Fault diagnosis method of intelligent mass flow controller

Abstract

The invention discloses a fault diagnosis method for an intelligent mass flow controller, and particularly relates to the technical field of fault diagnosis. Flow information controlled by a mass flow controller, running state information of a pump and opening feedback information of a valve are obtained, a flow peak variation coefficient, a pump running stability coefficient and a valve opening feedback coefficient are obtained, a fault diagnosis evaluation model is constructed, and a fault diagnosis screening index is generated; by analyzing the synergistic effect and mutual influence among multiple devices, similar symptoms caused by faults of the flow controller and faults of other devices can be effectively distinguished, when the fault type of the mass flow controller is not a single-point fault, a correspondingly generated fault diagnosis and discrimination index is obtained, and the standard deviation of the linkage interference concealment ratio is calculated. And comparing the standard deviation of the linkage interference covering ratio with a preset standard deviation threshold value, evaluating the linkage interference covering degree of the fault signal of the mass flow controller, and performing dynamic early warning according to the linkage interference covering degree.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis, and more specifically, to a fault diagnosis method for an intelligent mass flow controller. Background Art

[0002] In intelligent manufacturing and automated production lines, flow controllers, as one of the core components, are responsible for accurately controlling the flow of fluids to ensure the stability and accuracy of the production process. However, with the increasing complexity and automation of production lines, flow controller fault diagnosis faces more and more challenges. Flow controllers usually work closely with other devices (such as pumps, valves, sensors, robots, etc.), and their failures may affect each other, resulting in the superposition of equipment fault signals and complex abnormal performance; the signal of flow controller failure may produce similar abnormalities as the failure of other equipment. For example, a flow controller failure may cause flow fluctuations, while a pump or valve failure may also cause similar flow changes, resulting in the diagnostic system being unable to accurately distinguish the source of the fault. The mutual interference and masking of equipment fault signals further increase the difficulty of fault diagnosis. Secondly, the failure of multiple devices may have the characteristics of dynamic propagation. The failure of the flow controller may cause pressure changes, temperature fluctuations, etc., which may affect the performance of other devices, resulting in the superposition and confusion of fault signals, increasing the complexity of the diagnostic process. Summary of the invention

[0003] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a fault diagnosis method of an intelligent mass flow controller to solve the problems raised in the above-mentioned background technology.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] The fault diagnosis method of the intelligent mass flow controller includes the following steps:

[0006] Step S1, obtaining flow information controlled by a mass flow controller, and obtaining a flow peak variation coefficient according to the flow information;

[0007] Step S2, obtaining the operating status information of the pump, and obtaining the pump operating stability coefficient according to the operating status information;

[0008] Step S3, obtaining valve opening feedback information, and obtaining a valve opening feedback coefficient according to the valve opening feedback information;

[0009] Step S4, constructing a fault diagnosis evaluation model according to the flow peak variation coefficient, the pump operation stability coefficient, and the valve opening feedback coefficient, generating a fault diagnosis identification index, and evaluating whether the fault type of the mass flow controller is a single point fault;

[0010] Step S5, when the fault type of the mass flow controller is not a single point fault, obtain the corresponding generated fault diagnosis identification index, perform a ratio operation on it and a preset fault diagnosis identification index threshold, evaluate the linkage interference masking degree of the mass flow controller fault signal, and issue an early warning based on the linkage interference masking degree.

[0011] In a preferred embodiment, by acquiring the flow information controlled by the mass flow controller, analyzing the instantaneous abnormal deviation between the target flow value of the mass flow controller and the actual measured flow value, and obtaining the flow peak variation coefficient, the instantaneous abnormal deviation degree between the target flow value of the mass flow controller and the actual measured flow value is measured;

[0012] The logic for obtaining the peak flow variation coefficient is as follows:

[0013] The target flow value of the mass flow controller and the measured actual flow value are obtained within a preset time period T; the instantaneous flow deviation ΔLQ(t) is calculated, and the expression is as follows: ΔLQ(t) = |LQ-LQC(t)|, where LQ represents the target flow value, LQC(t) represents the actual flow value measured at the tth moment, t = {1, 2, ..., T}, T is a positive integer, and ΔLQ(t) represents the instantaneous flow deviation calculated at the tth moment; the mean value μΔLQ of the instantaneous flow deviation is calculated, and the expression is as follows: Calculate the standard deviation of the instantaneous flow rate deviation Calculate the kurtosis Fd of the instantaneous deviation of flow rate, the expression is as follows

[0014] Perform fast Fourier transform on the instantaneous flow deviation to convert the time domain signal of the instantaneous flow deviation into the frequency domain. The expression is as follows: ΔLQ(f) = FFT(ΔLQ(t)), where ΔLQ(f) represents the frequency domain signal of the instantaneous flow deviation, and FFT() represents fast Fourier transform; calculate the low-frequency energy Elow of the instantaneous flow deviation, and the expression is as follows Where flow represents the preset low-frequency band; the high-frequency energy Ehigh of the instantaneous deviation of flow is calculated as follows: Calculate the energy distribution ratio Rf, the expression is as follows

[0015] Calculate the peak flow rate variation coefficient Llfy, expressed as follows

[0016] In a preferred embodiment, the stability of the pump operation state is measured by acquiring the operation state information of the pump, analyzing the stability of the pump operation state, and acquiring the pump operation stability coefficient;

[0017] The logic for obtaining the pump operation stability coefficient is as follows:

[0018] Acquiring the operating status information of the pump within a preset time period T, wherein the operating status information includes but is not limited to pressure, speed, and vibration;

[0019] The running status information set is marked as X = {x i}={x1,x2,...,x I}, where x i represents the running status information data obtained at the i-th moment, I represents the total number of running status information data points, i={1,2,...,I}, I is a positive integer;

[0020] Constructing the autocorrelation function of the running status information data The function expression is as follows in represents the time delay, when the autocorrelation function When the output value of reaches the preset minimum value for the first time, the optimal time delay is obtained.

[0021] Calculate the running status information data point x i Its nearest neighbor x j The pseudo nearest neighbor ratio RPNN under the embedding dimension m is expressed as follows When the pseudo nearest neighbor ratio RPNN converges to the preset minimum value, the optimal embedding dimension m is obtained;

[0022] According to the optimal time delay and the optimal embedding dimension m to construct the phase space trajectory generate indivual is a vector;

[0023] exist dimensional phase space, and divide each dimension into Ngrid intervals to obtain Nwg grids. i In grid g, the nearest neighbor point x j If the state is in grid h, then grid g to grid h is marked as a state transition, and the transition probability P from grid g to grid h is calculated. gh , the expression is as follows Where C1 represents the number of state transitions from grid g to grid h, and C2 represents the total number of state transitions;

[0024] Calculate the pump operation stability coefficient Byxw, the expression is as follows Wherein g={1,2,...,G}, G is a positive integer, h={1,2,...,H}, H is a positive integer.

[0025] In a preferred embodiment, by acquiring valve opening feedback information, analyzing the matching degree between the valve opening control signal and the actual feedback opening, and acquiring the valve opening feedback coefficient, the matching degree between the valve opening control signal and the actual feedback opening is measured;

[0026] The logic for obtaining the valve opening feedback coefficient is as follows:

[0027] Get the valve control signal KZ within the preset time period T a and the actual feedback opening KD a ;

[0028] Calculate the control signal KZ of the valve a and the actual feedback opening KD a The correlation coefficient ρKZD is expressed as follows Among them KZ a Indicates the a-th control signal of the valve, KD a Indicates the ath actual feedback opening of the valve, a={1,2,...,A}, A is a positive integer, Represents the mean value of the control signal, and the calculation expression is as follows Represents the mean value of the actual feedback opening, and the calculation expression is as follows

[0029] Calculate the valve opening feedback coefficient Fmkd, the expression is as follows Fmkd = 1-|ρKZD|.

[0030] In a preferred embodiment, a fault diagnosis evaluation model is constructed based on the flow peak variation coefficient, the pump operation stability coefficient, and the valve opening feedback coefficient to generate a fault diagnosis discrimination index Gzdz. The model is based on the following formula: Wherein w1, w2, and w3 represent the preset proportional coefficients of the flow peak variation coefficient, the pump operation stability coefficient, and the valve opening feedback coefficient, respectively, and w1, w2, and w3 are all greater than 0.

[0031] In a preferred embodiment, the fault diagnosis discrimination index is compared with a preset fault diagnosis discrimination index threshold to evaluate whether the fault type of the mass flow controller is a single point fault, as follows:

[0032] If the fault diagnosis discrimination index is greater than the fault diagnosis discrimination index threshold, a single point fault warning signal of the mass flow controller is generated;

[0033] If the fault diagnosis discrimination index is less than or equal to the fault diagnosis discrimination index threshold, a linkage interference masking signal is generated.

[0034] In a preferred embodiment, when the fault type of the mass flow controller is not a single point fault, the corresponding generated fault diagnosis discrimination index is obtained, and the ratio operation is performed on it and the preset fault diagnosis discrimination index threshold to obtain the linkage interference masking ratio, and the linkage interference masking ratio calculated by generating linkage interference masking signals at different times is added to the linkage interference masking set, and the mean value μLD of the linkage interference masking ratio in the linkage interference masking set is calculated, and the expression is as follows: where grb b represents the bth linkage interference masking ratio in the linkage interference masking set, b={1,2,...,B}, B is a positive integer; calculate the standard deviation σLZ of the linkage interference masking ratio in the linkage interference masking set, the expression is as follows

[0035] In a preferred embodiment, the standard deviation of the linkage interference masking ratio is compared with a preset standard deviation threshold to evaluate the linkage interference masking degree of the mass flow controller fault signal. If the standard deviation of the linkage interference masking ratio is greater than the standard deviation threshold, a first-level linkage interference masking degree is generated; if the standard deviation of the linkage interference masking ratio is less than or equal to the standard deviation threshold, a second-level linkage interference masking degree is generated;

[0036] When the first-level linkage interference masking level is generated, respond to the emergency warning signal, and when the second-level linkage interference masking level is generated, respond to the conventional warning signal.

[0037] Technical effects and advantages of the present invention:

[0038] 1. The present invention obtains the flow information controlled by the mass flow controller, the operating status information of the pump, the opening feedback information of the valve, and obtains the flow peak variation coefficient, the pump operation stability coefficient, and the valve opening feedback coefficient to construct a fault diagnosis evaluation model, generate a fault diagnosis identification index, and can more comprehensively and accurately evaluate the fault state of the flow controller. By analyzing the synergy and mutual influence between multiple devices, it can effectively distinguish the similar symptoms caused by the flow controller failure and other equipment failures, and avoid misdiagnosis caused by similar fault manifestations between devices in traditional fault diagnosis methods. When the fault type of the mass flow controller is not a single point fault, the corresponding generated fault diagnosis identification index is obtained, and the ratio is calculated with the preset fault diagnosis identification index threshold to obtain the linkage interference masking ratio, and the standard deviation of the linkage interference masking ratio is compared with the preset standard deviation threshold, and the linkage interference masking degree of the mass flow controller fault signal is evaluated. Dynamic early warning is performed according to the linkage interference masking degree, so as to realize efficient monitoring and accurate diagnosis of complex equipment faults in the production line, improve the accuracy of fault detection, identify and handle faults in advance through an intelligent early warning mechanism, significantly improve the stability and efficiency of the production line, effectively reduce the downtime caused by equipment failure, optimize the production process, and reduce operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0040] Figure 1 The figure is a flow chart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] Example: Figure 1 A fault diagnosis method for an intelligent mass flow controller of the present invention is provided, comprising the following steps:

[0043] Step S1, obtaining flow information controlled by a mass flow controller, and obtaining a flow peak variation coefficient according to the flow information;

[0044] Step S2, obtaining the operating status information of the pump, and obtaining the pump operating stability coefficient according to the operating status information;

[0045] Step S3, obtaining valve opening feedback information, and obtaining a valve opening feedback coefficient according to the valve opening feedback information;

[0046] Step S4, constructing a fault diagnosis evaluation model according to the flow peak variation coefficient, the pump operation stability coefficient, and the valve opening feedback coefficient, generating a fault diagnosis identification index, and evaluating whether the fault type of the mass flow controller is a single point fault;

[0047] Step S5, when the fault type of the mass flow controller is not a single point fault, obtain the corresponding generated fault diagnosis identification index, perform a ratio operation on the fault diagnosis identification index with a preset fault diagnosis identification index threshold, evaluate the linkage interference masking degree of the mass flow controller fault signal, and issue an early warning according to the linkage interference masking degree;

[0048] Step S1, obtaining flow information controlled by a mass flow controller, and obtaining a flow peak variation coefficient according to the flow information;

[0049] The peak flow variation coefficient is the instantaneous abnormal deviation between the target flow value and the actual measured flow value of the mass flow controller. Its core is to capture the abnormal state of the mass flow controller during operation. By calculating the peak flow variation coefficient, the abnormal behavior of the flow controller can be quickly evaluated and the source characteristics of the fault signal can be preliminarily identified. When the peak flow variation coefficient is larger, it means that the abnormal fluctuation of the flow controller is significant, which may be caused by a single point failure of its internal components (such as sensors, actuators or signal processing modules). It is usually not directly related to other equipment (such as pumps and valves) and has independence. If the peak flow variation coefficient is small, the overall system shows abnormal flow fluctuations, which is more likely to be an indirect effect caused by the linkage failure of other equipment (pumps or valves). By calculating the peak flow variation coefficient, it is determined whether the flow controller fails independently, so that the faulty equipment can be quickly isolated, the accuracy of fault diagnosis can be improved, and the time for troubleshooting can be significantly shortened, which helps to maintain the efficient operation of intelligent manufacturing and automated production lines.

[0050] Therefore, by acquiring the flow information controlled by the mass flow controller, analyzing the instantaneous abnormal deviation between the target flow value of the mass flow controller and the actual measured flow value, and obtaining the flow peak value variation coefficient, the instantaneous abnormal deviation degree between the target flow value of the mass flow controller and the actual measured flow value is measured;

[0051] The logic for obtaining the peak flow variation coefficient is as follows:

[0052] The target flow value of the mass flow controller and the measured actual flow value are obtained within a preset time period T; the instantaneous flow deviation ΔLQ(t) is calculated, and the expression is as follows: ΔLQ(t) = |LQ-LQC(t)|, where LQ represents the target flow value, LQC(t) represents the actual flow value measured at the tth moment, t = {1, 2, ..., T}, T is a positive integer, and ΔLQ(t) represents the instantaneous flow deviation calculated at the tth moment; the mean value μΔLQ of the instantaneous flow deviation is calculated, and the expression is as follows: Calculate the standard deviation of the instantaneous flow rate deviation Calculate the kurtosis Fd of the instantaneous deviation of flow rate, the expression is as follows

[0053] Perform fast Fourier transform on the instantaneous flow deviation to convert the time domain signal of the instantaneous flow deviation into the frequency domain. The expression is as follows: ΔLQ(f) = FFT(ΔLQ(t)), where ΔLQ(f) represents the frequency domain signal of the instantaneous flow deviation, and FFT() represents fast Fourier transform; calculate the low-frequency energy Elow of the instantaneous flow deviation, and the expression is as follows Where flow represents the preset low-frequency band; the high-frequency energy Ehigh of the instantaneous deviation of flow is calculated as follows: Calculate the energy distribution ratio Rf, the expression is as follows

[0054] Calculate the peak flow rate variation coefficient Llfy, expressed as follows

[0055] Step S2, obtaining the operating status information of the pump, and obtaining the pump operating stability coefficient according to the operating status information;

[0056] The pump operation stability coefficient is used to quantify the stability of the pump's operating state, especially when working in conjunction with a mass flow controller. By analyzing the pump's operating state data, the pump's potential impact on the mass flow controller's abnormality is evaluated. The pump operation stability coefficient is based on the pump's operating state information, such as pressure fluctuations, speed fluctuations, and vibration amplitude. When the pump operation stability coefficient is high, it means that the pump's operating state is stable and may not significantly interfere with flow abnormalities; when the pump operation stability coefficient is low, it indicates that the pump may be in an abnormal state, which may cause a coordinated failure of the mass flow controller, thereby further concealing the authenticity of the mass flow controller. By calculating the pump operation stability coefficient, the pump's potential interference with the mass flow controller and the risk of coordinated failure can be effectively evaluated;

[0057] Therefore, by acquiring the operating status information of the pump, analyzing the stability of the pump operating status, and obtaining the pump operating stability coefficient, the stability of the pump operating status is measured;

[0058] The logic for obtaining the pump operation stability coefficient is as follows:

[0059] Acquiring the operating status information of the pump within a preset time period T, wherein the operating status information includes but is not limited to pressure, speed, and vibration;

[0060] The running status information set is marked as X = {x i}={x1,x2,...,x I}, where x i represents the running status information data obtained at the i-th moment, I represents the total number of running status information data points, i={1,2,...,I}, I is a positive integer;

[0061] Constructing the autocorrelation function of the running status information data The function expression is as follows in represents the time delay, when the autocorrelation function When the output value of reaches the preset minimum value for the first time, the optimal time delay is obtained.

[0062] Calculate the running status information data point x i Its nearest neighbor xj The pseudo nearest neighbor ratio RPNN under the embedding dimension m is expressed as follows When the pseudo nearest neighbor ratio RPNN converges to the preset minimum value, the optimal embedding dimension m is obtained;

[0063] It should be noted that the running status information data point x i The nearest neighbor point x j It refers to the running status information data point with the smallest Euclidean distance. Euclidean distance is a common distance calculation method and will not be described here.

[0064] According to the optimal time delay and the optimal embedding dimension m to construct the phase space trajectory generate indivual is a vector;

[0065] exist dimensional phase space, and divide each dimension into Ngrid intervals to obtain Nwg grids. i In grid g, the nearest neighbor point x j If the state is in grid h, then grid g to grid h is marked as a state transition, and the transition probability P from grid g to grid h is calculated. gh , the expression is as follows Where C1 represents the number of state transitions from grid g to grid h, and C2 represents the total number of state transitions;

[0066] Calculate the pump operation stability coefficient Byxw, the expression is as follows Where g={1,2,...,G}, G is a positive integer, h={1,2,...,H}, H is a positive integer;

[0067] Step S3, obtaining valve opening feedback information, and obtaining a valve opening feedback coefficient according to the valve opening feedback information;

[0068] The valve opening feedback coefficient is used to measure the matching degree between the valve opening control signal and the actual feedback opening. It reflects whether the actual opening state of the valve can accurately respond to the expected control target after executing the control instruction. By calculating the valve opening feedback coefficient, the potential impact of the valve on the abnormality of the mass flow controller is evaluated. When the valve opening feedback coefficient is high, it means that the matching degree between the valve opening control signal and the actual feedback opening is worse, which may cause the coordinated failure of the mass flow controller and further conceal the authenticity of the mass flow controller; when the valve opening feedback coefficient is low, it means that the valve opening change is highly consistent with the control signal, and the valve can accurately execute the set opening to ensure the flow control accuracy, thereby reducing the potential impact on the mass flow controller and reducing the possibility of coordinated failure;

[0069] Therefore, by obtaining the valve opening feedback information, analyzing the matching degree between the valve opening control signal and the actual feedback opening, and obtaining the valve opening feedback coefficient, the matching degree between the valve opening control signal and the actual feedback opening is measured;

[0070] The logic for obtaining the valve opening feedback coefficient is as follows:

[0071] Get the valve control signal KZ within the preset time period T a and the actual feedback opening KD a ;

[0072] The control signal refers to the desired opening of the valve;

[0073] Calculate the control signal KZ of the valve a and the actual feedback opening KD a The correlation coefficient ρKZD is expressed as follows Among them KZ a Indicates the a-th control signal of the valve, KD a Indicates the ath actual feedback opening of the valve, a={1,2,...,A}, A is a positive integer, Represents the mean value of the control signal, and the calculation expression is as follows Represents the mean value of the actual feedback opening, and the calculation expression is as follows

[0074] Calculate the valve opening feedback coefficient Fmkd, the expression is as follows Fmkd = 1-|ρKZD|;

[0075] Step S4, constructing a fault diagnosis evaluation model according to the flow peak variation coefficient, the pump operation stability coefficient, and the valve opening feedback coefficient, generating a fault diagnosis identification index, and evaluating whether the fault type of the mass flow controller is a single point fault;

[0076] A fault diagnosis evaluation model is constructed based on the flow peak variation coefficient, pump operation stability coefficient, and valve opening feedback coefficient to generate the fault diagnosis identification index Gzdz. The model is based on the following formula: Wherein w1, w2, and w3 represent the preset proportional coefficients of the flow peak variation coefficient, the pump operation stability coefficient, and the valve opening feedback coefficient, respectively, and w1, w2, and w3 are all greater than 0;

[0077] It should be noted that before constructing the fault diagnosis and evaluation model, it is necessary to ensure that the flow peak variation coefficient, pump operation stability coefficient, and valve opening feedback coefficient are all normalized. Common normalization methods include Min-Max normalization and Z-Score standardization. w1, w2, and w3 are set according to actual conditions. For example, the expert empowerment method is adopted, that is, experts in related fields are invited to determine the preset proportional coefficients of various indicators through professional opinion surveys and comprehensive evaluations.

[0078] It can be seen from the above calculation expression that the larger the flow peak variation coefficient, the larger the pump operation stability coefficient, and the smaller the valve opening feedback coefficient, the larger the fault diagnosis identification index, indicating that the probability that the mass flow controller's fault type is a single-point fault is greater; conversely, the smaller the flow peak variation coefficient, the smaller the pump operation stability coefficient, and the larger the valve opening feedback coefficient, the smaller the fault diagnosis identification index, indicating that the probability that the mass flow controller's fault type is a single-point fault is smaller;

[0079] The fault diagnosis discrimination index is compared with a preset fault diagnosis discrimination index threshold to evaluate whether the fault type of the mass flow controller is a single point fault, as follows:

[0080] If the fault diagnosis discrimination index is greater than the fault diagnosis discrimination index threshold, it indicates that the fault type of the mass flow controller is a single point fault, and a single point fault warning signal of the mass flow controller is generated;

[0081] If the fault diagnosis discrimination index is less than or equal to the fault diagnosis discrimination index threshold, it indicates that the fault type of the mass flow controller is not a single point fault, and it may be due to the coordinated fault of the mass flow controller caused by the pump and the valve, generating a linkage interference masking signal;

[0082] Step S5, when the fault type of the mass flow controller is not a single point fault, obtain the corresponding generated fault diagnosis identification index, perform a ratio operation on the fault diagnosis identification index with a preset fault diagnosis identification index threshold, evaluate the linkage interference masking degree of the mass flow controller fault signal, and perform a fault warning according to the linkage interference masking degree;

[0083] When the fault type of the mass flow controller is not a single point fault, the corresponding generated fault diagnosis discrimination index is obtained, and the ratio operation is performed on it and the preset fault diagnosis discrimination index threshold to obtain the linkage interference masking ratio. The linkage interference masking ratios calculated by generating linkage interference masking signals at different times are added to the linkage interference masking set, and the mean value μLD of the linkage interference masking ratios in the linkage interference masking set is calculated. The expression is as follows: where grb brepresents the bth linkage interference masking ratio in the linkage interference masking set, b={1,2,...,B}, B is a positive integer; calculate the standard deviation σLZ of the linkage interference masking ratio in the linkage interference masking set, the expression is as follows

[0084] The standard deviation of the linkage interference masking ratio is compared with a preset standard deviation threshold to evaluate the linkage interference masking degree of the mass flow controller fault signal. If the standard deviation of the linkage interference masking ratio is greater than the standard deviation threshold, it indicates that the linkage interference masking degree of the mass flow controller fault signal is extremely large, and a first-level linkage interference masking degree is generated; if the standard deviation of the linkage interference masking ratio is less than or equal to the standard deviation threshold, it indicates that the linkage interference masking degree of the mass flow controller fault signal is small, and a second-level linkage interference masking degree is generated;

[0085] When the first-level linkage interference masking level is generated, it responds to the emergency warning signal, and when the second-level linkage interference masking level is generated, it responds to the conventional warning signal;

[0086] It should be noted that the warning level of emergency warning signals is greater than that of regular warning signals;

[0087] The present invention acquires the flow information controlled by the mass flow controller, the operation status information of the pump, the opening feedback information of the valve, and acquires the flow peak variation coefficient, the pump operation stability coefficient, and the valve opening feedback coefficient to construct a fault diagnosis evaluation model, generate a fault diagnosis identification index, and can more comprehensively and accurately evaluate the fault state of the flow controller. By analyzing the synergy and mutual influence between multiple devices, it can effectively distinguish the similar symptoms caused by the flow controller failure and other equipment failures, and avoid misdiagnosis caused by similar fault manifestations between devices in the traditional fault diagnosis method. When the fault type of the mass flow controller is not a single point fault, the corresponding generated fault diagnosis identification index is acquired, and the ratio operation is performed on the fault diagnosis identification index with a preset fault diagnosis identification index threshold to obtain a linkage interference masking ratio, and the standard deviation of the linkage interference masking ratio is compared with the preset standard deviation threshold, and the linkage interference masking degree of the mass flow controller fault signal is evaluated. Dynamic early warning is performed according to the linkage interference masking degree, so as to realize efficient monitoring and accurate diagnosis of complex equipment failures in the production line, improve the accuracy of fault detection, identify and handle faults in advance through an intelligent early warning mechanism, significantly improve the stability and efficiency of the production line, effectively reduce the downtime caused by equipment failure, optimize the production process, and reduce operating costs.

[0088] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0089] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0090] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A fault diagnosis method for an intelligent mass flow controller, characterized in that: The steps include: Step S1, obtaining flow information controlled by a mass flow controller, and obtaining a flow peak variation coefficient according to the flow information; Step S2, obtaining the operating status information of the pump, and obtaining the pump operating stability coefficient according to the operating status information; Step S3, obtaining valve opening feedback information, and obtaining a valve opening feedback coefficient according to the valve opening feedback information; Step S4, constructing a fault diagnosis evaluation model according to the flow peak variation coefficient, the pump operation stability coefficient, and the valve opening feedback coefficient, generating a fault diagnosis identification index, and evaluating whether the fault type of the mass flow controller is a single point fault; Step S5, when the fault type of the mass flow controller is not a single point fault, obtain the corresponding generated fault diagnosis identification index, perform a ratio operation on it and a preset fault diagnosis identification index threshold, evaluate the linkage interference masking degree of the mass flow controller fault signal, and issue an early warning based on the linkage interference masking degree.

2. The fault diagnosis method of the intelligent mass flow controller according to claim 1 is characterized in that: By acquiring the flow information controlled by the mass flow controller, analyzing the instantaneous abnormal deviation between the target flow value of the mass flow controller and the actual measured flow value, and obtaining the flow peak variation coefficient, the instantaneous abnormal deviation degree between the target flow value of the mass flow controller and the actual measured flow value is measured; The logic for obtaining the peak flow variation coefficient is as follows: The target flow value of the mass flow controller and the measured actual flow value are obtained within a preset time period T; the instantaneous flow deviation ΔLQ(t) is calculated, and the expression is as follows: ΔLQ(t) = |LQ-LQC(t)|, where LQ represents the target flow value, LQC(t) represents the actual flow value measured at the tth moment, t = {1, 2, ..., T}, T is a positive integer, and ΔLQ(t) represents the instantaneous flow deviation calculated at the tth moment; the mean value μΔLQ of the instantaneous flow deviation is calculated, and the expression is as follows: Calculate the standard deviation of the instantaneous flow rate deviation Calculate the kurtosis Fd of the instantaneous deviation of flow rate, the expression is as follows Perform fast Fourier transform on the instantaneous flow deviation to convert the time domain signal of the instantaneous flow deviation into the frequency domain. The expression is as follows: ΔLQ(f)=FFT(ΔLQ(t)), where ΔLQ(f) represents the frequency domain signal of the instantaneous flow deviation, and FFT() represents fast Fourier transform; calculate the low-frequency energy Elow of the instantaneous flow deviation, and the expression is as follows Where flow represents the preset low-frequency band; the high-frequency energy Ehigh of the instantaneous deviation of flow is calculated as follows: Calculate the energy distribution ratio Rf, the expression is as follows Calculate the peak flow rate variation coefficient Llfy, expressed as follows 3. The fault diagnosis method of the intelligent mass flow controller according to claim 1 is characterized in that: By acquiring the operating status information of the pump, analyzing the stability of the pump operating status, and obtaining the pump operating stability coefficient, the stability of the pump operating status is measured; The logic for obtaining the pump operation stability coefficient is as follows: Acquiring the operating status information of the pump within a preset time period T, wherein the operating status information includes but is not limited to pressure, speed, and vibration; The running status information set is marked as X = {x i }={x1,x2,...,x I }, where x i represents the running status information data obtained at the i-th moment, I represents the total number of running status information data points, i={1,2,...,I}, I is a positive integer; Constructing the autocorrelation function of the running status information data The function expression is as follows in represents the time delay, when the autocorrelation function When the output value of reaches the preset minimum value for the first time, the optimal time delay is obtained. Calculate the running status information data point x i Its nearest neighbor x j The pseudo nearest neighbor ratio RPNN under the embedding dimension m is expressed as follows When the pseudo nearest neighbor ratio RPNN converges to the preset minimum value, the optimal embedding dimension m is obtained; According to the optimal time delay and the optimal embedding dimension m to construct the phase space trajectory generate m is a vector; Grid division is performed in the m-dimensional phase space, and each dimension is divided into Ngrid intervals to obtain Nwg grids. If the running status information data point x i In grid g, the nearest neighbor point x j If the state is in grid h, then grid g to grid h is marked as a state transition, and the transition probability P from grid g to grid h is calculated. gh , the expression is as follows Where C1 represents the number of state transitions from grid g to grid h, and C2 represents the total number of state transitions; Calculate the pump operation stability coefficient Byxw, the expression is as follows Wherein g={1,2,...,G}, G is a positive integer, h={1,2,...,H}, H is a positive integer.

4. The fault diagnosis method of the intelligent mass flow controller according to claim 1, characterized in that: By obtaining the valve opening feedback information, analyzing the matching degree between the valve opening control signal and the actual feedback opening, and obtaining the valve opening feedback coefficient, the matching degree between the valve opening control signal and the actual feedback opening is measured; The logic for obtaining the valve opening feedback coefficient is as follows: Get the valve control signal KZ within the preset time period T a and the actual feedback opening KD a ; Calculate the control signal KZ of the valve a and the actual feedback opening KD a The correlation coefficient ρKZD is expressed as follows Among them KZ a Indicates the a-th control signal of the valve, KD a Indicates the ath actual feedback opening of the valve, a={1,2,...,A}, A is a positive integer, Represents the mean value of the control signal, and the calculation expression is as follows Represents the mean value of the actual feedback opening, and the calculation expression is as follows Calculate the valve opening feedback coefficient Fmkd, the expression is as follows Fmkd = 1-|ρKZD|.

5. The fault diagnosis method of the intelligent mass flow controller according to claim 1 is characterized in that: A fault diagnosis evaluation model is constructed based on the flow peak variation coefficient, pump operation stability coefficient, and valve opening feedback coefficient to generate the fault diagnosis identification index Gzdz. The model is based on the following formula: Wherein w1, w2, and w3 represent the preset proportional coefficients of the flow peak variation coefficient, the pump operation stability coefficient, and the valve opening feedback coefficient, respectively, and w1, w2, and w3 are all greater than 0.

6. The fault diagnosis method of the intelligent mass flow controller according to claim 5 is characterized in that: The fault diagnosis discrimination index is compared with a preset fault diagnosis discrimination index threshold to evaluate whether the fault type of the mass flow controller is a single point fault, as follows: If the fault diagnosis discrimination index is greater than the fault diagnosis discrimination index threshold, a single point fault warning signal of the mass flow controller is generated; If the fault diagnosis discrimination index is less than or equal to the fault diagnosis discrimination index threshold, a linkage interference masking signal is generated.

7. The fault diagnosis method of the intelligent mass flow controller according to claim 6 is characterized in that: When the fault type of the mass flow controller is not a single point fault, the corresponding generated fault diagnosis discrimination index is obtained, and the ratio operation is performed on it and the preset fault diagnosis discrimination index threshold to obtain the linkage interference masking ratio. The linkage interference masking ratios calculated by generating linkage interference masking signals at different times are added to the linkage interference masking set, and the mean value μLD of the linkage interference masking ratios in the linkage interference masking set is calculated. The expression is as follows: where grb b represents the bth linkage interference masking ratio in the linkage interference masking set, b={1,2,...,B}, B is a positive integer; calculate the standard deviation σLZ of the linkage interference masking ratio in the linkage interference masking set, the expression is as follows 8. The fault diagnosis method of the intelligent mass flow controller according to claim 7, characterized in that: The standard deviation of the linkage interference masking ratio is compared with a preset standard deviation threshold to evaluate the linkage interference masking degree of the mass flow controller fault signal. If the standard deviation of the linkage interference masking ratio is greater than the standard deviation threshold, a first-level linkage interference masking degree is generated; if the standard deviation of the linkage interference masking ratio is less than or equal to the standard deviation threshold, a second-level linkage interference masking degree is generated; When the first-level linkage interference masking level is generated, respond to the emergency warning signal, and when the second-level linkage interference masking level is generated, respond to the conventional warning signal.

Images