Safety Instrumented System Risk Modeling for Faster Reliability Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complex machine systems, such as turbomachines, face challenges in predicting and maintaining the reliability of safety instrumented systems (SIS) due to their intricate nature and the time-consuming process of deriving performance measures like probability of failure on demand (PFD) and risk reduction factor (RRF), which are typically based on historical data and require new models for each system configuration.
Innovation Solution
A model-based reliability system (MRS) is developed, utilizing a library of reusable component models and a dynamic risk calculation engine (DRCE) that employs Markov models, Fault Tree Analysis, and other techniques to analyze and update SIS performance measures in real-time, allowing for efficient proof test scheduling and maintenance optimization across various configurations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional historical data-based methods are used to derive SIS performance measures, then reliability predictions can be made, but the process is time-consuming and requires new models for each system configuration
Solution Approach 1:
The patent pre-calculates and stores performance measures (PFD, RRF) for individual SIS components during the modeling phase. These pre-computed values are then rapidly aggregated using mathematical formulas during operations, eliminating the need to re-run complex simulations for each query and enabling fast reliability assessment.
Solution Approach 2:
The patent divides the SIS into discrete functional components (sensors, final elements, logic solvers) and models each separately. This segmentation allows independent analysis and pre-computation of each component's performance measures, which can then be combined to determine overall system reliability without analyzing the entire system from scratch.
2Reliability
If comprehensive SIS modeling is performed for each system configuration, then accurate reliability measures are obtained, but the complexity of modeling increases significantly
Solution Approach 1:
The patent segments the SIS into standardized functional components (sensors, final elements, logic solvers) with predefined modeling templates. Each component type has its own simplified model structure, reducing the overall complexity by breaking down the complex system into manageable, reusable segments with consistent modeling approaches.
Solution Approach 2:
The patent creates universal component models that can represent multiple SIS component types through standardized parameters and structures. These multi-functional models can be configured to represent different sensor types, final elements, or logic solvers by adjusting parameters, eliminating the need to create entirely new models for each component variant.
3Productivity
If real-time risk assessment is implemented, then maintenance optimization is improved, but computational resources and system complexity increase
Solution Approach 1:
The patent pre-computes and stores performance measures for all SIS components during the modeling and commissioning phases. During real-time operations, the system only needs to retrieve these pre-stored values and perform simple aggregations, avoiding complex real-time simulations and reducing computational burden while maintaining accurate risk assessment capabilities.
Solution Approach 2:
The patent implements a feedback mechanism where actual SIS performance data is continuously monitored and compared against the pre-computed performance measures. This feedback loop enables dynamic maintenance optimization by identifying deviations from expected performance, allowing the system to adapt maintenance schedules based on actual condition while using the pre-established model framework.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system may include a model library 22 configured to model a safety system, wherein the model library comprises a plurality of subsystem models 24, 26, 28, 30, 32, 34 and each of the plurality of subsystem models is configured to derive a reliability measure. The system further includes a fault tolerance input and a maintenance policy input. The system further includes a dynamic risk calculation engine (DRCE) 38 configured to use a user-defined set of the plurality of subsystem models, the fault tolerance input and the maintenance policy input, to derive a system risk for an apparatus 21.