PFD algorithm based on reliability block diagram with varying demand rate and detection scheme

Abstract

This invention discloses a PFD algorithm based on reliability block diagrams that varies with required rate and testing scheme. Considering engineering practice and the impact of changes in testing schemes and functional required rate, the algorithm can generate reliability block diagram reports from all components involved in the calculation process, including equipment information, failure data, and calculation results. This facilitates a clear display of the reliability block diagram of the SIF loop. By displaying PFD and MTTF, the impact of random failure probabilities of each component in the loop and architectural constraints on overall functionality is derived. Ultimately, the random failure probability (PFD), risk reduction factor (RRF), random failure risk level (SIL), architectural constraint risk level (SIL), and SIF loop SIL level are determined. For different voting structures (MooN), modeling and calculating the SIF loop of the safety instrumented system can effectively improve the accuracy of the calculated average failure probability (PFD) results for the SIF loop.

Description

Technical Field

[0001] This invention relates to the field of hardware reliability assessment technology for safety instrumented systems (SIS) in process industries such as petroleum, natural gas, chemical, power, metallurgy, pharmaceutical, refrigeration, and offshore platforms, specifically to a PFD algorithm based on reliability block diagrams that varies with required rate and detection scheme. Background Technology

[0002] Since 2000, with the release of the functional safety fundamental standards IEC 61508 and IEC 61511, functional safety technology has been widely applied internationally. Domestically, it has been transformed into GB / T 20438 and GB / T 21109, and subsequently, GB / T 35320, GB / T 32857, and other functional safety-related technical method standards have been released, providing guidance for industries such as petroleum, chemical, power, pharmaceutical, and railway to achieve safety assurance through the adoption of functional safety technologies. With the widespread application of GB / T 20438 and GB / T 21109, the concept of functional safety has been effectively popularized. In process industries such as petroleum, chemical, power, and pharmaceutical, functional safety research focuses on how to ensure that the protection layer, including safety instrumented systems, achieves sufficient risk management capabilities. Safety instrumented systems (SAS) are essential safety management technologies for process industries such as petroleum, chemical, power, and pharmaceutical. Their key feature is that while the process unit is in a dormant state during normal operation, it monitors for emerging or latent hazards during production, issues alarms, or directly executes predetermined procedures to immediately bring the unit to a safe state, preventing accidents and mitigating their harm and impact. A safety instrumented system consists of one or more safety instrument functions with specific protection scenarios.

[0003] After the basic process flow design is completed, hazard and risk analysis (such as Hazard and Operability Analysis, HAZOP) and SIL rating (such as Layer of Protection Analysis, LOPA) should be used to identify the risks in the current design, propose Safety Instrumented Function (SIF) setting requirements and SIL level requirements. After completing the SIS design and selection configuration, SIL verification is required to ensure that the design and configuration meet safety requirements. SIL verification is a quantitative analysis method that determines the actual SIL level of the loop through calculation. Commonly used SIL verification methods include reliability block diagram method, fault tree analysis method, and Markov model analysis method.

[0004] The commonly used SIL verification method, the reliability block diagram method, does not consider the impact of the time interval between the test coverage and functional requirements. This results in the SIF loop's calculated mean failure probability (PFD) being inaccurate and failing to reflect the actual situation. Summary of the Invention

[0005] The purpose of this invention is to provide a PFD algorithm based on reliability block diagrams that varies with the required rate and testing scheme. By combining engineering practice and considering the impact of changes in inspection and testing schemes and functional required rate, the accuracy of the SIF loop calculation results of safety instrumented systems is improved, thus making it closer to the actual situation. This enhances the reliability level of safety instrumented systems in the process industry and can effectively reduce safety accidents.

[0006] To solve the above technical problems, the present invention adopts the following technical solution: a PFD algorithm based on reliability block diagrams that varies with demand rate and detection scheme, comprising the following steps:

[0007] S110. All components in the Safety Instrumented Function (SIF) loop that constructs the reliability block diagram are divided into modules and classified, with each component corresponding to a sensing unit, logic unit, and execution unit.

[0008] S120. A multi-layer reliability block diagram model is built based on the voting structure MooN of each element in the sensing unit, logic unit, and execution unit; where if M out of N identical elements in MooN can operate normally, safety can be guaranteed.

[0009] S130. Calculate the safety failure score (SFF) and hardware fault margin (HFT) of each unit based on the failure data of each component in the sensing unit, logic unit, and execution unit, thereby obtaining the SIF loop architecture constraint (SIL).

[0010] S140. Calculate the random failure probability (PFD) required by the sensing unit, logic unit, and execution unit based on different voting structures and the parameters of the components.

[0011] S150. Based on the random failure probability PFD required by the sensing unit, logic unit and execution unit respectively, obtain the random failure probability PFD of the SIF loop, the risk reduction factor RRF, the SIL of random failure, the SIL of architectural constraints and the SIL level of the SIF loop.

[0012] S160. Based on the calculated random failure probability PFD, risk reduction factor RRF, random failure SIL, architectural constraint SIL and SIL level of the SIF loop, the software can automatically generate a reliability block diagram report and an SIF overview table for the SIL verification work.

[0013] The aforementioned PFD algorithm based on reliability block diagrams that varies with demand rate and detection scheme, wherein the voting structure MooN of each element in the sensing unit, logic unit and execution unit in step S120 is specifically: 1oo1, 1oo2, 2oo2, 1oo3, 3oo3, 1oo4, 4oo4, 1oo1, 1oo2, 2oo2, 1oo3, 2oo3, 3oo3, 1oo4, 2oo4, 3oo4, 4oo4, 5oo5, 6oo6, 1oo1D, 1oo2D, 2oo2D;

[0014] The multi-layer reliability block diagram model is represented by one or more combinations of several basic structures, such as series, parallel, and voting. The reliability block diagram consists of three parts: sensing unit, logic unit, and execution unit, which are connected in series. The logic unit in the reliability block diagram is a box. The sensing unit and execution unit are displayed according to the actual situation and can be any structure in the voting structure MooN above.

[0015] The aforementioned PFD algorithm based on reliability block diagrams that varies with demand rate and detection scheme, wherein step S130 specifically comprises:

[0016] S131. Failure data for each component of the sensing unit, logic unit, and execution unit are obtained from an internationally trusted database and product certification data provided by the owner. The specific failure data for each component is λ. DD、 λ DU、 λ SD and λ SU ;λ DD For the detected dangerous failure rate, λ DU For the rate of undetected dangerous failures, λ SD For the safety failure rate being detected, λ SU This represents the rate of undetected safety failures.

[0017] S132. Calculate the safety failure fraction SFF of each unit, using the following formula;

[0018] ,

[0019] Where λ DD For the detected dangerous failure rate, λ DU For the rate of undetected dangerous failures, λ SD For the safety failure rate being detected, λ SU The rate of undetected safety failures;

[0020] S133. Calculate the hardware fault margin (HFT) of each unit using the voting structure MooN of each element. The calculation formula is as follows:

[0021] HFT = N - M,

[0022] If we calculate the HFT of the voting structure 2oo3, then HFT = 3 - 2, and the result is 1;

[0023] S134. Determine whether the instrument type is Class A or Class B based on the actual instruments and equipment installed on site. Then, determine the value of the architecture constraint SIL based on the different categories of instruments, combined with SFF and HFT.

[0024] The aforementioned PFD algorithm based on reliability block diagrams that varies with demand rate and detection scheme, wherein step S140 specifically comprises:

[0025] S141. The commonly used voting structures MooN for sensing units and execution units include: 1oo1, 1oo2, 2oo2, 1oo3, 2oo3, 3oo3, 1oo4, 2oo4, 3oo4, 4oo4, 5oo5, and 6oo6. The random failure probability PFD of the sensing unit and execution unit is calculated based on the voting structure.

[0026] When the voting structure is 1oo1, the formula for calculating the probability of random failure (PFD) is:

[0027] ,

[0028] Where PTC is the inspection and testing coverage; TI is the inspection and testing time interval (h); Td is the time interval between requirements (h); MTTR is the mean recovery time (h); MRT is the mean repair time (h); λ DD The detected rate of dangerous failure; λ DU Undetected dangerous failure rate;

[0029]

[0030] in

[0031] t CE The equivalent average downtime (h) of the channels in the 1oo1, 1oo2, 2oo2, and 2oo3 structures;

[0032] t GE The equivalent average inactivity time (h) of the voting groups in the 1oo2 and 2oo3 structures;

[0033] t G2E The equivalent average inactivity time (h) of the voting groups in the 1oo3, 1oo4, and 2oo4 structures;

[0034] t G3E The equivalent average inactivity time (h) of the voting group in the 1oo4 structure.

[0035] When the voting structure is 1oo2, the formula for calculating the probability of random failure (PFD) is:

[0036]

[0037] Where β represents the number of undetected failures with a common cause; β D Failure scores that have been detected by diagnostic tests and have a common cause;

[0038] When the voting structure is 2oo2, the formula for calculating the probability of random failure (PFD) is:

[0039]

[0040] When the voting structure is 1oo3, the formula for calculating the probability of random failure (PFD) is:

[0041]

[0042] When the voting structure is 2oo3, the formula for calculating the probability of random failure (PFD) is:

[0043]

[0044] When the voting structure is 3oo3, the formula for calculating the probability of random failure (PFD) is:

[0045]

[0046] When the voting structure is 1004, the formula for calculating the probability of random failure (PFD) is:

[0047]

[0048] When the voting structure is 2oo4, the formula for calculating the probability of random failure (PFD) is:

[0049]

[0050] When the voting structure is 3oo4, the formula for calculating the probability of random failure (PFD) is:

[0051]

[0052] When the voting structure is 4oo4, the formula for calculating the probability of random failure (PFD) is:

[0053]

[0054] When the voting structure is 5005, the formula for calculating the probability of random failure (PFD) is:

[0055]

[0056] When the voting structure is 6006, the formula for calculating the probability of random failure (PFD) is:

[0057]

[0058] S142. The voting structure MooN in the low-requirement operation mode of the logic unit includes: 1oo1, 1oo1D, 1oo2, 1oo2D, 2oo2, 2oo2D, 1oo3, 2oo3 and 3oo3. The random failure probability PFD when the logic unit is required is calculated according to the voting structure.

[0059] When the logic unit voting structure is 1oo1, 1oo2, 2oo2, 1oo3, 2oo3 and 3oo3, the formula for calculating the random failure probability PFD is the corresponding formula in step S141.

[0060] When the voting structure is 1oo1D, the formula for calculating the probability of random failure (PFD) is:

[0061]

[0062] Where PTC is the inspection and testing coverage rate; TI is the inspection and testing time interval (h); Td is the time interval between requirements (h); MRT is the mean time to repair (h); λ DU Undetected dangerous failure rate;

[0063] When the voting structure is 1oo2D, the formula for calculating the probability of random failure (PFD) is:

[0064]

[0065] Where MTTR is the mean recovery time (h); λ DD The detected rate of dangerous failure; λ SD The safety failure rate being detected; The equivalent average downtime (h) of the channel in the 1oo2D structure; is the equivalent average downtime (h) of the voting group in the 1oo2D structure; K is the success percentage of the circuit tested only in the 1oo2D system, describing the efficiency of the comparison / switching mechanism of the internal channel. Since the comparison / switching mechanism of the channel is not 100% effective, the parameter K needs to be determined by FMEA. Here, K=0.98.

[0066] When the voting structure is 2oo2D, the formula for calculating the probability of random failure (PFD) is:

[0067]

[0068] Where β represents the number of undetected failures with a common cause; β DThese are the failure scores that have been detected by diagnostic tests and have a common cause.

[0069] The aforementioned PFD algorithm based on reliability block diagrams that varies with demand rate and detection scheme, wherein step S150 specifically comprises:

[0070] S151. Calculate the random failure probability (PFD) of the SIF loop:

[0071] The random failure probability PFD of the SIF loop is the sum of the random failure probabilities PFD of the sensing unit, logic unit, and execution unit.

[0072] S152. Calculate the risk reduction factor RRF, the formula is:

[0073] RRF = 1 / PFD

[0074] Where PFD is the random failure probability of the SIF loop;

[0075] S153. Based on the correspondence table between PFD values ​​and SIL levels, obtain the random failure SIL;

[0076] S154. Determine the SIF loop SIL level based on the random failure SIL and the architecture constraint SIL, specifically: the smaller of the random failure SIL and the architecture constraint SIL is the SIF loop SIL level.

[0077] The aforementioned PFD algorithm based on reliability block diagrams varies with the required rate and testing scheme, wherein step S160 specifically involves: exporting from the software a reliability block diagram report and a SIF overview table for the SIL verification work, which contains basic project information, loop structure, data analysis, reliability block diagram model, analysis and calculation conclusions, PFD distribution diagram and MTTF distribution diagram.

[0078] The aforementioned PFD algorithm based on reliability block diagrams varies with demand rate and detection scheme, wherein the sensing unit includes all components in the SIF loop from the field sensing element to the input of the logic controller; the logic unit includes input cards (AI / DI), output cards (AO / DO), power modules, and a controller; and the execution unit includes all components in the SIF loop from the output of the logic controller to the final field actuator.

[0079] The aforementioned PFD algorithm based on reliability block diagrams varies with required rate and detection scheme, wherein the software is self-developed, specifically: SIL verification based on reliability block diagrams, which incorporates the algorithm of this invention.

[0080] The aforementioned PFD algorithm based on reliability block diagrams that varies with required rate and detection scheme includes a component failure database management function in the SIL verification based on reliability block diagrams, which can perform add, delete, modify and query operations on the data in the component failure database.

[0081] Compared with the prior art, the advantages of the present invention are that,

[0082] 1. This invention is based on the calculation method of average probability of failure (PFD) in GB / T 20438 and GB / T 21109. Combined with engineering practice, it considers the impact of changes in inspection and testing schemes and functional requirements. For different voting structures (MooN), it performs modeling and calculation of the SIF loop of the safety instrumented system, which can effectively improve the accuracy of the calculated average probability of failure (PFD) of the SIF loop.

[0083] 2. This invention can generate a reliability block diagram report through the software system, which can collect equipment information, failure data and calculation results of all components involved in the entire calculation process. This makes it easy to clearly display the reliability block diagram of the SIF loop. By displaying PFD and MTTF, the impact of random failure probability of each component in the loop and architectural constraints on the overall function can be obtained. Finally, the random failure probability PFD, risk reduction factor RRF, SIL of random failure, SIL of architectural constraints and SIL level of SIF loop are determined.

[0084] 3. This invention can provide technical support for reliability assessment of safety instrumented systems in process industries such as petroleum, natural gas, chemical, power, metallurgy, pharmaceutical, refrigeration, and offshore platforms. It can complete the SIL verification of safety instrumented systems based on reliability block diagrams. Based on the verification results, users can reasonably and effectively set up safety instrumented systems to ensure the safe and stable operation of the equipment. Attached Figure Description

[0085] Figure 1 This is a flowchart illustrating the present invention;

[0086] Figure 2 This is the multi-layer reliability frame generated by the present invention. Figure 1 ;

[0087] Figure 3 This is the multi-layer reliability frame generated by the present invention. Figure 2 ;

[0088] Figure 4 This is a schematic diagram of the software page in this invention. Figure 1 ;

[0089] Figure 5 This is a schematic diagram of the software page in this invention. Figure 2 ;

[0090] Figure 6 This is a schematic diagram of the parameter setting page of the software in this invention;

[0091] Figure 7 This is a schematic diagram of the calculation results page of the software in this invention;

[0092] Figure 8 This is a schematic diagram of the PFD and MTTF distributions generated by the software in this invention;

[0093] Figure 9 This is a schematic diagram of the reliability block diagram report for the SIL verification work of the software in this invention;

[0094] Figure 10 This is a schematic representation of the SIF of the software in this invention;

[0095] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. Detailed Implementation

[0096] Embodiment 1 of the present invention: A PFD algorithm based on reliability block diagrams that varies with demand rate and detection scheme, comprising the following steps:

[0097] S110. All components in the Safety Instrumented Function (SIF) loop that constructs the reliability block diagram are divided into modules and classified, with each component corresponding to a sensing unit, logic unit, and execution unit.

[0098] S120. A multi-layer reliability block diagram model is built based on the voting structure MooN of each element in the sensing unit, logic unit, and execution unit; where MooN specifically refers to a voting structure in which safety can be guaranteed if M of N identical elements can operate normally.

[0099] S130. Calculate the safety failure score (SFF) and hardware fault margin (HFT) of each unit based on the failure data of each component in the sensing unit, logic unit, and execution unit, thereby obtaining the SIF loop architecture constraint (SIL).

[0100] S140. Calculate the random failure probability (PFD) required by the sensing unit, logic unit, and execution unit based on different voting structures and the parameters of the components.

[0101] S150. Based on the random failure probability PFD required by the sensing unit, logic unit and execution unit respectively, obtain the random failure probability PFD of the SIF loop, the risk reduction factor RRF, the SIL of random failure, the SIL of architectural constraints and the SIL level of the SIF loop.

[0102] S160. Based on the calculated random failure probability (PFD), risk reduction factor (RRF), random failure hazard level (SIL), architectural constraint SIL, and SIL level of the SIF loop, the software can automatically generate a reliability block diagram report and an SIF overview table for the SIL verification work. For details of the SIL verification process based on reliability block diagram modeling, please refer to [link to relevant documentation]. Figure 1 .

[0103] Embodiment 2 of the present invention: A PFD algorithm based on reliability block diagrams that varies with demand rate and detection scheme, comprising the following steps:

[0104] S110. All components in the Safety Instrumented Function (SIF) loop of the reliability block diagram are modularized and categorized. Each component is assigned to a sensing unit, logic unit, and execution unit. The sensing unit includes all components in the SIF loop from the field sensing elements to the input of the logic controller, including input safety barriers, AI modules, AI terminal boards, DI terminal boards, and DI modules. The specific components need to be determined based on the actual field installation. The logic unit includes input cards (AI / DI), output cards (AO / DO), power modules, and controllers. The execution unit includes all components in the SIF loop from the output of the logic controller to the final field actuator, including intermediate components (outputs), relays, solenoid valves, actuators, and valve bodies. The specific components need to be determined based on the actual field installation.

[0105] S120. A multi-layer reliability block diagram model is built based on the voting structure MooN of each element in the sensing unit, logic unit, and execution unit; where MooN specifically refers to a voting structure in which security can be guaranteed if M of N identical elements can operate normally.

[0106] The voting structure MooN of each element in the sensing unit, logic unit, and execution unit is as follows: 1oo1, 1oo2, 2oo2, 1oo3, 3oo3, 1oo4, 4oo4, 1oo1, 1oo2, 2oo2, 1oo3, 2oo3, 3oo3, 1oo4, 2oo4, 3oo4, 4oo4, 5oo5, 6oo6, 1oo1D, 1oo2D, 2oo2D;

[0107] A multi-layer reliability block diagram model is represented by one or more combinations of basic structures such as series, parallel, and voting. The reliability block diagram consists of three parts: sensing units, logic units, and execution units, connected in series. A logic unit in the reliability block diagram is represented by a box. The sensing and execution units are represented by structures that can be any structure from the MooN voting structure, depending on the specific requirements. This results in a multi-layer reliability block diagram, for example... Figure 2Specifically, the sensing unit has a 1001 voting structure, the sensing unit group has a 1002 voting structure, the execution unit has a 2002 voting structure, and the execution unit group has two 2002 voting structures. Figure 3 Specifically, the voting structure is as follows: the sensing unit is a 1oo2 voting structure, the sensing unit group has 1oo1 and 2oo3 voting structures, the execution unit is a 2oo2 voting structure, and the execution unit group has two 2oo2 voting structures.

[0108] S130. Calculate the safety failure score (SFF) and hardware fault margin (HFT) of each unit based on the failure data of each component in the sensing unit, logic unit, and execution unit, thereby obtaining the SIF loop architecture constraint (SIL).

[0109] S131. Failure data for each component of the sensing unit, logic unit, and execution unit are obtained from an internationally trusted database and product certification data provided by the owner. The specific failure data for each component is λ. DD、 λ DU、 λ SD and λ SU ;λ DD For the detected dangerous failure rate, λ DU For the rate of undetected dangerous failures, λ SD For the safety failure rate being detected, λ SU The rate of undetected safety failures;

[0110] S132. Calculate the safety failure fraction (SFF) for each unit using the following formula:

[0111] ,

[0112] Where λ DD For the detected dangerous failure rate, λ DU For the rate of undetected dangerous failures, λ SD For the safety failure rate being detected, λ SU The rate of undetected safety failures;

[0113] S133. Calculate the hardware fault margin (HFT) of each unit using the voting structure MooN of each element. The calculation formula is as follows:

[0114] HFT = N - M,

[0115] If we calculate the HFT of the voting structure 2oo3, then HFT = 3 - 2, and the result is 1;

[0116] S134. Determine whether the instrument type is Class A or Class B based on the actual instruments and equipment installed on site. Then, determine the value of the architecture constraint SIL based on the different types of instruments, combined with SFF and HFT.

[0117] A subsystem whose components are required to perform safety functions can be classified as Class A if it meets the following conditions.

[0118] a. The failure modes of all components are well defined; and

[0119] b. The behavior of the subsystem under fault conditions can be fully determined; and

[0120] c. Sufficient and reliable data obtained through field experience can demonstrate the failure rate of both detected and undetected hazardous failures that meet the stated requirements.

[0121] Typical Class A equipment includes actuators, valves, relays, or simple electronic modules consisting of resistors, capacitors, amplifiers, etc.

[0122] A subsystem whose components are required to perform safety functions can be classified as Class B if it meets the following conditions:

[0123] a. The failure mode of at least one component is not well defined; or

[0124] b. The behavior of the subsystem under fault conditions cannot be fully determined; or

[0125] c. Reliable data obtained through field experience is insufficient to demonstrate the failure rate of both detected and undetected hazardous failures as stated.

[0126] Typical Class B subsystems: microprocessor-based devices or devices with complex custom logic.

[0127] By combining the following table showing the correspondence between the architectural constraint SILac and HFT and SFF, the value of the architectural constraint SIL can be determined:

[0128]

[0129] S140. Calculate the random failure probability (PFD) required by the sensing unit, logic unit, and execution unit based on different voting structures and the parameters of the components.

[0130] S141. The commonly used voting structures MooN for sensing units and execution units include: 1oo1, 1oo2, 2oo2, 1oo3, 2oo3, 3oo3, 1oo4, 2oo4, 3oo4, 4oo4, 5oo5, and 6oo6. The random failure probability PFD of the sensing unit and execution unit is calculated based on the voting structure.

[0131] When the voting structure is 1oo1, the formula for calculating the probability of random failure (PFD) is:

[0132] ,

[0133] Where PTC is the inspection and testing coverage; TI is the inspection and testing time interval (h); Td is the time interval between requirements (h); MTTR is the mean recovery time (h); MRT is the mean repair time (h); λ DD The detected rate of dangerous failure; λ DU Undetected dangerous failure rate;

[0134]

[0135] in

[0136] t CE The equivalent average downtime (h) of the channels in the 1oo1, 1oo2, 2oo2, and 2oo3 structures;

[0137] t GE The equivalent average inactivity time (h) of the voting groups in the 1oo2 and 2oo3 structures;

[0138] t G2E The equivalent average inactivity time (h) of the voting groups in the 1oo3, 1oo4, and 2oo4 structures;

[0139] t G3E The equivalent average inactivity time (h) of the voting group in the 1oo4 structure.

[0140] When the voting structure is 1oo2, the formula for calculating the probability of random failure (PFD) is:

[0141]

[0142] Where β represents the number of undetected failures with a common cause; β D Failure scores that have been detected by diagnostic tests and have a common cause;

[0143] When the voting structure is 2oo2, the formula for calculating the probability of random failure (PFD) is:

[0144]

[0145] When the voting structure is 1oo3, the formula for calculating the probability of random failure (PFD) is:

[0146]

[0147] When the voting structure is 2oo3, the formula for calculating the probability of random failure (PFD) is:

[0148]

[0149] When the voting structure is 3oo3, the formula for calculating the probability of random failure (PFD) is:

[0150]

[0151] When the voting structure is 1004, the formula for calculating the probability of random failure (PFD) is:

[0152]

[0153] When the voting structure is 2oo4, the formula for calculating the probability of random failure (PFD) is:

[0154]

[0155] When the voting structure is 3oo4, the formula for calculating the probability of random failure (PFD) is:

[0156]

[0157] When the voting structure is 4oo4, the formula for calculating the probability of random failure (PFD) is:

[0158]

[0159] When the voting structure is 5005, the formula for calculating the probability of random failure (PFD) is:

[0160]

[0161] When the voting structure is 6006, the formula for calculating the probability of random failure (PFD) is:

[0162]

[0163] S142. The voting structure MooN in the low-requirement operation mode of the logic unit includes: 1oo1, 1oo1D, 1oo2, 1oo2D, 2oo2, 2oo2D, 1oo3, 2oo3 and 3oo3. The random failure probability PFD when the logic unit is required is calculated according to the voting structure.

[0164] When the logic unit voting structure is 1oo1, 1oo2, 2oo2, 1oo3, 2oo3 and 3oo3, the formula for calculating the random failure probability PFD is the corresponding formula in step S141.

[0165] When the voting structure is 1oo1D, the formula for calculating the probability of random failure (PFD) is:

[0166]

[0167] Where PTC is the inspection and testing coverage rate; TI is the inspection and testing time interval (h); Td is the time interval between requirements (h); MRT is the mean time to repair (h); λ DU Undetected dangerous failure rate;

[0168] When the voting structure is 1oo2D, the formula for calculating the probability of random failure (PFD) is:

[0169]

[0170] Where MTTR is the mean recovery time (h); λ DD The detected rate of dangerous failure; λ SD The safety failure rate being detected; The equivalent average downtime (h) of the channel in the 1oo2D structure; is the equivalent average downtime (h) of the voting group in the 1oo2D structure; K is the success percentage of the circuit tested only in the 1oo2D system, describing the efficiency of the comparison / switching mechanism of the internal channel. Since the comparison / switching mechanism of the channel is not 100% effective, the parameter K needs to be determined by FMEA. Here, K=0.98.

[0171] When the voting structure is 2oo2D, the formula for calculating the probability of random failure (PFD) is:

[0172]

[0173] Where β represents the number of undetected failures with a common cause; β D These are the failure scores that have been detected by diagnostic tests and have a common cause.

[0174] S150. Based on the random failure probability PFD required by the sensing unit, logic unit and execution unit respectively, obtain the random failure probability PFD of the SIF loop, the risk reduction factor RRF, the SIL of random failure, the SIL of architectural constraints and the SIL level of the SIF loop.

[0175] S151. Calculate the random failure probability (PFD) of the SIF loop:

[0176] The random failure probability PFD of the SIF loop is the sum of the random failure probabilities PFD of the sensing unit, logic unit, and execution unit.

[0177] S152. Calculate the risk reduction factor RRF, the formula is:

[0178] RRF = 1 / PFD

[0179] Where PFD is the random failure probability of the SIF loop;

[0180] S153. Based on the following table of correspondence between SIL levels and PFDs, the random failure SIL is obtained;

[0181] The correspondence between Safety Integrity Level (SILV) and the required Probability of Failure (PFD) and RRF:

[0182]

[0183] S154. Determine the SIF loop SIL level based on the random failure SIL and the architecture constraint SIL, specifically: the smaller of the random failure SIL and the architecture constraint SIL is the SIF loop SIL level.

[0184] S160. Based on the calculated random failure probability (PFD), risk reduction factor (RRF), random failure SIL, architectural constraint SIL, and SIL level of the SIF loop, the software can automatically generate a reliability block diagram report and an SIF overview table for SIL verification work, containing basic project information, loop composition, data analysis, reliability block diagram model, analysis and calculation conclusions, PFD distribution diagram, and MTTF distribution diagram. The software is independently developed and specifically refers to a SIL verification tool based on reliability block diagrams. The software incorporates the algorithm of this invention and includes a component failure database management function, which can perform add, delete, modify, and query operations on the data in the component failure database.

[0185] The working principle of one embodiment of the present invention is as follows: All components in the Safety Instrumented Function (SIF) loop, which forms the reliability block diagram, are modularized and categorized, with each component corresponding to a sensing unit, logic unit, and execution unit; a multi-layer reliability block diagram model is constructed based on the voting structure MooN of each component in the sensing unit, logic unit, and execution unit; the Safety Failure Score (SFF) and Hardware Failure Margin (HFT) of each unit are calculated based on the failure data of each component in the sensing unit, logic unit, and execution unit, thereby obtaining the SIF loop architecture constraint SIL; and calculations are performed based on different voting structures and the parameters of the components. The random failure probability (PFD) required for the sensing unit, logic unit, and execution unit is calculated. Based on the PFDs required for each sensing unit, logic unit, and execution unit, the random failure probability (PFD), risk reduction factor (RRF), random failure level (SIL), architectural constraint level (SIL), and SIL level of the SIF loop are obtained. The calculated random failure probability (PFD), risk reduction factor (RRF), random failure level (SIL), architectural constraint level (SIL), and SIL level of the SIF loop can be automatically generated by the software to produce a reliability block diagram report and an SIF overview table for SIL verification.

Claims

1. A PFD calculation method based on reliability block diagrams that varies with required rate and testing scheme, characterized in that, Includes the following steps: S110. All components in the Safety Instrumented Function (SIF) loop that constructs the reliability block diagram are divided into modules and classified, with each component corresponding to a sensing unit, logic unit, and execution unit. S120. A multi-layer reliability block diagram model built based on the voting structure MooN of each element in the sensing unit, logic unit, and execution unit. S130. Calculate the safety failure score (SFF) and hardware fault margin (HFT) of each unit based on the failure data of each component in the sensing unit, logic unit, and execution unit, thereby obtaining the SIF loop architecture constraint (SIL). S140. Calculate the random failure probability (PFD) required by the sensing unit, logic unit, and execution unit based on different voting structures and the parameters of the components. Step S140 specifically involves: S141. The commonly used voting structures MooN for sensing units and execution units include: 1oo1, 1oo2, 2oo2, 1oo3, 2oo3, 3oo3, 1oo4, 2oo4, 3oo4, 4oo4, 5oo5, and 6oo6. The random failure probability PFD of the sensing unit and execution unit is calculated based on the voting structure. S142. The voting structure MooN in the low-requirement operation mode of the logic unit includes: 1oo1, 1oo1D, 1oo2, 1oo2D, 2oo2, 2oo2D, 1oo3, 2oo3, and 3oo3. The random failure probability PFD when the logic unit requirement is calculated based on the voting structure. When the voting structure is 1oo1D, the formula for calculating the random failure probability PFD is: Where PTC is the inspection and testing coverage rate; TI is the inspection and testing time interval (h); Td is the time interval between requirements (h); MRT is the mean time to repair (h); λ DU The undetected hazardous failure rate; when the voting structure is 1oo2D, the random failure probability (PFD) is calculated using the following formula: Where MTTR is the mean recovery time (h); λ DD The detected rate of dangerous failure; λ SD The safety failure rate being detected; The equivalent average downtime (h) of the channel in the 1oo2D structure; The equivalent average downtime (h) of the voting group in the 1oo2D structure; K is the success percentage of the circuit tested only in the 1oo2D system, describing the efficiency of the comparison / switching mechanism of the internal channel. Since the comparison / switching mechanism of the channel is not 100% effective, the parameter K needs to be determined by FMEA, and here K=0.98; When the voting structure is 2oo2D, the random failure probability (PFD) is calculated as follows: Where β represents the number of undetected failures with a common cause; β D Failure scores that have been detected by diagnostic tests and have a common cause; S150. Based on the random failure probability PFD required by the sensing unit, logic unit and execution unit respectively, obtain the random failure probability PFD of the SIF loop, the risk reduction factor RRF, the SIL of random failure, the SIL of architectural constraints and the SIL level of the SIF loop. S160. Based on the calculated random failure probability PFD, risk reduction factor RRF, random failure SIL, architectural constraint SIL and SIL level of the SIF loop, the software automatically generates a reliability block diagram report and an SIF overview table for the SIL verification work.

2. The PFD calculation method based on reliability block diagrams as a function of required rate and detection scheme, as described in claim 1, is characterized in that, In step S120, the commonly used voting structures MooN for the sensing unit and execution unit include: 1oo1, 1oo2, 2oo2, 1oo3, 2oo3, 3oo3, 1oo4, 2oo4, 3oo4, 4oo4, 5oo5, and 6oo6; the structures of the voting structures MooN in the low-requirement operation mode of the logic unit include: 1oo1, 1oo1D, 1oo2, 1oo2D, 2oo2, 2oo2D, 1oo3, 2oo3, and 3oo3; The multi-layer reliability block diagram model is represented by one or more combinations of several basic structures, such as series, parallel, and voting. The reliability block diagram consists of three parts: sensing unit, logic unit, and execution unit, which are connected in series. The logic unit is represented by a box in the reliability block diagram. The sensing unit and execution unit are displayed according to their actual structures, and the displayed structures of the sensing unit and execution unit are any structures in the corresponding voting structure MooN.

3. The PFD calculation method based on reliability block diagrams as a function of required rate and detection scheme, as described in claim 2, is characterized in that... The specific steps of S130 are as follows: S131. Failure data for each component of the sensing unit, logic unit, and execution unit are obtained from an internationally trusted database and product certification data provided by the owner. The specific failure data for each component is λ. DD、 λ DU、 λ SD and λ SU ;λ DD For the detected dangerous failure rate, λ DU For the rate of undetected dangerous failures, λ SD For the safety failure rate being detected, λ SU The rate of undetected safety failures; S132. Calculate the safety failure fraction (SFF) for each unit using the following formula: ， Where λ DD For the detected dangerous failure rate, λ DU For the rate of undetected dangerous failures, λ SD For the safety failure rate being detected, λ SU The rate of undetected safety failures; S133. Calculate the hardware fault margin (HFT) of each unit using the voting structure MooN of each element. The calculation formula is as follows: HFT = N - M, For the HFT of the voting structure 2oo3, HFT = 3 - 2, and the result is 1; S134. Determine whether the instrument type is Class A or Class B based on the actual instruments and equipment installed on site. Then, determine the value of the architecture constraint SIL based on the different categories of instruments, combined with SFF and HFT.

4. The PFD calculation method based on reliability block diagrams as a function of required rate and detection scheme, as described in claim 3, is characterized in that... Step S141 specifically involves: When the voting structure is 1oo1, the formula for calculating the probability of random failure (PFD) is: ， Where PTC is the inspection and testing coverage; TI is the inspection and testing time interval (h); Td is the time interval between requirements (h); MTTR is the mean recovery time (h); MRT is the mean repair time (h); λ DD The detected rate of dangerous failure; λ DU Undetected dangerous failure rate; Among them, t CE The equivalent average downtime (h) of the channels in the 1oo1, 1oo2, 2oo2, and 2oo3 structures; t GE The equivalent average inactivity time (h) of the voting groups in the 1oo2 and 2oo3 structures; t G2E The equivalent average inactivity time (h) of the voting group in the 1oo3 and 2oo4 structures; t G3E The equivalent average inactivity time (h) of the voting group in the 1oo4 structure. When the voting structure is 1oo2, the formula for calculating the probability of random failure (PFD) is: Where β represents the number of undetected failures with a common cause; β D Failure scores that have been detected by diagnostic tests and have a common cause; When the voting structure is 2oo2, the formula for calculating the probability of random failure (PFD) is: When the voting structure is 1oo3, the formula for calculating the probability of random failure (PFD) is: When the voting structure is 2oo3, the formula for calculating the probability of random failure (PFD) is: When the voting structure is 3oo3, the formula for calculating the probability of random failure (PFD) is: When the voting structure is 1004, the formula for calculating the probability of random failure (PFD) is: When the voting structure is 2oo4, the formula for calculating the probability of random failure (PFD) is: When the voting structure is 3oo4, the formula for calculating the probability of random failure (PFD) is: When the voting structure is 4oo4, the formula for calculating the probability of random failure (PFD) is: When the voting structure is 5005, the formula for calculating the probability of random failure (PFD) is: When the voting structure is 6006, the formula for calculating the probability of random failure (PFD) is: ； Step S142 further includes: when the logic unit voting structure is 1oo1, 1oo2, 2oo2, 1oo3, 2oo3 and 3oo3, the formula for calculating the random failure probability PFD is the corresponding formula in step S141.

5. The PFD calculation method based on reliability block diagrams as a function of required rate and detection scheme, as described in claim 4, is characterized in that... The specific steps of S150 are as follows: S151. Calculate the random failure probability (PFD) of the SIF loop: The random failure probability PFD of the SIF loop is the sum of the random failure probabilities PFD of the sensing unit, logic unit, and execution unit. S152. Calculate the risk reduction factor RRF, the formula is: RRF = 1 / PFD Where PFD is the random failure probability of the SIF loop; S153. Based on the correspondence table between PFD values ​​and SIL levels, obtain the random failure SIL; S154. Determine the SIF loop SIL level based on the random failure SIL and the architecture constraint SIL, specifically: the smaller of the random failure SIL and the architecture constraint SIL is the SIF loop SIL level.

6. The PFD calculation method based on reliability block diagrams as a function of required rate and detection scheme, as described in claim 1, is characterized in that... Specifically, step S160 involves exporting a reliability block diagram report and a SIF overview table from the software, containing basic project information, loop structure, data analysis, reliability block diagram model, analysis and calculation conclusions, PFD distribution diagram, and MTTF distribution diagram.

7. The PFD calculation method based on reliability block diagrams as a function of required rate and detection scheme, as described in claim 1, is characterized in that, The sensing unit includes all components in the SIF loop from the field sensing element to the input terminal of the logic controller. The logic unit includes an input card, an output card, a power module, and a controller; The execution unit includes all components in the SIF loop from the output of the logic controller to the final field actuator.

8. The PFD calculation method based on reliability block diagrams as a function of required rate and detection scheme, as described in claim 1, is characterized in that... The software is independently developed and specifically refers to a SIL verification tool based on reliability block diagrams, which incorporates the PFD calculation method.

9. The PFD calculation method based on reliability block diagrams as a function of required rate and detection scheme, as described in claim 8, is characterized in that... The SIL verification tool based on reliability block diagrams includes a component failure database management function, which can perform add, modify, and query operations on the data in the component failure database.

Images