A method and system for detecting the open / close state of an emergency shut-off valve

Abstract

This invention discloses a method and system for detecting the on / off status of an emergency shut-off valve, specifically relating to the field of valve on / off status detection technology. The method includes: collecting valve on / off status data using multi-source sensors; constructing a temporal correlation map between valve motion parameters and fluid parameters based on the on / off status data; constructing a spatiotemporal alignment feature matrix of correlation coefficients based on the temporal correlation map; constructing a three-dimensional valve on / off status judgment coordinate system based on the spatiotemporal alignment feature matrix; judging the valve on / off status within a preset qualified area of ​​the three-dimensional coordinate system; and employing a dual-channel feedback mechanism to provide real-time feedback on the valve on / off status. This invention, by collecting valve on / off status data using multi-source sensors, establishing the correlation between valve motion parameters and fluid parameters, and establishing a spatiotemporal alignment feature matrix, solves the problems of insufficient data reliability, lack of dynamic data correlation, and low judgment accuracy in existing technologies.

Description

Technical Field

[0001] This invention relates to the field of valve on / off status detection technology, and more specifically, to a method and system for detecting the on / off status of an emergency shut-off valve. Background Technology

[0002] Emergency shut-off valves are crucial safety devices in industrial processes such as petrochemicals, natural gas transportation, and hazardous chemical storage. Their core function is to quickly and reliably cut off the flow of media upon detecting leaks, overpressure, fires, or other emergencies, preventing the accident from escalating and ensuring the safety of personnel, equipment, and the environment. Therefore, real-time and accurate monitoring of the on / off status of emergency shut-off valves is a key factor in ensuring their functional effectiveness and the overall safety of the system.

[0003] Existing emergency shut-off valve on / off status detection systems can basically meet the application requirements, but some shortcomings still exist: Firstly, existing methods for detecting the on / off status of emergency shut-off valves mainly rely on single parameters such as displacement or pressure, neglecting the comprehensive judgment of multi-source valve data, leading to insufficient reliability of the detection data. Secondly, existing methods independently analyze the valve's motion and fluid parameters, ignoring the dynamic correlation between them. Finally, existing methods for detecting the on / off status of emergency shut-off valves are not aligned in the spatiotemporal dimensions, affecting the accuracy of valve on / off status judgment. Therefore, a method is needed to address the problems of insufficient data reliability, lack of dynamic data correlation, and insufficient judgment accuracy. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a method and system for detecting the on / off status of an emergency shut-off valve, which solves the problems mentioned in the background art through the following solutions.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method and system for detecting the on / off status of an emergency shut-off valve, comprising:

[0006] S1: Preset a specified type of emergency shut-off valve as the target valve, and collect the switch status data of the target valve in real time by installing multi-source sensors. The switch status data includes the valve's motion parameter data and the fluid parameter data in the valve pipe.

[0007] S2: Based on the switch status data collected in S1, feature dimension expansion analysis is performed to obtain displacement change rate, angle change rate, differential pressure gradient, flow velocity pulsation coefficient and sealing pressure distribution entropy value. A time-series correlation map of valve motion parameters and fluid parameters is established through cross-correlation algorithm.

[0008] S3: Based on the temporal correlation map of valve motion parameters and fluid parameters in S2, the correlation coefficient between valve motion parameters and fluid parameters is obtained. The correlation coefficient is spatiotemporally registered to obtain the spatiotemporal alignment feature matrix of the correlation coefficient.

[0009] S4: Based on the spatiotemporal alignment feature matrix of the correlation coefficient in S3, a three-dimensional valve switching state judgment coordinate system is constructed. The coordinate system includes an X-axis representing the valve position achievement degree, a Y-axis representing the fluid state matching degree, and a Z-axis representing the parameter correlation degree. The valve switching state is judged by setting a qualified area in the coordinate system. A dual-channel feedback mechanism is adopted to provide real-time feedback on the valve switching state.

[0010] Preferably, the feature dimension expansion analysis in S2 includes feature extraction of the valve stem displacement data and the original displacement sequence. Perform a 5th-order polynomial least squares fit, the fitting formula is as follows: ,in This represents the original displacement data of the valve stem. The fitting coefficients for the original displacement data are used, and the goodness of fit needs to be controlled above 95%. This is achieved through... The rate of change of displacement is obtained by taking the first derivative. The calculation formula is ,right The displacement acceleration is obtained by taking the second derivative. The calculation formula is ; Extracting the angular features of the valve stem using angular sequence data. The formula for calculating the Fourier transform is as follows: ,in Indicates the frequency of data collection. This indicates that angle data is collected in increments of 10 degrees to obtain the rate of change of angle. The calculation formula is: ;

[0011] Differential pressure gradient calculation is performed on the internal pressure data of valves and pipelines, and the differential pressure data is processed through a sliding window. The window is segmented, with a sliding window size of 100. Step size is 50 Segment the differential pressure data and define the differential pressure gradient. The calculation formula is ,in For two points in time within the window, The value represents the change in pipe pressure per unit time; the velocity pulsation coefficient is calculated by decomposing the velocity data using the db4 wavelet basis, reconstructing the velocity pulsation components, and then calculating the velocity pulsation coefficient. The formula is ,in The standard deviation of the pulse component The average value of the flow velocity data; the sealing pressure distribution entropy value, which is obtained by calculating the distribution entropy of the collected pressure data. The specific calculation formula is as follows: ,in Indicates the first Pressure values ​​at each collection point, with a collection cycle of 100. .

[0012] Preferably, a system for detecting the on / off status of an emergency shut-off valve includes:

[0013] Sensor acquisition module: Real-time acquisition of the on / off status data of the target valve by installing multi-source sensors. The on / off status data includes valve displacement data, angle data, and fluid parameter data in the valve pipe.

[0014] Data analysis module: Based on the switch status data, feature dimension expansion analysis is performed to obtain displacement change rate, angle change rate, differential pressure gradient, flow velocity pulsation coefficient and sealing pressure distribution entropy value. The time series correlation map between valve motion parameters and fluid parameters is established through cross-correlation algorithm.

[0015] Data processing module: Based on the temporal correlation map of valve motion parameters and fluid parameters, the correlation coefficient between valve motion parameters and fluid parameters is obtained, and the correlation coefficient is spatiotemporally registered to obtain the spatiotemporal alignment feature matrix of the correlation coefficient.

[0016] Result Judgment and Feedback Module: Based on the spatiotemporal alignment feature matrix of correlation coefficient, a three-dimensional valve switch status judgment coordinate system is constructed. The valve switch status is judged by a preset qualified area in the three-dimensional valve switch status judgment coordinate system. A dual-channel feedback mechanism is adopted to provide real-time feedback on the valve switch status.

[0017] The technical effects and advantages of this invention are as follows:

[0018] 1. This invention solves the problem of insufficient reliability of valve switch status detection data by presetting a specified type of emergency shut-off valve as the target valve and collecting the switch status data of the target valve in real time by installing multi-source sensors.

[0019] 2. This invention solves the problem of lack of correlation in valve switch status detection data by expanding the feature dimensions of the collected switch status data and constructing a time-series correlation map;

[0020] 3. This invention solves the problem of spatiotemporal inconsistency in valve switching state detection by constructing a spatiotemporal aligned feature matrix through secondary calculation of the correlation degree of the temporal correlation graph and spatiotemporal registration;

[0021] 4. This invention achieves accurate judgment and real-time feedback of valve opening and closing status through preset qualified areas and dual-channel feedback mechanism, thereby improving the reliability and stability of the system. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the emergency shut-off valve on / off status detection method of the present invention.

[0023] Figure 2 This is a schematic diagram of the spatiotemporal alignment matrix structure of the present invention.

[0024] Figure 3 This is a schematic diagram of the three-dimensional determination coordinate system structure of the present invention.

[0025] Figure 4 This is a schematic diagram of the emergency shut-off valve switch status judgment structure of the present invention.

[0026] Figure 5 This is a schematic diagram of the on / off status detection system for the emergency shut-off valve of the present invention. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] Please see Figures 1-4 As shown, this embodiment of the invention provides a method for detecting the on / off state of an emergency shut-off valve. The method involves collecting on / off state data of the target valve, performing feature dimension expansion analysis on the collected data to construct a temporal correlation map between valve motion parameters and fluid parameters, constructing a spatiotemporal alignment feature matrix of correlation coefficients based on the temporal correlation map, and constructing a three-dimensional valve on / off state judgment coordinate system based on the spatiotemporal alignment feature matrix. This allows for the detection of the on / off state of the emergency shut-off valve. This embodiment of the invention discloses a method for detecting the on / off state of an emergency shut-off valve, including the following steps:

[0029] S1: Preset a specified type of emergency shut-off valve as the target valve, and collect the switch status data of the target valve in real time by installing multi-source sensors. The switch status data includes the valve's motion parameter data and the fluid parameter data in the valve pipe.

[0030] S2: Based on the switch status data collected in S1, feature dimension expansion analysis is performed to obtain displacement change rate, angle change rate, differential pressure gradient, flow velocity pulsation coefficient and sealing pressure distribution entropy value. A time-series correlation map of valve motion parameters and fluid parameters is established through cross-correlation algorithm.

[0031] S3: Based on the temporal correlation map of valve motion parameters and fluid parameters in S2, the correlation coefficient between valve motion parameters and fluid parameters is obtained. The correlation coefficient is spatiotemporally registered to obtain the spatiotemporal alignment feature matrix of the correlation coefficient.

[0032] S4: Based on the spatiotemporal alignment feature matrix of the correlation coefficient in S3, a three-dimensional valve switching state judgment coordinate system is constructed. The coordinate system includes an X-axis representing the valve position achievement degree, a Y-axis representing the fluid state matching degree, and a Z-axis representing the parameter correlation degree. The valve switching state is judged by setting a qualified area in the coordinate system. A dual-channel feedback mechanism is adopted to provide real-time feedback on the valve switching state.

[0033] In S1, a specified type of emergency shut-off valve is preset as the target valve. The switching data of the target valve is collected in real time by installing multi-source sensors. The switching status data includes the valve's displacement data, angle data, and fluid parameter data in the valve pipe.

[0034] Further explanation is needed regarding the following: Displacement data is collected using a laser displacement sensor, which measures the rate of change of valve stem displacement in both open and closed states. This sensor is specifically installed directly above the valve stem axis. Angle data is collected using a high-precision angle encoder, which measures the rate of change of valve stem angle in both open and closed states. This encoder is specifically installed around the valve stem at 90-degree intervals. Fluid parameter data includes fluid pressure and flow velocity data. A differential pressure transmitter is used to collect the differential pressure gradient change and sealing pressure distribution entropy values ​​within the pipe when the target valve is open and closed. This transmitter is specifically installed at the upstream inlet of the valve's pipeline. An electromagnetic flowmeter is used to collect the flow velocity pulsation coefficient within the pipe when the target valve is open and closed. This flowmeter is specifically installed at the downstream outlet of the valve's pipeline.

[0035] In S2, the collected valve opening and closing status data are subjected to feature dimension expansion analysis to provide calculation data for the subsequent establishment of a time-series correlation map between valve motion parameters and fluid parameters;

[0036] Feature extraction is performed on the displacement data of the valve stem, and the original displacement sequence is analyzed. Perform a 5th-order polynomial least squares fit, the fitting formula is as follows: ,in This represents the original displacement data of the valve stem. The fitting coefficients for the original displacement data are used, and the goodness of fit needs to be controlled above 95%. This is achieved through... The rate of change of displacement is obtained by taking the first derivative. The calculation formula is ,right The displacement acceleration is obtained by taking the second derivative. The calculation formula is ;

[0037] Extracting the angular features of the valve stem using angular sequence data. The formula for calculating the Fourier transform is as follows: ,in Indicates the frequency of data collection. This indicates that angle data is collected at a frequency of 10 degrees to obtain the rate of angle change. The calculation formula is: ;

[0038] Differential pressure gradient calculation is performed on the internal pressure data of valves and pipelines, and the differential pressure data is processed through a sliding window. The window is segmented, with a sliding window size of 100. Step size is 50 Segment the differential pressure data and define the differential pressure gradient. The calculation formula is ,in For two points in time within the window, This represents the change in pressure inside the pipe per unit time.

[0039] The velocity fluctuation coefficient was calculated by decomposing the velocity data using the db4 wavelet basis, reconstructing the velocity fluctuation components, and then calculating the velocity fluctuation coefficient. The formula is ,in The standard deviation of the pulse component This represents the average value of the flow velocity data;

[0040] The sealing pressure distribution entropy value is obtained by calculating the distribution entropy of the collected pressure data. The specific calculation formula is as follows: ,in Indicates the first Pressure values ​​at each collection point, with a collection cycle of 100. ;

[0041] Based on the numerical analysis of the extended feature dimensions, a time-series correlation map of valve motion parameters and fluid parameters is constructed, and the displacement change rate during the valve opening and closing process is selected. As a reference signal for calculating cross-correlation algorithms Select the flow velocity pulsation coefficient during the valve opening and closing process. As the target signal for calculating cross-correlation algorithms The correlation between the valve displacement change rate and the flow velocity pulsation coefficient over a time interval was calculated using a cross-correlation algorithm. ;

[0042] It should be further explained that the reference signal and the target signal are first preprocessed to extract the effective displacement change data and flow velocity data during the valve opening and closing movement period, and the displacement change rate is removed. and flow velocity fluctuation coefficient The mean of the values ​​is used as the data for calculating the cross-correlation function. The specific formula for calculating the function is shown below. in Indicates time delay, This represents the displacement change rate and flow velocity pulsation coefficient values ​​corresponding to the effective displacement change data and flow velocity data during the valve opening and closing movement period. Perform a correlation assessment when When a significant temporal correlation is found between the displacement change rate parameter and the flow velocity pulsation coefficient parameter, the correlation value is saved to construct a temporal correlation map. The correlation judgment threshold of 0.60 is obtained by selecting the same type of emergency shut-off valve and controlling the valve diameter to below DN500, collecting 300 sets of synchronous displacement change rate and flow velocity pulsation data, and calculating the cross-correlation function between the two. When the correlation data is 0.60, 98% of the valve opening and closing status detection parameters are effectively correlated. If the same type of emergency shut-off valve is selected and the valve diameter is controlled to above DN500, the cross-correlation function between the displacement change rate parameter and the flow velocity pulsation coefficient parameter needs to be recalculated.

[0043] The specific construction includes building a time-series correlation graph by calculating the correlation degree of the target valve's opening and closing data; using a cross-correlation algorithm to calculate the correlation degree between the valve stem displacement and angle data and the fluid parameter data in the pipe; constructing a directed graph from the obtained parameter correlation values, where the graph nodes represent the values ​​corresponding to each parameter; and including the valve movement nodes. and The fluid parameter data are arranged on the left side of the graph. and Arranged on the right side of the graph, the graph edge attribute is set to determine the correlation of the correlation calculation results. When there is a significant correlation between two parameters, a red edge line is used to indicate that they are not significantly correlated, and a gray edge line is used to indicate that they are not significantly correlated.

[0044] In S3, the construction of the spatiotemporal alignment feature matrix of correlation coefficients includes calculating the linear correlation coefficients of motion parameters and fluid parameters respectively using a construction method based on correlation maps, calculating the nonlinear correlation coefficients of sealing pressure distribution entropy and fluid parameters using grey relational analysis, and mapping the correlation coefficients to a unified spatiotemporal grid through spatiotemporal registration to form the spatiotemporal alignment matrix of correlation coefficients.

[0045] Further explanation is needed regarding the preliminary calculation of parameter correlation based on the correlation map. This involves performing quadratic linear and nonlinear correlation calculations between various parameters, improving the accuracy of the parameter correlation. Spatiotemporal registration processing is then performed based on the quadratic correlation calculation to construct a spatiotemporal aligned feature matrix. The quadratic linear correlation calculation uses the Pearson correlation coefficient to calculate the linear correlation between valve motion parameters and fluid parameters within the valve pipe. Here, displacement change rate and differential pressure gradient within the valve pipe are taken as examples to calculate their linear correlation. The specific calculation formula is as follows:

[0046] ,

[0047] Where V is the displacement change rate, G is the differential pressure gradient value, and i is the valve opening / closing state index. This represents the average value of the flow velocity data. The average value of the differential pressure gradient is represented by R, which ranges from ±1. A value greater than 0.7 indicates a strong correlation, while a value less than 0.3 indicates a weak correlation. Nonlinear correlation primarily processes fluid data within the valve pipe. Here, we focus on the nonlinear correlation between the sealing pressure distribution entropy and fluid parameters, selecting the velocity pulsation coefficient based on the fluid parameters. The specific calculation formula is as follows:

[0048] ,

[0049] Where H is the entropy value of the sealing pressure distribution. The value represents the flow velocity pulsation coefficient, and the valve on / off state (i) represents the sampling point index. This represents the average value of the entropy of the sealing pressure distribution. The average value of the flow velocity fluctuation coefficient, R, ranges from 0 to 1. The closer the R value is to 1, the stronger the correlation. Spatiotemporal registration is performed on the correlation results from the secondary calculation. Specifically, this includes time registration, which maps all parameter sampling points to a time axis with 1ms precision by establishing a unified time axis. Interpolation is then performed on the parameter data timestamps to ensure that the interpolated parameters have corresponding values ​​at the same time point, and the time error needs to be controlled within a certain range. Within milliseconds; spatial registration, with the valve center as the origin, defines three-dimensional spatial coordinates (x, y, z) and calibrates the sensor positions, records the installation position coordinates of each sensor in the spatial coordinate system, and maps the valve opening and closing status data parameter values ​​collected by sensors at different positions to a unified spatial grid to achieve spatial registration; a spatiotemporal alignment feature matrix is ​​constructed through spatiotemporal registration. The matrix is ​​defined by time dimension data, spatial dimension data, and valve opening and closing parameter dimension. The time dimension is divided at 0.1s intervals, taking the most recent 10s data for a total of 100 time nodes; the spatial dimension is divided according to the sensor installation position, taking the key monitoring points for a total of 8 spatial nodes; the parameter dimension includes 4 types of correlation parameters, forming 4 feature channels. The correlation coefficients after spatiotemporal registration are arranged according to the above dimensions to form a 100×8×4 three-dimensional spatiotemporal alignment matrix.

[0050] In S4, the parameters of the three-dimensional valve switching state judgment coordinate system are defined as follows: the X-axis represents the valve position achievement degree, the Y-axis represents the fluid state matching degree, and the Z-axis represents the parameter correlation degree. The position achievement degree includes the displacement change rate and the angle change rate. The fluid state matching degree includes the differential pressure gradient and the flow velocity pulsation coefficient. The correlation degree of valve switching state data parameters is the correlation coefficient in the spatiotemporal alignment feature matrix.

[0051] It should be further explained that, from the 100×8×4 three-dimensional spatiotemporal alignment matrix output from S3, data strongly correlated with the valve switching state are extracted. The qualified area for normal valve switching operation is calculated and divided using the parameters of each axis of the three-dimensional coordinate system. The X-axis represents the valve position achievement degree calculated by comparing the actual displacement change rate V collected by the laser displacement sensor with the theoretical full-stroke displacement L of the valve switching, and also by collecting the measured angular change rate from the angular displacement encoder. The theoretical fully closed valve angle A; the determination criteria for the X-axis are calculated according to the formula, the specific formula is as follows: The Y-axis represents the fluid state matching degree. The specific calculation formula is as follows: ,in The standard differential pressure and standard flow rate represent the valve opening and closing states, respectively. The Z-axis represents the correlation degree of the parameters with respect to the preprocessed spatiotemporal matrix. The average correlation coefficient of all valve parameters is calculated.

[0052] The valve opening / closing status is determined by a preset qualified area, and a dual-channel feedback mechanism is used to provide real-time feedback on the valve opening / closing status, including:

[0053] A qualified region is constructed based on historical data of valve switching status. This qualified region is determined based on statistical analysis of multi-condition experimental data and failure risk analysis. The qualified region is a three-dimensional cubic region with X-axis greater than 0.95, Y-axis greater than 0.90, and Z-axis greater than 0.75. The dual-channel feedback mechanism includes a physical channel and a communication channel. The physical channel receives the three-dimensional valve switching status judgment coordinates and status judgment results in real time. The communication channel transmits the valve switching status data to the gateway and feeds it back to the user in JSON format.

[0054] Further explanation is needed regarding the method for constructing the qualified region for valve switching status. This involves selecting a similar emergency shut-off valve under standard operating conditions (design pressure 1.6 MPa, medium is water or natural gas), simulating 1000 normal switching processes, and collecting raw parameters such as displacement change rate, angle change rate, and differential pressure gradient for each switch. Simultaneously, 500 typical failure scenarios (such as valve jamming, internal leakage, and sensor offset) are simulated, and the parameter values ​​at failure are recorded. For normal samples, the distribution ranges of the X, Y, and Z axes are calculated to determine the lower limit of the normal threshold. When (X, Y, Z) completely fall within the three-dimensional cube of the qualified region, the emergency shut-off valve is considered to be in normal working order. When X is less than 0.95, it indicates that the emergency shut-off valve position is not up to standard, caused by valve blockage or the valve not being fully opened or closed. When Y is less than 0.90, it indicates abnormal fluid in the valve pipeline, caused by pipe blockage or internal leakage. When Z is less than 0.75, it indicates disordered parameter correlation, caused by sensor malfunction or failure of the correlation between motion parameters and fluid parameters.

[0055] A dual-channel feedback mechanism is implemented. The physical channel uses an OLED display to show the coordinate values ​​for judging the three-dimensional valve switch status. The switch status of the emergency shut-off valve is determined by the status indicator light on the OLED display. When the indicator light is solid green, it indicates that the emergency shut-off valve is in a normal switch status. A flashing red light indicates that the X-axis is not up to standard because the emergency shut-off valve is not in the correct position. A flashing yellow light indicates that the Y-axis is not up to standard because the fluid parameters in the emergency shut-off valve pipeline are abnormal. A solid red light indicates that the Z-axis is not up to standard because the emergency shut-off valve switch status detection sensor is damaged or the parameter association coordination has failed. The communication channel uploads a data packet every 30 seconds when the emergency shut-off valve is in a normal switch status. The data packet contains the coordinate values ​​for judging the three-dimensional valve switch status and the corresponding timestamp. When the emergency shut-off valve is in an abnormal switch status, a data packet is uploaded immediately to achieve a millisecond-level response, and the abnormal status is fed back to the user in JSON format.

[0056] Please see Figure 5 As shown, this embodiment of the invention provides an on / off status detection system for an emergency shut-off valve, the specific modules of which include the following:

[0057] Sensor acquisition module: Real-time acquisition of the on / off status data of the target valve by installing multi-source sensors. The on / off status data includes valve displacement data, angle data, and fluid parameter data in the valve pipe.

[0058] Data analysis module: Based on the switch status data, feature dimension expansion analysis is performed to obtain displacement change rate, angle change rate, differential pressure gradient, flow velocity pulsation coefficient and sealing pressure distribution entropy value. The time series correlation map between valve motion parameters and fluid parameters is established through cross-correlation algorithm.

[0059] Data processing module: Based on the temporal correlation map of valve motion parameters and fluid parameters, the correlation coefficient between valve motion parameters and fluid parameters is obtained, and the correlation coefficient is spatiotemporally registered to obtain the spatiotemporal alignment feature matrix of the correlation coefficient.

[0060] Result Judgment and Feedback Module: Based on the spatiotemporal alignment feature matrix of correlation coefficient, a three-dimensional valve switch status judgment coordinate system is constructed. The valve switch status is judged by a preset qualified area in the three-dimensional valve switch status judgment coordinate system. A dual-channel feedback mechanism is adopted to provide real-time feedback on the valve switch status.

[0061] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other.

[0062] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for detecting the on / off status of an emergency shut-off valve, characterized in that, include: S1: Preset a specified type of emergency shut-off valve as the target valve, and collect the switch status data of the target valve in real time by installing multi-source sensors. The switch status data includes the valve's motion parameter data and the fluid parameter data in the valve pipe. S2: Based on the switch status data collected in S1, feature dimension expansion analysis is performed to obtain valve motion parameters including displacement change rate and angle change rate, and fluid parameters including differential pressure gradient, flow velocity pulsation coefficient and sealing pressure distribution entropy value. A time-series correlation map between valve motion parameters and fluid parameters is established through cross-correlation algorithm. S3: Based on the temporal correlation map of valve motion parameters and fluid parameters in S2, the correlation coefficient between valve motion parameters and fluid parameters is obtained. The correlation coefficient is spatiotemporally registered to obtain the spatiotemporal alignment feature matrix of the correlation coefficient. S4: Based on the spatiotemporal alignment feature matrix of the correlation coefficient in S3, a three-dimensional valve opening / closing state judgment coordinate system is constructed. The coordinate system includes an X-axis representing the valve position achievement degree, a Y-axis representing the fluid state matching degree, and a Z-axis representing the parameter correlation degree. The valve opening / closing state is judged by setting a qualified area in the coordinate system. A dual-channel feedback mechanism is adopted to provide real-time feedback on the valve opening / closing state. The valve position achievement degree includes the displacement change rate and the angle change rate. The fluid state matching degree includes the differential pressure gradient and the flow velocity pulsation coefficient. The parameter correlation degree is the correlation coefficient in the spatiotemporal alignment feature matrix.

2. The method for detecting the on / off status of an emergency shut-off valve according to claim 1, characterized in that, The motion parameter data of the valve in S1 includes: Displacement data is used to record the rate of change of valve stem displacement when the valve is open and closed, while angle data is used to record the rate of change of valve stem angle when the valve is open and closed.

3. The method for detecting the on / off status of an emergency shut-off valve according to claim 1, characterized in that, The fluid parameter data in S1 includes: The fluid pressure data and fluid velocity data are recorded. The fluid pressure data records the differential pressure gradient and sealing pressure distribution entropy value in the pipe when the target valve is in the open and closed states. The fluid velocity data records the velocity pulsation coefficient in the pipe when the target valve is in the open and closed states.

4. The method for detecting the on / off status of an emergency shut-off valve according to claim 1, characterized in that, The feature dimension expansion analysis in S2 includes: Displacement feature extraction, for the original displacement sequence Perform 5th-order polynomial least squares fitting, for The rate of change of displacement is obtained by taking the first derivative. ,right The displacement acceleration is obtained by performing second-order differentiation; angle feature extraction is performed on the angle sequence data. Perform a Fourier transform to obtain the rate of change of angle. Differential pressure gradient calculation uses a sliding window to process differential pressure data. Segmentation is performed to obtain the differential pressure gradient. The velocity fluctuation coefficient is calculated by decomposing the velocity data using a wavelet basis. The sealing pressure distribution entropy value is obtained by calculating the distribution entropy of the collected pressure data. .

5. The method for detecting the on / off status of an emergency shut-off valve according to claim 1, characterized in that, The cross-correlation algorithm in S2 establishes the time-series correlation graph, including: Select the displacement change rate during the valve opening and closing process As a reference signal for calculating cross-correlation algorithms Select the flow velocity pulsation coefficient during the valve opening and closing process. As the target signal for calculating cross-correlation algorithms The correlation between the valve displacement change rate and the flow velocity pulsation coefficient over a time interval was calculated using a cross-correlation algorithm. ,based on A directed graph structure is used to construct the association graph.

6. The method for detecting the on / off state of an emergency shut-off valve according to claim 1, characterized in that, The spatiotemporal alignment feature matrix for constructing the correlation coefficient in S3 includes: Based on the construction method of the correlation map in S2, the linear correlation coefficients of motion parameters and fluid parameters are calculated respectively. The nonlinear correlation coefficients of sealing pressure distribution entropy and fluid parameters are calculated by grey relational analysis. The correlation coefficients are mapped to a unified spatiotemporal grid through spatiotemporal registration to form a spatiotemporal alignment feature matrix of correlation coefficients.

7. The method for detecting the on / off state of an emergency shut-off valve according to claim 1, characterized in that, The qualified area and dual-channel feedback mechanism set in S4 include: A qualified region is constructed based on historical data of valve switching status. The qualified region is a three-dimensional cubic region with X-axis greater than 0.95, Y-axis greater than 0.90, and Z-axis greater than 0.

75. The dual-channel feedback mechanism includes a physical channel and a communication channel. The physical channel receives the coordinate values ​​and status judgment results of the three-dimensional valve switching status in real time, while the communication channel transmits the valve switching status data to the gateway and feeds it back to the user in JSON format.

8. A system for detecting the on / off status of an emergency shut-off valve, characterized in that, Includes the following modules: Sensor acquisition module: Real-time acquisition of the on / off status data of the target valve by installing multi-source sensors. The on / off status data includes valve displacement data, angle data, and fluid parameter data in the valve pipe. Data analysis module: Based on the switch status data, feature dimension expansion analysis is performed to obtain displacement change rate, angle change rate, differential pressure gradient, flow velocity pulsation coefficient and sealing pressure distribution entropy value. The time series correlation map between valve motion parameters and fluid parameters is established through cross-correlation algorithm. Data processing module: Based on the temporal correlation map of valve motion parameters and fluid parameters, the correlation coefficient between valve motion parameters and fluid parameters is obtained, and the correlation coefficient is spatiotemporally registered to obtain the spatiotemporal alignment feature matrix of the correlation coefficient. Result Judgment and Feedback Module: Based on the spatiotemporal alignment feature matrix of correlation coefficient, a three-dimensional valve switch status judgment coordinate system is constructed. The valve switch status is judged by a preset qualified area in the three-dimensional valve switch status judgment coordinate system. A dual-channel feedback mechanism is adopted to provide real-time feedback on the valve switch status.

Images