Fault prediction method and system for high-voltage switchgear

Abstract

The invention discloses a high-voltage switch equipment fault prediction method and system, and relates to the technical field of high-voltage switch monitoring, and the method comprises the steps: extracting key operation parameters of high-voltage switch equipment during high-load operation, and building a heat effect dynamic model through combining the structure information of the equipment, comprising a working condition classification unit, a heat conduction heat dissipation unit and a load-temperature coupling unit, key operation parameters are collected in real time and input into a heat effect dynamic model, a load-temperature change curve is generated, fault features are extracted, high-load operation fault prediction is carried out, and scientific decision response is provided by quantifying risk levels. According to the method and the device, the technical problems of inaccurate fault prediction and low equipment operation reliability caused by the fact that the existing high-voltage switch equipment fault prediction cannot dynamically reflect the heat effect change under high-load operation are solved, and the technical effect of improving the fault prediction accuracy and the operation reliability of the high-voltage switch equipment under the high-load operation condition is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of high-voltage switch monitoring, and in particular to a method and system for predicting faults of high-voltage switch equipment. Background Art

[0002] High-voltage switchgear is a key control and protection device in the power system, and its operating status directly affects the stability and safety of the power grid. During peak power consumption, high-voltage switchgear often faces the challenge of long-term high-load operation. In this case, the internal thermal effect of the equipment increases significantly, which may cause problems such as contactor overheating and insulation material aging. In severe cases, it may even cause equipment failure, threatening the normal operation of the power system.

[0003] Traditional equipment fault monitoring and prediction methods usually rely on fixed thresholds or single monitoring indicators, which cannot fully reflect the dynamic changes of equipment thermal effects under high-load operating conditions. In addition, most of the commonly used fault prediction models currently ignore the coupling relationship between equipment operating parameters and the internal heat conduction and heat dissipation characteristics of the equipment, resulting in insufficient fault prediction accuracy and difficulty in timely and accurate identification of potential risks. Summary of the invention

[0004] The present application provides a high-voltage switchgear fault prediction method and system for solving the technical problem that the existing high-voltage switchgear fault prediction cannot dynamically reflect the changes in thermal effects under high-load operation, resulting in inaccurate fault prediction and low equipment operation reliability.

[0005] In a first aspect of the present application, a method for predicting faults of high-voltage switchgear is provided, the method comprising: extracting key operating parameters of the high-voltage switchgear when operating at high load based on historical operating records of the high-voltage switchgear; performing thermal effect modeling based on the key operating parameters and equipment structure information, and establishing a thermal effect dynamic model, the thermal effect dynamic model comprising a working condition classification unit, a heat conduction and heat dissipation unit, and a load-temperature coupling unit; collecting operating parameter sequences of the high-voltage switchgear in real time, inputting the thermal effect dynamic model for thermal effect analysis, generating a load-temperature variation curve, and extracting equipment operation fault characteristics based on the load-temperature variation curve; using the equipment operation fault characteristics to predict high-load operation faults, and making decision responses based on the fault prediction results.

[0006] The second aspect of the present application provides a high-voltage switchgear fault prediction system, the system comprising: a key operating parameter extraction module, the key operating parameter extraction module is used to extract the key operating parameters of the high-voltage switchgear when operating at high load based on the historical operating records of the high-voltage switchgear; a thermal effect modeling module, the thermal effect modeling module is used to perform thermal effect modeling based on the key operating parameters and equipment structure information, and establish a thermal effect dynamic model, the thermal effect dynamic model includes a working condition classification unit, a heat conduction and heat dissipation unit and a load-temperature coupling unit; a thermal effect analysis module, the thermal effect analysis module is used to collect the operating parameter sequence of the high-voltage switchgear in real time, input the thermal effect dynamic model to perform thermal effect analysis, generate a load-temperature change curve, and extract the equipment operation fault characteristics based on the load-temperature change curve; a high-load operation fault prediction module, the high-load operation fault prediction module is used to use the equipment operation fault characteristics to perform high-load operation fault prediction, and make a decision response based on the fault prediction result.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: The high-voltage switchgear fault prediction method and system provided in the present application relate to the field of high-voltage switch monitoring technology. By extracting key operating parameters of the high-voltage switchgear during high-load operation, a thermal effect dynamic model is established in combination with the equipment structure information, and key operating parameters are collected in real time and input into the thermal effect dynamic model. A load-temperature change curve is generated and fault characteristics are extracted to predict high-load operation faults, and a decision response is provided based on the risk level. The technical problem that the existing high-voltage switchgear fault prediction cannot dynamically reflect the thermal effect changes under high-load operation, resulting in inaccurate fault prediction and low equipment operation reliability is solved. The technical effect of improving the fault prediction accuracy and operation reliability of the high-voltage switchgear under high-load operation conditions by constructing a thermal effect dynamic model to capture the equipment's operating status and fault characteristics in real time is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0009] Figure 1 A schematic flow chart of a method for predicting faults of high-voltage switchgear provided in an embodiment of the present application.

[0010] Figure 2 A schematic diagram of the structure of a high-voltage switchgear fault prediction system provided in an embodiment of the present application.

[0011] Explanation of the reference numerals: key operating parameter extraction module 11 , thermal effect modeling module 12 , thermal effect analysis module 13 , high-load operation fault prediction module 14 . DETAILED DESCRIPTION

[0012] The present application provides a high-voltage switchgear fault prediction method and system for solving the technical problem that the existing high-voltage switchgear fault prediction cannot dynamically reflect the changes in thermal effects under high-load operation, resulting in inaccurate fault prediction and low equipment operation reliability.

[0013] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0014] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices.

[0015] Embodiment 1, as Figure 1 As shown, the present application provides a method for predicting faults of high-voltage switchgear, the method comprising: P10: Based on the historical operation records of high-voltage switchgear, extract the key operating parameters of high-voltage switchgear when operating under high load.

[0016] Optionally, in the scenario where high-voltage switchgear operates under long-term high-load conditions, in order to effectively predict failures caused by thermal effects, it is first necessary to extract key operating parameters based on historical operating records. Historical operating records refer to the operating data of the equipment under different load conditions, including current, voltage, temperature, operating time, and environmental parameters (such as ambient temperature, humidity, and wind speed). These records provide a data basis for establishing the relationship between the operating status of the equipment and thermal effects.

[0017] First, filter the data of high-load operation status from the historical operation records. High-load operation usually refers to the stage when the equipment load reaches more than 80% of its rated value, such as when the current load Super 0.8 The equipment enters a high-load state when the load is high. Data analysis tools can be used to quickly locate the data segments in these stages to ensure that the extracted data is directly related to the high-load operation scenario.

[0018] Next, it is necessary to extract key operating parameters from this data. These include current, voltage, temperature, and equipment operation time, which directly reflect the equipment's operating status. These parameters can characterize the electrical characteristics and thermal effect trends of the equipment under high load. In addition, it is also necessary to extract parameters related to the operating environment, such as ambient temperature, humidity, and wind speed. These external conditions have a significant impact on the equipment's heat dissipation performance and operating stability.

[0019] In the process of data screening and extraction, data preprocessing is an indispensable step. The collected historical data needs to be cleaned to eliminate outliers caused by sensor failures or other external interference. The cleaned data is then subjected to noise reduction processing, such as using the sliding average method or wavelet transform to reduce the interference of high-frequency noise, thereby ensuring that the extracted parameters accurately reflect the actual operating status of the equipment.

[0020] Through the above steps, the extracted key operating parameters can not only accurately describe the performance characteristics of the equipment under high-load operation conditions, but also provide a solid data foundation for subsequent thermal effect modeling and fault prediction.

[0021] P20: Based on the key operating parameters and equipment structure information, thermal effect modeling is performed to establish a thermal effect dynamic model, wherein the thermal effect dynamic model includes a working condition classification unit, a heat conduction and heat dissipation unit, and a load-temperature coupling unit.

[0022] Furthermore, step P20 of the embodiment of the present application further includes: P21: Preprocess the key operating parameters to obtain a standard sample parameter set; P22: Extract the equipment structure information of the target high-voltage switchgear, including equipment physical parameters and operating status classification constraints; P23: Generate operating condition classification rules based on the standard sample parameter set and the operating status classification constraints, and construct the operating condition classification unit according to the operating condition classification rules; P24: Determine the heat conduction equation and the heat dissipation equation based on the equipment physical parameters of the target high-voltage switchgear, and construct a heat conduction and heat dissipation unit based on the heat conduction equation and the heat dissipation equation; P25: According to the standard sample parameter set, fit the coupling relationship between load and temperature rise, and establish a load-temperature coupling unit; P26: Connect the operating condition classification unit, the heat conduction and heat dissipation unit and the load-temperature coupling unit to generate the thermal effect dynamic model.

[0023] It should be understood that in order to effectively predict the thermal effect failure risk of high-voltage switchgear under high-load operating conditions, it is necessary to combine the operating characteristics and physical structure information of the equipment to establish a thermal effect dynamic model that can dynamically reflect the changes in thermal effects. The model consists of three core units: operating condition classification unit, heat conduction and heat dissipation unit, and load-temperature coupling unit, which are responsible for the modeling tasks of operating state identification, heat conduction and heat dissipation calculation, and dynamic relationship between load and temperature rise.

[0024] First, a unified set of standard sample parameters is generated by preprocessing the extracted key operating parameters. The key operating parameters include equipment operating current, voltage, temperature and environmental conditions. In order to eliminate the impact of inconsistent data distribution on modeling, these data need to be normalized and all parameters need to be scaled to the same dimensional range. In addition, possible erroneous data in historical records are eliminated through outlier detection and noise reduction techniques, such as using the sliding average method to reduce high-frequency noise interference, thereby improving data quality.

[0025] Next, the physical structure information and operating constraints of the equipment are extracted as one of the inputs of the model. The equipment structure information includes key physical parameters such as thermal conductivity, density, and specific heat capacity of the equipment, which determine the heat transfer characteristics of the equipment; while the operating constraints reflect the classification rules of the equipment under different conditions, such as high load, overload, or normal operation. Combining this information can better describe the operating status of the equipment.

[0026] Based on the standard sample parameter set and operating constraints, the operating condition classification rules are generated and the operating condition classification unit is constructed. The operating condition classification rules define the classification standards for different load conditions by analyzing historical data. For example, when the current load and temperature both exceed the rated value, it can be determined that the equipment is in an overload condition. The classification unit dynamically identifies the current operating condition by real-time monitoring of parameter inputs and transmits the results to other units.

[0027] When constructing a heat conduction and heat dissipation unit, it is necessary to determine the heat conduction and heat dissipation equations based on the physical parameters of the equipment. The heat conduction equation can describe the heat transfer behavior inside the equipment, and the heat dissipation equation describes the heat exchange process between the surface heat dissipation of the equipment and the environment. Combining these two equations, the heat conduction and heat dissipation unit can calculate the heat dissipation power and internal heat distribution of the equipment in real time. At the same time, based on the standard sample parameter set, the coupling relationship between load and temperature rise is fitted to establish a load-temperature coupling unit. Through nonlinear regression fitting, the relationship between load and temperature rise is obtained. This unit can calculate the comprehensive temperature rise based on the real-time load and equipment status, providing a basis for fault prediction.

[0028] Finally, the working condition classification unit, heat conduction and heat dissipation unit and load-temperature coupling unit are connected to form a complete thermal effect dynamic model. The working condition classification unit identifies the equipment status in real time and provides classification labels to other units; the heat conduction and heat dissipation unit and the load-temperature coupling unit work together to calculate the heat dissipation power and comprehensive temperature rise of the equipment, and finally output the thermal effect characteristics of the equipment. Through the above steps, the constructed thermal effect dynamic model can accurately reflect the thermal effect characteristics of the equipment under high-load operating conditions, providing a reliable basis for fault prediction and decision-making response.

[0029] Furthermore, step P24 of the embodiment of the present application also includes: P24-1: According to the physical parameters of the target high-voltage switchgear, extract the heat conduction related parameters and the heat dissipation related parameters, and obtain the equipment operating environment information; P24-2: According to the heat conduction related parameters, the heat dissipation related parameters and the equipment operating environment information, construct the heat conduction equation and the heat dissipation equation, where: The heat conduction equation is: ;in, is the temperature change rate, is the temperature distribution, is the thermal conductivity, is the heat generation rate per unit volume, is the material density, is the specific heat capacity; the heat dissipation equation is: ;in, is the total heat dissipation power, is the natural convection heat dissipation coefficient, is the forced convection heat dissipation coefficient, is the heat dissipation area, is the surface temperature of the equipment, is the ambient temperature.

[0030] In a possible embodiment of the present application, in the high-load operation scenario of the high-voltage switchgear, the thermal effect prediction of the equipment needs to build accurate heat conduction equations and heat dissipation equations based on the physical characteristics and operating environment of the equipment. The establishment of these equations can not only describe the transfer of heat inside the equipment, but also quantify the heat dissipation capacity, providing key support for the dynamic model of the thermal effect of the equipment.

[0031] First, it is necessary to extract the physical characteristics and operating environment information of the target high-voltage switchgear from the equipment, including heat conduction related parameters: such as thermal conductivity, which determines the efficiency of heat transfer inside the equipment. The heat generation rate per unit volume is related to the load of the equipment during operation, and is usually calculated by the operating current and the equivalent thermal resistance of the equipment material. Material density and specific heat capacity affect the heat absorption and transfer behavior of the equipment respectively, and the specific values ​​are determined according to the material characteristics of the equipment. Heat dissipation related parameters: such as natural convection heat dissipation coefficient, which is related to the temperature difference between the surface temperature of the equipment and the ambient temperature. Forced convection heat dissipation coefficient is affected by wind speed and radiator surface characteristics. Heat dissipation area reflects the effective heat dissipation area of ​​the equipment in contact with the environment. Equipment operating environment information, including ambient temperature, wind speed, and ambient humidity, these factors directly affect the heat dissipation capacity of the equipment.

[0032] Combining the above parameters, the core equation describing the thermal effect of the equipment is established: The heat conduction equation is: ; This equation describes the time variation of the internal temperature T of the device, where: is the temperature change rate, The heat conduction in the space inside the device depends on Thermal conductivity and temperature distribution . It is the heat generation and absorption per unit volume, which is determined by the operating load and the material characteristics of the equipment. is the temperature distribution, is the thermal conductivity, is the heat generation rate per unit volume, is the material density, is the specific heat capacity.

[0033] The heat dissipation equation is: ; This equation quantifies the total heat dissipation power of the device: where is the total heat dissipation power, Indicates natural convection heat dissipation, with temperature difference Increase and strengthen, Indicates forced convection cooling, which depends on the wind speed and the heat dissipation design of the equipment. is the natural convection heat dissipation coefficient, is the forced convection heat dissipation coefficient, is the heat dissipation area, is the surface temperature of the equipment, is the ambient temperature.

[0034] The heat conduction equation is used to describe the generation and transfer of heat inside the device, reflecting the distribution and changes of temperature inside the device. The heat dissipation equation is used to calculate the heat exchange between the device surface and the environment, reflecting the heat dissipation capacity of the device. By combining the two, the core heat conduction and heat dissipation unit in the thermal effect dynamic model can be established to achieve a comprehensive description of the internal and external thermal effects of the device, providing an accurate mathematical basis for the real-time thermal effect analysis of the device, thereby more effectively predicting the risk of failure in high-load operation scenarios.

[0035] Furthermore, step P25 of the embodiment of the present application also includes: P25-1: Perform fitting analysis based on the standard sample parameter set to build the coupling relationship between load and temperature change: ;in, Indicates the real-time temperature. represents the initial ambient temperature, Indicates the real-time current load, is the adjustment factor.

[0036] Optionally, in high-load operation scenarios, in order to accurately predict the temperature rise of the equipment, it is necessary to establish a coupling relationship between the current load and the temperature change. Based on the data extracted from the standard sample parameter set, the following basic formula is obtained through fitting analysis to describe the dynamic relationship between load and temperature change: ;in, Indicates the time at which the device Real-time temperature, Indicates the initial ambient temperature, that is, the ambient temperature when the equipment starts to operate, which is used as the basis for temperature rise calculation. Indicates the real-time current load, reflects the load condition of the equipment during operation, and is the core variable that affects the temperature rise. It is an adjustment coefficient, which indicates the degree of coupling between load and temperature rise. Its value is related to parameters such as equipment structure, thermal conductivity, and heat dissipation capacity.

[0037] The specific construction process can be, first extract the current load from the standard sample parameter set and the corresponding temperature The data points are collected to ensure that the data covers the high-load operation stage. Then the data is normalized and denoised to remove outliers and ensure the accuracy of the fitting results. Furthermore, the nonlinear regression method is used to obtain the formula As the model basis, fitting The fitting goal is to minimize the error between the actual value of temperature and the predicted value, calculated by the least squares method or machine learning optimization algorithm. Finally, an independent validation dataset was used to fit the optimal value of The values ​​were substituted into the formula to calculate the temperature rise and compared with the actual measured values ​​to verify the accuracy of the model.

[0038] In actual applications, under high load conditions, real-time acquisition and , and substitute it into the formula to dynamically predict the temperature rise of the equipment. If the temperature is higher than the threshold temperature allowed by the equipment, it can be judged that there may be an overheating risk in the current operating state, providing a basis for subsequent fault prediction. The load-temperature coupling relationship established through fitting analysis can effectively quantify the dynamic impact of current load on equipment temperature rise, laying an accurate data model foundation for thermal effect analysis and fault prediction of high-voltage switchgear.

[0039] P30: Collect the operating parameter sequence of the high-voltage switchgear in real time, input the thermal effect dynamic model to perform thermal effect analysis, generate a load-temperature change curve, and extract the equipment operation fault characteristics based on the load-temperature change curve.

[0040] Furthermore, step P30 of the embodiment of the present application further includes: P31: The operating condition classification unit receives the real-time operating parameter sequence, generates an operating condition type label and passes it to the heat conduction and heat dissipation unit and the load-temperature coupling unit; P32: The heat conduction and heat dissipation unit calculates the real-time heat dissipation power and heat dissipation temperature change according to the operating condition type, the real-time operating parameter sequence and the real-time environmental parameters, and feeds back to the load-temperature coupling unit; P33: The load-temperature coupling unit calculates the comprehensive temperature rise in combination with the load parameters, current data and the heat dissipation temperature change, and corrects the comprehensive temperature rise according to the real-time heat dissipation power to generate a load-temperature change curve.

[0041] It should be understood that in the operation of high-voltage switchgear, in order to realize dynamic monitoring of the thermal effect of the equipment, it is necessary to collect the operating parameters in real time, analyze them through the dynamic model of thermal effect, and generate a real-time load-temperature change curve. This process not only reflects the relationship between load and temperature rise, but also provides basic support for the subsequent extraction of equipment operation fault characteristics.

[0042] First, the operating parameter sequence is collected in real time through the monitoring system, including current, voltage, equipment surface temperature, ambient temperature, wind speed, etc. These data are input into the working condition classification unit, and the working condition type label is generated according to the classification rules. For example, when the current load I(t)>0.9Irated and the temperature When the critical value of the equipment is approached, it can be marked as a high load condition. The generated condition label is transmitted to the heat conduction and cooling unit and the load-temperature coupling unit in real time to adjust the calculation logic to make it more suitable for the current operating status.

[0043] After receiving the working condition type tag, the heat conduction and heat dissipation unit adjusts the calculation mode of heat conduction and heat dissipation parameters according to the current working condition type. Combined with the real-time collected operating parameters (such as equipment temperature, current, ambient temperature) and environmental parameters (such as wind speed, humidity), the unit calculates the real-time heat dissipation power, and based on the heat dissipation power calculation results, combined with the thermal conductivity of the equipment, predicts the change trend of the equipment surface temperature, and generates heat dissipation temperature change data. These calculation results are fed back to the load-temperature coupling unit as a correction factor for the subsequent comprehensive temperature rise calculation.

[0044] Furthermore, in the load-temperature coupling unit, the comprehensive temperature rise of the equipment is calculated by combining the real-time current load I(t), the operating condition classification results, and the heat dissipation temperature change fed back from the heat conduction heat dissipation unit. , ,in, It represents the change of heat dissipation temperature, and the correction factor reflects the impact of heat dissipation on the comprehensive temperature rise.

[0045] By correcting the comprehensive temperature rise, a real-time load-temperature change curve is finally generated. This curve directly reflects the thermal effect behavior of the equipment under the current load and environmental conditions, and provides the dynamic characteristics of real-time temperature rise for monitoring the operating status of the equipment. When an abnormal peak or a rapid rising trend appears in the temperature rise curve, it can be used as a basis for fault feature extraction to provide support for subsequent fault prediction.

[0046] Furthermore, step P30 of the embodiment of the present application further includes: P34: Define key fault features, which include temperature rise rate, load-temperature rise gain, temperature fluctuation amplitude and overheating time ratio; P35: Set a time window for feature extraction, which includes a short-time window and a long-time window; P36: Extract key fault features from the load-temperature change curve according to the short-time window and the long-time window, respectively, to obtain a short-time fault feature set and a long-time fault feature set; P37: Combine the short-time fault feature set and the long-time fault feature set to generate the equipment operation fault feature.

[0047] Optionally, after generating a real-time load-temperature change curve, in order to identify potential equipment operation failures, it is necessary to extract key failure features from the curve, classify and comprehensively analyze the features from short-term and long-term perspectives, and finally generate equipment operation failure features.

[0048] First, according to the thermal effect characteristics of high-voltage switchgear, the core indicators required for fault prediction, namely key fault characteristics, are clarified. These characteristics can reflect the changing trend and abnormal mode of thermal effects during equipment operation, including but not limited to temperature rise rate, load-temperature rise gain, temperature fluctuation amplitude and overheating time ratio.

[0049] In order to fully extract the operational fault characteristics of the equipment, it is necessary to observe the load-temperature change curve from different time scales, including short-term windows: the time range is from seconds to minutes, which is used to capture instantaneous load changes and temperature rise rates. The goal is to identify dynamic anomalies such as rapid temperature rise and short-term temperature fluctuations. Long-term windows: The time range is from hours to days, which are used to evaluate the operating trends of the equipment and the impact of long-term loads. The goal is to identify long-term operating characteristics such as the proportion of overheating time and the amplitude of temperature fluctuations. By setting multiple time windows, fault characteristics can be accurately captured from both local and overall levels.

[0050] Furthermore, within the set time window, the load-temperature change curve is analyzed and features are extracted to extract parameters such as the instantaneous temperature rise rate and the short-term temperature fluctuation amplitude to form a short-term fault feature set for identifying rapid changes in equipment operation. Features such as long-term load-temperature rise gain and overheating time ratio are extracted to form a long-term fault feature set for evaluating the equipment's operating trend and thermal effect accumulation.

[0051] A comprehensive analysis is performed on the short-term fault feature set and the long-term fault feature set to generate a complete equipment operation fault feature. Among them, the short-term feature pays more attention to real-time and is suitable for rapid fault warning, while the long-term feature pays more attention to trend and is suitable for equipment health status assessment. Different features can be assigned corresponding weights according to the prediction requirements, and the short-term and long-term features are combined by weight to form a unified fault feature vector, that is, the equipment operation fault feature is generated. The equipment operation fault feature can fully reflect the changes in the thermal effect of the equipment under high-load operation conditions, and provide data support for subsequent fault prediction and risk assessment.

[0052] P40: Utilize the equipment operation fault characteristics to predict high-load operation faults, and make decision responses based on the fault prediction results.

[0053] Furthermore, step P40 of the embodiment of the present application also includes: P41: Based on the historical operation records of high-voltage switchgear, short-term fault feature samples and long-term fault feature samples are extracted respectively and weight coefficients are assigned to train the fault prediction model; P42: The equipment operation fault characteristics are input into the fault prediction model for analysis, and the risk level is determined according to the preset threshold rules to obtain the fault prediction result.

[0054] Specifically, during the operation of high-voltage switchgear, the extracted equipment operation fault features contain the equipment's operating status and potential abnormal information at different time scales. Using these features for fault prediction can not only assess the risk level of the equipment under high-load operation, but also provide a scientific basis for subsequent operation and maintenance decisions.

[0055] The core of fault prediction is to build a prediction model that can effectively utilize short-term and long-term features to accurately determine the risk status of equipment under high-load operation. The specific steps are as follows: First, extract short-term fault feature samples of equipment in the high-load operation stage from historical operation records, such as temperature rise rate and short-term fluctuation amplitude, and extract long-term feature samples, such as load-temperature rise gain and overheating time ratio. These samples contain the correspondence between the historical equipment operation status and known fault events, providing basic data for model training. Short-term features reflect real-time abnormal changes and are suitable for rapid response; long-term features reflect operation trends and are suitable for trend prediction. According to the application scenario, weight coefficients are assigned to the two samples respectively, and the selection of weights can be based on the historical relevance of fault occurrence and the importance of data. Further, supervised learning algorithms are used to build fault prediction models (such as random forests, XGBoost or LSTM). The training goal is to minimize the error between the prediction results output by the model and the historical known fault events, and to improve the prediction accuracy by repeatedly iterating and optimizing the model parameters, and finally generate a fault prediction model.

[0056] After the fault prediction model training is completed, the real-time extracted equipment operation fault features, such as short-term features (such as the current temperature rise rate, short-term fluctuation amplitude) and long-term features (such as load-temperature rise gain, overheating time ratio) can be used as input vectors and input into the fault prediction model for real-time analysis. The model calculates and outputs the risk probability of equipment failure, indicating the possibility of equipment failure under current conditions. Then, according to the fault risk probability output by the model, it is compared with the preset risk threshold rules to divide the risk level, and formulate targeted operation and maintenance strategies based on the risk prediction results. For example, low risk means that the equipment is in normal operation and continuous monitoring, and no measures are required. Medium risk means that there are potential risks in the equipment, and it is recommended to strengthen monitoring and appropriately adjust the operating parameters (such as reducing the load or optimizing the heat dissipation conditions). High risk means that the equipment is already in a high risk state of failure, and emergency measures should be taken immediately, such as load reduction or shutdown inspection.

[0057] Through the above steps, combined with the comprehensive analysis of short-term and long-term fault characteristics, the fault prediction model can evaluate the operating risk of the equipment in real time, and provide a scientific decision-making basis for the operation and maintenance team through quantitative assessment of risk levels. This data-driven prediction and response mechanism can significantly improve the reliability and safety of high-voltage switchgear under high-load operating conditions.

[0058] In summary, the embodiments of the present application have at least the following technical effects: This application extracts the key operating parameters of high-voltage switchgear during high-load operation, establishes a thermal effect dynamic model in combination with the equipment structure information, collects the key operating parameters in real time and inputs them into the thermal effect dynamic model, generates a load-temperature change curve and extracts fault characteristics, predicts high-load operation faults, and provides decision responses based on risk levels.

[0059] The technical effect is achieved by building a dynamic model of thermal effects, capturing the operating status and fault characteristics of the equipment in real time, and improving the fault prediction accuracy and operating reliability of high-voltage switchgear under high-load operating conditions.

[0060] Embodiment 2, based on the same inventive concept as the method for predicting faults of high-voltage switchgear in the above embodiment, Figure 2 As shown, the present application provides a high-voltage switchgear fault prediction system, and the system and method embodiments in the present application embodiments are based on the same inventive concept. The system includes: The key operating parameter extraction module 11 is used to extract the key operating parameters of the high-voltage switchgear when it is operating at high load according to the historical operating records of the high-voltage switchgear.

[0061] The thermal effect modeling module 12 is used to perform thermal effect modeling based on the key operating parameters and equipment structure information, and establish a thermal effect dynamic model. The thermal effect dynamic model includes a working condition classification unit, a heat conduction and heat dissipation unit, and a load-temperature coupling unit.

[0062] The thermal effect analysis module 13 is used to collect the operating parameter sequence of the high-voltage switchgear in real time, input the thermal effect dynamic model to perform thermal effect analysis, generate a load-temperature change curve, and extract equipment operation fault characteristics based on the load-temperature change curve.

[0063] The high-load operation fault prediction module 14 is used to use the equipment operation fault characteristics to predict high-load operation faults and make a decision response based on the fault prediction results.

[0064] Furthermore, the thermal effect modeling module 12 is also used to perform the following steps: Preprocess the key operating parameters to obtain a standard sample parameter set; extract the equipment structure information of the target high-voltage switchgear, including equipment physical parameters and operating state classification constraints; generate operating condition classification rules based on the standard sample parameter set and the operating state classification constraints, and construct the operating condition classification unit according to the operating condition classification rules; determine the heat conduction equation and the heat dissipation equation according to the equipment physical parameters of the target high-voltage switchgear, and construct a heat conduction and heat dissipation unit according to the heat conduction equation and the heat dissipation equation; fit the coupling relationship between load and temperature rise according to the standard sample parameter set, and establish a load-temperature coupling unit; connect the operating condition classification unit, the heat conduction and heat dissipation unit and the load-temperature coupling unit to generate the thermal effect dynamic model.

[0065] Furthermore, the thermal effect modeling module 12 is also used to perform the following steps: According to the physical parameters of the target high-voltage switchgear, heat conduction related parameters and heat dissipation related parameters are extracted, and the equipment operating environment information is obtained; according to the heat conduction related parameters, heat dissipation related parameters and equipment operating environment information, a heat conduction equation and a heat dissipation equation are constructed, wherein: the heat conduction equation is: ;in, is the temperature change rate, is the temperature distribution, is the thermal conductivity, is the heat generation rate per unit volume, is the material density, is the specific heat capacity; the heat dissipation equation is: ;in, is the total heat dissipation power, is the natural convection heat dissipation coefficient, is the forced convection heat dissipation coefficient, is the heat dissipation area, is the surface temperature of the equipment, is the ambient temperature.

[0066] Furthermore, the thermal effect modeling module 12 is also used to perform the following steps: A fitting analysis is performed based on the standard sample parameter set to construct the coupling relationship between load and temperature change: ;in, Indicates the real-time temperature. represents the initial ambient temperature, Indicates the real-time current load, is the adjustment factor.

[0067] Furthermore, the thermal effect analysis module 13 is also used to perform the following steps: The working condition classification unit receives the real-time operating parameter sequence, generates a working condition type label and transmits it to the heat conduction heat dissipation unit and the load-temperature coupling unit; the heat conduction heat dissipation unit calculates the real-time heat dissipation power and heat dissipation temperature change according to the working condition type, the real-time operating parameter sequence and the real-time environmental parameters, and feeds them back to the load-temperature coupling unit; the load-temperature coupling unit calculates the comprehensive temperature rise in combination with the load parameters, current data and the heat dissipation temperature change, and corrects the comprehensive temperature rise according to the real-time heat dissipation power to generate a load-temperature change curve.

[0068] Furthermore, the thermal effect analysis module 13 is also used to perform the following steps: Define key fault features, the key fault features include temperature rise rate, load-temperature rise gain, temperature fluctuation amplitude and overheating time ratio; set a time window for feature extraction, the time window includes a short-time window and a long-time window; extract key fault features from the load-temperature change curve according to the short-time window and the long-time window, respectively, to obtain a short-time fault feature set and a long-time fault feature set; combine the short-time fault feature set and the long-time fault feature set to generate the equipment operation fault feature.

[0069] Furthermore, the high-load operation fault prediction module 14 is also used to perform the following steps: Based on the historical operation records of high-voltage switchgear, short-term fault feature samples and long-term fault feature samples are extracted and weight coefficients are assigned to train a fault prediction model; the equipment operation fault characteristics are input into the fault prediction model for analysis, and the risk level is determined according to the preset threshold rules to obtain the fault prediction result.

[0070] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0071] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0072] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A method for predicting faults of high-voltage switchgear, characterized in that: The method comprises: According to the historical operation records of high-voltage switchgear, key operating parameters of high-voltage switchgear under high-load operation are extracted; Based on the key operating parameters and equipment structure information, thermal effect modeling is performed to establish a thermal effect dynamic model, wherein the thermal effect dynamic model includes a working condition classification unit, a heat conduction and heat dissipation unit, and a load-temperature coupling unit; Collecting the operating parameter sequence of the high-voltage switchgear in real time, inputting the thermal effect dynamic model to perform thermal effect analysis, generating a load-temperature change curve, and extracting the equipment operation fault characteristics based on the load-temperature change curve; The equipment operation fault characteristics are used to predict high-load operation faults, and a decision response is made based on the fault prediction results.

2. The high-voltage switchgear fault prediction method according to claim 1, characterized in that: Based on the key operating parameters and equipment structure information, thermal effect modeling is performed to establish a thermal effect dynamic model, including: Preprocessing the key operating parameters to obtain a standard sample parameter set; Extracting the equipment structure information of the target high-voltage switchgear, including equipment physical parameters and operating status classification constraints; Based on the standard sample parameter set and the operating state classification constraint, generating an operating condition classification rule, and constructing the operating condition classification unit according to the operating condition classification rule; Determine a heat conduction equation and a heat dissipation equation according to the physical parameters of the target high-voltage switchgear, and construct a heat conduction heat dissipation unit according to the heat conduction equation and the heat dissipation equation; According to the standard sample parameter set, the coupling relationship between load and temperature rise is fitted to establish a load-temperature coupling unit; The working condition classification unit, the heat conduction and heat dissipation unit and the load-temperature coupling unit are connected to generate the thermal effect dynamic model.

3. The high-voltage switchgear fault prediction method according to claim 2, characterized in that: According to the physical parameters of the target high-voltage switchgear, a heat conduction equation and a heat dissipation equation are determined, including: Extracting heat conduction related parameters and heat dissipation related parameters according to the device physical parameters of the target high-voltage switchgear, and obtaining device operating environment information; According to the heat conduction related parameters, heat dissipation related parameters and equipment operating environment information, a heat conduction equation and a heat dissipation equation are constructed, wherein: The heat conduction equation is: ; in, is the temperature change rate, is the temperature distribution, is the thermal conductivity, is the heat generation rate per unit volume, is the material density, is the specific heat capacity; The heat dissipation equation is: ； in, is the total heat dissipation power, is the natural convection heat dissipation coefficient, is the forced convection heat dissipation coefficient, is the heat dissipation area, is the surface temperature of the equipment, is the ambient temperature.

4. The high-voltage switchgear fault prediction method according to claim 3, characterized in that: According to the standard sample parameter set, the coupling relationship between load and temperature rise is fitted to establish a load-temperature coupling unit, including: A fitting analysis is performed based on the standard sample parameter set to construct the coupling relationship between load and temperature change: ； in, Indicates the real-time temperature. represents the initial ambient temperature, Indicates the real-time current load, is the adjustment factor.

5. The high-voltage switchgear fault prediction method according to claim 4, characterized in that: The operating parameter sequence of the high-voltage switchgear is collected in real time, input into the thermal effect dynamic model for thermal effect analysis, and a load-temperature variation curve is generated, including: The working condition classification unit receives the real-time operating parameter sequence, generates a working condition type label and transmits it to the heat conduction and heat dissipation unit and the load-temperature coupling unit; The heat conduction heat dissipation unit calculates the real-time heat dissipation power and heat dissipation temperature change according to the working condition type, the real-time operation parameter sequence and the real-time environmental parameters, and feeds back to the load-temperature coupling unit; The load-temperature coupling unit combines load parameters, current data and the heat dissipation temperature change to calculate the comprehensive temperature rise, and corrects the comprehensive temperature rise according to the real-time heat dissipation power to generate a load-temperature change curve.

6. The high-voltage switchgear fault prediction method according to claim 1, characterized in that: Based on the load-temperature change curve, extract the equipment operation fault characteristics, including: defining key fault characteristics, wherein the key fault characteristics include temperature rise rate, load-temperature rise gain, temperature fluctuation amplitude, and overheating time ratio; Setting a time window for feature extraction, wherein the time window includes a short-term window and a long-term window; Extract key fault features from the load-temperature variation curve according to the short-time window and the long-time window respectively, to obtain a short-time fault feature set and a long-time fault feature set; The short-term fault feature set and the long-term fault feature set are combined to generate the equipment operation fault feature.

7. The high-voltage switchgear fault prediction method according to claim 6, characterized in that: Using the equipment operation fault characteristics, high-load operation fault prediction is performed, including: Based on the historical operation records of high-voltage switchgear, short-term fault feature samples and long-term fault feature samples are extracted and weight coefficients are assigned to train the fault prediction model; The equipment operation fault characteristics are input into the fault prediction model for analysis, and the risk level is determined according to a preset threshold rule to obtain the fault prediction result.

8. A high voltage switchgear fault prediction system, characterized in that: The system comprises: A key operating parameter extraction module, which is used to extract key operating parameters of the high-voltage switchgear when it is running under high load according to the historical operating records of the high-voltage switchgear; A thermal effect modeling module, wherein the thermal effect modeling module is used to perform thermal effect modeling based on the key operating parameters and equipment structure information, and establish a thermal effect dynamic model, wherein the thermal effect dynamic model includes a working condition classification unit, a heat conduction and heat dissipation unit, and a load-temperature coupling unit; A thermal effect analysis module, which is used to collect the operating parameter sequence of the high-voltage switchgear in real time, input the thermal effect dynamic model to perform thermal effect analysis, generate a load-temperature change curve, and extract equipment operation fault characteristics based on the load-temperature change curve; A high-load operation fault prediction module is used to use the equipment operation fault characteristics to predict high-load operation faults and make a decision response based on the fault prediction results.

Images