Fault early warning method and device for regulating valve, regulating valve and storage medium

By setting a vibration sensor in the regulating valve and analyzing data using deep learning models, the problem of electric regulating valve jamming in harsh environments is solved, and early warning of faults is achieved, interruption of fluid delivery is avoided, and the stability of the production process is improved. sex.

CN119043702APending Publication Date: 2024-11-29GUANGDONG SHOULONG FLUID CONTROL TECHNOLOGY CO LTD
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411055763.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-02
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

Existing electric regulating valves are prone to jamming problems in harsh environments, resulting in the inability to transport fluid normally. The existing valve position monitoring device can only detect when a fault occurs and cannot be warned in advance.

Method used

By setting up a vibration sensor in the regulating valve, collecting vibration data and opening data, using deep learning models to analyze vibration characteristics and valve adjustment characteristics, generate fault warning data, early warning and avoid interruptions in fluid delivery.

Benefits of technology

It realizes early warning of regulating valve failure, avoids interruption of fluid delivery, and improves the stability and reliability of the production process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119043702A_ABST
    Figure CN119043702A_ABST
Patent Text Reader

Abstract

The invention discloses a fault early warning method of a regulating valve, a fault early warning device of the regulating valve, the regulating valve and a storage medium, and belongs to the technical field of valves. First vibration data sent by a vibration sensor are received, first opening degree data of a valve body of a regulating valve are obtained, and the first vibration data and the first opening degree data are collected within the same time; determining vibration characteristic data according to the adjusting valve vibration data, and determining valve adjusting characteristic data according to the first opening degree data and first valve control data corresponding to the first opening degree data; inputting the vibration characteristic data and the valve adjustment characteristic data into a fault early warning model, outputting fault early warning data, and generating fault prompt information according to the fault early warning data; and the fault early warning model is a deep learning model, so that the effects of early warning the fault of the regulating valve and ensuring the normal transportation of the fluid are achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of valves, and in particular to a fault warning method of a regulating valve, a fault warning device of a regulating valve, a regulating valve and a storage medium. Background Art

[0002] Electric control valves are necessary components for fluid control in the production process and have a wide range of applications. For example, electric control valves are required for regulating water temperature and steam volume in thermal plants. Due to the large number of application scenarios, electric control valves may cause jamming problems in some harsh environments. At present, the commonly used solution is to set up a valve position monitoring device to receive real-time feedback signals from the electric valve position to determine whether there is a jamming problem. However, this method can only detect jamming, which makes it impossible to determine the failure of the control valve in advance, making it impossible to transport the fluid normally.

[0003] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0004] The main purpose of the present invention is to provide a fault warning method of a regulating valve, a fault warning device of a regulating valve, a regulating valve and a storage medium, aiming to warn of faults of the regulating valve and ensure normal transportation of the fluid.

[0005] To achieve the above object, the present invention provides a fault early warning method for a regulating valve, wherein the regulating valve is provided with a vibration sensor, and the fault early warning method for the regulating valve comprises the following steps:

[0006] receiving first vibration data sent by the vibration sensor, and acquiring first opening data of the regulating valve, wherein the first vibration data and the first opening data are collected at the same time;

[0007] Determine vibration characteristic data according to the regulating valve vibration data, and determine valve regulation characteristic data according to the first opening data and first valve control data corresponding to the first opening data;

[0008] Inputting the vibration characteristic data and the valve adjustment characteristic data into a fault warning model, outputting fault warning data, and generating fault prompt information according to the fault warning data;

[0009] The fault warning model is a deep learning model.

[0010] Optionally, before the step of inputting the vibration characteristic data and the valve adjustment characteristic data into the fault early warning model, the method further includes:

[0011] Acquire second vibration data and second opening data, where the second vibration data and the second opening data are data collected before the current moment;

[0012] A deep learning model is trained according to the second vibration data and the second opening data to obtain the fault warning model.

[0013] Optionally, the second vibration data includes a plurality of second sub-vibration data, the second opening data includes a plurality of second sub-opening data, the second sub-vibration data corresponds to the second sub-opening data one by one, and the step of training the deep learning model according to the second vibration data and the second opening data to obtain the fault warning model includes:

[0014] The second sub-vibration data and the corresponding second sub-opening data are grouped according to the second valve control data, and association parameters are set for a plurality of training samples according to the second valve control data to obtain input data of a plurality of training samples;

[0015] Taking the time difference data between the first sample time and the fault time corresponding to the training sample as the output data of the training sample;

[0016] The fault warning model is determined based on multiple training samples and a deep learning model.

[0017] Optionally, the step of determining the fault warning model according to a plurality of training samples and the deep learning model includes:

[0018] Divide multiple training samples into training sets and test sets;

[0019] Training the deep learning model according to the training set to obtain a first training model;

[0020] When the test set verifies that the result of the first training model meets the model target requirement, the first training model is used as the fault warning model;

[0021] When the test set verifies that the result of the first training model does not meet the model target requirements, modify the parameters for training the deep learning model, and return to execute the step of training the deep learning model according to the training set to obtain the first training model.

[0022] Optionally, the step of using the time difference data between the first sample time and the fault time corresponding to the training sample as the output data of the training sample includes:

[0023] Acquire time data corresponding to a plurality of second sub-vibration data in a training sample, and calculate an average value of all the time data as the first sample time;

[0024] Obtaining a fault time of an adjacent fault occurring after a first sample time, wherein the adjacent fault is a fault that does not exist between the first sample time and the fault time;

[0025] The time difference data between the first sample time and the fault time is calculated as the output data of the training sample.

[0026] Optionally, the step of determining vibration characteristic data according to the regulating valve vibration data comprises:

[0027] Using filtering to remove noise components of the regulating valve vibration data to obtain first vibration processing data;

[0028] Extracting the vibration mean, vibration standard deviation, vibration maximum value, and vibration minimum value of the first vibration processing data through time domain analysis;

[0029] Extracting a vibration spectrum of the first vibration processing data by frequency domain analysis;

[0030] extracting the vibration waveform form of the first vibration processing data through feature analysis;

[0031] The vibration mean, the vibration standard deviation, the vibration maximum value, the vibration minimum value, the vibration spectrum and the vibration waveform form are used as the vibration characteristic data.

[0032] Optionally, the step of determining valve adjustment characteristic data according to the first opening data and first valve control data corresponding to the first opening data comprises:

[0033] The change speed of the valve position is determined by calculation based on the first valve control data and the first opening data.

[0034] To achieve the above object, the present invention provides a fault warning device for a regulating valve, the fault warning device for the regulating valve comprising:

[0035] An acquisition module, used for receiving first vibration data sent by the vibration sensor, and acquiring first opening data of the regulating valve, wherein the first vibration data and the first opening data are collected at the same time;

[0036] an extraction module, configured to determine vibration characteristic data according to the regulating valve vibration data, and to determine valve regulation characteristic data according to the first opening data and first valve control data corresponding to the first opening data;

[0037] The analysis module is used to input the vibration characteristic data and the valve adjustment characteristic data into a fault warning model, output the fault warning data, and generate fault prompt information according to the fault warning data, wherein the fault warning model is a deep learning model.

[0038] To achieve the above-mentioned purpose, the present invention provides a regulating valve, which includes: a memory, a processor, and a fault warning program of the regulating valve stored in the memory and executable on the processor, wherein the fault warning program of the regulating valve is configured to implement the steps of the fault warning method of the regulating valve described in any one of the above-mentioned items.

[0039] To achieve the above objectives, the present invention provides a storage medium, on which a fault warning program for a regulating valve is stored. When the fault warning program for a regulating valve is executed by a processor, the steps of any of the above-mentioned methods for warning the fault of a regulating valve are implemented.

[0040] The present invention proposes a fault warning method for a regulating valve. The method can obtain relevant data for analyzing and predicting faults by receiving first vibration data sent by the vibration sensor and acquiring first opening data of the valve of the regulating valve. Vibration characteristic data is determined by the vibration data of the regulating valve. Valve adjustment characteristic data is determined according to the first opening data and first valve control data. Characteristics of the regulating valve during operation can be obtained. The vibration characteristic data and the valve adjustment characteristic data are input into a fault warning model, and fault warning data is output, thereby realizing the prediction of regulating valve faults and avoiding the situation where fluid cannot be transported normally. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a structural schematic diagram of a regulating valve in a hardware operating environment involved in an embodiment of the present invention;

[0042] Figure 2 It is a schematic flow chart of a first embodiment of a fault warning method for a regulating valve according to the present invention;

[0043] Figure 3 It is a flow chart of a second embodiment of a fault warning method for a regulating valve according to the present invention;

[0044] Figure 4 It is a schematic flow chart of the third embodiment of the fault warning method of the regulating valve of the present invention.

[0045] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0046] 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.

[0047] Reference Figure 1 , Figure 1 The schematic diagram of the regulating valve structure of the hardware operating environment involved in the embodiment of the present invention is shown in FIG.

[0048] like Figure 1 As shown, the regulating valve may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, an interactive device 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The interactive device 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional interactive device 1003 may also be connected to the communication bus through a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (WIreless-FIdelity, WI-FI) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM) memory, or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0049] It should be noted that the valve body structure of the regulating valve can be connected to the processor 1001, such as the central processing unit (CPU), the communication bus 1002, the interactive device 1003, the network interface 1004, and the memory 1005 through a wireless network, that is, the above components can be set separately from the valve body. In addition, the structure of the regulating valve may include: a valve core, a valve seat, an actuator, a positioner, a spring, etc., and may also include a vibration sensor.

[0050] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the regulating valve, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0051] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a data storage module, a network communication module, a user interface module and a fault warning program of the regulating valve.

[0052] exist Figure 1 In the control valve shown, the network interface 1004 is mainly used for data communication with other devices; the interactive device 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the control valve of the present invention can be set in the control valve, and the control valve calls the fault warning program of the control valve stored in the memory 1005 through the processor 1001, and executes the fault warning method of the control valve provided in the embodiment of the present invention.

[0053] The embodiment of the present invention provides a fault warning method for a regulating valve, referring to Figure 2 , Figure 2 The figure is a flow chart of a first embodiment of a method for early warning of a fault of a regulating valve according to the present invention. In this embodiment, the method for early warning of a fault of a regulating valve comprises:

[0054] Step S1, receiving first vibration data sent by the vibration sensor, and acquiring first opening data of the regulating valve, wherein the first vibration data and the first opening data are collected at the same time;

[0055] In this embodiment, optionally, the vibration sensor here can be set on the valve body of the regulating valve to detect the vibration caused by the fluid flow, or it can be set at the valve stem. Since the valve stem is a component connecting the actuator and the valve core, it can be installed at the valve stem. In addition to monitoring the vibration caused by the fluid flow, it can also facilitate monitoring the vibration of the valve stem, and the state of the valve stem can be determined by the vibration information. If it is set on the valve stem, it is necessary to ensure good sealing. Of course, it can also be set at the elbow or tee near the valve. The specific method of obtaining the first opening data of the valve of the regulating valve can be determined by receiving data feedback from the valve position transmitter or the valve positioner. It should be noted that in order to ensure that the collected vibration data and opening data can accurately reflect the state of the regulating valve, the first vibration data and the first opening data need to be collected at the same time.

[0056] Step S2, determining vibration characteristic data according to the regulating valve vibration data, and determining valve regulation characteristic data according to the first opening data and first valve control data corresponding to the first opening data;

[0057] The vibration characteristic data here include: average vibration amplitude, vibration standard deviation, maximum value of vibration data and minimum value of vibration data, vibration spectrum diagram, the vibration spectrum diagram includes each vibration frequency and corresponding intensity, and in addition, it can also include the waveform characteristics of vibration. Specifically, the vibration characteristic data is determined by processing the regulating valve vibration data through a data characteristic analysis method. It should be noted that the first opening data reflects the actual position of the valve, and the first valve control data reflects the position to which the valve needs to be adjusted in the system. When the valve is adjusted according to the first valve control data, due to various factors, the adjustment of the valve often has a delay, so that the change speed of the valve position of the valve can be determined according to the first opening data and the first valve control data. In this embodiment, the valve adjustment characteristic data here can be the change speed of the valve position, or it can be the data of a change speed curve composed of multiple change speeds and the curve characteristics corresponding to the change speed curve.

[0058] Step S3, inputting the vibration characteristic data and the valve adjustment characteristic data into a fault warning model, outputting fault warning data, and generating fault prompt information according to the fault warning data, wherein the fault warning model is a deep learning model;

[0059] Optionally, the vibration characteristic data and the valve adjustment characteristic data are normalized, and the vibration characteristic data and the valve adjustment characteristic data are input into a fault warning model to output fault warning data, where the fault warning data may be the time to the fault, the probability of the fault, etc. When the fault warning data is greater than or equal to a preset fault data threshold, the generated fault information indicates that there is a fault risk and maintenance is recommended; when the fault warning data is less than a preset fault data threshold, the generated fault information indicates that there is no fault risk; of course, the fault warning data here may also include fault warning data of different types of faults.

[0060] In this embodiment, by receiving the first vibration data sent by the vibration sensor and acquiring the first opening data of the valve of the control valve, relevant data for analyzing and predicting faults can be obtained, and vibration characteristic data is determined by the vibration data of the control valve. Valve adjustment characteristic data is determined according to the first opening data and the first valve control data. The characteristics of the control valve during operation can be obtained, and the vibration characteristic data and the valve adjustment characteristic data are input into a fault warning model, and fault warning data is output, thereby realizing the prediction of the control valve fault and avoiding the situation where the fluid cannot be transported normally.

[0061] Further, based on the first embodiment, a second embodiment of the fault warning method of the regulating valve of the present invention is proposed. In this embodiment, referring to Figure 3 , before the step of inputting the vibration characteristic data and the valve adjustment characteristic data into the fault warning model, it also includes:

[0062] Step S301, acquiring second vibration data and second opening data, where the second vibration data and the second opening data are data collected before the current moment;

[0063] It should be noted that the second vibration data and the first vibration data here need to be data collected by a vibration sensor set at the same position. The second opening data is obtained in the same manner as the first opening data. Optionally, the second vibration data and the second opening data can be obtained from a storage medium that records the vibration data and the opening data, and the second vibration data and the second opening data include data collected at multiple times.

[0064] Step S302: training a deep learning model according to the second vibration data and the second opening data to obtain the fault warning model.

[0065] Specifically, by preprocessing the second vibration data and the second opening data, the preprocessing here includes data cleaning, i.e. removing duplicate records, processing missing values, identifying and correcting errors or outliers, etc., the second vibration data and the second opening data are normalized, and after preprocessing, the features of the second vibration data and the second opening data are extracted, and labeling and other operations are performed to obtain a data set for training, and a deep learning model is trained according to the data set to obtain a fault warning model.

[0066] In this embodiment, the second vibration data and the second opening data collected before the current moment are used, and a deep learning model is trained based on the second vibration data and the second opening data, so that a model for early warning can be obtained, thereby realizing fault early warning of the regulating valve.

[0067] Further, based on the second embodiment, a third embodiment of the fault warning method of the regulating valve of the present invention is proposed. In this embodiment, referring to Figure 4 , the second vibration data includes a plurality of second sub-vibration data, the second opening data includes a plurality of second sub-opening data, the second sub-vibration data corresponds to the second sub-opening data one by one, and the step of training the deep learning model according to the second vibration data and the second opening data to obtain the fault warning model includes:

[0068] Step S3021, grouping the second sub-vibration data and the corresponding second sub-opening data according to the second valve control data, and setting association parameters for a plurality of training samples according to the second valve control data to obtain input data of a plurality of training samples;

[0069] In this embodiment, the second valve control data here is used to control the valve, including regulating the valve state, maintaining the state, and the regulating valve state here includes the open sub-state of opening the door and the closed sub-state of closing. In this embodiment, in addition to grouping the second sub-vibration data and the corresponding second sub-opening data according to the second valve control data, the second sub-vibration data and the corresponding second sub-opening data can also be grouped according to the standard time length. Thus, the input data of multiple training samples of the standard time length can be obtained. It should be noted that in this embodiment, due to the large amount of data, it is not directly input into the model training here, but the features of the input data are extracted and then input into the model training. The specific extraction method needs to correspond to the method of determining the vibration feature data according to the regulating valve vibration data, and determining the valve adjustment feature data according to the first opening data and the first valve control data corresponding to the first opening data, so as to ensure that the input of the fault warning model is consistent with the vibration feature data and the valve adjustment feature data.

[0070] Of course, in other embodiments, the input data may also be directly input into the deep learning model training.

[0071] Step S3022, taking the time difference data between the first sample time and the fault time corresponding to the training sample as the output data of the training sample;

[0072] In this embodiment, by using the time difference data as the output data of the training sample, the time when the fault occurs is actually associated with the input data of the training sample. This is because the valve stem or valve core movement is blocked due to the jamming, fouling, corrosion, etc. of the valve core in the regulating valve, which does not appear suddenly, and often there have been changes that affect the vibration and opening before. It can effectively predict the time when it may occur. Of course, the effect may be relatively general for the fault of foreign body blockage. In other embodiments, the possibility data of the fault can also be determined by judging the fouling data, corrosion data, wear data and foreign body blockage data in the regulating valve through routine maintenance, and the possibility data can be used as the output data of the training sample.

[0073] Step S3023: determine the fault warning model based on multiple training samples and the deep learning model.

[0074] In this embodiment, the type of deep learning model is not limited, and may be a multilayer perceptron (MLP), a convolutional neural network (CNN), a recurrent neural network (RNN), and a deep residual neural network (ResNet), etc. Specifically, the weights of each unit in the deep learning model are adjusted through the loss function and reverse transfer.

[0075] In this embodiment, the second sub-vibration data and the corresponding second sub-opening data are grouped according to the second valve control data, and the associated parameters are set for multiple training samples according to the second valve control data, so as to obtain the input data of multiple training samples, and the time difference data between the first sample time and the fault time corresponding to the training sample is used as the output data of the training sample, thereby obtaining a model that can predict the fault time by inputting vibration characteristic data and opening data. Since the time data is accurate, the possibility of fault judgment can be improved compared with other embodiments that need to be judged through maintenance, and the time data is easier to obtain than the maintenance data, and more training data can be obtained under the same other conditions.

[0076] Furthermore, the step of using the time difference data between the first sample time and the fault time corresponding to the training sample as the output data of the training sample includes:

[0077] Acquire time data corresponding to a plurality of second sub-vibration data in a training sample, and calculate an average value of all the time data as the first sample time;

[0078] Obtaining a fault time of an adjacent fault occurring after a first sample time, wherein the adjacent fault is a fault that does not exist between the first sample time and the fault time;

[0079] The time difference data between the first sample time and the fault time is calculated as the output data of the training sample.

[0080] Specifically, the time difference data is obtained by subtracting the first sample time from the fault time. In other embodiments, the latest time among the time data corresponding to the plurality of second sub-vibration data in the training sample may be used as the first sample time.

[0081] In this embodiment, the accuracy of the time difference data can be improved by calculating the average value of all the time data as the first sample time.

[0082] Further, based on the third embodiment, a fourth embodiment of the fault warning method of the regulating valve of the present invention is proposed. In this embodiment, the step of determining the fault warning model according to the plurality of training samples and the deep learning model includes:

[0083] Divide multiple training samples into training sets and test sets;

[0084] Training the deep learning model according to the training set to obtain a first training model;

[0085] When the test set verifies that the result of the first training model meets the model target requirement, the first training model is used as the fault warning model;

[0086] When the test set verifies that the result of the first training model does not meet the model target requirements, modify the parameters for training the deep learning model, and return to execute the step of training the deep learning model according to the training set to obtain the first training model.

[0087] Specifically, the ratio of the training set to the test set can be 60% and 40% respectively. The model target requirement here can be that the accuracy of the prediction of the first training model is higher than the set accuracy threshold, or that the recall rate of the first training model is higher than the set recall rate threshold, etc.

[0088] In this embodiment, the deep learning model is trained according to the training set to obtain a first training model; when the test set verifies that the result of the first training model meets the model target requirements, the first training model is used as the fault warning model; when the test set verifies that the result of the first training model does not meet the model target requirements, the parameters for training the deep learning model are modified, and the step of training the deep learning model according to the training set to obtain the first training model is returned to execute, thereby ensuring the accuracy of the prediction of the fault warning model.

[0089] Further, based on any of the above embodiments, a fifth embodiment of the fault warning method of the regulating valve of the present invention is proposed. In this embodiment, the step of determining the vibration characteristic data according to the vibration data of the regulating valve includes:

[0090] Using filtering to remove noise components of the regulating valve vibration data to obtain first vibration processing data;

[0091] Extracting the vibration mean, vibration standard deviation, vibration maximum value, and vibration minimum value of the first vibration processing data through time domain analysis;

[0092] Extracting a vibration spectrum of the first vibration processing data by frequency domain analysis;

[0093] extracting the vibration waveform form of the first vibration processing data through feature analysis;

[0094] The vibration mean, the vibration standard deviation, the vibration maximum value, the vibration minimum value, the vibration spectrum and the vibration waveform form are used as the vibration characteristic data.

[0095] In this embodiment, the vibration mean, vibration standard deviation, vibration maximum value, and vibration minimum value of the first vibration processing data can be directly extracted through time domain analysis. Frequency domain analysis is specifically Fourier transform or fast Fourier transform to obtain the vibration spectrum diagram. Different filters can also be set to filter the regulating valve vibration data, and a vibration spectrum diagram is generated according to the filtering results, and the data of the vibration spectrum diagram, the vibration mean, the vibration standard deviation, the vibration maximum value, the vibration minimum value, and the vibration waveform form are used as data for the input model.

[0096] In this embodiment, the vibration mean, the vibration standard deviation, the vibration maximum value, the vibration minimum value, the vibration spectrum diagram and the vibration waveform form are used as the vibration characteristic data, so as to improve the accuracy of the model output result.

[0097] Furthermore, the step of determining valve adjustment characteristic data according to the first opening data and first valve control data corresponding to the first opening data includes:

[0098] The change speed of the valve position is determined by calculation based on the first valve control data and the first opening data.

[0099] In addition, an embodiment of the present invention further provides a fault warning device for a regulating valve, the fault warning device for the regulating valve comprising:

[0100] An acquisition module, used for receiving first vibration data sent by the vibration sensor, and acquiring first opening data of the regulating valve, wherein the first vibration data and the first opening data are collected at the same time;

[0101] an extraction module, configured to determine vibration characteristic data according to the regulating valve vibration data, and to determine valve regulation characteristic data according to the first opening data and first valve control data corresponding to the first opening data;

[0102] The analysis module is used to input the vibration characteristic data and the valve adjustment characteristic data into a fault warning model, output the fault warning data, and generate fault prompt information according to the fault warning data, wherein the fault warning model is a deep learning model.

[0103] In addition, an embodiment of the present invention also proposes a regulating valve, which includes: a memory, a processor, and a fault warning program of the regulating valve stored in the memory and executable on the processor, wherein the fault warning program of the regulating valve is configured to implement the steps of an embodiment of the fault warning method of the regulating valve described in any one of the above items.

[0104] In addition, an embodiment of the present invention further proposes a storage medium, on which a fault warning program for a regulating valve is stored. When the fault warning program for a regulating valve is executed by a processor, the steps of any of the above-mentioned embodiments of the method for fault warning of a regulating valve are implemented.

[0105] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0106] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0107] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0108] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A fault warning method for a regulating valve, characterized in that: The regulating valve is provided with a vibration sensor, and the regulating valve failure early warning method comprises the following steps: receiving first vibration data sent by the vibration sensor, and acquiring first opening data of the regulating valve, wherein the first vibration data and the first opening data are collected at the same time; Determine vibration characteristic data according to the regulating valve vibration data, and determine valve regulation characteristic data according to the first opening data and first valve control data corresponding to the first opening data; The vibration characteristic data and the valve adjustment characteristic data are input into a fault warning model, fault warning data is output, and fault prompt information is generated according to the fault warning data. The fault warning model is a deep learning model.

2. The fault warning method of the regulating valve according to claim 1, characterized in that: Before the step of inputting the vibration characteristic data and the valve adjustment characteristic data into the fault early warning model, the method further includes: Acquire second vibration data and second opening data, where the second vibration data and the second opening data are data collected before the current moment; A deep learning model is trained according to the second vibration data and the second opening data to obtain the fault warning model.

3. The fault warning method of the regulating valve according to claim 2, characterized in that: The second vibration data includes a plurality of second sub-vibration data, the second opening data includes a plurality of second sub-opening data, the second sub-vibration data corresponds to the second sub-opening data one by one, and the step of training the deep learning model according to the second vibration data and the second opening data to obtain the fault warning model includes: The second sub-vibration data and the corresponding second sub-opening data are grouped according to the second valve control data, and association parameters are set for a plurality of training samples according to the second valve control data to obtain input data of a plurality of training samples; Taking the time difference data between the first sample time and the fault time corresponding to the training sample as the output data of the training sample; The fault warning model is determined based on multiple training samples and a deep learning model.

4. The fault warning method of the regulating valve according to claim 3, characterized in that: The step of determining the fault warning model according to a plurality of training samples and the deep learning model comprises: Divide multiple training samples into training sets and test sets; Training the deep learning model according to the training set to obtain a first training model; When the test set verifies that the result of the first training model meets the model target requirement, the first training model is used as the fault warning model; When the test set verifies that the result of the first training model does not meet the model target requirements, modify the parameters for training the deep learning model, and return to execute the step of training the deep learning model according to the training set to obtain the first training model.

5. The fault warning method of the regulating valve according to claim 3, characterized in that: The step of using the time difference data between the first sample time and the fault time corresponding to the training sample as the output data of the training sample comprises: Acquire time data corresponding to a plurality of second sub-vibration data in a training sample, and calculate an average value of all the time data as the first sample time; Obtaining a fault time of an adjacent fault occurring after a first sample time, wherein the adjacent fault is a fault that does not exist between the first sample time and the fault time; The time difference data between the first sample time and the fault time is calculated as the output data of the training sample.

6. The method for early warning of a failure of a regulating valve according to claim 1, characterized in that: The step of determining vibration characteristic data according to the vibration data of the regulating valve comprises: Using filtering to remove noise components of the regulating valve vibration data to obtain first vibration processing data; Extracting the vibration mean, vibration standard deviation, vibration maximum value, and vibration minimum value of the first vibration processing data through time domain analysis; Extracting a vibration spectrum of the first vibration processing data by frequency domain analysis; extracting the vibration waveform form of the first vibration processing data through feature analysis; The vibration mean, the vibration standard deviation, the vibration maximum value, the vibration minimum value, the vibration spectrum and the vibration waveform form are used as the vibration characteristic data.

7. The fault warning method for a regulating valve according to any one of claims 1 to 6, characterized in that: The step of determining valve adjustment characteristic data according to the first opening data and the first valve control data corresponding to the first opening data comprises: The change speed of the valve position is determined by calculation based on the first valve control data and the first opening data.

8. A fault warning device for a regulating valve, characterized in that: The fault warning device of the regulating valve comprises: An acquisition module, used for receiving first vibration data sent by the vibration sensor, and acquiring first opening data of the regulating valve, wherein the first vibration data and the first opening data are collected at the same time; an extraction module, configured to determine vibration characteristic data according to the regulating valve vibration data, and to determine valve regulation characteristic data according to the first opening data and first valve control data corresponding to the first opening data; The analysis module is used to input the vibration characteristic data and the valve adjustment characteristic data into a fault warning model, output the fault warning data, and generate fault prompt information according to the fault warning data, wherein the fault warning model is a deep learning model.

9. A regulating valve, characterized in that: The regulating valve comprises: a memory, a processor, and a fault warning program of the regulating valve stored in the memory and executable on the processor, wherein the fault warning program of the regulating valve is configured to implement the steps of the fault warning method of the regulating valve as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores a fault warning program of the regulating valve, and when the fault warning program of the regulating valve is executed by the processor, the steps of the fault warning method of the regulating valve according to any one of claims 1 to 7 are implemented.

Citation Information

Cited By

  • Intelligent control method and system of electric actuating mechanism

    CN119617164A

  • Fault diagnosis method, device and system suitable for intelligent regulating valve

    CN121298234A

  • Fault diagnosis method, device and system suitable for intelligent regulating valve

    CN121298234B