A data-driven control valve intelligent diagnosis method and system

Abstract

The present invention relates to the field of control valve diagnosis technology, and specifically discloses a data-driven control valve intelligent diagnosis method and system, the intelligent diagnosis method comprising: obtaining control valve operating status data detected by multiple sensors, wherein the control valve operating status data includes control valve stroke data and control valve pressure data; diagnosing multiple fault items of the control valve according to the control valve operating status data to obtain multiple fault item diagnosis results of the control valve; and outputting multiple fault item diagnosis results of the control valve. The data-driven control valve intelligent diagnosis method provided by the present invention can obtain in-depth information on the operating status of the control valve and generate a diagnosis report to monitor and identify the fault condition of the valve, thereby reducing the error of the fault diagnosis result.

Description

Technical Field

[0001] The present invention relates to the technical field of control valve diagnosis, and more specifically, to a data-driven intelligent diagnosis method for control valves and a data-driven intelligent diagnosis system for control valves. Background Art

[0002] Control valves are the most widely used type of final control element. The safety and reliability of valves during use play an important role in the personal safety of company employees and production efficiency. In the intelligent automation control system of modern factories, timely and accurate diagnosis of control valves is the guarantee for the safe and orderly operation of the factory. However, most factories rely too much on traditional diagnostic methods, manual experience, or solve problems only after they occur, resulting in large errors in fault diagnosis results. Summary of the invention

[0003] In order to solve the deficiencies in the prior art, the present invention provides a data-driven control valve intelligent diagnosis method and system, which can reduce the error of fault diagnosis results.

[0004] As a first aspect of the present invention, a data-driven control valve intelligent diagnosis method is provided, comprising: Step S1: Acquire control valve operation status data detected by multiple sensors, wherein the control valve operation status data includes control valve stroke data and control valve pressure data; Step S2: diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain diagnosis results of the multiple fault items of the control valve; Step S3: outputting the diagnosis results of multiple fault items of the control valve.

[0005] Furthermore, before performing step S2, the method further includes: Calibrate the stroke sensor of the control valve according to the control valve operation status data, and after the stroke sensor is calibrated, execute step S2; wherein the process of calibrating the stroke sensor of the control valve includes: (1) Send a 4mA current signal to the control valve and keep it for 10 seconds; then record the control valve stroke data X1 detected by the stroke sensor at this time; (2) Send a 20mA current signal to the control valve and keep it for 10s; then record the control valve stroke data X2 detected by the stroke sensor at this time; (3) Calculate the actual travel data ΔX of the control valve. The calculation formula is as follows: ΔX=X2-X1 (4) Calculate the rated travel deviation data of the control valve. The calculation formula is as follows: Rated stroke deviation data = actual stroke data ΔX / rated stroke Among them, the rated stroke is the inherent property parameter of the control valve.

[0006] Further, the diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: Performing IP hysteresis diagnosis of the control valve according to the control valve operation status data to obtain the IP hysteresis diagnosis result of the control valve; wherein the process of IP hysteresis diagnosis of the control valve includes: (1) Send a 7.2mA current signal to the control valve to make the control valve opening reach 20%, and then determine whether the control valve pressure data detected by the pressure sensor at this time meets the preset pressure range; if so, maintain it for 10 seconds and record the control valve pressure data when the control valve opening is 20%; (2) Perform diagnosis on the four measuring points of the control valve opening of 35%, 50%, 65%, and 80% in sequence according to the method in (1) above; (3) Perform diagnosis in sequence at five measuring points with control valve openings of 80%, 65%, 50%, 35%, and 20% in the manner described in (1) above; (4) Calculate the hysteresis and repeatability error values ​​to obtain the IP hysteresis diagnosis results of the control valve.

[0007] Further, the diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: A resolution diagnosis of the control valve is performed according to the control valve operation status data to obtain a resolution diagnosis result of the control valve; wherein the process of the resolution diagnosis of the control valve includes: (1) Send a 12mA current signal to the control valve and keep it for 15s. After the control valve output is stable, determine whether the valve position is equal to 0.5*rated stroke. If so, enter the upstream current value of the first measuring point and keep it for the set stabilization time. Then enter the downstream current value of the first measuring point and keep it for the set stabilization time. Otherwise, check again. (2) The following measurement points are also measured up and down until all measurement points are completed.

[0008] Further, the diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: A hysteresis diagnosis of the control valve is performed according to the control valve operation status data to obtain a hysteresis diagnosis result of the control valve; wherein the process of the hysteresis diagnosis of the control valve includes: (1) Send a 5.6 mA current signal to the control valve, so that the control valve opening reaches 10% and is maintained for 10 seconds. Then, the control valve stroke data detected by the stroke sensor is recorded. (2) Perform diagnosis in sequence at eight measuring points where the control valve opening is 20%, 30%, 40%, 50%, 60%, 70%, 80%, and 90% in the manner described in (1) above; (3) Perform diagnosis in turn at nine measuring points with control valve openings of 90%, 80%, 70%, 60%, 50%, 40%, 30%, 20%, and 10% in the manner described in (1) above; (4) Calculate the hysteresis and repeatability error values ​​to obtain the hysteresis diagnosis results of the control valve.

[0009] Further, the diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: A step response diagnosis of the control valve is performed based on the control valve operating status data to obtain a step response diagnosis result of the control valve; wherein the process of the step response diagnosis of the control valve includes: recording the control valve stroke data detected by the stroke sensor under each control valve opening during the entire step process, and calculating the deviation value, hysteresis value, the time T63 required for the control valve to run from 0 to 63%, the time T86 required for the control valve to run from 0 to 86%, basic error, hysteresis and overshoot to obtain a step response diagnosis result of the control valve.

[0010] Further, the diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: A frequency response diagnosis of the control valve is performed according to the control valve operation status data to obtain a frequency response diagnosis result of the control valve; wherein the process of the frequency response diagnosis of the control valve includes: (1) Click Start, substitute t as 0 and calculate the input current value, i.e. the initial valve position current value. After the initial valve position is stable, check whether the valve position is within the allowable error. If so, proceed to (2), otherwise proceed to check; (2) Increase t by 0.1 and substitute the input current value. Each time the valve position moves to within the allowable error or exceeds the minimum valve position error value, the time difference between this arrival time and the time when the input current is input is taken as the increased seconds. Keep one decimal place, record the seconds of arrival and the percentage of arrival, and calculate. Keep two decimal places, calculate the phase shift value, and convert the increased seconds into readings. The calculation formula is 360°*frequency value*increased seconds. (3) Calculate the value of t to be increased according to the number of cycles and frequency. For example, if the frequency is 0.1 and the number of cycles is 10, then t should be increased by 0.1 to 100, that is, 10 / 0.1=100. (4) After one frequency is completed, increase the frequency by 0.1 Hz. The formula becomes 12 + 1.2 * sin (360 ° * 0.2 * t). Continue to substitute t until the number of seconds is 100 + 10 / 0.2 = 150s. Repeat until the cutoff frequency.

[0011] Further, the diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: According to the control valve operation status data, start and end point deviation diagnosis, dead zone diagnosis and stroke diagnosis of the control valve are respectively performed to obtain the start and end point deviation diagnosis result, dead zone diagnosis result and stroke diagnosis result of the control valve accordingly; The process of diagnosing the start and end point deviation of the control valve includes: calculating the start point deviation and the end point deviation to obtain the diagnosis result of the start and end point deviation of the control valve; The process of dead zone diagnosis of the control valve includes: conducting tests at three measuring points of 25%, 50% and 75% of the input signal range; taking the largest dead zone value among the three measuring points of 25%, 50% and 75% as the dead zone value of the control valve to obtain the dead zone diagnosis result of the control valve; The process of the stroke diagnosis of the control valve includes: controlling the current signal to transfer from 3mA to 21mA and then back to 3mA, and simultaneously monitoring the control valve stroke data and the control valve pressure data to obtain the control valve stroke diagnosis result.

[0012] As a second aspect of the present invention, a data-driven control valve intelligent diagnosis system is provided, comprising: A diagnostic analysis module, used to obtain control valve operation status data detected by multiple sensors, wherein the control valve operation status data includes control valve stroke data and control valve pressure data; and diagnose multiple fault items of the control valve according to the control valve operation status data to obtain diagnosis results of the multiple fault items of the control valve; The data sending and receiving module is used to output the diagnosis results of multiple fault items of the control valve.

[0013] Furthermore, it also includes a sensor calibration module, a data storage module, a real-time data module and a historical data module; wherein the sensor calibration module is used to calibrate the travel sensor of the control valve according to the control valve operating status data; the data storage module is used to store data; the real-time data module is used to display the fault item diagnosis results in real time; the historical data module is used to view the historical fault item diagnosis results.

[0014] The data-driven control valve intelligent diagnosis method and system provided by the present invention have the following advantages: through software and hardware design, a variety of sensors collect process data of the valve operating status in real time, and then transmit the data to the diagnosis platform, effectively utilize the data, dig out the hidden information of the control valve itself from aspects such as pressure, stroke, temperature, etc., and combine the intelligent diagnosis software to analyze the data, so as to obtain in-depth information on the operating status of the control valve, and generate a diagnosis report to monitor and identify the fault condition of the valve, thereby reducing the error of the fault diagnosis result. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present invention, but do not constitute a limitation of the present invention.

[0016] Figure 1 A flow chart of the data-driven control valve intelligent diagnosis method provided by the present invention.

[0017] Figure 2 A flow chart of the stroke sensor calibration provided by the present invention.

[0018] Figure 3A-3H A diagnostic flowchart of multiple fault items provided by the present invention.

[0019] Figure 4 This is a structural diagram of the data-driven control valve intelligent diagnosis system provided by the present invention.

[0020] Figure 5 This is a workflow diagram of the data-driven control valve intelligent diagnosis system provided by the present invention. DETAILED DESCRIPTION

[0021] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation method, structure, characteristics and effects of the data-driven control valve intelligent diagnosis method and system proposed by the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments. Obviously, the described embodiments are part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the protection scope of the present invention.

[0022] In this embodiment, a data-driven control valve intelligent diagnosis method is provided. Figure 1 As shown, the data-driven control valve intelligent diagnosis method includes: Step S1: Acquire control valve operation status data detected by multiple sensors, wherein the control valve operation status data includes control valve stroke data and control valve pressure data; Step S2: diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain diagnosis results of the multiple fault items of the control valve; Preferably, if Figure 2 As shown, before performing step S2, it also includes: Calibrate the stroke sensor of the control valve according to the control valve operation status data, and after the stroke sensor is calibrated, execute step S2; wherein the process of calibrating the stroke sensor of the control valve includes: (1) Send a 4mA current signal to the control valve and keep it for 10 seconds; then record the control valve stroke data X1 detected by the stroke sensor at this time; (2) Send a 20mA current signal to the control valve and keep it for 10s; then record the control valve stroke data X2 detected by the stroke sensor at this time; (3) Calculate the actual travel data ΔX of the control valve. The calculation formula is as follows: ΔX=X2-X1 (4) Calculate the rated travel deviation data of the control valve. The calculation formula is as follows: Rated stroke deviation data = actual stroke data ΔX / rated stroke Among them, the rated stroke is the inherent property parameter of the control valve, and the holding time can be set manually.

[0023] Preferably, if Figure 3A As shown, the method of diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: Performing IP hysteresis diagnosis of the control valve according to the control valve operation status data to obtain the IP hysteresis diagnosis result of the control valve; wherein the process of IP hysteresis diagnosis of the control valve includes: (1) Send a 7.2mA current signal to the control valve to make the control valve opening reach 20%, and then determine whether the control valve pressure data detected by the pressure sensor at this time meets the preset pressure range; if so, maintain it for 10 seconds and record the control valve pressure data when the control valve opening is 20%; (2) Perform diagnosis on the four measuring points of the control valve opening of 35%, 50%, 65%, and 80% in sequence according to the method in (1) above; (3) Perform diagnosis in sequence at five measuring points with control valve openings of 80%, 65%, 50%, 35%, and 20% in the manner described in (1) above; (4) Calculate the hysteresis and repeatability error values ​​to obtain the IP hysteresis diagnosis results of the control valve.

[0024] It should be noted that IP hysteresis diagnosis is used to determine the accuracy and repeatability of the pressure sensor driving the valve. A series of equal step cycles will be performed between 0% and 100% of the valve opening, and the calculation method is the same as the hysteresis test.

[0025] Specifically, the valve opening is tested at five points: 20%, 35%, 50%, 65%, and 80%. The entire process displays the IP pressure percentage (up and down) of the output valve - input signal curve, and the hysteresis and repeatability are calculated.

[0026] Preferably, if Figure 3B As shown, the method of diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: A resolution diagnosis of the control valve is performed according to the control valve operation status data to obtain a resolution diagnosis result of the control valve; wherein the process of the resolution diagnosis of the control valve includes: (1) Set according to the frequency point; send a 12mA current signal to the control valve and keep it for 15s. After the control valve output is stable, determine whether the valve position is equal to 0.5*rated stroke. If so, enter the up-current value of the first measuring point and keep the set stabilization time, then enter the down-current value of the first measuring point and keep the set stabilization time; otherwise, check; (2) The following measurement points are also measured up and down until all measurement points are completed.

[0027] It should be noted that during the entire process of resolution diagnosis, the input current signal (starting position), stabilization time and change rate per step can be set manually.

[0028] Flow chart of resolution diagnosis: X-axis is time point, Y-axis is dual-axis display, the left side is the opening percentage, unit %, and the right side is the input current, unit mA; Resolution diagnosis step point description: For example, a 5% step point means an increase of (20-4)*0.05=0.8mA. If the starting point is 12mA, the upstream input current is 12+0.8=12.8mA, and the downstream input current is 12-0.8=11.2mA.

[0029] Preferably, if Figure 3C As shown, the method of diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: A hysteresis diagnosis of the control valve is performed according to the control valve operation status data to obtain a hysteresis diagnosis result of the control valve; wherein the process of the hysteresis diagnosis of the control valve includes: (1) Send a 5.6 mA current signal to the control valve, so that the control valve opening reaches 10% and is maintained for 10 seconds. Then, the control valve stroke data detected by the stroke sensor is recorded. (2) Perform diagnosis in sequence at eight measuring points where the control valve opening is 20%, 30%, 40%, 50%, 60%, 70%, 80%, and 90% in the manner described in (1) above; (3) Perform diagnosis in turn at nine measuring points with control valve openings of 90%, 80%, 70%, 60%, 50%, 40%, 30%, 20%, and 10% in the manner described in (1) above; (4) Calculate the hysteresis and repeatability error values ​​to obtain the hysteresis diagnosis results of the control valve.

[0030] It should be noted that the hysteresis diagnosis: the valve opening is diagnosed at 9 points such as 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, etc. The whole process displays the curve data such as stroke percentage and input signal, and calculates the hysteresis value and repeatability error value. Note: The test point, holding time, and number of cycles can be set manually.

[0031] Among them, the return difference = |a–b|: After one up-and-down cycle, the absolute value of the difference between the valve range percentage a when going from bottom to top and the valve range percentage b when going from top to bottom at each test point; Wherein, repeatability error value = |Maxv-Minv|: the absolute value of the difference between the maximum value Maxv and the minimum value Minv of the valve position range percentage of each measuring point after all back and forth cycles are completed.

[0032] Preferably, if Figure 3D As shown, the method of diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: The step response diagnosis of the control valve is performed according to the control valve operation status data to obtain the step response diagnosis result of the control valve; wherein, the process of the step response diagnosis of the control valve includes: recording the control valve stroke data detected by the stroke sensor under each control valve opening during the entire step process, and calculating the deviation value, hysteresis value, the time T63 required for the control valve to run from 0 to 63%, the time T86 required for the control valve to run from 0 to 86%, basic error, hysteresis and overshoot to obtain the step response diagnosis result of the control valve. Specifically as follows: Deviation value: The difference between the valve position and the theoretical valve position after the valve is at the set point and maintains stability for a certain period of time; Hysteresis value: the time from when the input signal changes to when the valve position starts to change; T63: The time difference (including hysteresis) from the input signal change to the actual valve position reaching 63.2% of the stable position; T86: The time difference (including hysteresis) from the input signal change to the actual valve position reaching 86.5% of the stable position; Basic error: MAX [(actual stroke of each measuring point - difference of theoretical stroke) / rated stroke × 100%]; Hysteresis: After one up-and-down cycle, the absolute value of the percentage of the valve range when each measuring point moves from bottom to top - the percentage of the valve range when it moves from top to bottom; Overshoot: The value by which the actual stroke exceeds the theoretical stroke.

[0033] Preferably, if Figure 3E As shown, the method of diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: A frequency response diagnosis of the control valve is performed according to the control valve operation status data to obtain a frequency response diagnosis result of the control valve; wherein the process of the frequency response diagnosis of the control valve includes: (1) Click Start, substitute t as 0 and calculate the input current value, i.e. the initial valve position current value. After the initial valve position is stable, check whether the valve position is within the allowable error. If so, proceed to (2), otherwise proceed to check; (2) Increase t by 0.1 and substitute the input current value. Each time the valve position moves to within the allowable error or exceeds the minimum valve position error value, the time difference between this arrival time and the time when the input current is input is taken as the increased seconds. Keep one decimal place, record the seconds of arrival (t + increased seconds) and the percentage of arrival position, and calculate, keep two decimal places, calculate the phase shift value, and convert the increased seconds into readings. The calculation formula is 360°*frequency value*increased seconds. (3) Calculate the value of t to be increased according to the number of cycles and frequency. For example, if the frequency is 0.1 and the number of cycles is 10, then t should be increased by 0.1 to 100, that is, 10 / 0.1=100. (4) After one frequency is completed, increase the frequency by 0.1 Hz. The formula becomes 12 + 1.2 * sin (360 ° * 0.2 * t). Continue to substitute t until the number of seconds is 100 + 10 / 0.2 = 150s. Repeat until the cutoff frequency.

[0034] It should be noted that in frequency response diagnosis: Cut-off frequency: The upper limit of frequency change. After one frequency is tested, the frequency is increased by 0.1Hz until the cut-off frequency is reached. The current input value calculation formula is: x+y*sin(360°*z*t), x is the input current value of the initial setting valve position; y is the maximum input amplitude; z is the frequency value; t is the time, the second; for example: 12+1.2*sin(360°*0.1*t), the current value is rounded to 3 decimal places.

[0035] The gain value formula is: 20*log10(output % / input %).

[0036] Preferably, if Figure 3F-3H As shown, the method of diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: According to the control valve operation status data, start and end point deviation diagnosis, dead zone diagnosis and stroke diagnosis of the control valve are respectively performed to obtain the start and end point deviation diagnosis result, dead zone diagnosis result and stroke diagnosis result of the control valve accordingly; Among them, Figure 3F As shown, the process of diagnosing the start and end point deviation of the control valve includes: calculating the start point deviation and the end point deviation to obtain the diagnosis result of the start and end point deviation of the control valve; It should be noted that the deviation diagnosis of the starting and ending points: the step signal, stabilization time, and the standard of the change amount can be set manually during the whole process, and the starting and ending pressure values ​​are recorded. Note: The current is converted into a percentage.

[0037] Among them, Figure 3G As shown, the process of dead zone diagnosis of the control valve includes: performing tests at three measuring points of 25%, 50%, and 75% of the input signal range; taking the largest dead zone value among the three measuring points of 25%, 50%, and 75% as the dead zone value of the control valve, so as to obtain the dead zone diagnosis result of the control valve; Example: (1) At the 25% measurement point, slowly change (increase or decrease) the input signal and hold it for 10 seconds (manually settable) until a noticeable displacement change of the valve is detected. Record the input signal value a at this time (which may be larger or smaller than 25%). There is no need to return to the measurement point 25%. Input the signal in the opposite direction (decrease or increase) and hold it for 10 seconds (manually settable) until a noticeable displacement change of the valve is detected. Record the input signal value b at this time (which may be larger or smaller than 25%). (2) The upward dead zone value of the 25% point is equal to the absolute value of the difference between a and b. The downward dead zone value is calculated in the same way. Finally, the maximum value of the two is taken as the dead zone value of the 25% measurement point. (3) The measurement method of 50% and 75% is the same as that of the 25% measurement point. (4) Finally, the largest dead zone value among the three measurement points is taken as the dead zone value of the valve.

[0038] It should be noted that the measurement point, input signal, perceptible displacement change, and holding time are all parameters that can be set manually.

[0039] Among them, Figure 3H As shown, the process of the stroke diagnosis of the control valve includes: the control current signal is transferred from 3mA to 21mA and returned to 3mA, and the control valve stroke data and the control valve pressure data are monitored at the same time to obtain the stroke diagnosis result of the control valve. Among them, the control signal, position and pressure are all plotted in two directions. At the end of the test, the friction curve is calculated and plotted. Note: The holding time can be set manually.

[0040] Step S3: outputting the diagnosis results of multiple fault items of the control valve.

[0041] The data-driven intelligent diagnosis method for control valves provided by the present invention uses a variety of sensors to collect process data of the valve operating status in real time, and then transmits the data to a diagnosis platform, effectively utilizing the data to mine the hidden information of the control valve itself from aspects such as pressure, stroke, and temperature, and combines the intelligent diagnosis software to analyze the data, thereby obtaining in-depth information on the operating status of the control valve, and generating a diagnosis report to monitor and identify the fault condition of the valve, thereby reducing the error of the fault diagnosis result.

[0042] As another embodiment of the present invention, a data-driven control valve intelligent diagnosis system is provided. Figure 4 As shown, the data-driven control valve intelligent diagnosis system includes: A diagnostic analysis module, used to obtain control valve operation status data detected by multiple sensors, wherein the control valve operation status data includes control valve stroke data and control valve pressure data; and diagnose multiple fault items of the control valve according to the control valve operation status data to obtain diagnosis results of the multiple fault items of the control valve; The data sending and receiving module is used to output the diagnosis results of multiple fault items of the control valve.

[0043] Preferably, it also includes a sensor calibration module, a data storage module, a real-time data module and a historical data module; wherein the sensor calibration module is used to calibrate the stroke sensor of the control valve according to the control valve operating status data; the data storage module is used to store data; the real-time data module is used to display the fault item diagnosis results in real time; the historical data module is used to view the historical fault item diagnosis results.

[0044] like Figure 5 As shown, the working process of the data-driven control valve intelligent diagnosis system of the present invention is as follows: (1) The diagnostic system issues diagnostic instructions; (2) After receiving the diagnostic instruction, the acquisition card collects the detection data of the relevant sensors; (3) The acquisition card sends the collected test data to the diagnostic system; (4) When the diagnosis begins, the diagnostic system receives and analyzes the test data in real time, and visually displays the diagnostic process curve chart data through the PC or tablet; (5) After the diagnosis is completed, the diagnostic system visually displays the diagnostic result curve chart data through the PC or tablet according to the diagnostic rules; (6) The diagnostic system stores diagnostic process curve chart data and diagnostic result curve chart data, and issues a diagnostic report.

[0045] It should be noted that, based on the diagnostic process curve chart data and the diagnostic result curve chart data, a preliminary diagnosis can be made on the positioner, pressure reducing valve, and cylinder problems, and a diagnosis can also be made on the valve core and seat, valve sticking, packing friction, actuator leakage, gas line leakage, etc.

[0046] The data-driven intelligent diagnosis system for control valves provided by the present invention has a hardware part which is a set of independently designed diagnostic equipment, which is composed of a pressure sensor, a valve position sensor, a data acquisition card, a data transmission line, etc. Among them, the acquisition card device adopts a dot matrix LCD screen, which is a DC voltage measurement terminal and is a highly integrated product for DC power parameter measurement applications. At the same time, the device has an RS485 interface and supports the MODBUS-RTU communication protocol, and can upload the measured and collected data to the main server. The software part is an intelligent diagnosis software independently developed based on the web, mainly based on the development environment of the c# net5.0 system, from the front-end data visualization to the data interface layer and then to the data access layer, and finally the software program is deployed to the PC or tablet end, and the intelligent diagnosis software can be used for diagnosis.

[0047] The data-driven intelligent control valve diagnosis system provided by the present invention combines data acquisition, storage services, data analysis and graphic visualization services to realize automatic display of test process and results. The supporting service program for collecting and storing data acquisition storage services uses data acquisition cards to collect data, analyzes, calculates and displays curves on the collected data, supports simulated visualization of system data, and uses database management for data storage; supports image display of real-time icons, and supports data services such as queries for multiple needs.

[0048] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this profession can make some changes or modifications to equivalent embodiments of equivalent changes using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention still falls within the scope of the technical solution of the present invention.

Claims

1. A data-driven intelligent diagnosis method for control valves, characterized in that: include: Step S1: Acquire control valve operation status data detected by multiple sensors, wherein the control valve operation status data includes control valve stroke data and control valve pressure data; Step S2: diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain diagnosis results of the multiple fault items of the control valve; Step S3: outputting the diagnosis results of multiple fault items of the control valve.

2. The data-driven intelligent diagnosis method for control valves according to claim 1 is characterized in that: Before performing step S2, the method further includes: Calibrate the stroke sensor of the control valve according to the control valve operation status data, and after the stroke sensor is calibrated, execute step S2; wherein the process of calibrating the stroke sensor of the control valve includes: (1) Send a 4mA current signal to the control valve and keep it for 10 seconds; then record the control valve stroke data X1 detected by the stroke sensor at this time; (2) Send a 20mA current signal to the control valve and keep it for 10s; then record the control valve stroke data X2 detected by the stroke sensor at this time; (3) Calculate the actual travel data ΔX of the control valve. The calculation formula is as follows: ΔX=X2-X1 (4) Calculate the rated travel deviation data of the control valve. The calculation formula is as follows: Rated stroke deviation data = actual stroke data ΔX / rated stroke Among them, the rated stroke is the inherent property parameter of the control valve.

3. The data-driven intelligent control valve diagnosis method according to claim 1 is characterized in that: The diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: Performing IP hysteresis diagnosis of the control valve according to the control valve operation status data to obtain the IP hysteresis diagnosis result of the control valve; wherein the process of IP hysteresis diagnosis of the control valve includes: (1) Send a 7.2mA current signal to the control valve to make the control valve opening reach 20%, and then determine whether the control valve pressure data detected by the pressure sensor at this time meets the preset pressure range; if so, maintain it for 10 seconds and record the control valve pressure data when the control valve opening is 20%; (2) Perform diagnosis on the four measuring points of the control valve opening of 35%, 50%, 65%, and 80% in sequence according to the method in (1) above; (3) Perform diagnosis in sequence at five measuring points with control valve openings of 80%, 65%, 50%, 35%, and 20% in the manner described in (1) above; (4) Calculate the hysteresis and repeatability error values ​​to obtain the IP hysteresis diagnosis results of the control valve.

4. The data-driven intelligent diagnosis method for control valves according to claim 1 is characterized in that: The diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: A resolution diagnosis of the control valve is performed according to the control valve operation status data to obtain a resolution diagnosis result of the control valve; wherein the process of the resolution diagnosis of the control valve includes: (1) Send a 12mA current signal to the control valve and keep it for 15s. After the control valve output is stable, determine whether the valve position is equal to 0.5*rated stroke. If so, enter the upstream current value of the first measuring point and keep it for the set stabilization time. Then enter the downstream current value of the first measuring point and keep it for the set stabilization time. Otherwise, check again. (2) The following measurement points are also measured up and down until all measurement points are completed.

5. The data-driven intelligent diagnosis method for control valves according to claim 1 is characterized in that: The diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: A hysteresis diagnosis of the control valve is performed according to the control valve operation status data to obtain a hysteresis diagnosis result of the control valve; wherein the process of the hysteresis diagnosis of the control valve includes: (1) Send a 5.6 mA current signal to the control valve, so that the control valve opening reaches 10% and is maintained for 10 seconds. Then, the control valve stroke data detected by the stroke sensor is recorded. (2) Perform diagnosis in sequence at eight measuring points where the control valve opening is 20%, 30%, 40%, 50%, 60%, 70%, 80%, and 90% in the manner described in (1) above; (3) Perform diagnosis in turn at nine measuring points with control valve openings of 90%, 80%, 70%, 60%, 50%, 40%, 30%, 20%, and 10% in the manner described in (1) above; (4) Calculate the hysteresis and repeatability error values ​​to obtain the hysteresis diagnosis results of the control valve.

6. The data-driven intelligent diagnosis method for control valves according to claim 1 is characterized in that: The diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: A step response diagnosis of the control valve is performed based on the control valve operating status data to obtain a step response diagnosis result of the control valve; wherein the process of the step response diagnosis of the control valve includes: recording the control valve stroke data detected by the stroke sensor under each control valve opening during the entire step process, and calculating the deviation value, hysteresis value, the time T63 required for the control valve to run from 0 to 63%, the time T86 required for the control valve to run from 0 to 86%, basic error, hysteresis and overshoot to obtain a step response diagnosis result of the control valve.

7. The data-driven intelligent diagnosis method for control valves according to claim 1 is characterized in that: The diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: A frequency response diagnosis of the control valve is performed according to the control valve operation status data to obtain a frequency response diagnosis result of the control valve; wherein the process of the frequency response diagnosis of the control valve includes: (1) Click Start, substitute t as 0 and calculate the input current value, i.e. the initial valve position current value. After the initial valve position is stable, check whether the valve position is within the allowable error. If so, proceed to (2), otherwise proceed to check; (2) Increase t by 0.1 and substitute the input current value. Each time the valve position moves to within the allowable error or exceeds the minimum valve position error value, the time difference between this arrival time and the time when the input current is input is taken as the increased seconds. Keep one decimal place, record the seconds of arrival and the percentage of arrival, and calculate. Keep two decimal places, calculate the phase shift value, and convert the increased seconds into readings. The calculation formula is 360°*frequency value*increased seconds. (3) Calculate the value of t to be increased according to the number of cycles and frequency. For example, if the frequency is 0.1 and the number of cycles is 10, then t should be increased by 0.1 to 100, that is, 10 / 0.1=100. (4) After one frequency is completed, increase the frequency by 0.1 Hz. The formula becomes 12 + 1.2 * sin (360 ° * 0.2 * t). Continue to substitute t until the number of seconds is 100 + 10 / 0.2 = 150s. Repeat until the cutoff frequency.

8. The data-driven intelligent diagnosis method for control valves according to claim 1 is characterized in that: The diagnosing multiple fault items of the control valve according to the control valve operation status data to obtain multiple fault item diagnosis results of the control valve also includes: According to the control valve operation status data, start and end point deviation diagnosis, dead zone diagnosis and stroke diagnosis of the control valve are respectively performed to obtain the start and end point deviation diagnosis result, dead zone diagnosis result and stroke diagnosis result of the control valve accordingly; The process of diagnosing the start and end point deviation of the control valve includes: calculating the start point deviation and the end point deviation to obtain the diagnosis result of the start and end point deviation of the control valve; The process of dead zone diagnosis of the control valve includes: conducting tests at three measuring points of 25%, 50% and 75% of the input signal range; taking the largest dead zone value among the three measuring points of 25%, 50% and 75% as the dead zone value of the control valve to obtain the dead zone diagnosis result of the control valve; The process of the stroke diagnosis of the control valve includes: controlling the current signal to transfer from 3mA to 21mA and then back to 3mA, and simultaneously monitoring the control valve stroke data and the control valve pressure data to obtain the control valve stroke diagnosis result.

9. A data-driven control valve intelligent diagnosis system, used to implement the data-driven control valve intelligent diagnosis method according to any one of claims 1 to 8, characterized in that: The data-driven control valve intelligent diagnosis system includes: A diagnostic analysis module, used to obtain control valve operation status data detected by multiple sensors, wherein the control valve operation status data includes control valve stroke data and control valve pressure data; and diagnose multiple fault items of the control valve according to the control valve operation status data to obtain diagnosis results of the multiple fault items of the control valve; The data sending and receiving module is used to output the diagnosis results of multiple fault items of the control valve.

10. The data-driven control valve intelligent diagnosis system according to claim 9, characterized in that: It also includes a sensor calibration module, a data storage module, a real-time data module and a historical data module; wherein the sensor calibration module is used to calibrate the travel sensor of the control valve according to the control valve operating status data; the data storage module is used to store data; the real-time data module is used to display the fault item diagnosis results in real time; and the historical data module is used to view the historical fault item diagnosis results.

Images