Operation and maintenance method of complete switchgear based on DMC materials

Abstract

The present invention relates to the field of operation and maintenance technology of complete switchgear, and discloses an operation and maintenance method of complete switchgear based on DMC materials, including: experimentally measuring the insulation strength, thermal conductivity, arc resistance and other key performance parameters of DMC materials, and constructing an aging characteristic model; calculating the material health index based on real-time collected operating status data, and dynamically generating degradation trends and remaining life in combination with a time series prediction algorithm; generating a maintenance plan through an optimization model, and introducing a risk weight coefficient to achieve a dynamic balance between maintenance costs and equipment reliability. The present invention significantly improves the accuracy and response speed of equipment status monitoring through dynamic fusion of material performance modeling and multi-source data; through the multi-dimensional collaboration of trend prediction and maintenance optimization strategies, a closed-loop management of the entire life cycle is formed, and intelligent upgrades of operation and maintenance under complex working conditions are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of operation and maintenance of complete switchgear, and in particular to an operation and maintenance method of a complete switchgear based on DMC materials, which is used to improve the intelligence level of equipment status monitoring, degradation trend prediction and maintenance optimization. Background Art

[0002] With the rapid development of the power system, complete sets of switchgear, as an important device to ensure the transmission and distribution of electric energy, have a direct impact on the stability and safety of the power grid due to its operational reliability and maintenance efficiency. In order to achieve safe and efficient operation of equipment, the industry usually adopts a combination of condition monitoring and regular maintenance. Condition monitoring mainly uses sensors to collect equipment operating data, such as current, voltage, temperature and other parameters, combined with simple threshold judgment or offline analysis models to determine whether the equipment is abnormal. Regular maintenance is based on the statistical laws of equipment usage time or operating environment to formulate maintenance plans to minimize equipment failure rates. However, these traditional methods have the following problems:

[0003] 1. Limitations of condition monitoring: Existing condition monitoring methods mostly use static threshold judgments, which cannot reflect the operating status and degradation trend of the equipment in real time, resulting in difficulty in timely identification of potential problems. Especially under complex working conditions (such as high temperature and high humidity environments), equipment performance changes have significant nonlinear characteristics, and traditional methods are difficult to adapt dynamically.

[0004] 2. Uncertainty in life prediction: In order to extend equipment life and optimize maintenance efficiency, the industry has gradually introduced life prediction models based on historical data. However, such models are usually based on fixed mathematical relationships or a single data source, lack dynamic feedback on real-time operating data, and cannot accurately capture the accelerated deterioration trend of equipment or the risk of sudden failures. The accuracy and reliability of the prediction results are low.

[0005] 3. Low adaptability of maintenance plans: Most existing maintenance plans are based on statistical laws or empirical rules, and fail to fully consider the actual operating status and failure risks of the equipment. This may lead to excessive maintenance and increased costs, or equipment failure due to insufficient maintenance, further increasing the risk of power grid operation.

[0006] In order to solve the above problems, the industry has gradually tried to introduce data-driven dynamic optimization methods. For example, sensor data is analyzed through artificial intelligence algorithms to dynamically generate trend forecasts and maintenance plans. However, although these methods have improved certain prediction capabilities, their prediction results still have large errors due to the failure to combine the aging characteristic model of the material itself. In addition, these methods have limited ability to integrate multi-source data, fail to establish a closed-loop management system throughout the life cycle, and are difficult to meet the high-precision status monitoring and reliability requirements under complex working conditions.

[0007] Therefore, how to comprehensively improve the accuracy and adaptability of equipment status monitoring, life prediction and maintenance optimization by combining multi-source data fusion, dynamic prediction and intelligent optimization has become a technical problem to be solved by the present invention. Summary of the invention

[0008] The technical problem solved by the present invention is to provide an operation and maintenance method for a complete set of switchgear based on DMC materials in response to the defects existing in the above-mentioned prior art, so as to solve the problems of inaccurate status monitoring, unreliable trend prediction and inflexible maintenance optimization proposed in the above-mentioned background technology.

[0009] In order to solve the above technical problems, the technical solution adopted by the present invention is as follows:

[0010] A method for operating and maintaining a complete set of switchgear based on DMC materials, comprising the following steps:

[0011] Step 1, determination and modeling of key performance parameters of DMC materials:

[0012] The key performance parameters of DMC materials were determined experimentally, and the key performance parameters included:

[0013] Dielectric strength: Measure the breakdown voltage of the material by using a withstand voltage test device while gradually increasing the voltage, record the maximum breakdown voltage value, and calculate its ratio to the material thickness;

[0014] Thermal conductivity: Use a steady-state thermal conductivity tester to measure the heat flux density of the material under constant temperature gradient conditions and calculate the thermal conductivity;

[0015] Arc resistance: measured by a high-voltage arc test device in the current range of 50 mA to 500 mA, and the ratio of arc action time to surface breakdown time is recorded;

[0016] The key performance parameters are associated with experimental environmental conditions, including temperature, humidity and mechanical stress, to establish an aging characteristic model of DMC materials. The calculation formula of the model is:

[0017] in: For time Key performance parameter values ​​at the time of is the initial performance value; is the aging factor, which indicates that the material properties change with temperature. ,humidity and mechanical stress the rate of decay due to changes in is a time variable, which refers to the time interval from the start of the experiment to the observation time; the aging factor is determined by the following steps:

[0018] Experimental data collection: Accelerated aging tests were performed at different temperatures ranging from 20°C to 100°C, different humidities ranging from 30% to 90%, and different mechanical stresses ranging from 0 to 10 MPa to measure the changes in key performance parameters over time;

[0019] Data fitting: The experimental data were processed using the nonlinear regression algorithm of the least squares method to fit the specific values ​​of the aging factors. The correlation coefficient of the fitting results was not less than 0.95.

[0020] Step 2: Equipment operation status data collection:

[0021] Sensors are arranged at key locations of the complete switchgear to collect real-time operating status data of the equipment, including: operating current and operating voltage, collected by current sensors and voltage sensors; ambient temperature and humidity, collected by temperature and humidity sensors; mechanical vibration parameters, collected by triaxial acceleration sensors; local temperature rise of DMC materials, collected by thermistors; and surface electric field strength of materials, collected by surface electric field detectors;

[0022] Step 3, material health index calculation:

[0023] Based on the aging characteristic model of the DMC material and the real-time collected operating status data, the health index of the DMC material is calculated. The health index is referred to as MHI, and the calculation formula is:

[0024] in: For current time The key performance parameter values ​​at , calculated by the aging characteristic model; is the initial performance value; It is the local temperature rise of DMC material, collected by thermistor; It is the maximum temperature rise allowed by the material, which is determined by durability testing in a high temperature environment and is a constant value; It is the fluctuation value of the electric field intensity on the surface of the material, collected by the surface electric field detector; It is the maximum surface electric field strength allowed by the material, which is determined by high electric field breakdown test and is a constant value;

[0025] Step 4, Deterioration trend prediction and remaining life calculation:

[0026] Based on the historical data and real-time updated data of the health index, a time series prediction model is constructed to predict the degradation trend and remaining service life of the DMC material. The remaining service life is calculated by the following formula:

[0027] in: The remaining useful life of the material; is the current health index; It is the safety threshold of the health index, which is set by the equipment operation safety standard; is the rate of change of the health index over time, calculated using the difference formula using historical health index data;

[0028] Step 5, Intelligent maintenance plan generation:

[0029] Dynamically generate a maintenance plan based on the degradation trend prediction and the remaining life calculation results, wherein the maintenance plan includes maintenance time, maintenance location and maintenance measures;

[0030] The maintenance plan is generated with the following optimization goals:

[0031] in: Maintenance cost is calculated based on the equipment maintenance history data and weighted by failure frequency and repair cost; For failure risk, the failure risk value is calculated using the failure probability model based on the health index and degradation trend;

[0032] is the risk weight coefficient, which is obtained by fitting the multivariate linear regression model of historical failure data;

[0033] Step 6: Cloud platform monitoring and feedback:

[0034] Upload the equipment's operating status data, health index, and degradation trend prediction results to the cloud platform through the communication module;

[0035] In the cloud platform: the uploaded data is normalized and stored, and the degradation trend is updated in real time based on the time series prediction model; the maintenance plan is automatically adjusted according to the updated degradation trend and remaining life, and sent to the device end through the communication module to guide equipment operation and maintenance.

[0036] As a further solution of the present invention, during the measurement of the thermal conductivity of the DMC material, the temperature gradient is controlled in the range of 10°C to 50°C, and the measurement accuracy of the heat flux is not less than 0.01 W / m².

[0037] As a further embodiment of the present invention, the aging factor The fitting process further includes a data cleaning step, wherein the data cleaning includes abnormal data removal, noise signal filtering, and data normalization processing to improve the accuracy of the fitting model.

[0038] As a further aspect of the present invention, the material health index During the calculation process, the maximum temperature rise allowed for the material It is determined by long-term durability tests conducted in the range of 80°C to 150°C for a minimum of 100 hours.

[0039] As a further solution of the present invention, the maximum surface electric field strength allowed by the DMC material is It is measured by gradually increasing the electric field strength at a constant voltage until breakdown, with the rate of increase of the electric field strength being 10 kV / mm.

[0040] As a further aspect of the present invention, the remaining life During the calculation process, the rate of change of the health index over time It is calculated by the linear difference method of the historical data of the health index, with a difference interval of 1 hour.

[0041] As a further solution of the present invention, the failure risk value The calculation process uses a probability evaluation model based on a Bayesian network, and the input parameters of the model include the health index , Deterioration trend prediction results and remaining life .

[0042] As a further solution of the present invention, during the generation of the maintenance plan, the maintenance cost It is calculated by the following formula:

[0043] in: is the probability of failure occurrence, which is determined by the failure risk value Decide; The cost of repairing the fault; For preventive maintenance costs.

[0044] As a further embodiment of the present invention, the risk weight coefficient It is obtained by multivariate linear regression fitting of historical failure data and maintenance cost data, and satisfies the following fitting equation:

[0045] in: , , is the fitting coefficient, obtained by fitting historical data; The cost of repairing the fault; is the failure risk value.

[0046] Compared with the prior art, the beneficial effects of the present invention are as follows: by constructing an aging characteristic model based on the key performance parameters of DMC materials, a dynamic quantitative relationship between the physical properties of the material and the environmental conditions (such as temperature, humidity and mechanical stress) is realized, and the monitoring capability of the material degradation behavior is significantly improved. At the same time, combined with the calculation method of the health index, multi-source operating status data (such as local temperature rise and surface electric field strength) is organically combined with material performance modeling to generate a comprehensive index that can accurately characterize the health status of the equipment in real time; on the basis of the material health index, the degradation trend and remaining life of the equipment are dynamically generated through a time series prediction algorithm and a change rate calculation model; the trend prediction model not only improves the accuracy of the prediction through the deep integration of real-time data and historical data, but also can predict the degradation trend and remaining life of the equipment in the operating environment or equipment status. The prediction results can be adjusted quickly when there are drastic changes, ensuring the reliability and adaptability of the prediction results; by building an intelligent maintenance optimization model based on health index, degradation trend and remaining life, a dynamic adjustment mechanism of risk weight coefficient is introduced to achieve a precise balance between maintenance cost and equipment reliability; this optimization strategy can not only dynamically adjust maintenance time, location and measures according to the real-time operation status of the equipment, but also effectively reduce the waste of maintenance resources and the possibility of risk out of control through mathematical models, so that the generation of maintenance plans is upgraded from the traditional fixed rule method to a dynamic decision-making mode, which can accurately match maintenance needs in different operation scenarios; and through the closed-loop management of the entire life cycle based on the cloud platform, centralized management and closed-loop optimization of equipment operation status, health assessment, trend prediction and maintenance decision-making are achieved. Cloud-based data processing not only improves the efficiency of monitoring and analysis, but also realizes real-time response and adaptive maintenance throughout the life cycle of the equipment through dynamic update functions.

[0047] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0049] Figure 1 This is a block diagram of the operation and maintenance system of a complete set of switchgear based on DMC materials of the present invention.

[0050] Figure 2 This is a closed-loop management flow chart of the cloud platform of the present invention. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] See also Figure 1 — Figure 2 In an embodiment of the present invention, a method for operating and maintaining a complete set of switchgear based on DMC materials comprises the following steps:

[0053] Step 1, determination and modeling of key performance parameters of DMC materials:

[0054] The key performance parameters of DMC materials were determined experimentally, and the key performance parameters included:

[0055] Dielectric strength: Measure the breakdown voltage of the material by using a withstand voltage test device while gradually increasing the voltage, record the maximum breakdown voltage value, and calculate its ratio to the material thickness;

[0056] Thermal conductivity: Use a steady-state thermal conductivity tester to measure the heat flux density of the material under constant temperature gradient conditions and calculate the thermal conductivity;

[0057] Arc resistance: measured by a high-voltage arc test device in the current range of 50 mA to 500 mA, and the ratio of arc action time to surface breakdown time is recorded;

[0058] The key performance parameters are associated with the experimental environmental conditions, including temperature, humidity and mechanical stress, and an aging characteristic model of the DMC material is established. The calculation formula of the model is:

[0059] in: For time Key performance parameter values ​​at the time of is the initial performance value; is the aging factor, which indicates that the material properties change with temperature. ,humidity and mechanical stress the rate of decay due to changes in is a time variable, which refers to the time interval from the start of the experiment to the observation time; the aging factor is determined by the following steps:

[0060] Experimental data collection: Accelerated aging tests were performed at different temperatures ranging from 20°C to 100°C, different humidities ranging from 30% to 90%, and different mechanical stresses ranging from 0 to 10 MPa to measure the changes in key performance parameters over time;

[0061] Data fitting: The experimental data were processed using the nonlinear regression algorithm of the least squares method to fit the specific values ​​of the aging factors. The correlation coefficient of the fitting results was not less than 0.95.

[0062] Step 2: Equipment operation status data collection:

[0063] Sensors are arranged at key parts of the complete set of switchgear to collect the operating status data of the equipment in real time. The data include: working current and working voltage, which are collected by current sensors and voltage sensors; ambient temperature and humidity, which are collected by temperature and humidity sensors; mechanical vibration parameters, which are collected by triaxial acceleration sensors; local temperature rise of DMC materials, which are collected by thermistors; material surface electric field strength, which is collected by surface electric field detectors; all the above sensors are integrated in the sensor acquisition module, which is responsible for the management and data collection of various sensors to ensure the real-time and accurate transmission of equipment operating status data. The sensor acquisition module not only monitors various parameters in real time, but also performs preliminary processing and filtering on the data to ensure the reliability and validity of the data. Subsequently, the data will be transmitted to the data processing module of the cloud platform for further analysis and processing; that is, after the data upload module is transmitted to the cloud platform, the data will be further processed by the cloud platform data processing module. The processing methods include storing, filtering, formatting, and preliminary analysis of the collected data, as well as removing abnormal data and noise to provide clean raw data for subsequent modules such as health index calculation and degradation trend prediction.

[0064] Step 3, material health index calculation:

[0065] Based on the aging characteristic model of the DMC material and the real-time collected operating status data, the health index of the DMC material is calculated. The health index is referred to as MHI, and the calculation formula is:

[0066] in: For current time The key performance parameter values ​​at , calculated by the aging characteristic model; is the initial performance value; It is the local temperature rise of DMC material, collected by thermistor; It is the maximum temperature rise allowed by the material, which is determined by durability testing in a high temperature environment and is a constant value; It is the fluctuation value of the electric field intensity on the surface of the material, collected by the surface electric field detector; The maximum surface electric field strength allowed by the material is determined by the high electric field breakdown test and is a constant value. In this step, after being processed by the cloud platform data processing module, the data will be input into the health index calculation module, which calculates the health index (MHI) of the device based on the predetermined aging model and real-time operation data. By comprehensively considering factors such as temperature, humidity, and mechanical stress, the health index calculation module reflects the health status of the device in real time and helps identify potential failure risks.

[0067] Step 4, Deterioration trend prediction and remaining life calculation:

[0068] Based on the historical data and real-time updated data of the health index, a time series prediction model is constructed to predict the degradation trend and remaining service life of the DMC material. The remaining service life is calculated by the following formula:

[0069]

[0070] in: The remaining useful life of the material; is the current health index; It is the safety threshold of the health index, which is set by the equipment operation safety standard; The rate of change of the health index over time is calculated using the differential formula through the historical health index data; in this step, the degradation trend prediction module combines the health index data generated by the health index calculation module, that is, the degradation trend prediction module predicts the future degradation trend of the equipment through time series analysis and regression model. This module can predict the change trend of equipment performance based on historical data and real-time updated data, and provide data support for the remaining life assessment; then the remaining service life (RUL) of the equipment can be calculated through the remaining life assessment module based on the prediction results output by the degradation trend prediction module. This module estimates the remaining life of the equipment based on the change of the health index of the equipment, the degradation rate and the operating conditions of the equipment, and issues a maintenance warning.

[0071] Step 5, Intelligent maintenance plan generation:

[0072] Dynamically generate a maintenance plan based on the degradation trend prediction and the remaining life calculation results, wherein the maintenance plan includes maintenance time, maintenance location and maintenance measures;

[0073] The maintenance plan is generated with the following optimization goals:

[0074] in: Maintenance cost is calculated based on the equipment maintenance history data and weighted by failure frequency and repair cost; For failure risk, the failure risk value is calculated using the failure probability model based on the health index and degradation trend; is the risk weight coefficient, which is obtained by fitting the multivariate linear regression model of historical failure data. In this step, the intelligent maintenance optimization module can generate an optimized maintenance plan based on the equipment's health index, degradation trend, and remaining life assessment results.

[0075] Step 6: Cloud platform monitoring and feedback:

[0076] Upload the equipment's operating status data, health index, and degradation trend prediction results to the cloud platform through the communication module;

[0077] In the cloud platform: the uploaded data is normalized and stored, and the degradation trend is updated in real time based on the time series prediction model; the maintenance plan is automatically adjusted according to the updated degradation trend and remaining life, and sent to the device side through the communication module to guide the operation and maintenance of the equipment; in this step, after the maintenance plan is generated, the maintenance plan is transmitted to the device side, and the equipment status is monitored in real time, and the operating parameters of the equipment (such as temperature, operating voltage, etc.) are automatically adjusted to ensure that the equipment operates according to the predetermined maintenance plan, forming an on-site equipment control and feedback mechanism, and during the operation of the equipment, the feedback data will be continuously uploaded to the cloud platform to optimize the next maintenance decision.

[0078] As a further solution of the present invention, during the measurement of the thermal conductivity of the DMC material, the temperature gradient is controlled in the range of 10°C to 50°C, and the measurement accuracy of the heat flux is not less than 0.01 W / m².

[0079] As a further embodiment of the present invention, the aging factor The fitting process further includes a data cleaning step, wherein the data cleaning includes abnormal data removal, noise signal filtering, and data normalization processing to improve the accuracy of the fitting model.

[0080] As a further aspect of the present invention, the material health index During the calculation process, the maximum temperature rise allowed for the material It is determined by long-term durability tests conducted in the range of 80°C to 150°C for a minimum of 100 hours.

[0081] As a further solution of the present invention, the maximum surface electric field strength allowed by the DMC material is It is measured by gradually increasing the electric field strength at a constant voltage until breakdown, with the rate of increase of the electric field strength being 10 kV / mm.

[0082] As a further aspect of the present invention, the remaining life During the calculation process, the rate of change of the health index over time It is calculated by the linear difference method of the historical data of the health index, with a difference interval of 1 hour.

[0083] As a further solution of the present invention, the failure risk value The calculation process uses a probability evaluation model based on a Bayesian network, and the input parameters of the model include the health index , Deterioration trend prediction results and remaining life .

[0084] As a further solution of the present invention, during the generation of the maintenance plan, the maintenance cost It is calculated by the following formula:

[0085] in: is the probability of failure occurrence, which is determined by the failure risk value Decide; The cost of repairing the fault; For preventive maintenance costs.

[0086] As a further embodiment of the present invention, the risk weight coefficient It is obtained by multivariate linear regression fitting of historical failure data and maintenance cost data, and satisfies the following fitting equation:

[0087] in: , , is the fitting coefficient, obtained by fitting historical data; The cost of repairing the fault; is the failure risk value.

[0088] Embodiment 1:

[0089] This embodiment is based on a typical power distribution system scenario, and specifically describes the practical application of the operation and maintenance method of a complete set of switchgear based on DMC materials in the present invention, showing how to achieve the entire process of equipment status monitoring, degradation trend prediction and intelligent maintenance optimization under complex operating conditions.

[0090] For example, in the power distribution system of a large industrial park, in order to ensure the reliability of continuous power supply, it is necessary to perform real-time status monitoring and maintenance optimization on multiple sets of switchgear. The operating environment of the power distribution system is complex, including high temperature (50°C), high humidity (90% relative humidity) and frequent mechanical vibration. Traditional methods are difficult to reflect the operating status and degradation trend of the equipment in real time, and it is difficult to balance maintenance costs and risk control. This embodiment specifically implements the technical solution of the present invention through the following steps:

[0091] First, material performance measurement and aging characteristics modeling, that is, experimental measurement of key performance parameters such as insulation strength, thermal conductivity and arc resistance of DMC (unsaturated polyester glass fiber reinforced molding compound) material used in switchgear; insulation strength: the breakdown voltage of the material under gradually increasing voltage conditions is measured by voltage resistance testing equipment, and the experimental environment temperature is set to 25°C to 100°C, and the ratio of breakdown voltage to material thickness is used as a performance indicator; thermal conductivity: a steady-state thermal conductivity tester is used to measure the thermal conductivity of the material in the temperature range of 20°C to 50°C; arc resistance: a high-voltage arc test device is used to measure the arc breakdown time of the material under different current conditions; and based on the above experimental data, combined with environmental conditions (including temperature, humidity and mechanical stress), an aging characteristic model of DMC material is established by least squares fitting. The model can dynamically quantify the degradation law of material performance with environmental changes.

[0092] Then, the operation status data is collected, and a sensor array is arranged at the key parts of the switchgear, including: temperature and humidity sensors: real-time collection of ambient temperature and humidity data; current and voltage sensors: monitoring the operating current and voltage changes of the equipment; three-axis acceleration sensors: capturing mechanical vibration signals during equipment operation; thermistors and surface electric field detectors: respectively measuring the local temperature rise and surface electric field strength of the DMC material. Through the Internet of Things communication module, the collected data is uploaded to the cloud platform in real time to form a dynamic database of the equipment's operating status.

[0093] The health index is then calculated, that is, the material health index (MHI) is calculated based on the collected operating status data and aging characteristic model. The specific steps include: combining real-time data with the aging model to dynamically update the current performance value of the material; using the health index formula to comprehensively consider the impact of local temperature rise and surface electric field intensity fluctuations on material performance. The health index ranges from 0 to 1. The closer the value is to 1, the closer the material performance is to the initial state. Under high temperature and high humidity experimental conditions, the health index dynamically reflects the performance degradation trend of the equipment in harsh environments, providing an accurate basis for subsequent predictions.

[0094] Then, the degradation trend prediction and remaining life assessment are carried out, that is, by using the time series prediction model and the change rate calculation method, the degradation trend of the material is predicted and the remaining life (RUL) is assessed; the degradation trend is generated based on the historical health index data and the real-time change data, and is dynamically adjusted by the time series algorithm; the remaining life calculation is assessed by the health index change rate formula. For example, after 72 consecutive hours of operation, the prediction results show that the rate of decline of the health index of the equipment has increased significantly, and the remaining life assessment is 30 hours, at which time the maintenance warning is automatically triggered. And the intelligent maintenance plan is generated and optimized, that is, the intelligent maintenance plan is generated according to the health index, degradation trend and remaining life data. The maintenance optimization model introduces a risk weight coefficient to dynamically balance the maintenance cost and the failure risk: the maintenance cost is calculated by the historical failure frequency and repair cost; the failure risk is assessed by the Bayesian network model, and the risk value is adjusted in real time in combination with the degradation trend and the health index; the optimization objective function dynamically weighs the maintenance cost and the failure risk, and outputs the specific maintenance time, location and measures. For example, when the equipment health index drops to 0.3, the system automatically generates a maintenance plan, and recommends partial replacement of specific parts within the next 24 hours to avoid large-scale equipment failure.

[0095] In addition, the cloud platform can be used for closed-loop management, that is, the cloud platform can analyze and store the equipment's operating status, health index and prediction results in real time. The platform integrates data normalization module, time series analysis module and dynamic optimization module: the data normalization module standardizes multi-source data to improve analysis accuracy; the time series analysis module is used to update degradation trend prediction in real time; the dynamic optimization module adjusts the maintenance plan according to the latest data and sends it to the on-site operators through the Internet of Things communication module. In addition, the platform also has a historical data comparison function to further optimize the fitting accuracy of the aging model.

[0096] In this embodiment, through the dynamic fusion of multi-source data and the closed-loop management of the cloud platform, the technical solution of the present invention significantly improves the accuracy and real-time performance of equipment status monitoring. Compared with the traditional solution based on static models, the accuracy of degradation trend prediction is improved by more than 30%, and the warning time of the remaining life of the equipment is advanced by about 25%. At the same time, the introduction of risk weight coefficients makes the maintenance plan more flexible and efficient, reduces the maintenance cost by about 20%, and significantly improves the reliability of system operation.

[0097] Embodiment 2:

[0098] This embodiment uses the operation and maintenance of switchgear in a certain power company's UHV transmission line as a specific application scenario to verify the actual effect and superiority of the technical solution of the present invention. By conducting comparative experiments with the prior art methods, the experimental background and settings are as follows: During the operation of the UHV transmission line, the complete set of switchgear is subjected to high current, high voltage and frequent mechanical vibrations for a long time, which is prone to degradation of insulation performance due to aging of DMC materials, thereby causing equipment failure or line power outages. Traditional maintenance methods mostly use preventive maintenance plans based on fixed time intervals, or passive response methods based on simple status monitoring thresholds, but these methods are difficult to accurately predict degradation trends and remaining life, and have high maintenance costs and insufficient reliability.

[0099] In order to verify the technical effect of the present invention, three sets of switchgear in actual operation were selected for experiments:

[0100] Equipment A: adopts the traditional maintenance method based on fixed maintenance cycle;

[0101] Equipment B: adopts maintenance method based on condition monitoring and static life model;

[0102] Equipment C: adopts the operation and maintenance method based on DMC material aging characteristic modeling, health index calculation, degradation trend prediction and intelligent maintenance optimization described in the present invention.

[0103] The experimental steps include status monitoring and data collection: the same type of sensors are arranged for the three devices to collect operating environment data (such as temperature and humidity, mechanical vibration) and equipment operating status data (such as current, voltage, local temperature rise and surface electric field strength). The sensor acquisition frequency is set to once per minute, and the data is uploaded to the experimental monitoring system through the Internet of Things communication module. The monitoring data of device C is combined with the aging characteristic model of the present invention to calculate the health index (MHI), which reflects the performance status changes of the DMC material in real time.

[0104] It also includes degradation trend prediction and life assessment, that is, for equipment B, a static life model is used to predict degradation trends based on historical operating data; for equipment C, a combination of real-time monitoring data and historical health index data is used to predict degradation trends using time series analysis and dynamically assess the remaining life (RUL).

[0105] Then, a maintenance plan is generated, i.e., for equipment A, maintenance is performed according to a fixed cycle (e.g., every 6 months); for equipment B, the maintenance time and scope are determined based on the prediction results of the static life model; for equipment C, a maintenance plan is generated based on the intelligent maintenance optimization model of the present invention, the health index, degradation trend and risk weight coefficient are comprehensively considered, and the maintenance time and maintenance location are dynamically adjusted on the basis of balancing maintenance cost and failure risk.

[0106] Experimental results and analysis: After 6 consecutive months of operation experiments, the failure rate, maintenance cost and prediction error during the operation of the equipment were recorded. The results show that: Equipment C dynamically reflects the operating status of the equipment through real-time health index calculation, which can more accurately identify the degradation trend of equipment performance compared with traditional static models or stateless monitoring methods. During the 6-month operation, the average health index of equipment C was always higher than 0.8, indicating that the equipment performance remained stable, while the state degradation of equipment A and equipment B was not discovered in time; the remaining life prediction error of equipment C was only ±5%, significantly better than ±20% of equipment B. This high-precision prediction comes from the dynamic fusion of health index, time series analysis and real-time data, which can capture potential failure risks in advance; and the intelligent maintenance plan of equipment C effectively avoids over-maintenance or under-maintenance while balancing maintenance costs and failure risks. Compared with equipment A, the maintenance cost of equipment C was reduced by 33%, and failures during operation were completely avoided; and during the experiment, equipment C achieved zero failure rate through degradation trend prediction and precise maintenance optimization. However, due to the failure to accurately predict the degradation trend, equipment A and equipment B experienced 3 and 2 failures respectively, resulting in the interruption of the transmission line.

[0107] This experiment fully verifies that the dynamic monitoring of the health index in a complex operating environment of the present invention significantly improves the accuracy of status assessment and effectively makes up for the shortcomings of the static model in the prior art; high-precision degradation trend prediction and remaining life assessment make maintenance decisions more reliable; intelligent maintenance optimization strategy achieves reduced maintenance costs and improved reliability.

[0108] Embodiment 3:

[0109] In the present embodiment, the error control and influence range of the experimental environment parameters are further clarified for the determination of the key performance parameters of the DMC material: temperature control range: 20°C to 100°C, with a fluctuation error of no more than ±0.5°C. Humidity control range: 30% to 90% relative humidity, with a fluctuation error of no more than ±2%. Mechanical stress loading range: 0 to 10 MPa, with a loading accuracy of 0.01 MPa. Through the above experimental conditions, it is ensured that during the determination of the key performance parameters, the influence of the experimental environment on the fitting results of the aging factor is minimized, thereby improving the accuracy and repeatability of the fitting.

[0110] At the same time, in order to improve the accuracy of multi-source data collection, a dynamic calibration mechanism is introduced in the embodiment during the arrangement of each sensor, such as calibration frequency: once every 48 hours; calibration method: using a reference sensor to collect environmental data, comparing the reference data with the experimental data, and using an error normalization method (see the formula below) to correct the collected data:

[0111] in is the calibrated data, The original data collected is is the deviation between the measured value and the reference value, This method ensures that the accuracy of multi-source data such as temperature, humidity and electric field strength is within ±1%.

[0112] And to further optimize the health index calculation formula, a dynamic error compensation factor can be used ( ):

[0113]

[0114] in: ; ;

[0115] Sensor error ( ): Errors caused by insufficient sensor acquisition accuracy;

[0116] Model error ( ): The error caused by the aging factor fitting accuracy being lower than 0.95.

[0117] Environmental error ( ): Error introduced by temperature and humidity fluctuations.

[0118] By introducing the compensation factor , dynamically correct the error sources in the calculation of the health index and make the changing trend of the health index more in line with the actual situation.

[0119] At the same time, in order to solve the problem of insufficient data processing in the aging factor fitting process, this embodiment adds a data cleaning step, which is divided into the following three parts, namely, abnormal data removal: identifying and removing abnormal values ​​in the experimental data through the 3σ rule; noise signal filtering: using a Gaussian filter to smooth the experimental data and remove high-frequency noise signals; normalization processing: normalizing the experimental data according to the following formula:

[0120] in, is the original experimental data, and are the minimum and maximum values ​​of the data respectively; after the above processing, the correlation coefficient of the fitting model is increased to 0.98, which significantly enhances the reliability of the aging factor model; this embodiment introduces the LSTM (long short-term memory) network model to replace the traditional time series model, thereby enhancing the dynamic adaptability of degradation trend prediction, namely, input data: historical health index data and real-time update data; model structure: 3-layer LSTM unit, time window is set to 24 hours; output result: degradation trend curve and remaining life prediction value; experimental results show that the prediction error of the LSTM model is reduced to ±3%, which is significantly improved compared with the traditional time series algorithm (error ±10%); it is suitable for complete sets of switchgear operating in industrial parks, which need to monitor the performance changes of DMC materials in real time; it is suitable for high electric field strength environments such as ultra-high voltage transmission line switchgear operation.

[0121] Embodiment 4:

[0122] In order to solve the problem of decreased accuracy caused by the accumulation of errors in multi-source data in the calculation of health index, this embodiment introduces a dynamic error compensation factor , compensate for the error source and optimize the calculation formula. The calculation formula for the compensation factor is:

[0123] in: ; : Sensor acquisition error, determined by experimental calibration value; : Model fitting error, based on the fitting correlation coefficient calculate; : Environmental errors, including dynamic fluctuations in temperature, humidity and mechanical stress.

[0124] After introducing the compensation factor, the calculation formula of the health index is updated as follows:

[0125] in, : The performance parameter value at the current time t, calculated by the aging characteristic model; : Initial performance value, derived from standard performance testing of the material; : Local temperature rise of the material, measured by thermistor; : The maximum allowable temperature rise of the material, determined by long-term durability tests; : The fluctuation value of the electric field intensity on the surface of the material is collected in real time by the electric field detector; : The maximum electric field strength allowed by the material, determined by a high electric field breakdown test. The addition of ensures that the accuracy of health index calculation is further improved, especially in high humidity or high vibration environments, the error is reduced by more than 30%.

[0126] Based on the aforementioned time series prediction model, this embodiment introduces a long short-term memory (LSTM) network model to enhance the dynamic adaptability of degradation trend prediction. The model fusion steps are as follows:

[0127] Data input: Historical health index data ; Real-time updated environmental parameters , , ; Current material performance parameters Model structure: Use three layers of LSTM units, each layer contains 256 neurons; the time window is set to 24 hours. Output result: Degradation trend curve ; Remaining life prediction value The experimental results show that the integration of LSTM network and time series prediction model further improves the accuracy of degradation trend prediction and reduces the prediction error to ±3%.

[0128] And in the generation of maintenance plans, the risk weight coefficient can be improved to dynamically balance maintenance costs and failure risks By combining historical operation data with real-time fault assessment, dynamic adjustment The value range of is as follows:

[0129] in: : Fault repair cost, determined by historical maintenance data; : Failure risk value, evaluated based on Bayesian network model; : Regression fitting coefficients are obtained by training with historical data; the dynamic adjustment mechanism of risk weight coefficients optimizes the objective function It is more flexible and the adaptability of maintenance plans is significantly enhanced; and the maintenance plan optimization process can be further refined into the following steps, namely data analysis: the health index and degradation trend prediction results are updated to the cloud platform in real time; the model parameters are corrected in combination with historical data; plan generation: the maintenance time, location and measures are determined by optimizing the objective function; sensors are preferentially arranged at key equipment locations to reduce monitoring costs; implementation feedback: the cloud platform automatically sends the maintenance plan to the equipment end, and feedbacks the data after the maintenance is completed for further optimization of the model.

[0130] Embodiment 5:

[0131] The operation and maintenance method of the complete switchgear based on DMC materials in this embodiment further improves the efficiency and reliability of equipment maintenance by enhancing real-time monitoring of the equipment operation status, accurate prediction of degradation trends and intelligent maintenance decisions. For example, in order to ensure that the aging characteristic model of the DMC material is more accurate, the key performance parameters of the material are more detailed experimentally designed and measured in the embodiment. By conducting accelerated aging tests under different environmental conditions (temperature: 20°C to 100°C, humidity: 30% to 90%, mechanical stress: 0 to 10 MPa), real-time data collection, and least squares method for data fitting, the fitting coefficient is ensured to be no less than 0.98, thereby improving the accuracy of the aging factor.

[0132] In the calculation of the material health index, for example, the calculation method of the health index MHI is optimized, and this embodiment further describes the source, meaning and function of each variable. For example, The local temperature rise of DMC material is collected by thermistor. It is the maximum temperature rise allowed for the material, and the value is determined through long-term durability tests (80°C to 150°C, test time is not less than 100 hours). is the fluctuation value of the electric field intensity on the material surface, measured in real time by the surface electric field detector. It is the maximum electric field strength of the material, measured in the high electric field breakdown test.

[0133] The accurate calculation of aging factors, for example, in the fitting process of aging factors, not only experimental data collection and data fitting are introduced, but also data cleaning steps are added, Gaussian filtering is used to remove noise signals, and the 3σ rule is used to eliminate abnormal data. This can effectively ensure the accuracy of the fitting process and make the prediction ability of the aging model more accurate.

[0134] In its degradation trend prediction and remaining life assessment, for example, by combining the health index with the time series analysis model, the degradation trend of the DMC material is dynamically predicted. The calculation method of the remaining life RUL clarifies the implementation path of each step: by differential calculation of historical health index data and real-time data, the rate of change of the health index over time is accurately evaluated, ensuring the accuracy of the remaining life assessment; and in the generation of intelligent maintenance plans, based on the degradation trend prediction and remaining life assessment results, this embodiment proposes an intelligent maintenance plan generation system. When generating a maintenance plan, we not only consider the maintenance cost of the equipment, but also combine the failure risk for dynamic balance. The optimization goal is to minimize the comprehensive maintenance cost and failure risk. The maintenance cost is calculated based on factors such as the equipment's fault repair cost and maintenance frequency, while the failure risk is derived through the equipment health index and degradation trend prediction.

[0135] In order to dynamically balance maintenance costs and failure risks, we introduced a risk weight coefficient, which was obtained through historical data analysis and reflects the relative importance between maintenance costs and failure risks. Specifically, its value is obtained through regression analysis of historical data, which takes into account the impact of equipment repair costs and failure risks on maintenance plans. This method can dynamically adjust according to the actual operation of the equipment, thereby optimizing the maintenance plan.

[0136] In terms of closed-loop management of the cloud platform, real-time monitoring and feedback of equipment are achieved through the cloud platform. All collected equipment operation status data, health index and degradation trend prediction results are uploaded to the cloud platform, where data normalization and storage are performed. Through real-time updates based on time series analysis, maintenance plans are dynamically adjusted and sent to the equipment end to ensure the accuracy and efficiency of equipment maintenance.

[0137] And in this embodiment, the data acquisition module is responsible for real-time acquisition of parameters such as the working current, voltage, temperature and humidity, mechanical vibration, local temperature rise and electric field strength of the equipment; the model calculation module uses the aging characteristic model to calculate the health index of the DMC material based on the collected data; the prediction and evaluation module combines the health index data and uses the time series prediction model to calculate the degradation trend and remaining life; the optimization and decision-making module generates an intelligent maintenance plan based on the maintenance cost and failure risk using the optimization algorithm; the cloud platform management module processes and stores the data and dynamically adjusts the maintenance strategy based on the real-time data.

[0138] Through this embodiment, the technology solves many technical problems in traditional condition monitoring and maintenance methods, significantly improving the accuracy of equipment condition monitoring, the reliability of degradation trend prediction, and the intelligence level of maintenance plans. By combining data-driven and intelligent optimization, this solution provides a closed-loop management solution throughout the life cycle, which can effectively improve equipment operating efficiency, reduce maintenance costs, and ensure equipment reliability.

[0139] Embodiment 6:

[0140] In order to improve the calculation accuracy of the DMC material health index, this embodiment introduces an error compensation mechanism in the health index calculation process to eliminate the impact of environmental changes (such as temperature, humidity and mechanical stress) on sensor data. The health index reflects the overall operating status of the device, and the sensor data is corrected to ensure that the health index accurately reflects the actual performance of the DMC material.

[0141] For example, the error compensation mechanism, the sensor data such as temperature rise and electric field strength collected in real time may be affected by environmental fluctuations and sensor errors. Therefore, in this embodiment, the error compensation factor is calculated by comparing the sensor data with the reference value in real time. This compensation factor is dynamically adjusted according to environmental changes to ensure that the measured data can more accurately reflect the actual state of the material. For example, the temperature rise compensation factor will be corrected in real time according to the change of ambient temperature, thereby ensuring that the calculation result of the health index is more accurate.

[0142] In the correction and calculation of its health index, the correction of the health index not only depends on the sensor data of temperature rise and electric field strength, but also combines the dynamic adjustment of environmental data. By comparing the real-time collected data with the initial performance value of the equipment, combined with the dynamic correction factor, the current health index is finally calculated. For example, during the operation of the equipment, by comparing the difference between each data collection value and the initial value, and then correcting it according to the compensation factor, it ensures that the health index can accurately reflect the health status of the equipment.

[0143] In terms of degradation trend prediction and remaining life assessment, this embodiment further optimizes the degradation trend prediction model, uses comprehensive data sources for multi-dimensional analysis, and ensures that the prediction of degradation trend and remaining life is more accurate. That is, in order to more accurately predict the degradation trend of the equipment, this embodiment adopts a time series analysis method, combined with historical health index and real-time environmental data, to dynamically predict the degradation trend of the equipment. Through real-time updates and continuous learning of historical data, the prediction model can be adjusted in time to avoid error accumulation in long-term predictions and provide more flexible prediction capabilities. For example, changes in temperature and humidity will affect the operating status of the equipment. The model will automatically adjust and feed back to the prediction results in real time to ensure that the degradation trend is more accurate; the calculation of the remaining life is based on the rate of change of the health index and the real-time operating status of the equipment. By analyzing the historical health index change data, combined with environmental factors such as temperature, humidity, and mechanical stress, the remaining life is dynamically evaluated. The calculation of the remaining life not only considers the changes in the health index, but also the influence of environmental factors, so as to obtain a more accurate life prediction. This process can help equipment maintenance personnel predict the risk of equipment failure in advance and take effective measures.

[0144] In the process of generating an intelligent maintenance plan, this embodiment combines the health index, fault risk assessment and equipment degradation trend, and adopts a dynamic optimization algorithm to make the maintenance plan more intelligent and adjust the maintenance time and maintenance measures according to the real-time health status of the equipment. Traditional maintenance methods are often based on fixed cycles for maintenance, which easily leads to waste of resources or insufficient maintenance. However, this embodiment generates a more accurate maintenance plan by combining the health index, degradation trend and fault risk. The plan dynamically adjusts the maintenance time, maintenance parts and maintenance measures through multi-dimensional optimization. For example, when the health index of the equipment drops to a predetermined threshold, the system automatically calculates and generates a maintenance plan, accurately points out the parts that need maintenance, and arranges maintenance operations in advance; this embodiment introduces a dynamic adjustment mechanism for the risk weight coefficient. By combining historical fault data with the real-time health index, the failure risk of each component is dynamically calculated. The risk weight coefficient is calculated through a multivariate regression model and can be adjusted according to the current health status of the equipment and historical data, so that the maintenance plan achieves the best balance between risk and cost. For example, for a key component, if its failure risk is high, the system will prioritize its maintenance.

[0145] At the same time, in order to improve the prediction accuracy and data quality, this embodiment has carried out a detailed design of the data cleaning process to ensure the accuracy and consistency of the data, that is, after data collection, all sensor data are first screened through the outlier removal step. The 3σ rule is used to remove data that does not conform to the normal range, and the signal is smoothed by a Gaussian filter to remove high-frequency noise to ensure the stability and validity of the data; due to the different dimensions of different sensor data, it may affect the comparison and analysis of the data. Therefore, normalization is used in data processing to convert all data into the same dimension for subsequent processing and analysis. The process of data normalization is carried out through a standardized method to ensure that various types of data can be compared fairly during the analysis process.

[0146] And in this embodiment, on the other hand, in terms of hardware requirements, it is necessary to deploy a variety of sensors at key parts of the equipment, including temperature and humidity sensors, current and voltage sensors, vibration sensors, temperature rise sensors, and electric field strength sensors. Among them, key parts refer to components that directly affect operational safety and performance, such as electrical connection parts, insulating material areas, mechanical moving parts, cooling systems, and battery energy storage units. These parts need to be deployed with sensors to monitor their working status in real time. These sensors will continue to monitor the operating status of the equipment in real time and transmit data to the cloud platform. The data collected by the sensors of the equipment will be uploaded to the cloud platform through the Internet of Things communication module. The cloud platform will process and analyze the data in real time, and calculate indicators such as health index and degradation trend. At the same time, the system will update the maintenance plan of the equipment according to the real-time data, and send the updated results to the on-site operators. In the cloud platform, based on the real-time collected data, combined with historical health indexes and environmental factors, the system will automatically generate and adjust the maintenance plan. The maintenance plan includes specific maintenance parts, time, and measures to ensure that maintenance operations can effectively prevent equipment failures and reduce maintenance costs.

[0147] Embodiment 7:

[0148] In order to solve the data error problem in the current technical solution regarding the calculation of the DMC material health index (MHI) and the prediction of the degradation trend, this embodiment proposes an optimization solution to improve the accuracy of the health index and the decision-making efficiency of equipment operation and maintenance through a dynamic error compensation mechanism and an advanced prediction model. The optimization solution includes the following steps:

[0149] Step 1: Introduction and optimization of error compensation factors. In order to solve the sensor data errors caused by environmental factors such as temperature, humidity, and mechanical stress, a dynamic error compensation factor is introduced in the embodiment. The calculation of this factor is carried out by the following formula:

[0150] in, It is composed of the following errors: : The error of the sensor itself, obtained through regular calibration and accuracy testing; : The fitting error of the model is evaluated by the correlation coefficient of the aging factor fitting to ensure that its fitting degree is not less than 0.98; :The environmental error caused by temperature and humidity fluctuations is compensated by real-time monitoring of environmental conditions and dynamic adjustments; the introduction of this compensation factor effectively reduces the measurement error caused by environmental factors and further improves the accuracy of health index calculation.

[0151] Step 2: Optimize the calculation of health index. In order to fully consider the influence of various errors in the calculation of health index MHI, the embodiment optimizes the health index formula. Combined with the original health index formula, a new calculation formula is obtained:

[0152] In this formula: Indicates the performance parameter value at the current time t, which is calculated by the aging characteristic model of DMC material; is the initial property value of the material, derived from standard testing; It is the local temperature rise of the material, collected by the thermistor sensor; is the maximum allowable temperature rise of the material, determined by high temperature durability testing; It is the fluctuation value of the electric field intensity on the material surface, which is collected in real time by the surface electric field detector; The maximum electric field strength of the material is determined by a high electric field breakdown test. Through this optimized formula, the health index not only takes into account the changes in the physical properties of the material, but also effectively corrects the deviations caused by environmental fluctuations and sensor errors, thereby ensuring the accuracy of the health index.

[0153] Step 3: Data cleaning and model optimization. In order to further improve the accuracy of the data, this embodiment also cleans and preprocesses the data involved in the aging factor fitting process of the DMC material. First, the 3σ rule is used to identify and eliminate abnormal data to ensure the representativeness and authenticity of the data; then, a Gaussian filter is used to smooth the collected high-frequency noise signal; finally, data normalization is performed to unify sensor data from different sources into a standardized form to improve the comparability of the data and the accuracy of subsequent analysis. The optimized aging factor fitting process ensures that the correlation coefficient of the fitting results is not less than 0.98, significantly improves the accuracy of the aging model, and ensures the reliability of the prediction results.

[0154] Step 4: Optimization of degradation trend and remaining life prediction. In order to further improve the prediction accuracy of degradation trend, this embodiment adopts LSTM (Long Short-Term Memory) network model instead of traditional time series prediction model. LSTM network has powerful nonlinear modeling ability and can better adapt to the degradation characteristics of DMC materials under complex working conditions. The model structure includes three layers of LSTM units, each layer contains 256 neurons, and the time window is set to 24 hours. By taking historical health index data, real-time environmental data and current material performance parameters as input, the LSTM model can update the degradation trend of the equipment in real time and accurately predict the remaining life (RUL). Through the application of LSTM model, the prediction error is reduced from ±10% of traditional methods to ±3%, which significantly improves the accuracy of degradation trend prediction and provides equipment failure warning in advance.

[0155] Step 5: Intelligent maintenance plan generation and optimization. Based on the optimized health index and degradation trend prediction results, this embodiment generates a maintenance plan through an intelligent maintenance optimization algorithm. The algorithm comprehensively considers the balance between maintenance cost and failure risk, and introduces a dynamically adjusted risk weight coefficient. , in the maintenance plan, optimize the selection of maintenance time, location and measures. Maintenance cost The calculation formula is:

[0156] in: is the probability of failure, which is determined by the failure risk value Decide; The cost of repairing the fault; is the preventive maintenance cost. The risk weight coefficient obtained by fitting the multivariate linear regression model Balance the relationship between maintenance costs and failure risks to optimize maintenance plans and ensure efficiency and reliability in different equipment operating environments.

[0157] Step 6: Cloud platform closed-loop management and feedback. In this embodiment, all device data and health index are uploaded to the cloud platform through the IoT communication module. The cloud platform analyzes and stores data in real time, and continuously updates the degradation trend and remaining life prediction through time series analysis and dynamic optimization models. The maintenance plan is also adjusted according to the latest prediction results and sent to the device side through the cloud platform. The device side performs corresponding maintenance operations based on the real-time data fed back by the cloud platform, forming a closed-loop management of the entire life cycle.

[0158] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the present invention can be implemented in other specific forms without departing from the spirit or essential features of the present invention. Therefore, the embodiments should be considered exemplary and non-restrictive in all respects, and the scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present invention.

Claims

1. A method for operating and maintaining a complete set of switchgear based on DMC materials, characterized in that: The following steps are involved: Step 1, determination and modeling of key performance parameters of DMC materials: The key performance parameters of DMC materials are determined experimentally, and the key performance parameters are associated with experimental environmental conditions, including temperature, humidity and mechanical stress, to establish an aging characteristic model of DMC materials. The calculation formula of the model is: in: For time Key performance parameter values ​​at the time of is the initial performance value; is the aging factor, which indicates that the material properties change with temperature. ,humidity and mechanical stress the rate of decay due to changes in is a time variable, which refers to the time interval from the start of the experiment to the observation time; Step 2: Equipment operation status data collection: Arrange sensors at key locations of the complete switchgear to collect equipment operation status data in real time; Step 3, material health index calculation: based on the aging characteristic model of the DMC material and the real-time collected operating status data, the health index of the DMC material is calculated. The health index is referred to as MHI, and the calculation formula is: in: For current time The key performance parameter values ​​at , calculated by the aging characteristic model; is the initial performance value; It is the local temperature rise of DMC material; is the maximum temperature rise allowed for the material; is the fluctuation value of the electric field intensity on the material surface; is the maximum surface electric field strength allowed by the material; Step 4, Deterioration trend prediction and remaining service life calculation: Based on the historical data and real-time update data of the health index, a time series prediction model is constructed to predict the degradation trend and remaining service life of the DMC material. The remaining service life is calculated by the following formula: in: The remaining useful life of the material; is the current health index; It is the safety threshold of the health index, which is set by the equipment operation safety standard; is the rate of change of the health index over time, calculated using the difference formula using historical health index data; Step 5, intelligent maintenance plan generation: Based on the degradation trend prediction and the remaining life calculation results, a maintenance plan is dynamically generated, the maintenance plan includes maintenance time, maintenance parts and maintenance measures; the maintenance plan is generated by the following optimization objectives: in: Maintenance cost is calculated based on the equipment maintenance history data and weighted by failure frequency and repair cost; For failure risk, the failure risk value is calculated using the failure probability model based on the health index and degradation trend; is the risk weight coefficient, which is obtained by fitting the multivariate linear regression model of historical failure data; Step 6, cloud platform monitoring and feedback: upload the equipment's operating status data, health index and degradation trend prediction results to the cloud platform through the communication module; in the cloud platform: normalize and store the uploaded data, and update the degradation trend in real time based on the time series prediction model; automatically adjust the maintenance plan according to the updated degradation trend and remaining life, and send it to the device end through the communication module to guide equipment operation and maintenance.

2. A method for operating and maintaining a complete set of switchgear based on DMC materials according to claim 1, characterized in that: The key performance parameters include: insulation strength: measuring the breakdown voltage of the material by a withstand voltage test device under gradually increasing voltage, recording the maximum breakdown voltage value, and calculating its ratio to the material thickness; thermal conductivity: using a steady-state thermal conductivity tester to measure the heat flux density of the material and calculate the thermal conductivity; arc resistance: measured by a high-voltage arc test device, recording the ratio of the arc action time to the surface breakdown time; the aging factor is determined by the following steps: experimental data collection: carried out under different temperatures from 20°C to 100°C, different humidities from 30% to 90%, and different mechanical stresses from 0 to 10 MPa. Accelerated aging test, measuring the changes of key performance parameters over time; data fitting: using the least squares nonlinear regression algorithm to process experimental data and fit the specific value of the aging factor; and arranging sensors at key parts of the complete set of switchgear to collect the operating status data of the equipment in real time, including: working current and working voltage, collected by current sensors and voltage sensors; ambient temperature and humidity, collected by temperature and humidity sensors; mechanical vibration parameters, collected by triaxial acceleration sensors; local temperature rise of DMC materials, collected by thermistors; material surface electric field strength, collected by surface electric field detectors; and in step 3, Collected by thermistor; Determined by durability test in high temperature environment, it is a constant value; Collected by surface electric field detector; Measured by high electric field breakdown test, it is a constant value.

3. The method for operating and maintaining a complete set of switchgear based on DMC materials according to claim 1, characterized in that: Aging Factors The fitting process further includes a data cleaning step, which includes abnormal data removal, noise signal filtering and data normalization. During the measurement of the thermal conductivity of the DMC material, the temperature gradient is controlled in the range of 10°C to 50°C, and the measurement accuracy of the heat flux is not less than 0.01 W / m².

4. The method for operating and maintaining a complete set of switchgear based on DMC materials according to claim 1, characterized in that: Material Health Index During the calculation process, the maximum temperature rise allowed by the material It is determined by long-term durability tests conducted in the range of 80°C to 150°C, with the test time not less than 100 hours.

5. The method for operating and maintaining a complete set of switchgear based on DMC materials according to claim 1, characterized in that: Maximum surface electric field strength allowed by DMC materials It is measured by gradually increasing the electric field strength at a constant voltage until breakdown, with the rate of increase of the electric field strength being 10 kV / mm.

6. The method for operating and maintaining a complete set of switchgear based on DMC materials according to claim 1, characterized in that: Remaining life During the calculation process, the rate of change of health index over time It is calculated by the linear difference method of the historical data of the health index, with a difference interval of 1 hour.

7. The method for operating and maintaining a complete set of switchgear based on DMC materials according to claim 1, characterized in that: Failure risk value The calculation process uses a probability evaluation model based on a Bayesian network, and the input parameters of the model include the health index , Deterioration trend prediction results and remaining life .

8. The method for operating and maintaining a complete set of switchgear based on DMC materials according to claim 1, characterized in that: During the generation of the maintenance plan, the maintenance cost It is calculated by the following formula: in: is the probability of failure occurrence, which is determined by the failure risk value Decide; The cost of repairing the fault; For preventive maintenance costs.

9. The method for operating and maintaining a complete set of switchgear based on DMC materials according to claim 1, characterized in that: Risk Weight Factor It is obtained by multivariate linear regression fitting of historical failure data and maintenance cost data, and satisfies the following fitting equation: in: , , is the fitting coefficient, obtained by fitting historical data; The cost of repairing the fault; is the failure risk value.

Images