Switch machine fault detection method and device, computer equipment and storage medium

Abstract

The invention relates to a point switch fault detection method and device, computer equipment and a storage medium. Relates to the technical field of rail transit. The method comprises the following steps: acquiring an analog current signal in a running process of a to-be-detected switch machine in real time, and converting the analog current signal into a digital current signal; filtering and denoising the digital current signal, and generating a standardized current curve based on the filtered and denoised digital current signal; extracting time domain features and frequency domain features in the standardized current curve; and matching the time domain feature and the frequency domain feature with a preset normal working condition feature library to obtain a fault detection result of the to-be-detected point switch. By adopting the method, the accuracy of the switch machine fault detection method can be improved.

Description

Technical Field

[0001] This application relates to the field of rail transit technology, and in particular to a method, apparatus, computer equipment, and storage medium for detecting faults in switch machines. Background Technology

[0002] As the core equipment for switching tracks in rail transit systems, switch machines play an irreplaceable role in ensuring the safe and efficient operation of trains. Their operational reliability directly affects the safety and stability of train operation; a malfunction in a switch machine could potentially lead to serious accidents such as train derailment. Therefore, detecting switch machine malfunctions has become a pressing technical problem to be solved.

[0003] However, current fault detection methods for switch machines mainly rely on manual inspections, which are resource-intensive and cannot provide real-time information on the machine's operating status. Furthermore, they typically rely on a single threshold current amplitude to determine fault conditions, which can easily lead to misjudgments. Therefore, current switch machine fault detection methods suffer from low accuracy. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for detecting switch machine faults, addressing the technical problem of low accuracy in the aforementioned switch machine fault detection methods.

[0005] Firstly, this application provides a method for detecting faults in a switch machine, including:

[0006] The analog current signal of the switch machine under test is collected in real time during operation and converted into a digital current signal.

[0007] The digital current signal is filtered and denoised, and a standardized current curve is generated based on the filtered and denoised digital current signal.

[0008] Extract the time-domain and frequency-domain features from the standardized current curve;

[0009] The time-domain features and frequency-domain features are matched with a preset normal operating condition feature library to obtain the fault detection result of the switch machine to be detected.

[0010] In one embodiment, matching the time-domain features and the frequency-domain features with a preset normal operating condition feature library to obtain the fault detection result of the switch machine to be detected includes: training a model based on the data of the preset normal operating condition feature library to obtain a classification model that can distinguish between normal and fault states; inputting the time-domain features and the frequency-domain features into the classification model, and obtaining a normal fault detection result for the switch machine to be detected if the output of the classification model is normal, and obtaining a fault detection result for the switch machine to be detected if the output of the classification model is abnormal.

[0011] In one embodiment, before inputting the time-domain features and the frequency-domain features into the classification model, the method further includes: acquiring the time-domain features and frequency-domain features of a normal switch machine under different operating conditions; and constructing the preset normal operating condition feature library based on multiple sets of the time-domain features and the frequency-domain features.

[0012] In one embodiment, the real-time acquisition of the analog current signal during the operation of the switch machine under test and the conversion of the analog current signal into a digital current signal includes: acquiring the analog current signal during the operation of the switch machine under test in real time based on a current sensor; amplifying, filtering and level conversion of the analog current signal based on a signal conditioning circuit to obtain a conditioned analog current signal; and converting the conditioned analog current signal into the digital current signal based on an analog-to-digital converter.

[0013] In one embodiment, after obtaining the fault detection result of the switch machine under test, the method further includes: acquiring historical fault records of the switch machine under test, the degradation trend of time-domain and frequency-domain characteristics in the standardized current curves corresponding to each fault, and the cumulative number of operations; obtaining the health index of the switch machine under test based on the historical fault records, the degradation trend, and the cumulative number of operations; and determining the remaining usable time of the switch machine under test based on the health index.

[0014] In one embodiment, if the fault detection result of the switch machine under test is no fault, the time-domain features and frequency-domain features of the standardized current curve corresponding to the switch machine under test are added to the preset normal operating condition feature library to update the preset normal operating condition feature library.

[0015] Secondly, this application also provides a switch machine fault detection device, comprising:

[0016] The signal acquisition module is used to acquire the analog current signal during the operation of the switch machine under test in real time, and convert the analog current signal into a digital current signal.

[0017] The signal processing module is used to filter and denoise the digital current signal and generate a standardized current curve based on the filtered and denoised digital current signal.

[0018] The feature extraction module is used to extract the time-domain and frequency-domain features from the standardized current curve;

[0019] The result acquisition module is used to match the time-domain features and the frequency-domain features with a preset normal operating condition feature library to obtain the fault detection result of the switch machine to be detected.

[0020] Thirdly, this application also provides a computer device, the computer device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0021] The analog current signal of the switch machine under test is collected in real time during operation and converted into a digital current signal.

[0022] The digital current signal is filtered and denoised, and a standardized current curve is generated based on the filtered and denoised digital current signal.

[0023] Extract the time-domain and frequency-domain features from the standardized current curve;

[0024] The time-domain features and frequency-domain features are matched with a preset normal operating condition feature library to obtain the fault detection result of the switch machine to be detected.

[0025] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0026] The analog current signal of the switch machine under test is collected in real time during operation and converted into a digital current signal.

[0027] The digital current signal is filtered and denoised, and a standardized current curve is generated based on the filtered and denoised digital current signal.

[0028] Extract the time-domain and frequency-domain features from the standardized current curve;

[0029] The time-domain features and frequency-domain features are matched with a preset normal operating condition feature library to obtain the fault detection result of the switch machine to be detected.

[0030] Fifthly, this application also provides a computer program product, which includes a computer program that, when executed by a processor, performs the following steps:

[0031] The analog current signal of the switch machine under test is collected in real time during operation and converted into a digital current signal.

[0032] The digital current signal is filtered and denoised, and a standardized current curve is generated based on the filtered and denoised digital current signal.

[0033] Extract the time-domain and frequency-domain features from the standardized current curve;

[0034] The time-domain features and frequency-domain features are matched with a preset normal operating condition feature library to obtain the fault detection result of the switch machine to be detected.

[0035] The aforementioned switch machine fault detection method, device, computer equipment, storage medium, and computer program product, in the process of switch machine fault detection, firstly acquires the analog current signal during the operation of the switch machine under test in real time, and converts the analog current signal into a digital current signal; then, the digital current signal is filtered and denoised, and a standardized current curve is generated based on the filtered and denoised digital current signal; next, the time-domain features and frequency-domain features in the standardized current curve are extracted; finally, the time-domain features and frequency-domain features are matched with a preset normal operating condition feature library to obtain the fault detection result of the switch machine under test. In the above process, the continuous analog current signal is converted into a calculable digital current signal through analog-to-digital conversion. At the same time, the filtering and denoising process effectively removes noise such as mechanical vibration clutter from the digital current signal, avoiding feature distortion caused by noise. Furthermore, by generating standardized current curves, the signal benchmark of switch machines of different models and operating environments is unified, which significantly reduces the probability of misjudgment. Moreover, by simultaneously extracting the time-domain and frequency-domain features of the standardized current curves, the operating status of the switch machine is accurately obtained, which is beneficial for subsequent fault detection. Therefore, this method can improve the accuracy of switch machine fault detection methods. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a flowchart illustrating a switch machine fault detection method in one embodiment;

[0038] Figure 2 This is a flowchart illustrating the switch machine fault detection steps in another embodiment;

[0039] Figure 3 This is a flowchart illustrating a method for determining the remaining usable time of a switch machine in one embodiment;

[0040] Figure 4 This is a detailed flowchart illustrating the switch machine fault detection steps in one embodiment;

[0041] Figure 5 This is a structural block diagram of a switch machine fault detection device in one embodiment;

[0042] Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0044] As a core piece of equipment in the rail transit system that enables turnout switching, the switch machine plays an irreplaceable and vital role in ensuring the safe and efficient operation of trains. Its operational reliability is directly related to the safety and stability of train operation. Once the switch machine malfunctions, it is highly likely to cause serious accidents such as train derailment and delays, posing a huge risk to rail transit operations.

[0045] However, traditional methods for monitoring the health of switch machines have significant limitations. Currently, they mainly rely on regular manual inspections. This method is not only costly in terms of manpower and resources, but also fails to provide real-time updates on equipment status during inspection intervals, making it difficult to detect potential problems promptly. Furthermore, current amplitude judgment methods based on a single threshold are rather crude, failing to comprehensively capture the dynamic characteristics of current curves as they change with time, load, and other factors. Moreover, manual inspections are prone to misdiagnosing equipment faults, misinterpreting normal current fluctuations as malfunctions, or missing detections, thus failing to identify potential faults in a timely manner and posing a threat to the safe operation of rail transit. Therefore, the current methods for detecting switch machine faults suffer from low accuracy.

[0046] To address the issue of low accuracy in the aforementioned switch machine fault detection methods, in an exemplary embodiment, such as... Figure 1 As shown, a method for detecting faults in a switch machine is provided. Taking the application of this method to a terminal as an example, the method includes the following steps S102 to S108. Wherein:

[0047] Step S102: Real-time acquisition of analog current signals during the operation of the switch machine under test, and conversion of analog current signals into digital current signals.

[0048] Among them, the switch machine is the core equipment of the railway turnout control system. It is used to drive the turnout to switch, such as from straight track to curved track or vice versa, lock the turnout position and reflect the turnout status to ensure the safe operation of trains. The switch machine under test is the target switch machine that is in operation and needs to be judged for faults through subsequent steps. The analog current signal is the working current of the motor and other components of the switch machine under test in the form of a continuously changing electrical signal when it is running. The amplitude, frequency and other parameters of the signal change continuously with time. It is an analog signal with high accuracy but is susceptible to electromagnetic interference.

[0049] More specifically, digital current signals are discrete signals obtained by converting analog current signals to digital signals. They are binary data consisting of 0s and 1s that can be recognized by computers. The conversion process can be achieved by acquiring instantaneous values ​​of analog signals at fixed time intervals and converting the sampled values ​​into finite-bit digital numbers.

[0050] Step S104: Filter and denoise the digital current signal, and generate a standardized current curve based on the filtered and denoised digital current signal.

[0051] The filtering and denoising process is mainly due to the interference noise that may be mixed into the digital current signal, such as useless signals generated by motor vibration, electromagnetic radiation, and sensor errors. The filtering and denoising process weakens or removes noise through methods such as mean filtering, Kalman filtering, and wavelet filtering, while retaining the effective components in the digital current signal that actually reflect the operating state of the switch machine under test, thereby improving signal quality. The standardized current curve is a curve generated after normalizing and calibrating the filtered digital current signal. Time can be used as the horizontal axis and current value as the vertical axis. The standardization process can eliminate the influence of individual differences of different switch machines or environmental factors on the current curve, which facilitates feature extraction and fault analysis of the switch machine under test.

[0052] Step S106: Extract the time-domain and frequency-domain features from the standardized current curve.

[0053] Among them, time-domain features are feature parameters extracted from the time dimension of the current curve, which are used to reflect the law of signal change over time. They can include statistical features, such as peak current, valley current, average current, variance, peak factor, etc.; trend features, such as current rise rate, fall rate, duration of steady phase, etc.; frequency-domain features are features extracted after converting the time-domain signal to the frequency domain through methods such as Fourier transform and wavelet transform, which are used to reflect the distribution law of different frequency components in the signal. They can include the frequency with the strongest energy in the signal, frequency spectrum peak, spectral entropy, etc.

[0054] Step S108: Match the time-domain features and frequency-domain features with the preset normal operating condition feature library to obtain the fault detection results of the switch machine to be detected.

[0055] The preset normal operating condition feature library is a feature set built through statistical analysis or machine learning based on a large amount of historical data of switch machines in normal operation. It includes time-domain features, frequency-domain features and their distribution models under normal conditions. It serves as the benchmark for judging whether the switch machine under test is faulty and needs to be updated regularly to adapt to long-term changes such as equipment aging. The fault detection results can be obtained by matching the time-domain and frequency-domain features of the switch machine under test with the normal operating condition feature library. The final conclusion usually includes normal, faulty and specific fault type.

[0056] In the aforementioned switch machine fault detection method, the analog current signal during the operation of the switch machine under test is first acquired in real time and converted into a digital current signal. Then, the digital current signal is filtered and denoised, and a standardized current curve is generated based on the filtered and denoised digital current signal. Next, the time-domain and frequency-domain features of the standardized current curve are extracted. Finally, the time-domain and frequency-domain features are matched with a preset normal operating condition feature library to obtain the fault detection result of the switch machine under test. In the above process, the continuous analog current signal is converted into a calculable digital current signal through analog-to-digital conversion. Simultaneously, the filtering and denoising process effectively removes noise such as mechanical vibration clutter from the digital current signal, avoiding feature distortion caused by noise. Furthermore, the generation of a standardized current curve unifies the signal benchmark for switch machines of different models and operating environments, significantly reducing the probability of misjudgment. Moreover, by simultaneously extracting the time-domain and frequency-domain features of the standardized current curve, accurate acquisition of the switch machine's operating status is achieved, which is beneficial for subsequent fault detection. Therefore, this method can improve the accuracy of switch machine fault detection.

[0057] In one exemplary embodiment, such as Figure 2 As shown, the time-domain features and frequency-domain features are matched with a preset normal operating condition feature library to obtain the fault detection result of the switch machine to be detected, including steps S202 to S204:

[0058] Step S202: Train the model based on the data of the preset normal working condition feature library to obtain a classification model that can distinguish between normal and fault states; Step S204: Input the time domain features and frequency domain features into the classification model. If the output of the classification model is normal, the fault detection result of the switch machine to be detected is normal. If the output of the classification model is abnormal, the fault detection result of the switch machine to be detected is fault.

[0059] The training model is a process of learning from data using machine learning algorithms. During training, the model parameters are continuously adjusted to improve the model's accuracy in identifying normal and fault states. The classification model is an algorithm model that can output whether a state is normal or faulty based on the input time-domain and frequency-domain features after training. The time-domain and frequency-domain features are two key features extracted from the standardized current curves of the switch machine under test. They are the input data for the classification model. The combination of time-domain and frequency-domain features can comprehensively characterize the operating state of the switch machine, providing multi-dimensional basis for the classification model's judgment.

[0060] More specifically, the output of a classification model is the result obtained after processing the input time-domain and frequency-domain features. It can include either normal or abnormal, or more, without limitation here. The output result is the conclusion of the classification model based on the rules learned during the training phase, judging the input time-domain and frequency-domain features. For example, if the difference between the input time-domain and frequency-domain features and the normal operating condition feature library is within the model's allowable range, the output is normal; if the difference between the input time-domain and frequency-domain features and the normal operating condition feature library exceeds the threshold or matches the fault sample feature pattern, the output is abnormal.

[0061] Therefore, the fault detection results are based on the final conclusions output by the classification model, which directly correspond to the operating status of the switch machine. When the classification model output is normal, the switch machine under test is normal; when the classification model output is abnormal, the switch machine under test is faulty.

[0062] In this embodiment, a classification model is trained by pre-setting a normal operating condition feature library to reduce human error and improve detection efficiency and accuracy. The classification model can quickly output the normal or fault results of the switch machine to be tested, and can promptly detect hidden dangers such as jamming and mechanical wear, reduce the risk of fault delays, ensure the safe and stable operation of the railway turnout system, and improve the reliability of railway transportation.

[0063] In one embodiment, before inputting the time-domain features and frequency-domain features into the classification model, the method further includes:

[0064] Obtain the time-domain and frequency-domain characteristics of a normal switch machine under different operating conditions; construct a pre-defined normal operating condition feature library based on multiple sets of time-domain and frequency-domain characteristics.

[0065] Among them, a normal switch machine is a switch machine in a fault-free operating state, whose mechanical structure and electrical components are all working normally, and can complete actions such as turnout switching and locking according to instructions, and whose operating parameters meet the design standards; different operating conditions refer to the diverse environments and working scenarios that the switch machine may encounter during operation, including but not limited to operating environments with differences in high temperature, low temperature, humidity, dust, and electromagnetic interference intensity; fluctuations in power supply voltage, switching speed, load changes, such as resistance differences caused by different degrees of turnout wear; normal aging states of different service years, and inherent characteristics of different models of switch machines; multiple sets of time-domain features and frequency-domain features are a large collection of feature samples collected under different operating conditions. Multiple sets emphasize the quantity and diversity of samples to obtain sufficient data; the pre-built normal operating condition feature library is a database formed after statistical analysis, normalization, or clustering of multiple sets of time-domain features and frequency-domain features.

[0066] In this embodiment, by collecting the characteristics of a normal switch machine under different operating conditions, the comprehensiveness and representativeness of the preset normal operating condition feature library are ensured, avoiding judgment bias caused by a single sample. The constructed preset normal operating condition feature library serves as a standard, laying a reliable foundation for the intelligent detection of switch machines to be tested, reducing reliance on human experience and improving the accuracy of fault identification.

[0067] Furthermore, in one embodiment, the analog current signal during the operation of the switch machine under test is acquired in real time, and the analog current signal is converted into a digital current signal, including:

[0068] The analog current signal during the operation of the switch machine under test is acquired in real time by a current sensor; the analog current signal is amplified, filtered and level-converted by a signal conditioning circuit to obtain a conditioned analog current signal; and the conditioned analog current signal is converted into a digital current signal by an analog-to-digital converter.

[0069] Among them, the current sensor is a device used to detect the magnitude of the current during the operation of the switch machine under test. It can convert the measured current signal into an easily measurable electrical signal. Common types include Hall current sensors and shunts. Real-time acquisition involves synchronously collecting the current signal according to the time process of the switch machine under test, which can ensure the time continuity and timeliness of the signal. The conditioned analog current signal has a higher signal-to-noise ratio, a more suitable amplitude and level range, and better meets the input requirements of analog-to-digital conversion. The analog-to-digital converter is an electronic component that converts continuous analog current signals into discrete digital signals.

[0070] In this embodiment, the operating current of the switch machine under test is acquired in real time by a current sensor, which can ensure the complete recording of the dynamic characteristics of key stages such as motor start-up, operation, and braking, laying the foundation for subsequent analysis. The signal conditioning circuit improves the resolution by amplifying weak signals, filtering out electromagnetic interference and noise, and level conversion to adapt to subsequent equipment, thereby improving signal quality and reducing distortion. The overall process effectively improves the accuracy of fault detection of the switch machine under test.

[0071] In one exemplary embodiment, such as Figure 3 As shown, after obtaining the fault detection result of the switch machine to be tested, steps S302 to S306 are also included:

[0072] Step S302: Obtain the historical fault records of the switch machine under test, the degradation trend of the time domain characteristics and frequency domain characteristics in the standardized current curves corresponding to each fault, and the cumulative number of operations; Step S304: Obtain the health index of the switch machine under test based on the historical fault records, degradation trend and cumulative number of operations; Step S306: Determine the remaining usable time of the switch machine under test based on the health index.

[0073] Among them, the historical fault record is the fault-related information of the switch machine under test in the past, including the fault occurrence time, fault type, fault cause, maintenance measures and repair effect, etc., which reflects the weak links of the switch machine under test and is an important basis for assessing its long-term health status; the degradation trend refers to the change law of the time domain characteristics and frequency domain characteristics of the standardized current curve gradually deviating from the normal range with time or number of operations before and after each fault occurrence.

[0074] More specifically, the cumulative number of runs refers to the total number of turnout switching actions completed by the switch machine under test since it was put into use, and is a key indicator for measuring the degree of equipment wear and tear; the health index is a value that reflects the current health status of the switch machine, calculated by a quantitative model by comprehensively considering historical fault records, characteristic degradation trends, and cumulative number of runs; and the remaining usable time is the time that the switch machine under test can still operate safely before a fault occurs, predicted based on the health index and combined with the equipment degradation pattern.

[0075] In this embodiment, by combining historical fault records, the inherent defects and fault modes of the switch machine under test can be obtained, providing historical basis for health assessment; analyzing the characteristic degradation trend can dynamically capture the performance degradation law and accurately reflect the process of the switch machine under test from normal to fault; by accumulating the number of runs, the wear and tear of mechanical and electrical components can be quantified, and finally, based on the remaining usable time predicted by the health index, it can prevent sudden failures from causing railway shutdowns.

[0076] More specifically, in one embodiment, if the fault detection result of the switch machine under test is no fault, the time-domain and frequency-domain features of the standardized current curve corresponding to the switch machine under test are added to a preset normal operating condition feature library to update the preset normal operating condition feature library.

[0077] Among them, updating the preset normal operating condition feature library involves integrating newly added fault-free features into the original feature library. Through statistical updates, such as adjusting the feature distribution range and supplementing typical features under new operating conditions, the feature library covers a wider range of normal state scenarios, dynamically optimizes the completeness and timeliness of the feature library, avoids misjudgment of normal changes due to the solidification of the feature library, improves the adaptability of the feature library to diverse scenarios, provides a more accurate benchmark for subsequent fault detection, and reduces the risk of missed detection and false detection.

[0078] This application provides a method for detecting faults in a switch machine. To better understand the process of the above-described method for detecting switch machine faults, combined with... Figure 4 As shown below, the specific process of a switch machine fault detection method of this application is described in detail, including the following steps:

[0079] Step S402: Data acquisition and signal preprocessing.

[0080] The data acquisition process includes real-time acquisition of analog current signals during the operation of the switch machine, signal conditioning of the analog current signals to convert them into digital current signals, and synchronous recording of operation time, switching direction and ambient temperature parameters. The signal preprocessing process includes filtering and denoising the digital current signals to eliminate high-frequency interference and baseline drift, and time alignment and normalization of the filtered digital current signals to generate standardized current curves.

[0081] Step S404: Extract time-domain and frequency-domain features from the standardized current curve and perform anomaly detection.

[0082] The time-domain features include action time, peak current, rising slope, falling slope, and curve fluctuation variance. The frequency-domain features include the spectral energy distribution and the amplitude ratio of the main frequency component obtained through fast Fourier transform. Anomaly detection involves dynamically comparing and matching the time-domain and frequency-domain features with a preset normal operating condition feature library to identify sudden changes, oscillations, or deviations in the current curve and generate anomaly type labels.

[0083] Step S406: Perform a health status assessment on the switch machine to be tested and output the life prediction results.

[0084] Among them, the health status assessment of the switch machine to be tested can be carried out by calculating the health index of the switch machine using a weighted scoring model based on historical abnormal records, characteristic parameter degradation trends and cumulative number of actions. The health index is divided into three levels: normal, warning and fault. Finally, the remaining service life prediction result can be output based on the health index.

[0085] Step S408 involves updating the normal operating condition feature library online and optimizing the dynamic threshold parameters, then iteratively training the anomaly detection model using newly added sample data to adapt to feature drift caused by aging, seasonal changes, and mechanical wear of the switch machine under test.

[0086] Through the above embodiments, refined monitoring of the switch machine's operation process is achieved by extracting multi-dimensional current curve features and performing dynamic threshold analysis, significantly improving the sensitivity and accuracy of anomaly detection. Combined with a health assessment model driven by historical data, the system can quantify the degree of equipment performance degradation and predict remaining lifespan, supporting preventative maintenance decisions. Furthermore, the system has adaptive learning capabilities, which can optimize the detection logic for different switch machine models and environmental conditions, reducing false alarm rates. Through anomaly-level alarm and visual diagnosis, fault diagnosis time can be significantly shortened, reducing the risk of operational interruption.

[0087] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0088] Based on the same inventive concept, this application also provides a switch machine fault detection device for implementing the switch machine fault detection method described above. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more switch machine fault detection device embodiments provided below can be found in the limitations of the switch machine fault detection method described above, and will not be repeated here.

[0089] In one exemplary embodiment, such as Figure 5 As shown, a switch machine fault detection device is provided, comprising: a signal acquisition module 501, a signal processing module 502, a feature extraction module 503, and a result acquisition module 504, wherein:

[0090] The signal acquisition module 501 is used to acquire the analog current signal during the operation of the switch machine under test in real time and convert the analog current signal into a digital current signal.

[0091] The signal processing module 502 is used to filter and denoise the digital current signal and generate a standardized current curve based on the filtered and denoised digital current signal.

[0092] The feature extraction module 503 is used to extract time-domain and frequency-domain features from the standardized current curve.

[0093] The result acquisition module 504 is used to match the time-domain features and frequency-domain features with a preset normal operating condition feature library to obtain the fault detection results of the switch machine to be detected.

[0094] Furthermore, in one embodiment, the result acquisition module 504 is also used to train a model based on a preset normal operating condition feature library to obtain a classification model that can distinguish between normal and fault states; input time-domain features and frequency-domain features into the classification model, and if the output of the classification model is normal, the fault detection result of the switch machine to be detected is normal; if the output of the classification model is abnormal, the fault detection result of the switch machine to be detected is fault.

[0095] Furthermore, in one embodiment, the result acquisition module 504 is also used to acquire the time-domain and frequency-domain features of the normal switch machine under different operating conditions; and to construct a preset normal operating condition feature library based on multiple sets of time-domain and frequency-domain features.

[0096] Furthermore, in one embodiment, the signal acquisition module 501 is also used to acquire the analog current signal during the operation of the switch machine under test in real time based on the current sensor; to amplify, filter and level-convert the analog current signal based on the signal conditioning circuit to obtain the conditioned analog current signal; and to convert the conditioned analog current signal into a digital current signal based on the analog-to-digital converter.

[0097] Furthermore, in one embodiment, the result acquisition module 504 is also used to acquire the historical fault records of the switch machine under test, the degradation trend of the time domain characteristics and frequency domain characteristics in the standardized current curves corresponding to each fault, and the cumulative number of runs; obtain the health index of the switch machine under test based on the historical fault records, degradation trend and cumulative number of runs; and determine the remaining usable time of the switch machine under test based on the health index.

[0098] Furthermore, in one embodiment, the result acquisition module 504 is also used to add the time-domain features and frequency-domain features in the standardized current curve corresponding to the switch machine under test to a preset normal operating condition feature library when the fault detection result of the switch machine under test is no fault, so as to update the preset normal operating condition feature library.

[0099] Each module in the aforementioned switch machine fault detection device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0100] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 6 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores switch machine fault detection data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a switch machine fault detection method.

[0101] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0102] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0103] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0104] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0105] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0106] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0107] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0108] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for detecting faults in a switch machine, characterized in that, The method includes: The analog current signal of the switch machine under test is collected in real time during operation and converted into a digital current signal. The digital current signal is filtered and denoised, and a standardized current curve is generated based on the filtered and denoised digital current signal. Extract the time-domain and frequency-domain features from the standardized current curve; The time-domain features and frequency-domain features are matched with a preset normal operating condition feature library to obtain the fault detection result of the switch machine to be detected.

2. The method according to claim 1, characterized in that, The step of matching the time-domain features and the frequency-domain features with a preset normal operating condition feature library to obtain the fault detection result of the switch machine to be detected includes: Based on the data training model of the preset normal operating condition feature library, a classification model that can distinguish between normal and fault states is obtained. The time-domain features and the frequency-domain features are input into the classification model. If the output of the classification model is normal, the fault detection result of the switch machine to be detected is normal. If the output of the classification model is abnormal, the fault detection result of the switch machine to be detected is faulty.

3. The method according to claim 2, characterized in that, Before inputting the time-domain features and the frequency-domain features into the classification model, the method further includes: Obtain the time-domain and frequency-domain characteristics of a normal switch machine under different operating conditions; The preset normal operating condition feature library is constructed based on multiple sets of time-domain features and frequency-domain features.

4. The method according to claim 1, characterized in that, The real-time acquisition of analog current signals during the operation of the switch machine under test, and the conversion of the analog current signals into digital current signals, includes: The analog current signal during the operation of the switch machine under test is acquired in real time based on the current sensor. The analog current signal is amplified, filtered, and level-converted using a signal conditioning circuit to obtain a conditioned analog current signal. The conditioned analog current signal is then converted into a digital current signal using an analog-to-digital converter.

5. The method according to claim 1, characterized in that, After obtaining the fault detection result of the switch machine under test, the method further includes: The historical fault records of the switch machine under test, the degradation trend of time-domain and frequency-domain features in the standardized current curves corresponding to each fault, and the cumulative number of runs are obtained. The health index of the switch machine to be tested is obtained based on the historical fault records, the degradation trend, and the cumulative number of operations. The remaining usable time of the switch machine to be tested is determined based on the health index.

6. The method according to claim 1, characterized in that, The method further includes: If the fault detection result of the switch machine under test is no fault, the time-domain and frequency-domain features of the standardized current curve corresponding to the switch machine under test are added to the preset normal operating condition feature library to update the preset normal operating condition feature library.

7. A switch machine fault detection device, characterized in that, The device includes: The signal acquisition module is used to acquire the analog current signal during the operation of the switch machine under test in real time, and convert the analog current signal into a digital current signal. The signal processing module is used to filter and denoise the digital current signal and generate a standardized current curve based on the filtered and denoised digital current signal. The feature extraction module is used to extract the time-domain and frequency-domain features from the standardized current curve; The result acquisition module is used to match the time-domain features and the frequency-domain features with a preset normal operating condition feature library to obtain the fault detection result of the switch machine to be detected.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images