A control system functional safety quantitative evaluation system based on digital twin technology

Abstract

The present invention relates to the technical field of data analysis, and specifically discloses a control system functional safety quantitative evaluation system based on digital twin technology, including an evaluation module, an analysis module and an optimization module. Specifically: Evaluation module: Establish a digital twin model of the control system, extract the operation data of the control system based on the digital twin model, and obtain the functional safety quantitative score of the control system based on the operation data; Analysis module: Obtain sub-curves on the fitting curve and screen reference points according to the sub-curves; Optimization module: Adjust the maintenance cycle according to the distribution of the reference points. The present invention realizes the dynamic monitoring of the safety state of the control system.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and particularly to a control system functional safety quantitative evaluation system based on digital twin technology. Background Art

[0002] Digital twin is a simulation process that fully utilizes data such as physical models, sensors, and operation history, integrates multiple disciplines, multiple physical quantities, multiple scales, and multiple probabilities, and completes mapping in the virtual space to reflect the entire life cycle process of the corresponding physical equipment.

[0003] In traditional control system functional safety assessment practices, single evaluations are mainly carried out: through methods such as hazard analysis, fault trees, and quantitative probability calculations, the safety integrity level or probability index is calculated, and then corresponding treatment measures are taken accordingly. However, this method only stays at a certain static time point and lacks the dynamic mining and utilization of the long-term operation history data of the system. Summary of the Invention

[0004] The purpose of the present invention is to provide a control system functional safety quantitative evaluation system based on digital twin technology to solve the following technical problems:

[0005] In traditional control system functional safety assessment practices, single evaluations are mainly carried out: through methods such as hazard analysis, fault trees, and quantitative probability calculations, the safety integrity level or probability index is calculated, and then corresponding treatment measures are taken accordingly. However, this method only stays at a certain static time point and lacks the dynamic mining and utilization of the long-term operation history data of the system.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A control system functional safety quantitative evaluation system based on digital twin technology includes an evaluation module, an analysis module, and an optimization module. Specifically:

[0008] Evaluation module: Establish a digital twin model of the control system, extract the operation data of the control system based on the digital twin model, and obtain the functional safety quantitative score of the control system based on the operation data;

[0009] Analysis module: Periodically obtain the functional safety quantitative score within a preset monitoring period, generate a fitting curve, obtain the monotonically decreasing part of the fitting curve, denoted as the sub-curve. When the length of the domain of the sub-curve is greater than a preset duration threshold T and the functional safety quantitative score corresponding to the starting point of the sub-curve is less than a preset score threshold, use the starting point of the sub-curve as the reference point;

[0010] Optimization module: Sort the reference points in the order of the time axis, calculate the time interval corresponding to the reference points at two adjacent sorted positions, and denote it as the target interval;

[0011] Statistical difference between the target interval and a preset time interval, denote the difference less than the preset difference threshold as the abnormal difference, calculate the proportion C of the abnormal difference in the difference, and calculate the new maintenance cycle JC = (1 - C) / jc, where jc represents the preset maintenance cycle.

[0012] As a further solution of the present invention: The process of obtaining the functional safety quantification score of the control system based on the operation data includes:

[0013] Normalize the operation data into positive data, and the positive data is positively correlated with the functional safety of the control system;

[0014] Determine the weights of the positive data based on Sobol sensitivity analysis, and perform weighted summation on the positive data to obtain the functional safety quantification score of the control system.

[0015] As a further solution of the present invention: The process of generating the fitting curve includes:

[0016] Generate coordinate points (t b , P b ), where P b represents the functional safety quantification score obtained at the b-th time, and t b represents the time point when the functional safety quantification score is obtained at the b-th time. Fit the coordinate points to obtain the fitting curve f(t), where t represents time.

[0017] As a further solution of the present invention: The process of obtaining the reference point further includes:

[0018] Obtain the time point T1 corresponding to the end point of the sub-curve Z1. When the point A corresponding to the time point T1 - T is on the sub-curve Z1 and the functional safety quantification score corresponding to the point A is less than the preset value, use the point A as the reference point.

[0019] As a further solution of the present invention: In the optimization module, when the maintenance cycle JC ≤ T2, set the maintenance cycle JC = T2, where T2 represents the shortest length of the preset maintenance cycle.

[0020] As a further solution of the present invention: In the analysis module, the following steps are further included:

[0021] When the number of reference points is less than the preset number threshold, do not adjust the maintenance cycle.

[0022] Advantages of the present invention: Compared with the prior art:

[0023] 1) Real-time dynamic evaluation based on the digital twin model can continuously collect and quantify functional safety indicators throughout the entire system operation process, breaking through the limitations of traditional single static evaluation and realizing refined and real-time monitoring of the safety status of the control system. This invention can timely reflect the potential risks of the system through in-depth mining of operating data, shorten the response time of safety blind spots, improve the accuracy of fault identification, provide a scientific basis for operation and maintenance decisions, and significantly enhance the system's early warning capabilities for sudden anomalies, thereby effectively reducing the probability of safety accidents and improving overall safety and reliability;

[0024] 2) Use the analysis module to generate a fitting curve and extract a monotonically decreasing sub-curve as a reference point. By presetting the duration threshold and the quantitative score judgment, the starting point of functional safety decline can be accurately located to achieve early capture of the downward trend of system safety performance. This method can periodically track potential fault hazards, avoid the subjectivity of manual experience judgment, and improve prediction accuracy; at the same time, by monitoring the evolution of historical curves, it can guide operation and maintenance personnel to optimize maintenance strategies in a targeted manner, reduce human intervention and blind maintenance, and not only improve maintenance efficiency, but also effectively control maintenance costs;

[0025] 3) The optimization module automatically calculates the new maintenance cycle JC based on the difference between the reference point time interval and the preset maintenance cycle, and realizes dynamic adaptive adjustment of the maintenance strategy. This mechanism can automatically correct the maintenance frequency according to the actual operation performance of the system, so that the maintenance cycle is neither too frequent to waste resources nor too long to accumulate safety risks; in addition, the maintenance frequency is adjusted by statistically analyzing the abnormal difference ratio C, taking into account both system availability and operational safety, minimizing downtime losses, and continuously improving the maintenance plan during long-term operation, avoiding sudden downtime caused by "over-guarantee operation" and preventing "excessive maintenance" from wasting resources, achieving simultaneous optimization of maintenance costs and safety and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The present invention will be further described below in conjunction with the accompanying drawings.

[0027] Figure 1 It is a structural schematic diagram of a control system functional safety quantitative evaluation system based on digital twin technology of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] See also Figure 1As shown in the figure, the present invention is a control system functional safety quantitative evaluation system based on digital twin technology, including an evaluation module, an analysis module, and an optimization module. Specifically:

[0030] Evaluation module: Establish a digital twin model of the control system, extract the operation data of the control system based on the digital twin model, and obtain the functional safety quantitative score of the control system based on the operation data;

[0031] In a preferred embodiment of the present invention, the process of obtaining the functional safety quantitative score of the control system based on the operation data includes:

[0032] Normalize the operation data into positive data, and the positive data is positively correlated with the functional safety of the control system;

[0033] Determine the weights of the positive data based on Sobol sensitivity analysis, and perform weighted summation on the positive data to obtain the functional safety quantitative score of the control system;

[0034] It should be noted that the entire digital twin construction process uses an object-oriented modeling language (such as Modelica) as a carrier and performs virtual mapping according to the three-level structure of the device layer, the sensing layer, and the control logic layer. First, in the Modelica environment, standardized components such as gates, pump stations, level-flow sensors, and PID regulators are called, and the model framework is built based on physical equations and parametric interfaces to achieve the organic coupling between subsystems. Secondly, the measured data for the past 90 days (including upstream level, downstream flow, valve opening change curves, etc.) are segmented and input into the model, and key parameters are optimized and calibrated through parameter identification algorithms (such as the least squares method or adaptive gradient descent). The identification process relies on automated script-driven iterative solution. After each simulation, the mean square error between the simulation output and the measured sequence is calculated and fed back to the optimization module to adjust the parameter search step size and direction until the global error converges within 2%, ensuring that the model can accurately reproduce the real system dynamic response under different working conditions;

[0035] In the model verification stage, the historical data is divided into a training set and a verification set using the holdout method. While the error on the verification set is maintained within 2%, the stability requirements under fault injection and extreme working conditions also need to be met. By comparing the simulation results with the measured fluctuation curves, the ability of the model to capture sudden changes (such as high-flow shocks or rapid level changes) can be further evaluated. When the model shows high consistency under both normal and abnormal working conditions, its high-fidelity characteristics can be confirmed, providing a reliable basis for subsequent functional safety quantitative evaluation. This high-fidelity digital twin model can not only output real-time safety indicators but also be used to carry out "what-if" scenario simulations to simulate the impact of different maintenance strategies or working condition changes on system safety, thereby supporting decision-making optimization;

[0036] Subsequently, n (the specific value can be freely set) key process variables (such as upstream water level, gate opening, pump outlet pressure, etc.) are captured from the 1Hz running data stream of the twin for real-time playback. Each dimension is normalized to 0-1 within the interval [lower limit, upper limit], and the direction is unified: if the increase of the variable represents an improvement in safety, the positive direction is maintained, otherwise its opposite number is taken (it can also be other methods, which are not limited here), so that all data is monotonically positively correlated with functional safety; immediately afterwards, the Sobol global sensitivity analysis is performed on the normalized matrix, and the first-order and total effect indices are extracted and normalized into a weight vector. For example, the upstream water level W1 = 0.21, the gate opening W2 = 0.15, the pump outlet pressure W3 = 0.09, etc., and the sum of all weights is 1; finally, the weighted sum of the positive data within the same monitoring period is calculated according to the corresponding weights and mapped to the 0-100 interval to obtain the current functional safety quantification score, such as 83.6 points. The lower the value, the smaller the safety margin, thus providing a single, comparable and traceable metric benchmark for subsequent trend analysis and maintenance optimization.

[0037] Analysis module: During the preset monitoring period, the functional safety quantification score is periodically obtained, a fitting curve is generated, and the monotonically decreasing part of the fitting curve is obtained and denoted as a sub-curve. When the length of the domain of the sub-curve is greater than the preset duration threshold T and the functional safety quantification score corresponding to the starting point of the sub-curve is less than the preset score threshold, the starting point of the sub-curve is used as a reference point;

[0038] In another preferred embodiment of the present invention, the process of generating the fitting curve includes:

[0039] Generate coordinate points (t b , P b ), where P b represents the functional safety quantification score obtained at the b-th time, and t b represents the time point when the functional safety quantification score is obtained at the b-th time. The coordinate points are fitted to obtain the fitting curve f(t), where t represents time;

[0040] In another preferred embodiment of the present invention, the process of obtaining the reference point further includes:

[0041] Obtain the time point T1 corresponding to the end point of the sub-curve Z1. When the point A corresponding to the time point T1 - T is on the sub-curve Z1 and the functional safety quantification score corresponding to the point A is less than the preset value, the point A is used as the reference point;

[0042] It is worth noting that, first, a monitoring period of, for example, 24 hours and a sampling period of 60 seconds are set, and the functional safety quantification scores are read from the evaluation module at regular intervals to form a discrete point sequence (tb, Pb); subsequently, cubic splines are used to perform least squares fitting on all coordinate points to generate a continuous function f(t), and sub-curves and reference points are extracted;

[0043] Exemplarily, a section of pump station operation data shows a continuous decline during the period from 10:15 to 11:05. Δt = 50 min > T, and the score at t1 = 10:15 is 72.3 < 75, then 10:15 is recorded as a reference point; if it still falls within the monotonically decreasing interval when tracing back T minutes from the end point T1 of the same sub-curve, and the score at this point is also lower than 75, then this point is also supplemented as a reference point. Through this "sampling - fitting - segmentation - threshold filtering" process, the system can, without manual intervention, capture in real time the sections where the functional safety deteriorates continuously and has a significant time span, and output accurate time coordinates, providing a reliable basis for subsequent optimization of the maintenance cycle.

[0044] Optimization module: Sort the reference points in the order of the time axis, calculate the time interval corresponding to the reference points at two adjacent sorted positions, and denote it as the target interval;

[0045] Statistically analyze the difference between the target interval and the preset time interval, denote the difference less than the preset difference threshold as the abnormal difference, statistically analyze the proportion C of the abnormal difference in the difference, and calculate the new maintenance cycle JC = (1 - C) / jc, where jc represents the preset maintenance cycle;

[0046] In a preferred embodiment of the present invention, in the optimization module, when the maintenance cycle JC ≤ T2, let the maintenance cycle JC = T2, where T2 represents the shortest length of the preset maintenance cycle;

[0047] It should be noted that in the analysis module, the following steps are further included:

[0048] When the number of reference points is less than the preset number threshold, the maintenance cycle is not adjusted.

[0049] It can be understood that by arranging the reference points in sequence, calculating the adjacent time intervals and comparing them with the expected intervals, the system can gain real-time insight into the speed change of the functional safety deterioration; if the proportion of short intervals increases, indicating that the risk accumulation accelerates, the algorithm automatically compresses the new maintenance cycle, and vice versa, it maintains or moderately extends it. At the same time, the shortest cycle constraint and the reference point quantity threshold jointly suppress the misjudgment caused by accidental noise, prevent the waste of resources due to overly frequent maintenance, and also avoid sudden shutdowns caused by maintenance lags. This mechanism enables the maintenance decision to dynamically expand and contract with the health status, establishing an adaptive balance between reliability and economy, and ultimately significantly improving the safety margin and full-life efficiency of the control system.

[0050] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data and performing software simulations to get a formula that is closest to the actual situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0051] The above has described in detail one embodiment of the present invention, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the application of the present invention shall still fall within the scope covered by the patent of the present invention.

Claims

1. A control system functional safety quantitative evaluation system based on digital twin technology, characterized in that It includes an evaluation module, an analysis module, and an optimization module. Specifically: Evaluation module: Establish a digital twin model of the control system, extract the operation data of the control system based on the digital twin model, and obtain the functional safety quantification score of the control system based on the operation data; Analysis module: Periodically obtain the functional safety quantification score within a preset monitoring period, generate a fitting curve, obtain the monotonically decreasing part of the fitting curve, denoted as the sub-curve. When the length of the domain of the sub-curve is greater than the preset time threshold T and the functional safety quantification score corresponding to the starting point of the sub-curve is less than the preset score threshold, use the starting point of the sub-curve as the reference point; Optimization module: Sort the reference points in chronological order of the time axis, calculate the time interval corresponding to adjacent reference points in the sorted positions, denoted as the target interval; Statistically analyze the difference between the target interval and the preset time interval, denote the difference less than the preset difference threshold as the abnormal difference, statistically analyze the proportion C of the abnormal difference in the difference, and calculate the new maintenance cycle JC = (1 - C) / jc, where jc represents the preset maintenance cycle.

2. The functional safety quantitative evaluation system for a control system based on digital twin technology according to claim 1, characterized in that, The process of obtaining the functional safety quantification score of the control system based on the operation data includes: Normalize the operation data into positive data, and the positive data is positively correlated with the functional safety of the control system; Determine the weights of the positive data based on Sobol sensitivity analysis, and perform weighted summation on the positive data to obtain the functional safety quantification score of the control system.

3. The functional safety quantitative evaluation system of a control system based on digital twin technology according to claim 1, wherein The process of generating the fitting curve includes: Generate coordinate points (t b , P b ), where P b represents the functional safety quantification score obtained at the b-th time, and t b represents the time point when the functional safety quantification score is obtained at the b-th time. A fitting curve f(t) is obtained by fitting the coordinate points, where t represents time.

4. A functional safety quantitative evaluation system for a control system based on digital twin technology according to claim 1, characterized in that, The process of obtaining the reference point also includes: Obtain the time point T1 corresponding to the end point of the sub-curve. When the point A corresponding to the time point T1 - T is on the sub-curve and the functional safety quantification score corresponding to point A is less than the preset value, use point A as the reference point.

5. A functional safety quantitative evaluation system for a control system based on digital twin technology according to claim 1, characterized in that, In the optimization module, when the maintenance cycle JC ≤ T2, set the maintenance cycle JC = T2, where T2 represents the shortest length of the preset maintenance cycle.

6. A control system functional safety quantitative evaluation system based on digital twin technology according to claim 1, characterized in that, In the analysis module, the following steps are also included: When the number of reference points is less than the preset number threshold, do not adjust the maintenance cycle.

Images