the early warning module;

[0066] When the emergency power-off device detects a risk of electric shock, it immediately cuts off the power supply to prevent people from being injured by electric shock;

[0067] The early warning system sends out warnings to users in various forms such as sound alarms, light prompts, and mobile phone notifications, ensuring that users are aware of the risk of electric shock in a timely manner and take protective measures.

[0068] Furthermore, the formula for the high frequency impedance measurement technique is as follows:

[0069]

[0070] Among them, represents the impedance of the ground wire at frequency f, and the impedance value represents the degree of resistance of the ground wire to AC current. The impedance value of Z(f) measured at different frequencies is used to evaluate the health status of the ground wire and whether there is any abnormality; R represents the resistance of the ground wire, which is one of the inherent electrical characteristics of the ground wire and reflects the degree of resistance of the ground wire to DC current. By measuring the resistance, it can be determined whether the material and connection of the ground wire are good; f represents the signal frequency, which is the frequency of the AC signal applied to the ground wire. Measuring the impedance of the ground wire at different frequencies can provide more information about the state of the ground wire, such as the impedance change at high frequency can reflect the aging degree and contact condition of the ground wire; L represents the inductance of the ground wire, and the inductance represents the degree of resistance of the ground wire to the change of current. The measurement of the inductance of the ground wire helps to determine the response characteristics of the ground wire at high frequency, which is very important for detecting the health status and aging degree of the ground wire; C represents the capacitance of the ground wire, and the capacitance reflects the response characteristics of the ground wire to voltage changes. By measuring the capacitance of the ground wire, the performance and health status of the ground wire at high frequency can be understood, especially to determine whether the ground wire has leakage risks.

[0071] The edge computing module is used to locally process the current data, utilize real-time current detection algorithm and Fourier transform fast data analysis to extract the current phase angle and current waveform information, and implement instantaneous current limiting operation through parallel high-speed relays and reactors to ensure that the power supply is quickly cut off when the current is abnormal to avoid electric shock accidents.

[0072] The operation process of the edge computing module includes the following steps:

[0073] receiving real-time current and voltage data from the leakage detection module, including current and voltage values ​​of the switchgear, and resistance and frequency data of the ground wire;

[0074] Preprocess the collected data, including denoising, filtering and data smoothing, and use wavelet transform or Kalman filtering to remove noise;

[0075] Performing frequency domain analysis on the preprocessed data by fast Fourier transform and extracting current characteristics, wherein the current characteristics include current phase angle and waveform information, current peak value, root mean square value, harmonic components and phase difference;

[0076] It should be noted that the phase angle and waveform information: the frequency spectrum of the current signal is obtained through FFT, and the phase angle and waveform information of the current are extracted.

[0077] Current peak value: Calculate the peak value of the current signal to determine the maximum current intensity.

[0078] Root Mean Square (RMS): Calculates the root mean square value of current to evaluate the effective value of current.

[0079] Harmonic components: Analyze the harmonic components in the current signal and evaluate the current quality.

[0080] Phase difference: Calculate the phase difference between different signals to determine the synchronization of the signals.

[0081] Based on the extracted current characteristics, instantaneous current limiting operation is achieved through high-speed relays and reactors; ensuring rapid response in the event of current overload or electric shock to protect personnel safety;

[0082] It should be noted that high-speed relays are used to achieve fast switching operations, and the action of the relay is determined according to the current characteristic value. The reactor is used to perform instantaneous current limiting operations to reduce current shocks and protect the safety of the power system. The system involves a reactor and two high-speed relays, which are connected in parallel. Due to the frequent operation of the high-speed relay, the system adds an equivalent high-speed relay as a backup, so the system uses two high-speed relays.

[0083] All processed data is stored locally and backed up to the cloud regularly; ensuring the integrity and security of the data so that rapid response and processing can be carried out when electric shock hazards occur again;

[0084] Furthermore, the formula of the fast Fourier transform is as follows:

[0085]

[0086] Among them, X(k) represents the spectrum value at frequency k, which is used to analyze the frequency domain characteristics of current and voltage; X(n) represents the sampled current data or voltage data at the nth time point; N represents the total number of sampling points, that is, the number of data points collected within a certain time window; k represents the frequency sequence number, from 0 to N-1, which is used to determine each frequency component; j represents the imaginary unit, which is used to calculate the complex spectrum value.

[0087] The analysis and control module is used to receive and analyze the data processed by the edge computing module, use the deep learning algorithm to evaluate the current recovery in real time, and perform predictive analysis on the leakage risk based on the real-time monitoring data and historical records;

[0088] The operation process of the analysis control module includes the following steps:

[0089] Receive and analyze the current and voltage data of the edge computing module, including the current phase angle, waveform information, peak value, root mean square value, harmonic component and phase difference;

[0090] The parsed data is input into a deep learning algorithm model that combines the self-attention mechanism with a temporal convolutional network, and the data at different time points are weighted through the self-attention mechanism;

[0091] Specifically, the formula of the attention mechanism is as follows:

[0092]

[0093] Where X represents the input current data feature matrix, including phase angle, waveform information, peak value, RMS value, harmonic component and phase difference; W is the learnable weight matrix in the self-attention mechanism; d k Represents the dimension of the current data feature vector; softmax is an activation function used to calculate the relative importance of weights. The specific calculation method is as follows:

[0094]

[0095] Among them, z i Represents the i-th element in the input vector, and N is the length of the input vector.

[0096] The temporal convolutional network is used to perform convolution operations on the data processed by the self-attention mechanism, extract the current recovery features in the time series, and generate feature vectors;

[0097] Based on the generated feature vectors, conduct leakage risk assessment and prediction analysis, combine real-time monitoring data and historical records, identify potential leakage hazards through model training and parameter optimization, and calculate the leakage risk probability;

[0098] Generate a detailed analysis report, including the severity of leakage risk, potential hidden danger locations and potential causes.

[0099] It should be noted that the severity level is to assess the risk of electric leakage and give a specific risk level.

[0100] Hidden danger location: Identify the location of equipment or lines that may have leakage hazards.

[0101] Potential causes: Analyze the potential causes of leakage risks and provide targeted treatment suggestions.

[0102] Furthermore, the formula of the deep learning algorithm model is as follows:

[0103]

[0104] Where y is the probability of leakage risk, which predicts the probability value of leakage risk; X represents the input current data feature vector, including phase angle, waveform information, peak value, root mean square value, harmonic component and phase difference; W represents the learnable weight matrix in the self-attention mechanism; d k Represents the dimension of the current data feature vector; W c Represents the convolution kernel of the temporal convolutional network, which is used to extract local features in the time series; b c Represents the bias of the temporal convolutional network, which is used to adjust the output of the convolution operation; W o and b o Represents the weight and bias of the output layer, which converts the output of the convolutional network into the final leakage risk probability and is used to adjust the final output value; σ and softmax represent activation functions.

[0105] The hierarchical warning module is used to implement a multi-level warning strategy based on the analysis results of the analysis and control module, including: low-level warning, medium-level warning and high-level warning;

[0106] Among them, when the low-level warning specifically detects a slight leakage, the system sends a prompt to the user through the APP or mini program, suggesting that the equipment and lines be checked;

[0107] When the intermediate warning specifically detects a moderate leakage, the system performs local current limiting through the edge computing module and sends an emergency alert to the user, suggesting immediate inspection and repair;

[0108] When the advanced warning specifically detects a serious leakage, the system automatically cuts off the power supply, implements an emergency power-off through the edge computing module, and sends an emergency notification to the user and relevant safety management personnel, requiring immediate processing.

[0109] It should be noted that corresponding early warning strategies are formulated according to the leakage risk level, including low-level warning, medium-level warning and high-level warning.

[0110] Low-level warning strategy:

[0111] Trigger condition: A slight leakage is detected (for example, the leakage current is lower than the set threshold).

[0112] Warning content: The system sends prompt information to users through APP or mini-programs, suggesting that users check electrical equipment and lines to ensure safety.

[0113] Notification method: Push notification is used, and the message content includes the leakage location, time and recommended inspection measures.

[0114] Intermediate warning strategy:

[0115] Trigger condition: Moderate leakage is detected (for example, the leakage current is higher than the low warning threshold but lower than the high warning threshold).

[0116] Warning content: The system performs local current limiting operations through the edge computing module and sends emergency alerts to users, suggesting that users check and repair immediately.

[0117] Notification method: In addition to push notifications, users can also be reminded via SMS or phone calls. The message content includes detailed leakage information and emergency treatment suggestions.

[0118] Advanced early warning strategy:

[0119] Trigger condition: severe leakage is detected (for example, the leakage current is higher than the set advanced warning threshold).

[0120] Warning content: The system automatically cuts off the power supply, implements emergency power-off operation through the edge computing module, and sends emergency notifications to users and relevant safety managers, requiring them to immediately deal with the leakage problem.

[0121] Notification method: Multi-channel notification, including APP push, SMS, phone and email. The notification content details the leakage situation, the execution of the power-off operation and emergency treatment suggestions.

[0122] According to the formulated early warning strategy, specific early warning operations are implemented to ensure accurate transmission of early warning information, and automated operations are taken when necessary.

[0123] Generate specific warning information based on the warning strategy, including leakage location, time, leakage current value, and recommended inspection and treatment measures. The information sending module sends the warning information to relevant users and managers through various communication methods (such as APP push, SMS, phone and email).

[0124] When an intermediate warning is triggered, the edge computing module implements local current limiting operations by controlling high-speed relays and reactors to reduce the risk of leakage and ensure the safety of the power system.

[0125] When the advanced warning is triggered, the system automatically cuts off the power supply and implements an emergency power-off operation through the edge computing module to prevent further spread of leakage and ensure the safety of personnel and equipment.

[0126] At the same time, there is an arc extinguishing function when the advanced warning is triggered. The arc extinguishing function eliminates or reduces the arc at the moment of disconnection in the circuit. When the circuit is disconnected, the current is suddenly interrupted, and the current passes through the air or other media to form a high-temperature, high-pressure ionization channel, which may cause fire or equipment damage.

[0127] By real-time monitoring of the sending and execution status of warning information, the warning operation is ensured to be timely and effective. Users and managers can feedback the warning processing status through the APP or mini program, and the system will adjust and optimize according to the feedback information.

[0128] The system management module is used to manage and monitor the entire system. It uses the Internet of Things technology to monitor the ground contact of system equipment and switch boxes at all levels, the aging degree of the ground wire, and the signal reception status in real time, and notifies users in real time through APP or mini-program push;

[0129] Wherein, the system management module includes a device management unit and a visualization unit;

[0130] The device management unit is used to manage and maintain the hardware devices of the system, including status monitoring, fault diagnosis, maintenance records and remote control of the equipment;

[0131] The visualization unit is used to push real-time notifications to users via an APP or mini-program.

[0132] In summary, this application realizes real-time monitoring of the current, voltage, impedance and frequency of the switchgear and ground wire through the intelligent sensing unit and ground wire monitoring unit of the leakage detection module, which can quickly and accurately detect leakage and ensure timely response of the system. The edge computing module uses fast Fourier transform (FFT) and wavelet transform and other technologies to pre-process and analyze the current data in the frequency domain, which can quickly extract current characteristics and improve the efficiency and accuracy of data processing. The analysis and control module uses a deep learning algorithm that combines the self-attention mechanism with a temporal convolutional network to evaluate the current recovery in real time and predict the leakage risk, thereby improving the intelligence level and prediction ability of the system. The system sets low-level, medium-level and high-level warning strategies, and performs corresponding warning and processing operations according to the leakage risk level to ensure that leakage hazards are discovered and handled in time, reducing the safety risks of the power system. Short-distance and long-distance data transmission is achieved through wireless communication modules (Zigbee and LoRa), ensuring the stability and reliability of data transmission, which is suitable for monitoring areas of different ranges. The system management module uses IoT technology to monitor device status, ground contact and signal reception status in real time, and pushes real-time notifications through APP or mini-programs, allowing users to understand the system operation status and processing status at any time.

[0133] The above implementation modes are merely descriptions of the preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary engineering and technical personnel in the field shall fall within the protection scope determined by the claims of the present invention.

Claims

1. An intelligent anti-leakage monitoring and protection system, characterized in that: It includes a leakage detection module, an edge computing module, an analysis and control module, a hierarchical warning module and a system management module which are sequentially connected in communication; The leakage detection module is used to monitor the leakage status of the switch device and the ground wire using a sensor network, and collect current data at the same time; the current data includes the current and voltage of the switch device, and the resistance and frequency of the ground wire; The edge computing module is used to locally process the current data, use real-time current detection algorithm and Fourier transform fast data analysis to extract the current phase angle and current waveform information, and realize instantaneous current limiting operation through parallel high-speed relays and reactors; The analysis and control module is used to receive and analyze the data processed by the edge computing module, use the deep learning algorithm to evaluate the current recovery in real time, and perform predictive analysis on the leakage risk based on the real-time monitoring data and historical records; The hierarchical warning module is used to implement a multi-level warning strategy based on the analysis results of the analysis and control module, including: low-level warning, medium-level warning and high-level warning; The system management module is used to manage and monitor the entire system. It uses the Internet of Things technology to monitor the ground contact of system equipment and switch boxes at all levels, the aging degree of the ground wire, and the signal reception status in real time, and notifies users in real time through APP or mini-program push; The operation process of the analysis control module includes the following steps: Receive and analyze the current and voltage data of the edge computing module, including the current phase angle, waveform information, peak value, root mean square value, harmonic component and phase difference; The parsed data is input into a deep learning algorithm model that combines the self-attention mechanism with a temporal convolutional network, and the data at different time points are weighted through the self-attention mechanism; The temporal convolutional network is used to perform convolution operations on the data processed by the self-attention mechanism, extract the current recovery features in the time series, and generate feature vectors; Based on the generated feature vectors, conduct leakage risk assessment and prediction analysis, combine real-time monitoring data and historical records, identify potential leakage hazards through model training and parameter optimization, and calculate the leakage risk probability; Generate detailed analysis reports, including the severity of leakage risk, potential hidden danger locations and potential causes; The formula of the deep learning algorithm model is as follows: ； Where y is the leakage risk probability, X represents the input current data feature vector, including phase angle, waveform information, peak value, RMS value, harmonic component and phase difference; W represents the learnable weight matrix in the self-attention mechanism; Represents the dimension of the current data feature vector; Represents the convolution kernel of the temporal convolutional network; Represents the bias of the temporal convolutional network; and Represents the weights and biases of the output layer; and Represents the activation function.

2. The intelligent anti-leakage monitoring and protection system according to claim 1 is characterized in that: The leakage detection module includes an intelligent sensing unit and a ground monitoring unit; The intelligent sensing unit is used to monitor the current and voltage of the switch device, collect electrical parameter data in real time, and transmit the data to the edge computing module through wireless communication; The ground wire monitoring unit is used to monitor the resistance and frequency of the ground wire using high-frequency impedance measurement technology, and transmit the monitoring data to the edge computing module.

3. The intelligent anti-leakage monitoring and protection system according to claim 2 is characterized in that: The formula for the high frequency impedance measurement technique is as follows: ； in, It represents the impedance of the ground wire at frequency f; R represents the resistance of the ground wire; f represents the signal frequency; L represents the inductance of the ground wire, and the inductance represents the degree of resistance of the ground wire to the change of current; C represents the capacitance of the ground wire.

4. The intelligent anti-leakage monitoring and protection system according to claim 1 is characterized in that: The operation process of the edge computing module includes the following steps: receiving real-time current and voltage data from the leakage detection module, including current and voltage values ​​of the switchgear, and resistance and frequency data of the ground wire; Preprocess the collected data, including denoising, filtering and data smoothing, and use wavelet transform or Kalman filtering to remove noise; Performing frequency domain analysis on the preprocessed data by fast Fourier transform and extracting current characteristics, wherein the current characteristics include current phase angle and waveform information, current peak value, root mean square value, harmonic components and phase difference; Based on the extracted current characteristics, instantaneous current limiting operation is achieved through high-speed relays and reactors; All processed data is stored locally and backed up to the cloud regularly.

5. The intelligent anti-leakage monitoring and protection system according to claim 4 is characterized in that: The formula for the fast Fourier transform is as follows: ； Among them, X(k) represents the spectrum value at frequency k, which is used to analyze the frequency domain characteristics of current and voltage; x(n) represents the sampled current data or voltage data at the nth time point; N represents the total number of sampling points, that is, the number of data points collected within a certain time window; k represents the frequency sequence number, from 0 to N-1, which is used to determine each frequency component; j represents the imaginary unit, which is used to calculate the complex spectrum value.

6. The intelligent anti-leakage monitoring and protection system according to claim 1 is characterized in that: When the low-level warning specifically detects a slight leakage, the system sends a prompt to the user through the APP or mini program, suggesting that the device and line be checked; When the intermediate warning specifically detects a moderate leakage, the system performs local current limiting through the edge computing module and sends an emergency alert to the user, suggesting immediate inspection and repair; When the advanced warning specifically detects a serious leakage, the system automatically cuts off the power supply, implements an emergency power-off through the edge computing module, and sends an emergency notification to the user and relevant safety management personnel, requiring immediate processing.

7. The intelligent anti-leakage monitoring and protection system according to claim 1 is characterized in that: The system management module includes a device management unit and a visualization unit; The device management unit is used to manage and maintain the hardware devices of the system, including status monitoring, fault diagnosis, maintenance records and remote control of the equipment; The visualization unit is used to push real-time notifications to users via an APP or mini-program.

Images