Weak signal multi-module networking adaptive monitoring system for power transmission and transformation construction

Abstract

The invention provides a weak signal multi-module networking self-adaptive monitoring system for power transmission and transformation construction, and relates to the technical field of multi-module networking, and the system comprises a multi-mode module which is used for deploying a multi-mode sensing unit at a key monitoring point of power transmission and transformation equipment, collecting vibration, partial discharge, leakage current and environmental parameter signals, and transmitting the signals to a monitoring center; the signals are processed through packaging of an electromagnetic shielding shell and a built-in harmonic filter circuit so as to obtain original monitoring signal data; and the detection module is used for inputting the original monitoring signal data into the chaotic enhanced weak signal detection device, and extracting a fault feature vector by capturing an attractor phase diagram abrupt change and a time domain waveform threshold. Through the multi-mode sensing and self-adaptive networking technology, monitoring of power transmission and transformation weak signals and timely extraction of fault features are achieved, and the fault early warning accuracy and the self-adaptive capacity are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-module networking, and particularly to an adaptive monitoring system for weak-signal multi-module networking in power transmission and transformation construction. Background Art

[0002] In power transmission and transformation construction projects, traditional technologies have some limitations when facing weak signals in complex environments. For example, during the construction of a new substation in mountainous areas, due to complex terrain and insufficient base station coverage, traditional single-mode signal transmission technologies are prone to signal interruption or transmission delay problems. When construction workers use traditional monitoring equipment to record tower foundation pouring data during the key link of installing high-voltage transmission towers, unstable signals result in incomplete transmission of some data, and construction management personnel cannot obtain accurate construction progress and quality information in a timely manner.

[0003] In addition, in terms of networking flexibility, traditional systems also have deficiencies. In the construction scenario of mountainous area substations, when the construction area expands or the equipment position changes, traditional fixed networking modes are difficult to quickly adapt to the changes. If a new temporary cable laying monitoring point is added, due to the lack of adaptive networking capabilities of traditional systems, a large number of lines need to be re-laid and complex equipment debugging is required, consuming a lot of time and labor costs, resulting in the new monitoring point being unable to be connected to the system in a timely manner, there are monitoring blind spots, and the entire construction process cannot be comprehensively and real-time monitored. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide an adaptive monitoring system for weak-signal multi-module networking in power transmission and transformation construction, which realizes stable transmission in a weak-signal environment and dynamic adaptive adjustment of the networking topology through multi-module collaboration, and improves the real-time performance of the whole-process monitoring of power transmission and transformation construction.

[0005] To solve the above technical problems, the technical solution of the present invention is as follows:

[0006] In a first aspect, an adaptive monitoring system for weak-signal multi-module networking in power transmission and transformation construction includes:

[0007] A multi-modal module, which is used to deploy multi-modal sensing units at key monitoring points of power transmission and transformation equipment, collect vibration, partial discharge, leakage current and environmental parameter signals, and process the signals through electromagnetic shielding enclosures and built-in harmonic filtering circuits to obtain original monitoring signal data;

[0008] A detection module, which is used to input the original monitoring signal data into a chaotic enhanced weak-signal detection device, and extract fault feature vectors by capturing the mutation of the attractor phase diagram and the time-domain waveform threshold;

[0009] A transmission module, which is used to transmit the fault feature vector through a wired and wireless dual-channel hot standby transmission device. When the signal attenuation reaches a preset threshold, the transmission device automatically switches the transmission channel, and at the same time triggers the relay node to perform gain compensation on the fault feature vector;

[0010] A planning module, which is used to plan the transmission path for the fault feature vector data by adopting a clustered multi-hop ad hoc network architecture. The cluster head node realizes the autonomous repair of the wireless mesh network breakpoint based on the dynamic routing strategy, and transmits the fault feature vector data to the aggregation node;

[0011] A diagnosis module, which is used to perform spatio-temporal correlation fusion on the multi-source fault feature vectors according to the aggregation node, match the diagnosis results based on the pre-trained fault mode library, dynamically adjust the sampling rate, transmission power and filtering parameters according to the diagnosis results, and generate alarm events at the same time;

[0012] A regulation module, which is used to automatically reconstruct the multi-module network topology structure based on the fault diagnosis results and network status feedback, dynamically adjust the working modes of each sensing unit, and realize the adaptive maintenance and adjustment of the multi-module network monitoring.

[0013] Furthermore, input the original monitoring signal data into a chaotic enhanced weak signal detection device, and extract the fault feature vector by capturing the sudden change of the attractor phase diagram and the time-domain waveform threshold, including:

[0014] Perform normalization processing on the original monitoring signal data to eliminate the dimensional difference between different sensor signals, obtain the standardized signal data with a unified dimension, and input the standardized signal data as an external drive signal into the chaotic detection structure with preset parameters;

[0015] Calculate the dynamic stability index of the chaotic detection structure in real time, judge the change trend of the current state according to the dynamic stability index, and track the morphological pattern of the attractor trajectory formed in the state space;

[0016] When it is detected that the attractor trajectory suddenly changes from a chaotic state to a regular periodic state, record the moment when the state transition occurs. At the same time, in the time-domain waveform of the standardized signal, analyze the signal amplitude and frequency changes corresponding to before and after the transition moment;

[0017] Identify the waveform segments that meet the preset amplitude threshold and frequency change characteristics as the signal segments with faults, and perform wavelet packet decomposition on the potential fault signal segments to obtain the energy distribution in different frequency bands;

[0018] Based on the energy proportion of each frequency band, construct a frequency-domain energy distribution vector for characterizing the signal characteristics, and use the frequency-domain energy distribution vector as the fault feature vector reflecting the operating state of the device.

[0019] Further, the fault feature vector is transmitted through a wired and wireless dual-channel hot standby transmission device. When the signal attenuation reaches a preset threshold, the transmission device automatically switches the transmission channel, and at the same time triggers the relay node to perform gain compensation on the fault feature vector, including:

[0020] The main controller monitors the quality index of the currently activated primary transmission channel. If the primary channel is a wired channel, the bit error rate is monitored; if the primary channel is a wireless channel, the received signal strength indication value is monitored. When it is detected that the quality index of the primary channel is less than the corresponding preset threshold, it is determined that the channel signal attenuation has reached the preset threshold, and a channel switching instruction and a gain compensation trigger signal are generated;

[0021] Based on the channel switching instruction and the gain compensation trigger signal, the transmission device performs a switching operation from the current primary channel to the standby channel. At the same time, the gain compensation trigger signal activates the specified relay node on the transmission path to obtain the fault feature vector data packet whose transmission is interrupted due to the disconnection of the primary channel;

[0022] For the fault feature vector data packet, based on the signal attenuation data, predict the signal attenuation trajectory of the current transmission link, and according to the signal attenuation rate, dynamically calculate the compensation gain value required for power amplification of the fault feature vector data packet.

[0023] Further, for the fault feature vector data packet, based on the signal attenuation data, predict the signal attenuation trajectory of the current transmission link, and according to the signal attenuation rate, dynamically calculate the compensation gain value required for power amplification of the fault feature vector data packet, including:

[0024] Obtain the channel attenuation data and environmental interference characteristics of the current transmission link, and based on the correlation relationship between the time characteristics of the channel attenuation data and the environmental interference characteristics, predict the change trend of the signal attenuation amplitude with the transmission time and distance;

[0025] Combine the signal attenuation change trend with the real-time collected link impedance fluctuation parameters and the instantaneous value of the electromagnetic interference intensity to dynamically generate the predicted value of the real-time signal attenuation rate at each discrete position point in the transmission path;

[0026] According to the distribution of the real-time signal attenuation rate prediction values, divide the transmission path of the fault feature vector data packet into continuous sections according to the attenuation rate mutation points, and calculate the corresponding theoretical signal attenuation degree at each section node;

[0027] Compare the theoretical signal attenuation degree of the section node with the actual attenuation degree monitored by the sensor in real time to obtain the attenuation degree deviation amount at each node;

[0028] According to the attenuation degree deviation amount, combined with the influence degree ratio of the transmission distance of the corresponding section on the signal attenuation, dynamically generate the compensation gain value required for power amplification of the fault feature vector data packet.

[0029] Further, for the fault feature vector data, a clustered multi-hop ad-hoc network architecture is adopted to plan the transmission path. The cluster head node realizes autonomous repair of the wireless mesh network breakpoints based on the dynamic routing strategy and transmits the fault feature vector data to the aggregation node, including:

[0030] Receive the fault feature vector data packet after relay node gain compensation, and based on the node geographical location information and real-time link quality data required for data packet transmission, establish the topological connection structure between the nodes within the cluster in the clustered multi-hop ad-hoc network to form the initial transmission path;

[0031] Based on the topological connection structure, the cluster head node periodically detects the connectivity status of each neighbor node, and according to the detection results, updates the network routing table in real time, and for each path recorded in the table, comprehensively evaluates the performance based on the signal strength and the remaining energy of the node;

[0032] When the connectivity status monitoring indicates that there is a breakpoint in the current transmission path, the cluster head node starts the dynamic routing strategy for autonomous repair;

[0033] After the path repair is completed, the cluster head node relays and transmits the fault feature vector data to the aggregation node through the repaired multi-hop path.

[0034] Further, according to the aggregation node, perform spatio-temporal correlation fusion on the multi-source fault feature vectors, match the diagnosis results based on the pre-trained fault mode library, dynamically adjust the sampling rate, transmission power and filtering parameters according to the diagnosis results, and generate alarm events at the same time, including:

[0035] The aggregation node extracts the timestamps and device location tags of each fault feature vector, and based on the timestamps and location tags, aligns the signals from different devices through a sliding time window; after alignment, use the weighted evidence theory to fuse the fault features associated in the spatio-temporal dimension to generate a fused feature vector;

[0036] Input the fused feature vector into the pre-trained fault mode library for similarity matching, and according to the matching results, obtain the device fault type and probability value. When the probability value is greater than the preset diagnosis threshold, generate a deterministic diagnosis result;

[0037] According to the diagnosis results, dynamically adjust the parameters. If the fault probability in the diagnosis results is greater than the preset probability threshold, increase the signal sampling rate of the relevant device; if the current link quality index is less than the preset quality threshold, increase the node transmission power; if a frequency band interference signal is detected in the fused feature, adaptively enhance the suppression parameter of the filter;

[0038] Real-time monitor the voltage signal in the fused feature. When the detected transient voltage mutation amplitude is greater than the safety limit, immediately trigger the hierarchical alarm event generation mechanism.

[0039] Further, input the fused feature vector into the pre-trained fault mode library for similarity matching, and based on the matching results, obtain the device fault type and probability value. When the probability value is greater than the preset diagnosis threshold, generate a deterministic diagnosis result, including:

[0040] For each fault mode feature vector in the pre-trained fault mode library, calculate the Euclidean distance from it to the fused feature vector in the feature space respectively, generate a set of distance metric values including the distance values corresponding to all fault modes, and perform normalization processing on the set of distance metric values to obtain a set of normalized distance values;

[0041] Input the set of normalized distance values into a preset similarity converter, and dynamically generate the matching probability values of the fused feature vector corresponding to each fault mode through the built-in distance-probability inverse relationship in the converter;

[0042] Arrange the matching probability values in descending order numerically, and compare them item by item with the preset diagnosis threshold to determine the probability values greater than the threshold and the associated fault modes, forming a set of candidate fault modes;

[0043] Extract the fault mode ranked first in the set of candidate fault modes, and parse the device fault type code and the affected component identifier corresponding to the mode;

[0044] Based on the fault type code and the affected component identifier, and in combination with the preset confidence level grading rules, generate a deterministic diagnosis result including the specific fault location identifier, the fault type description, and the confidence level rating.

[0045] Further, based on the fault diagnosis result and the network status feedback, automatically reconstruct the multi-module network topology structure, dynamically adjust the working modes of each sensing unit, and realize the adaptive maintenance and adjustment of the multi-module network monitoring, including:

[0046] Based on the fault location identifier and the confidence level rating in the deterministic diagnosis result, and in combination with the network status feedback data collected in real time, including the node connectivity status, the link quality index, and the remaining energy of the node, generate a multi-module network topology reconstruction instruction;

[0047] According to the topology reconstruction instruction, re-divide the cluster areas and determine the cluster head nodes in the clustered multi-hop ad hoc network, isolate the fault areas with a confidence level rating exceeding the preset threshold, and at the same time construct redundant transmission paths bypassing the fault areas;

[0048] Based on the reconstructed topology structure and the influence range of the fault type in the deterministic diagnosis result, dynamically adjust the working modes of the sensing units in the associated areas;

[0049] Periodically verify the effectiveness of the reconstructed topology according to the trend of link quality change in the network status feedback, and implement the adaptive operating state of the monitoring network based on the verification results.

[0050] In a second aspect, a computing device includes:

[0051] One or more processors;

[0052] A storage device for storing one or more programs, which when executed by the one or more processors cause the one or more processors to implement the described system.

[0053] In a third aspect, a computer-readable storage medium stores a program that, when executed by a processor, implements the described system.

[0054] The above solution of the present invention has at least the following beneficial effects:

[0055] By deploying multiple sensors in a multi-modal manner to collect vibration and partial discharge signals, and using a chaotic enhanced weak signal detection device to capture the attractor phase diagram mutation and the time-domain waveform threshold, weak fault features can be accurately extracted, and early potential faults can be effectively identified. Compared with traditional monitoring methods, the probability of accidents is reduced. The dual-channel transmission adopts a wired and wireless hot standby transmission mechanism. When the signal attenuation reaches the preset threshold, the channel is automatically switched, and the relay node is triggered to perform gain compensation. The compensation gain value is dynamically calculated based on the signal attenuation prediction to ensure the continuity and accuracy of data transmission, and effectively avoid the monitoring failure caused by transmission interruption. The networking planning uses a cluster-based multi-hop ad hoc network architecture. The cluster head node realizes the autonomous repair of the wireless mesh network breakpoint based on the dynamic routing strategy. It can quickly respond to network topology changes, automatically bypass the fault area and reconstruct the transmission path, and the network recovery time is shortened to the second level, improving the network stability and fault tolerance. The fault diagnosis performs spatio-temporal correlation fusion on multi-source fault feature vectors, combines the pre-trained fault mode library to match the diagnosis results, and generates a deterministic diagnosis report through Euclidean distance measurement and probability conversion. The adaptive regulation is based on fault diagnosis and network status feedback, automatically reconstructs the multi-module networking topology structure, and dynamically adjusts the working mode of the sensing unit. It can quickly adapt to equipment failures and environmental changes, and improve the intelligent level and operation and maintenance efficiency of the power transmission and transformation construction monitoring system. Description of the Drawings

[0056] Figure 1 It is a schematic diagram of a weak signal multi-module networking adaptive monitoring system for power transmission and transformation construction provided by an embodiment of the present invention.

[0057] Figure 2It is a schematic flow chart of a process that inputs a fused feature vector into a pre-trained fault mode library for similarity matching, and obtains the device fault type and probability value according to the matching result. When the probability value is greater than a preset diagnosis threshold, a deterministic diagnosis result is generated. Detailed implementation manners

[0058] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art.

[0059] As Figure 1 shown, an embodiment of the present invention provides a multi-module network self-adaptive monitoring system for weak signals in power transmission and transformation construction, including:

[0060] A multi-modal module, which is used to deploy multi-modal sensing units at key monitoring points of power transmission and transformation equipment, collect vibration, partial discharge, leakage current and environmental parameter signals, and perform anti-interference processing on the signals through electromagnetic shielding enclosures and built-in harmonic filtering circuits to obtain original monitoring signal data;

[0061] A detection module, which is used to input the original monitoring signal data into a chaotic enhanced weak signal detection device, and extract fault feature vectors by capturing the mutation of the attractor phase diagram and the time-domain waveform threshold;

[0062] A transmission module, which is used to transmit the fault feature vectors through a wired and wireless dual-channel hot standby transmission device. When the signal attenuation reaches a preset threshold, the transmission device automatically switches the transmission channel, and at the same time triggers the relay node to perform gain compensation on the fault feature vectors;

[0063] A planning module, which is used to plan the transmission path for the fault feature vector data by adopting a clustered multi-hop ad hoc network architecture. The cluster head node realizes autonomous repair of the wireless mesh network breakpoint based on a dynamic routing strategy, and transmits the fault feature vector data to the aggregation node;

[0064] A diagnosis module, which is used to perform spatio-temporal correlation fusion on multi-source fault feature vectors according to the aggregation node, match the diagnosis result based on a pre-trained fault mode library, dynamically adjust the sampling rate, transmission power and filtering parameters according to the diagnosis result, and generate alarm events at the same time;

[0065] A regulation module, which is used to automatically reconstruct the multi-module network topology structure based on the fault diagnosis result and network status feedback, dynamically adjust the working mode of each sensing unit, and realize the adaptive maintenance and adjustment of multi-module network monitoring.

[0066] In the embodiments of the present invention, multi-modal sensing units are deployed at key monitoring points of power transmission and transformation equipment to achieve synchronous acquisition of multi-dimensional signals such as vibration, partial discharge, leakage current, and environmental parameters, comprehensively covering the key indicators of the equipment operation state and avoiding the limitations of single-signal monitoring. Anti-interference processing is carried out through electromagnetic shielding enclosures and built-in harmonic filtering circuits to reduce the influence of external electromagnetic interference and harmonic noise from the signal acquisition source, improving the purity and reliability of the original monitoring signal data. A chaotic enhanced weak signal detection device is adopted, taking advantage of the sensitivity of the chaotic system to weak signal changes to break through the accuracy bottleneck of traditional detection methods and being able to capture extremely small abnormal signal changes during equipment operation. By combining the attractor phase diagram mutation with the time-domain waveform threshold criterion, fault feature vectors can be effectively extracted, and even in the early fault stage of the equipment, weak fault signals can be accurately identified, improving the timeliness and accuracy of fault warning.

[0067] A wired and wireless dual-channel hot standby transmission mechanism is used to construct a redundant data transmission path. When the quality of the primary transmission channel drops to a preset threshold due to signal attenuation or environmental interference, the transmission device automatically and quickly switches to the standby channel to ensure uninterrupted transmission of fault feature vector data. At the same time, the relay node is triggered to perform gain compensation on the data, dynamically adjusting the transmission power according to the signal attenuation situation, effectively overcoming the distance limitation and attenuation problems during signal transmission, ensuring the stability and integrity of data transmission, and avoiding monitoring failure caused by transmission interruption. The transmission path is planned based on a cluster multi-hop self-organizing network architecture, fully adapting to the characteristics of wide distribution and complex environment of power transmission and transformation equipment, and flexibly constructing a network topology. The cluster head node adopts a dynamic routing strategy, real-time monitoring the connectivity status of network nodes. When a breakpoint in the wireless mesh network occurs, it can quickly and autonomously repair the path, automatically bypass the fault area and re-plan the data transmission route, reducing manual troubleshooting and intervention, improving the fault tolerance and self-healing ability of the network, and ensuring that the fault feature vector data can be efficiently and stably transmitted to the aggregation node.

[0068] Perform spatio-temporal correlation fusion on the multi-source fault feature vectors collected by the sink node, comprehensively consider the signal information in different devices, different times, and spatial dimensions, deeply explore the potential correlations between signals, and avoid misjudgments and missed judgments in single-signal diagnosis. Combine with a pre-trained fault mode library for matching diagnosis, which can quickly and accurately determine the type, probability, and location of equipment faults, and dynamically adjust the sampling rate, transmission power, and filtering parameters according to the diagnosis results to achieve intelligent optimal allocation of monitoring resources. At the same time, generate alarm events in a timely manner to help maintenance personnel quickly locate faults and improve the efficiency of fault handling. According to the fault diagnosis results and network status feedback, automatically reconstruct the multi-module network topology structure, isolate the fault area, re-divide the cluster area and determine the cluster head node, construct redundant transmission paths that bypass the fault area, and enhance the reliability and fault tolerance of the network. Dynamically adjust the working modes of each sensing unit, such as increasing the sensor sampling frequency in high-fault areas and optimizing the transmission strategy during network congestion, so that the entire multi-module network monitoring system can quickly adapt to equipment faults, environmental changes, and network status fluctuations, and achieve adaptive maintenance and adjustment.

[0069] In a preferred embodiment of the present invention, the original monitoring signal data is input into a chaotic enhanced weak signal detection device, and by capturing the sudden change of the attractor phase diagram and the time-domain waveform threshold, the fault feature vector can be extracted, which may include:

[0070] Normalize the original monitoring signal data to eliminate the dimension difference between different sensor signals, obtain the standardized signal data with a unified dimension, and use the standardized signal data as an external driving signal and input it into a chaotic detection structure with preset parameters;

[0071] Calculate the dynamic stability index of the chaotic detection structure in real time, judge the change trend of the current state according to the dynamic stability index, and track the morphological pattern of the attractor trajectory formed in the state space;

[0072] When it is detected that the attractor trajectory suddenly changes from a chaotic state to a regular periodic state, record the moment when the state transition occurs. At the same time, in the time-domain waveform of the standardized signal, analyze the signal amplitude and frequency changes corresponding before and after the transition moment;

[0073] Identify the waveform segments that meet the preset amplitude threshold and frequency change characteristics as signal segments with faults, and perform wavelet packet decomposition on the potential fault signal segments to obtain the energy distribution in different frequency bands;

[0074] Based on the energy proportion of each frequency band, construct a frequency-domain energy distribution vector for characterizing the signal characteristics, and use the frequency-domain energy distribution vector as the fault feature vector reflecting the operating state of the equipment.

[0075] In the embodiments of the present invention, the original monitoring signal data is collected by vibration, partial discharge, and leakage current multimodal sensors. Due to different physical quantities, there are dimensional differences among the signals (for example, vibration is acceleration m / s², and current is ampere A). For unified processing, a linear normalization method is adopted to process the original signals of each sensor for calculation. Suppose the minimum value of a sensor signal is , and the maximum value is . The normalization formula is , which maps the signal to the interval [0, 1]. Through calculation, the influence of dimension is eliminated, making the signals of different sensors comparable, generating standardized signal data, and serving as the external drive signal of the chaos detection structure. The chaos detection structure with preset parameters has a specific dynamic equation (such as the Lorenz equation and other chaos system equations), and its state changes with time. To judge the state change trend, the dynamic stability index is calculated in real time. Taking the Lyapunov exponent as an example, it describes the average exponential divergence or convergence rate of adjacent trajectories in the phase space. When calculating, the dynamic equation of the chaos detection structure is linearized, and the Jacobian matrix is calculated through iteration, and then the Lyapunov exponent is solved. If the exponent is positive, it indicates a chaotic state, and the adjacent trajectories are exponentially separated with time; if the exponent tends to zero, it changes to a stable state. At the same time, the attractor trajectory form is tracked in the state space. The attractor is the final form of the long-term evolution of the system, reflecting the stable state or change trend.

[0076] Continuously monitor the attractor trajectory. When it is detected that it suddenly changes from a chaotic state (the trajectory shows an irregular and complex winding shape) to a regular periodic state (the trajectory shows a simple shape of periodic repetition), record the moment when the state transition occurs. In the time-domain waveform of the standardized signal, focus on the signals before and after the transition moment, and analyze the amplitude and frequency changes. The root mean square value (RMS) is used for amplitude calculation, and the formula is , where is the number of sampling points, is the The signal values of the sampling points; for frequency analysis, the Fast Fourier Transform (FFT) is used to convert the time-domain signal into the frequency domain to obtain the frequency components and corresponding amplitudes of the signal. According to the preset amplitude threshold and frequency change characteristics (such as the amplitude suddenly rising above the threshold or the enhancement of specific frequency components), the signal segments with faults are identified. The potential fault signal segments are decomposed by wavelet packet decomposition, which is a method of decomposing the signal into different frequency bands. It gradually subdivides the signal into narrower frequency bands through multi-layer decomposition. For example, by performing three-layer wavelet packet decomposition, the signal can be decomposed into 8 different frequency bands. During the decomposition process, the convolution operation is performed between the wavelet function and the signal to obtain the coefficients of each frequency band, and then the energy distribution of different frequency bands is obtained. The energy is calculated as the sum of the squares of the frequency band coefficients. Based on the energy proportion of each frequency band, a frequency-domain energy distribution vector is constructed. Calculate the proportion of the energy of each frequency band in the total energy. Suppose there are frequency bands, the energy of the th frequency band is , the total energy is , then the proportion of the energy of the th frequency band is . Arrange these energy proportions in the order of frequency bands to form a frequency-domain energy distribution vector , and this vector serves as a fault feature vector reflecting the operating state of the device.

[0077] By normalizing the data, the dimension differences are eliminated to ensure the analysis of multi-modal signals on the same scale, avoiding analysis biases caused by different dimensions, and laying a foundation for accurately extracting fault features. The chaotic detection structure is sensitive to the changes of weak signals. By combining dynamic stability indicators such as Lyapunov exponents to analyze the sudden changes of attractor trajectories, it can capture the early fault signals of the device that are difficult to detect by traditional methods, improving the accuracy of fault feature extraction. Comprehensive fault judgment is carried out by combining the sudden changes of the attractor phase diagram with the amplitude and frequency changes of the time-domain waveform. The signal is analyzed from two dimensions of the phase space and the time domain, avoiding the limitations of single-dimensional judgment, improving the comprehensiveness and accuracy of fault judgment, and effectively reducing the probabilities of misjudgment and missed judgment. Wavelet packet decomposition can adaptively decompose the signal into different frequency bands. In view of the complex frequency components of the fault signals of power transmission and transformation equipment, it comprehensively obtains the energy distribution of each frequency band, accurately depicts the characteristics of the fault signals. The constructed frequency-domain energy distribution vector contains rich signal feature information, calculates the dynamic stability indicators in real time and tracks the attractor trajectories, and can timely detect the changes of fault signals to realize the real-time monitoring of faults. Through a rigorous calculation process and multi-step verification (such as amplitude threshold judgment, frequency analysis, etc.), it is ensured that the extracted fault feature vector is true and reliable, providing a solid monitoring guarantee for the stable operation of power transmission and transformation equipment, and reducing the economic losses and safety risks caused by equipment failures.

[0078] In a preferred embodiment of the present invention, the fault feature vector is transmitted through a wired and wireless dual-channel hot standby transmission device. When the signal attenuation reaches a preset threshold, the transmission device automatically switches the transmission channel, and at the same time triggers the relay node to perform gain compensation on the fault feature vector, which may include:

[0079] The main controller monitors the quality index of the currently activated primary transmission channel. If the primary channel is a wired channel, the bit error rate is monitored; if the primary channel is a wireless channel, the received signal strength indication value is monitored. When it is detected that the quality index of the primary channel is less than the corresponding preset threshold, it is determined that the channel signal attenuation has reached the preset threshold, and a channel switching instruction and a gain compensation trigger signal are generated;

[0080] Based on the channel switching instruction and the gain compensation trigger signal, the transmission device performs a switching operation from the current primary channel to the standby channel. At the same time, the gain compensation trigger signal activates the specified relay node on the transmission path to obtain the fault feature vector data packet whose transmission is interrupted due to the disconnection of the primary channel;

[0081] For the fault feature vector data packet, based on the signal attenuation data, predict the signal attenuation trajectory of the current transmission link, and according to the signal attenuation rate, dynamically calculate the compensation gain value required for power amplification of the fault feature vector data packet. Specifically, it includes: obtaining the channel attenuation data and environmental interference characteristics of the current transmission link, predicting the change trend of the signal attenuation amplitude with the transmission time and distance based on the correlation between the time characteristics of the channel attenuation data and the environmental interference characteristics; combining the signal attenuation change trend with the real-time collected link impedance fluctuation parameters and the instantaneous value of the electromagnetic interference intensity, dynamically generating the predicted value of the real-time signal attenuation rate at each discrete position point in the transmission path; according to the distribution of the real-time signal attenuation rate predicted value, dividing the transmission path of the fault feature vector data packet into continuous sections according to the attenuation rate mutation points, and calculating the corresponding theoretical signal attenuation degree at each section node; comparing the theoretical signal attenuation degree at the section node with the actual attenuation degree monitored by the sensor in real time, obtaining the attenuation degree deviation amount at each node; according to the attenuation degree deviation amount, combined with the influence degree ratio of the transmission distance on the signal attenuation in the corresponding section, dynamically generating the compensation gain value required for power amplification of the fault feature vector data packet.

[0082] In the embodiment of the present invention, the main controller monitors the fault feature vector data transmitted through the wired channel in real time. During the data transmission process, for every certain number of data packets transmitted (set as pieces), count the number of data packets with errors among them (set as pieces). The calculation formula for the bit error rate is . For example, if 1000 data packets are transmitted and 5 of them have errors, the bit error rate is = 0.5%. When the calculated bit error rate is less than the preset bit error rate threshold for the wired channel (such as 1%), it is determined that the signal attenuation of the wired channel has reached the preset threshold. For the wireless channel, the master controller continuously reads the received signal strength indication value. The RSSI value reflects the strength of the received signal, and the unit is usually dBm. For example, the RSSI value read at a certain moment is -70 dBm. Compare the real-time obtained RSSI value with the preset RSSI threshold for the wireless channel (such as -80 dBm). When the RSSI value is less than this threshold, it is determined that the signal attenuation of the wireless channel has reached the preset threshold. Once the quality indicators of the primary channel meet the above threshold conditions, the master controller immediately generates a channel switching instruction and a gain compensation trigger signal.

[0083] After the transmission device receives the channel switching instruction, it first stops the data transmission operation on the primary channel. Then, it performs the initialization settings for channel switching, including configuring the communication parameters of the standby channel (such as the frequency band of the wireless channel, the port protocol of the wired channel, etc.), and establishing a connection with the standby channel. After the connection is established, the transmission device switches the fault feature vector data packet to the standby channel for transmission. The gain compensation trigger signal is sent to the preset designated relay node on the transmission path. After receiving the trigger signal, the relay node starts the data reception program and begins to obtain the fault feature vector data packet whose transmission was interrupted due to the disconnection of the primary channel. The relay node will cache the received data packet and wait for gain compensation processing. The master controller collects the channel attenuation data of the current transmission link (such as the bit error rate or the change record of the RSSI value in different time periods) and the environmental interference characteristics (such as the distribution of surrounding electromagnetic devices, environmental factors affecting signal transmission such as weather conditions, etc.). By analyzing the time series characteristics of the channel attenuation data, for example, using the moving average method, calculate the average attenuation amount in the past period of time (set as ), and use this to predict the change trend of the signal attenuation amplitude with the transmission time. At the same time, correct the prediction result in combination with the environmental interference characteristics. For example, when it is detected that a strong electromagnetic device is turned on around, appropriately increase the predicted signal attenuation amplitude. In addition, consider the influence of the transmission distance on the signal attenuation, and adjust the attenuation prediction value according to the length of the transmission path, so as to obtain the comprehensive change trend of the signal attenuation amplitude with the transmission time and distance.

[0084] Combine the signal attenuation change trend with the real-time collected link impedance fluctuation parameters (reflecting the electrical characteristic changes of the transmission line) and the instantaneous value of electromagnetic interference intensity. For example, when the link impedance suddenly increases, it indicates that there may be a poor contact problem in the transmission line, which will lead to an increase in signal attenuation. At this time, according to the impedance change amplitude and data experience, the predicted signal attenuation rate is increased accordingly; when the instantaneous value of electromagnetic interference intensity increases, the predicted value of the signal attenuation rate is also adjusted. In this way, the real-time signal attenuation rate prediction values of each discrete position point in the transmission path are dynamically generated. According to the distribution of the real-time signal attenuation rate prediction values, the transmission path of the fault feature vector data packet is divided into continuous sections by the attenuation rate mutation points. For each segmentation node, calculate the theoretical signal attenuation degree, which can be comprehensively calculated according to factors such as the length of this section, the predicted attenuation rate, and the initial signal strength. At the same time, the sensor monitors the actual attenuation degree at each node in real time. Compare the theoretical signal attenuation degree of the segmentation node with the actual attenuation degree to obtain the attenuation degree deviation amount at each node. According to the attenuation degree deviation amount, combined with the influence degree ratio of the transmission distance of the corresponding section on signal attenuation (for example, the longer the distance, the higher the influence weight of the distance factor on attenuation), dynamically generate the compensation gain value required for power amplification of the fault feature vector data packet. For example, if the theoretical attenuation of a certain section is 10 dB, the actual attenuation is 12 dB, the deviation amount is 2 dB, and the distance of this section is relatively long, and the weight of the distance factor accounts for 60%, then according to the preset calculation rule, the final determined compensation gain value is 3 dB to ensure that the signal can maintain sufficient strength during transmission.

[0085] The dual-channel hot standby transmission mechanism provides a redundant transmission path for the fault feature vector data. When the primary channel deteriorates due to signal attenuation, equipment failure, or environmental interference, it can automatically and quickly switch to the standby channel to avoid data transmission interruption. Even in a complex and changing power transmission and transformation environment (such as strong electromagnetic interference, adverse weather affecting wireless signals, or physical damage to wired lines), it can ensure continuous and stable data transmission. By real-time monitoring the quality indicators of the primary channel and making dynamic judgments and decisions based on preset thresholds, it can intelligently adjust the transmission strategy according to the actual transmission situation. Whether it is the abnormal bit error rate of the wired channel or the insufficient signal strength of the wireless channel, it can trigger the corresponding processing mechanism in a timely manner. The gain compensation function of the relay node can dynamically adjust the compensation gain value according to the signal attenuation trajectory and the real-time attenuation rate, accurately matching the signal enhancement requirements of different transmission sections, effectively overcoming the distance limitation and attenuation problems during signal transmission, ensuring that the data still maintains good quality when transmitted over long distances and in complex environments, and improving the stability and effectiveness of transmission. It has the capabilities of automatic channel switching and gain compensation, without the need for manual detection and switching of the transmission channel, nor manual calculation and adjustment of the signal gain. When a problem occurs in the transmission channel, it can independently complete fault response and repair, reducing the workload and maintenance cost of the operation and maintenance personnel. At the same time, the fast fault response mechanism can shorten the data transmission interruption time and reduce the impact on the monitoring and fault diagnosis of power transmission and transformation equipment caused by data loss or transmission delay. In the power transmission and transformation environment, electromagnetic interference and line aging factors are likely to cause signal attenuation and transmission faults. The dual-channel transmission and dynamic gain compensation mechanism can effectively cope with these interferences and faults. The existence of the standby channel provides a fault tolerance space, while the gain compensation of the relay node enhances the resistance to signal attenuation. Even when a problem occurs in a local transmission link, it can maintain data transmission through channel switching and gain adjustment, ensuring the normal operation of the monitoring system and improving the anti-interference and fault tolerance performance of the weak signal multi-module network adaptive monitoring system for power transmission and transformation construction.

[0086] In a preferred embodiment of the present invention, for the fault feature vector data, a cluster-based multi-hop ad hoc network architecture is used to plan the transmission path. The cluster head node can autonomously repair the breakpoint of the wireless mesh network based on the dynamic routing strategy and transmit the fault feature vector data to the aggregation node, which may include:

[0087] Receive the fault feature vector data packet after gain compensation by the relay node, and based on the node geographical location information and real-time link quality data required for data packet transmission, establish a topological connection structure between the nodes within the cluster in the cluster-based multi-hop ad hoc network to form an initial transmission path;

[0088] Based on the topological connection structure, the cluster head node periodically detects the connectivity status of each neighbor node, and according to the detection results, updates the network routing table in real time. For each path recorded in the table, the performance is evaluated by synthesizing the signal strength and the remaining energy of the node.

[0089] When the connectivity status monitoring indicates that there is a breakpoint in the current transmission path, the cluster head node starts a dynamic routing strategy for autonomous repair.

[0090] After completing the path repair, the cluster head node relays the fault feature vector data to the sink node through the repaired multi-hop path.

[0091] In the embodiment of the present invention, the cluster head node receives the fault feature vector data packet after gain compensation from the relay node. The data packet contains the data itself and the attached meta-information, such as the geographical location coordinates (latitude and longitude information) of the sending node and the data packet generation time. The cluster head node parses this information and extracts the node geographical location information required for data packet transmission. The cluster head node sends a detection signal (such as a low-power broadcast signal) to each node in the cluster. After each node receives the detection signal, it calculates the link quality index between itself and the cluster head node according to the received signal strength and signal-to-noise ratio parameters. Common link quality index calculation methods are based on the received signal strength indication (RSSI). The RSSI value is mapped to a quality score range of 0-100. For example, when the RSSI value is between -50dBm and -70dBm, it corresponds to a score of 80-60. The node feeds back the calculated link quality score to the cluster head node.

[0092] Based on the extracted node geographical location information and link quality data, the cluster head node uses the method in graph theory to construct the topological connection structure between the nodes in the cluster. Each node is regarded as a vertex in the graph, the connection relationship between nodes is regarded as an edge, and the weight of the edge is determined by the link quality score. The higher the score, the lower the weight (indicating better link quality). Through the shortest path algorithm, the shortest path from each node to the cluster head node (based on the link quality weight) is calculated, thus forming an initial transmission path. For example, if there are three nodes H, M, and G in the cluster, the link quality score between H and the cluster head node is 80, M is 70, and G is 60. After algorithm calculation, the transmission paths of H-cluster head, M-H-cluster head, and G-M-H-cluster head may be determined. The cluster head node periodically (such as every 10 seconds) sends a connectivity detection data packet (containing simple identification information) to the neighbor nodes. After the neighbor nodes receive the detection data packet, they immediately reply with a confirmation data packet. The cluster head node judges the connectivity status of the neighbor nodes according to whether it receives the confirmation data packet. If it does not receive the confirmation within the specified time (such as 2 seconds), it is determined that the neighbor node is in a disconnected state.

[0093] According to the connectivity detection results, the cluster head node updates the network routing table in real time. The routing table records information about each reachable node, including the node ID, geographical location, link quality score with this node, and next-hop node information. When the connectivity state of a certain node is detected to change, the relevant information of the corresponding node is updated; if a new reachable node is found, a new record is added; if a node is disconnected, the record of this node is deleted or marked. For each path recorded in the routing table, the cluster head node comprehensively evaluates the performance based on the signal strength and the remaining energy of the node. The signal strength is represented by the link quality score; the remaining energy of the node is measured by the percentage of battery power regularly reported by the node. By adjusting the weights, the signal strength or the remaining energy of the node can be emphasized according to actual requirements. The higher the score of a path, the better its performance. During data transmission, if the cluster head node does not receive an acknowledgment reply from a certain next-hop node within the specified time after sending a data packet to it, and there is still no response after multiple retransmissions (such as 3 times), it is determined that there is a breakpoint in the current transmission path. When a breakpoint is detected, the cluster head node starts a dynamic route discovery strategy. First, nodes that are adjacent to the breakpoint node and in a connected state are selected from the network routing table as candidate nodes. Then, with the cluster head node as the starting point, the candidate nodes as intermediate nodes, and the sink node as the end point, the shortest path algorithm (based on the updated link quality weight and node remaining energy weight) is used again to calculate a new path from the cluster head node to the sink node. During the calculation process, the path segments that have been determined to be faulty are excluded, and paths with higher performance evaluation indicators are preferentially selected. After the path repair is completed, the cluster head node relays the fault feature vector data along the repaired multi-hop path. The cluster head node sends the data packet to the next-hop node of the new path. After receiving the data packet, this node forwards it to the next-hop according to the routing table information, and relays it in turn until the data packet is transmitted to the sink node. When each node forwards a data packet, it checks the integrity of the data packet (such as by checksum calculation). If the data packet is found to be damaged, it is discarded and the previous-hop node is notified to retransmit.

[0094] The clustered multi-hop ad-hoc network architecture effectively addresses the complex and changing network conditions in the power transmission and transformation environment by establishing an in-cluster topological connection structure, combined with a dynamic routing strategy and a path repair mechanism. When a breakpoint occurs in the transmission path, it can quickly and autonomously repair the path, avoiding data transmission interruption caused by a single-point failure. Even in a harsh environment (such as strong electromagnetic interference causing signal interruption of some nodes), it can maintain the network connectivity, ensure the stable transmission of fault feature vector data to the aggregation node, and improve the stability and reliability of the network. By comprehensively evaluating the path based on signal strength and node remaining energy, a path with good signal quality and sufficient node energy is selected when planning the transmission path, avoiding overusing nodes with low energy, and effectively balancing the load of each node in the network. The cluster head node periodically detects the connectivity status of neighbor nodes and updates the routing table in real time, enabling the network to quickly sense environmental changes (such as device position changes, new node addition, or old node withdrawal). The dynamic routing strategy ensures that the network can adjust the transmission path in a timely manner according to environmental changes, adapt to the complex and changing power transmission and transformation site environment without manual intervention, enhancing the self-adaptability and flexibility of the network, and meeting the monitoring requirements of wide device distribution and complex environment in power transmission and transformation construction. The clustered multi-hop transmission method decomposes long-distance data transmission into multiple short-distance relay transmissions, combines dynamic routing to select the final path, and can effectively reduce data transmission delay. At the same time, the fast response mechanism of path repair enables data to quickly switch to a new path for transmission in case of a fault, avoiding long waiting or data accumulation, ensuring the real-time nature of monitoring data, and providing strong support for the rapid fault diagnosis and timely treatment of power transmission and transformation equipment. The clustered multi-hop ad-hoc network architecture facilitates the expansion and upgrade of the network.

[0095] In a preferred embodiment of the present invention, for the multi-source fault feature vectors, spatio-temporal correlation fusion is performed according to the aggregation node, and the diagnostic results are matched based on a pre-trained fault mode library. According to the diagnostic results, the sampling rate, transmission power, and filtering parameters are dynamically adjusted, and at the same time, an alarm event is generated, which may include:

[0096] The aggregation node extracts the timestamps and device position tags of each fault feature vector, and based on the timestamps and position tags, aligns the signals from different devices through a sliding time window; after alignment, the weighted evidence theory is used to fuse the fault features associated in the spatio-temporal dimension to generate a fused feature vector;

[0097] The fused feature vector is input into the pre-trained fault mode library for similarity matching, and based on the matching results, the device fault type and probability value are obtained. When the probability value is greater than the preset diagnosis threshold, a deterministic diagnosis result is generated, which specifically includes: for each fault mode feature vector in the pre-trained fault mode library, calculate the Euclidean distance from it to the fused feature vector in the feature space respectively, generate a set of distance metric values including the distance values corresponding to all fault modes, and perform a normalization process on the set of distance metric values to obtain a set of normalized distance values; input the set of normalized distance values into a preset similarity converter, and dynamically generate the matching probability values of the fused feature vector corresponding to each fault mode through the built-in distance-probability inverse relationship of the converter.

[0098] Arrange the matching probability values in descending order of numerical value, and compare them item by item with the preset diagnosis threshold to determine the probability values greater than the threshold and the associated fault modes, forming a set of candidate fault modes; extract the fault mode ranked first in the set of candidate fault modes, and parse the device fault type code and the identifier of the affected component corresponding to the mode; based on the fault type code and the identifier of the affected component, combined with the preset confidence level grading rule, generate a deterministic diagnosis result including the specific fault location identifier, the fault type description, and the confidence level rating.

[0099] According to the diagnosis result, adjust the parameters dynamically. If the fault probability in the diagnosis result is greater than the preset probability threshold, increase the signal sampling rate of the relevant device; if the current link quality index is less than the preset quality threshold, increase the node transmission power; if a frequency band interference signal is detected in the fused features, adaptively enhance the suppression parameter of the filter.

[0100] Monitor the voltage signal in the fused features in real time. When the monitored transient voltage mutation amplitude is greater than the safety limit, immediately trigger the hierarchical alarm event generation mechanism.

[0101] In the embodiment of the present invention, the aggregation node receives the fault feature vectors from different devices, and each vector records the signal acquisition time (accurate to milliseconds) and the device installation location (including the device number and geographical coordinates). The aggregation node analyzes the data format and extracts these time and location information separately to prepare for subsequent unified processing. First, set the length of the time window (such as 500 milliseconds) and the interval of each slide (such as 100 milliseconds). Based on the signal acquisition time, regard the signals collected by different devices within the same time window as the data at the same moment. If the acquisition time of a device signal is inconsistent with the start time of the window, map the signal to the appropriate time point within the window through a reasonable estimation method to achieve the alignment of signals from different devices in time. At the same time, according to the device location information, associate the signals of devices with similar locations to ensure the consistency in the time and space dimensions.

[0102] According to the importance of the equipment (for example, the equipment on critical transmission lines is more important) and the accuracy of signal acquisition (the signals collected by high-precision sensors are more reliable), a weight is assigned to the fault characteristics of each equipment. The sum of the weights of all equipment is equal to 1. Then, the values of each equipment in each characteristic dimension (such as vibration frequency, partial discharge intensity, etc.) are added according to the weights to obtain the fused characteristic values. These fused characteristic values are combined together to form a fused characteristic vector. The fused characteristic vector is compared with each fault mode characteristic vector in the pre-trained fault mode library to judge their similarity. The comparison method is to calculate the distance between the two. The closer the distance, the higher the similarity. First, adjust the calculated distances so that all distances are in the range of 0 to 1. Then, according to the corresponding relationship between the distance and the probability, the closer the fault mode is to the calculated distance, the higher the matching probability value is assigned to it.

[0103] Sort the calculated matching probability values from largest to smallest and compare them with the set diagnostic criteria (such as 0.7) in turn. Screen out the fault modes with probability values exceeding the standard to form a candidate list. From this list, select the fault mode with the highest probability value and analyze the corresponding equipment fault type and affected components. Then, combined with the pre-set credibility criteria (such as a probability of 0.7 - 0.8 with low credibility, 0.8 - 0.9 with medium credibility, and above 0.9 with high credibility), generate a diagnostic result including the specific location of the fault, the description of the fault type, and the credibility rating. Compare the fault probability in the diagnostic result with the set probability standard (such as 0.6). If the fault probability exceeds the standard, it means that the equipment fault risk is relatively high and more detailed signal information needs to be obtained. At this time, increase the signal sampling frequency of the relevant equipment by a certain proportion and send the new sampling frequency setting to the equipment. After receiving it, the equipment will automatically adjust the sampling frequency. Monitor the quality of the current signal transmission in real time (comprehensively evaluated by indicators such as bit error rate, signal strength, etc.) and compare it with the set quality standard (such as 80 points). If the transmission quality is lower than the standard, it means that the signal transmission effect is not good. According to the degree of quality decline, increase the transmission power of the node proportionally and send an adjustment instruction to the node. The node will enhance the signal transmission power according to the instruction.

[0104] Analyze the fused feature vector to check for the presence of frequency band interference signals (set the interference frequency band range and determine whether the signal strength within this range is abnormal). If interference is detected, according to the severity of the interference, proportionally enhance the filter's suppression ability for the interference frequency band and send the new parameter settings to the filter, which will filter out more interference signals accordingly. Extract the voltage signal part in the fused feature vector in real time, compare the voltage values at adjacent times, and calculate the change amplitude of the voltage signal. Compare the calculated voltage change amplitude with the set safety standard (such as 20% of the rated voltage). If the change amplitude exceeds the standard, trigger different levels of alarms according to the exceeding degree.

[0105] By simultaneously considering the time and space information of multi-device signals, the limitations of diagnosing solely relying on a single device signal are avoided, improving the accuracy and comprehensiveness of fault diagnosis. The method of fusing by assigning weights according to device importance and signal quality makes the diagnostic results more reliable. Automatically adjusting the sampling rate, transmission power, and filtering parameters according to the diagnostic results realizes the reasonable allocation of resources. Automatically increasing the signal acquisition frequency when the device failure risk is high, enhancing the transmission power when the signal transmission quality is poor, and optimizing the filtering effect when interference occurs, can automatically adjust according to the actual operating conditions, improving the monitoring efficiency and data quality, and avoiding resource waste. Real-time monitoring of the voltage signal in the fused feature can quickly detect abnormal changes in voltage and issue different levels of alarms according to the change amplitude. This hierarchical alarm method allows the operation and maintenance personnel to quickly judge the severity of the fault, prioritize the handling of high-risk faults, shorten the fault handling time, reduce the risk of equipment damage or power grid accidents caused by voltage abnormalities, and ensure the safe and stable operation of the power transmission and transformation system. The combination of multi-source data fusion and dynamic parameter adjustment can adapt to the complex and changeable operating environment of power transmission and transformation. Whether it is equipment failure, network fluctuation, or environmental interference, it can maintain stable operation through its own diagnosis and adjustment mechanism, reduce the impact of external factors on the monitoring and diagnostic results, enhance the anti-interference ability and stability, and improve the reliability and availability of the entire monitoring system.

[0106] In a preferred embodiment of the present invention, based on the fault diagnosis result and network status feedback, automatically reconstruct the multi-module network topology structure, dynamically adjust the working modes of each sensing unit, and realize the adaptive maintenance and adjustment of multi-module network monitoring, which may include:

[0107] Based on the fault location identifier and confidence rating in the deterministic diagnosis result, combined with the network status feedback data collected in real time, including node connectivity status, link quality index, and node remaining energy, generate a multi-module network topology reconstruction instruction;

[0108] According to the topology reconstruction instruction, re-partition the cluster area and determine the cluster head nodes in the clustered multi-hop ad hoc network, isolate the fault area with a confidence rating exceeding the preset threshold, and at the same time construct a redundant transmission path around the fault area;

[0109] Based on the reconstructed topology structure and the influence range of the fault type in the deterministic diagnosis result, dynamically adjust the working mode of the sensing units in the associated area;

[0110] According to the change trend of the link quality in the network status feedback, periodically verify the effectiveness of the reconstructed topology, and realize the adaptive operating state of the monitoring network based on the verification result.

[0111] In the embodiment of the present invention, obtain the deterministic diagnosis result output by the fault diagnosis module, and extract the fault location identifier (such as device number, geographical coordinates) and the confidence rating (indicating the reliability of the diagnosis result, with a value range of 0-1) therein. At the same time, real-time collect the network status feedback data, including the connectivity status of each node (judge whether the node is online by sending a probe signal and receiving a reply), the link quality index (calculated by integrating parameters such as the bit error rate, signal strength, signal-to-noise ratio, etc., and the higher the value, the better the link quality), and the remaining energy of the node (determined by the node reporting the battery power percentage or the remaining capacity of the power supply regularly).

[0112] If the fault confidence rating in a certain area exceeds the preset threshold (such as 0.8), and the connectivity status of the nodes in this area deteriorates, the link quality index drops to a certain extent (such as below 60 points), and the remaining energy of the nodes is low (such as less than 30%), then it is determined that this area needs topology reconstruction. According to these conditions, generate a multi-module network topology reconstruction instruction containing information such as the fault area range and the reconstruction target (such as isolating the fault area, optimizing the transmission path).

[0113] Based on the clustered multi-hop ad-hoc network architecture, with the fault area as the center and combined with the geographical location distribution of network nodes, the cluster area is re-planned. The distance metric method is used to calculate the distance from each node to the boundary of the fault area, and the nodes that are far from the fault area and have good connectivity are divided into new clusters. For example, a distance threshold (such as 50 meters) is set, and the nodes that are more than the threshold away from the fault area and whose link quality scores between nodes are higher than a certain standard (such as 70 points) are grouped into the same cluster. For each newly divided cluster, a cluster head node is elected. The election criteria comprehensively consider factors such as the remaining energy of the node (the higher the energy, the higher the priority), processing capacity (such as CPU performance, memory capacity), link quality (the average link quality score with other nodes in the cluster), etc. The cluster head node is determined through a voting mechanism or a weighted scoring method. For example, the remaining energy of the node accounts for 40%, the processing capacity accounts for 30%, and the link quality accounts for 30%. Calculate the comprehensive score of each node, and the node with the highest score is elected as the cluster head node. For the fault area with a confidence rating exceeding the preset threshold, cut off the connection link between this area and other normal areas to achieve physical isolation. At the same time, based on the remaining normal nodes, use the path search algorithm in graph theory (such as a variant of Dijkstra's algorithm) to build multiple redundant transmission paths from the source node (a normal node close to the fault area) to the target node (the aggregation node or other key nodes), avoiding the fault area. During the construction process, paths with good link quality and high remaining energy of nodes are preferred.

[0114] Determine its scope of influence according to the fault type in the definite diagnosis result. For example, if the fault type is partial discharge fault, analyze the equipment and surrounding areas that may be affected by this fault; if it is a network link fault, determine the affected transmission path and related nodes. For the sensing units in the associated area, adjust the working mode according to the degree of fault influence and network status. If the fault influence scope is large, increase the sampling frequency of the sensing unit (such as from once per minute to five times per minute) to collect data more frequently; if the network link quality is poor, reduce the data transmission frequency of the sensing unit (such as from once per second to once every five seconds) to reduce the data transmission pressure. For the sensing units close to the fault area but not directly affected, turn on the early warning mode to enhance the monitoring sensitivity to abnormal signals. According to the change trend of link quality in the network status feedback, collect data such as bit error rate, signal strength, and transmission delay of each transmission path periodically (such as every 10 minutes). Calculate the comprehensive performance index of each path. The comprehensive performance index is weighted by 30% of the bit error rate, 40% of the signal strength, and 30% of the transmission delay. Compare the calculated comprehensive performance index with the preset performance standard (such as the comprehensive score needs to reach more than 75 points). If the comprehensive performance index of a certain path or the entire reconstructed topology does not meet the standard, it indicates that there is a problem with the reconstructed topology. At this time, re-evaluate the network status and fault situation, and execute the topology reconstruction process again, adjusting the cluster area division, cluster head node selection, or redundant path construction strategy until the reconstructed topology meets the performance requirements and realizes the adaptive operation state of the monitoring network.

[0115] By automatically reconstructing the multi-module network topology, the fault area is isolated in time and redundant transmission paths are constructed, effectively avoiding the spread of faults and network paralysis. Even if there are emergencies such as equipment failure and link interruption in the power transmission and transformation environment, the network structure can be quickly adjusted to maintain the continuity of data transmission, ensuring the stable collection and transmission of monitoring data. The working mode of the sensor unit is dynamically adjusted to reasonably allocate resources according to the fault impact range and network status. The sampling frequency is increased in fault-prone areas to ensure the acquisition of detailed fault information; the transmission frequency is reduced when the network is congested or the link quality is poor to reduce the network burden. This intelligent resource allocation method not only improves the effectiveness of monitoring data, but also extends the service life of sensor units and network nodes, reducing overall energy consumption and operation and maintenance costs. Based on real-time network status feedback and fault diagnosis results, it can quickly perceive changes in the power transmission and transformation environment (such as equipment failure, environmental interference leading to decreased link quality) and respond quickly. By periodically verifying the effectiveness of the reconstructed topology, the monitoring network always maintains the final operating state, enhances the adaptive ability to complex and changing environments, and meets the needs of long-term stable operation of power transmission and transformation construction. The isolation mechanism for high-confidence fault areas effectively prevents the spread of faults and reduces the scope of the fault's impact on the entire network. At the same time, the construction of redundant transmission paths ensures that when some links fail, data can still be transmitted through other paths, shortening the network recovery time. The adaptive maintenance and adjustment mechanism realizes the automation and intelligence of network management, reducing the workload of manual intervention and manual configuration. Operation and maintenance personnel do not need to monitor the network status in real time and manually adjust the topology structure. They only need to perform targeted processing based on the generated alarms and diagnostic results, which improves the efficiency and accuracy of operation and maintenance management and promotes the development of power transmission and transformation construction and operation and maintenance in the direction of intelligence.

[0116] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0117] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0118] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A weak signal multi-module networking adaptive monitoring system for power transmission and transformation construction, characterized in that, Including: A multimodal module, which is used to deploy multimodal sensing units at key monitoring points of power transmission and transformation equipment, collect vibration, partial discharge, leakage current and environmental parameter signals, and process the signals through electromagnetic shielding enclosures and built-in harmonic filtering circuits to obtain original monitoring signal data; A detection module, which is used to input the original monitoring signal data into a chaotic enhanced weak signal detection device, and extract fault feature vectors by capturing the mutation of the attractor phase diagram and the time-domain waveform threshold; A transmission module, which is used to transmit the fault feature vectors through a wired and wireless dual-channel hot standby transmission device. When the signal attenuation reaches a preset threshold, the transmission device automatically switches the transmission channel, and at the same time triggers the relay node to perform gain compensation on the fault feature vectors; A planning module, which is used to plan the transmission path for the fault feature vector data using a clustered multi-hop ad hoc network architecture. The cluster head node realizes the autonomous repair of the wireless mesh network breakpoint based on the dynamic routing strategy, and transmits the fault feature vector data to the aggregation node; A diagnosis module, which is used to perform spatio-temporal correlation fusion on multi-source fault feature vectors according to the aggregation node, match the diagnosis results based on a pre-trained fault mode library, dynamically adjust the sampling rate, transmission power and filtering parameters according to the diagnosis results, and generate alarm events at the same time; A regulation module, which is used to automatically reconstruct the multi-module network topology structure based on the fault diagnosis results and network status feedback, dynamically adjust the working modes of each sensing unit, and realize the adaptive maintenance and adjustment of multi-module network monitoring.

2. The weak signal multi-module networking adaptive monitoring system for power transmission and transformation construction according to claim 1, characterized in that Inputting the original monitoring signal data into a chaotic enhanced weak signal detection device, and extracting fault feature vectors by capturing the mutation of the attractor phase diagram and the time-domain waveform threshold, including: Performing normalization processing on the original monitoring signal data to eliminate the dimension difference between different sensor signals, obtaining standardized signal data with a unified dimension, and using the standardized signal data as an external driving signal to input it into a chaotic detection structure with preset parameters; Calculating the dynamic stability index of the chaotic detection structure in real time, judging the change trend of the current state according to the dynamic stability index, and tracking the morphological pattern of the attractor trajectory formed in the state space; When it is detected that the attractor trajectory suddenly changes from a chaotic state to a regular periodic state, record the moment when the state transition occurs. At the same time, in the time-domain waveform of the standardized signal, analyze the signal amplitude and frequency changes corresponding to before and after the transition moment; Identifying the waveform segments that meet the preset amplitude threshold and frequency change characteristics as signal segments with faults, and performing wavelet packet decomposition on the potential fault signal segments to obtain the energy distribution in different frequency bands; Based on the energy proportion of each frequency band, constructing a frequency-domain energy distribution vector for characterizing the signal characteristics, and using the frequency-domain energy distribution vector as a fault feature vector reflecting the operating state of the equipment.

3. The weak signal multi-module networking adaptive monitoring system for power transmission and transformation construction according to claim 2, characterized in that, Transmitting the fault feature vectors through a wired and wireless dual-channel hot standby transmission device. When the signal attenuation reaches a preset threshold, the transmission device automatically switches the transmission channel, and at the same time triggers the relay node to perform gain compensation on the fault feature vectors, including: The main controller monitors the quality indicators of the currently active primary transmission channel. If the primary channel is a wired channel, it monitors the bit error rate; if the primary channel is a wireless channel, it monitors the received signal strength indicator value. When it is detected that the quality indicator of the primary channel is less than the corresponding preset threshold, it is determined that the channel signal attenuation has reached the preset threshold, and a channel switching instruction and a gain compensation trigger signal are generated. Based on the channel switching instruction and the gain compensation trigger signal, the transmission device performs the switching operation from the current primary channel to the backup channel. Meanwhile, the gain compensation trigger signal activates the specified relay node on the transmission path to obtain the fault feature vector data packet whose transmission is interrupted due to the disconnection of the primary channel. For the fault feature vector data packet, based on the signal attenuation data, predict the signal attenuation trajectory of the current transmission link, and dynamically calculate the compensation gain value required for power amplification of the fault feature vector data packet according to the signal attenuation rate.

4. The weak signal multi-module networking adaptive monitoring system for power transmission and transformation construction according to claim 3, characterized in that, For the fault feature vector data packet, based on the signal attenuation data, predict the signal attenuation trajectory of the current transmission link, and dynamically calculate the compensation gain value required for power amplification of the fault feature vector data packet, including: Obtain the channel attenuation data and environmental interference characteristics of the current transmission link, and predict the change trend of the signal attenuation amplitude with the transmission time and distance based on the correlation between the time characteristics of the channel attenuation data and the environmental interference characteristics. Combine the signal attenuation change trend with the real-time collected link impedance fluctuation parameters and the instantaneous value of the electromagnetic interference intensity to dynamically generate the predicted value of the real-time signal attenuation rate at each discrete position point in the transmission path. According to the distribution of the real-time signal attenuation rate prediction values, divide the transmission path of the fault feature vector data packet into continuous sections at the attenuation rate mutation points, and calculate the corresponding theoretical signal attenuation degree at each section node. Compare the theoretical signal attenuation degree of the section node with the actual attenuation degree monitored by the sensor in real time to obtain the attenuation degree deviation amount at each node. According to the attenuation degree deviation amount, combined with the influence degree ratio of the transmission distance on the signal attenuation in the corresponding section, dynamically generate the compensation gain value required for power amplification of the fault feature vector data packet.

5. The weak signal multi-module networking adaptive monitoring system for power transmission and transformation construction according to claim 4, characterized in that For the fault feature vector data, use a clustered multi-hop ad hoc network architecture to plan the transmission path. The cluster head node realizes the autonomous repair of the wireless mesh network breakpoint based on the dynamic routing strategy, and transmits the fault feature vector data to the aggregation node, including: Receive the fault feature vector data packet after gain compensation by the relay node, and establish the topological connection structure between the nodes within the cluster in the clustered multi-hop ad hoc network based on the node geographical location information and real-time link quality data required for data packet transmission to form the initial transmission path. Based on the topological connection structure, the cluster head node periodically detects the connectivity status of each neighbor node, and according to the detection result, updates the network routing table in real time, and performs performance evaluation on each path recorded in the table by comprehensively considering the signal strength and the remaining energy of the node. When the connectivity status monitoring indicates that there is a breakpoint in the current transmission path, the cluster head node starts the dynamic routing strategy for autonomous repair. After the path repair is completed, the cluster head node relays the fault feature vector data to the aggregation node through the repaired multi-hop path.

6. The weak signal multi-module networking adaptive monitoring system for power transmission and transformation construction according to claim 5, characterized in that, Based on the aggregation node, perform spatio-temporal correlation fusion on the multi-source fault feature vectors, match the diagnostic results based on the pre-trained fault mode library, dynamically adjust the sampling rate, transmission power, and filtering parameters according to the diagnostic results, and simultaneously generate alarm events, including: The aggregation node extracts the timestamps and device location tags of each fault feature vector, and based on the timestamps and location tags, aligns the signals from different devices through a sliding time window; after alignment, the weighted evidence theory is used to fuse the fault features associated in the spatio-temporal dimensions to generate a fused feature vector; Input the fused feature vector into the pre-trained fault mode library for similarity matching, and based on the matching results, obtain the device fault type and probability value. When the probability value is greater than the preset diagnostic threshold, generate a deterministic diagnostic result; According to the diagnostic results, dynamically adjust the parameters. If the fault probability in the diagnostic result is greater than the preset probability threshold, increase the signal sampling rate of the relevant device; if the current link quality index is less than the preset quality threshold, increase the node transmission power; if a frequency band interference signal is detected in the fused features, adaptively enhance the suppression parameter of the filter; Monitor the voltage signal in the fused features in real time. When the monitored transient voltage mutation amplitude is greater than the safety limit, immediately trigger the hierarchical alarm event generation mechanism.

7. The multi-module network adaptive monitoring system for weak signals in power transmission and transformation construction according to claim 6, wherein Input the fused feature vector into the pre-trained fault mode library for similarity matching, and based on the matching results, obtain the device fault type and probability value. When the probability value is greater than the preset diagnostic threshold, generate a deterministic diagnostic result, including: For each fault mode feature vector in the pre-trained fault mode library, calculate the Euclidean distance in the feature space with the fused feature vector respectively, generate a set of distance metric values including the distance values corresponding to all fault modes, and perform normalization processing on the set of distance metric values to obtain a set of normalized distance values; Input the set of normalized distance values into the preset similarity converter, and dynamically generate the matching probability values of the fused feature vector corresponding to each fault mode through the built-in distance-probability inverse relationship in the converter; Arrange the matching probability values in descending order of numerical value, and compare them item by item with the preset diagnostic threshold to determine the probability values greater than the threshold and the associated fault modes, forming a candidate fault mode set; Extract the fault mode ranked first in the candidate fault mode set, and analyze the device fault type code and the identification of the affected components corresponding to the mode; Based on the fault type code and the identification of the affected components, combined with the preset confidence level grading rules, generate a deterministic diagnostic result including the specific fault location identification, fault type description, and confidence level rating.

8. The weak signal multi-module networking adaptive monitoring system for power transmission and transformation construction according to claim 7, characterized in that, Based on the fault diagnosis results and network status feedback, automatically reconstruct the multi-module network topology structure, dynamically adjust the working modes of each sensing unit, and realize the adaptive maintenance and adjustment of multi-module network monitoring, including: Based on the fault location identification and confidence level rating in the deterministic diagnostic result, combined with the network status feedback data collected in real time, including node connectivity status, link quality index, and node remaining energy, generate a multi-module network topology reconstruction instruction; According to the topology reconstruction instruction, re-divide the cluster area and determine the cluster head nodes in the clustered multi-hop ad hoc network, isolate the fault areas with confidence ratings exceeding the preset threshold, and simultaneously construct redundant transmission paths that bypass the fault areas; Based on the topology structure after reconstruction and the influence scope of the fault types in the deterministic diagnosis results, dynamically adjust the working modes of the sensing units in the associated areas; According to the link quality change trend in the network status feedback, periodically verify the effectiveness of the reconstructed topology, and based on the verification results, achieve the adaptive operating state of the monitoring network.

9. A computing device, characterized in that, Comprising: One or more processors; A storage device for storing one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the system according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, A program is stored in the computer-readable storage medium, and when the program is executed by a processor, the system according to any one of claims 1 to 8 is implemented.

Images