A DCS measuring point redundancy fault-tolerant and fault identification method

Abstract

The application belongs to the technical field of DCS fault detection, and particularly relates to a DCS measuring point redundancy fault tolerance and fault identification method, which comprises the following steps: collecting signals of each measuring point in the DCS in real time and obtaining quality state information; generating quality evaluation coefficients of each measuring point according to the quality state information; performing consistency test on the measurement values of each measuring point in the same redundancy group and calculating deviation amounts; determining the comprehensive confidence weight of each measuring point according to the quality evaluation coefficients and the deviation amounts; performing weighted fusion on the measurement values in the redundancy group based on the comprehensive confidence weight to generate a fusion output value; identifying the fault type of the measuring point according to the quality evaluation coefficients, the deviation amounts and historical state data; and performing redundancy switching and fault isolation according to the fault type and the comprehensive confidence weight. The application can finely evaluate the quality of the redundancy measuring point signals, accurately identify the fault of the measuring point and dynamically process the fault tolerance, thereby effectively improving the reliability and availability of the DCS system measuring point signals.

Description

Technical Field

[0001] This invention belongs to the field of DCS fault detection technology, specifically a DCS measurement point redundancy tolerance and fault identification method. Background Technology

[0002] Distributed Control Systems (DCS) are core automation control systems widely used in process industries such as power, petroleum, chemical, and metallurgy. DCS systems acquire signals from field instruments and sensors through numerous input / output (I / O) measurement points to achieve comprehensive monitoring and control of the production process. Due to the complex and harsh industrial environment, measurement signals are susceptible to various interferences and faults during acquisition, transmission, and processing, leading to inaccurate or even completely invalid measurement data. This can potentially cause control deviations, equipment malfunctions, and even production safety accidents.

[0003] To improve the reliability of DCS systems, redundancy configuration schemes are commonly used in existing technologies. For critical process parameters and important protection signals, two or three redundant measurement points are typically configured. The reliability of measurement results is improved through mutual verification and voting among these redundant measurement points. Currently, mainstream redundancy processing methods include "two out of three" logical judgment and "middle out of three" value selection. The "two out of three" method is mainly suitable for digital signals; the condition is considered valid if two or more of the three redundant signals meet the criteria. The "middle out of three" method is suitable for analog signals, selecting the median value from the three redundant measurements as the final output. However, existing redundancy processing methods have the following shortcomings: First, the existing redundant voting method uses equal weights for each redundant measurement point, which fails to fully consider the differences in the actual signal quality of each measurement point. When a measurement point experiences partial performance degradation but has not yet completely failed, the unreliable data of that measurement point still participates in the voting with the same weight as the normal measurement point, which may pollute the final fusion result.

[0004] Second, existing methods are rather crude in identifying faults at measurement points, mainly relying on simple over-limit judgments or channel self-tests. They lack the ability to comprehensively analyze the multi-dimensional characteristics of measurement point signals, making it difficult to effectively warn of early deterioration trends at measurement points.

[0005] Third, the existing methods have relatively rigid redundancy switching strategies after measurement point failures, with single switching conditions, and cannot perform dynamic and refined fault-tolerant processing based on the real-time health status of each measurement point.

[0006] Fourth, the redundancy processing logic for analog measurement points and switch measurement points in existing methods is independent of each other, lacking a unified and scalable redundancy fault-tolerant framework.

[0007] To address the aforementioned issues, there is an urgent need for a DCS measurement point redundancy fault tolerance and fault identification method capable of performing refined quality assessment, multi-dimensional fault identification, and dynamic fault tolerance processing for redundant measurement points. Summary of the Invention

[0008] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0009] The technical solution adopted by this invention to solve its technical problem is: a DCS measurement point redundancy tolerance and fault identification method, comprising the following steps: S1. Signal acquisition and quality status acquisition.

[0010] The signals from each measuring point in the DCS are acquired in real time to obtain the measured value and corresponding quality status information of each measuring point.

[0011] Specifically, the DCS controller periodically scans and acquires the measurement point signals on each I / O card channel according to a preset sampling period. The acquired raw signals are then converted from analog to digital to obtain the measured values ​​of the measurement points. .

[0012] Preferably, the acquisition process also obtains quality status information for each measuring point. This quality status information includes: measuring point communication status, measuring point data verification status, measuring point signal amplitude status, and measuring point signal rate of change status.

[0013] The measurement point communication status is used to indicate whether the communication link between the measurement point and the DCS controller is normal. When the communication link is normal, the measurement point can report data to the controller according to the preset cycle; when the communication link is abnormal, the controller cannot receive the measurement point data within the specified timeout period.

[0014] The measurement point data verification status is used to indicate whether the cyclic redundancy check of the measurement point data frame has passed. If a bit error occurs during the transmission of the data frame, the verification will fail.

[0015] The amplitude status of the measuring point signal is used to indicate whether the measured value of the measuring point is within the preset range. Exceeding the range usually indicates that there is an abnormality in the sensor or signal conditioning circuit.

[0016] The measurement point signal change rate status is used to indicate whether the change rate of the measured value of the measurement point exceeds the preset change rate threshold. Abnormal change rate often indicates signal jump or abnormal sensor response.

[0017] S2. Signal Quality Assessment at Measurement Points. Based on the quality status information of each measurement point, the signal quality of each measurement point is assessed, generating a quality assessment coefficient for each measurement point. Quality assessment is a quantitative representation of the reliability of the signal at the measurement point. This invention comprehensively assesses the signal quality of the measurement points from multiple dimensions. For the i-th measurement point, its quality assessment coefficient... Determined in the following manner: in, Let i be the signal integrity coefficient at measurement point i. Let i be the time delay stability coefficient at measurement point i. Let be the historical reliability coefficient of measurement point i; α, β, and γ are the corresponding weighting coefficients, and α + β + γ = 1, with values ​​ranging from [0, 1]. Signal integrity coefficient. The value is determined based on the verification results of the measurement point data frames. Specifically, within a preset time window, the ratio of the number of data frames that successfully pass verification at measurement point i to the total number of received data frames is calculated; this ratio is the percentage of the total number of data frames received. . The closer the value is to 1, the better the integrity of the data transmission from the measurement point. (Time Delay Stability Coefficient) The time is determined based on the deviation between the actual arrival time and the expected arrival time of the signal at the measuring point. Specifically, within a preset time window, the deviation between the actual arrival time and the expected arrival time of each signal at measuring point i is statistically analyzed, and the reciprocal of the deviation is calculated and normalized to obtain the result. . The closer the value is to 1, the more stable the signal transmission delay at the measurement point. Historical reliability coefficient. The value is determined based on the ratio of the number of faults at the measuring point within a preset time period to the total number of samples taken. Specifically, ,in This represents the number of times a fault occurred at measurement point i within a preset time period. This represents the total number of samples.

[0018] S3. Data Consistency Check. A data consistency check is performed on the measured values ​​of multiple measuring points belonging to the same redundancy group, and the deviation between the measured values ​​of each measuring point is calculated. In actual industrial settings, multiple redundant measuring points are typically configured for the same process parameter. These measuring points are installed in the same or nearby process locations, and theoretically, their measured values ​​should maintain a high degree of consistency. When the measured value of a certain measuring point deviates significantly from the measured values ​​of other redundant measuring points, that measuring point may be faulty. For a redundancy group containing n measuring points, the deviation between the measured values ​​of each measuring point is determined as follows: in, Let i be the measured value of the i-th measuring point. Let j be the measured value of the j-th measuring point. Let be the deviation between measuring point i and measuring point j. The average deviation of all measuring points within the redundant group is determined as follows: in, This represents the average deviation of measurement point i relative to other measurement points within the redundant group. The larger the value, the greater the deviation of the measured value of measuring point i from other measuring points, and the lower the reliability of its data.

[0019] S4. Determination of Comprehensive Reliability Weights. Based on the quality assessment coefficients and deviations of each measurement point, the comprehensive reliability weights for each measurement point are determined. The determination of the comprehensive reliability weights takes into account both the signal quality of the measurement point itself and its data consistency relative to other measurement points within the redundancy group. The comprehensive reliability weight for measurement point i is... Determined in the following manner: in, Let i be the quality evaluation coefficient for measuring point i. Let be the average deviation of measurement point i, and λ be the deviation penalty coefficient, where λ > 0. This formula means that the overall reliability of a measurement point is directly proportional to its own quality assessment coefficient and inversely proportional to its deviation relative to other measurement points. The greater the deviation of a measurement point's measured value from other measurement points within the redundancy group, the greater the penalty on its overall reliability. The overall reliability weights of each measurement point are then normalized. in, The normalized comprehensive confidence weight is denoted by , where n is the total number of test points within the redundant group, and . Normalization ensures that the sum of the weights of all measurement points is 1, which facilitates subsequent weighted fusion calculations.

[0020] S5. Weighted Fusion Output. Based on the comprehensive reliability weight of each measurement point, the measured values ​​of each measurement point within the redundancy group are weighted and fused to generate the fused output value of the redundancy group. Determined in the following manner: in, Let i be the measured value of the i-th measuring point. Let be the normalized comprehensive confidence weight for measurement point i, and n be the total number of measurement points in the redundant group. Compared to the traditional "median-of-three" method, which only uses the measurement value of a single measurement point as the output, the weighted fusion method comprehensively utilizes the information of all measurement points in the redundant group, and the contribution of each measurement point is dynamically adjusted according to its real-time confidence. When the confidence of a certain measurement point decreases, its impact on the fusion output automatically decreases, thereby effectively suppressing the contamination of the output results by faulty measurement points. S6, Fault Type Identification.

[0021] Based on the quality assessment coefficient, average deviation, and historical status data of each measuring point, the fault type of each measuring point is identified.

[0022] This invention categorizes measurement point faults into the following five types: Signal loss fault: When the signal integrity coefficient of the measuring point is... Below the first threshold When the signal is lost, it indicates that the measurement point data cannot reach the controller normally, which is determined to be a signal loss fault. Signal loss faults are usually caused by communication link interruption, I / O card failure, or sensor power supply failure. Signal deviation fault: When the average deviation of the measurement point... Greater than the second threshold And quality assessment coefficient Greater than the third threshold When this occurs, it is determined to be a signal deviation fault. In this case, the signal integrity of the measuring point itself is still acceptable. (Higher), but its measured value continues to deviate from other redundant measurement points, indicating that the sensor or signal conditioning circuit at this measurement point may have calibration deviation problems such as zero drift or gain variation. Signal jump fault: When the measured value of the measurement point changes within two consecutive sampling periods... Greater than the fourth threshold When this occurs, it is determined to be a signal jump fault. Signal jump faults are usually caused by electromagnetic interference, poor connector contact, or intermittent sensor malfunctions, and are characterized by a large change in the measured value within a very short time. Signal saturation fault: When the measured value at the measuring point... Continuously at the preset upper limit value Or preset lower limit value When the value falls within a preset range, it is considered a signal saturation fault. Signal saturation faults are usually caused by improper sensor range selection, sensor damage, or saturation of the signal conditioning circuit. Abnormal signal noise fault: When the variance of the measured value at the measuring point exceeds a preset time period... Greater than the fifth threshold When this occurs, it is determined to be a signal noise abnormality fault. Signal noise abnormality faults are usually caused by increased electromagnetic interference, sensor aging, or poor grounding, and are characterized by a significant increase in the random fluctuation amplitude of the measured value.

[0023] The above thresholds to It can be tuned and optimized according to the actual working conditions and measurement point types of different industrial sites.

[0024] Preferably, a fault diagnosis record is also generated based on the fault type of each measuring point. The fault diagnosis record includes: fault measuring point identifier, fault type, fault occurrence time, fault duration, and fault level.

[0025] The fault level is determined comprehensively based on the overall reliability weight of the measuring point and the severity of the fault type, and is divided into Level 1, Level 2, and Level 3 faults, with Level 1 faults having the highest severity and Level 3 faults having the lowest severity. The generation of fault diagnosis records facilitates maintenance personnel to promptly grasp the health status of each measuring point and provides data support for subsequent preventive maintenance.

[0026] S7. Redundancy switching and fault isolation.

[0027] Based on the fault type and the overall confidence weight, perform measurement point redundancy switching and fault isolation operations.

[0028] Specifically, when the comprehensive reliability weight of a certain measurement point within the redundancy group Less than the sixth threshold If the reliability of a measurement point has dropped to an unacceptable level, the measurement point is marked as unusable, and a fault isolation operation is performed to exclude its measurements from the fusion calculation. After fault isolation, the data from that measurement point will no longer participate in subsequent weighted fusion, thus avoiding contamination of the fusion results by low-quality data.

[0029] When the number of measurement points marked as unavailable within a redundant group reaches a preset number, it indicates that most of the measurement points in the current redundant group have failed, triggering a redundancy switchover operation to switch the current redundant group to a backup redundant group. The backup redundant group can be another set of redundant measurement points with the same process parameters (e.g., redundant sensor groups installed in different physical locations), or it can be a backup I / O channel of the same measurement point group.

[0030] When the overall reliability weight of all measuring points in the redundant group is less than the sixth threshold If the redundant group fails, a fault alarm signal is output, and the output switches to a preset safety value. The preset safety value is set in advance according to specific process requirements, and is usually a value that can make the controlled object tend to a safe state.

[0031] The beneficial effects of the present invention are as follows: 1. The DCS measurement point redundancy tolerance and fault identification method of the present invention, by introducing quality evaluation coefficient and comprehensive credibility weight, realizes multi-dimensional fine evaluation of the signal quality of redundant measurement points, overcomes the defects of medium weight processing in the prior art, and enables high-quality measurement points to occupy a larger weight in the fusion output and effectively suppresses the negative impact of low-quality measurement points.

[0032] 2. The DCS measurement point redundancy fault tolerance and fault identification method described in this invention subdivides measurement point faults into five types: signal loss, signal deviation, signal jump, signal saturation, and abnormal signal noise. Through comprehensive judgment using multi-dimensional features, it achieves refined identification of measurement point faults, which helps maintenance personnel accurately locate the cause of the fault.

[0033] 3. The DCS measurement point redundancy fault tolerance and fault identification method described in this invention performs dynamic fault isolation and redundancy switching based on comprehensive confidence weights. The switching conditions are adaptively adjusted according to the real-time health status of each measurement point. Compared with the switching strategy of fixing a single threshold in the prior art, it has higher flexibility and reliability.

[0034] 4. The DCS measurement point redundancy fault tolerance and fault identification method described in this invention, through the dynamic update mechanism of the comprehensive confidence weight, enables the weight of each measurement point to be smoothly adjusted with the change of signal quality, which not only ensures the system's rapid response to changes in the measurement point status, but also avoids drastic fluctuations in the fused output.

[0035] 5. The DCS measurement point redundancy and fault identification method described in this invention provides a unified redundancy and fault tolerance framework, which can be applied to both analog measurement points and switch measurement points, and has good versatility and scalability. Attached Figure Description

[0036] The invention will now be further described with reference to the accompanying drawings.

[0037] Figure 1 This is an overall flowchart of the DCS measurement point redundancy tolerance and fault identification method of the present invention; Figure 2 This is a flowchart of the fault type identification process for measurement points in this invention; Figure 3 This is a flowchart of the redundancy switching and fault isolation operation in this invention. Detailed Implementation

[0038] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments. Example

[0039] like Figures 1 to 3 As shown in the figure, this embodiment takes the measurement of the main steam temperature of a thermal power plant as an example to describe the method of the present invention in detail.

[0040] Main steam temperature is one of the key monitoring parameters for boiler operation in thermal power plants. Typically, three redundant thermocouple measuring points are configured, installed at different locations to ensure measurement reliability. Assume the three measuring points are numbered A, B, and C, forming a redundant group (n=3).

[0041] S1. Signal Acquisition and Quality Status Measurement. The DCS controller acquires signals from three thermocouple measuring points in real time with a sampling period of 200ms. In a certain sampling, the measured values ​​of the three measuring points are as follows: =541.2℃ =540.8℃ =539.5℃. At the same time, the quality status information of each measuring point was obtained: the communication of measuring point A was normal, the calibration was passed, the amplitude was within the range, and the rate of change was normal; the communication of measuring point B was normal, the calibration was passed, the amplitude was within the range, and the rate of change was normal; the communication of measuring point C was normal, the calibration was passed, the amplitude was within the range, but the rate of change was slightly higher than the normal value.

[0042] S2. Signal Quality Assessment at Measurement Points. Calculate the quality assessment coefficients based on the quality status information of each measurement point. Assume the weighting coefficients are set to α=0.4, β=0.3, and γ=0.3. The calculated signal integrity coefficient for measurement point A is... =0.98, time delay stability coefficient =0.95, historical reliability coefficient =0.99, then =0.4×0.98+0.3×0.95+0.3×0.99=0.974; Measuring point B =0.97、 =0.96、 =0.98, then =0.4×0.97+0.3×0.96+0.3×0.98=0.970; Measuring point C =0.96、 =0.88 (abnormal rate of change leads to decreased time delay stability) =0.97, then =0.4×0.96+0.3×0.88+0.3×0.97=0.939.

[0043] S3, Data Consistency Check.

[0044] Calculate the deviation between each measuring point: =|541.2-540.8|=0.4℃ =|541.2-539.5|=1.7℃ =|540.8-539.5|=1.3℃ Average deviation at each measuring point: =(0.4+1.7) / 2=1.05℃ =(0.4+1.3) / 2=0.85℃ =(1.7+1.3) / 2=1.50℃ S4. Determining the overall credibility weight.

[0045] Set the deviation penalty coefficient λ=0.5, and calculate the overall reliability weight of each measuring point: =0.974 / (1+0.5×1.05)=0.639 =0.970 / (1+0.5×0.85)=0.681 =0.939 / (1+0.5×1.50)=0.537 Normalization process: =0.639 / (0.639+0.681+0.537)=0.344 =0.681 / 1.857=0.367 =0.537 / 1.857=0.289 As can be seen, measurement point C has the lowest overall reliability weight (0.289) due to its abnormal rate of change and large data deviation; while measurement point B has the highest overall reliability (0.367).

[0046] S5, Weighted fusion output. Fusion output value: =0.344×541.2+0.367×540.8+0.289×539.5=540.56℃ This fusion value is output to the control system as the final measured value of the main steam temperature.

[0047] S6. Fault type identification. Set various thresholds: =0.85、 =1.2℃ =0.90、 =5℃ / s =2.0、 =600℃ =400℃. Based on the judgment: measuring point C... =1.50℃> ,and =0.939> ,but =0.96> The system determined that a "signal deviation fault" existed at measuring point C, possibly caused by slight zero drift in the thermocouple or the influence of local airflow on the installation location. No faults were detected at measuring points A and B. A fault diagnosis record was generated: measuring point C, signal deviation fault, occurrence time, duration, and fault level: Level 2.

[0048] S7. Redundancy switching and fault isolation.

[0049] Set a sixth threshold =0.20. Currently, the weights of all measuring points are greater than 0.20, so no isolation or switching is required. However, the system records the deviation and fault information of measuring point C and continuously monitors its status change trend. Example

[0050] like Figures 1 to 3 As shown in the figure, this embodiment takes the pressure protection switch signal of a chemical plant reactor as an example to illustrate the application of the present invention in the redundancy and fault tolerance of switch measurement points.

[0051] Reactor overpressure protection is a critical safety feature in chemical production, typically configured with three redundant pressure switch measuring points. The signals from the three measuring points are as follows: , , The value is either 0 or 1, where 0 indicates normal pressure and 1 indicates an overpressure alarm. Unlike analog measurement points, the measured values ​​of digital measurement points are discrete, and their deviation calculations require adaptive adjustments. For digital signals, the deviation between measurement points is defined as the XOR result of the signal values: ,in , ∈{0,1}. In a certain sampling, the values ​​of the three measurement points are: =0、 =0、 =1. Average deviation: =(0+1) / 2=0.5、 =(0+1) / 2=0.5、 =(1+1) / 2=1.0. Regarding quality assessment, the communication status of measuring point C is normal, but the signal switching frequency is high; its quality assessment coefficient is... The values ​​are lower than those at measurement points A and B. The calculated normalized overall confidence weights for the three measurement points are as follows: =0.38、 =0.38、 =0.24. For switch signals, the fused output uses a weighted voting method: the weighted sum is calculated as follows: Output 1 if the weighted sum is greater than 0.5, otherwise output 0. In this example, the weighted sum... =0.38×0+0.38×0+0.24×1=0.24<0.5, therefore the fused output is 0, meaning the pressure is normal. Although measuring point C reported excessively high pressure, due to its low reliability and inconsistency with the other two measuring points, the system ultimately adopted the judgment results of measuring points A and B, effectively avoiding malfunctions caused by false alarms from a single measuring point. Measuring point C was weighted based on its overall reliability. =0.24 greater than (0.20), temporarily not isolated, but continuously monitored. If the weight of subsequent measurement point C continues to decrease to below... If so, then isolate it. Example

[0052] like Figures 1 to 3 As shown, this embodiment takes a severe failure scenario of a key parameter in a nuclear power plant's DCS as an example to illustrate the overall failure handling mechanism of the present invention.

[0053] Suppose a redundancy group contains three measurement points. Due to a power failure in the I / O card housing the redundancy group, all three measurement points simultaneously experience signal loss. The signal integrity coefficient of the three measurement points is... All dropped sharply to below 0.1, quality assessment coefficients All decreased to below 0.2, with average deviation. The calculated value is abnormal (because there is no valid data at any of the three measuring points).

[0054] The comprehensive confidence weight calculation results show that the normalized weights of the three measurement points are... , , All are less than the sixth threshold (0.20) triggers the "redundancy group overall failure" judgment. The system immediately outputs a fault alarm signal and switches the output value of this parameter to the preset safe value. For reactor protection parameters, the safe value is usually set to the value that triggers the protection action to ensure safety. At the same time, the system sends a detailed fault diagnosis report to the host computer, including information such as fault type (signal loss fault), fault measurement point identification, and fault occurrence time, so as to facilitate maintenance personnel to quickly locate and troubleshoot the fault.

[0055] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for redundancy tolerance and fault identification of DCS measurement points, characterized in that: Includes the following steps: S1. Real-time acquisition of signals from each measuring point in the DCS to obtain the measured value and corresponding quality status information of each measuring point; S2. Based on the quality status information of each measuring point, evaluate the signal quality of each measuring point and generate the quality evaluation coefficient for each measuring point. S3. Perform data consistency checks on the measured values ​​of multiple measuring points belonging to the same redundant group, and calculate the deviation and average deviation between the measured values ​​of each measuring point. S4. Determine the comprehensive reliability weight of each measuring point based on the quality assessment coefficient and average deviation of each measuring point. S5. Based on the comprehensive confidence weight of each measurement point, the measured values ​​of each measurement point in the redundancy group are weighted and fused to generate the fused output value of the redundancy group. S6. Identify the fault type of each measuring point based on the quality assessment coefficient, average deviation, and historical status data of each measuring point. S7. Based on the fault type and the comprehensive confidence weight, perform measurement point redundancy switching and fault isolation operations.

2. The DCS measurement point redundancy tolerance and fault identification method according to claim 1, characterized in that: In step S2, the quality assessment coefficient is determined as follows: For the i-th measurement point, its quality evaluation coefficient Represented as: in, Let i be the signal integrity coefficient at measurement point i. Let i be the time delay stability coefficient at measurement point i. Let be the historical reliability coefficient of measurement point i; α, β, and γ are the corresponding weighting coefficients, and α+β+γ=1, with the values ​​of α, β, and γ all ranging from [0,1].

3. The DCS measurement point redundancy tolerance and fault identification method according to claim 1, characterized in that: In step S3, for a redundant group containing n measuring points, the deviation between the measured values ​​of each measuring point is determined as follows: in, Let i be the measured value of the i-th measuring point. Let j be the measured value of the j-th measuring point. Let be the deviation between measuring point i and measuring point j; the average deviation of all measuring points within the redundant group is determined as follows: in, This represents the average deviation of measurement point i relative to other measurement points within the redundant group.

4. A DCS measurement point redundancy tolerance and fault identification method according to claim 2 or 3, characterized in that: In step S4, the comprehensive confidence weight of measurement point i Determined in the following manner: in, Let i be the quality evaluation coefficient for measuring point i. Let be the average deviation of measuring point i, and λ be the deviation penalty coefficient, where λ > 0; The overall reliability weights of each measurement point are normalized: in, The normalized comprehensive confidence weight is denoted by , where n is the total number of test points within the redundant group, and . .

5. The DCS measurement point redundancy tolerance and fault identification method according to claim 1, characterized in that: In step S5, the fused output value of the redundant group Determined in the following manner: in, Let i be the measured value of the i-th measuring point. Let i be the normalized comprehensive confidence weight of measurement point i, and n be the total number of measurement points in the redundant group.

6. The DCS measurement point redundancy tolerance and fault identification method according to claim 1, characterized in that: In step S6, the fault types include signal loss fault, signal deviation fault, signal jump fault, signal saturation fault, and abnormal signal noise fault. For measurement point i, the fault type is identified as follows: When the signal integrity coefficient of the measuring point Below the first threshold When this occurs, it is determined to be a signal loss fault; When the average deviation of the measuring point Greater than the second threshold And quality assessment coefficient Greater than the third threshold When this occurs, it is determined to be a signal deviation fault; When the change in the measured value of the measuring point within two consecutive sampling periods Greater than the fourth threshold When this occurs, it is determined to be a signal transition fault; When the measured value of this measuring point Continuously at the preset upper limit value Or preset lower limit value If the signal is within a preset range, it is determined to be a signal saturation fault; When the variance of the measured value at this measuring point within a preset time period Greater than the fifth threshold When this occurs, it is determined to be a signal noise abnormality fault.

7. The DCS measurement point redundancy tolerance and fault identification method according to claim 1, characterized in that: In step S7, the measurement point redundancy switching and fault isolation operation includes: When the comprehensive reliability weight of a certain measurement point in the redundant group Less than the sixth threshold When this happens, mark the measuring point as an unavailable measuring point, perform a fault isolation operation, and exclude the measured value of the measuring point from the fusion calculation; When the number of measurement points marked as unavailable in the redundancy group reaches the preset number, the redundancy switching operation is triggered, and the current redundancy group is switched to the standby redundancy group. When the overall reliability weight of all measuring points in the redundant group is less than the sixth threshold When the redundancy group fails, it is determined that the entire group has failed, and a fault alarm signal is output and the output is switched to the preset safe value.

8. The DCS measurement point redundancy tolerance and fault identification method according to claim 1, characterized in that: In step S1, the quality status information includes the measurement point communication status, measurement point data verification status, measurement point signal amplitude status, and measurement point signal rate of change status. The communication status of the measuring point is used to indicate whether the communication link between the measuring point and the DCS controller is normal. The measurement point data verification status is used to indicate whether the cyclic redundancy check of the measurement point data frame has passed; The amplitude status of the measuring point signal is used to indicate whether the measured value of the measuring point is within the preset range. The rate of change status of the measuring point signal is used to indicate whether the rate of change of the measured value of the measuring point exceeds a preset rate of change threshold.

9. The DCS measurement point redundancy tolerance and fault identification method according to claim 1, characterized in that: In step S6, a fault diagnosis record is also generated based on the fault type of each measuring point. The fault diagnosis record includes the fault measuring point identifier, fault type, fault occurrence time, fault duration, and fault level. The fault levels are divided into Level 1, Level 2, and Level 3 faults, with Level 1 faults being the most severe and Level 3 faults being the least severe.

Images