Substation channel cover plate data acquisition and intelligent analysis management system and method

Abstract

The invention relates to the technical field of transformer substations, and particularly discloses a transformer substation channel cover plate data acquisition and intelligent analysis management system and method, and the method comprises the steps: collecting cover plate vibration data and strain data through an acceleration sensor and a resistive strain gauge; vibration direction change, vibration attenuation and cover plate deformation springback time are extracted from acceleration data, the maximum amplitude, the change rate and the strain gradient are extracted from strain data, and the stress deformation characteristics of the cover plate are comprehensively represented; based on the stress deformation and springback characteristics of the cover plate, a damage evaluation model is established by adopting a neural network model, and intelligent judgment of the damage degree of the cover plate is realized; predicting a damage change trend by using a neural network, and calculating a cover plate damage acceleration trend; and according to the damage evaluation result and the damage acceleration trend, calculating the remaining to-be-maintained time of the cover plate, and automatically triggering operation and maintenance early warning when the remaining to-be-maintained time is lower than an early warning threshold value. According to the method, intelligent monitoring, prediction and active operation and maintenance management of the health state of the cover plate can be realized.

Description

Technical Field

[0001] This application relates to the technical field of substations, and specifically to a data acquisition and intelligent analysis management system and method for substation channel cover plates. Background Technique

[0002] Substation channel cover plates are an important part of the power system infrastructure, mainly used to cover cable channels and protect the safe operation of internal cable facilities. These cover plates are usually made of composite materials and are subject to the influence of vehicle rolling, personnel trampling, and environmental factors for a long time.

[0003] In actual operation and maintenance management, the status monitoring of substation channel cover plates mainly relies on manual inspection. Maintenance personnel judge the usage status by visually inspecting obvious defects such as cracks and damages on the surface of the cover plates. This traditional detection method can detect obvious damages that have occurred, but it is difficult to detect minor damages and performance degradation inside the cover plates in a timely manner. During the long-term use of the cover plates, their bearing capacity and structural performance will gradually decline. If not detected and replaced in time, it may cause the cover plates to suddenly break, affecting the safe operation of the substation.

[0004] Therefore, how to accurately evaluate the damage status of substation channel cover plates and how to scientifically predict the maintenance time of the cover plates have become technical problems that need to be solved in substation operation and maintenance management.

[0005] In view of this, this application proposes a data acquisition and intelligent analysis management system and method for substation channel cover plates. Summary of the Invention

[0006] To achieve the above object, this application provides a data acquisition and intelligent analysis management system and method for substation channel cover plates, and the specific technical solutions are as follows:

[0007] A data acquisition and intelligent analysis management method for substation channel cover plates, including:

[0008] Embedding an acceleration sensor and a resistive strain gauge in the cover plate to collect vibration data and strain data of the cover plate;

[0009] Extracting the change in vibration direction and vibration decay time of the acceleration sensor, calculating the deformation rebound time after the cover plate stops being stressed, and extracting the characteristic values of the strain data, including the maximum strain amplitude, strain change rate, and strain gradient in the stress concentration area, to characterize the stress deformation characteristics of the cover plate under the stress state;

[0010] Based on the stress deformation characteristics and rebound characteristics of the cover plate, constructing a cover plate damage assessment model based on stress deformation and deformation rebound time using a neural network model to evaluate the damage degree of the cover plate;

[0011] Fuse the characteristics of vibration data and strain data, predict the change trend of damage degree based on the machine learning model, and calculate the acceleration trend of the cover plate damage;

[0012] Based on the currently evaluated damage degree of the cover plate and the acceleration trend of the cover plate damage, calculate the remaining maintenance time of the cover plate. When the remaining maintenance time is lower than the set threshold, trigger the operation and maintenance warning information.

[0013] Preferably, install the acceleration sensor at the geometric center of the lower surface of the cover plate, form a rigid connection with the cover plate, and be used to collect the vibration data of the cover plate;

[0014] Set a strain rosette composed of multiple resistive strain gauges at the geometric center of the cover plate, and arrange unidirectional strain gauges along the diagonal direction at the four corners of the cover plate respectively, so as to synchronously obtain the strain data of the central area and the corner area of the cover plate.

[0015] Preferably, after removing the high-frequency noise from the collected strain data of the cover plate by using a Butterworth low-pass filter, traverse and search within a fixed time window to obtain the maximum strain amplitude time series, and use the multi-scale central difference method to calculate the corresponding strain change rate, and characterize the magnitude and rate of the force deformation of the cover plate through the maximum strain amplitude and the strain change rate.

[0016] Preferably, establish an interpolation strain field according to the strain values of the four corner measurement points, and obtain the strain gradients in each direction after fitting by the least square method;

[0017] Construct a covariance matrix for the acceleration data and perform eigenvalue decomposition to determine the main vibration direction, obtain the vibration decay time through the free vibration envelope decay analysis, and then integrate the acceleration twice and combine it with high-pass filtering to determine the cover plate deformation rebound time.

[0018] Preferably, combine the maximum strain amplitude, the strain change rate, the corner strain gradient, the vibration direction angle, the vibration decay time and the deformation rebound time into a multi-dimensional feature vector; use the comprehensive damage index as the supervision label, and adopt the support vector regression algorithm to construct the cover plate damage assessment model.

[0019] Preferably, optimize the penalty coefficient and kernel function width of the support vector regression model through grid search combined with cross-validation, use the root mean square error as the performance evaluation index, and use the trained support vector regression model as the cover plate damage assessment model;

[0020] Establish a confidence evaluation mechanism. When the cover plate damage assessment model outputs the prediction result of the cover plate damage, calculate the confidence of the cover plate damage prediction result according to the threshold of the distance between the real-time sample and the support vector in the training set.

[0021] Preferably, the comprehensive damage index is stored in chronological order to form a damage time series; a long short-term memory network (LSTM) model is used to construct a damage trend prediction model. The input of the LSTM model is the historical damage time series, and the output is the sequence of the damage trend of the cover plate in the future time period.

[0022] Preferably, the damage acceleration factor is obtained by calculating the damage growth rate and the change rate of the cover plate.

[0023] When the change of the damage growth rate exceeds the set threshold within a continuous number of moments, it is determined as the acceleration stage; if the prediction errors in a continuous number of moments exceed the preset threshold, the online incremental learning mechanism is triggered to update the parameters of the long short-term memory network (LSTM) model using the latest monitoring data.

[0024] Preferably, the current damage rate is linearly fitted according to the comprehensive damage index within the preset recent time window, a damage development function is constructed, and the remaining maintenance time for the cover plate to reach the maintenance threshold is obtained.

[0025] The remaining maintenance time is corrected based on the damage acceleration factor and by introducing a safety margin coefficient, and multi-level warning thresholds are set.

[0026] When the corrected remaining maintenance time is not greater than the corresponding warning threshold, a hierarchical operation and maintenance warning message including the cover plate identification, the current damage degree, the damage acceleration trend, and the recommended maintenance time window is generated, and a time interval is set to suppress repeated warnings for the same cover plate.

[0027] A data acquisition and intelligent analysis management system for the substation channel cover plate, which is used for the data acquisition and intelligent analysis management method of the substation channel cover plate, includes: a data acquisition module, a feature extraction module, a cover plate damage assessment module, a cover plate damage trend prediction module, and a warning module;

[0028] The data acquisition module embeds an acceleration sensor and a resistive strain gauge in the cover plate to collect the vibration data and strain data of the cover plate.

[0029] The feature extraction module extracts the change of the vibration direction and the vibration decay time of the acceleration sensor, calculates the deformation rebound time after the cover plate stops being stressed, and extracts the characteristic values of the strain data, including the maximum strain amplitude, the strain change rate, and the strain gradient in the stress concentration area, to characterize the stress deformation characteristics of the cover plate under the stressed state.

[0030] The cover plate damage assessment module constructs a cover plate damage assessment model based on the stress deformation and the deformation rebound time based on a neural network model according to the stress deformation characteristics and the rebound characteristics of the cover plate to evaluate the damage degree of the cover plate.

[0031] The cover plate damage trend prediction module integrates the characteristics of vibration data and strain data, predicts the change trend of the damage degree based on a machine learning model, and calculates the cover plate damage acceleration trend.

[0032] The warning module calculates the remaining maintenance time of the cover plate based on the currently evaluated damage degree of the cover plate and the cover plate damage acceleration trend, and triggers an operation and maintenance warning message when the remaining maintenance time is lower than the set threshold.

[0033] Advantages of this application: By embedding sensors to collect the vibration and strain data of the cover plate, this application realizes real-time and accurate monitoring of the stress, vibration and deformation states of the cover plate, provides a comprehensive and reliable raw data basis for subsequent analysis, and effectively improves the comprehensiveness and accuracy of data collection.

[0034] Through multi-dimensional feature extraction of acceleration and strain signals, this application realizes in-depth quantitative analysis of the stress response, deformation process and rebound performance of the cover plate, provides rich physical quantity feature supports for damage characteristic discrimination, and enhances the refined characterization ability of the cover plate structure state.

[0035] By using a machine learning model to model the stress deformation and rebound characteristics of the cover plate, this application realizes quantitative and intelligent evaluation of the damage degree of the cover plate, can timely detect early signs of deterioration of the cover plate performance, and improves the accuracy and intelligent level of damage identification.

[0036] By integrating multi-source features and performing trend prediction, this application can identify the accelerating development trend of cover plate damage in advance, realize dynamic tracking of the cover plate health state and risk prediction, and provide a scientific basis for subsequent operation and maintenance decisions.

[0037] Based on the damage degree and acceleration trend, this application intelligently calculates the remaining maintenance time of the cover plate and actively pushes a warning at the critical moment, effectively realizing the active management and risk prevention of cover plate operation and maintenance, and greatly improving the foresight and efficiency of operation and maintenance work.

[0038] This application realizes the integration of substation channel cover plate state monitoring, health assessment and intelligent operation and maintenance management; the technical solution of this application can timely and accurately master the stress and damage states of the cover plate, predict potential structural risks of the cover plate in advance, significantly improve the scientificity, timeliness and economy of cover plate maintenance, effectively ensure the safety of substation equipment and operation and maintenance personnel, and provide strong support for the intelligent management and refined operation and maintenance of power facilities. Description of the Drawings

[0039] Figure 1 It is a flow chart of the substation channel cover plate data collection and intelligent analysis management method provided by this application.

[0040] Figure 2Flow chart for extracting strain data and vibration data features provided by this application;

[0041] Figure 3 Flow chart for constructing a cover plate damage assessment model provided by this application;

[0042] Figure 4 Flow chart for constructing a cover plate damage trend prediction provided by this application;

[0043] Figure 5 Flow chart for calculating and warning the remaining maintenance time of the cover plate provided by this application;

[0044] Figure 6 Structural diagram of the substation channel cover plate data acquisition and intelligent analysis management system provided by this application. Detailed implementation manners

[0045] To make the above objects, features, and advantages of this application more obvious and understandable, the following will provide a detailed description of the specific implementation manners of this application in conjunction with the accompanying drawings of the specification.

[0046] In the following description, many specific details are set forth to facilitate a full understanding of this application. However, this application can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of this application. Therefore, this application is not limited by the specific embodiments disclosed below.

[0047] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of this application. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that excludes other embodiments.

[0048] Embodiment 1:

[0049] Refer to Figures 1 to 5 , the first embodiment of this application, provides a substation channel cover plate data acquisition and intelligent analysis management method.

[0050] Step 1: Install an acceleration sensor and a resistive strain gauge on the cover plate to collect the vibration data and strain data of the cover plate.

[0051] Install a high-precision triaxial acceleration sensor at the geometric center position of the lower surface of the cover plate to ensure that the vibration response of the cover plate when subjected to external forces can be accurately captured; Exemplarily, the acceleration sensor is rigidly connected to the cover plate through epoxy resin adhesive or a preset card slot to ensure the lossless transmission of vibration signals.

[0052] A resistive strain gauge is arranged on the lower surface of the cover plate. Exemplarily, a foil strain gauge with a resistance value of 350 Ω is used. A strain rosette is arranged at the geometric center position of the cover plate, which is composed of three resistive strain gauges arranged in the directions of 0°, 45°, and 90° for measuring the principal strain direction and magnitude at the geometric center point. The arrangement of the strain rosette can accurately obtain the two-dimensional strain state in the central area of the cover plate, and the principal strain value can be obtained through calculation. and , where , in the formula 、 are the normal strains in the directions of 0° and 90° respectively, , is the shear strain in the 45° direction; the complex stress state at the center of the cover plate can be comprehensively reflected through the arrangement method of the strain rosette.

[0053] For the key stress concentration areas of the cover plate, taking the four corner areas of the cover plate as stress concentration points, a single-direction strain gauge is arranged at the edge position of each corner of the cover plate, and the long axis direction of the strain gauge is arranged along the diagonal direction of the cover plate. By arranging the strain gauges at the edge of the cover plate corners, the stress concentration phenomenon in the corner areas of the cover plate can be effectively monitored, because the corners are often the positions where damage first appears when the cover plate is loaded.

[0054] In this step, sensors are arranged on the cover plate material and data is collected to establish a complete monitoring system for the force deformation and vibration response of the cover plate, which is used to monitor the quantitative characterization of the mechanical behavior of the cover plate under load, providing high-quality basic data support for subsequent damage assessment and operation and maintenance management.

[0055] Step 2: Extract the change in the vibration direction and the vibration decay time of the acceleration sensor, calculate the deformation rebound time after the cover plate stops being stressed, and extract the characteristic values of the strain data, including the maximum strain amplitude, strain change rate, and strain gradient in the stress concentration area, to characterize the force deformation characteristics of the cover plate under the stressed state; see Figure 2 , which is the flow chart for extracting the characteristic values of the strain data and vibration data in this step.

[0056] The collected original strain data is preprocessed, and a Butterworth low-pass filter is used to filter out high-frequency noise to ensure the complete retention of effective strain data while eliminating measurement noise interference.

[0057] For the filtered strain data , search for the maximum strain amplitude within the time interval , where is the independent variable representing time; is the length of the analysis time window, is the subscript index; the maximum strain amplitude is defined as , the maximum strain amplitude time series is obtained by traversing the entire monitoring period , where is the th maximum strain amplitude in the maximum strain amplitude time series. The eigenvalue of the maximum strain amplitude directly reflects the maximum deformation degree of the cover plate in each time period, providing a quantitative index for evaluating the bearing capacity of the cover plate.

[0058] The central difference method is used to calculate the strain rate of change. For the th discrete sampled strain data point , the strain rate of change is calculated by the formula , where is the sampling time interval, and is the difference step parameter.

[0059] To improve the calculation accuracy, the multi-scale difference method is adopted to calculate the strain rate of change at respectively, and then the final strain rate of change is obtained through weighted average, where is the strain rate of change at the th scale, and is the weight coefficient and satisfies . The magnitude of the strain rate of change reflects the severity of the load action. An excessive strain rate of change indicates that the cover plate is subjected to impact loads, and the impact load form has a more significant impact on the damage of the cover plate.

[0060] Calculate the strain gradient in the stress concentration area. Using the strain measurement point data at the four corners of the cover plate, a local coordinate system is established, where the origin is located at the geometric center of the cover plate. The strain field function is constructed by bilinear interpolation, where the coefficient is obtained by least squares fitting. Tests are carried out at different positions in the cover plate to obtain the actual strain values and the positions of each strain measurement point at the four corners, and the parameter calculation data in the least squares fitting strain field function is obtained.

[0061] The component of the strain gradient in the direction is , and the component in the direction is ; A square calculation domain with a side length of is defined in each corner area. Multiple calculation points are evenly selected in the square calculation domain, and the strain gradient of each point is calculated. The average value of is taken as the strain gradient eigenvalue of this corner; A high strain gradient value indicates that there is a stress concentration phenomenon in this area, which is a potential damage development area.

[0062] Analyze the change in the vibration direction of the cover plate by analyzing the acceleration components measured by a triaxial acceleration sensor through a covariance matrix , and construct a covariance matrix , where , is the analysis duration; perform eigenvalue decomposition on the covariance matrix , where is the eigenvector matrix, is the eigenvalue diagonal matrix; the eigenvector corresponding to the largest eigenvalue is the main vibration direction, are the three components of the eigenvector , corresponding to the X, Y, and Z axes of the triaxial acceleration respectively

[0063] Define the vibration direction angle to include and , , is the angle with the vertical direction, and , is the horizontal projection azimuth angle; continuously monitor the changes in the vibration direction angle and , where represents the change in the angle between the main vibration direction at the current moment and the main vibration direction at the initial moment in the z-axis direction, represents the change in the direction of the main vibration direction at the current moment and the main vibration direction at the initial moment in the horizontal projection plane (x-y plane), represents the inclination angle of the main vibration direction in the z-axis direction at the current moment, represents the inclination angle of the main vibration direction in the z-axis direction at the initial moment, represents the direction angle of the main vibration direction in the x-y plane at the current moment, represents the direction angle of the main vibration direction in the x-y plane at the initial moment; thus, identify the change in the support condition of the cover plate and the non-uniform degradation of the structural stiffness

[0064] Extract the vibration decay time based on the free vibration response analysis. At the moment when the load removal is identified, define as the moment when the acceleration amplitude drops from the peak to the threshold , where is the threshold coefficient, represents the acceleration peak. Start extracting the free vibration segment data from and calculate the vibration envelope ; at the same time, the vibration envelope satisfies the exponential decay form , perform logarithmic transformation on the vibration envelope , is the initial value of the vibration envelope, and the damping ratio is determined by linear regression and the natural frequency .

[0065] Define the vibration decay time as the time required for the vibration envelope to decay to times the initial value, that is , where is the vibration decay time, is the decay proportionality coefficient; an increase in the vibration decay time indicates an increase in structural damping, usually caused by an increase in energy dissipation due to internal damage to the cover plate.

[0066] Calculate the cover plate deformation rebound time, and perform integral processing on the acceleration data to obtain the displacement response; use the trapezoidal integration method for numerical integration, the velocity , the displacement , where and are the initial velocity and initial displacement, is the total number of discrete time points, is the acceleration value sampled at the th time point, at the th time point the sampled velocity value.

[0067] To eliminate integral drift, use a high-pass filter to process the displacement signal and identify the moment when the load is completely removed, and define as the moment when the acceleration returns to the static noise level, and start monitoring the displacement recovery process from moment.

[0068] Define the static equilibrium position as , where is the waiting time, is the average time window; define the deformation rebound time as the time required for the displacement to recover from the maximum value to , that is , where is the convergence judgment threshold, is the deformation rebound time, is the time required for the displacement to recover from the maximum value to . The extension of the deformation rebound time reflects the decline in the elastic recovery ability of the cover plate and is an important indicator of the degradation of the cover plate material properties.

[0069] This step realizes the comprehensive quantitative characterization of the static strain characteristics, dynamic vibration characteristics and elastic recovery characteristics of the cover plate; it reflects the mechanical performance state of the cover plate from the perspective of multi-dimensional characteristic parameters.

[0070] Step 3: Based on the force-induced deformation characteristics and resilience characteristics of the cover plate, construct a cover plate damage assessment model based on the force-induced deformation and deformation resilience time using a neural network model to evaluate the degree of cover plate damage; see Figure 3 , which is the flowchart for constructing the cover plate damage assessment model in this step.

[0071] Combine the maximum strain amplitude , strain change rate , strain gradient , vibration direction change angle and , vibration decay time and the deformation resilience time to form a multi-dimensional feature vector .

[0072] To establish a supervised learning model, define a quantitative index for the degree of cover plate damage, and use the comprehensive damage index as the damage label, defined as , where is the ultimate strain value of the cover plate, is the standard deformation resilience time of the new cover plate, is the critical value of the strain gradient, is the weight coefficient and satisfies ; the value range of the comprehensive damage index is , where indicates no damage, indicates severe damage. Through a combination of expert evaluation and on-site inspection, accurately label the damage labels for historical data to form a training dataset , where is the number of samples.

[0073] Use the support vector regression SVR model to construct a cover plate damage assessment model, and select the radial basis function (RBF) as the kernel function; the SVR model can effectively handle non-linear relationships and has good robustness to noise, making it suitable for dealing with complex mapping relationships in cover plate damage assessment.

[0074] Use the method of grid search combined with cross-validation to optimize the hyperparameters of the SVR model, uniformly sample on the logarithmic scale to form a parameter grid; use k-fold cross-validation to evaluate the performance of each group of parameters, and the performance index is the root mean square error RMSE; select the parameter combination that minimizes the cross-validation RMSE as the optimal parameter.

[0075] Extract the multi-dimensional feature vector from the real-time collected cover plate monitoring data, and input it into the trained SVR model to obtain the damage prediction value , where is the th support vector, is the kernel function, represents the inner product between the support vector and the input multi-dimensional feature vector, is the bias term, is the Lagrange multiplier, is the number of support vectors, is the support vector index.

[0076] To verify the reliability of the cover plate damage assessment model, a confidence evaluation mechanism is established. By analyzing the distance between a new sample and the support vectors of the training set when it indicates that the new sample deviates from the training data distribution. is the distance threshold between the new sample and the support vectors of the training set, and a low-confidence warning is issued. Constructing a confidence evaluation mechanism can identify the applicable boundary of the model and avoid incorrect damage assessment results under abnormal working conditions.

[0077] In this step, by constructing a cover plate damage assessment model based on machine learning, an accurate mapping from multi-dimensional monitoring data of the cover plate to the damage degree is realized, overcoming the limitations of traditional threshold judgment methods. The constructed cover plate damage assessment model can comprehensively consider the static deformation characteristics and dynamic response characteristics of the cover plate and accurately identify different degrees of cover plate damage.

[0078] Step 4: Integrate the characteristics of vibration data and strain data, predict the change trend of the damage degree based on a machine learning model, and calculate the acceleration trend of the cover plate damage; see Figure 4 for the flow chart of constructing the cover plate damage trend prediction in this step.

[0079] Store the comprehensive damage index in chronological order to form a damage time series , where is the th monitoring time, is the total number of monitorings.

[0080] To accurately capture the damage evolution law, a long short-term memory network (LSTM) model is used to construct a damage trend prediction model; the input of the LSTM model is the historical damage time series, and the output of the LSTM is the sequence of the cover plate damage trend in the future time period.

[0081] Construct an input feature matrix , where is of length Historical damage time series; the LSTM network can effectively learn the long-term dependence relationship of damage development and overcome the gradient disappearance problem of traditional time series models.

[0082] Analyze the damage acceleration trend based on the second derivative of the damage growth rate and calculate the damage growth rate , and the damage acceleration growth rate is defined as ; when consecutive time points satisfy , it is determined as the damage acceleration growth state, where is the acceleration growth rate threshold; define the damage acceleration factor , where is the average damage growth rate of the current period, is the average damage growth rate of the reference period; when , it indicates that the cover plate damage enters the rapid growth stage, is the damage acceleration threshold.

[0083] To improve the adaptive ability of the prediction model, an online learning mechanism is constructed. Define the prediction error , when the prediction errors of consecutive time points satisfy , is the actual damage value at time , is the predicted damage value at time , is the error threshold, triggering model update. Adopt an incremental learning strategy, add the latest data points in the time series to the training set, and at the same time remove the earliest data points in the training set time series to keep the training set size constant.

[0084] This step constructs a damage trend prediction, damage judgment, and failure risk assessment based on deep learning, realizing accurate prediction of the future health state of the cover plate and risk quantification assessment. The method of this step can not only predict the development trend of damage but also identify the critical period of cover plate damage acceleration.

[0085] Step 5: Calculate the remaining maintenance time of the cover plate based on the currently evaluated cover plate damage degree and the cover plate damage acceleration trend. When the remaining maintenance time is lower than the set threshold, trigger the operation and maintenance warning information; see Figure 5 , which is the flow chart of the calculation and warning of the remaining maintenance time of the cover plate in this step.

[0086] Based on the current damage degree and the damage acceleration trend , construct a calculation model for the remaining maintenance time of the cover plate.

[0087] ​​Exemplarily, define the damage development function , where is the current damage rate, represents the time period starting from the current moment; the current damage rate is obtained by linearly fitting the damage data of the last time sampling points, , where and are the mean values of time and damage respectively, is the time sampling point index; set the damage threshold that the cover plate needs to maintain as , and obtain the remaining maintenance time by solving the equation :

[0088]

[0089] The calculation method based on the physical degradation law can relatively accurately predict the time point when the cover plate reaches the maintenance standard, providing a reliable basis for formulating the maintenance plan.

[0090] Considering the uncertainty of damage development, introduce a safety margin coefficient to correct the remaining maintenance time. Define the corrected remaining maintenance time , where is the safety margin coefficient, and the value range is .

[0091] The safety margin coefficient is dynamically adjusted according to the damage acceleration trend, , where is the historical maximum damage acceleration; when the damage acceleration trend is large, the safety margin coefficient is small, and the remaining maintenance time is correspondingly shortened to ensure that maintenance can be carried out in time when the cover plate degrades rapidly. The dynamic adjustment mechanism based on the safety margin coefficient can improve the safety and adaptability of the cover plate maintenance time prediction.

[0092] Set the time threshold for triggering the operation and maintenance warning, and the threshold comprehensively considers the maintenance preparation time and the safety reserve time; when the corrected remaining maintenance time is lower than the warning threshold , that is , trigger the operation and maintenance warning.

[0093] The warning information includes the cover plate identification, the current damage degree , the damage acceleration trend , the estimated remaining maintenance time and the recommended maintenance time window , where , , is the half-width of the time window.

[0094] Exemplarily, the warning levels can be divided into three levels according to the urgency of the remaining maintenance time: emergency warning ( ), important warning ( ), and general warning ( ), where is the classification threshold. The hierarchical warning mechanism can help the operation and maintenance personnel reasonably arrange the maintenance priorities and optimize the resource allocation.

[0095] After the warning is triggered, the warning information is pushed to the operation and maintenance management platform. To avoid information overload caused by frequent warnings, a warning suppression mechanism is set up to filter out repeated warnings for the same cover plate within the time interval , and only the warning information of the highest level is retained; the intelligent warning release mechanism ensures that the operation and maintenance personnel can obtain key information in a timely and accurate manner and can also avoid unnecessary disturbances caused by information redundancy.

[0096] In this step, through the technical solutions of calculating the remaining maintenance time, considering the time correction of the safety margin, triggering warnings by comparing with the classification threshold, and intelligently providing warning information, the accurate prediction and timely warning of the maintenance requirements of the cover plate are realized; it can effectively avoid the safety risks brought by the failure of the cover plate, and at the same time optimize the scheduling of maintenance resources through a reasonable warning mechanism, significantly improving the operation and maintenance management efficiency and reliability of the substation channel cover plate.

[0097] Embodiment 2:

[0098] Referring to Figure 6 , for the second embodiment of the present application, a data acquisition and intelligent analysis management system for substation channel cover plates is provided.

[0099] The system includes: a data acquisition module, a feature extraction module, a cover plate damage assessment module, a cover plate damage trend prediction module, and a warning module.

[0100] The data acquisition module embeds an acceleration sensor and a resistive strain gauge in the cover plate to collect the vibration data and strain data of the cover plate.

[0101] The feature extraction module extracts the vibration direction change and vibration decay time of the acceleration sensor, calculates the deformation rebound time after the cover plate stops being stressed, and extracts the characteristic values of the strain data, including the maximum strain amplitude, strain change rate, and strain gradient in the stress concentration area, to characterize the stress deformation characteristics of the cover plate under the stressed state.

[0102] The cover plate damage assessment module constructs a cover plate damage assessment model based on the stress deformation and deformation rebound time based on a neural network model according to the stress deformation characteristics and rebound characteristics of the cover plate, and evaluates the degree of cover plate damage.

[0103] The cover plate damage trend prediction module fuses the characteristics of vibration data and strain data, predicts the change trend of the damage degree based on a machine learning model, and calculates the cover plate damage acceleration trend.

[0104] The warning module calculates the remaining maintenance time of the cover plate based on the currently evaluated damage degree of the cover plate and the cover plate damage acceleration trend, and triggers an operation and maintenance warning message when the remaining maintenance time is lower than the set threshold.

[0105] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For another example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some communication interfaces. The indirect coupling or communication connection of the devices or units can be in an electrical, mechanical or other form.

[0106] The embodiments of the present application have been described above with reference to the accompanying drawings. However, the present application is not limited to the above specific implementation manners. The above specific implementation manners are only illustrative and not restrictive. Under the inspiration of the present application, those of ordinary skill in the art can also make changes, modifications, substitutions and variations to the above embodiments without departing from the purpose of the present application and the scope protected by the claims. These all belong to the protection scope of the present application.

Claims

1. A data acquisition and intelligent analysis management method for substation channel covers, characterized in that, Including: Embed an acceleration sensor and a resistive strain gauge in the cover plate to collect the vibration data and strain data of the cover plate; Extract the vibration direction change and vibration decay time of the acceleration sensor, calculate the deformation rebound time after the cover plate stops being stressed, and extract the characteristic values of the strain data, including the maximum strain amplitude, strain change rate, and strain gradient in the stress concentration area, to characterize the force deformation characteristics of the cover plate under the stressed state; According to the force deformation characteristics and rebound characteristics of the cover plate, construct a cover plate damage assessment model based on force deformation and deformation rebound time based on a neural network model to evaluate the damage degree of the cover plate; Fuse the characteristics of vibration data and strain data, predict the change trend of the damage degree based on a machine learning model, and calculate the cover plate damage acceleration trend; Based on the currently evaluated damage degree of the cover plate and the cover plate damage acceleration trend, calculate the remaining maintenance time of the cover plate. When the remaining maintenance time is lower than the set threshold, trigger an operation and maintenance warning message.

2. The method for data acquisition and intelligent analysis management of the substation trench cover plate according to claim 1, wherein Install the acceleration sensor at the geometric center of the lower surface of the cover plate to form a rigid connection with the cover plate for collecting the vibration data of the cover plate; Set a strain rosette composed of multiple resistive strain gauges at the geometric center of the cover plate, and arrange unidirectional strain gauges along the diagonal direction at the four corners of the cover plate to synchronously obtain the strain data of the central area and the corner area of the cover plate.

3. The data acquisition and intelligent analysis management method for the substation trench cover plate according to claim 2, characterized in that, After removing the noise from the collected cover plate strain data using a Butterworth low-pass filter, traverse and search within a fixed time window to obtain the maximum strain amplitude time series, and use the multi-scale central difference method to calculate the corresponding strain change rate. The magnitude and rate of force deformation of the cover plate are characterized by the maximum strain amplitude and the strain change rate.

4. The data acquisition and intelligent analysis management method for the substation trench cover plate according to claim 3, characterized in that, Establish an interpolation strain field based on the strain values at the four corner measurement points, and obtain the strain gradients in each direction after fitting by the least squares method; Construct a covariance matrix for the acceleration data and perform eigenvalue decomposition to determine the main vibration direction. Obtain the vibration decay time through the free vibration envelope decay analysis, and then perform double integration on the acceleration and combine it with high-pass filtering to determine the deformation rebound time of the cover plate.

5. The data acquisition and intelligent analysis management method for the substation trench cover plate according to claim 4, characterized in that, Combine the maximum strain amplitude, strain change rate, corner strain gradient, vibration direction angle, vibration decay time, and deformation rebound time into a multi-dimensional feature vector; Use the comprehensive damage index as the supervision label and adopt the support vector regression algorithm to construct a cover plate damage assessment model.

6. The method for data acquisition and intelligent analysis management of the substation trench cover plate according to claim 5, characterized in that, Optimize the penalty coefficient and kernel function width of the support vector regression model through grid search combined with cross-validation, use the root mean square error as the performance evaluation index, and use the trained support vector regression model as the cover plate damage assessment model; Establish a confidence evaluation mechanism. When the cover plate damage assessment model outputs the prediction result of cover plate damage, calculate the confidence of the cover plate damage prediction result according to the threshold of the distance between the real-time sample and the support vector in the training set.

7. The method for data acquisition, intelligent analysis and management of the substation trench cover according to claim 6, characterized in that Store the comprehensive damage index in chronological order to form a damage time series; adopt a long short-term memory network (LSTM) model to construct a damage trend prediction model. The input of the LSTM model is the historical damage time series, and the output is the sequence of the cover plate damage trend in the future time period.

8. The method for data acquisition and intelligent analysis management of the substation trench cover according to claim 7, wherein, Obtain the damage acceleration factor by calculating the cover plate damage growth rate and change rate; When the change rate of damage growth exceeds the set threshold within consecutive multiple moments, it is determined as the acceleration stage; if the prediction error for consecutive multiple moments exceeds the preset threshold, the online incremental learning mechanism is triggered to update the parameters of the long short-term memory network (LSTM) model using the latest monitoring data.

9. The method for data acquisition, intelligent analysis and management of the substation trench cover according to claim 8, characterized in that, Based on the linear fitting of the comprehensive damage index within the preset recent time window, the current damage rate is obtained, a damage development function is constructed, and the remaining maintenance time until the cover plate reaches the maintenance threshold is calculated. The remaining maintenance time is corrected according to the damage acceleration factor and by introducing a safety margin coefficient, and multi-level warning thresholds are set. When the corrected remaining maintenance time is not greater than the corresponding warning threshold, a hierarchical operation and maintenance warning message including the cover plate identification, current damage degree, damage acceleration trend, and recommended maintenance time window is generated, and a time interval is set to suppress repeated warnings for the same cover plate.

10. A substation channel cover data acquisition and intelligent analysis management system for implementing the substation channel cover data acquisition and intelligent analysis management method according to any one of claims 1 to 9, characterized in that It includes: a data acquisition module, a feature extraction module, a cover plate damage assessment module, a cover plate damage trend prediction module, and a warning module; The data acquisition module embeds an acceleration sensor and a resistive strain gauge in the cover plate to collect the vibration data and strain data of the cover plate. The feature extraction module extracts the change in the vibration direction and the vibration decay time of the acceleration sensor, calculates the deformation rebound time after the cover plate stops being stressed, and extracts the characteristic values of the strain data, including the maximum strain amplitude, strain change rate, and strain gradient in the stress concentration area, to characterize the stress deformation characteristics of the cover plate under the stressed state. The cover plate damage assessment module constructs a cover plate damage assessment model based on stress deformation and deformation rebound time based on a neural network model according to the stress deformation characteristics and rebound characteristics of the cover plate to evaluate the damage degree of the cover plate. The cover plate damage trend prediction module fuses the characteristics of the vibration data and strain data, predicts the change trend of the damage degree based on a machine learning model, and calculates the cover plate damage acceleration trend. The warning module calculates the remaining maintenance time of the cover plate based on the currently evaluated damage degree of the cover plate and the cover plate damage acceleration trend, and triggers an operation and maintenance warning message when the remaining maintenance time is lower than the set threshold.

Images