A Fault Diagnosis and Prediction Method and Device for a Control Valve

Abstract

The present application relates to a method and device for fault diagnosis and prediction of a regulating valve, including: obtaining first-type parameters of a specified regulating valve collected, and determining the state information of the regulating valve according to the first-type parameters; according to the current state information of the regulating valve, sending a collection instruction matching the state information to a collection module, so that the collection module performs data collection according to the collection instruction; analyzing the first-type parameters and the thresholds corresponding to the first-type parameters to obtain a first analysis result; when the first analysis result is an abnormal result, sending a collection instruction for obtaining second-type parameters to the collection module, and performing simulation processing based on the first-type parameters and the second-type parameters to obtain simulation data, and obtaining a second analysis result based on the simulation data; determining the diagnosis result of the regulating valve based on the first analysis result and the second analysis result. The method in the present application reduces the interference of human experience on the fault prediction of the regulating valve, and realizes the full-cycle diagnosis and prediction of the faults of the regulating valve.

Description

Technical Field

[0001] The present invention relates to the technical field of fault identification, and in particular, to a method and device for fault diagnosis and prediction of a control valve. Background Art

[0002] The control valve is one of the most important devices in the control loop of the energy and chemical industries. The working stability of the control valve directly affects the safety and stability of the control loop, thereby affecting the operation of the entire device. Once a fault occurs in the control valve, it will cause great safety hazards and economic losses to the factory production. Therefore, if the fault of the control valve can be predicted to prevent the occurrence of faults, it is beneficial to the safe and stable production of the factory.

[0003] Currently, the faults of the control valve mainly rely on on-site inspections by staff, which requires a large amount of manpower and time costs. At the same time, fault identification also depends on the experience level of the staff. Potential faults are easily overlooked and are often not discovered until the faults are relatively serious, which has a certain lag, is not conducive to production safety, and also increases the maintenance cost. To sum up, for the fault identification and prediction of the control valve, currently, the opening threshold of the control valve is also set to monitor the control valve during stable operation. That is to say, when the opening of the control valve exceeds the threshold range, it is considered that the control valve may have a fault.

[0004] There is currently a method for predicting control valve faults, specifically: by collecting other data such as the operation of the control valve, training a fault diagnosis or prediction model, and validating it with test data. However, this method has the following defects: 1. The setting of the opening threshold of the control valve depends on human experience and is prone to deviation, resulting in missed faults; 2. The action frequency of some control valves is low, and its duration is relatively short compared to stable operation. There is little data that can be collected, and relying only on the action process for fault identification may have a lag; similarly, relying only on the data during steady-state operation will miss the relevant faults that will only be exposed during the action process; 3. Based on data-driven diagnosis, a large amount of various data needs to be collected, and different data detection and collection devices often need to be added to the loop, which not only increases the cost but also may affect the operation of the loop; although some data can be read from the control valve itself, when there are many parameters to be read, the reading period will increase; in addition, the data model without theoretical guidance depends on many parameters, and there is a problem of difficult model maintenance. Summary of the Invention

[0005] (1) Technical Problems to be Solved

[0006] In view of the above-mentioned disadvantages and deficiencies of the prior art, the present application provides a method and device for fault diagnosis and prediction of a control valve.

[0007] (2) Technical Solutions

[0008] To achieve the above object, the present application adopts the following technical solutions:

[0009] In a first aspect, the present application provides a method for fault diagnosis and prediction of a regulating valve, including:

[0010] S10. Obtain the first type of parameters of a specified regulating valve collected, and determine the status information of the regulating valve according to the first type of parameters;

[0011] S20. According to the status information of the current regulating valve, send a collection instruction matching the status information to the collection module, so that the collection module performs data collection according to the collection instruction;

[0012] S30. Analyze the first type of parameters and the thresholds corresponding to the first type of parameters to obtain a first analysis result;

[0013] S40. When the first analysis result is an abnormal result, send a collection instruction for obtaining the second type of parameters to the collection module, and perform simulation processing based on the first type of parameters and the second type of parameters to obtain simulation data, and obtain a second analysis result based on the simulation data;

[0014] S50. Determine the diagnosis result of the regulating valve based on the first analysis result and the second analysis result.

[0015] Optionally, it further includes:

[0016] S60. If the diagnosis result of the regulating valve is in a normal state, perform fitting based on the historical data of the first type of parameters to obtain prediction data, and diagnose the faults of the regulating valve in a future time period based on the prediction data.

[0017] Optionally, it further includes:

[0018] The first type of parameters includes: valve position set value, valve position feedback, adjustment rate, and valve position deviation;

[0019] The S10 includes:

[0020] All parameters collected and transmitted by the collection module carry time information when the parameters are collected;

[0021] If the valve position set value changes, it is determined that the regulating valve is in an operating state;

[0022] For the regulating valve in the operating state, when the change range of the valve position feedback is less than a preset value, it is determined that the regulating valve enters a stable state.

[0023] Optionally, the S20 includes:

[0024] If the regulating valve is in a stable state, send a first acquisition instruction to the acquisition module, where the first acquisition instruction carries a first acquisition frequency and an identifier of the parameter to be acquired;

[0025] If the regulating valve is in an operating state, send a second acquisition instruction to the acquisition module, where the second acquisition instruction carries a second acquisition frequency and an identifier of the parameter to be acquired;

[0026] The first acquisition frequency is less than the second acquisition frequency.

[0027] Optionally, the S30 includes:

[0028] S301. Obtain the thresholds of each parameter in the first type of parameters according to the historical data of the first type of parameters in a preset time period;

[0029] S302. Compare the first type of parameters with the corresponding thresholds. If at least one parameter exceeds the threshold, determine that the first analysis result is an abnormal result.

[0030] Optionally, the S301 includes:

[0031] For the valve position deviation parameter, the threshold of the valve position deviation is S th = E S ±σ S ;

[0032] where E S is the mean of the valve position deviation historical data, and σ S is the relative standard deviation corresponding to the valve position deviation;

[0033] For the regulation rate parameter, the threshold of the regulation rate is V th = E V ±σ V ;

[0034] where E V is the mean of the regulation rate historical data, and σ V is the relative standard deviation corresponding to the regulation rate.

[0035] Optionally, the S40 includes:

[0036] The second type of parameters includes: supply air pressure, medium temperature, valve inlet pressure, valve outlet pressure, control signal, and working mode;

[0037] The simulation data includes: the third type of parameter information corresponding to each time point during the simulation process, where the third type of parameter is the same parameter as the first type of parameter;

[0038] Analyze the third type of parameter information and the threshold corresponding to the first type of parameter to obtain a second analysis result, where the threshold corresponding to the first type of parameter is the updated and adjusted threshold.

[0039] Optionally, the S40 includes:

[0040] Compare the third type of parameter information with the threshold corresponding to the first type of parameter. If at least one parameter exceeds the threshold, determine that the second analysis result is an abnormal result.

[0041] Optionally, the S50 includes:

[0042] If the first analysis result is a normal result, determine that the regulating valve is in a normal state;

[0043] If the first analysis result is an abnormal result and the second analysis result is a normal result, determine that the regulating valve is in a faulty state;

[0044] If both the first analysis result and the second analysis result are abnormal results, adjust the parameter threshold of the regulating valve and repeat the steps of S30.

[0045] In a second aspect, the present application provides a device for fault diagnosis and prediction of a regulating valve, including:

[0046] The device includes a control device and a collection module. The control device is used to receive the parameters of a specified regulating valve collected in real time by the collection module;

[0047] The control device includes:

[0048] A state judgment module, configured to receive the first type of parameters of a specified regulating valve sent by the collection module, and determine the state information of the regulating valve according to the first type of parameters;

[0049] A data collection module, configured to send a collection instruction matching the state information to the collection module according to the current state information of the regulating valve, so that the collection module performs data collection according to the collection instruction;

[0050] A data processing module, configured to analyze the first type of parameters and the threshold corresponding to the first type of parameters to obtain a first analysis result;

[0051] A data simulation module, configured to send a collection instruction for obtaining the second type of parameters to the collection module when the first analysis result is an abnormal result, and perform simulation processing based on the first type of parameters and the second type of parameters to obtain simulation data, and obtain a second analysis result according to the simulation data;

[0052] A fault diagnosis module, configured to determine the diagnosis result of the regulating valve based on the first analysis result and the second analysis result;

[0053] A fault prediction module, which is used to perform fitting based on the historical data of the first type of parameters if the diagnostic result of the regulating valve is in a normal state, obtain prediction data, and diagnose the faults of the regulating valve in a future time period based on the prediction data.

[0054] (III) Beneficial effects

[0055] First, determining the parameter threshold based on the historical data of the regulating valve parameters reduces the interference of human experience, and thus reduces the missed diagnosis or misdiagnosis of faults;

[0056] Second, by collecting different regulating valve parameter data in a frequency division manner, the acquisition pressure of the parameters is reduced, and full-cycle fault identification and prediction of the regulating valve are realized in both the action and stable operation modes.

[0057] Finally, by performing fault verification on the regulating valve, the misdiagnosis of faults is reduced, and thus the accuracy of the regulating valve fault diagnosis is improved. Description of the drawings

[0058] This application is described with the aid of the following drawings:

[0059] Figure 1 It is a schematic flowchart of the regulating valve fault diagnosis method;

[0060] Figure 2 It is a schematic flowchart of the regulating valve parameter threshold calculation method;

[0061] Figure 3 It is a schematic structural diagram of the regulating valve fault diagnosis and prediction device. Detailed implementation manners

[0062] In order to better explain the present invention for easy understanding, the present invention will be described in detail below in conjunction with the drawings through specific implementation manners. It can be understood that the specific embodiments described below are only used to explain the related invention and do not limit the invention. In addition, it should be noted that without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other; for the convenience of description, only the parts related to the invention are shown in the drawings.

[0063] Embodiment 1 provides a regulating valve fault diagnosis method, as Figure 1 shown, and the specific method steps are as follows:

[0064] S10. Obtain the first type of parameters of the specified regulating valve collected, and determine the state information of the regulating valve according to the first type of parameters.

[0065] In this embodiment, the regulating valve can be applied to petrochemical and other common production fields, which is not limited herein. For example, by setting the regulating valve at the raw material input device, the loop of the raw material input quantity can be controlled.

[0066] In this embodiment, the first type of parameters includes: valve position set value, valve position feedback, regulation rate, and valve position deviation.

[0067] In this embodiment, the state of the regulating valve is divided into a stable state and an operating state.

[0068] S20. According to the current state information of the regulating valve, send a collection instruction matching the state information to the collection module, so that the collection module performs data collection according to the collection instruction.

[0069] S30. Analyze the first type of parameters and the thresholds corresponding to the first type of parameters to obtain a first analysis result.

[0070] S40. When the first analysis result is an abnormal result, send a collection instruction for obtaining the second type of parameters to the collection module, perform simulation processing based on the first type of parameters and the second type of parameters to obtain simulation data, and obtain a second analysis result based on the simulation data.

[0071] In this embodiment, the second type of parameters includes: supply air pressure, medium temperature, valve inlet pressure, valve outlet pressure, control signal, and working mode.

[0072] In this embodiment, the simulation data includes: the information of the third type of parameters corresponding to each time point during the simulation process, where the third type of parameters is the same as the first type of parameters.

[0073] S50. Determine the diagnosis result of the regulating valve based on the first analysis result and the second analysis result.

[0074] In the method described in the first embodiment, determining the parameter threshold based on the historical data of the regulating valve parameters reduces the interference of human experience, and thus reduces the missed diagnosis or misdiagnosis of faults; by collecting different regulating valve parameter data in a frequency division manner, the collection pressure of the parameters is reduced, and full-cycle fault diagnosis of the regulating valve is realized in both the action and stable operation modes. By performing fault verification on the regulating valve, the misdiagnosis of faults is reduced, and thus the accuracy of the regulating valve fault diagnosis is improved.

[0075] The second embodiment provides a method for fault diagnosis and prediction of a regulating valve. The specific method steps are as follows:

[0076] S10. Obtain the first type of parameters of the specified regulating valve collected, and determine the state information of the regulating valve according to the first type of parameters.

[0077] Based on step S10, it should be noted that:

[0078] In this embodiment, after the acquisition module acquires the parameter data of the specified regulating valve, for all the acquired parameter data, the control device can timestamp the parameter data, so that all the acquired parameter data carry the time information at the time of parameter acquisition.

[0079] In this embodiment, the control device determines the current state information of the regulating valve according to the valve position set value and valve position feedback in the first type of parameters. Among them, the valve position set value and valve position feedback can be directly read from the upper computer of the device without prior acquisition, which reduces the workload of the regulating valve fault diagnosis to a certain extent;

[0080] For better understanding, the following method can be used to determine the state information of the regulating valve:

[0081] S10-1. If the valve position set value changes, it is determined that the regulating valve is in the operating state.

[0082] S10-2. For the regulating valve in the operating state, when the change range of the valve position feedback is less than the preset value, it is determined that the regulating valve enters the stable state.

[0083] In this embodiment, the preset value is not a fixed value and can be set according to the specific working conditions.

[0084] S20. According to the current state information of the regulating valve, send an acquisition instruction matching the state information to the acquisition module, so that the acquisition module performs data acquisition according to the acquisition instruction.

[0085] Based on step S20, it should be noted that:

[0086] In this embodiment, for better understanding, the following method can be used to acquire the parameter data of the regulating valve:

[0087] S20-1. If the current regulating valve is in the stable state, the control device sends a first acquisition instruction to the acquisition module. The first acquisition instruction carries the first acquisition frequency and the identifier of the parameter to be acquired.

[0088] S20-2. If the current regulating valve is in the operating state, the control device sends a second acquisition instruction to the acquisition module. The second acquisition instruction carries the second acquisition frequency and the identifier of the parameter to be acquired.

[0089] In this embodiment, the first acquisition frequency is less than the second acquisition frequency, and the specific values of the first and second acquisition frequencies are not fixedly limited. The specific frequency values are determined according to the specific working conditions.

[0090] In this embodiment, the acquisition module collects parameters of the regulating valve in the operating state at a high frequency and collects parameters of the regulating valve in the stable state at a low frequency, which can reduce the pressure of parameter acquisition, and then complete the fault identification and prediction of the whole cycle for the regulating valve.

[0091] S30. Analyze the first type of parameters and the thresholds corresponding to the first type of parameters to obtain the first analysis result.

[0092] Based on step S30, it should be noted that:

[0093] In this embodiment, for better understanding, the following method can be used to obtain and analyze the first analysis result:

[0094] S30-1. The control device obtains the thresholds corresponding to the adjustment rate and valve position deviation according to the historical data of the adjustment rate and valve position deviation in the preset time period;

[0095] S30-2. The control device compares the real-time data of the collected adjustment rate and valve position deviation with the corresponding thresholds. If at least one parameter exceeds the threshold, it is determined that the first analysis result is an abnormal result.

[0096] In this embodiment, for the valve position deviation parameter, the threshold of the valve position deviation is:

[0097] S th = E S ±σ S

[0098] Wherein, E S is the mean value of the historical data of the valve position deviation, and σ S is the relative standard deviation corresponding to the valve position deviation;

[0099] In this embodiment, for the adjustment rate parameter, the threshold of the adjustment rate is:

[0100] V th = E V ±σ V

[0101] Wherein, E V is the mean value of the historical data of the adjustment rate, and σ V is the relative standard deviation corresponding to the adjustment rate.

[0102] In this embodiment, the historical data of the valve position deviation and the adjustment rate are directly used as the data support for fault diagnosis and prediction. Since the interference of human experience is reduced, the missed diagnosis and misdiagnosis of faults are reduced.

[0103] S40. When the first analysis result is an abnormal result, send a collection instruction for obtaining the second type of parameters to the collection module, perform simulation processing based on the first type of parameters and the second type of parameters to obtain simulation data, and obtain a second analysis result based on the simulation data.

[0104] Based on step S40, it should be noted that:

[0105] In this embodiment, if the first analysis result is an abnormal result, the control device will analyze the adjustment rate and valve position deviation obtained based on the simulation data with the thresholds corresponding to the adjustment rate and valve position deviation to obtain the second analysis result.

[0106] In this embodiment, the thresholds corresponding to the adjustment rate and valve position deviation in S40 are the thresholds corresponding to the adjustment rate and valve position deviation re-determined based on the simulation data.

[0107] In this embodiment, the control device will compare the adjustment rate and valve position deviation obtained based on the simulation data with the thresholds corresponding to the adjustment rate and valve position deviation. If at least one parameter exceeds the threshold, it is determined that the second analysis result is an abnormal result.

[0108] In this embodiment, after completing the fault diagnosis of the specified regulating valve, performing fault verification on the regulating valve based on parameter data simulation can reduce false fault diagnosis and further improve the diagnosis accuracy.

[0109] S50. Determine the diagnosis result of the regulating valve based on the first analysis result and the second analysis result.

[0110] Based on step S50, it should be noted that:

[0111] In this embodiment, if the first analysis result is a normal result, the control device determines that the regulating valve is in a normal state in this embodiment;

[0112] In this embodiment, if the first analysis result is an abnormal result and the second analysis result is a normal result, the control device determines that the regulating valve is in a fault state;

[0113] In this embodiment, if both the first analysis result and the second analysis result are abnormal results, the control device adjusts the parameter thresholds of the regulating valve and repeats the steps of S30.

[0114] S60. If the diagnosis result of the regulating valve is a normal state, perform fitting based on the historical data of the first type of parameters to obtain prediction data, diagnose the faults of the regulating valve in the future time period based on the prediction data, and give a fault prediction curve.

[0115] Based on step S60, it should be noted that:

[0116] For better understanding, the following method can be used to perform fault prediction on a specified regulating valve:

[0117] An optional processing method is as follows: If the regulating valve is in the action mode, the control device fits the change of the adjustment rate and the number of adjustments in a preset time period, predicts the adjustment rate when the valve undergoes N actions in the future; and determines the adjustment rate threshold when the valve undergoes the Nth action in the future based on the historical data of the adjustment rate and the action data from the 1st to the (N - 1)th actions obtained through prediction.

[0118] Another optional processing method is as follows: If the regulating valve is in the steady state mode, the control device fits the change of the valve position deviation with the number of data points in a preset time period, predicts the valve position deviation of the Nth steady state data after the end of the preset time, and determines the valve position deviation threshold in the future time period based on the historical data of the valve position deviation and the data from the 1st to the (N - 1)th data obtained through prediction.

[0119] In this embodiment, based on the obtained predicted values of the adjustment rate and the valve position deviation, as well as the new threshold data, the possible faults of the regulating valve in the future time period are diagnosed, and the possible future fault conditions are published.

[0120] In this embodiment, the method of fitting the historical data of the diagnostic parameters is not specifically limited, and it can be a common fitting algorithm such as the least squares method.

[0121] In the method described in Embodiment 2, determining the parameter threshold based on the historical data of the regulating valve parameters reduces the interference of human experience, thereby reducing the missed diagnosis or misdiagnosis of faults; collecting different regulating valve parameter data through frequency division reduces the acquisition pressure of the parameters, and realizes the full-cycle fault identification and prediction of the regulating valve in both the action and stable operation modes. By performing fault verification on the regulating valve, the misdiagnosis of faults is reduced, thereby improving the accuracy of regulating valve fault diagnosis.

[0122] Embodiment 3 provides a method for calculating the parameter threshold of a regulating valve, as Figure 2 shown, and the specific method is as follows:

[0123] 1) Calculate the average value E and the standard deviation σ of the regulating valve parameters;

[0124] 2) Based on calculate the relative standard deviation δ%, and determine whether δ% is less than 5%;

[0125] 3) If δ% is greater than or equal to 5%, set the data window translation size, and use the average data of each data window as the data of a single parameter, then go back to the above step 2) until δ% is less than 5%;

[0126] 4) If δ% is less than 5%, calculate the threshold of the regulating valve parameter based on the average value E and the standard deviation σ of the regulating valve parameter.

[0127] The calculation methods of the valve position deviation threshold and the regulation rate threshold are described in detail below, specifically:

[0128] In this embodiment, when the regulating valve parameter is the valve position deviation, the valve position deviation threshold S th is calculated as follows:

[0129] 1) Based on S i = OP i - MV i calculate the valve position deviation S i ;

[0130] where, OP i is the historical operation data of the valve position set value, and MV i is the historical operation data of the valve position feedback, specifically:

[0131] OP i = {OP1, OP2,... OP N}, MV i = {MV1, MV2,... MV N}, i = {1, 2,..., N};

[0132] In this embodiment, based on the control device, the calculation process of the valve position deviation S i in the above 1) is completed, and thus there is no need to calculate separately, and the valve position deviation S i can be directly obtained through the upper computer;

[0133] 2) Based on calculate the mean value E of N historical valve position deviations S ;

[0134] 3) Based on calculate the standard deviation σ of N historical valve position deviations S ;

[0135] 4) Based on calculate the relative standard deviation δ S %;

[0136] 5) If the relative standard deviation δ S % is greater than 5%, then based on calculate the adjusted valve position deviation S of the kth group w,k ; Similarly, obtain all the adjusted valve position deviations S w ;

[0137] where, k = {1, 2,..., N w}, Nw = N - n + 1, where N w is the number of groups of currently shared data windows, and n is the number of data in each data window;

[0138] 6) Based on all the adjusted valve position deviations S w , recalculate the relative standard deviation δ S % through the above 2) to 4), until δ S % is less than 5%; if the relative standard deviation δ S % is less than 5%, the valve position deviation threshold S th is expressed as: S th = E S ± σ S .

[0139] In this embodiment, when the regulating valve parameter is the regulation rate, the calculation method of the regulation rate threshold V th is specifically as follows:

[0140] 1) Based on calculate the regulation rate V i ;

[0141] where Δt is the time interval from the start of timing when the valve position set value changes until the relative average deviation of the valve position feedback within the data window is less than 5%; ΔMV is the valve position feedback difference within the time period Δt;

[0142] In this embodiment, based on the control device to complete the calculation process of the regulation rate V i in the above 1), and thus there is no need to calculate separately, and the regulation rate V i can be directly obtained through the upper computer;

[0143] 2) Based on calculate the mean value E V of N historical regulation rates;

[0144] 3) Based on calculate the standard deviation σ V of N historical regulation rates;

[0145] 4) Based on calculate the relative standard deviation δ V %;

[0146] 5) If the relative standard deviation δ V % is greater than 5%, then based on calculate the adjusted regulation rate V w,k of the k-th group; similarly, obtain all the adjusted regulation rates V w ;

[0147] where k = {1, 2,..., N w}, N w = N - n + 1, N w is the number of groups of the currently shared data windows, and n is the number of data in each data window;

[0148] 6) Based on all the adjusted adjustment rates V w , recalculate the relative standard deviation δ V %, until δ V % is less than 5%; if the relative standard deviation δ V % is less than 5%, the adjustment rate threshold V th is expressed as: V th = E V ± σ V .

[0149] In the third embodiment, the historical data of the valve position deviation and the adjustment rate are directly used as the data support for fault diagnosis and prediction. Since the interference of human experience is reduced, the missed diagnosis and misdiagnosis of faults are also reduced; in addition, the historical data of the valve position deviation and the adjustment rate are directly obtained from the device host computer, and the first analysis result can be obtained without collecting data additionally, which reduces the workload to a certain extent.

[0150] The fourth embodiment provides a fault diagnosis and prediction device for a regulating valve, as Figure 3 shown, specifically:

[0151] In this embodiment, the fault diagnosis and prediction device includes a control device and a collection module. Among them, the control device is used to receive the parameters of the specified regulating valve collected by the collection module in real time.

[0152] In this embodiment, the control device includes a state judgment module, a data collection module, a data processing module, a data simulation module, a fault diagnosis module, and a fault prediction module.

[0153] In this embodiment, the state judgment module is used to receive the first type of parameters of the specified regulating valve sent by the collection module, and determine the state information of the regulating valve according to the first type of parameters.

[0154] In this embodiment, the data collection module is used to send a collection instruction matching the state information to the collection module according to the current state information of the regulating valve, so that the collection module performs data collection according to the collection instruction.

[0155] In this embodiment, the data processing module is used to analyze the first type of parameters and the threshold corresponding to the first type of parameters to obtain a first analysis result.

[0156] In this embodiment, the data simulation module is configured to, when the first analysis result is an abnormal result, send a collection instruction for obtaining the second type of parameters to the collection module, perform simulation processing based on the first type of parameters and the second type of parameters to obtain simulation data, and obtain a second analysis result based on the simulation data.

[0157] In this embodiment, the fault diagnosis module is configured to determine the diagnosis result of the regulating valve based on the first analysis result and the second analysis result.

[0158] In this embodiment, the fault prediction module is configured to, if the diagnosis result of the regulating valve is in a normal state, perform fitting based on the historical data of the first type of parameters to obtain prediction data, and diagnose the faults of the regulating valve in a future time period based on the prediction data.

[0159] In the fault diagnosis and prediction device described in Embodiment 4, based on the state judgment module, data collection module, data processing module, data simulation module, fault diagnosis module, and fault prediction module in the equipment management platform, the judgment of the state of the regulating valve, the collection of the parameter data of the regulating valve, the processing of the parameter data of the regulating valve, the simulation of the operating conditions of the regulating valve, the diagnosis and prediction of the faults of the regulating valve are realized.

[0160] It should be noted that in the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. In addition, it should be noted that in the description of this specification, the description of terms such as "one embodiment", "some embodiments", "embodiment", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0161] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications after learning the basic creative concepts. Therefore, the claims should be construed to include the preferred embodiments as well as all changes and modifications falling within the scope of the present invention.

[0162] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention should also include these modifications and variations.

Claims

1. A fault diagnosis and prediction method for a control valve, characterized in that, The method includes: S10. Obtain the first type of parameters of the specified regulating valve collected, and determine the status information of the regulating valve according to the first type of parameters; S20. According to the status information of the current regulating valve, send a collection instruction matching the status information to the collection module, so that the collection module performs data collection according to the collection instruction; If the regulating valve is in a stable state, send a first collection instruction to the collection module, and the first collection instruction carries a first collection frequency and an identifier of the parameter to be collected; If the regulating valve is in an operating state, send a second collection instruction to the collection module, and the second collection instruction carries a second collection frequency and an identifier of the parameter to be collected; The first collection frequency is less than the second collection frequency; S30. Analyze the first type of parameters and the threshold corresponding to the first type of parameters to obtain a first analysis result; the threshold is a threshold determined based on the historical data of the regulating valve parameters; S40. When the first analysis result is an abnormal result, send a collection instruction to obtain the second type of parameters to the collection module, and perform simulation processing based on the first type of parameters and the second type of parameters to obtain simulation data, and obtain a second analysis result based on the simulation data; S50. Determine the diagnosis result of the regulating valve based on the first analysis result and the second analysis result.

2. The method according to claim 1, wherein It further includes: S60. If the diagnosis result of the regulating valve is a normal state, perform fitting based on the historical data of the first type of parameters to obtain prediction data, and diagnose the faults of the regulating valve in the future time period based on the prediction data.

3. The method according to claim 1, wherein It further includes: The first type of parameters includes: valve position set value, valve position feedback, adjustment rate, and valve position deviation; S10 includes: All parameters collected and transmitted by the collection module carry time information when the parameters are collected; If the valve position set value changes, it is determined that the regulating valve is in an operating state; For the regulating valve in an operating state, when the change amplitude of the valve position feedback is less than a preset value, it is determined that the regulating valve enters a stable state.

4. The method according to claim 1, wherein S30 includes: S301. Obtain the thresholds of each parameter in the first type of parameters according to the historical data of the first type of parameters in a preset time period; S302. Compare the first type of parameters with the corresponding thresholds. If at least one parameter exceeds the threshold, it is determined that the first analysis result is an abnormal result.

5. The method according to claim 4, wherein S301 includes: For the valve position deviation parameter, the threshold of the valve position deviation is ; Among them, is the mean of the historical data of the valve position deviation, is the relative standard deviation corresponding to the valve position deviation; For adjusting the rate parameter, the threshold of the adjustment rate is ; Among them, is the mean value of the adjustment rate historical data, is the relative standard deviation corresponding to the adjustment rate.

6. The method according to claim 1, wherein S40 includes: The second type of parameters includes: gas supply pressure, medium temperature, valve inlet pressure, valve outlet pressure, control signal, and working mode; The simulation data includes: third type of parameter information corresponding to each time point during the simulation process, where the third type of parameter is the same parameter as the first type of parameter; Analyze the third type of parameter information and the threshold corresponding to the first type of parameter to obtain a second analysis result, where the threshold corresponding to the first type of parameter is the updated and adjusted threshold.

7. The method according to claim 6, wherein S40 includes: Compare the third type of parameter information with the threshold corresponding to the first type of parameter. If at least one parameter exceeds the threshold, it is determined that the second analysis result is an abnormal result.

8. The method according to claim 1, wherein S50 includes: If the first analysis result is a normal result, determine that the regulating valve is in a normal state; If the first analysis result is an abnormal result and the second analysis result is a normal result, determine that the regulating valve is in a faulty state; If both the first analysis result and the second analysis result are abnormal results, adjust the parameter threshold of the regulating valve and repeat the steps of S30.

9. A device for fault diagnosis and prediction of a regulating valve, characterized in that the device includes a control device and a collection module, and the control device is used to receive the parameters of a specified regulating valve collected in real time by the collection module; the control device includes: a state judgment module, configured to receive the first type of parameters of the specified regulating valve sent by the collection module, and determine the state information of the regulating valve according to the first type of parameters; a data collection module, configured to send a collection instruction matching the state information to the collection module according to the current state information of the regulating valve, so that the collection module performs data collection according to the collection instruction; if the regulating valve is in a stable state, send a first collection instruction to the collection module, and the first collection instruction carries a first collection frequency and an identifier of the parameter to be collected; if the regulating valve is in an operating state, send a second collection instruction to the collection module, and the second collection instruction carries a second collection frequency and an identifier of the parameter to be collected; the first collection frequency is less than the second collection frequency; a data processing module, configured to analyze the first type of parameters and the threshold corresponding to the first type of parameters to obtain a first analysis result; the threshold is a threshold determined based on the historical data of the regulating valve parameters; a data simulation module, configured to send a collection instruction for obtaining the second type of parameters to the collection module when the first analysis result is an abnormal result, perform simulation processing based on the first type of parameters and the second type of parameters to obtain simulation data, and obtain a second analysis result based on the simulation data; a fault diagnosis module, configured to determine the diagnosis result of the regulating valve based on the first analysis result and the second analysis result; a fault prediction module, configured to, if the diagnosis result of the regulating valve is a normal state, perform fitting based on the historical data of the first type of parameters to obtain prediction data, and diagnose the regulating valve fault in a future time period based on the prediction data.

Images