Power transmission and transformation equipment status assessment method and system based on digital twin

Abstract

The present invention discloses a method and system for evaluating the status of power transmission and transformation equipment based on digital twins. The method comprises the following steps: collecting and preprocessing power transmission and transformation equipment data; using an LSTM algorithm model to predict the operating status of the equipment; and using a fault tree analysis module to evaluate possible fault types and probabilities. Furthermore, an interaction mechanism is established between the LSTM algorithm model and the fault tree analysis module. When the LSTM algorithm model predicts an abnormal operating status of the equipment, the relevant data and prediction results are input into the fault tree analysis module to guide the fault tree to locate the fault. When the fault tree analysis module discovers a potential fault risk, the relevant information is fed back to the LSTM algorithm model. Based on the confidence level of the LSTM algorithm prediction results and the fault tree analysis results, a weighted fusion is performed on the two to obtain an evaluation result. The present invention establishes a model interaction mechanism to enable information sharing and collaborative work between the two, enhancing the ability to capture abnormal equipment status and potential fault risks.

Description

Technical Field

[0001] The present invention belongs to the field of power systems and relates to power system transmission and transformation equipment status monitoring and management technology, and specifically to a power transmission and transformation equipment status assessment method and system based on digital twins. Background Art

[0002] Digital twin technology is an emerging technology that uses digital means to construct virtual models of physical entities. In the field of power transmission and transformation equipment management, by building digital twin models of power transmission and transformation equipment, various operating data of the equipment can be mapped to the virtual model in real time, enabling real-time monitoring and comprehensive perception of the equipment's operating status. Leveraging advanced data analysis algorithms and simulation technologies, it is possible to accurately predict the future operating status of the equipment, identify potential failure risks in advance, and provide a scientific basis for equipment health management and maintenance decisions.

[0003] Currently, the commonly used architecture for predictive maintenance of power transmission and distribution equipment based on digital twin technology includes a data acquisition layer, a data processing layer, a digital twin model layer, a state prediction layer, and a health management layer. The data acquisition layer collects real-time data from various sensors on the equipment, including electrical parameters (such as voltage, current, and power), mechanical parameters (such as vibration and temperature), and environmental parameters (such as humidity and air pressure). The collected data is transmitted to the data processing layer via wired or wireless communication. The data processing layer performs cleaning and preprocessing on the transmitted data to remove noise and outliers and fill in missing data to improve data quality. The processed data is then transmitted to the digital twin model layer. This layer uses the cleaned data, combined with the physical structure, operating principles, and historical fault data of the power transmission and distribution equipment, to construct a digital twin model of the equipment, providing a virtual representation of the equipment's operating status. The state prediction layer inputs real-time data into the digital twin model and uses time series prediction algorithms to predict the equipment's operating status over a period of time. It also incorporates fault tree analysis to assess the potential fault types and probabilities. The health management layer automatically generates equipment health reports based on health prediction results, provides a graded assessment of equipment health, and provides maintenance personnel with repair recommendations and decision support when potential equipment failures occur. Data exchange between each layer is achieved through standard data interfaces and communication protocols, ensuring smooth data transmission and system collaboration.

[0004] However, this current architecture suffers from poor synergy between time series prediction algorithms and fault tree analysis methods. Time series prediction algorithms (such as LSTM algorithms) focus on capturing patterns in the time series of equipment operating data to predict future states, while fault tree analysis methods focus on analyzing fault causes and probabilities based on logical relationships. The two differ in data processing logic and application scenario emphasis. In practical applications, LSTM may predict an abnormal equipment operating state, but fault tree analysis may be unable to quickly locate the corresponding failure mode and cause. Alternatively, fault tree analysis may identify potential failure risks, but the LSTM model may fail to reflect the relevant trends in its predictions. This leads to inconsistencies between state predictions and fault risk assessments, hindering accurate judgment of equipment status. Summary of the Invention

[0005] Purpose of the invention: In order to solve the problem of poor synergy between time series prediction algorithm and fault tree analysis method, a method and system for power transmission and transformation equipment status assessment based on digital twin is provided.

[0006] Technical solution: To achieve the above objectives, the present invention provides a method for status assessment of power transmission and transformation equipment based on digital twins, comprising the following steps:

[0007] S1: Collect data from power transmission and transformation equipment and perform preprocessing;

[0008] S2: Input the preprocessed data into the LSTM algorithm model and the fault tree analysis module respectively;

[0009] The LSTM algorithm model predicts the operating status of the device in the future based on the relationship between the operating status and various parameters learned from the device's historical operating data, and outputs the changing trends of each parameter and possible abnormal situations.

[0010] The fault tree analysis module combines the equipment's historical operating data and fault data, uses fault tree analysis methods to determine the top and bottom events of the fault tree, and combines expert experience with historical data to determine the probability of each event. It calculates the probability of equipment failure through the logical relationship of the fault tree, analyzes the impact of various fault causes on equipment failure, and evaluates possible fault types and failure probabilities.

[0011] S3: Establish an interactive mechanism between the LSTM algorithm model and the fault tree analysis module. When the LSTM algorithm model predicts an abnormal equipment operating status, the relevant data and prediction results are input into the fault tree analysis module, guiding the fault tree to quickly locate possible faults. When the fault tree analysis module discovers a potential fault risk, it feeds the relevant information back to the LSTM algorithm model to help adjust the prediction direction.

[0012] S4: Based on the confidence level of the LSTM algorithm prediction results and the fault tree analysis results, the two are weightedly fused to obtain the evaluation results.

[0013] Furthermore, the data collected in step S1 includes electrical parameters, mechanical parameters and environmental parameters.

[0014] Furthermore, the preprocessing in step S1 includes preliminary screening, outlier detection, noise removal, missing data filling, and construction of a unified data feature framework.

[0015] Furthermore, in the preprocessing of step S1, constructing a unified data feature framework includes:

[0016] In the process of constructing electrical parameter features, in addition to using the time series data of each electrical parameter as the basic feature, the correlation between the parameters and each fault event in the fault tree analysis is analyzed;

[0017] Introducing the parameter-event correlation matrix, It represents the correlation between the i-th electrical parameter and the j-th fault event, which is calculated based on the frequency and intensity of electrical parameter anomalies and fault events in historical data. The formula is:

[0018]

[0019] in, Indicates that the i-th electrical parameter exceeds the threshold Under the condition, the probability of the jth fault event occurring is, represents the probability of occurrence of the jth fault event in the entire sample;

[0020] Construct an associated feature for each electrical parameter at time point t :

[0021]

[0022] in, represents the probability of occurrence of the jth fault event at time point t;

[0023] In the process of constructing the features of mechanical parameters and environmental parameters, the time series data of the parameters are used as the basic features;

[0024] The vibration signal is converted from the time domain to the frequency domain using short-time Fourier transform. Let the vibration signal of the parameter signal at time t be s(t), and the frequency domain transformation result is

[0025]

[0026] Where S(f) is the amplitude at frequency f;

[0027] In the frequency range The extracted features Represents the signal energy within the frequency band, and the calculation formula is

[0028] .

[0029] Furthermore, the LSTM algorithm model in step S2 includes an input gate, a forget gate, and an output gate;

[0030] The calculation formula of the forget gate is:

[0031]

[0032] Among them, f t represents the output of the forget gate, σ is the Sigmoid activation function, W f is the weight matrix of the forget gate, b f is the bias term of the forget gate, is the hidden state of the previous time step, x t is the input of the current time step;

[0033] The calculation formula of the input gate is:

[0034]

[0035] in, Represented as the input gate output, is the input gate weight matrix, represents the input gate bias term;

[0036] The candidate value of the new information is given by the candidate state vector It means that its calculation formula is:

[0037]

[0038] in, represents the weight matrix of the candidate state vector, Represents the bias term of the candidate state vector;

[0039] The update formula of the cell state at the current time step is:

[0040]

[0041] Where ⊙ represents the Hadamard product of element-wise multiplication;

[0042] The calculation formula of the output gate is:

[0043]

[0044] in, represents the output of the output gate, represents the weight matrix of the output gate, Represents the bias term of the output gate;

[0045] The update formula for the hidden state is:

[0046] .

[0047] Furthermore, the operation of the fault tree analysis module in step S2 includes:

[0048] Let the top event be E, which corresponds to a critical failure of the equipment; let the bottom event set be , represents each basic fault event that may lead to the occurrence of the top event; according to the logical relationship of the fault tree, if there is an "AND gate" logic between the top event and the bottom event, the probability of the top event occurring is:

[0049]

[0050] If there is an "OR gate" logic between the top event and the bottom event, the probability of the top event occurring is:

[0051]

[0052] Among them, P(E) represents the probability of occurrence of the top event, represents the probability of occurrence of the i-th bottom event.

[0053] Furthermore, in step S3, when the LSTM algorithm model predicts that the equipment operating status is abnormal, at time point t, the abnormal parameter set predicted by the LSTM model is ,in represents the observed value of the i-th parameter at time point t; for each abnormal parameter, the LSTM model gives the abnormality degree information, and the abnormality degree is set to ,in Indicates the abnormality of the i-th parameter, and the calculation formula is

[0054]

[0055] in, Representation parameters The mean of Representation parameters The standard deviation of

[0056] After receiving the data, the fault tree analysis module locates the fault mode in the fault tree logic structure it has built based on the parameter abnormality information; let the top event of the fault tree be E, and its corresponding bottom event set be , representing the basic events that may cause equipment failure; define a mapping function , represents the bottom event set that each abnormal parameter may cause; for abnormal parameters , if the abnormal degree of the parameter exceeds the set threshold θ, the bottom event associated with the parameter is regarded as a possible fault event, and its probability calculation formula is:

[0057]

[0058] The fault tree analysis module transfers the occurrence probability of all related bottom events to the top event, and obtains the final probability of the top event through logical operations; if the final probability exceeds the set risk threshold, it is judged that the equipment has a potential failure risk.

[0059] Furthermore, in step S3, when the fault tree analysis module finds a potential fault risk, it feeds back the key bottom events and related parameter information that lead to the fault risk to the LSTM algorithm model; let the feedback bottom event set be , the corresponding parameter set is ;After receiving feedback information, the LSTM algorithm model adjusts the weight parameters within the model;

[0060] Assume the hidden state of the LSTM algorithm model is h t , its update formula is adjusted to:

[0061]

[0062] in, Represents the output of the output gate, α is the adjustment coefficient, Representation parameters The weights in the model.

[0063] Furthermore, the step S4 specifically includes:

[0064] Assume that the prediction result of the LSTM algorithm model is , the result of fault tree analysis is , and The confidence levels are and , the calculation formula is:

[0065]

[0066]

[0067] in, Indicates the accuracy of the LSTM method for positive samples, Indicates the accuracy of the FTA method for positive samples, Indicates the recall rate of the LSTM method for positive samples, Represents the recall rate of the FTA method for positive samples;

[0068] Based on the confidence level, a dynamic weight mechanism is introduced; let the equipment's operating time be T, the cumulative number of failures be N, and the failure risk level at the current time point be R; define the dynamic weight and The calculation formula is:

[0069]

[0070]

[0071] Parameters β, γ, and δ are adjustment coefficients, which are used to control the sensitivity of the weight to risk, operating time, and number of failures, respectively;

[0072] The final comprehensive evaluation result is the weighted sum of the two parts, and the formula is:

[0073]

[0074] Where y represents the final result of equipment status prediction and failure risk assessment.

[0075] The present invention also provides a power transmission and transformation equipment status assessment system based on digital twins, comprising:

[0076] Data acquisition and preprocessing module, used to collect data from power transmission and transformation equipment and perform preprocessing;

[0077] The LSTM algorithm module is used to predict the operating status of the device in the future by learning the relationship between the operating status and various parameters based on the device's historical operating data through the LSTM algorithm model, and output the changing trends of each parameter and possible abnormal situations.

[0078] The fault tree analysis module is used to determine the top and bottom events of the fault tree using the fault tree analysis method, and to determine the probability of each event by combining expert experience and historical data. It calculates the probability of equipment failure through the logical relationship of the fault tree, analyzes the impact of various fault causes on equipment failure, and evaluates the possible fault types and failure probabilities.

[0079] The interactive module, when the LSTM algorithm model predicts an abnormal equipment operating status, inputs relevant data and prediction results into the fault tree analysis module, guiding the fault tree to quickly locate possible faults. When the fault tree analysis module finds potential fault risks, it feeds relevant information back to the LSTM algorithm model to help adjust the prediction direction.

[0080] The comprehensive evaluation module is used to perform weighted fusion of the LSTM algorithm prediction results and the fault tree analysis results based on their confidence levels to obtain the evaluation results.

[0081] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0082] Improving prediction accuracy: By building a unified data feature representation framework, the LSTM algorithm and fault tree analysis method can analyze data based on similar understandings, reducing prediction bias caused by differences in data processing logic. A model interaction mechanism is established to enable information sharing and collaboration between the two methods, enhancing the ability to capture abnormal equipment conditions and potential failure risks. A comprehensive evaluation algorithm is designed to weightedly integrate the results of the two methods, fully leveraging the advantages of both methods to improve the accuracy of equipment condition prediction and fault risk assessment, and avoid misjudgments caused by inconsistent prediction results.

[0083] Improved O&M efficiency: Accurate equipment status prediction and fault risk assessment help O&M personnel promptly identify potential equipment issues, develop maintenance plans in advance, and rationally allocate maintenance resources. This reduces unnecessary equipment inspections and repairs, lowers O&M costs, improves O&M efficiency, and ensures stable power system operation.

[0084] Enhance system reliability: Timely and accurate early warning mechanisms enable operation and maintenance personnel to respond quickly to equipment anomalies or failures and take effective measures to deal with them, thereby avoiding further expansion of equipment failures, reducing the occurrence of power outages, improving the reliability of power transmission and transformation equipment, and ensuring the safe and stable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] Figure 1 It is a system framework diagram of the present invention;

[0086] Figure 2 This is an interaction diagram of the interaction module. DETAILED DESCRIPTION

[0087] The present invention is further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, modifications of various equivalent forms of the present invention made by those skilled in the art all fall within the scope defined by the claims attached to this application.

[0088] Example 1:

[0089] like Figure 1 As shown, this embodiment provides a method for evaluating the state of power transmission and transformation equipment based on digital twins, comprising the following steps:

[0090] S1: Collect and pre-process data from power transmission and transformation equipment:

[0091] The collected data include electrical parameters, mechanical parameters and environmental parameters; electrical parameters include voltage, current, frequency, phase angle, etc., mechanical parameters include vibration, displacement, speed, temperature, etc., and environmental parameters include humidity, air pressure, wind speed, temperature, etc.

[0092] Preprocessing includes preliminary screening, outlier detection, noise removal, missing data filling, and building a unified data feature framework;

[0093] Preliminary screening: The collected data is preliminarily screened to remove obviously erroneous or unreasonable data. The screening criteria are based on physical constraints and the normal operating range of equipment parameters. For example, for voltage parameters, if the voltage value at a certain moment is much higher than the rated voltage of the equipment or significantly lower than the normal operating voltage range, the data is considered to have obvious errors and is eliminated.

[0094] Outlier detection includes: using statistical outlier detection methods to identify and eliminate abnormal data; determining the range of abnormal data based on the mean and standard deviation of the data; setting the data sample set to

[0095]

[0096] The sample mean is:

[0097]

[0098] The sample standard deviation is

[0099]

[0100] The range of abnormal data is defined as:

[0101]

[0102] Among them, k is the coefficient of abnormality judgment, which usually ranges from 2 to 3. This method has a strong ability to identify data that deviates significantly from the average value in the data set;

[0103] Noise removal involves using the DBSCAN algorithm, a density-based spatial clustering algorithm, to remove noise points. The DBSCAN algorithm identifies high-density data clusters in the dataset by setting a radius parameter ϵ and a density threshold MinPts, and treats isolated points or sparse areas as noise and removes them. For any data point p in the dataset, if the number of points within a neighborhood with a radius of ϵ centered on p is not less than MinPts, then the point p is marked as a core point. All points that are directly or indirectly connected to the core point belong to the same cluster, and unclassified points are treated as noise points and removed.

[0104] Missing data filling includes: using the K nearest neighbor interpolation algorithm based on machine learning to complete the data; assuming that the position where a parameter in the data set has a missing value is x i , use the K nearest neighbor algorithm to find x i The K most similar sample points in the feature space:

[0105]

[0106] The missing values ​​are estimated by weighted average method. The specific calculation formula is:

[0107]

[0108] Among them, w j is the jth sample point and x i The inverse of the distance between, that is:

[0109]

[0110] Among them, d(x i ,x j ) represents the sample x i with x j The distance between them is calculated using Euclidean distance:

[0111]

[0112] Among them, M is the feature dimension.

[0113] Building a unified data feature framework includes:

[0114] In the process of constructing electrical parameter features, in addition to taking the time series data of each electrical parameter as the basic feature, the correlation between the parameters and each fault event in the fault tree analysis is analyzed; let the observation value of the i-th electrical parameter at time point t be , and its historical observation data sequence is , this sequence is used for trend feature extraction in time series models;

[0115] At the same time, in order to further extract the features associated with the fault event, the parameter-event association matrix is ​​introduced, where It represents the correlation between the i-th electrical parameter and the j-th fault event, which is calculated based on the frequency and intensity of electrical parameter anomalies and fault events in historical data. The formula is:

[0116]

[0117] in, Indicates that the i-th electrical parameter exceeds the threshold Under the condition, the probability of the jth fault event occurring is, Indicates the probability of occurrence of the jth fault event in the overall sample; correlation The larger the value, the more likely the abnormality of the electrical parameter will cause a corresponding fault event;

[0118] Based on the correlation matrix, a correlation feature is constructed for each electrical parameter at time point t. :

[0119]

[0120] in, represents the probability of occurrence of the jth fault event at time point t; this probability is obtained based on the event deduction results in the fault tree analysis module; finally, the parameter Time series characteristics and their correlation characteristics Input the model together to improve the accuracy of equipment status prediction and failure risk assessment;

[0121] In the process of constructing the characteristics of mechanical parameters and environmental parameters, the time series data of the parameters are first used as the basic characteristics; let the observation value of the i-th parameter at time point t be , and its historical observation data sequence is , this sequence is used to extract time trend features;

[0122] On this basis, in order to further extract the correlation characteristics between parameters and specific fault events, it is necessary to combine the sensitivity of related fault events to these parameters in the fault tree analysis. Assume that a certain mechanical fault is associated with a specific frequency range of the vibration signal, and use short-time Fourier transform to convert the vibration signal from the time domain to the frequency domain; suppose the vibration signal of the parameter signal at time t is s(t), and its frequency domain transformation result is

[0123]

[0124] Where S(f) is the amplitude at frequency f, indicating the signal strength at that frequency;

[0125] In a specific frequency range The extracted features Represents the signal energy within the frequency band, and the calculation formula is

[0126]

[0127] This feature is used to characterize the vibration characteristics of mechanical or environmental parameters associated with fault events. Ultimately, the time series characteristics and energy characteristics within the sensitive frequency range are simultaneously input into the model to improve the accuracy of equipment status prediction and fault risk assessment.

[0128] S2: Input the preprocessed data into the LSTM algorithm model and the fault tree analysis module respectively;

[0129] The LSTM algorithm model predicts the operating status of the device in the future based on the relationship between the operating status and various parameters learned from the device's historical operating data, and outputs the changing trends of each parameter and possible abnormal situations.

[0130] The fault tree analysis module combines the equipment's historical operating data and fault data, uses fault tree analysis methods to determine the top and bottom events of the fault tree, and combines expert experience with historical data to determine the probability of each event. It calculates the probability of equipment failure through the logical relationship of the fault tree, analyzes the impact of various fault causes on equipment failure, and evaluates possible fault types and failure probabilities.

[0131] The LSTM algorithm model includes input gate, forget gate and output gate;

[0132] The calculation formula of the forget gate is:

[0133]

[0134] Among them, f t represents the output of the forget gate, σ is the Sigmoid activation function, W f is the weight matrix of the forget gate, b f is the bias term of the forget gate, is the hidden state of the previous time step, x t is the input of the current time step;

[0135] The calculation formula of the input gate is:

[0136]

[0137] in, Represented as the input gate output, is the input gate weight matrix, represents the input gate bias term;

[0138] The candidate value of the new information is given by the candidate state vector It means that its calculation formula is:

[0139]

[0140] in, represents the weight matrix of the candidate state vector, Represents the bias term of the candidate state vector;

[0141] The update formula of the cell state at the current time step is:

[0142]

[0143] Where ⊙ represents the Hadamard product of element-wise multiplication;

[0144] The calculation formula of the output gate is:

[0145]

[0146] in, represents the output of the output gate, represents the weight matrix of the output gate, Represents the bias term of the output gate;

[0147] The update formula for the hidden state is:

[0148]

[0149] The operation of the Fault Tree Analysis module includes:

[0150] Let the top event be E, which corresponds to a critical failure of the equipment; let the bottom event set be , represents each basic fault event that may lead to the occurrence of the top event; according to the logical relationship of the fault tree, if there is an "AND gate" logic between the top event and the bottom event, the probability of the top event occurring is:

[0151]

[0152] If there is an "OR gate" logic between the top event and the bottom event, the probability of the top event occurring is:

[0153]

[0154] Among them, P(E) represents the probability of occurrence of the top event, represents the probability of occurrence of the i-th bottom event.

[0155] S3: If Figure 2 As shown in the figure, an interactive mechanism is established between the LSTM algorithm model and the fault tree analysis module. When the LSTM algorithm model predicts an abnormal equipment operating status, the relevant data and prediction results are input into the fault tree analysis module to guide the fault tree to quickly locate possible faults. When the fault tree analysis module finds a potential fault risk, the relevant information is fed back to the LSTM algorithm model to help it adjust its prediction direction and enhance its ability to capture specific fault trends.

[0156] When the LSTM algorithm model predicts that the equipment operating status is abnormal, at time point t, the abnormal parameter set predicted by the LSTM model is ,in represents the observed value of the i-th parameter at time point t; for each abnormal parameter, the LSTM model gives the abnormality degree information, and the abnormality degree is set to ,in It represents the abnormality of the i-th parameter, which can be measured by the degree by which the parameter deviates from the normal range. The calculation formula is:

[0157]

[0158] in, Representation parameters The mean of Representation parameters The larger the abnormality value, the greater the deviation of the parameter from the normal range;

[0159] After receiving the data, the fault tree analysis module quickly locates the possible fault mode in the fault tree logic structure it has built based on the parameter abnormality information; let the top event of the fault tree be E, and its corresponding bottom event set be , represents the basic events that may cause equipment failure; in order to quickly match the association between abnormal parameters and failure events, a mapping function is defined , represents the bottom event set that each abnormal parameter may cause; for abnormal parameters , if the abnormal degree of the parameter exceeds the set threshold θ, the bottom event associated with the parameter is regarded as a possible fault event, and its probability calculation formula is:

[0160]

[0161] The fault tree analysis module transfers the occurrence probability of all related bottom events to the top event, and obtains the final probability of the top event through logical operations; if the final probability exceeds the set risk threshold, it is judged that the equipment has a potential failure risk.

[0162] When the fault tree analysis module finds a potential fault risk, it will feed back the key bottom events and related parameter information that lead to the fault risk to the LSTM algorithm model; let the feedback bottom event set be , the corresponding parameter set is After receiving feedback, the LSTM algorithm model adjusts the weight parameters within the model based on the historical change trends of these parameters to enhance its focus on specific fault trends.

[0163] Assume the hidden state of the LSTM algorithm model is h t , its update formula is adjusted to:

[0164]

[0165] in, Represents the output of the output gate, α is the adjustment coefficient, Representation parameters Weights in the model; by introducing feedback parameters and their weights, the LSTM algorithm model can focus on the changing trends of potential risk parameters and further improve the prediction accuracy of specific fault trends.

[0166] S4: Based on the confidence level of the LSTM algorithm prediction results and the fault tree analysis results, a weighted fusion is performed on the two to obtain the evaluation results;

[0167] Assume that the prediction result of the LSTM algorithm model is , the result of fault tree analysis is , and their respective confidence levels are and , the confidence is calculated based on the accuracy and recall of each method in historical prediction and evaluation, and the calculation formula is:

[0168]

[0169]

[0170] in, Indicates the accuracy of the LSTM method for positive samples, Indicates the accuracy of the FTA method for positive samples, Indicates the recall rate of the LSTM method for positive samples, represents the recall rate of the FTA method for positive samples; the harmonic mean of the two is used as the confidence value to measure the reliability of each method;

[0171] Based on the confidence level, a dynamic weight mechanism is introduced. The weight setting is related to the operating status of the equipment. Let the equipment's operating time be T, the cumulative number of failures be N, and the failure risk level at the current time point be R. Define the dynamic weight and The calculation formula is:

[0172]

[0173]

[0174] Parameters β, γ, and δ are adjustment coefficients, which are used to control the sensitivity of the weight to risk, operating time, and number of failures, respectively. As the failure risk R increases, rise, Decrease; when the equipment operation time is short and the cumulative number of failures is small, wLSTM is larger;

[0175] The final comprehensive evaluation result is the weighted sum of the two parts, and the formula is:

[0176]

[0177] Here, y represents the final result of equipment status prediction and fault risk assessment. It can favor the prediction results of the LSTM algorithm when the equipment is operating stably, and place more emphasis on the fault tree analysis results when the equipment is approaching a high-failure period, thereby obtaining more accurate and consistent prediction and assessment conclusions.

[0178] S5: Based on the equipment status prediction and fault risk assessment results after comprehensive evaluation, the system automatically generates an equipment status prediction report and a fault risk assessment report, which records in detail the predicted operating status of the equipment, possible fault types and fault probabilities; if the evaluation results show that the equipment is in an abnormal or faulty state, the system will issue a warning signal in a timely manner.

[0179] Example 2:

[0180] Based on the evaluation method provided in Example 1, this embodiment provides a power transmission and transformation equipment status evaluation system based on digital twins, including:

[0181] Data acquisition and preprocessing module, used to collect data from power transmission and transformation equipment and perform preprocessing;

[0182] The LSTM algorithm module is used to predict the operating status of the device in the future by learning the relationship between the operating status and various parameters based on the device's historical operating data through the LSTM algorithm model, and output the changing trends of each parameter and possible abnormal situations.

[0183] The fault tree analysis module is used to determine the top and bottom events of the fault tree using the fault tree analysis method, and to determine the probability of each event by combining expert experience and historical data. It calculates the probability of equipment failure through the logical relationship of the fault tree, analyzes the impact of various fault causes on equipment failure, and evaluates the possible fault types and failure probabilities.

[0184] The interactive module, when the LSTM algorithm model predicts an abnormal equipment operating status, inputs relevant data and prediction results into the fault tree analysis module, guiding the fault tree to quickly locate possible faults. When the fault tree analysis module discovers potential fault risks, it feeds relevant information back to the LSTM algorithm model to help it adjust its prediction direction and enhance its ability to capture specific fault trends.

[0185] The comprehensive evaluation module is used to perform weighted fusion of the LSTM algorithm prediction results and the fault tree analysis results based on their confidence levels to obtain the evaluation results.

[0186] Example 3:

[0187] To verify the effectiveness of the solution of the present invention, this embodiment conducted a large-scale experimental test. The experimental data was collected from the actual operation data of a transmission and substation in a certain area, and different operating conditions were simulated to evaluate the system's prediction accuracy and fault warning capabilities.

[0188] 1. Data Collection

[0189] Acquisition equipment: transformers, circuit breakers, power cables, lightning arresters, mutual inductors;

[0190] Monitoring parameters:

[0191] Electrical parameters: voltage (kV), current (A), power factor, frequency (Hz), power quality harmonic content;

[0192] Mechanical parameters: vibration frequency (Hz), equipment temperature (°C), equipment noise (dB), oil level change (mm);

[0193] Environmental parameters: humidity (%RH), temperature (°C), wind speed (m / s), air pressure (hPa);

[0194] Collection cycle: 1 minute / time, continuous operation for 180 days;

[0195] Data storage capacity: over 15TB;

[0196] 2. State Prediction Experiment

[0197] The LSTM model is used for time series prediction and the prediction error is evaluated, as shown in Table 1.

[0198] Table 1

[0199]

[0200] 3. Fault Warning Accuracy

[0201] By comparing the system warning results with the actual fault events, the warning accuracy is calculated.

[0202] Total equipment failures: 50;

[0203] Number of successful system warnings: 46;

[0204] Early warning accuracy: 92%;

[0205] False positive rate: 3.2%;

[0206] Missing report rate: 4.8%;

[0207] Experimental data and equipment test results show that the solution of the present invention can effectively improve the accuracy of equipment status monitoring, identify potential failure risks in advance, improve the reliability of power transmission and transformation equipment, reduce operation and maintenance costs, and provide important technical support for smart grid management.

Claims

1. A method for status assessment of power transmission and transformation equipment based on digital twins, characterized in that: The steps include: S1: Collect data from power transmission and transformation equipment and perform preprocessing; S2: Input the preprocessed data into the LSTM algorithm model and the fault tree analysis module respectively; The LSTM algorithm model predicts the operating status of the device based on the relationship between the operating status and various parameters learned from the device's historical operating data, and outputs the changing trends of each parameter and possible abnormal situations. The fault tree analysis module combines the equipment's historical operating data and fault data, uses fault tree analysis methods to determine the top and bottom events of the fault tree, and combines expert experience with historical data to determine the probability of each event. It calculates the probability of equipment failure through the logical relationship of the fault tree, analyzes the impact of various fault causes on equipment failure, and evaluates possible fault types and failure probabilities. S3: Establish an interactive mechanism between the LSTM algorithm model and the fault tree analysis module. When the LSTM algorithm model predicts an abnormal equipment operating status, the relevant data and prediction results are input into the fault tree analysis module to guide the fault tree to locate the fault. When the fault tree analysis module discovers a potential fault risk, the relevant information is fed back to the LSTM algorithm model to help adjust the prediction direction. S4: Based on the confidence level of the LSTM algorithm prediction results and the fault tree analysis results, a weighted fusion is performed on the two to obtain the evaluation results; In step S3, when the fault tree analysis module finds a potential fault risk, the key bottom events that lead to the fault risk and their related parameter information are fed back to the LSTM algorithm model; let the feedback bottom event set be E f ={e f1 ,e f2 ,...,e fk }, the corresponding parameter set is X f ={x f1 ,x f2 ,...,x fk After receiving feedback information, the LSTM algorithm model adjusts the weight parameters within the model; Assume the hidden state of the LSTM algorithm model is h t , its update formula is adjusted to: Among them, O t represents the output of the output gate, α is the adjustment coefficient, w fj Represents parameter x fj weights in the model; Step S4 specifically includes: Assume that the prediction result of the LSTM algorithm model is y LSTM , the result of fault tree analysis is y FTA ,y LSTM and y FTA The confidence levels are c LSTM and c FTA , the calculation formula is: Among them, P LSTM Indicates the accuracy of the LSTM method for positive samples, P FTA Indicates the accuracy of the FTA method for positive samples, R LSTM Represents the recall rate of the LSTM method for positive samples, R FTA Represents the recall rate of the FTA method for positive samples; Based on the confidence level, a dynamic weight mechanism is introduced. Let the running time of the equipment be T, the cumulative number of failures be N, and the failure risk level at the current time point be R. Define the dynamic weight w LSTM and w FTA The calculation formula is: Parameters β, γ, and δ are adjustment coefficients, which are used to control the sensitivity of the weight to risk, operating time, and number of failures, respectively; The final comprehensive evaluation result is the weighted sum of the two parts, and the formula is: y=w LSTM ·y LSTM +w FTA ·y FTA Where y represents the final result of equipment status prediction and failure risk assessment.

2. A method for evaluating the state of power transmission and transformation equipment based on digital twins according to claim 1, characterized in that: The data collected in step S1 include electrical parameters, mechanical parameters and environmental parameters.

3. The method for status assessment of power transmission and transformation equipment based on digital twin according to claim 2, characterized in that: The preprocessing in step S1 includes preliminary screening, outlier detection, noise removal, missing data filling, and building a unified data feature framework.

4. A method for status assessment of power transmission and transformation equipment based on digital twins according to claim 3, characterized in that: In the preprocessing of step S1, building a unified data feature framework includes: In the process of constructing electrical parameter features, in addition to using the time series data of each electrical parameter as the basic feature, the correlation between the parameters and each fault event in the fault tree analysis is analyzed; Introducing parameter-event correlation matrix, a ij It represents the correlation between the i-th electrical parameter and the j-th fault event, which is calculated based on the frequency and intensity of electrical parameter anomalies and fault events in historical data. The formula is: Among them, P(e j |x i >θ i ) indicates that the i-th electrical parameter exceeds the threshold θ i Under the condition, the probability of the jth fault event occurring, P(e j ) represents the probability of occurrence of the jth fault event in the entire sample; Construct the associated feature z for each electrical parameter at time point t ti : Among them, p ij represents the probability of occurrence of the jth fault event at time point t; In the process of constructing the features of mechanical parameters and environmental parameters, the time series data of the parameters are used as the basic features; The vibration signal is converted from the time domain to the frequency domain using short-time Fourier transform. Let the vibration signal of the parameter signal at time t be s(t), and the frequency domain transformation result is Where S(f) is the amplitude at frequency f; In the frequency range [f low ,f high ], the extracted feature z ti Represents the signal energy within the frequency band, and the calculation formula is 5. The method for status assessment of power transmission and transformation equipment based on digital twin according to claim 1, characterized in that: The LSTM algorithm model in step S2 includes an input gate, a forget gate, and an output gate; The calculation formula of the forget gate is: f t =σ(W f ·[h t-1 ,x t ]+b f ) Among them, f t represents the output of the forget gate, σ is the Sigmoid activation function, W f is the weight matrix of the forget gate, b f is the bias term of the forget gate, h t-1 is the hidden state of the previous time step, x t is the input of the current time step; The calculation formula of the input gate is: i t =σ(W i ·[h t-1 ,x t ]+b i ) Among them, i t Represented as input gate output, W i is the input gate weight matrix, b i represents the input gate bias term; The candidate value of the new information is given by the candidate state vector It means that its calculation formula is: Among them, W C Represents the weight matrix of the candidate state vector, b C Represents the bias term of the candidate state vector; The update formula of the cell state at the current time step is: Where ⊙ represents the Hadamard product of element-wise multiplication; The calculation formula of the output gate is: the t =σ(W o ·[h t-1 ,x t ]+b o ) Among them, t Represents the output of the output gate, W o represents the weight matrix of the output gate, b o Represents the bias term of the output gate; The update formula for the hidden state is: h t =o t ☉tan h(C t )。 6. A method for status assessment of power transmission and transformation equipment based on digital twins according to claim 4, characterized in that: The operation of the fault tree analysis module in step S2 includes: Let the top event be E, which corresponds to a critical failure of the equipment; let the bottom event set be {e1,e2,...,e n }, representing the various basic fault events that may lead to the occurrence of the top event; according to the logical relationship of the fault tree, if there is an "AND gate" logic between the top event and the bottom event, the probability of the top event occurring is: If there is an "OR gate" logic between the top event and the bottom event, the probability of the top event occurring is: Among them, P(E) represents the probability of occurrence of the top event, P(e i ) represents the probability of occurrence of the i-th bottom event.

7. A method for status assessment of power transmission and transformation equipment based on digital twins according to claim 6, characterized in that: In step S3, when the LSTM algorithm model predicts that the equipment operation status is abnormal, the abnormal parameter set predicted by the LSTM model at time point t is X t ={x t1 ,x t2 ,...,x ti }, where x ti represents the observed value of the i-th parameter at time point t; for each abnormal parameter, the LSTM model gives the abnormality degree information, and the abnormality degree is set to D t ={d t1 ,d t2 ,...,d ti }, where d ti Indicates the abnormality of the i-th parameter, and the calculation formula is Among them, μ i Represents parameter x ti The mean of i Represents parameter x ti The standard deviation of After receiving the data, the fault tree analysis module locates the fault mode in the fault tree logic structure it has built based on the parameter abnormality information; let the top event of the fault tree be E, and its corresponding bottom event set be {e1, e2, ..., e n }, representing the basic events that may cause equipment failure; define a mapping function f:X t →{e1,e2,...,e n }, represents the bottom set of events that may be triggered by each abnormal parameter; for abnormal parameter x ti , if the abnormal degree of the parameter exceeds the set threshold θ, the bottom event associated with the parameter is regarded as a possible fault event, and its probability calculation formula is: The fault tree analysis module transfers the occurrence probability of all related bottom events to the top event, and obtains the final probability of the top event through logical operations; if the final probability exceeds the set risk threshold, it is judged that the equipment has a potential failure risk.

8. A power transmission and transformation equipment status assessment system based on digital twins, characterized in that: include: Data acquisition and preprocessing module, used to collect data from power transmission and transformation equipment and perform preprocessing; The LSTM algorithm module is used to predict the operating status of the device based on the relationship between the operating status and various parameters learned from the device's historical operating data through the LSTM algorithm model, and output the changing trends of each parameter and possible abnormal situations; The fault tree analysis module is used to determine the top and bottom events of the fault tree using the fault tree analysis method, and to determine the probability of each event by combining expert experience and historical data. It calculates the probability of equipment failure through the logical relationship of the fault tree, analyzes the impact of various fault causes on equipment failure, and evaluates the possible fault types and failure probabilities. Interaction module: When the LSTM algorithm model predicts that the equipment operating status is abnormal, the relevant data and prediction results are input into the fault tree analysis module to guide the fault tree to locate the fault; When the fault tree analysis module discovers a potential fault risk, it feeds the relevant information back to the LSTM algorithm model to help adjust the prediction direction, including: When the fault tree analysis module finds a potential fault risk, it will feed back the key bottom events and related parameter information that lead to the fault risk to the LSTM algorithm model; let the feedback bottom event set be E f ={e f1 ,e f2 ,...,e fk }, the corresponding parameter set is X f ={x f1 ,x f2 ,...,x fk After receiving feedback information, the LSTM algorithm model adjusts the weight parameters within the model; Assume the hidden state of the LSTM algorithm model is h t , its update formula is adjusted to: Among them, t represents the output of the output gate, α is the adjustment coefficient, w fj Represents parameter x fj weights in the model; The comprehensive evaluation module is used to perform weighted fusion of the LSTM algorithm prediction results and the fault tree analysis results based on their confidence levels to obtain the evaluation results, including: Assume that the prediction result of the LSTM algorithm model is y LSTM , the result of fault tree analysis is y FTA ,y LSTM and y FTA The confidence levels are c LSTM and c FTA , the calculation formula is: Among them, P LSTM Indicates the accuracy of the LSTM method for positive samples, P FTA Indicates the accuracy of the FTA method for positive samples, R LSTM Represents the recall rate of the LSTM method for positive samples, R FTA Represents the recall rate of the FTA method for positive samples; Based on the confidence level, a dynamic weight mechanism is introduced. Let the running time of the equipment be T, the cumulative number of failures be N, and the failure risk level at the current time point be R. Define the dynamic weight w LSTM and w FTA The calculation formula is: Parameters β, γ, and δ are adjustment coefficients, which are used to control the sensitivity of the weight to risk, operating time, and number of failures, respectively; The final comprehensive evaluation result is the weighted sum of the two parts, and the formula is: y=w LSTM ·y LSTM +w FTA ·y FTA Where y represents the final result of equipment status prediction and failure risk assessment.

Images