CN-121783538-B - Safety valve on-line checking and performance evaluation system

CN121783538BCN 121783538 BCN121783538 BCN 121783538BCN-121783538-B

Abstract

The invention relates to the technical field of safety valve testing and monitoring, and particularly discloses an online safety valve checking and performance evaluating system. The system comprises a field data acquisition module, an edge calculation and feature extraction module, a cloud performance evaluation and decision module and a man-machine interaction module. The multi-dimensional characteristics are extracted by synchronously collecting pressure, acoustic emission, vision and process data, and comprehensive evaluation and decision are performed based on dynamic performance base lines, multi-parameter fusion degradation models and system cooperativity analysis. The invention realizes the span from single parameter verification to system level performance prediction evaluation, and provides a solution for the predictive maintenance and full life cycle management of the safety valve.

Inventors

KOU YONGGANG
Lv kaixuan
XUE FEI
WANG XUAN
ZHANG LIQUAN
Yin Xumiao
FU LEI
ZHANG LILI
ZHAO XIANGKUI
JIANG LIYUAN

Assignees

内蒙古容特检测科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260204

Claims (9)

1. The safety valve on-line verification and performance evaluation system is characterized by comprising: The system comprises a safety valve, a field data acquisition module, a control module and a control module, wherein the field data acquisition module is used for synchronously acquiring multi-source heterogeneous real-time process data in an online verification process of the safety valve and comprises a high-dynamic pressure sensor array, a non-contact acoustic emission sensor, a high-frame-rate industrial vision unit and a process interface unit; the edge calculation and feature extraction module is deployed in the field side industrial gateway equipment and is used for carrying out real-time preprocessing and primary feature extraction on the original data acquired by the field data acquisition module; The cloud performance evaluation and decision module is used as a core analysis engine of the system, receives the preprocessed multi-dimensional characteristic data set uploaded by the edge calculation and characteristic extraction module and performs deep performance modeling and evaluation; the human-computer interaction and report generation module is used for displaying the evaluation result to the user and providing an interaction interface; The cloud performance evaluation and decision module comprises a dynamic performance baseline library, a multi-parameter fusion degradation model, a system collaborative analysis unit and a risk evaluation and decision unit; The dynamic performance baseline library stores the reference performance feature vector established by the system after each controlled safety valve is initially commissioned or last offline overhaul; the multi-parameter fusion degradation model is used for quantitatively evaluating the performance attenuation degree of the safety valve, firstly, the mahalanobis distance between the feature vector extracted in the current verification period and the corresponding reference vector in the dynamic performance baseline library is calculated as a comprehensive deviation index, time sequence analysis is introduced to arrange the comprehensive deviation indexes of the past verification in time sequence, and an exponential weighted moving average algorithm is adopted to fit the change trend of the comprehensive deviation index to calculate the degradation slope representing the performance attenuation rate; The system cooperativity analysis unit is used for evaluating the response cooperativity of the safety valve in the whole safety instrument loop, and analyzing the time difference between the reading of the upstream pressure transmitter acquired from the process interface unit and the detection of the starting movement of the valve rod by the visual characteristic analysis unit as system response delay, and simultaneously analyzing the time difference between the output of the action command by the logic controller and the actual take-off of the valve as valve execution delay, and evaluating the stability and the change trend of the delay time by comparing historical data; the risk assessment and decision unit carries out comprehensive decision based on the comprehensive deviation degree and the degradation slope output by the multi-parameter fusion degradation model and the delay time stability index output by the system collaborative analysis unit, and the unit presets a plurality of decision thresholds; Generating an early warning signal and suggesting a shortened verification period when the integrated departure is greater than a1 st threshold but the degradation slope is not greater than a2 nd threshold, generating a performance degradation alarm and suggesting preventive maintenance when the integrated departure and degradation slope are both greater than their respective thresholds, and generating a system cooperativity alarm when the system response delay or the fluctuation coefficient of the valve execution delay is greater than a 3 rd threshold.
2. The safety valve on-line verification and performance evaluation system of claim 1, wherein the high dynamic pressure sensor array continuously collects pressure pulsation signals of the safety valve inlet pipeline; the non-contact acoustic emission sensor is used for collecting high-frequency acoustic emission signals generated in the whole valve action process; the high frame rate industrial vision unit captures displacement track images of the valve rod; the process interface unit obtains upstream pressure transmitter readings associated with the safety valve, output status signals of the logic controller, and temperature and flow data of the process medium in real time from a process control system of the plant via an industrial bus protocol.
3. The system for online verification and performance evaluation of a safety valve according to claim 2, wherein the edge computing and feature extraction module comprises a signal synchronization and time scale alignment unit, a time domain and frequency domain feature extraction unit and a visual feature analysis unit; The signal synchronization and time mark alignment unit receives data streams from the high-dynamic pressure sensor, the acoustic emission sensor, the industrial visual unit and the process interface unit, and unifies all the data streams to a time reference with the same millisecond precision based on a hardware time stamp and a software interpolation algorithm; The time domain and frequency domain feature extraction unit processes the synchronized pressure signal and acoustic emission signal, extracts a jump pressure instantaneous value, a pressure rising rate maximum value, a steady discharge pressure average value, a seat returning pressure instantaneous value and a pressure falling rate for the pressure signal, performs fast Fourier transform on the pressure signal to extract main frequency energy and 3 rd, 5 th and 7 th harmonic component amplitude values, and extracts event counting rate, absolute energy, ringing count, amplitude distribution and each sub-band energy entropy obtained through wavelet packet decomposition for the acoustic emission signal; the visual characteristic analysis unit detects and tracks the sub-pixel level edge of the valve rod displacement image sequence, calculates the displacement-time curve of the valve rod from rest to full-open stroke, and extracts the average movement speed, the maximum acceleration and the action delay time of the valve rod from the displacement-time curve.
4. The safety valve on-line verification and performance evaluation system of claim 3, wherein the human-machine interaction and report generation module comprises a multi-dimensional data visualization interface, a structured report generator, and a maintenance worksheet docking interface; the multidimensional data visual interface displays key parameters, comprehensive deviation indexes, degradation trend curves and a system delay analysis chart of the current verification in real time in a form of an instrument panel; the structured report generator automatically integrates the original data, the extracted characteristics, the model evaluation result and the decision advice of each verification to generate a verification and evaluation report which accords with the industry standard; the maintenance work order interfacing interface automatically pushes equipment information, fault descriptions and recommended actions to a computerized maintenance management system of the factory to trigger a work order creation process when the risk assessment and decision unit generates maintenance recommendations.
5. The system for online verification and performance evaluation of a safety valve according to claim 4, wherein the following process is performed for calculating the mahalanobis distance in the multiparameter fusion degradation model: Firstly, a reference feature vector of a current safety valve and a covariance matrix thereof are called from a dynamic performance baseline library; and finally, multiplying the difference value vector of the feature vector and the reference feature vector by the inverse matrix of the reference covariance matrix, and multiplying the inverse matrix by the transpose of the difference value vector, wherein the square root of the obtained scalar value is the Mahalanobis distance.
6. The online check and performance evaluation system of claim 5, wherein the update mechanism of the dynamic performance baseline library is as follows: after the safety valve completes the offline disassembly maintenance or the replacement of key components after the safety valve is confirmed to be qualified once, executing a complete online verification process; Taking the mean value of the feature vector extracted by the verification as a new reference feature vector, and taking the covariance of the feature vector repeatedly measured by the verification as a new reference covariance matrix, thereby updating the dynamic performance baseline of the valve.
7. The on-line check and performance evaluation system of a safety valve according to claim 6, wherein the high dynamic pressure sensor array in the field data acquisition module employs a plurality of pressure sensors distributed in equidistant fashion along the safety valve inlet pipe axis; the signal synchronization and time scale alignment unit performs weighted average processing on pressure signals of the pressure sensors to eliminate the influence of local flow field disturbance, and a synthetic signal representing average pressure of the inlet section of the valve is obtained.
8. The safety valve on-line verification and performance evaluation system according to claim 7, wherein the system is deployed in an industrial internet platform in a micro-service architecture; The field data acquisition module and the edge calculation and feature extraction module form an edge micro-service; the cloud performance evaluation and decision module and the man-machine interaction and report generation module form a cloud micro service; and the edge micro-service and the cloud micro-service perform data asynchronous communication through a message queue encrypted by adopting a transmission layer security protocol.
9. The system for online verification and performance evaluation of a safety valve according to claim 8, wherein the visual feature analysis unit extracts a valve stem contour by combining a Canny edge detection algorithm with a morphological closing operation, and calculates a valve stem displacement-time curve by using a sub-pixel registration algorithm based on phase correlation.

Description

Safety valve on-line checking and performance evaluation system Technical Field The invention belongs to the technical field of safety valve testing and monitoring, and particularly relates to an online safety valve checking and performance evaluating system. Background In the field of industrial safety, safety instrument systems are a core line of defense for ensuring safe operation of critical process facilities (e.g., petrochemical industry, energy production). Safety valves are critical final actuators in safety instrumented systems, and the reliability of their performance is directly related to the safety integrity level of the overall system. The safety valve is checked and evaluated regularly and accurately, and is a necessary technical means for ensuring the function of the safety valve to be effective and preventing catastrophic accidents. The online verification technology of the safety valve aims at testing and verifying key performance parameters of the valve under the condition of not interrupting the process. The traditional online checking method mainly focuses on measuring the tripping pressure and the seat returning pressure of the safety valve, namely checking whether the safety valve can accurately act at a set pressure point. The core goal of this approach is to confirm the immediate accuracy of the valve mechanical action. The safety valve is generally regarded as an independent mechanical part for verification in the prior art, and the verification logic and evaluation system of the safety valve have the following problems. The verification process only collects and focuses on a single operating pressure data, and completely ignores the overall performance attribute that the safety valve should possess as an organic component in the safety instrumented system circuit. This results in an inability to evaluate the valve's performance decay tendency, stability of response time over long term operation, and its system reliability in conjunction with upstream sensors and logic controllers. The 'pass/fail' binary conclusion provided by the traditional method lacks quantitative analysis on the residual life of the valve, potential failure modes and contribution to the reliability of the whole safety loop, so that the preventive maintenance strategy of the factory lacks data support, and the risk-based inspection is difficult to realize. Accordingly, a solution that enables systematic online verification and comprehensive performance assessment of safety valves is desired. Disclosure of Invention The invention aims to provide an online check and performance evaluation system for a safety valve, which is used for solving the contradiction that the prior art can only check the jump and seat returning pressure of the safety valve and cannot quantitatively evaluate the overall performance, long-term reliability and system cooperativity of the safety valve as a loop component of a safety instrument system. The invention provides an online check and performance evaluation system of a safety valve, which comprises the following components: The field data acquisition module is used for synchronously acquiring multi-source heterogeneous real-time process data in the online verification process of the safety valve; the edge calculation and feature extraction module is deployed in the field side industrial gateway equipment and is used for carrying out real-time preprocessing and primary feature extraction on the original data acquired by the field data acquisition module; The cloud performance evaluation and decision module is used as a core analysis engine of the system, receives the preprocessed multi-dimensional characteristic data set uploaded by the edge calculation and characteristic extraction module and performs deep performance modeling and evaluation; and the man-machine interaction and report generation module is used for displaying the evaluation result to the user and providing an interaction interface. Preferably, the field data acquisition module comprises a high dynamic pressure sensor array, a non-contact acoustic emission sensor, a high frame rate industrial vision unit and a process interface unit; the high dynamic pressure sensor array continuously collects pressure pulsation signals of an inlet pipeline of the safety valve; the non-contact acoustic emission sensor is arranged at a specific position outside the valve body and is used for collecting high-frequency acoustic emission signals generated in the whole valve action process; the high frame rate industrial vision unit captures displacement track images of the valve rod; the process interface unit obtains upstream pressure transmitter readings associated with the safety valve, output status signals of the logic controller, and temperature and flow data of the process medium in real time from a process control system of the plant via an industrial bus protocol. Preferably, the edge computing and feature extracting module compri