New energy high-stability control valve intelligent detection method and system

Abstract

The invention relates to the technical field of control valves, in particular to an intelligent detection method and system for a new energy high-stability control valve, and the method comprises the following steps: collecting pressure sensor data of an inlet and an outlet of a control valve, monitoring the current inlet pressure and the current outlet pressure of the valve, and calculating the pressure difference between the inlet pressure and the outlet pressure through continuous data collection; and analyzing whether the difference value exceeds the range of the conventional operation data or not. Earlier fault diagnosis and early warning are realized by analyzing fine changes of pressure difference and sound frequency, so that energy loss is reduced, the operation efficiency of a valve path system is improved, intelligent processing of pressure difference and sound data is integrated, the problem of the control valve can be quickly reflected, the working state of the control valve can be accurately judged, and the safety of the control valve is improved. Therefore, the gas supply efficiency and safety of the hydrogen energy automobile are optimized, faults can be predicted and prevented by continuously monitoring the function state of the control valve, the service life of the control valve path is prolonged, and the working efficiency of the valve path is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of control valve circuits, and particularly to an intelligent detection method and system for a new energy high-stability control valve circuit. Background Art

[0002] The control valve circuit technology involves using various types of control valves to regulate fluid flow, including the flow rate, pressure, and temperature of gases, liquids, or steam. These control valves, such as solenoid valves, proportional valves, or rotary valves, are usually integrated into a broader automation system to optimize the efficiency and safety of industrial processes. The design and implementation of the control valve circuit need to consider the response speed, accuracy of the valve, and compatibility with other parts of the system.

[0003] Among them, the intelligent detection method for a new energy high-stability control valve circuit is a technology for intelligently detecting and controlling a control valve circuit in a new energy vehicle system. This method monitors the performance of the control valve by integrating sensor technology and data analysis software to ensure the optimization of valve operation, thereby improving the stability and efficiency of the entire hydrogen energy vehicle system. It has a wide range of applications, including improving hydrogen supply efficiency, reducing energy losses, preventing equipment failures, and extending the life of hydrogen energy vehicles through continuous monitoring.

[0004] The prior art cannot provide sufficient real-time data monitoring and instant fault analysis, resulting in slow response when facing fluid dynamic changes, inability to adjust in real time, easy reduction of the valve circuit system efficiency, and even safety accidents. If the abnormal changes in hydrogen flow cannot be recognized and adjusted in real time, it will lead to gas leakage or excessive pressure, increasing the operation risk and maintenance cost of the valve circuit system. The prior art is insufficient in early fault detection and accurate assessment of the valve circuit system state, often intervening only after the problem develops to a serious stage, increasing the valve circuit complexity and potential fault repair costs. Summary of the Invention

[0005] The purpose of the present invention is to solve the disadvantages existing in the prior art, and propose an intelligent detection method and system for a new energy high-stability control valve circuit. [[ID=2??]]

[0006] To achieve the above purpose, the present invention adopts the following technical solution: An intelligent detection method for a new energy high-stability control valve circuit includes the following steps. S1: Collect the pressure sensor data at the inlet and outlet of the control valve, monitor the current inlet pressure and outlet pressure of the valve, calculate the pressure difference between the two through continuous data acquisition, analyze whether the difference exceeds the range of conventional operation data, and record the deviation in the data to obtain the pressure difference offset data. S2: Collect acoustic wave data during the operation of the control valve through an acoustic sensor, continuously monitor the acoustic wave frequency and intensity, perform acoustic spectrum analysis on the collected acoustic wave data, identify abnormal points inconsistent with the conventional acoustic wave characteristics, and obtain abnormal acoustic wave indicators; S3: Analyze the differential pressure offset data, calculate the differential pressure change rate, evaluate the frequency and amplitude of the differential pressure fluctuation within a continuous time period, identify the key time intervals of the differential pressure fluctuation, and perform gradient analysis on the data in the time intervals to determine whether there is an abnormal differential pressure change trend, and obtain the differential pressure fluctuation trend; S4: Based on the abnormal acoustic wave indicators, analyze the abnormal signal intensity and duration, determine the signal intervals that conform to the abnormal acoustic characteristic pattern, calculate the abnormal signal frequency within the intervals and its distribution ratio during the operation process, judge potential control valve structure problems, and obtain acoustic anomaly diagnosis data.

[0007] The improvements of the present invention are that the differential pressure offset data includes inlet pressure deviation, outlet pressure deviation, and pressure offset index, the abnormal acoustic wave indicators include frequency deviation value, intensity deviation value, and abnormal spectrum identification, the differential pressure fluctuation trend includes fluctuation rate index, frequency fluctuation index, and amplitude fluctuation index, and the acoustic anomaly diagnosis data includes acoustic signal intensity index, acoustic duration index, and acoustic frequency distribution.

[0008] The improvements of the present invention are that the acquisition steps of the differential pressure offset data are specifically as follows: S111: Collect the pressure sensor data at the inlet and outlet of the control valve, continuously collect the inlet pressure value and outlet pressure value within a certain time, record the corresponding time series, and then calculate the instantaneous pressure difference at each time point to generate an instantaneous pressure difference sequence; S112: Based on the instantaneous pressure difference sequence, perform standardization processing on the sequence, using the formula: ; Calculate the standardized pressure difference at each time point , and obtain the standardized pressure difference trend value, where represents the inlet pressure at the th time point, represents the outlet pressure at the th time point, and are the mean and standard deviation of the instantaneous pressure difference sequence respectively; S113: Call the standardized pressure difference trend value, judge whether it exceeds the normal operation range, screen the data that exceeds the normal range, and record the deviation situation in the data to obtain the differential pressure offset data.

[0009] The improvements of the present invention are that the acquisition steps of the abnormal acoustic wave indicators are specifically as follows: S211: Collect the acoustic wave data during the operation of the control valve through an acoustic sensor, continuously monitor the acoustic wave frequency and intensity, organize them into time series data, and generate an acoustic wave time series; S212: Based on the acoustic wave time series, analyze the frequency offset and intensity change, and use the formula: ; Calculate the acoustic wave feature offset degree at each time point , where represents the acoustic wave frequency at the th time point, represents the acoustic wave intensity at the th time point, and are the mean and standard deviation of the acoustic wave frequency respectively, and are the mean and standard deviation of the acoustic wave intensity respectively; S213: Based on the acoustic wave feature offset degree, identify the abnormal points with offsets exceeding the normal range, screen the abnormal acoustic wave data, and extract the corresponding key acoustic parameters to obtain the abnormal acoustic wave index.

[0010] The improvement of the present invention is that the step of obtaining the differential pressure fluctuation trend is specifically as follows: S311: Analyze the differential pressure offset data, collect the differential pressure values at each time point from the data source, and calculate the differential pressure change rate between adjacent time points; S312: Based on the differential pressure change rate, analyze the fluctuation frequency and amplitude within a continuous time period, and use the formula: ; Calculate the standard deviation of the fluctuation amplitude , to obtain the key time interval of differential pressure fluctuation, where represents the differential pressure value at the th time point, represents the average value of the differential pressure at the time point, represents the total number of time points; S313: Based on the key time interval of differential pressure fluctuation, perform gradient analysis on the differential pressure data within its range, judge whether there is an abnormal trend, and extract the trend characteristics to obtain the differential pressure fluctuation trend.

[0011] The improvement of the present invention is that the step of obtaining the acoustic anomaly diagnosis data is specifically as follows: S411: Based on the abnormal acoustic wave index, analyze the abnormal signal intensity and duration, judge the signal interval that conforms to the abnormal acoustic feature pattern, calculate the abnormal signal frequency within the interval, and use the formula: ; Calculate the proportion of abnormal signal distribution , where represents the frequency of abnormal signals in the th time period, represents the total number of measurements during the total operation period; S412: Based on the proportion of abnormal signal distribution, combined with the intensity and duration of abnormal signals, analyze the distribution characteristics of acoustic abnormal signals, judge the existing control valve structure problems, and obtain acoustic abnormal diagnosis data.

[0012] The improvement of the present invention is that the step further includes: S5: Based on the pressure difference fluctuation trend and acoustic abnormal diagnosis data, analyze the data correlation, identify the current state of the control valve path, calculate the degree of matching with known fault modes, and judge the warning level corresponding to the current state to obtain the current state of the control valve path; The current state of the control valve path includes fault mode consistency and warning level index.

[0013] The improvement of the present invention is that the specific steps for obtaining the current state of the control valve path are: S511: Based on the pressure difference fluctuation trend and acoustic abnormal diagnosis data, analyze the time series characteristics of each parameter, screen the characteristic data in the corresponding time interval to obtain the time-corresponding characteristic data; S512: Call the time-corresponding characteristic data, identify the current state of the control valve path, compare with the characteristic data of known fault modes, and use the formula: ; Calculate the degree of matching with known fault modes to obtain the current state fault similarity , where represents the th dimensional time-corresponding characteristic data of the current state, represents the th dimensional characteristic value of the corresponding known fault mode, represents the total number of characteristic dimensions; S513: According to the current state fault similarity, compare with the warning levels of fault modes, judge the warning level corresponding to the control valve path, and obtain the current state of the control valve path.

[0014] A new energy high-stability control valve path intelligent detection system, the system includes: The pressure difference monitoring module collects the pressure sensor data at the inlet and outlet of the control valve, monitors the current inlet pressure and outlet pressure of the valve, calculates the pressure difference between the two, analyzes whether the difference exceeds the range of conventional operation data, and records the deviation in the data to obtain the pressure difference offset data; The acoustic wave analysis module collects acoustic wave data during the operation of the control valve through an acoustic sensor, continuously monitors the acoustic wave frequency and intensity, performs spectrogram analysis on the collected acoustic wave data, identifies abnormal points inconsistent with the conventional acoustic wave characteristics, and obtains abnormal acoustic wave indicators; The differential pressure fluctuation analysis module analyzes the differential pressure offset data, evaluates the frequency and amplitude of the differential pressure fluctuation within a continuous time period, identifies the key time intervals of the differential pressure fluctuation, determines whether there is an abnormal differential pressure change trend, and obtains the differential pressure fluctuation trend; The acoustic anomaly diagnosis module, based on the abnormal acoustic wave indicators, analyzes the abnormal signal intensity and duration, determines the signal interval that conforms to the abnormal acoustic characteristic pattern, calculates the abnormal signal frequency within the interval and its distribution ratio during the operation process, judges potential control valve structure problems, and obtains acoustic anomaly diagnosis data; The status evaluation module, based on the differential pressure fluctuation trend and the acoustic anomaly diagnosis data, identifies the current state of the control valve path, calculates the degree of matching with known failure modes, and judges the warning level corresponding to the current state, and obtains the current state of the control valve path.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In the present invention, by analyzing the subtle changes in pressure difference and sound frequency, earlier fault diagnosis and warning are realized, thereby reducing energy loss and improving the operation efficiency of the valve path system. The intelligent processing of comprehensive differential pressure and sound data can not only quickly respond to control valve problems but also accurately judge the working state of the control valve, thereby optimizing the gas supply efficiency and safety of hydrogen energy vehicles. By continuously monitoring the functional state of the control valve, faults can be predicted and prevented, thereby prolonging the service life of the control valve path and improving the working efficiency of the valve path. Description of the Drawings

[0016] Figure 1 It is a flow chart of the intelligent detection method for a new energy high-stability control valve path proposed by the present invention; Figure 2 It is a flow chart of the acquisition steps of the differential pressure offset data in the present invention; Figure 3 It is a flow chart of the acquisition steps of the abnormal acoustic wave indicators in the present invention; Figure 4 It is a flow chart of the acquisition steps of the differential pressure fluctuation trend in the present invention; Figure 5 It is a flow chart of the acquisition steps of the acoustic anomaly diagnosis data in the present invention; Figure 6 It is a flow chart of the acquisition steps of the current state of the control valve path in the present invention. Detailed Embodiments

[0017] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0018] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by terms such as "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention. In addition, in the description of the present invention, "a plurality of" means two or more, unless otherwise specifically defined. Embodiment

[0019] Please refer to Figure 1 , the present invention provides a technical solution: a new energy high-stability control valve path intelligent detection method, including the following steps: S1: Collect the pressure sensor data at the inlet and outlet of the control valve, monitor the current inlet pressure and outlet pressure of the valve, calculate the pressure difference between the two through continuous data acquisition, analyze whether the difference exceeds the range of conventional operation data, and record the deviation in the data to obtain the pressure difference offset data; S2: Collect the acoustic wave data during the operation of the control valve through an acoustic sensor, continuously monitor the acoustic wave frequency and intensity, perform spectrogram analysis on the collected acoustic wave data, and identify the abnormal points inconsistent with the conventional acoustic wave characteristics to obtain the abnormal acoustic wave index; S3: Analyze the pressure difference offset data, calculate the pressure difference change rate, evaluate the frequency and amplitude of the pressure difference fluctuation within a continuous time period, identify the key time intervals of the pressure difference fluctuation, and perform gradient analysis on the data in the time intervals to determine whether there is an abnormal pressure difference change trend to obtain the pressure difference fluctuation trend; S4: Based on the abnormal acoustic wave index, analyze the abnormal signal intensity and duration, determine the signal interval that conforms to the abnormal acoustic characteristic pattern, calculate the abnormal signal frequency within the interval and its distribution ratio during the operation process, and judge the potential control valve structure problem to obtain the acoustic anomaly diagnosis data; S5: Based on the pressure difference fluctuation trend and the acoustic anomaly diagnosis data, analyze the data correlation, identify the current state of the control valve path, calculate the degree of matching with the known fault modes, and judge the warning level corresponding to the current state to obtain the current state of the control valve path.

[0020] The differential pressure offset data includes the inlet pressure deviation, the outlet pressure deviation, and the pressure offset index. The abnormal acoustic wave indicators include the frequency deviation value, the intensity deviation value, and the abnormal frequency spectrum identifier. The differential pressure fluctuation trend includes the fluctuation rate index, the frequency fluctuation index, and the amplitude fluctuation index. The acoustic anomaly diagnosis data includes the acoustic signal intensity index, the acoustic duration index, and the acoustic frequency distribution. The current state of the control valve circuit includes the fault mode consistency and the warning level index.

[0021] Please refer to Figure 2 , and the steps for obtaining the differential pressure offset data are specifically as follows: S111: Collect the pressure sensor data at the inlet and outlet of the control valve, continuously acquire the inlet pressure value and the outlet pressure value within a certain period of time, record the corresponding time series, and then calculate the instantaneous pressure difference at each time point to generate an instantaneous pressure difference sequence; Collect the pressure sensor data at the inlet and outlet of the control valve. The pressure sensors of the control valve can be installed at the inlet and outlet positions of the pipeline to measure the fluid pressure entering the valve and the fluid pressure flowing out of the valve respectively. The sensors will continuously acquire the pressure data within a certain time period and record the inlet pressure and the outlet pressure at each moment according to the time stamp. The time interval of data acquisition depends on the requirements of the control system. For example, it is recorded in seconds or milliseconds. The inlet pressure and the outlet pressure are stored through the data recording system to obtain a complete time series data set. For each time point , calculate the corresponding instantaneous pressure difference, that is , where represents the instantaneous pressure difference at a certain time point. After calculating the instantaneous pressure differences at all time points, organize them into a continuous pressure difference data sequence. For example, if the inlet pressure and the outlet pressure of a certain valve are recorded as 5.2 MPa and 4.8 MPa respectively, the instantaneous pressure difference is 0.4 MPa. As time goes by, multiple instantaneous pressure differences will form a time series, and this series can be used for further analysis of the valve operation condition to generate an instantaneous pressure difference sequence.

[0022] S112: Based on the instantaneous pressure difference sequence, perform normalization processing on the sequence, using the formula: ; Calculate the normalized pressure difference at each time point , and obtain the normalized pressure difference trend value, where represents the inlet pressure at the th time point, represents the outlet pressure at the th time point, and are the mean and standard deviation of the instantaneous pressure difference sequence respectively; Based on the instantaneous pressure difference sequence, perform standardization on it. The purpose of standardization is to eliminate the fluctuation influence of pressure data at different time points and make the data more comparable. First, calculate the mean of the instantaneous pressure difference sequence and the standard deviation. The calculation method of the mean is as follows: , and the calculation method of the standard deviation is as follows: , where represents the total number of data points, represents the th instantaneous pressure difference value at a time point. After obtaining the mean and the standard deviation, calculate the standardized pressure difference value at each time point. The instantaneous pressure difference values collected by a certain control valve at 3 time points are as follows (unit: MPa): ; Calculate the mean : ; Calculate the standard deviation : ; Calculate the sum of squared differences: ; ; ; Sum of squared differences: ; ; Calculate the standardized pressure difference value at each time point: ; ; ; After calculating the standardized pressure difference values of all data points, obtain the complete standardized pressure difference trend value: ; This trend value can reflect the stability of the pressure change of the control valve, and obtain the standardized pressure difference trend value.

[0023] S113: Call the standardized pressure difference trend value, judge whether it exceeds the normal operation range, screen the data that exceeds the normal range, and record the deviation situation in the data to obtain the differential pressure offset data; Call the standardized pressure difference trend value and judge whether it exceeds the normal operation range. Set the thresholds and of the normal operation range. If the standardized pressure difference trend value Out of range , then the pressure difference at that time point is abnormal, filter out the time points that are beyond the normal range, and record the data deviation at that time point. For example, set the normal range to , if the standardized pressure difference calculated at a certain moment , then the data point exceeds the normal range and is recorded as an abnormal deviation point. After screening all data points that exceed the threshold, the number of deviation time points and the deviation amplitude are counted, the overall deviation degree is calculated, and the pressure difference deviation data is obtained.

[0024] See also Figure 3 , the specific steps for obtaining abnormal sound wave indicators are: S211: collecting acoustic wave data during the operation of the control valve through an acoustic sensor, continuously monitoring the frequency and intensity of the acoustic waves, and arranging them into time series data to generate an acoustic wave time series; The acoustic sensor is installed near the control valve to capture the acoustic wave data generated during operation in real time. The sampling frequency of the sensor should be set according to the normal vibration frequency range of the control valve. For example, if the valve vibration frequency is mainly concentrated in the range of 100Hz-5kHz, the sampling rate of the sensor should be at least 10kHz to ensure data integrity. The collected data includes the acoustic wave frequency and intensity and is stored in a time series format. The data storage time interval is set to 1 millisecond to ensure sufficient time resolution. For example, if 1000 data points are recorded in a certain period of time, a corresponding acoustic wave time series of 1 second in length can be obtained. The specific data format of this time series is usually a two-dimensional array. The first column records the timestamp, the second column records the sound wave frequency, and the third column records the sound wave intensity. After collecting the data, preprocessing is required. First, the data is denoised. For example, the sliding average method is used to eliminate instantaneous mutation values. Assuming that the sound wave frequency measured at a certain moment is 1200Hz, but its adjacent data points are all around 1100Hz, the average value of the five data points around it can be calculated and used to replace the abnormal point. Subsequently, the processed data is sorted into a time series to ensure that the timestamps are continuous and arranged in chronological order, and finally a complete sound wave time series is obtained.

[0025] S212: Based on the acoustic wave time series, analyze the frequency offset and intensity change using the formula: ; Calculate the acoustic wave characteristic deviation at each time point ,in, Representative The frequency of the sound wave at a point in time, Representative The intensity of the sound wave at a certain time point, and are the mean and standard deviation of the acoustic wave frequencies respectively, and are the mean and standard deviation of the acoustic wave intensities respectively; Analyze the frequency offset and intensity change at each time point. The frequency offset refers to the deviation of the frequency at a certain time point from the overall frequency mean, and the intensity change refers to the degree of deviation of the intensity at this time point from the overall mean. First, calculate the mean and standard deviation of the overall data. The calculation method is to calculate the mean of the frequencies at all time points to obtain and calculate the standard deviation of the frequencies at all time points to obtain Similarly, perform the same operation on the intensity data to obtain and Then calculate the standardized offset for each time point. The collected acoustic wave data is: Time point 1: frequency 1000 Hz, intensity 78 dB; Time point 2: frequency 1100 Hz, intensity 82 dB; Time point 3: frequency 1200 Hz, intensity 85 dB; Calculate the mean: ; ; Calculate the standard deviation: ; ; ; ; Calculate the acoustic wave feature deviation at each time point: For time point : ; ; For time point : ; For time point : ; The results show that among the three time points, the deviation degree of the acoustic wave characteristics at the first time point is 1.77, and that at the third time point is 1.69, both of which are higher than 0.11 at the second time point. This indicates that the degree of deviation of the acoustic wave frequency and intensity from the overall mean at the first and third time points is relatively large, while the data at the second time point is close to the overall mean and the deviation is small. This shows that at the first and third time points, abnormal fluctuations occurred in the acoustic wave characteristics of the control valve, while the acoustic state at the second time point was relatively stable.

[0026] S213: Identify abnormal points where the deviation exceeds the normal range based on the deviation degree of acoustic wave characteristics, screen abnormal acoustic wave data, and extract the corresponding key acoustic parameters to obtain abnormal acoustic wave indicators; Based on the deviation degree of acoustic wave characteristics, it is necessary to identify abnormal acoustic wave data. First, set the deviation degree threshold , and the setting of the threshold is based on historical data statistics or engineering experience values. For example, it can be set to , that is, points with a deviation degree exceeding 3 are regarded as abnormal points. In actual calculations, if the calculated value at a certain time point is greater than , then the data at that time point is marked as abnormal, and the corresponding acoustic parameters are recorded, including its acoustic wave frequency, intensity, and timestamp. For example, if the at a certain time point is

[0027] and the frequency is 1300 Hz and the intensity is 90 dB, then this point is marked as an abnormal point and stored in the abnormal data list. After summarizing all abnormal point data, extract its key parameters to obtain abnormal acoustic wave indicators. Figure 4 Please refer to S311: Analyze the differential pressure deviation data, collect the differential pressure values at each time point from the data source, and calculate the differential pressure change rate between adjacent time points; Collect the differential pressure values at each time point from the data source. During the collection process, select data points within a specific time range and ensure the integrity and continuity of the data. Set the data sampling interval, determine the pressure values at each time point, and fill in the missing data by interpolation. After the data collection is completed, sort the data by time to ensure continuous time points, and calculate the differential pressure change amount between adjacent time points. The calculation method is to take the difference between the differential pressure value at each time point and the differential pressure value at the previous time point to obtain the instantaneous differential pressure change amount at each time point. On this basis, calculate the differential pressure change rate, that is, divide the differential pressure change amount by the time interval between adjacent time points. If the differential pressure value at a certain time point is 5.3 kPa, the differential pressure value at the previous time point is 5.0 kPa, and the time interval between the two time points is 2 seconds, then the differential pressure change rate at this time point is kPa / s, and calculate the data at all time points in turn to generate the differential pressure change rate.

[0028] S312: Based on the pressure difference change rate, analyze the fluctuation frequency and fluctuation amplitude within a continuous time period, using the formula: ; Calculate the standard deviation of fluctuations , used to measure the fluctuation amplitude of the pressure difference value within a certain period of time and obtain the key time interval of pressure difference fluctuation, where, Representative The pressure difference value at a time point, represents the average value of the pressure difference at a given time point, Represents the total number of time points; First, determine the number of data points in the statistical time window. Assume that the time window contains 3 data points, which are kPa, kPa, kPa, calculate the average pressure difference during this time period, the formula is: ; Calculate the squared deviation for each data point and sum them: ; Calculate the standard deviation of the fluctuations: ; This result indicates that the pressure differential fluctuation within this time window is 0.2 kPa. The standard deviation measures the dispersion of the pressure differential values; a larger value indicates more severe pressure differential fluctuations within this time window. If this value exceeds the set fluctuation threshold (for example, 0.15 kPa), it indicates that the pressure differential fluctuation within this time period is abnormal, and this time window is marked as a critical pressure differential fluctuation time period.

[0029] S313: Based on the key time interval of pressure differential fluctuation, gradient analysis is performed on the pressure differential data within the range to determine whether there is an abnormal trend, and trend features are extracted to obtain the pressure differential fluctuation trend; According to the key time interval of pressure difference fluctuation, the pressure difference data within its range is subjected to gradient analysis, and the pressure difference gradient between adjacent time points is calculated, that is, the pressure difference change rate at each time point compared with the rate change value at the previous time point. The gradient calculation method is: ,in For the The rate of change of pressure difference at a time point, is the pressure difference change rate at the previous time point. If the pressure difference change rate at a certain time point is 0.12 kPa / s, and the pressure difference change rate at the previous time point is 0.09 kPa / s, then the gradient value at that time point is kPa / s², after calculating the gradient values at all time points, filter out the data points whose gradients exceed the preset gradient threshold as abnormal data points. The set gradient threshold is calculated based on historical data or experimental test results. For example, under normal operating conditions, the gradient usually ranges from 0.01 kPa / s² to 0.05 kPa / s². If the gradient of a certain data point exceeds 0.06 kPa / s², it is marked as an abnormal point, and a differential pressure fluctuation trend is generated.

[0030] Please refer to Figure 5 , and the specific steps for obtaining acoustic anomaly diagnosis data are as follows: S411: Based on the abnormal acoustic wave indicators, analyze the abnormal signal intensity and duration, determine the signal interval that conforms to the abnormal acoustic feature pattern, calculate the abnormal signal frequency within the interval, and use the formula: ; Calculate the proportion of abnormal signal distribution , where represents the th time period of the abnormal signal frequency, that is, the number of abnormal signals detected within this time period, represents the total number of measurements during the total operation period; Obtain and record the signal intensity at each time point. This signal intensity can be quantified by detecting the amplitude of the acoustic signal through an acoustic wave sensor. For example, set the signal amplitude range within a monitoring period to be to , and convert the amplitude signal into numerical data for storage. Calculate the intensity change rate between adjacent time points, that is, for two consecutive time points and , calculate their signal amplitude change rate . This calculation method can be used to quantify the signal intensity change. Further, count the signal duration, that is, accumulate the time periods when the signal exceeds a certain threshold to obtain the total duration of the abnormal signal. For example, if an abnormal signal lasts for 2 s and the threshold is set to 1.5 s, then this signal is identified as a long-term abnormal signal. Then, determine the signal interval that conforms to the abnormal acoustic feature pattern, and filter out the time periods that meet the amplitude feature threshold and the duration as the abnormal signal interval. Calculate the abnormal signal frequency within this interval, and use statistical methods to calculate the number of occurrences of abnormal signals per unit time. For example, if the duration of an interval is 10 s and 5 abnormal signals are detected, then its abnormal signal frequency is 0.5 Hz, obtain the proportion of abnormal signal distribution, and the measurement duration is the total duration during the entire monitoring period. This duration represents the time range of data collection during the entire measurement process. For example, set , and divide it into 6 time periods, and count the abnormal signal frequency within each time period , which represents the ratio of the number of abnormal signals detected in the time period to the length of the time period. If the measured abnormal signal frequency data is as follows: , , , , , ; , substitute the measured data into the calculation: ; ; Calculate the distribution ratio of abnormal signals , which represents the average frequency of abnormal signals during the entire measurement period, that is, approximately 0.3833 abnormal signals occur per second. This result can be used to further analyze the distribution characteristics of abnormal signals and, combined with other acoustic data, to determine the existence of control valve structural problems.

[0031] S412: Analyze the distribution characteristics of the abnormal acoustic signal based on the abnormal signal distribution ratio, combined with the abnormal signal strength and duration, determine the existing control valve structure problem, and obtain acoustic abnormality diagnosis data; Obtain abnormal signal strength statistics, including the peak amplitude of the abnormal signal and mean amplitude For example, if the amplitude of the abnormal signal ranges from 5dB to 20dB, its peak amplitude is 20dB and its mean amplitude is 12dB. Further obtain the statistical data of the signal duration, including the shortest duration of the abnormal signal. and maximum duration For example, if the duration of the detected abnormal signal is between 0.5s and 3s, then and , combined with the distribution ratio of abnormal signals, the data is trend analyzed to determine whether the abnormal signal shows periodic or sudden changes. For periodic changes, it is determined whether the signal recurs at fixed intervals. For example, if the abnormal signal is detected to appear once every 30 seconds, the signal is judged to be periodic. On the contrary, if the abnormal signal distribution is random, it is judged to be a sudden abnormal signal, and the acoustic abnormality diagnosis data is obtained.

[0032] See also Figure 6 , the steps for obtaining the current status of the control valve are as follows: S511: Based on the pressure difference fluctuation trend and acoustic anomaly diagnosis data, analyze the time series characteristics of each parameter, filter the characteristic data of the time corresponding interval, and obtain the time corresponding characteristic data; Collect the differential pressure fluctuation data of the control valve circuit, including the recorded values of the pressure sensor at different time points. The data is collected by the equipment monitoring system at fixed time intervals, for example, recorded once per second. The obtained data points form a pressure fluctuation curve. The acoustic anomaly diagnosis data comes from the acoustic sensor installed on the equipment. This sensor records the acoustic wave signals generated by the airflow vibration inside the valve. The signals are subjected to spectral analysis through fast Fourier transform (FFT) to extract the intensity information of different frequency components. Subsequently, for the collected differential pressure fluctuation data, the sliding window method is used to calculate its time series characteristics, such as eigenvalue features like mean, variance, skewness, kurtosis, etc. Among them, the size of the sliding window can be adjusted according to the valve operation cycle. For example, it is set as a 10-second window and slides 1 second each time. At the same time, feature extraction is performed on the acoustic signal, and the power spectral density of the signal in different frequency intervals is calculated to analyze its energy distribution in a specific frequency range. For example, the signal power from 100 Hz to 1 kHz can reflect the airflow disturbance inside the valve. Next, screen the feature data in the corresponding time interval, that is, align the differential pressure fluctuation data and the acoustic signal data in time. If an abnormal event occurs at time T, then the differential pressure fluctuation eigenvalues and acoustic eigenvalues 5 seconds before and after T need to be selected, and the timestamps of the two data sets are matched to ensure that both correspond to the same operating state, obtaining the time-corresponding feature data.

[0033] S512: Call the time-corresponding feature data to identify the current state of the control valve circuit. Compare with the feature data of known fault modes and use the formula: ; Calculate the matching degree with the known fault modes to obtain the current state fault similarity , where, represents the -dimensional time-corresponding feature data of the current state, represents the -dimensional eigenvalue corresponding to the known fault mode, represents the total number of feature dimensions; Read the fault mode data, such as valve jamming fault, seal leakage fault, air source fault, etc. Each fault mode includes a set of typical differential pressure fluctuation characteristics and acoustic eigenvalue features. For example, the valve jamming fault is manifested as an increase in the power of the acoustic signal at a specific frequency (300 Hz), while the seal leakage fault is manifested as an increase in the power of the high-frequency (above 5 kHz) signal. Subsequently, calculate the similarity between the current state and each fault mode. The typical eigenvalue features of a certain fault mode are: the average pressure fluctuation amplitude is 2.5 kPa, and the power spectral density of the acoustic signal at 300 Hz is 0.8 dB. The measured values of the current state are 2.0 kPa and 0.6 dB respectively. Then the characteristic ratios are: , ; Set , then the similarity calculation is as follows: ; This value indicates a relatively high degree of matching between the current state and this fault mode. By comparing the similarities calculated with different fault modes and selecting the mode with the highest similarity, the current state fault similarity is obtained.

[0034] S513: According to the current state fault similarity, compare it with the warning level of the fault mode, determine the warning level corresponding to the control valve path, and obtain the current state of the control valve path; According to the current state fault similarity, compare it with the warning level of the fault mode. First, set the warning level threshold. For example, if the current state fault similarity is greater than 0.8, it is determined as a serious fault; if it is between 0.5 and 0.8, it is determined as a minor fault; if it is less than 0.5, it is determined as a normal state. Assume that the current state fault similarity calculated above is 0.60125, which falls within the minor fault range. Therefore, it is judged that the warning level of the control valve path is a minor fault, and the current state of the control valve path is obtained.

[0035] A new energy high-stability control valve path intelligent detection system, the system includes: The differential pressure monitoring module collects the pressure sensor data at the inlet and outlet of the control valve, monitors the current inlet pressure and outlet pressure of the valve, calculates the pressure difference between the two, analyzes whether the difference exceeds the range of conventional operation data, and records the deviation in the data to obtain the differential pressure offset data; The acoustic wave analysis module collects the acoustic wave data during the operation of the control valve through an acoustic sensor, continuously monitors the acoustic wave frequency and intensity, performs a spectrogram analysis on the collected acoustic wave data, and identifies the abnormal points inconsistent with the conventional acoustic wave characteristics to obtain the abnormal acoustic wave index; The differential pressure fluctuation analysis module analyzes the differential pressure offset data, evaluates the frequency and amplitude of the differential pressure fluctuation within a continuous time period, identifies the key time intervals of the differential pressure fluctuation, and determines whether there is an abnormal differential pressure change trend to obtain the differential pressure fluctuation trend; The acoustic anomaly diagnosis module, based on the abnormal acoustic wave index, analyzes the abnormal signal intensity and duration, determines the signal interval that conforms to the abnormal acoustic characteristic mode, calculates the abnormal signal frequency within the interval and its distribution ratio during the operation process, and judges potential control valve structure problems to obtain the acoustic anomaly diagnosis data; The state evaluation module, based on the differential pressure fluctuation trend and the acoustic anomaly diagnosis data, identifies the current state of the control valve path, calculates the degree of matching with known fault modes, and judges the warning level corresponding to the current state to obtain the current state of the control valve path.

[0036] The above are only the preferred embodiments of the present invention and do not limit the present invention in other forms. Any person skilled in the relevant art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A new energy high-stability control valve circuit intelligent detection method, characterized in that, It includes the following steps: S1: Collect the pressure sensor data at the inlet and outlet of the control valve, monitor the current inlet pressure and outlet pressure of the valve, calculate the pressure difference between the two through continuous data acquisition, analyze whether the difference exceeds the range of normal operation data, record the deviation in the data, and obtain the differential pressure offset data; S2: Collect the acoustic wave data when the control valve is operating through an acoustic sensor, continuously monitor the acoustic wave frequency and intensity, perform spectrogram analysis on the collected acoustic wave data, identify the abnormal points inconsistent with the normal acoustic wave characteristics, and obtain the abnormal acoustic wave index; S3: Analyze the differential pressure offset data, calculate the differential pressure change rate, evaluate the frequency and amplitude of the differential pressure fluctuation within a continuous time period, identify the key time intervals of the differential pressure fluctuation, and perform gradient analysis on the data in the time intervals to determine whether there is an abnormal differential pressure change trend, and obtain the differential pressure fluctuation trend; S4: Based on the abnormal acoustic wave index, analyze the abnormal signal intensity and duration, determine the signal interval that conforms to the abnormal acoustic characteristic pattern, calculate the abnormal signal frequency within the interval and its distribution ratio during the operation process, judge the potential structural problems of the control valve, and obtain the acoustic anomaly diagnosis data.

2. The intelligent detection method for a new energy highly stable control valve circuit according to claim 1, characterized in that The differential pressure offset data includes the inlet pressure deviation, the outlet pressure deviation, and the pressure offset index. The abnormal acoustic wave index includes the frequency deviation value, the intensity deviation value, and the abnormal spectrum identifier. The differential pressure fluctuation trend includes the fluctuation rate index, the frequency fluctuation index, and the amplitude fluctuation index. The acoustic anomaly diagnosis data includes the acoustic signal intensity index, the acoustic duration index, and the acoustic frequency distribution.

3. The intelligent detection method for the new energy high-stability control valve circuit according to claim 1, characterized in that The specific steps for obtaining the differential pressure offset data are as follows: S111: Collect the pressure sensor data at the inlet and outlet of the control valve, continuously collect the inlet pressure value and the outlet pressure value within a certain period of time, record the corresponding time series, and then calculate the instantaneous pressure difference at each time point to generate an instantaneous pressure difference sequence; S112: Based on the instantaneous pressure difference sequence, perform normalization processing on the sequence, using the formula: ； Calculate the normalized pressure difference at each time point , and obtain the normalized pressure difference trend value, where represents the inlet pressure at the th time point, represents the outlet pressure at the th time point, and are the mean and standard deviation of the instantaneous pressure difference sequence respectively; S113: Call the normalized pressure difference trend value, judge whether it exceeds the normal operation range, screen the data that exceeds the normal range, and record the deviation in the data to obtain the differential pressure offset data.

4. The intelligent detection method for a new energy high-stability control valve circuit according to claim 1, wherein The specific steps for obtaining the abnormal acoustic wave index are as follows: S211: Collect the acoustic wave data when the control valve is operating through an acoustic sensor, continuously monitor the acoustic wave frequency and intensity, and organize them into time series data to generate an acoustic wave time series; S212: Based on the acoustic wave time series, analyze the frequency offset amount and the intensity change amount, using the formula: ； Calculate the deviation degree of the acoustic wave characteristics at each time point , where represents the acoustic wave frequency at the th time point, represents the acoustic wave intensity at the th time point, and are the mean and standard deviation of the acoustic wave frequency respectively, and are the mean and standard deviation of the acoustic wave intensity respectively; S213: According to the acoustic wave characteristic offset degree, identify the abnormal points whose offset exceeds the normal range, screen the abnormal acoustic wave data, and extract the corresponding key acoustic parameters to obtain the abnormal acoustic wave index.

5. The intelligent detection method for a new energy high-stability control valve circuit according to claim 1, wherein, The specific steps for obtaining the differential pressure fluctuation trend are as follows: S311: Analyze the differential pressure offset data, collect the differential pressure value at each time point from the data source, and calculate the differential pressure change rate between adjacent time points; S312: Based on the differential pressure change rate, analyze the fluctuation frequency and fluctuation amplitude within a continuous time period, using the formula: ； Calculate the standard deviation of the fluctuation amplitude , and obtain the key time interval of the differential pressure fluctuation, where represents the differential pressure value at the th time point, represents the average value of the differential pressure at the time points, represents the total number of time points; S313: Based on the critical time interval of the differential pressure fluctuation, perform gradient analysis on the differential pressure data within this range, determine whether there is an abnormal trend, extract trend features, and obtain the differential pressure fluctuation trend.

6. The intelligent detection method for a new energy high-stability control valve circuit according to claim 1, wherein The specific steps for obtaining the acoustic anomaly diagnosis data are as follows: S411: Based on the abnormal acoustic wave index, analyze the abnormal signal intensity and duration, determine the signal interval that conforms to the abnormal acoustic feature pattern, calculate the abnormal signal frequency within the interval, and use the formula: ； Calculate the proportion of abnormal signal distribution , where represents the frequency of abnormal signals in the th time period, represents the total number of measurements during the total operation period; S412: Based on the abnormal signal distribution ratio, combined with the abnormal signal intensity and duration, analyze the distribution characteristics of the acoustic anomaly signal, determine the existing control valve structure problems, and obtain the acoustic anomaly diagnosis data.

7. The intelligent detection method for a new energy high-stability control valve circuit according to claim 1, characterized in that The steps further include: S5: Based on the differential pressure fluctuation trend and the acoustic anomaly diagnosis data, analyze the data correlation, identify the current state of the control valve circuit, calculate the degree of matching with the known fault modes, and determine the warning level corresponding to the current state to obtain the current state of the control valve circuit; The current state of the control valve circuit includes the fault mode consistency and the warning level index.

8. The intelligent detection method for a new energy highly stable control valve circuit according to claim 7, wherein The specific steps for obtaining the current state of the control valve circuit are as follows: S511: Based on the differential pressure fluctuation trend and the acoustic anomaly diagnosis data, analyze the time series characteristics of each parameter, screen the characteristic data in the corresponding time interval, and obtain the time-corresponding characteristic data; S512: Call the time-corresponding characteristic data, identify the current state of the control valve circuit, compare with the characteristic data of the known fault modes, and use the formula: ； Calculate the matching degree with known fault modes to obtain the fault similarity of the current state , where represents the -dimensional time-corresponding feature data of the current state, represents the -dimensional eigenvalue corresponding to the known fault mode, represents the total number of feature dimensions; S513: According to the similarity of the current state to the fault, compare with the warning level of the fault mode, determine the warning level corresponding to the control valve circuit, and obtain the current state of the control valve circuit.

9. An intelligent detection system for a new energy high-stability control valve circuit, characterized in that, Executed according to the new energy high-stability control valve circuit intelligent detection method according to any one of claims 1-8, the system includes: The differential pressure monitoring module collects the pressure sensor data at the inlet and outlet of the control valve, monitors the current inlet pressure and outlet pressure of the valve, calculates the pressure difference between the two, analyzes whether the difference exceeds the range of normal operation data, and records the deviation in the data to obtain the differential pressure offset data; The acoustic wave analysis module collects the acoustic wave data during the operation of the control valve through an acoustic sensor, continuously monitors the acoustic wave frequency and intensity, performs spectrogram analysis on the collected acoustic wave data, and identifies the abnormal points that are inconsistent with the normal acoustic wave characteristics to obtain the abnormal acoustic wave index; The differential pressure fluctuation analysis module analyzes the differential pressure offset data, evaluates the frequency and amplitude of the differential pressure fluctuation within a continuous time period, identifies the critical time interval of the differential pressure fluctuation, and determines whether there is an abnormal differential pressure change trend to obtain the differential pressure fluctuation trend; The acoustic anomaly diagnosis module, based on the abnormal acoustic wave index, analyzes the abnormal signal intensity and duration, determines the signal interval that conforms to the abnormal acoustic feature pattern, calculates the abnormal signal frequency within the interval and its distribution ratio during the operation process, and determines the potential control valve structure problems to obtain the acoustic anomaly diagnosis data; Based on the differential pressure fluctuation trend and the acoustic anomaly diagnosis data, the status evaluation module identifies the current status of the control valve path, calculates the degree of matching with known fault modes, and determines the warning level corresponding to the current status, thereby obtaining the current status of the control valve path.

Images