A method and system for testing a backup power supply device under simulated power system operating conditions
By collecting and processing the operating parameters of the automatic transfer switch (ATS) device under high altitude and low air pressure conditions, and constructing a multi-layer simulation test model using air pressure and viscosity change formulas, the adaptability and accuracy of the ATS device test method were solved, the reliability of fault prediction and equipment performance evaluation was improved, and the safety and stability of the power system were ensured.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG HUAXIN ELECTRIC
- Filing Date
- 2025-01-21
- Publication Date
- 2026-05-12
AI Technical Summary
Existing technologies for testing automatic start-up devices in high-altitude and low-pressure environments suffer from poor adaptability, complex physical model construction, inconvenient parameter adjustment, and a lack of integrated testing equipment, resulting in inaccurate and cumbersome test results.
The operating parameters of the automatic start-up device are collected by the data acquisition system. The data is cleaned and features are extracted using the Z-score method. Combined with the formulas for changes in air pressure and viscosity, a multi-layer simulation test model is constructed. The reference range of lubricating oil viscosity is dynamically adjusted to perform systematic comparison and fault prediction.
It improves the performance evaluation and fault prediction capabilities of automatic transfer switches in extreme environments, reduces the risk of equipment failure, and ensures the safe and stable operation of the power system.
Smart Images
Figure CN119962228B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power system testing, and in particular to a method and system for simulating automatic switching test of power system operating conditions. Background Technology
[0002] In modern power systems, automatic transfer switch (ATS) devices are critical equipment for ensuring power supply continuity. With the increasing scale and complexity of power systems, especially in extreme environments (such as high-altitude, low-pressure areas), the performance requirements for ATS devices are becoming increasingly stringent. However, current testing of ATS devices faces numerous challenges:
[0003] 1. Poor adaptability to special environments: Traditional testing methods mainly focus on normal operating conditions and fail to fully consider the impact of extreme environments such as high altitude and low air pressure on the performance of automatic start-up devices.
[0004] 2. Complex physical model construction: Building a physical model similar to the actual power system requires the connection of a large number of electrical devices, such as multiple real power sources, circuit breakers, relays, and complex secondary control circuits. This construction process is not only time-consuming and labor-intensive, but also prone to wiring errors due to human negligence, which seriously affects the accuracy and repeatability of the test, especially when simulating high-altitude and low-pressure environments.
[0005] 3. Inconvenient parameter adjustment: A comprehensive evaluation of the performance of an automatic transfer switch (ATS) device requires simulating various fault conditions and operating modes, including different degrees of voltage dips, frequency variations, phase differences, and different types of fault modes. However, in high-altitude and low-pressure environments, existing testing methods struggle to accurately adjust these parameters, making it impossible to efficiently and comprehensively test the ATS device.
[0006] 4. Lack of integrated testing equipment: There is a lack of integrated testing equipment specifically for automatic transfer switches (ATS) on the market, especially in extreme environments. Existing testing equipment either only performs single-function testing or requires the use of other instruments, making the testing process cumbersome. Data compatibility and synchronization between different instruments are poor, failing to provide a systematic and comprehensive testing solution. This makes it difficult to meet the high-quality testing requirements of modern power systems, especially in high-altitude and low-pressure environments. Summary of the Invention
[0007] To overcome the poor reliability of automatic transfer switch (ATS) test methods, this invention provides an ATS test method and system that simulates power system operating conditions.
[0008] The technical solution of this invention is: a method for simulating automatic switching test under power system operating conditions, comprising the following steps:
[0009] S1: In high-altitude, low-pressure environments, the operating parameter data of the automatic start-up device (ASD) is collected through a data acquisition system, and rigorous data cleaning, feature extraction, and standardization are performed using the Z-score method. The operating parameter data is divided into key feature data and auxiliary feature data. The key feature data specifically refers to parameters directly related to the failure caused by changes in lubricating oil viscosity, namely, the response delay and reduced sensitivity of the ASD. The auxiliary feature data are factors that mask the failure caused by changes in lubricating oil viscosity, namely, ambient temperature fluctuation data and power load change data.
[0010] S2: Using mathematical models and simulation tools, based on the formulas for air pressure variation with altitude and viscosity variation, the range of lubricating oil viscosity variation of the automatic start-up device under high altitude and low air pressure environment is calculated and defined as the viscosity reference range; data samples close to the fault point are selected from auxiliary feature data through data analysis algorithms, and the samples are analyzed in detail and compared with key feature data;
[0011] S3: Construct a multi-layer simulation test model, including a baseline layer, a feature selection layer, and a simulation layer; the baseline layer defines a viscosity baseline range, the feature selection layer is used to select representative auxiliary feature data samples, and the simulation layer uses computer simulation and numerical simulation to systematically compare the extracted data with the key feature data, and evaluate the differences between the two and the reasons behind them.
[0012] Preferably, in the high-altitude, low-pressure environment, the operation parameter data of the automatic start-up device is collected through a data acquisition system, and the Z-score method is used for rigorous data cleaning, feature extraction, and standardization processing, including: collecting the operation parameter data of the automatic start-up device under different altitudes and pressures, extracting lubricating oil viscosity change data and automatic start-up device fault data; and analyzing the severity of the automatic start-up device fault data and the fluctuation of the lubricating oil viscosity change data based on the ridge regression algorithm to obtain the relationship between the automatic start-up device faults and lubricating oil viscosity changes under different altitudes and pressures.
[0013] Preferably, the operating parameter data is divided into key feature data and auxiliary feature data, including: first, determining the key feature dataset by calculating the correlation between each feature and the change in lubricating oil viscosity. , in, For key feature datasets, This refers to parameters related to the viscosity variation of lubricating oil. Features and The Pearson correlation coefficient between them The set relevant threshold; auxiliary feature dataset The following is confirmed: in, For auxiliary feature datasets, To determine the remaining features after removing key features from all runtime parameter datasets, Features With fault labels The Pearson correlation coefficient between them The set fault-related threshold.
[0014] Preferably, the step of calculating the range of lubricating oil viscosity variation of the automatic start-up device under high altitude and low air pressure environments based on the formulas for air pressure variation with altitude and viscosity variation, and defining this range as a viscosity reference interval, includes: obtaining the air pressure value using the formula for air pressure variation with altitude, and obtaining the range of lubricating oil viscosity variation using the viscosity variation formula; the formula for air pressure variation with altitude is as follows: in, The calculated air pressure value, Altitude The molecular weight of air. It is the acceleration due to gravity. This refers to sea level temperature.
[0015] Preferably, obtaining the viscosity variation range of the lubricating oil using the viscosity variation formula includes: the viscosity variation formula is as follows, in, For lubricating oil at temperature and air pressure The viscosity below, As a pre-index factor, For activation energy, Let be the ideal gas constant. Absolute temperature The current air pressure. For reference air pressure, This is an empirical constant and depends on the type of lubricating oil.
[0016] Preferably, the step of selecting data samples close to the fault point from auxiliary feature data using data analysis algorithms, performing detailed analysis on the samples, and comparing them with key feature data includes:
[0017] Through linear correlation analysis, the relationship between the failure of the automatic start-up device and the change in lubricating oil viscosity was initially screened and defined as the primary relationship.
[0018] Based on the results of the initial screening, enhanced linear correlation analysis was used to amplify the relationship between the faults of the standby automatic switching device and the auxiliary characteristic data, and this relationship was defined as the second relationship.
[0019] The relationship between the faults and auxiliary characteristic data of the automatic switching device is obtained by using nonlinear analysis methods and defined as the third relationship.
[0020] Compare the first, second, and third relationships.
[0021] Preferably, the comparison of the first relationship, the second relationship, and the third relationship includes:
[0022] Auxiliary feature data from the third relation is selected for comparison with key feature data from the first relation that do not conform to the fault of the automatic switching device;
[0023] Auxiliary feature data from the second relation is selected for comparison with key feature data from the first relation that indicate a fault in the automatic switching device.
[0024] Preferably, the construction of the multi-layer simulation test model includes a baseline layer, a feature selection layer, and a simulation layer, comprising:
[0025] The comparison results of the auxiliary feature data in the third relationship and the key feature data in the first relationship that do not conform to the fault of the automatic start-up device are obtained. If the comparison results show that the auxiliary feature data masked the fault of the automatic start-up device caused by the change in lubricating oil viscosity, the viscosity reference range is dynamically adjusted and the range of lubricating oil viscosity change is extracted as the first narrowed range of the viscosity reference range.
[0026] The auxiliary feature data in the second relation is compared with the key feature data of the backup automatic start device failure in the first relation. If the comparison results show that the auxiliary feature data does not mask the fact that the lubricating oil viscosity change caused the backup automatic start device failure, the viscosity reference range is dynamically adjusted and the lubricating oil viscosity change range is extracted as the first expanded range of the viscosity reference range.
[0027] Preferably, the dynamically adjusted viscosity reference range includes:
[0028] The auxiliary feature data within the first narrowed range and the first expanded range are used as the first simulation data for simulating the power system operating conditions;
[0029] Key feature data within the first narrowed range and the first expanded range are used as second simulation data to simulate power system operating conditions.
[0030] Auxiliary feature data within a first narrowed range and auxiliary feature data within a first expanded range are selected. If, during the continuous simulation of power system operating conditions, auxiliary feature data within the first narrowed range appears continuously while auxiliary feature data within the first expanded range appears intermittently, then auxiliary feature data with a fluctuation level similar to that of lubricating oil is selected from the remaining auxiliary feature data as the first simulation data. The first simulation data is a portion of the auxiliary feature dataset and is selected as the preferred simulation data in the simulation of power system operating conditions.
[0031] A backup automatic transfer test system simulating power system operating conditions includes:
[0032] The data acquisition and preprocessing module collects high-altitude operating parameters, classifies and labels key and auxiliary feature data;
[0033] The mathematical modeling and data analysis module calculates the viscosity variation range, defines the baseline interval, and analyzes fault samples.
[0034] The relationship analysis and comparison module analyzes three types of relationships, compares and selects appropriate features, and optimizes the simulation data.
[0035] A multi-layer simulation model building module is used to construct a three-layer model, evaluate differences, and optimize the selection of the first simulation data.
[0036] The dynamic adjustment and verification module dynamically adjusts the viscosity reference range to optimize adaptability to actual working conditions.
[0037] Beneficial effects: This invention ensures the accuracy and completeness of data through comprehensive data acquisition and preprocessing. Addressing the impact of high-altitude, low-pressure environments on the performance of automatic start-up devices, it utilizes precise environmental adaptability modeling and, based on formulas for changes in air pressure and viscosity, accurately calculates the range of lubricating oil viscosity variations, defining a viscosity benchmark interval, significantly enhancing the practicality and reliability of the test results.
[0038] Furthermore, a multi-layer simulation test model was constructed to evaluate data differences and their causes from multiple levels. By dynamically adjusting the viscosity reference range, the model responded in real time to changes in different operating conditions, thereby improving the reliability and comprehensiveness of fault prediction.
[0039] By introducing auxiliary feature data from the first narrowed range and the first expanded range as simulation data, optimizing data selection, and focusing on the continuity and discontinuity of auxiliary feature data, the accuracy and reliability of the simulation are improved.
[0040] Finally, through dynamic adjustment and verification mechanisms, the viscosity reference range was optimized in real time, enhancing the model's adaptability and robustness, and ensuring the system's flexibility and stability. The method significantly improves the performance evaluation and fault prediction capabilities of automatic transfer switch (ATS) devices in complex environments, reduces equipment failure risks, ensures the safe and stable operation of the power system, and enhances the system's reliability and practicality. Attached Figure Description
[0041] Figure 1 This is a flowchart of the automatic switching test method for simulating power system operating conditions according to the present invention;
[0042] Figure 2 This is a schematic diagram of the automatic switching test system for simulating power system operating conditions according to the present invention. Detailed Implementation
[0043] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0044] Example 1: A method for simulating automatic switching test under power system operating conditions, such as... Figure 1 As shown, it includes the following steps:
[0045] S1: In high-altitude, low-pressure environments, the operating parameter data of the automatic start-up device (ASD) is collected through a data acquisition system, and rigorous data cleaning, feature extraction, and standardization are performed using the Z-score method. The operating parameter data is divided into key feature data and auxiliary feature data. The key feature data specifically refers to parameters directly related to the failure caused by changes in lubricating oil viscosity, namely, the response delay and reduced sensitivity of the ASD. The auxiliary feature data are factors that mask the failure caused by changes in lubricating oil viscosity, namely, ambient temperature fluctuation data and power load change data.
[0046] S2: Using mathematical models and simulation tools, based on the formulas for air pressure variation with altitude and viscosity variation, the range of lubricating oil viscosity variation of the automatic start-up device under high altitude and low air pressure environment is calculated and defined as the viscosity reference range; data samples close to the fault point are selected from auxiliary feature data through data analysis algorithms, and the samples are analyzed in detail and compared with key feature data;
[0047] S3: Construct a multi-layer simulation test model, including a baseline layer, a feature selection layer, and a simulation layer; the baseline layer defines a viscosity baseline range, the feature selection layer is used to select representative auxiliary feature data samples, and the simulation layer uses computer simulation and numerical simulation to systematically compare the extracted data with the key feature data, and evaluate the differences between the two and the reasons behind them.
[0048] In high-altitude, low-pressure environments, the operating parameter data of the automatic start-up device (APS) is collected through a data acquisition system. Ridge regression is then used for rigorous data cleaning, feature extraction, and standardization. This includes collecting APS operating parameter data under different altitudes and pressures, extracting lubricating oil viscosity variation data and APS fault data, and analyzing the severity of APS fault data and the fluctuation of lubricating oil viscosity variation data based on ridge regression algorithm to obtain the relationship between APS faults and lubricating oil viscosity variations under different altitudes and pressures.
[0049] Further explanation is that rigorous data cleaning and standardization improve the accuracy of operating parameters, laying the foundation for subsequent analysis. Combining ridge regression algorithm analysis with lubricating oil viscosity changes and fault data enhances the accuracy and reliability of fault diagnosis. It is particularly adaptable to extreme environments such as high altitudes and low air pressure, ensuring comprehensive and accurate performance evaluation and enhancing system robustness. It predicts potential problems in advance, optimizes maintenance strategies, reduces fault risks, and ensures the safe and stable operation of the power system.
[0050] The runtime parameter data is divided into key feature data and auxiliary feature data, including:
[0051] First, the key feature dataset is determined by calculating the correlation between each feature and the change in lubricating oil viscosity. , in, For key feature datasets, This refers to parameters related to the viscosity variation of lubricating oil. Features and The Pearson correlation coefficient between them The set relevant threshold; auxiliary feature dataset The following is confirmed: in, For auxiliary feature datasets, To determine the remaining features after removing key features from all runtime parameter datasets, Features With fault labels The Pearson correlation coefficient between them The set fault-related threshold.
[0052] To further explain, the key feature dataset middle, This is a dataset for all runtime parameters. This is the key feature dataset, which consists of parameters directly related to changes in lubricating oil viscosity. This refers to parameters related to the viscosity variation of lubricating oil. Features With parameters of lubricating oil viscosity change The Pearson correlation coefficient between two variables. It measures the strength and direction of the linear relationship between the two variables. A threshold value is set to distinguish which features are significantly correlated with changes in lubricating oil viscosity. The formula is used to filter out features highly correlated with changes in lubricating oil viscosity, forming a key feature dataset. Specifically, if a certain feature With parameters of lubricating oil viscosity change The absolute value of the Pearson correlation coefficient is greater than the set threshold. Auxiliary feature dataset middle, This is a dataset for all runtime parameters. This is the key feature dataset (already determined in the previous step). The remaining features after removing key features from all running parameter datasets are the non-key features. This is an auxiliary feature dataset, which masks factors that cause failures due to changes in lubricating oil viscosity. Features With fault labels The Pearson correlation coefficient between features. It measures the characteristics. The strength of the linear relationship between the automatic transfer switch (ATS) fault and the fault label. The formula is used to further filter from the remaining non-critical features those related to the fault label. Features with low correlation form an auxiliary feature dataset. Specifically, if a non-critical feature With fault labels The absolute value of the Pearson correlation coefficient is less than the set threshold. The main function of the two formulas is to aggregate all running parameter datasets. Divided into two categories: 1. Key feature datasets Includes features highly correlated with changes in lubricating oil viscosity. Screening criteria: Features With parameters of lubricating oil viscosity change Pearson correlation coefficient Greater than the set threshold 2. Auxiliary feature dataset : Includes fault tags Features with low relevance. Filtering criteria: Features With fault labels Pearson correlation coefficient Less than the set threshold This approach allows for more precise identification and analysis of key factors affecting the performance of automatic transfer switch (ATS) devices, improving the accuracy and reliability of fault diagnosis.
[0053] Based on the formulas for air pressure variation with altitude and viscosity variation, the viscosity variation range of the lubricating oil in the automatic start-up device under high altitude and low air pressure environments is calculated and defined as a viscosity reference range. This includes: obtaining the air pressure value using the formula for air pressure variation with altitude, and obtaining the viscosity variation range of the lubricating oil using the viscosity variation formula. The formula for air pressure variation with altitude is as follows: in, The calculated air pressure value, Altitude The molecular weight of air. It is the acceleration due to gravity. This refers to sea level temperature.
[0054] The viscosity variation range of lubricating oil can be obtained using viscosity change formulas, including: the viscosity change formulas are as follows. in, For lubricating oil at temperature and air pressure The viscosity below, As a pre-index factor, For activation energy, Let be the ideal gas constant. Absolute temperature The current air pressure. For reference air pressure, This is an empirical constant and depends on the type of lubricating oil.
[0055] To explain further, This indicates the height of the measurement point relative to sea level, affecting the degree of air pressure reduction. This reflects the influence of air composition on air pressure. The influence of Earth's gravity on air pressure distribution was taken into account. This reflects the effect of temperature on air pressure. It provides the foundation for the gas law. Sea level pressure (unit: Pa) is used as a reference. Used for subsequent viscosity calculations. For lubricating oil at temperature and air pressure The viscosity (in Pa·s) is used to calculate the viscosity of the lubricating oil, which is then used to evaluate its performance under different operating conditions. Depending on the specific chemical composition of the lubricating oil, it reflects the inherent characteristics of the lubricating oil. It reflects the energy required for lubricating oil molecules to change from one state to another, and determines the degree to which temperature affects viscosity. This reflects the effect of temperature on viscosity, and how it affects viscosity changes. The current air pressure (unit: Pa) is derived from the formula for air pressure variation with altitude. As a benchmark for comparison. Depending on the type of lubricating oil, it reflects the effect of air pressure on viscosity and the sensitivity of the lubricating oil type to changes in air pressure.
[0056] Data samples close to the fault point are selected from auxiliary feature data using data analysis algorithms. These samples are then analyzed in detail and compared with key feature data, including:
[0057] Through linear correlation analysis, the relationship between the failure of the automatic start-up device and the change in lubricating oil viscosity was initially screened and defined as the primary relationship.
[0058] Based on the results of the initial screening, enhanced linear correlation analysis was used to amplify the relationship between the faults of the standby automatic switching device and the auxiliary characteristic data, and this relationship was defined as the second relationship.
[0059] The relationship between the faults and auxiliary characteristic data of the automatic switching device is obtained by using nonlinear analysis methods and defined as the third relationship.
[0060] Compare the first, second, and third relationships.
[0061] Further explanation is as follows: Linear correlation analysis first identifies the direct relationship between automatic transfer switch (ATS) failures and lubricating oil viscosity changes, defined as the first relationship, quickly identifying significant correlations. Next, based on preliminary results, enhanced linear correlation analysis amplifies the relationship between failures and auxiliary feature data, defined as the second relationship, uncovering deeper linear associations. Subsequently, nonlinear analysis is introduced to evaluate the complex relationship between failures and auxiliary features, defined as the third relationship, capturing interactions that linear methods cannot reveal. Finally, the results of the first, second, and third relationships are compared to comprehensively evaluate the results of each analysis method, verifying their effectiveness and complementarity, ensuring the accuracy and reliability of the conclusions. This method improves failure prediction accuracy, optimizes maintenance strategies, reduces equipment failure risks, and ensures the safe and stable operation of the power system.
[0062] The comparison includes the first, second, and third relations, including:
[0063] Auxiliary feature data from the third relation is selected for comparison with key feature data from the first relation that do not conform to the fault of the automatic switching device;
[0064] Auxiliary feature data from the second relation is selected for comparison with key feature data from the first relation that indicate a fault in the automatic switching device.
[0065] A multi-layered simulation model is constructed, including a baseline layer, a feature selection layer, and a simulation layer, comprising:
[0066] The comparison results of the auxiliary feature data in the third relationship and the key feature data in the first relationship that do not conform to the fault of the automatic start-up device are obtained. If the comparison results show that the auxiliary feature data masked the fault of the automatic start-up device caused by the change in lubricating oil viscosity, the viscosity reference range is dynamically adjusted and the range of lubricating oil viscosity change is extracted as the first narrowed range of the viscosity reference range.
[0067] The auxiliary feature data in the second relation is compared with the key feature data of the backup automatic start device failure in the first relation. If the comparison results show that the auxiliary feature data does not mask the fact that the lubricating oil viscosity change caused the backup automatic start device failure, the viscosity reference range is dynamically adjusted and the lubricating oil viscosity change range is extracted as the first expanded range of the viscosity reference range.
[0068] Further explanation involves selecting auxiliary feature data from the third relationship and comparing it with key feature data from the first relationship that does not conform to the automatic start-up device (AS / RS) fault. This step aims to identify factors that mask the fault caused by changes in lubricating oil viscosity. Auxiliary feature data from the second relationship is then selected and compared with key feature data from the first relationship that conforms to the AS / RS fault to verify which auxiliary feature data are indeed related to the fault. The process includes: a baseline layer defining a viscosity baseline range as the basis for evaluation; a feature selection layer selecting representative auxiliary feature data samples to ensure the effectiveness of the comparison; a simulation layer systematically comparing extracted data with key feature data through computer simulation and numerical simulation to evaluate differences and their causes; and dynamically adjusting the viscosity baseline range to handle cases where the fault does not conform: if the auxiliary feature data from the third relationship masks the fault, the viscosity baseline range is dynamically adjusted, extracting the lubricating oil viscosity change range as the first narrowed range to ensure the model accurately reflects actual operating conditions. Conversely, if the auxiliary feature data from the second relationship does not mask the fault, the viscosity baseline range is dynamically adjusted, extracting the lubricating oil viscosity change area as the first expanded range to improve the system's robustness and flexibility.
[0069] Dynamically adjust the viscosity reference range, including:
[0070] The auxiliary feature data within the first narrowed range and the first expanded range are used as the first simulation data for simulating the power system operating conditions;
[0071] Key feature data within the first narrowed range and the first expanded range are used as second simulation data to simulate power system operating conditions.
[0072] Auxiliary feature data within a first narrowed range and auxiliary feature data within a first expanded range are selected. If, during the continuous simulation of power system operating conditions, auxiliary feature data within the first narrowed range appears continuously while auxiliary feature data within the first expanded range appears intermittently, then auxiliary feature data with a fluctuation level similar to that of lubricating oil is selected from the remaining auxiliary feature data as the first simulation data. The first simulation data is a portion of the auxiliary feature dataset and is selected as the preferred simulation data in the simulation of power system operating conditions.
[0073] Further explanation involves defining the simulation data as follows: First simulation data: Auxiliary characteristic data from both the first narrowed and first expanded ranges are used as the first simulation data to simulate power system operating conditions, capturing potential influencing factors related to changes in lubricating oil viscosity. Second simulation data: Key characteristic data within the same range are used as the second simulation data, directly correlated with the failure status of the automatic transfer switch (ATS). Optimal simulation data selection: Continuity and discontinuity assessment: During continuous simulation, observe whether the auxiliary characteristic data within the first narrowed range appears continuously, and whether the auxiliary characteristic data within the first expanded range appears discontinuously. This comparison helps identify data that better reflects changes in actual operating conditions. Preferred simulation data: Fluctuation matching: If it is found that the auxiliary characteristic data within the first narrowed range appears continuously, while the data within the first expanded range appears discontinuously, then the data among the remaining auxiliary characteristic data whose fluctuation level is similar to that of the lubricating oil viscosity is selected as the first simulation data. This step ensures that the selected data can more accurately simulate subtle changes in actual operating conditions, improving the accuracy and reliability of the simulation. Through these steps, the selection of simulation data is optimized, ensuring that the simulation process truly reflects the operating status of the ATS under high altitude and low air pressure conditions, enhancing the ability for performance evaluation and fault prediction.
[0074] Example 2: Based on Example 1, a backup automatic transfer test system simulating power system operating conditions includes:
[0075] The data acquisition and preprocessing module collects high-altitude operating parameters, classifies and labels key and auxiliary feature data;
[0076] The mathematical modeling and data analysis module calculates the viscosity variation range, defines the baseline interval, and analyzes fault samples.
[0077] The relationship analysis and comparison module analyzes three types of relationships, compares and selects appropriate features, and optimizes the simulation data.
[0078] A multi-layer simulation model building module is used to construct a three-layer model, evaluate differences, and optimize the selection of the first simulation data.
[0079] The dynamic adjustment and verification module dynamically adjusts the viscosity reference range to optimize adaptability to actual working conditions.
[0080] The present application has been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present application. Therefore, the content of this specification should not be construed as a limitation of the present application.
Claims
1. A method for simulating automatic transfer switching test under power system operating conditions, characterized in that, Includes the following steps: S1: In high-altitude, low-pressure environments, the operating parameter data of the standby automatic switching device are collected through the data acquisition system, and the Z-score method is used for rigorous data cleaning, feature extraction and standardization. The operating parameter data is divided into key feature data and auxiliary feature data, including: First, by calculating the correlation between each feature and the change in lubricating oil viscosity, the key feature dataset is determined. , ； in, For key feature datasets, This refers to parameters related to the viscosity variation of lubricating oil. Features With parameters of lubricating oil viscosity change The Pearson correlation coefficient between them The set relevant threshold; auxiliary feature dataset The following is confirmed: ； in, For auxiliary feature datasets, To determine the remaining features after removing key features from all runtime parameter datasets, Features With fault labels The Pearson correlation coefficient between them The set fault-related thresholds; the key feature data specifically refers to parameters directly related to the fault caused by changes in lubricating oil viscosity, namely, the response delay and reduced sensitivity of the automatic start-up device; the auxiliary feature data are factors that mask the fault caused by changes in lubricating oil viscosity, namely, ambient temperature fluctuation data and power load change data; S2: Using mathematical models and simulation tools, based on the formulas for air pressure changes with altitude and viscosity changes, calculate the range of lubricating oil viscosity changes of the automatic start-up device in high-altitude and low-pressure environments, and define it as the viscosity reference range; select data samples close to the fault point from the auxiliary feature data through data analysis algorithms, perform detailed analysis on the samples and compare them with the key feature data, including: preliminary screening through linear correlation analysis. The relationship between the automatic start-up device (AS / RS) malfunction and the change in lubricating oil viscosity is defined as the first relationship. Based on the preliminary screening results, enhanced linear correlation analysis is used to amplify the relationship between the AS / RS malfunction and auxiliary feature data, which is defined as the second relationship. A nonlinear analysis method is used to obtain the relationship between the AS / RS malfunction and auxiliary feature data, which is defined as the third relationship. The first, second, and third relationships are compared. This comparison includes: selecting auxiliary feature data from the third relationship for comparison with key feature data in the first relationship that does not conform to the AS / RS malfunction; and selecting auxiliary feature data from the second relationship for comparison with key feature data in the first relationship that conforms to the AS / RS malfunction. S3: Construct a multi-layer simulation test model, including a baseline layer, a feature selection layer, and a simulation layer. This includes: obtaining the comparison results of auxiliary feature data in the third relationship with key feature data in the first relationship that do not conform to the fault of the automatic transfer switch (ATS). If the comparison results show that the auxiliary feature data masks the lubricating oil viscosity change leading to the ATS fault, then dynamically adjust the viscosity baseline interval and extract the lubricating oil viscosity change interval as the first narrowed range of the viscosity baseline interval; obtaining the comparison results of auxiliary feature data in the second relationship with key feature data in the first relationship that conform to the fault of the ATS. If the comparison results show that the auxiliary feature data does not mask the lubricating oil viscosity change leading to the ATS fault, then dynamically adjust the viscosity baseline interval and extract the lubricating oil viscosity change interval as the first expanded range of the viscosity baseline interval; the dynamic adjustment of the viscosity baseline interval includes: using the auxiliary feature data within the first narrowed range and the first expanded range as the simulated power system... The system uses the following simulation methods: First, it uses the key feature data from the first narrowed range and the first expanded range as second, simulating the power system operating conditions. Second, it selects auxiliary feature data from the first narrowed range and the first expanded range. If, during continuous simulation of the power system operating conditions, auxiliary feature data from the first narrowed range appears continuously while auxiliary feature data from the first expanded range appears intermittently, then the auxiliary feature data with a fluctuation level similar to the lubricating oil viscosity among the remaining auxiliary feature data is selected as the first simulation data. The first simulation data is a portion of the auxiliary feature dataset and is the preferred data selected for simulation in the power system operating conditions simulation. The benchmark layer defines a viscosity benchmark range, the feature selection layer is used to select representative auxiliary feature data samples, and the simulation layer uses computer simulation and numerical simulation to systematically compare the extracted data and the key feature data, evaluating the differences between them and the underlying reasons.
2. The method for simulating automatic transfer switching test under power system operating conditions as described in claim 1, characterized in that, In high-altitude, low-pressure environments, the system collects operating parameter data of the automatic start-up device (ASD) through a data acquisition system, and performs rigorous data cleaning, feature extraction, and standardization using the Z-score method. This includes: collecting operating parameter data of the ASD under different altitudes and pressures; extracting lubricating oil viscosity change data and ASD fault data; and analyzing the severity of ASD fault data and the fluctuation of lubricating oil viscosity change data based on the ridge regression algorithm to obtain the relationship between ASD faults and lubricating oil viscosity changes under different altitudes and pressures.
3. The method for simulating automatic transfer switching test under power system operating conditions as described in claim 1, characterized in that, The formulas for air pressure variation with altitude and viscosity variation are used to calculate the range of lubricating oil viscosity variation of the automatic start-up device under high altitude and low air pressure environments, and define it as a viscosity reference range, including: The air pressure value is obtained using the formula for air pressure changing with altitude, and the viscosity range of the lubricating oil is obtained using the viscosity change formula. The formula for air pressure changing with altitude is as follows: ； in, The calculated air pressure value, Altitude The molecular weight of air. It is the acceleration due to gravity. Sea level temperature; The method of obtaining the viscosity variation range of lubricating oil using the viscosity variation formula includes: the viscosity variation formula is as follows. ； in, For lubricating oil at temperature and air pressure The viscosity below, As a pre-index factor, For activation energy, Let be the ideal gas constant. Absolute temperature The current air pressure. For reference air pressure, This is an empirical constant and depends on the type of lubricating oil.
4. A backup automatic transfer test system simulating power system operating conditions, used in the backup automatic transfer test method simulating power system operating conditions as described in any one of claims 1-3, characterized in that, Including: The data acquisition and preprocessing module collects high-altitude operating parameters, classifies and labels key and auxiliary feature data; The mathematical modeling and data analysis module calculates the viscosity variation range, defines the baseline interval, and analyzes fault samples. The relationship analysis and comparison module analyzes three types of relationships, compares and selects appropriate features, and optimizes the simulation data. The multi-layer simulation test model construction module builds a three-layer model, evaluates differences, and optimizes the selection of the first simulation data; the dynamic adjustment and verification module dynamically adjusts the viscosity reference range and optimizes the adaptability to actual working conditions.