State online monitoring method for steam isolation valve system

Abstract

The invention discloses a state online monitoring method for a steam isolation valve system, and belongs to the technical field of valve fault diagnosis. The method specifically comprises the following steps: 1) constructing a steam isolation valve system fault tree and associating corresponding monitoring quantities; 2) taking data when a system leaves a factory as reference data, performing segmentation processing, and judging whether the data has tendency, periodicity and autocorrelation; 3) combining working conditions and technical requirements to obtain a first threshold value of partial monitoring quantity; acquiring a tendency threshold, a periodicity threshold or an autocorrelation threshold for the monitoring quantity with tendency, periodicity or autocorrelation; calculating a variance and a mean value of multiple groups of reference data of all the monitoring quantities to obtain a second threshold value; 4) comparing the data in the actual operation process with a corresponding threshold value, and judging whether abnormity occurs or not; and 5) associating and indexing the abnormal monitoring quantity and the fault tree event, and outputting a fault diagnosis result. According to the method, specific faults of the system can be rapidly detected through online monitoring, and safe and stable operation of the system is guaranteed.

Description

Technical Field

[0001] The invention belongs to the technical field of valve fault diagnosis, and in particular relates to an online status monitoring method for a steam isolation valve system. Background Art

[0002] Steam isolation valves are important safety equipment in nuclear power plants. In the event of an accident such as a steam pipeline rupture, the steam isolation valve must be quickly closed within 3 seconds. If a steam isolation valve fails during operation, it will cause major events such as unit shutdown, seriously threatening the reliability of pipeline components. Its safe and stable operation has a significant impact on the safety and economic benefits of nuclear power plants.

[0003] Since the steam isolation valve system involves complex gas-liquid linkage actuators, has many hydraulic equipment, complex control systems, multiple working modes, special working environments, insufficient data and other characteristics, it is difficult to diagnose the faults of the steam isolation valve system. At present, the domestic foundation in the research, fault diagnosis, and professional detection technology of steam isolation valve systems is relatively weak. The actual detection mostly relies on the monthly or quarterly maintenance of nuclear power plants, and it is necessary to disassemble and test the components, which is time-consuming and labor-intensive, and it is difficult to conduct a comprehensive evaluation of their performance.

[0004] Therefore, giving the steam isolation valve the ability to perceive its own status and fault environment, and then realize online fault diagnosis, is a problem that needs to be solved in the operation and maintenance of steam isolation valves today. Summary of the invention

[0005] The object of the present invention is to provide a method for online monitoring of the status of a steam isolation valve system in view of the problem that the existing online monitoring and fault diagnosis methods of the steam isolation valve system are imperfect.

[0006] The specific technical solutions adopted by the present invention are as follows:

[0007] The present invention provides a method for online monitoring of the status of a steam isolation valve system, and the specific steps are as follows:

[0008] S1: Construct a fault tree for the steam isolation valve system and list all potential faults; associate the top event, intermediate event and bottom event of the fault tree with corresponding monitoring quantities; the monitoring quantities include the monitoring quantity of the gas-liquid linkage actuator and the monitoring quantity of the steam isolation valve body;

[0009] S2: taking the data of the monitoring quantity described in step S1 when the steam isolation valve system leaves the factory as the benchmark data; taking the time point of the control signal change as the point of change of the working state of the steam isolation valve system, and segmenting the benchmark data according to the time point; and performing feature classification on the benchmark data of each monitoring quantity segmentation according to basic characteristics; the basic characteristics include trend, periodicity and autocorrelation;

[0010] S3: Combine the working condition requirements and technical requirements to obtain the fixed threshold of some monitoring quantities as the first threshold of the monitoring quantity; if the benchmark data of the monitoring quantity has trend, periodicity or autocorrelation, add a certain margin on the basis of the trend index, periodicity peak or autocorrelation index of the benchmark data, and use them as the trend threshold, periodicity threshold or autocorrelation threshold of the monitoring quantity respectively; calculate the variance and mean of multiple groups of benchmark data of all monitoring quantities, use the mean as the benchmark and the variance as the margin, and obtain the second threshold of the monitoring quantity;

[0011] S4: pre-processing the monitoring quantity monitored during the actual operation of the steam isolation valve system to obtain pre-processed monitoring quantity data;

[0012] S5: Calculate the trend index, periodic peak value and autocorrelation index of the monitoring quantity data after different preprocessing obtained in step S4;

[0013] S6: According to different working states of the steam isolation valve system, the pre-processed monitoring data is compared with the first threshold value or the second threshold value, and the trend index, periodic peak value or autocorrelation index calculated in step S5 is compared with the corresponding trend threshold value, periodic threshold value or autocorrelation threshold value; if any data falls outside the corresponding threshold range, it is determined that the steam isolation valve system is abnormal, otherwise it is determined that the steam isolation valve system is operating normally;

[0014] S7: Associate each monitoring quantity judged abnormal in step S6 with the intermediate events and bottom events in the steam isolation valve system fault tree, and index them in the fault tree according to the coupling relationship between each monitoring quantity, trace them back to the specific faulty component, and output the fault diagnosis result.

[0015] Preferably, in step S1, the monitoring quantities of the gas-liquid linkage actuator include the hydraulic oil temperature, the oil pressure of the upper and lower chambers of the hydraulic cylinder, the accumulator air pressure, the accumulator oil pressure, the solenoid valve control current and the thrust of the gas-liquid linkage actuator; the monitoring quantities of the steam isolation valve body include the valve body thrust, the packing bolt force, the valve action time and the valve displacement.

[0016] Preferably, in step S1, the top event of the fault tree is a steam isolation valve system failure event; the intermediate events of the fault tree include valve position change events and valve position failure events, the valve position change events are associated with valve action time and valve displacement, and the valve position failure events are associated with valve action time, valve displacement, and valve stem thrust; the bottom events of the fault tree include hydraulic cylinder internal leakage events and valve seal failure events, wherein the hydraulic cylinder internal leakage events are associated with the hydraulic cylinder upper chamber oil pressure and the hydraulic cylinder lower chamber oil pressure, and the valve seal failure event is associated with the packing bolt force.

[0017] Preferably, in step S2, the trend of the monitoring quantity benchmark data is calculated using the moving average method, and a suitable window size is selected. The data trend is determined by calculating the data mean in the window. If the data mean gradually increases, it indicates that the data is on an upward trend. If the data mean gradually decreases, it indicates that the data is on a downward trend. In addition, the trend index is set as the sum of the difference between the data means in the front and rear windows divided by the range of allowable fluctuations of the data. If the trend index is less than 0.5, it is determined that the monitoring quantity does not have a trend, otherwise it has a trend.

[0018] The periodicity of the monitoring quantity benchmark data is measured using the autocorrelation function. If the autocorrelation function graph of the monitoring quantity has regular peaks at different delays, the monitoring quantity is judged to have periodicity, otherwise it does not have periodicity. If the monitoring quantity has periodicity, the periodic peak value of the monitoring quantity is obtained according to the autocorrelation function graph.

[0019] The autocorrelation of the benchmark data of the monitoring quantity is measured by the autocorrelation index; the autocorrelation coefficient r of multiple groups of benchmark data of the same monitoring quantity is solved k The average value is taken to obtain the autocorrelation index of the benchmark data of the monitoring quantity. If the autocorrelation index is close to 1, it is considered that the monitoring quantity has autocorrelation, otherwise it does not have autocorrelation.

[0020] Furthermore, the autocorrelation function takes the lag period k as the independent variable, and the autocorrelation coefficient r k Plotted as the dependent variable; autocorrelation coefficient r k The calculation formula is as follows:

[0021]

[0022] In the formula, x j represents the jth element of the monitoring quantity; N represents the total number of elements contained in the monitoring quantity x, represents the mean of N monitored quantities; k represents the number of lag periods.

[0023] Preferably, the partial monitoring quantities include hydraulic oil temperature, hydraulic cylinder upper chamber oil pressure, hydraulic cylinder lower chamber oil pressure, accumulator air pressure, accumulator oil pressure, solenoid valve control current and valve body thrust; wherein the first threshold range of the hydraulic oil temperature is 10°C to 55°C, the first threshold values ​​of the hydraulic cylinder upper chamber oil pressure and the hydraulic cylinder lower chamber oil pressure are <25MPa, the first threshold range of the accumulator air pressure is 14MPa to 30MPa, the first threshold range of the accumulator oil pressure is 14MPa to 30MPa, the first threshold range of the solenoid valve control current is 0 to 40mA, and the first threshold range of the valve body thrust is 50kN to 3000kN.

[0024] Preferably, the preprocessing in step S4 includes data cleaning, noise reduction and data format standardization.

[0025] Preferably, the threshold determination in step S6 is specifically as follows:

[0026] If the monitored quantity has a first threshold, the preprocessed monitored quantity data is compared with the first threshold and the second threshold of the monitored quantity, and the trend index, periodicity peak value or autocorrelation index calculated in step S5 is respectively compared with the trend threshold, periodicity threshold or autocorrelation threshold of the monitored quantity;

[0027] If the monitored quantity does not have the first threshold, the preprocessed monitored quantity data is directly compared with the second threshold of the monitored quantity, and the trend index, periodicity peak value or autocorrelation index calculated in step S5 is respectively compared with the trend threshold, periodicity threshold or autocorrelation threshold of the monitored quantity;

[0028] If the preprocessed monitoring data falls outside the first threshold or the second threshold, or the calculated trend index, periodic peak or autocorrelation index falls outside the corresponding trend threshold, periodic threshold, autocorrelation threshold, it is determined that the steam isolation valve system is abnormal; otherwise, it is determined that the steam isolation valve system is operating normally.

[0029] Preferably, in step S7, specific faulty components are traced by constructing a seasonal difference autoregressive moving average model.

[0030] Preferably, the working state of the steam isolation valve system is divided into a stable state, a fast closing state, a slow opening state and a slow closing state;

[0031] If the steam isolation valve system is in a stable state, the monitoring quantities involved include the oil pressure in the upper chamber of the hydraulic cylinder, the oil pressure in the lower chamber of the hydraulic cylinder, the air pressure in the accumulator, the oil pressure in the accumulator, the valve action time and the valve displacement;

[0032] If the steam isolation valve system is in the fast closing state, the monitoring quantities involved include the solenoid valve control current, accumulator air pressure, accumulator oil pressure, hydraulic cylinder upper chamber oil pressure, hydraulic cylinder lower chamber oil pressure, hydraulic oil temperature, actuator thrust, packing bolt force, valve body thrust, valve action time and valve displacement;

[0033] If the steam isolation valve system is in a slow opening or slow closing state, the monitoring quantities involved include the solenoid valve control current, the oil pressure in the upper chamber of the hydraulic cylinder, the oil pressure in the lower chamber of the hydraulic cylinder, the hydraulic oil temperature, the actuator thrust, the packing bolt force, the valve body thrust, the valve action time and the valve displacement.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] For the steam isolation valve system with complex gas-liquid linkage actuator, by making full use of the factory data and fault tree model of the steam isolation valve system, analyzing multiple monitoring data and their characteristics, selecting the threshold of the valve monitoring quantity, judging and analyzing the data coupling relationship through the threshold, and providing timely feedback, the online monitoring of the status of the steam isolation valve system is realized, and the fault is located to the scope of component events through diagnosis, which provides a new idea for the online monitoring of the status of the steam isolation valve system and fault diagnosis, and also provides a basis for the subsequent predictive maintenance of the steam isolation valve system. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a flow chart of the method for online monitoring of the status of the steam isolation valve system;

[0037] Figure 2 Fault tree model for steam isolation valve system;

[0038] Figure 3 is an autocorrelation function (ACF) diagram of the oil pressure in the upper chamber of the hydraulic cylinder in this embodiment;

[0039] Figure 4 is an autocorrelation function (ACF) diagram of the oil pressure in the lower chamber of the hydraulic cylinder in this embodiment;

[0040] Figure 5 : is the autocorrelation function (ACF) diagram of the accumulator oil pressure in this embodiment. DETAILED DESCRIPTION

[0041] The present invention is further described and illustrated below in conjunction with the accompanying drawings and specific embodiments. The technical features of each embodiment of the present invention can be combined accordingly without conflicting with each other.

[0042] like Figure 1 As shown, it is an implementation flow chart of an online status monitoring method for a steam isolation valve system provided by the present invention.

[0043] The steam isolation valve system consists of two parts: the valve body and the actuator. The actuator is used to provide power for the steam isolation valve system to open and close the valve. The steam isolation valve system has four main working states, namely stable, fast closing, slow opening, and slow closing: (1) The stable state is the state in which the valve maintains the fully open valve position unchanged. At this time, the monitoring quantities that can be collected include the oil pressure in the upper chamber of the hydraulic cylinder, the oil pressure in the lower chamber of the hydraulic cylinder, the accumulator air pressure, the accumulator oil pressure, the valve action time and displacement; (2) The fast closing state is the valve performing the fast closing function. The fast closing solenoid valve of the actuator receives the fast closing signal and opens, the control circuit is discharged, and then the oil inlet and return logic valve is opened. The accumulator energy is released, pushing the hydraulic oil into the upper chamber of the hydraulic cylinder, and finally pushing the valve to close. The monitoring quantities that can be collected in this process include the solenoid valve control current, the accumulator air pressure, and the pressure at the valve. , accumulator oil pressure, hydraulic cylinder upper chamber oil pressure, hydraulic cylinder lower chamber oil pressure, hydraulic oil temperature, actuator thrust, packing bolt force, valve body thrust, valve action time and displacement; (3) The slow opening state and slow closing state are the slow opening and slow closing of the valve respectively. The two share the same circuit. The driving force of the hydraulic cylinder is provided by the motor and pump. When the slow opening signal is received, the motor and pump start working. At this time, the pump rotates forward, the solenoid valve maintains the original valve position, and the hydraulic oil enters the upper chamber of the hydraulic cylinder to push the valve to close slowly. When the slow closing signal is received, the motor and pump start working. At this time, the pump reverses and the solenoid valve position changes. The hydraulic oil enters the lower chamber of the hydraulic cylinder and drives the valve to open slowly. The monitoring quantities that can be collected during the slow opening and slow closing process include the solenoid valve control current, hydraulic cylinder upper chamber oil pressure, hydraulic cylinder lower chamber oil pressure, hydraulic oil temperature, actuator thrust, packing bolt force, valve body thrust, valve action time and displacement.

[0044] This embodiment provides a method for online monitoring of the status of a steam isolation valve system, and the specific steps are as follows:

[0045] (1) Based on the above principles and combined with the typical failure modes of the steam isolation valve system, a fault tree model of the steam isolation valve system is established, such as Figure 2 As shown. Associate the top event, middle event and bottom event of the fault tree with the corresponding monitoring quantities respectively. By associating events with monitoring quantities, the monitoring quantities that need to be paid attention to when different faults occur can be obtained. It should be noted that the monitoring quantities include the monitoring quantities of the gas-liquid linkage actuator and the monitoring quantities of the steam isolation valve body. Among them, the monitoring quantities of the gas-liquid linkage actuator include the hydraulic oil temperature, the oil pressure in the upper chamber of the hydraulic cylinder, the oil pressure in the lower chamber of the hydraulic cylinder, the accumulator gas pressure, the accumulator oil pressure, the solenoid valve control current and the thrust of the gas-liquid linkage actuator. The monitoring quantities of the steam isolation valve body include the thrust of the valve body, the packing bolt force, the valve action time and the valve displacement.

[0046] For example, the top event of the fault tree is the failure of the steam isolation valve system; the intermediate events of the fault tree include valve position change events, valve position failure events, etc., among which the valve position change events are related to the valve action time and valve displacement, and the valve position failure events are related to the valve action time, valve displacement, and valve stem thrust; the bottom events of the fault tree include hydraulic cylinder leakage events, valve seal failure events, etc., among which the hydraulic cylinder leakage events are related to the hydraulic cylinder upper chamber oil pressure and the hydraulic cylinder lower chamber oil pressure, and the valve seal failure events are related to the packing bolt force.

[0047] (2) The data of the monitoring quantity obtained in step (1) when the steam isolation valve system leaves the factory is used as the baseline data; the time point when the control signal changes is used as the point at which the working state of the steam isolation valve system changes, and the baseline data is segmented according to the time point; and the baseline data of each monitoring quantity after segmentation is subjected to feature classification according to basic characteristics, wherein the basic characteristics include trend, periodicity and autocorrelation.

[0048] This embodiment selects the stable state of the steam isolation valve system. First, the stable state segment of the monitored quantity is taken according to the control signal, and the benchmark data is characterized and classified:

[0049] ① The trend of the benchmark data of the monitoring quantity is calculated using the moving average method. The appropriate window size is selected, and the data trend is judged by calculating the mean value of the data in the window. If the data mean gradually increases (the number of times the mean value increases exceeds 5 times), it means that the data is on an upward trend. If the data mean gradually decreases (the number of times the mean value decreases exceeds 5 times), it means that the data is on a downward trend. In addition, considering that the data itself has certain fluctuations during the actual operation process, the sum of the difference between the mean values ​​of the data in the previous and next windows is divided by the range of data fluctuations allowed to obtain the trend index. If the trend index is less than 0.5, it is determined that the monitoring quantity does not have a trend; otherwise, it is determined that the monitoring quantity has a trend.

[0050] In this embodiment, the trend indicators of the hydraulic cylinder upper chamber oil pressure, the hydraulic cylinder lower chamber oil pressure, the accumulator air pressure, the accumulator oil pressure, and the valve displacement are 1.5×10 -5 , 2.4×10 -6 , 0, 0, 0, are all far less than 0.5, so they do not have trends.

[0051] ② The periodicity of the monitoring quantity benchmark data is measured using the autocorrelation function. If the autocorrelation function graph of the monitoring quantity has obvious regular peaks at different delays, the monitoring quantity is considered to have periodicity. If the monitoring quantity has periodicity, the periodic peak of the monitoring quantity can be obtained based on the autocorrelation function graph.

[0052] Autocorrelation coefficient r k The calculation formula is as follows:

[0053]

[0054] In the formula, x j represents the jth element of the monitoring quantity; N represents the total number of elements contained in the monitoring quantity x, represents the mean of N monitored quantities; k represents the number of lag periods.

[0055] Taking the lag period k as the independent variable, the autocorrelation coefficient r k As the dependent variable, draw the autocorrelation function graph of the monitored quantity. Figure 3 is the autocorrelation function diagram of the oil pressure in the upper chamber of the hydraulic cylinder in this embodiment, Figure 4 and Figure 5 They are the autocorrelation function diagrams of the lower chamber oil pressure of the hydraulic cylinder and the accumulator oil pressure. It can be seen that the upper chamber oil pressure of the hydraulic cylinder does not have an obvious regular peak in the autocorrelation function diagram, so it does not have periodicity. The autocorrelation function diagrams of the lower chamber oil pressure of the hydraulic cylinder, the accumulator air pressure, the accumulator oil pressure, and the valve displacement are stable or rapidly drop to the confidence interval state, so they do not have periodicity either.

[0056] ③ The autocorrelation of the monitoring quantity benchmark data is measured by the autocorrelation index. Solve the autocorrelation coefficient r of multiple groups of benchmark data of the same monitoring quantity k The average value is taken to obtain the autocorrelation index of the baseline data of the monitoring quantity. If the autocorrelation index is close to 1, it is determined that the monitoring quantity has autocorrelation; otherwise, it is determined that the monitoring quantity does not have autocorrelation.

[0057] In this embodiment, the autocorrelation indexes of the hydraulic cylinder upper chamber oil pressure, hydraulic cylinder lower chamber oil pressure, accumulator air pressure, accumulator oil pressure and valve displacement are 1, 1, 0.1864, 0.1533 and -0.0889 respectively. This indicates that the hydraulic cylinder upper chamber oil pressure and the hydraulic cylinder lower chamber oil pressure have strong autocorrelation, while the accumulator air pressure, accumulator oil pressure and valve displacement do not have autocorrelation. At this point, the solution and analysis of the data characteristics of the steam isolation valve system when it leaves the factory are completed.

[0058] (3) According to the working conditions and technical requirements of the steam isolation valve system, it can be obtained that under the stable working state, the first threshold values ​​of the oil pressure in the upper chamber of the hydraulic cylinder and the oil pressure in the lower chamber of the hydraulic cylinder are both <25MPa, the first threshold value range of the accumulator air pressure is 14MPa~30MPa, and the first threshold value range of the accumulator oil pressure is 14MPa~30MPa.

[0059] If the benchmark data of the monitored quantity has trend, periodicity or autocorrelation, a certain margin will be added to the trend index, periodicity peak or autocorrelation index of the benchmark data, which will be used as the trend threshold, periodicity threshold or autocorrelation threshold of the monitored quantity respectively; calculate the variance and mean of multiple groups of benchmark data of all monitored quantities, take the mean as the benchmark and the variance as the margin, and obtain the second threshold of the monitored quantity.

[0060] Through the above threshold selection, each monitoring quantity can obtain at least one threshold range (there must be a second threshold). For monitoring quantities with two thresholds at the same time, once the monitoring quantity monitored during the actual operation exceeds one of the thresholds, it is considered to be abnormal.

[0061] The thresholds of the upper chamber oil pressure of the hydraulic cylinder, the lower chamber oil pressure of the hydraulic cylinder, the accumulator air pressure, the accumulator oil pressure, and the valve displacement of the steam isolation valve system of this embodiment under a stable working state are shown in Table 1.

[0062] Table 1 Thresholds of various monitoring quantities of the fast-closing channel under stable working conditions

[0063]

[0064] In this embodiment, only the upper chamber oil pressure of the hydraulic cylinder and the lower chamber oil pressure of the hydraulic cylinder have autocorrelation, so there is an autocorrelation threshold. The above-mentioned first threshold, autocorrelation threshold and second threshold together constitute the threshold under the stable working state of the steam isolation valve system.

[0065] Calculate the cross-correlation statistics between different monitoring quantities and obtain the cross-correlation threshold.

[0066] The Pearson correlation coefficient is used to characterize the statistical value of the mutual correlation between different monitoring quantities, and the coupling relationship between different monitoring quantities is obtained. xy The calculation formula is as follows:

[0067]

[0068] Where: x j represents the jth element of the monitoring quantity x; y j represents the jth element of the monitoring quantity y; N represents the total number of elements contained in the monitoring quantity x or y; represents the mean of N monitoring quantities x; Represents the mean of N monitored quantities y.

[0069] In this embodiment, according to the above Pearson correlation coefficient r xy The calculation formula is used to obtain the cross-correlation between different monitoring quantities. It can be obtained that the upper chamber oil pressure of the hydraulic cylinder is negatively correlated with the lower chamber oil pressure of the hydraulic cylinder (the coefficient is -1), and there is no obvious correlation between the upper chamber oil pressure of the hydraulic cylinder and the lower chamber oil pressure of the hydraulic cylinder and the accumulator air pressure, accumulator oil pressure, and valve displacement respectively (the coefficient is between [-0.01, 0.01]), the accumulator air pressure and the accumulator oil pressure are positively correlated (the coefficient is 1), and there is no obvious correlation between the accumulator air pressure and the accumulator oil pressure and the valve displacement respectively (the coefficient is unstable).

[0070] (4) Preprocess the monitoring quantity monitored during the actual operation of the steam isolation valve system to obtain preprocessed monitoring quantity data. The preprocessing includes data cleaning and noise reduction and data format standardization.

[0071] (5) Calculate the trend index, periodic peak value and autocorrelation index of the monitoring quantity data after different preprocessing obtained in step (4). The specific calculation method is the same as that in step (2).

[0072] (6) In this embodiment, the stable state of the steam isolation valve system is selected. If the monitoring quantity has a first threshold, the preprocessed monitoring quantity data is compared with the first threshold and the second threshold of the monitoring quantity, and the trend index, periodicity peak value or autocorrelation index calculated in step (5) is compared with the trend threshold, periodicity threshold or autocorrelation threshold of the monitoring quantity respectively;

[0073] If the monitored quantity does not have the first threshold, directly compare the preprocessed monitored quantity data with the second threshold of the monitored quantity, and compare the trend index, periodicity peak value or autocorrelation index calculated in step (5) with the trend threshold, periodicity threshold or autocorrelation threshold of the monitored quantity respectively;

[0074] If the pre-processed monitoring data falls outside the first threshold or the second threshold, or the calculated trend index, periodic peak value or autocorrelation index falls outside the corresponding trend threshold, periodic threshold, autocorrelation threshold, it is determined that the steam isolation valve system is abnormal; otherwise, it is determined that the steam isolation valve system is operating normally. In the case of abnormal detection, the next step of the fault diagnosis program is entered to determine the specific faulty components of the system.

[0075] Specifically, in this embodiment, the data window is taken as 2000, the lag period is 20, and the characteristic values ​​(trend indicators, periodic peak values ​​or autocorrelation indicators) of the preprocessed monitoring data (oil pressure in the upper chamber of the hydraulic cylinder, oil pressure in the lower chamber of the hydraulic cylinder, accumulator air pressure, accumulator oil pressure and valve displacement) are calculated respectively, and the characteristic values ​​are compared with the threshold range.

[0076] The monitoring data of the hydraulic cylinder upper chamber oil pressure and the hydraulic cylinder lower chamber oil pressure are both within the first threshold and the second threshold, but the trend indicators are 0.7058 and 0.7471 respectively, and it is judged that both have a downward trend, so they have a trend. However, it does not match the factory data characteristics (the benchmark data of the steam isolation valve system at the factory shows that the hydraulic cylinder upper chamber oil pressure and the hydraulic cylinder lower chamber oil pressure do not have a trend), and the output hydraulic cylinder upper chamber oil pressure abnormality and hydraulic cylinder lower chamber oil pressure abnormality results.

[0077] The valve displacement was normal in the early stage, and reached 0.1031mm in the later stage, exceeding the upper limit of the second threshold, and the trend index was 1.9046, which was judged to have an upward trend and a trend. However, it did not match the factory data characteristics (the benchmark data of the steam isolation valve system at the factory showed that the valve displacement did not have a trend), so the output valve displacement abnormal result. All indicators of the accumulator air pressure and accumulator oil pressure were within the threshold range, so the accumulator air pressure and accumulator oil pressure were normal.

[0078] At this time, the oil pressure in the upper chamber of the hydraulic cylinder is positively correlated with the oil pressure in the lower chamber of the hydraulic cylinder (the coefficient is 1), the oil pressure in the upper and lower chambers of the hydraulic cylinder is negatively correlated with the valve displacement (the coefficient is -1), the accumulator air pressure is positively correlated with the accumulator oil pressure (the coefficient is 1), and the rest has no obvious correlation. So far, the abnormality detection of the current data is completed.

[0079] (7) Associate each monitoring quantity judged to be abnormal in step (6) with the intermediate events and bottom events in the steam isolation valve system fault tree, and index them in the fault tree according to the coupling relationship between each monitoring quantity, trace them back to the specific faulty component, and output the fault diagnosis result.

[0080] In this embodiment, the currently detected abnormal monitoring quantities, namely, the hydraulic cylinder upper chamber oil pressure, the hydraulic cylinder lower chamber oil pressure and the valve displacement, are used as inputs, and all events that may be related to the current abnormality are retrieved in the fault tree, namely, valve position change, hydraulic loss, valve gate breakage, valve stud thread failure, valve nut thread failure, pipeline / pipe fitting leakage, overflow valve leakage, hydraulic cylinder internal leakage, and hydraulic cylinder external leakage.

[0081] Among them, valve position change is the first-level intermediate event, hydraulic loss is the second-level intermediate event, and the rest are bottom events and can be fully indexed to the top event of steam isolation valve system failure. However, for the bottom events of valve gate breakage, valve stud thread failure, and valve nut thread failure, although the valve position will exceed the limit, its valve position does not have a trend, and there is no significant correlation between the oil pressure in the upper and lower chambers of the hydraulic cylinder and the valve displacement;

[0082] For pipeline / pipe fitting leakage, overflow valve leakage and hydraulic cylinder leakage events, the constructed SARIMA (Seasonal Autoregressive Moving Average Model) model can be used to predict the oil pressure in the upper chamber of the hydraulic cylinder and the oil pressure in the lower chamber of the hydraulic cylinder. The oil pressure on one side will tend to atmospheric pressure, while the oil pressure on both sides of the hydraulic cylinder leakage event will tend to be consistent. Therefore, by indexing from the bottom event to the top event according to the fault tree, the final diagnostic result output can be obtained as follows: hydraulic cylinder leakage, hydraulic loss, abnormal valve position change, steam isolation valve system failure. If the SARIMA model is not used, the final diagnostic result can be output as follows: hydraulic cylinder leakage, pipeline / pipe fitting leakage, overflow valve leakage, hydraulic cylinder leakage, hydraulic loss, abnormal valve position change, steam isolation valve system failure. It is recommended to combine manual detection. All diagnostic procedures are completed at this point.

[0083] The above-described embodiment is only a preferred solution of the present invention, but it is not intended to limit the present invention. A person skilled in the relevant technical field may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, any technical solution obtained by equivalent replacement or equivalent transformation falls within the protection scope of the present invention.

Claims

1. A method for online monitoring of the status of a steam isolation valve system, characterized in that: The specific steps are as follows: S1: Construct a fault tree for the steam isolation valve system and list all potential faults; associate the top event, intermediate event and bottom event of the fault tree with corresponding monitoring quantities; the monitoring quantities include the monitoring quantity of the gas-liquid linkage actuator and the monitoring quantity of the steam isolation valve body; S2: taking the data of the monitoring quantity described in step S1 when the steam isolation valve system leaves the factory as the benchmark data; taking the time point of the control signal change as the point of change of the working state of the steam isolation valve system, and segmenting the benchmark data according to the time point; and performing feature classification on the benchmark data of each monitoring quantity segmentation according to basic characteristics; the basic characteristics include trend, periodicity and autocorrelation; S3: combining the working condition requirements and technical requirements to obtain a fixed threshold value of some monitoring quantities as the first threshold value of the monitoring quantities; If the baseline data of the monitored quantity has trend, periodicity or autocorrelation, a certain margin is added to the trend index, periodicity peak or autocorrelation index of the baseline data, which are used as the trend threshold, periodicity threshold or autocorrelation threshold of the monitored quantity respectively; Calculate the variance and mean of multiple groups of benchmark data of all monitoring quantities, take the mean as the benchmark and the variance as the margin, and obtain the second threshold value of the monitoring quantity; S4: pre-processing the monitoring quantity monitored during the actual operation of the steam isolation valve system to obtain pre-processed monitoring quantity data; S5: Calculate the trend index, periodic peak value and autocorrelation index of the monitoring quantity data obtained in step S4 after different preprocessing; S6: According to different working states of the steam isolation valve system, the pre-processed monitoring data is compared with the first threshold value or the second threshold value, and the trend index, periodic peak value or autocorrelation index calculated in step S5 is compared with the corresponding trend threshold value, periodic threshold value or autocorrelation threshold value; if any data falls outside the corresponding threshold range, it is determined that the steam isolation valve system is abnormal, otherwise it is determined that the steam isolation valve system is operating normally; S7: Associate each monitoring quantity judged abnormal in step S6 with the intermediate events and bottom events in the steam isolation valve system fault tree, and index them in the fault tree according to the coupling relationship between each monitoring quantity, trace them back to the specific faulty component, and output the fault diagnosis result.

2. The method for online status monitoring of a steam isolation valve system according to claim 1, characterized in that: In step S1, the monitoring quantities of the gas-liquid linkage actuator include the hydraulic oil temperature, the oil pressure of the upper and lower chambers of the hydraulic cylinder, the accumulator air pressure, the accumulator oil pressure, the solenoid valve control current and the thrust of the gas-liquid linkage actuator; the monitoring quantities of the steam isolation valve body include the valve body thrust, the packing bolt force, the valve action time and the valve displacement.

3. The method for online status monitoring of a steam isolation valve system according to claim 1, characterized in that: In step S1, the top event of the fault tree is a steam isolation valve system failure event; the intermediate events of the fault tree include valve position change events and valve position failure events, the valve position change events are associated with valve action time and valve displacement, and the valve position failure events are associated with valve action time, valve displacement, and valve stem thrust; the bottom events of the fault tree include hydraulic cylinder internal leakage events and valve seal failure events, wherein the hydraulic cylinder internal leakage events are associated with the hydraulic cylinder upper chamber oil pressure and the hydraulic cylinder lower chamber oil pressure, and the valve seal failure events are associated with the packing bolt force.

4. The method for online status monitoring of a steam isolation valve system according to claim 1, characterized in that: In step S2, the trend of the monitoring quantity benchmark data is calculated using the moving average method, and an appropriate window size is selected. The data trend is determined by calculating the data mean in the window. If the data mean gradually increases, it indicates that the data is on an upward trend. If the data mean gradually decreases, it indicates that the data is on a downward trend. In addition, the trend index is set as the sum of the difference between the data means in the previous and next windows divided by the range of data fluctuation allowed. If the trend index is less than 0.5, it is determined that the monitoring quantity does not have a trend, otherwise it has a trend. The periodicity of the monitoring quantity benchmark data is measured using the autocorrelation function. If the autocorrelation function graph of the monitoring quantity has regular peaks at different delays, the monitoring quantity is judged to have periodicity, otherwise it does not have periodicity. If the monitoring quantity has periodicity, the periodic peak value of the monitoring quantity is obtained according to the autocorrelation function graph. The autocorrelation of the benchmark data of the monitoring quantity is measured by the autocorrelation index; the autocorrelation coefficient r of multiple groups of benchmark data of the same monitoring quantity is solved k The average value is taken to obtain the autocorrelation index of the benchmark data of the monitoring quantity. If the autocorrelation index is close to 1, it is considered that the monitoring quantity has autocorrelation, otherwise it does not have autocorrelation.

5. The method for online status monitoring of a steam isolation valve system according to claim 4, characterized in that: The autocorrelation function takes the lag period k as the independent variable, and the autocorrelation coefficient r k Plotted as the dependent variable; autocorrelation coefficient r k The calculation formula is as follows: In the formula, x j represents the jth element of the monitoring quantity; N represents the total number of elements contained in the monitoring quantity x, represents the mean of N monitored quantities; k represents the number of lag periods.

6. The method for online status monitoring of a steam isolation valve system according to claim 1, characterized in that: The some monitored quantities include hydraulic oil temperature, hydraulic cylinder upper chamber oil pressure, hydraulic cylinder lower chamber oil pressure, accumulator air pressure, accumulator oil pressure, solenoid valve control current and valve body thrust; among them, the first threshold range of hydraulic oil temperature is 10℃~55℃, the first threshold range of hydraulic cylinder upper chamber oil pressure and hydraulic cylinder lower chamber oil pressure is <25MPa, the first threshold range of accumulator air pressure is 14MPa~30MPa, the first threshold range of accumulator oil pressure is 14MPa~30MPa, the first threshold range of solenoid valve control current is 0~40mA, and the first threshold range of valve body thrust is 50kN~3000kN.

7. The method for online status monitoring of a steam isolation valve system according to claim 1, characterized in that: The preprocessing in step S4 includes data cleaning, noise reduction and data format standardization.

8. The method for online status monitoring of a steam isolation valve system according to claim 1, characterized in that: The threshold determination in step S6 is specifically as follows: If the monitored quantity has a first threshold, the preprocessed monitored quantity data is compared with the first threshold and the second threshold of the monitored quantity, and the trend index, periodicity peak value or autocorrelation index calculated in step S5 is respectively compared with the trend threshold, periodicity threshold or autocorrelation threshold of the monitored quantity; If the monitored quantity does not have the first threshold, the preprocessed monitored quantity data is directly compared with the second threshold of the monitored quantity, and the trend index, periodicity peak value or autocorrelation index calculated in step S5 is respectively compared with the trend threshold, periodicity threshold or autocorrelation threshold of the monitored quantity; If the preprocessed monitoring data falls outside the first threshold or the second threshold, or the calculated trend index, periodic peak or autocorrelation index falls outside the corresponding trend threshold, periodic threshold, autocorrelation threshold, it is determined that the steam isolation valve system is abnormal; otherwise, it is determined that the steam isolation valve system is operating normally.

9. The method for online status monitoring of a steam isolation valve system according to claim 1, characterized in that: In step S7, the specific faulty component is traced back through the constructed seasonal difference autoregressive moving average model.

10. The method for online status monitoring of a steam isolation valve system according to claim 1, characterized in that: The working states of the steam isolation valve system are divided into a stable state, a fast closing state, a slow opening state and a slow closing state; If the steam isolation valve system is in a stable state, the monitoring quantities involved include the oil pressure in the upper chamber of the hydraulic cylinder, the oil pressure in the lower chamber of the hydraulic cylinder, the air pressure in the accumulator, the oil pressure in the accumulator, the valve action time and the valve displacement; If the steam isolation valve system is in the fast closing state, the monitoring quantities involved include the solenoid valve control current, accumulator air pressure, accumulator oil pressure, hydraulic cylinder upper chamber oil pressure, hydraulic cylinder lower chamber oil pressure, hydraulic oil temperature, actuator thrust, packing bolt force, valve body thrust, valve action time and valve displacement; If the steam isolation valve system is in a slow opening or slow closing state, the monitoring quantities involved include the solenoid valve control current, hydraulic cylinder oil pressure, hydraulic oil temperature, actuator thrust, packing bolt force, valve body thrust, valve action time and valve displacement.

Images