A solenoid valve fault intelligent detection method and system based on multi-parameter analysis

Abstract

The present invention relates to the field of industrial automation detection technology, and specifically to a method and system for intelligent detection of solenoid valve faults based on multi-parameter analysis, comprising the following steps: S1: synchronously collecting its dynamic current waveform, vibration spectrum information, and ambient temperature data; S2: extracting current characteristic values ​​and vibration energy distribution, and calculating temperature drift; S3: inputting the current characteristic values, vibration energy distribution, and temperature drift into a multi-parameter correlation matrix to generate a coupling deviation coefficient that characterizes the coupling relationship between parameters; S4: calculating the comprehensive health index of the solenoid valve based on the coupling deviation coefficient; S5: matching the comprehensive health index with a preset fault feature library, and outputting the corresponding fault type and location results. By fusing multi-source characteristic parameters and introducing a health index evaluation mechanism, the present invention achieves accurate identification and location of solenoid valve faults, improving detection accuracy and response efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial automation detection, and in particular to a solenoid valve fault intelligent detection method and system based on multi-parameter analysis. Background Art

[0002] In key control systems such as rail transit, petrochemical plants, and industrial automation, solenoid valves, as core actuator components, have operational stability that is directly related to system safety and control accuracy. However, due to the long-term operation of solenoid valves in high-frequency start-stop, mechanical wear, and high-temperature environments, they are prone to failures such as coil short circuits, valve core sticking, and spring fatigue. If a failure is not discovered in time, it may cause linkage control anomalies or even equipment damage.

[0003] Traditional solenoid valve status monitoring methods are often based on a single signal source and rely on simple thresholds for identification. This makes it difficult to accurately identify multiple types of faults and their evolution, and thus fails to meet the practical needs of high-reliability operation and maintenance. Therefore, an intelligent solenoid valve fault detection method and system based on multi-parameter analysis is urgently needed to address these issues. Summary of the Invention

[0004] Based on the above objectives, the present invention provides a solenoid valve fault intelligent detection method and system based on multi-parameter analysis.

[0005] An intelligent detection method for solenoid valve faults based on multi-parameter analysis includes the following steps:

[0006] S1: During the operation of the solenoid valve, its dynamic current waveform, vibration spectrum information and ambient temperature data are synchronously collected;

[0007] S2: Extract the current characteristic value from the dynamic current waveform, extract the vibration energy distribution from the vibration spectrum characteristics, and calculate the temperature drift from the ambient temperature parameters;

[0008] S3: Input the current characteristic value, vibration energy distribution and temperature drift into the multi-parameter correlation matrix to generate a coupling deviation coefficient that characterizes the coupling relationship between the parameters;

[0009] S4: Calculate the comprehensive health index of the solenoid valve based on the coupling deviation coefficient;

[0010] S5: Match the comprehensive health index with the preset fault feature library and output the corresponding fault type and location result.

[0011] Optionally, the S1 specifically includes:

[0012] S11: Use the open-loop Hall current sensor HCT-20A to collect the current signal in real time in the solenoid valve power supply circuit. The sampling frequency is set to 10kHz and the resolution is 16 bits to generate dynamic current waveform raw data.

[0013] S12: Install an ICP type acceleration sensor at the axial center point of the solenoid valve body housing. Use an anti-aliasing filter to intercept the 0-10kHz frequency band signal, obtain the vibration time domain waveform at a 12.8kHz sampling rate, and generate the vibration spectrum characteristics through fast Fourier transform.

[0014] S13: A PT1000 platinum resistance temperature sensor is mounted on the surface of the solenoid valve coil housing. The original temperature signal is obtained at a frequency of 100 Hz through a four-wire constant current source circuit, and linear correction is performed to output the ambient temperature parameter value.

[0015] S14: Using the rising edge of the solenoid valve power supply as the hardware trigger signal, the FPGA synchronization controller sends a synchronous acquisition instruction to the current sensor, acceleration sensor and temperature sensor, with a timing deviation of ≤1μs.

[0016] Optionally, the formula for performing linearization correction is: ,in, is the corrected ambient temperature; is the measured resistance value; The reference resistance value at 0℃; is the temperature coefficient of PT1000, which is .

[0017] Optionally, the current characteristic values ​​include a peak current deviation rate, a rising edge slope, and a steady-state current fluctuation variance.

[0018] Optionally, the vibration energy distribution is decomposed by wavelet packets to extract the energy proportion entropy in the 0-5 kHz frequency band.

[0019] Optionally, calculating the temperature drift includes:

[0020] Set the average temperature within 10 seconds after power on as the initial temperature ;

[0021] Collect the current real-time temperature, recorded as ;

[0022] The calculation formula for temperature drift is: .

[0023] Optionally, the S3 specifically includes:

[0024] S31: Construct the three extracted parameters into a three-dimensional feature vector ; and for the vector Implement minimum-maximum normalization to obtain a normalized vector ;

[0025] S32: Call the covariance matrix constructed by pre-stored normal state feature data ;

[0026] S33: Normalized real-time feature vector Compare with the normal sample mean vector and calculate the Mahalanobis distance , the formula is: ,in, is the normalized mean vector of normal samples, with dimension ; is the inverse of the covariance matrix;

[0027] S34: Mahalanobis distance Perform linear normalization to generate the coupling deviation coefficient, which is calculated as: ,in, is the coupling deviation coefficient, the value range is ; is the maximum distance threshold.

[0028] Optionally, the S4 specifically includes:

[0029] S41: Read the pre-stored baseline health coefficient ;

[0030] S42: Coupling deviation coefficient Input the exponential decay function to calculate the comprehensive health index. The formula is: ,in, is the comprehensive health index; is the exponential decay factor; Expressed as a natural constant An exponential function with base ;

[0031] S43: Comprehensive health index Set limits to prevent extreme values ​​from affecting diagnostic stability. Specifically, when >0.98, the output is forced to be 0.98; when When <0.01, the output is forced to be 0.01; in other cases, the output value is rounded to three decimal places.

[0032] Optionally, the S5 specifically includes:

[0033] S51: Based on the comprehensive health index The value of the current device status is divided by the size of , it is judged to be in a healthy state; if , it is judged as a warning state; if , then the fault diagnosis process is triggered;

[0034] S52: When the fault diagnosis process is triggered, according to the coupling deviation coefficient , vibration energy entropy , temperature drift Compare the predefined rules in the fault feature library to determine the fault type. When the coil is short-circuited, ,and When the valve core is stuck, ,and ℃, it is judged as spring fatigue; if all conditions of any fault type are met, the corresponding fault code will be output;

[0035] S53: Determine the diagnostic location from the physical location mapping table according to the matched fault type;

[0036] S54: The fault diagnosis result is structured and encapsulated into a JSON format for output, where the output content includes the fault code and the fault location.

[0037] An intelligent solenoid valve fault detection system based on multi-parameter analysis is used to implement the above-mentioned intelligent solenoid valve fault detection method based on multi-parameter analysis, including the following modules:

[0038] Signal acquisition module: used to collect the dynamic current waveform of the power supply circuit, the vibration spectrum characteristics of the solenoid valve housing, and the ambient temperature parameters of the coil surface during the operation of the solenoid valve;

[0039] Feature extraction module: connected to the signal acquisition module, extracts corresponding characteristic parameters based on the three types of signals collected, including current characteristic values, vibration energy distribution characteristics and temperature drift;

[0040] Coupling analysis module: connected to the feature extraction module, used to construct the extracted feature parameters into a standardized parameter vector, match it with the historical normal sample covariance matrix, and calculate the current coupling deviation coefficient based on the Mahalanobis distance;

[0041] Health assessment module: connected to the coupling analysis module, used to input the coupling deviation coefficient into the exponential decay function, output the comprehensive health index, and implement dynamic range constraints on the comprehensive health index value;

[0042] Fault identification module: Connected to the health assessment module, it performs status classification based on the comprehensive health index. When the health index is lower than the preset threshold, it calls the fault feature library to match the fault type, locates the physical location of the fault based on the matching results, and finally outputs the diagnosis results in a structured format.

[0043] Beneficial effects of the present invention:

[0044] The present invention realizes high-precision real-time monitoring of the operating status of the solenoid valve by synchronously collecting current, vibration and temperature signals through multiple parameters, and can comprehensively reflect the dynamic characteristics of the equipment and environmental changes; combined with multi-source feature fusion and coupling analysis, it can accurately identify different types of fault conditions and quantify the severity of the fault.

[0045] The present invention, by adopting an exponentially decaying health index evaluation method and a graded judgment mechanism, effectively improves the accuracy of early abnormality warning and fault location, and realizes structured output of fault type and physical location, thereby improving the intelligence level of solenoid valve fault detection and actual operation and maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0047] Figure 1 Schematic diagram of an intelligent detection method for solenoid valve faults according to an embodiment of the present invention;

[0048] Figure 2 Schematic diagram of an intelligent solenoid valve fault detection system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. It is also noted that, to provide a more detailed description, the following embodiments are best and preferred embodiments, and those skilled in the art may employ alternative methods for implementing certain known technologies. Furthermore, the accompanying drawings are intended only to provide a more detailed description of the embodiments and are not intended to limit the present invention.

[0050] It should be noted that references in the specification to "one embodiment," "an embodiment," "exemplary embodiments," "some embodiments," etc. indicate that the described embodiments may include specific features, structures, or characteristics, but not necessarily every embodiment will include such specific features, structures, or characteristics. Furthermore, when specific features, structures, or characteristics are described in conjunction with an embodiment, it is within the knowledge of persons skilled in the relevant art to implement such features, structures, or characteristics in conjunction with other embodiments (whether or not explicitly described).

[0051] In general, terms can be understood, at least in part, from their use in context. For example, depending at least in part on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending at least in part on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0052] like Figure 1 As shown, a solenoid valve fault intelligent detection method based on multi-parameter analysis includes the following steps:

[0053] S1: During the operation of the solenoid valve, its dynamic current waveform, vibration spectrum information and ambient temperature data are synchronously collected;

[0054] S2: Extract the current characteristic value from the dynamic current waveform, extract the vibration energy distribution from the vibration spectrum characteristics, and calculate the temperature drift from the ambient temperature parameters;

[0055] S3: Input the current characteristic value, vibration energy distribution and temperature drift into the multi-parameter correlation matrix to generate a coupling deviation coefficient that characterizes the coupling relationship between the parameters;

[0056] S4: Calculate the comprehensive health index of the solenoid valve based on the coupling deviation coefficient;

[0057] S5: Match the comprehensive health index with the preset fault feature library and output the corresponding fault type and location result.

[0058] S1 specifically includes:

[0059] S11: Use the open-loop Hall current sensor HCT-20A to collect the current signal in real time in the solenoid valve power supply circuit. The sampling frequency is set to 10kHz and the resolution is 16 bits to generate dynamic current waveform raw data.

[0060] S12: Install an ICP accelerometer (model PCB352C33) at the axial center of the solenoid valve housing. Use an anti-aliasing filter to intercept the 0-10 kHz frequency band signal. Obtain the vibration time domain waveform at a 12.8 kHz sampling rate. Generate the vibration spectrum characteristics through fast Fourier transform.

[0061] S13: A PT1000 platinum resistance temperature sensor is mounted on the surface of the solenoid valve coil housing. The original temperature signal is obtained at a frequency of 100 Hz through a four-wire constant current source circuit, and linear correction is performed to output the ambient temperature parameter value.

[0062] S14: Using the rising edge of the solenoid valve power supply as the hardware trigger signal, the FPGA synchronization controller (Xilinx Artix-7) sends synchronous acquisition instructions to the current sensor, acceleration sensor and temperature sensor, with a timing deviation of ≤1μs. Through the synchronous acquisition of the three physical quantities of current, vibration and temperature, combined with high-precision sampling, frequency domain processing and hardware triggering mechanism, the consistency of multi-source signals and the ability to capture the dynamic evolution process of faults are effectively improved, providing a highly reliable data foundation for subsequent feature extraction and state identification.

[0063] The formula for performing linearization correction is: ,in, is the corrected ambient temperature (unit: °C); is the measured resistance value; The reference resistance value at 0℃; is the temperature coefficient of PT1000, which is .

[0064] The current characteristic values ​​include the peak current deviation rate, rising edge slope and steady-state current fluctuation variance.

[0065] The steps to extract the current characteristic value are as follows:

[0066] First, take the first 50 milliseconds of the current waveform starting from the moment the solenoid valve is powered on, and then calculate the following three characteristic indicators:

[0067] Peak current deviation rate: ,in, is the peak current deviation rate; is the maximum current value measured; is the rated peak current;

[0068] Rising slope: When the current waveform reaches to Perform linear fitting within the amplitude range and take the slope of the fitting line. The formula is: ,in, is the rising edge slope; For the current from to changes in amplitude; is the corresponding time interval;

[0069] Steady-state current fluctuation variance: Extract 100 sampling points within the time period and calculate the variance of the current value. The formula is: , For the The current value of each sampling point; for The mean of the sampling points; .

[0070] The vibration energy distribution is decomposed by wavelet packets to extract the energy proportion entropy in the 0-5kHz frequency band.

[0071] The specific steps are as follows:

[0072] First, a 4-layer wavelet packet decomposition is performed on the vibration spectrum signal to divide the 0-5kHz frequency band into 16 equal-width sub-bands;

[0073] Then, the energy proportion of each sub-band is calculated as follows: ,in, For the The energy proportion of each sub-band; For the Sub-band The spectrum amplitude of the point; For the The number of spectrum points in the sub-band;

[0074] Finally, the energy proportion entropy is calculated based on the energy proportion: ,in, is the energy proportion entropy; is the sub-band energy ratio.

[0075] Calculating the temperature drift includes:

[0076] Set the average temperature within 10 seconds after power on as the initial temperature ;

[0077] Collect the current real-time temperature, recorded as ;

[0078] The calculation formula for temperature drift is: .

[0079] Through the multiple feature extraction of the above-mentioned current peak, rising characteristics and steady-state fluctuations, as well as the quantitative calculation of the multi-band energy structure and temperature change trend, the accurate expression of the multi-dimensional abnormal characteristics of the solenoid valve working state is achieved, providing a high-resolution input basis for coupling correlation analysis and fault mode identification.

[0080] S3 specifically includes:

[0081] S31: Construct the three extracted parameters into a three-dimensional feature vector , where the current eigenvalue scalar Calculated by the following weighted combination: ,in, is the peak current deviation rate; is the rising edge slope; is the base slope; is the steady-state current fluctuation variance; vibration energy distribution value , that is, energy proportion entropy; temperature drift ; and for the vector Implement minimum-maximum normalization to obtain a normalized vector , where each component is calculated as follows: ,in, For the Item original features; : No. The minimum and maximum values ​​of the item features;

[0082] S32: Call the covariance matrix constructed by pre-stored normal state feature data , which is defined as follows: ,in, The three-dimensional normalized feature vector set consisting of 200 groups of normal samples has a dimension of ; represents the covariance operation; The result is The real symmetric positive definite matrix is ​​stored in Flash;

[0083] S33: Normalized real-time feature vector Compare with the normal sample mean vector and calculate the Mahalanobis distance , the formula is: ,in, is the normalized mean vector of normal samples, with dimension ; is the inverse of the covariance matrix; is the Mahalanobis distance, which reflects the overall deviation between the current state and the normal state;

[0084] S34: Mahalanobis distance Perform linear normalization to generate the coupling deviation coefficient, which is calculated as: ,in, is the coupling deviation coefficient, the value range is ; is the maximum distance threshold, specifically defined as 3 times the standard deviation of the normal sample Mahalanobis distance distribution, that is: ,in 、 are the mean and standard deviation of the Mahalanobis distance of normal samples respectively; the above steps realize the dynamic quantification of the coupling relationship between parameters by constructing multi-source feature vectors and introducing normalization processing, covariance matrix matching and Mahalanobis distance calculation, so that the system can accurately identify various types of hidden anomalies that deviate from the normal state.

[0085] S4 specifically includes:

[0086] S41: Read the pre-stored baseline health coefficient , this coefficient is calculated from the mean value of the coupling deviation coefficient under historical normal conditions and serves as a reference benchmark for the comprehensive health index;

[0087] S42: Coupling deviation coefficient Input the exponential decay function to calculate the comprehensive health index. The formula is: ,in, is the comprehensive health index; is the exponential decay factor; Expressed as a natural constant An exponential function with base ;

[0088] S43: Comprehensive health index Set limits to prevent extreme values ​​from affecting diagnostic stability. Specifically, when >0.98, the output is forced to be 0.98; when When the value is less than 0.01, the output is forced to be 0.01; in other cases, the output value is rounded to three decimal places. The above steps adopt a health index evaluation method based on the exponential decay model, so that the system can output a state evaluation value of a nonlinear response when facing different degrees of parameter deviation, thereby enhancing the ability to identify abnormalities in the early stage. By setting a dynamic range limit, the output stability and controllability are further guaranteed.

[0089] S5 specifically includes:

[0090] S51: Based on the comprehensive health index The value of the current device status is divided by the size of , it is judged to be in a healthy state; if , it is judged as a warning state; if , then the fault diagnosis process is triggered;

[0091] S52: When the fault diagnosis process is triggered, according to the coupling deviation coefficient , vibration energy entropy , temperature drift Compare the predefined rules in the fault feature library to determine the fault type. When the coil is short-circuited, ,and When the valve core is stuck, ,and ℃, it is judged as spring fatigue; if all conditions of any fault type are met, the corresponding fault code will be output;

[0092] S53: According to the matched fault type, the diagnostic location is determined from the physical location mapping table. The mapping relationship is as follows: coil short circuit → electromagnetic coil winding; valve core stuck → valve core movement chamber; spring fatigue → reset spring assembly;

[0093] S54: The fault diagnosis results are structured and encapsulated in JSON format for output, including the fault code and fault location. By constructing a grading strategy based on health index thresholds and integrating coupling parameters with additional physical condition criteria for fault identification and component location, the accuracy and interpretability of fault diagnosis are effectively improved. At the same time, the use of structured output enhances the system's support for subsequent maintenance scheduling and human-machine interface.

[0094] like Figure 2 As shown, a solenoid valve fault intelligent detection system based on multi-parameter analysis is used to implement the above-mentioned solenoid valve fault intelligent detection method based on multi-parameter analysis, including the following modules:

[0095] Signal acquisition module: used to collect the dynamic current waveform of the power supply circuit, the vibration spectrum characteristics of the solenoid valve housing, and the ambient temperature parameters of the coil surface during the operation of the solenoid valve;

[0096] Feature extraction module: This module is connected to the signal acquisition module and extracts corresponding characteristic parameters based on the three types of signals collected, including current characteristic values, vibration energy distribution characteristics, and temperature drift. The current characteristic values ​​are obtained through weighted calculation of the peak deviation rate, rising edge slope, and steady-state fluctuation variance. The vibration energy distribution characteristics are represented by the energy proportion entropy after wavelet packet decomposition. The temperature drift is the difference between the real-time temperature and the initial reference temperature.

[0097] Coupling Analysis Module: This module is connected to the Feature Extraction Module and is used to construct the extracted feature parameters into a standardized parameter vector and match it with the covariance matrix of historical normal samples. The coupling deviation coefficient is calculated based on the Mahalanobis distance. The coupling deviation coefficient represents the comprehensive deviation degree between multi-source signals.

[0098] Health assessment module: connected to the coupling analysis module, used to input the coupling deviation coefficient into the exponential decay function, output the comprehensive health index, and implement dynamic range constraints on the comprehensive health index value to form a stable and controllable health status assessment result;

[0099] Fault identification module: Connected to the health assessment module, it performs status classification based on the comprehensive health index. When the health index is lower than the preset threshold, it calls the fault feature library to match the fault type, locates the physical location of the fault based on the matching results, and finally outputs the diagnosis results in a structured format.

[0100] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail below in connection with the preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0101] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. An intelligent detection method for electromagnetic valve faults based on multi-parameter analysis, characterized in that: The following steps are involved: S1: During the operation of the solenoid valve, its dynamic current waveform, vibration spectrum information and ambient temperature data are synchronously collected; S2: Extract the current characteristic value from the dynamic current waveform, extract the vibration energy distribution from the vibration spectrum characteristics, and calculate the temperature drift from the ambient temperature parameters; S3: Input the current characteristic value, vibration energy distribution and temperature drift into the multi-parameter correlation matrix to generate a coupling deviation coefficient that characterizes the coupling relationship between the parameters; The S3 specifically includes: S31: Construct the three extracted parameters into a three-dimensional feature vector ; and for the vector Implement minimum-maximum normalization to obtain a normalized vector ; S32: Call the covariance matrix constructed by pre-stored normal state feature data ; S33: Normalized real-time feature vector Compare with the normal sample mean vector and calculate the Mahalanobis distance , the formula is: ,in, is the normalized mean vector of normal samples, with dimension ; is the inverse of the covariance matrix; S34: Mahalanobis distance Perform linear normalization to generate the coupling deviation coefficient, which is calculated as: ,in, is the coupling deviation coefficient, the value range is ; is the maximum distance threshold; S4: Calculate the comprehensive health index of the solenoid valve based on the coupling deviation coefficient; The S4 specifically includes: S41: Read the pre-stored baseline health coefficient ; S42: Coupling deviation coefficient Input the exponential decay function to calculate the comprehensive health index. The formula is: ,in, is the comprehensive health index; is the exponential decay factor; Expressed as a natural constant An exponential function with base ; S43: Comprehensive health index Set limits to prevent extreme values ​​from affecting diagnostic stability. Specifically, when >0.98, the output is forced to be 0.98; when When <0.01, the output is forced to be 0.01; in other cases, the output value is rounded to three decimal places; S5: Match the comprehensive health index with the preset fault feature library and output the corresponding fault type and location result.

2. The intelligent detection method for electromagnetic valve faults based on multi-parameter analysis according to claim 1 is characterized in that: Said S1 specifically includes: S11: Use the open-loop Hall current sensor HCT-20A to collect the current signal in real time in the solenoid valve power supply circuit. The sampling frequency is set to 10kHz and the resolution is 16 bits to generate dynamic current waveform raw data. S12: Install an ICP type acceleration sensor at the axial center point of the solenoid valve body housing. Use an anti-aliasing filter to intercept the 0-10kHz frequency band signal, obtain the vibration time domain waveform at a 12.8kHz sampling rate, and generate the vibration spectrum characteristics through fast Fourier transform. S13: A PT1000 platinum resistance temperature sensor is mounted on the surface of the solenoid valve coil housing. The original temperature signal is obtained at a frequency of 100 Hz through a four-wire constant current source circuit, and linear correction is performed to output the ambient temperature parameter value. S14: Using the rising edge of the solenoid valve power supply as the hardware trigger signal, the FPGA synchronization controller sends a synchronous acquisition instruction to the current sensor, acceleration sensor and temperature sensor, with a timing deviation of ≤1μs.

3. The intelligent detection method for electromagnetic valve faults based on multi-parameter analysis according to claim 2 is characterized in that: The formula for performing linearization correction is: ,in, is the corrected ambient temperature; is the measured resistance value; The reference resistance value at 0℃; is the temperature coefficient of PT1000, which is .

4. The intelligent detection method for electromagnetic valve faults based on multi-parameter analysis according to claim 1 is characterized in that: The current characteristic values ​​include peak current deviation rate, rising edge slope and steady-state current fluctuation variance.

5. The intelligent detection method for electromagnetic valve faults based on multi-parameter analysis according to claim 1 is characterized in that: The vibration energy distribution is decomposed by wavelet packets to extract the energy proportion entropy in the 0-5kHz frequency range.

6. The intelligent detection method for electromagnetic valve faults based on multi-parameter analysis according to claim 1 is characterized in that: The calculation of the temperature drift comprises: Set the average temperature within 10 seconds after power on as the initial temperature ; Collect the current real-time temperature, recorded as ; The calculation formula for temperature drift is: .

7. The intelligent detection method for electromagnetic valve faults based on multi-parameter analysis according to claim 1 is characterized in that: The S5 specifically includes: S51: Based on the comprehensive health index The value of the current device status is divided by the size of , it is judged to be in a healthy state; if , it is judged as a warning state; if , then the fault diagnosis process is triggered; S52: When the fault diagnosis process is triggered, according to the coupling deviation coefficient , vibration energy entropy , temperature drift Compare the predefined rules in the fault feature library to determine the fault type. When the coil is short-circuited, ,and When the valve core is stuck, ,and ℃, it is judged as spring fatigue; if all conditions of any fault type are met, the corresponding fault code will be output; S53: Determine the diagnostic location from the physical location mapping table according to the matched fault type; S54: The fault diagnosis result is structured and encapsulated into a JSON format for output, where the output content includes the fault code and the fault location.

8. An intelligent solenoid valve fault detection system based on multi-parameter analysis, used to implement the intelligent solenoid valve fault detection method based on multi-parameter analysis according to any one of claims 1 to 7, characterized in that: Includes the following modules: Signal acquisition module: used to collect the dynamic current waveform of the power supply circuit, the vibration spectrum characteristics of the solenoid valve housing, and the ambient temperature parameters of the coil surface during the operation of the solenoid valve; Feature extraction module: connected to the signal acquisition module, extracts corresponding characteristic parameters based on the three types of signals collected, including current characteristic values, vibration energy distribution characteristics and temperature drift; Coupling analysis module: connected to the feature extraction module, used to construct the extracted feature parameters into a standardized parameter vector, match it with the historical normal sample covariance matrix, and calculate the current coupling deviation coefficient based on the Mahalanobis distance; Health assessment module: connected to the coupling analysis module, used to input the coupling deviation coefficient into the exponential decay function, output the comprehensive health index, and implement dynamic range constraints on the comprehensive health index value; Fault identification module: Connected to the health assessment module, it performs status classification based on the comprehensive health index. When the health index is lower than the preset threshold, it calls the fault feature library to match the fault type, locates the physical location of the fault based on the matching results, and finally outputs the diagnosis results in a structured format.

Images