(147ad) Path-Sampling and Machine Learning for Rare Un-postulated Abnormal Events

Broadly speaking, my research interests focus around the intersection of process safety, process modeling and simulation, data science, optimization, and robotics. Specifically, I'm interested in developing novel methods and tools to improve safety and reliability of chemical manufacturing processes.

In the realm of process safety and reliability, through my research, I aim to explore new aims to analyze and predict the likelihood of rare unanticipated abnormal events, using process modeling and simulation methods such as Monte Carlo techniques, to better understand the behavior of complex chemical systems and identify potential risks.

In addition, I've always been fascinated by data science and am interested in leveraging its power to enhance the reliability of chemical processes. Through effective data-mining techniques and application of machine learning algorithms to process historian data, I aim to develop predictive models that can alert operators to potential issues and suggest corrective actions. Furthermore, I see tremendous potential in developing physics-informed neural networks for analysis of rare abnormal events, given our understanding of the theoretical formalisms underlying rare-event phenomena and the limited availability of rare-event data.

Finally, I'm also excited by the future of robotics and automation in the process industry - despite having higher capital costs, autonomous robots can be utilized efficiently for tasks that are dangerous for humans (e.g., working at heights, cleaning and maintenance of toxic chemical tanks, etc.), although the risks involved in human-robot interaction and safety need to be studied elaborately.

Abstract

Chemical manufacturing processes can present significant dangers, and therefore, it is vital to incorporate safety and reliability measures during their design. To reduce the probability of catastrophic accidents, which can have grave consequences on human life and the environment, comprehensive instrumentation such as control systems, alarms, and automated Safety Instrumented Systems (SIS) are regularly utilized in chemical processes. Additionally, common reliability assessment methods such as failure mode and effect analysis (FMEA), fault-tree analysis (FTA), reliability-centered maintenance (RCM), root-cause analysis (RCA), and the like, have proven to be effective in identifying and handling postulated abnormal events that have occurred previously or are more likely to occur, based on process historian data. However, it is difficult to predict and counteract the impact of infrequent and unforeseeable un-postulated abnormal events in real-time, which, when not considered during process design, can lead to the most serious consequences. Hence, existing reliability/safety systems, alone, might prove to be insufficient in monitoring and alerting the operator for un-postulated abnormal events.

Previously, we developed an advisory system for analyzing and monitoring process reliability, consisting of novel, multivariable alarms and reliability systems introduced using process modeling and path-sampling for un-postulated abnormal events (Sudarshan et al., 2021; Sudarshan et al., 2022). Its purpose is to augment and support existing reliability systems, suggesting actions when unanticipated reliability/quality events are approached. Our analyses were demonstrated initially on an exothermic CSTR process and led to promising alarm thresholds and reliability response actions (Sudarshan et al., 2023a). Herein, we extend our analyses to a more complex case study; i.e., a polystyrene CSTR exhibiting a free-radical polymerization (FRP) mechanism. Polymerization reactors are known to exhibit complex, nonlinear behavior; e.g., output multiplicity, input multiplicity, and isolas — often, it is desirable to operate at the intermediate unstable region, with potential abnormal transitions to multiple undesirable operating regions, and to avoid non-minimum phase behavior leading to inverse response. This is a control, safety, and reliability problem, and hence, a good application for our analyses. Given multiple undesirable operating regions for the polystyrene CSTR, the path sampling and dynamic risk analyses are extended to develop multi-directional alarm and reliability systems.

Next, simple rationalization strategies were introduced, wherein the acceptability of every alarm threshold and response action was evaluated, with the alarm thresholds and/or response actions modified accordingly, based on key statistical metrics — seeking to ensure that every alarm is a quality alarm, and its response action is justified appropriately. For the exothermic CSTR, our strategies resulted in a significant reduction in the number of nuisance alarms, focusing on only quality alarms, which, if ignored, were more likely to result in an abnormal shift in operation to the undesirable regions (Sudarshan et al., 2023b). Similar rationalization strategies are being applied to the advisory system developed for the polystyrene CSTR.

Additionally, the real-time performance of the rationalized alarms and reliability systems is evaluated using dynamic risk assessment, in which, the risk associated is analyzed by estimating the failure probabilities of the reliability systems (Pariyani et al., 2012b; Moskowitz et al., 2015), based on multiple dynamic simulations for the process, inclusive of control, alarms and reliability systems. Expectedly, the failure probability distribution developed for the exothermic CSTR had a much lower variance as compared to one developed using a flat prior — similar low-variance probability distributions are constructed for the rationalized alarms and safety/reliability systems for the polystyrene CSTR.

Keywords: Un-postulated Abnormal Events, Path-Sampling, Advisory System, Alarms, Dynamic Risk Assessment

References:

Sudarshan, V., Seider, W.D., Patel, A.J., Arbogast, J.E., 2021. Understanding rare safety and reliability events using forward-flux sampling. Computers and Chemical Engineering 153.

Sudarshan, V. , Seider, W. D. , Patel, A. J. , Oktem, U. G. , Arbogast, J. E. , 2022. Alarm and Safety System Design Using Forward Flux Sampling. AIChE Annual Meeting Conference Proceedings.

Sudarshan, V., Seider, W.D., Patel, A.J., Oktem, U.G., Arbogast, J.E., 2023a. Reliability Path Sampling and Dynamic Risk Analysis – Part I – Developing an Advisory System To Recognize Rare Unpostulated Reliability Events – in preparation.

Sudarshan, V., Seider, W.D., Patel, A.J., Oktem, U.G., Arbogast, J.E., 2023b. Reliability Path Sampling and Dynamic Risk Analysis – Part II – Alarm Rationalization Strategies – in preparation.

Pariyani, A., Seider, W.D., Oktem, U.G., Soroush, M., 2012b. Dynamic risk analysis using alarm databases to improve process safety and product quality: Part II-Bayesian analysis. AIChE Journal 58

Moskowitz, I.H., Seider, W.D., Soroush, M., Oktem, U.G., Arbogast, J.E., 2015. Chemical process simulation for dynamic risk analysis: A steam-methane reformer case study. Ind. Eng. Chem. Res 54, 4347–4359.