Data collection method for power equipment status monitoring based on Beidou navigation system

Abstract

The present invention relates to the field of power equipment monitoring technology, and specifically to a power equipment status monitoring data collection method based on the Beidou navigation system, comprising the following steps: S1, equipment status data collection: real-time collection of equipment status data; S2, collaborative learning mechanism between equipment: establishment of collaborative learning mechanism between power equipment; S3, data fusion and standardization: standardization and fusion of status data collected by each device; S4, equipment status correlation analysis: identification of status correlation and potential failure modes between equipment; S5, status evaluation and feedback mechanism: evaluation of equipment status according to the analysis results of equipment status correlation, and feedback of the evaluation results to the collaborative learning mechanism between equipment. The present invention can dynamically optimize equipment monitoring and maintenance strategies, improve the accuracy and efficiency of equipment health management, reduce the risk of fault propagation, and extend the service life of equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment monitoring, and in particular to a method for collecting power equipment status monitoring data based on a Beidou navigation system. Background Art

[0002] With the continuous expansion of the scale of power systems and the improvement of their intelligence, the operating status monitoring and fault warning of power equipment have become key links in the safe operation of power systems. To ensure the stable operation of power equipment, existing monitoring methods rely on sensors to collect real-time data on equipment status and predict potential faults through data analysis technology. These technologies have improved the reliability of equipment operation to a certain extent, reduced the frequency of sudden failures, and optimized equipment maintenance strategies. However, with the complexity of power network structure and the increase in the number of equipment, traditional monitoring methods have become difficult to fully and timely capture the correlation between equipment and potential failure modes.

[0003] Existing technologies mainly rely on the mode of single-device monitoring and independent analysis, which makes it difficult to achieve efficient information sharing and collaborative work between devices. This method cannot fully utilize the correlation data between devices, resulting in insufficient accuracy and response speed of fault prediction. At the same time, existing equipment monitoring is mostly focused on the collection and analysis of single or a few status data, and cannot form a comprehensive equipment status assessment. It lacks dynamic feedback and optimization mechanisms, and is difficult to deal with potential correlated failures between multiple devices in complex power systems. In addition, data transmission and security issues also limit information sharing and the transfer of fault experience between devices. Existing methods also have shortcomings in the integration and analysis of multi-dimensional status data.

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a method for collecting data on the status monitoring of power equipment based on the Beidou navigation system, which can not only improve the accuracy of fault prediction and the scientific nature of maintenance decisions, but also reduce the failure risk of equipment through an optimized feedback mechanism, thereby improving the overall safety and operation efficiency of the power system. Summary of the invention

[0005] The present invention provides a method for collecting data for monitoring the state of electric power equipment based on the Beidou navigation system.

[0006] The method for collecting data of power equipment status monitoring based on Beidou navigation system includes the following steps:

[0007] S1, equipment status data collection: locate power equipment through the Beidou navigation system, and use a variety of sensors to collect equipment status data in real time, including temperature, vibration, humidity and equipment load;

[0008] S2, collaborative learning mechanism between devices: Establish a collaborative learning mechanism between power devices, allowing adjacent devices to share monitoring data and fault experience, thereby forming a collective knowledge base;

[0009] S3, data fusion and standardization: standardize and integrate the status data collected by each device to form a unified data set;

[0010] S4, equipment status correlation analysis: Analyze the fused status data to identify the status correlation and potential failure modes between devices, including:

[0011] S41, feature extraction: extract features from the fused status data to identify key indicators, including temperature fluctuations, vibration frequency, humidity levels, and equipment load changes;

[0012] S42, correlation modeling: Based on the identified key indicators, establish a state correlation model between devices, and identify potential correlations and mutual influences by analyzing the state data of different devices;

[0013] S43, Fault mode identification: Based on the analysis results of the state association model, identify potential fault modes, evaluate the fault propagation risk between devices, and generate corresponding fault warnings;

[0014] S5, status evaluation and feedback mechanism: Based on the analysis results associated with the equipment status, the equipment status is evaluated and the evaluation results are fed back to the collaborative learning mechanism between devices to continuously optimize the equipment monitoring and maintenance strategy.

[0015] Optionally, the device status data collection in S1 includes:

[0016] S11, equipment positioning: locate the power equipment through the Beidou navigation system to obtain the geographical coordinate information of the equipment;

[0017] S12, sensor installation: installing various types of sensors on the power equipment, including temperature sensors, vibration sensors, humidity sensors, and load sensors;

[0018] S13, status data collection: collect the status data of the equipment in real time through various types of installed sensors, including temperature, vibration, humidity and equipment load.

[0019] Optionally, the inter-device collaborative learning mechanism in S2 includes:

[0020] S21, Data Sharing Agreement Establishment: Develop a data sharing agreement between power equipment to clarify the data transmission format, frequency and security measures;

[0021] S22, collective knowledge base construction: Based on shared status data and experience, a collective knowledge base is constructed to organize and analyze historical fault cases and monitoring data to form a knowledge system for reference by all devices.

[0022] Optionally, the establishment of the data sharing agreement in S21 includes:

[0023] S211, formulate data transmission format: formulate a unified data transmission format for monitoring data of power equipment, including data type, data structure and encoding method;

[0024] S212, determine the data transmission frequency: determine the frequency of data transmission, and set two modes: scheduled transmission and event-triggered transmission;

[0025] S213, Establish security measures: Establish security measures for data sharing, including data encryption, identity authentication, and access control.

[0026] Optionally, the collective knowledge base construction in S22 includes:

[0027] S221, Data Collection and Collation: Based on the status data and fault experience shared among power equipment, collect and collate historical fault cases and related monitoring data;

[0028] S222, Data Analysis: Analyze the sorted historical failure cases and related monitoring data to identify failure modes, influencing factors and early warning indicators, including:

[0029] S2221, Identify failure modes: classify different failure modes by calculating failure rate and mean repair time as the quantitative basis of failure modes;

[0030] S2222, Identify influencing factors: Identify the relationship between the status data and the occurrence of the fault by calculating the Pearson correlation coefficient, and determine the influencing factors;

[0031] S2223, Identify early warning indicators: Set early warning indicators based on the historical distribution and trends of monitoring data;

[0032] S223, knowledge system construction: store the analysis results and troubleshooting experience in a structured manner in the knowledge base to form a knowledge system for reference by all devices.

[0033] Optionally, the data fusion and standardization in S3 includes:

[0034] S31, data standardization processing: standardizing the status data (including temperature, vibration, humidity, and load) collected from each power device;

[0035] S32, data fusion: use weighted average method to fuse the standardized state data.

[0036] Optionally, the feature extraction in S41 includes:

[0037] S411, Temperature Fluctuation: Extract temperature fluctuations by calculating the mean square error to quantify the temperature volatility;

[0038] S412, vibration frequency: analyze by Fourier transform (FFT) to extract the main frequency components in the spectrum;

[0039] S413, humidity level: extracting humidity level by calculating the rate of change of humidity data;

[0040] S414, Equipment load change: Use linear regression model to analyze the change trend of equipment load.

[0041] Optionally, the state association model in S42 adopts a graph neural network (GCN) model, and the graph neural network (GCN) model includes:

[0042] S421, equipment status feature initialization: In the power equipment status monitoring, the multi-dimensional status data of each power equipment is initialized, including temperature, vibration, humidity, equipment load, and defined as feature vector Represents the initial state of the device;

[0043] S422, constructing an adjacency matrix: the adjacency matrix is ​​weighted according to the physical connection of the devices, the state similarity, and the relevance of the historical state data;

[0044] S423, feature propagation and aggregation: In the propagation process of each layer, the state features of the power equipment are propagated through the adjacency matrix and weightedly aggregated with the state data of the surrounding power equipment. At the same time, bias terms and activation functions are added to enhance the nonlinear representation ability of the model;

[0045] S424, multi-scale fusion of device status: a multi-scale feature fusion method is used to cascade the features of different convolutional layers to form a device status representation;

[0046] S425, state association output: Through the last layer of graph convolution, combined with multi-scale features, the state association score between devices is output to identify the potential correlation and mutual influence between devices.

[0047] Optionally, the fault mode identification in S43 includes:

[0048] S431, fault correlation strength calculation: using the output result Y of the state correlation model iTo represent the state association score of device i, the fault propagation risk R between devices is calculated based on the state association score ij ;

[0049] S432, Fault propagation probability assessment: Use Bayesian network to evaluate the fault propagation probability P(F) between each power equipment. j |F i ) for assessment;

[0050] S432, Fault warning generation: Based on the calculated fault propagation risk R ij and the fault propagation probability P(F j |F i ), when the fault propagation risk exceeds the set propagation risk threshold R threshold And the fault propagation probability exceeds the set propagation probability threshold P threshold , automatically generate fault warnings.

[0051] Optionally, the status evaluation and feedback mechanism in S5 includes:

[0052] S51, Equipment status assessment: Output Y based on the status association model i , score the status of each device and associate the score Y with the device status i The health index H of the equipment is calculated by taking the weighted average of the equipment failure history data i ;

[0053] S52, feedback to the collaborative learning mechanism: the health index H of the device i Feedback to the collaborative learning mechanism between devices to update the learning weights w between devices i ;

[0054] S53, equipment monitoring and maintenance strategy optimization: According to the changing trend of the equipment health index, dynamically adjust the equipment monitoring frequency and maintenance strategy. When the equipment health index is lower than the preset health threshold H threshold When the fault occurs, high-frequency monitoring and advance maintenance are automatically triggered.

[0055] Beneficial effects of the present invention:

[0056] The present invention realizes real-time monitoring and data collection of the status of power equipment through precise positioning based on the Beidou navigation system, and effectively improves the accuracy and timeliness of equipment status data by using a variety of sensors. The systematic monitoring method not only improves the reliability of equipment monitoring, but also can timely reflect the equipment operation status, ensuring that the equipment is warned before potential failures occur, reducing maintenance costs and optimizing operating efficiency, thereby enhancing the overall safety and reliability of power equipment.

[0057] The present invention realizes data sharing and transmission of fault experience between power equipment through a collaborative learning mechanism between devices, effectively constructs a collective knowledge base, and the mutual collaboration and information sharing between devices significantly improve the accuracy of fault prediction. The equipment can quickly take effective response measures based on historical fault data, thereby significantly reducing the occurrence rate of faults, enhancing the system's collaborative ability and intelligent management level, and further optimizing maintenance decisions and equipment operation strategies.

[0058] The present invention uses a graph neural network model to perform equipment status association analysis, identify potential failure modes between equipment and evaluate the risk of fault propagation. By combining failure mode recognition, status evaluation and feedback mechanism, the system can dynamically optimize equipment monitoring and maintenance strategies, improve the accuracy and efficiency of equipment health management, reduce the risk of fault propagation, extend the service life of equipment, and ensure the stability and efficiency of power system operation through fault warning mechanisms and optimized maintenance strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0060] Figure 1 This is a flow chart of a collection method according to an embodiment of the present invention;

[0061] Figure 2 Schematic diagram of device status association analysis according to an embodiment of the present invention. DETAILED DESCRIPTION

[0062] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0063] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0064] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0065] like Figure 1-Figure 2 As shown, the power equipment status monitoring data collection method based on the Beidou navigation system includes the following steps:

[0066] S1, equipment status data collection: locate power equipment through the Beidou navigation system, and use a variety of sensors to collect equipment status data in real time, including temperature, vibration, humidity and equipment load;

[0067] S2, collaborative learning mechanism between devices: Establish a collaborative learning mechanism between power devices, allowing adjacent devices to share monitoring data and fault experience, thereby forming a collective knowledge base;

[0068] S3, data fusion and standardization: standardize and integrate the status data collected by each device to form a unified data set;

[0069] S4, equipment status correlation analysis: Analyze the fused status data to identify the status correlation and potential failure modes between devices, including:

[0070] S41, feature extraction: extract features from the fused status data to identify key indicators, including temperature fluctuations, vibration frequency, humidity levels, and equipment load changes;

[0071] S42, correlation modeling: Based on the identified key indicators, establish a state correlation model between devices, and identify potential correlations and mutual influences by analyzing the state data of different devices;

[0072] S43, Fault mode identification: Based on the analysis results of the state association model, identify potential fault modes, evaluate the fault propagation risk between devices, and generate corresponding fault warnings;

[0073] S5, status evaluation and feedback mechanism: Based on the analysis results associated with the equipment status, the equipment status is evaluated and the evaluation results are fed back to the collaborative learning mechanism between equipment to continuously optimize the equipment monitoring and maintenance strategy;

[0074] Through the above content, efficient collection, intelligent analysis and optimized feedback of equipment status data are achieved, which not only improves the accuracy and timeliness of equipment monitoring, but also promotes information sharing and collaboration between equipment, significantly reduces the risk of failure, and thus improves the overall reliability and maintenance efficiency of power equipment.

[0075] The equipment status data collection in S1 includes:

[0076] S11, equipment positioning: locate the power equipment through the Beidou navigation system to obtain the geographical coordinate information of the equipment;

[0077] S12, sensor installation: installing various types of sensors on the power equipment, including temperature sensors, vibration sensors, humidity sensors, and load sensors;

[0078] S13, status data collection: collect the status data of the equipment in real time through various types of installed sensors, including temperature, vibration, humidity and equipment load;

[0079] Through the above content, precise positioning is achieved to ensure the accuracy of monitoring. At the same time, a variety of sensors are combined to collect key data such as temperature, vibration and humidity in real time, which can timely reflect the operating status of the equipment. This systematic monitoring method not only improves the reliability and timeliness of the data, but also helps to detect potential faults at an early stage, reduce equipment maintenance costs, improve overall operational efficiency, and thus enhance the safety and reliability of power equipment.

[0080] The inter-device collaborative learning mechanism in S2 includes:

[0081] S21, Data Sharing Protocol Establishment: Develop a data sharing protocol between power equipment to ensure that adjacent equipment can exchange monitoring data and fault experience safely and efficiently, and clarify the data transmission format, frequency and safety measures;

[0082] S22, collective knowledge base construction: Based on shared status data and experience, a collective knowledge base is constructed to organize and analyze historical fault cases and monitoring data to form a knowledge system for reference by all devices, so as to improve the learning ability of each device and the accuracy of fault prediction;

[0083] Through the above content, information exchange and experience sharing between adjacent power equipment are promoted, and the fault prediction and response capabilities are significantly improved. This mechanism can effectively integrate the monitoring data and historical fault cases of each device to form a comprehensive knowledge system, so that the equipment can quickly take effective measures when facing similar problems, thereby reducing the occurrence of failures, improving the reliability and maintenance efficiency of the overall system, and ultimately realizing the intelligent management and optimized operation of power equipment.

[0084] The establishment of data sharing agreement in S21 includes:

[0085] S211, formulate data transmission format: formulate a unified data transmission format for the monitoring data of power equipment, including data type, data structure and encoding method, to ensure the compatibility and readability of data between various devices;

[0086] S212, determine the frequency of data transmission: clarify the frequency of data transmission, set two modes of scheduled transmission and event-triggered transmission to ensure the real-time update and accuracy of monitoring data to meet the needs of equipment operation;

[0087] S213, Establish security measures: Establish security measures for data sharing, including data encryption, identity authentication and access control, to ensure the security and integrity of data shared between devices and prevent data leakage and tampering;

[0088] Through the above content, efficient and reliable data exchange between devices can be effectively realized. This standardized protocol not only improves the compatibility and real-time performance of data transmission and ensures the accurate update of monitoring data, but also protects the integrity of data through security measures to prevent potential security risks. This mechanism not only improves the collaborative working ability of equipment, but also provides a solid data foundation for fault prediction and maintenance decisions, thereby enhancing the overall safety and reliability of the power system.

[0089] The collective knowledge base construction in S22 includes:

[0090] S221, Data Collection and Collation: Based on the status data and fault experience shared among power equipment, collect and collate historical fault cases and related monitoring data to ensure the comprehensiveness and accuracy of the information;

[0091] S222, Data Analysis: Analyze the sorted historical failure cases and related monitoring data to identify failure modes, influencing factors and early warning indicators, including:

[0092] S2221, Identify failure modes: By calculating the failure rate and mean repair time as the quantitative basis of the failure mode, different failure modes are classified as follows:

[0093]

[0094] Among them, F r is the failure rate, F n is the number of failures, T is the total operating time;

[0095]

[0096] Among them, T r represents the total repair time, N f Indicates the number of failures;

[0097] S2222, Identify influencing factors: By calculating the Pearson correlation coefficient, identify the relationship between the status data and the fault occurrence, and determine the influencing factors, which are expressed as:

[0098]

[0099] Among them, r is the Pearson correlation coefficient, ranging between -1 and 1, indicating the correlation between two variables, X i is the value of variable X in the i-th sample, is the mean of variable X, Y i is the value of variable Y in the i-th sample, is the mean of variable Y;

[0100] S2223, Identify early warning indicators: Based on the historical distribution and trend of monitoring data, set early warning indicators, expressed as:

[0101]

[0102] in, is the mean of the state data, θ is the warning threshold, k is the set multiple (2 or 3), and σ is the standard deviation;

[0103] S223, knowledge system construction: store the analysis results and troubleshooting experience in a structured manner in the knowledge base to form a knowledge system for reference by all devices;

[0104] Through the above content, the shared status data and historical failure experience of the equipment are systematically organized and analyzed to form a structured knowledge system, which not only improves the collaborative work efficiency between equipment, but also effectively improves the accuracy of fault prediction and the scientific nature of maintenance decisions through the summary of failure modes and experience sharing. At the same time, the continuous updating of the collective knowledge base can continuously optimize equipment operation and maintenance strategies, reduce downtime, reduce maintenance costs, and enhance the level of intelligence of equipment management.

[0105] Data fusion and standardization in S3 include:

[0106] S31, data standardization processing: standardize the status data (including temperature, vibration, humidity, load) collected by each power device, expressed as:

[0107]

[0108] Among them, X' is the standardized data, X is the original data, and X min and X max are the minimum and maximum values ​​of the variable respectively;

[0109] S32, data fusion: The weighted average method is used to fuse the standardized state data, which is expressed as:

[0110]

[0111] Among them, X fused is the fused data, X i ' is the standardized data of the i-th device, w i is the weight of each device, n is the number of devices;

[0112] Through the above content, the dimensional differences of data between different devices are eliminated, ensuring the comparability and consistency of the data. At the same time, through the weighted average fusion method, combined with the importance of each device and the credibility of the data, a unified and comprehensive status data set is generated, which can effectively improve the quality of data integration and provide a more accurate basis for subsequent analysis and decision-making, help improve the overall efficiency of equipment monitoring, enhance the accuracy of fault detection, and optimize maintenance and early warning strategies.

[0113] Feature extraction in S41 includes:

[0114] S411, Temperature Fluctuation: The temperature fluctuation is extracted by calculating the mean square error to quantify the temperature volatility, expressed as:

[0115]

[0116] Among them, T i is the temperature value at the i-th time point, is the mean temperature, m is the number of time points;

[0117] S412, vibration frequency: Analyze through Fourier transform (FFT) to extract the main frequency components in the spectrum, expressed as:

[0118]

[0119] Among them, x(t) is the time domain signal, X(f) is the frequency component in the frequency domain, and f is the vibration frequency;

[0120] S413, Humidity level: Extract the humidity level by calculating the rate of change of humidity data, expressed as:

[0121]

[0122] Among them, H max and H min are the maximum and minimum humidity values, respectively, and T is the time of the collection period;

[0123] S414, equipment load change: Use the linear regression model to analyze the change trend of equipment load, expressed as:

[0124] Y=β0+β1X+∈;

[0125] Among them, Y is the equipment load value, X is the state data, β0 and β1 are regression coefficients, and ∈ is the error term;

[0126] Through the above content, comprehensive and accurate feedback on the equipment operation status can be provided, which improves the real-time monitoring capability of the equipment status, improves the efficiency of equipment maintenance and reduces the failure rate.

[0127] The state association model in S42 adopts the graph neural network (GCN) model, which includes:

[0128] S421, equipment status feature initialization: In the power equipment status monitoring, the multi-dimensional status data of each power equipment is initialized, including temperature, vibration, humidity, equipment load, and defined as feature vector Represents the initial state of the device, expressed as:

[0129]

[0130] Among them, T i is the temperature data of device i, V i is the vibration data, H i is the humidity data, L i It is the equipment load data;

[0131] S422, constructing an adjacency matrix: The adjacency matrix is ​​weighted according to the physical connection of the device, the state similarity, and the relevance of the historical state data, and is expressed as:

[0132]

[0133] Among them, A ij is the association strength between device i and device j, is the similarity of device states, calculated using cosine similarity, is the connection relationship in the physical topology, α is the weight factor, which controls the influence ratio of state similarity and physical connection;

[0134] S423, feature propagation and aggregation: In the propagation process of each layer, the state features of the power equipment are propagated through the adjacency matrix and weightedly aggregated with the state data of the surrounding power equipment. At the same time, bias terms and activation functions are added to enhance the nonlinear representation ability of the model to ensure that complex equipment state changes can be captured, which is expressed as:

[0135]

[0136] in, is the state characteristic of device i at layer l+1, A ij is the adjacency matrix element of device i and device j, is the set of neighbor nodes of device i, D ii is the diagonal element of the degree matrix, representing the degree of device i, D jj is the diagonal element of the degree matrix, representing the degree of device j, W (l) is the weight matrix of the lth layer, b (l) is the bias term, σ is the ReLU activation function;

[0137] S424, multi-scale fusion of device status: A multi-scale feature fusion method is used to cascade the features of different convolutional layers to form a device status representation, which is expressed as:

[0138]

[0139] in, is the multi-scale feature representation of device i, They are the features of the 1st, 2nd, ..., Lth layers respectively. concat means concatenating multiple layers of features together.

[0140] S425, state association output: Through the last layer of graph convolution, combined with multi-scale features, the state association score between devices is output to identify the potential correlation and mutual influence between devices, expressed as:

[0141]

[0142] Among them, Y i is the final association score of device i, W final and b final are the weights and biases of the last layer, and σ is the activation function;

[0143] Through the above content, the complex correlation and mutual influence between devices can be captured more accurately. By combining the state similarity and the improved adjacency matrix of physical topology, the model not only considers the physical connection of the equipment, but also dynamically reflects the changes in the operating status of the equipment. At the same time, multi-scale feature fusion enhances the ability to capture state information at different levels, making the correlation analysis more comprehensive and accurate, improving the accuracy of equipment fault prediction, and optimizing equipment monitoring and maintenance strategies, which helps to reduce the failure rate, extend equipment life, and improve the overall stability and operation efficiency of the power system.

[0144] Fault mode identification in S43 includes:

[0145] S431, fault correlation strength calculation: using the output result Y of the state correlation model iTo represent the state association score of device i, the fault propagation risk R between devices is calculated based on the state association score ij , expressed as:

[0146] R ij =Y i ·P i ;

[0147] Among them, R ij represents the risk of fault propagation from device i to device j, P i is the failure probability of device i, Y i is the state association score of device i;

[0148] S432, Fault propagation probability assessment: Use Bayesian network to evaluate the fault propagation probability P(F) between each power equipment. j |F i ) is evaluated and expressed as:

[0149]

[0150] Among them, P(F j |F i ) is the probability that device j will fail if device i fails, P(F i |F j ) is the conditional probability that device j affects device i, P(F j ) and P(F i ) are the failure probabilities of device j and device i respectively;

[0151] S432, Fault warning generation: Based on the calculated fault propagation risk R ij and the fault propagation probability P(F j |F i ), when the fault propagation risk exceeds the set propagation risk threshold R threshold And the fault propagation probability exceeds the set propagation probability threshold P threshold , automatically generate fault warnings;

[0152] Transmission risk threshold R threshold Based on the statistical distribution of fault propagation risk in historical operation data, percentiles are used to set thresholds, including:

[0153] Collect the risk R of fault propagation between devices ij Historical data;

[0154] Calculate the distribution of these risk values ​​and determine the 90% percentile;

[0155] The risk value corresponding to this percentile is taken as the transmission risk threshold R threshold , expressed as:

[0156] R threshold =P 90 (R ij );

[0157] Among them, P 90 (R ij ) represents the fault propagation risk R ij 90% quantile value of

[0158] Propagation probability threshold P threshold It is set based on the average propagation probability in historical data, including:

[0159] The probability of fault propagation between collection devices P(F j |F i )’s historical data;

[0160] Calculate the mean of these propagation probabilities

[0161] Set the safety factor k, and set the transmission probability threshold to the mean plus a certain multiple of the standard deviation, expressed as:

[0162]

[0163] Among them, σ P is the standard deviation of the fault propagation probability;

[0164] Through the above content, it is possible to more accurately identify potential failure modes between equipment, quantify the possibility of fault propagation, set reasonable propagation risk thresholds and propagation probability thresholds, and ensure that warnings are triggered only when the risk is high and the probability of fault propagation is high, thereby avoiding false alarms or missed alarms. By comprehensively analyzing and predicting the equipment status, the accuracy of fault warnings is improved, the system's protection capabilities are enhanced, and it is helpful to take preventive measures in a timely manner, reduce the risk of equipment fault propagation, and ultimately improve the stability and operating efficiency of the power system.

[0165] The status evaluation and feedback mechanism in S5 includes:

[0166] S51, Equipment status assessment: Output Y based on the status association model i , score the status of each device and associate the score Y with the device status i The health index H of the equipment is calculated by taking the weighted average of the equipment failure history data i , expressed as:

[0167]

[0168] Among them, H i is the health index of device i, Y iis the state association score of the device, P(F k ) is the historical probability of equipment failure, N f is the number of fault occurrences, α and β are weighting coefficients used to adjust the relative importance of state association and historical fault data;

[0169] S52, feedback to the collaborative learning mechanism: the health index H of the device i Feedback to the collaborative learning mechanism between devices to update the learning weights w between devices i , expressed as:

[0170]

[0171] in, is the learning weight of the current device i at the next time step t+1, is the weight of the current device i in the collaborative learning mechanism, H target is the preset target health index, η is the learning rate, which is used to control the adjustment amplitude of feedback on the device learning weight;

[0172] S53, equipment monitoring and maintenance strategy optimization: According to the changing trend of the equipment health index, dynamically adjust the equipment monitoring frequency and maintenance strategy. When the equipment health index is lower than the preset health threshold H threshold Automatically trigger high-frequency monitoring and advance maintenance;

[0173] Health threshold H threshold Set according to the historical data and performance indicators of the equipment, including:

[0174] Collect historical equipment health index data: Calculate the historical health index H based on the equipment status data (such as temperature, vibration, humidity, load). i , and conduct statistical analysis on them;

[0175] Mean and standard deviation calculation: Calculate the mean of the health index and standard deviation σ H ;

[0176] Set the threshold: Select the safety factor k, and set the health index threshold to the historical mean minus k times the standard deviation, where k is 1 or 2, expressed as:

[0177]

[0178] in, is the historical mean of the equipment health index, σ H is the standard deviation of the equipment health index, k is the safety factor;

[0179] Through the above content, the operating status of the equipment can be dynamically evaluated and potential fault risks can be discovered in time. At the same time, the mechanism will feed back the evaluation results to the collaborative learning system, continuously optimize the information sharing and learning weights between devices, and thus continuously improve the monitoring and maintenance strategies of the equipment. This closed-loop feedback mechanism can significantly improve the accuracy of fault warning and maintenance efficiency, avoid excessive maintenance or fault propagation, extend the service life of equipment, and enhance the stability of the power system and the overall operating efficiency.

[0180] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0181] The above is only 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 principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for collecting data on power equipment status monitoring based on the Beidou navigation system, characterized in that: The following steps are involved: S1, equipment status data collection: locate power equipment through the Beidou navigation system, and use a variety of sensors to collect equipment status data in real time, including temperature, vibration, humidity and equipment load; S2, collaborative learning mechanism between devices: Establish a collaborative learning mechanism between power devices, allowing adjacent devices to share monitoring data and fault experience, thereby forming a collective knowledge base; S3, data fusion and standardization: standardize and integrate the status data collected by each device to form a unified data set; S4, equipment status correlation analysis: Analyze the fused status data to identify the status correlation and potential failure modes between devices, including: S41, feature extraction: extract features from the fused status data to identify key indicators, including temperature fluctuations, vibration frequency, humidity levels, and equipment load changes; S42, correlation modeling: Based on the identified key indicators, a state correlation model between devices is established, and potential correlations and mutual influences are identified by analyzing the state data of different devices. The state correlation model adopts a graph neural network model, and the graph neural network model includes: S421, equipment status feature initialization: In the power equipment status monitoring, the multi-dimensional status data of each power equipment is initialized, including temperature, vibration, humidity, equipment load, and defined as feature vector Represents the initial state of the device; S422, constructing an adjacency matrix: the adjacency matrix is ​​weighted according to the physical connection of the devices, the state similarity, and the relevance of the historical state data; S423, feature propagation and aggregation: In the propagation process of each layer, the state features of the power equipment are propagated through the adjacency matrix and weightedly aggregated with the state data of the surrounding power equipment. At the same time, bias terms and activation functions are added to enhance the nonlinear representation ability of the model; S424, multi-scale fusion of device status: a multi-scale feature fusion method is used to cascade the features of different convolutional layers to form a device status representation; S425, state association output: Through the last layer of graph convolution, combined with multi-scale features, the state association score between devices is output to identify the potential correlation and mutual influence between devices; S43, Fault mode identification: Based on the analysis results of the state association model, identify potential fault modes, evaluate the fault propagation risk between devices, and generate corresponding fault warnings; S5, status evaluation and feedback mechanism: Based on the analysis results associated with the equipment status, the equipment status is evaluated and the evaluation results are fed back to the collaborative learning mechanism between devices to continuously optimize the equipment monitoring and maintenance strategy.

2. The method for collecting data for power equipment status monitoring based on the Beidou navigation system according to claim 1 is characterized in that: The equipment status data collection in S1 includes: S11, equipment positioning: locate the power equipment through the Beidou navigation system to obtain the geographical coordinate information of the equipment; S12, sensor installation: installing various types of sensors on the power equipment, including temperature sensors, vibration sensors, humidity sensors, and load sensors; S13, status data collection: collect the status data of the equipment in real time through various types of installed sensors, including temperature, vibration, humidity and equipment load.

3. The method for collecting data for power equipment status monitoring based on the Beidou navigation system according to claim 1 is characterized in that: The inter-device collaborative learning mechanism in S2 includes: S21, Data Sharing Agreement Establishment: Develop a data sharing agreement between power equipment to clarify the data transmission format, frequency and security measures; S22, collective knowledge base construction: Based on shared status data and experience, a collective knowledge base is constructed to organize and analyze historical fault cases and monitoring data to form a knowledge system for reference of all power equipment.

4. The method for collecting data for power equipment status monitoring based on the Beidou navigation system according to claim 3 is characterized in that: The data sharing agreement establishment in S21 includes: S211, formulate data transmission format: formulate a unified data transmission format for monitoring data of power equipment, including data type, data structure and encoding method; S212, determine the data transmission frequency: determine the frequency of data transmission, and set two modes: scheduled transmission and event-triggered transmission; S213, Establish security measures: Establish security measures for data sharing, including data encryption, identity authentication, and access control.

5. The method for collecting data for power equipment status monitoring based on the Beidou navigation system according to claim 4 is characterized in that: The collective knowledge base construction in S22 includes: S221, Data Collection and Collation: Based on the status data and fault experience shared among power equipment, collect and collate historical fault cases and related monitoring data; S222, Data Analysis: Analyze the sorted historical failure cases and related monitoring data to identify failure modes, influencing factors and early warning indicators, including: S2221, Identify failure modes: classify different failure modes by calculating failure rate and mean repair time as the quantitative basis of failure modes; S2222, Identify influencing factors: Identify the relationship between the status data and the occurrence of the fault by calculating the Pearson correlation coefficient, and determine the influencing factors; S2223, Identify early warning indicators: Set early warning indicators based on the historical distribution and trends of monitoring data; S223, knowledge system construction: store the analysis results and troubleshooting experience in a structured manner in the knowledge base to form a knowledge system for reference by all devices.

6. The method for collecting data for power equipment status monitoring based on Beidou navigation system according to claim 1 is characterized in that: The data fusion and standardization in S3 include: S31, data standardization processing: standardizing the status data collected from each power device; S32, data fusion: use weighted average method to fuse the standardized state data.

7. The method for collecting data for power equipment status monitoring based on Beidou navigation system according to claim 1 is characterized in that: The feature extraction in S41 includes: S411, Temperature Fluctuation: Extract temperature fluctuations by calculating the mean square error to quantify the temperature volatility; S412, vibration frequency: analyzed by Fourier transform to extract the main frequency components in the spectrum; S413, humidity level: extracting humidity level by calculating the rate of change of humidity data; S414, Equipment load change: Use linear regression model to analyze the change trend of equipment load.

8. The method for collecting data for power equipment status monitoring based on Beidou navigation system according to claim 1 is characterized in that: The fault mode identification in S43 includes: S431, fault correlation strength calculation: using the output result Y of the state correlation model i To represent the state association score of device i, the fault propagation risk R between devices is calculated based on the state association score ij ; S432, Fault propagation probability assessment: Use Bayesian network to evaluate the fault propagation probability P(F) between each power equipment. j |F i ) for assessment; S432, Fault warning generation: Based on the calculated fault propagation risk R ij and the fault propagation probability P(F j |F i ), when the fault propagation risk exceeds the set propagation risk threshold R threshold And the fault propagation probability exceeds the set propagation probability threshold P threshold , automatically generate fault warnings.

9. The method for collecting data for power equipment status monitoring based on the Beidou navigation system according to claim 8 is characterized in that: The state evaluation and feedback mechanism in S5 includes: S51, Equipment status assessment: Output Y based on the status association model i , score the status of each device and associate the score Y with the device status i The health index H of the equipment is calculated by taking the weighted average of the equipment failure history data i ; S52, feedback to the collaborative learning mechanism: the health index H of the device i Feedback to the collaborative learning mechanism between devices to update the learning weights w between devices i ; S53, equipment monitoring and maintenance strategy optimization: According to the changing trend of the equipment health index, dynamically adjust the equipment monitoring frequency and maintenance strategy. When the equipment health index is lower than the preset health threshold H threshold When the fault occurs, high-frequency monitoring and advance maintenance are automatically triggered.

Images