Integration of model predictive control and process safety

We present a method that combines offline trajectory optimization and online Model Predictive Control (MPC), generating robust controllers for complex periodic behavior in domains with unilateral constraints (e.g., contact with the environment).MPC offers robust and adaptive control even in high-dimensional domains; however, the online optimization gets stuck in local minima when the domains has discontinuous dynamics.Some methods of trajectory optimization that are immune to such problems, but these are often too slow to be applied online.In this paper, we use offline optimization to find the limit-cycle solution of an infinite-horizon average-cost optimal-control task.We then compute a local quadratic approximation of the Value function around this limit cycle.Finally, we use this quadratic approximation as the terminal cost of an online MPC.This combination of an offline solution of the infinite-horizon problem with an online MPC controller is known as Infinite Horizon Model Predictive Control (IHMPC), and has previously been applied only to simple stabilization objectives.Here we extend IHMPC to tackle periodic tasks, and demonstrate the power of our approach by synthesizing hopping behavior in a simulated robot.IHMPC involves a limited computational load, and can be executed online on a standard laptop computer.The resulting behavior is extremely robust, allowing the hopper to recover from virtually any perturbation.In real robotic domains, modeling errors are inevitable.We show how IHMPC is robust to modeling errors by altering the morphology of the robot; the same controller remains effective, even when the underlying infinite-horizon solution is no longer accurate.

This paper studies the application of Infinite Horizon Model predictive Control (MPC) and model reduction by means of Hankel norm to chemical process of interest in the field of control of large, complex and networked systems. In this paper we first describes the model of the complete process which is composed by three reactors and three distillation columns. Later, we show the main aspects about model reduction by means of Hankel norm and Infinite Horizon MPC and finally the model obtained through numerical linearization and model reduction is used to design centralized Infinite Horizon Model predictive Control (MPC).

Off-site hydrogen refueling stations (HRS) handle large volumes of high-pressure hydrogen, requiring precise and systematic risk-reduction strategies. In this study, a Hazard and Operability (HAZOP) analysis was performed for an off-site HRS, and Layer of Protection Analysis (LOPA) was conducted for four risk-level-4 events using two different approaches. The Functional Safety only LOPA, based on IEC 61511, and the All Safeguards LOPA, developed according to the Center for Chemical Process Safety (CCPS) guideline, were both applied. The Functional Safety only approach, which considers only automated protection layers, required Safety Integrity Level (SIL) ratings of 1 and 2, whereas the All Safeguards approach, accounting for mechanical and procedural protection layers, achieved the Target Mitigated Event Likelihood (TMEL) in all scenarios without additional SIL requirements. Consequently, it was confirmed that the definition of protection layer scope significantly influences the required SIL, design cost, and system complexity. This study proposes a hybrid approach in which all safeguards are considered during the early design stage, while in the final design stage, protection measures are evaluated from a functional safety perspective in accordance with IEC 61511 to ensure both design efficiency and safety integrity.

Model predictive control for nonlinear batch processes with asymptotically perfect tracking

Using Dynsimr to study the implementation of advanced control in a Propylene/Propane Splitter

An implementable stabilizing model predictive controller applied to a rotary flexible link: An experimental case study

Layer of Protection Analysis (LOPA) has been widely adopted as a method of organizing and quantifying hazard rates. Frequently, an initial LOPA may indicate that a risk target has not been met and that a scenario requires additional layers of protection. Unlike Fault Tree Analysis, the inherent simplicity of the LOPA format makes it conducive for use directly as a problem solving tool in a team environment to close risk gaps. Teams at Air Products with representatives from operations, process controls, process design and process safety, have effectively used the LOPA format in problem‐solving sessions to identify and improve components of the analysis involving human error, controls architecture, and inherently safer process designs. This multifunctional team approach has resulted in broader ownership of the safety analysis by the full team of stakeholders, and this has ultimately led to more cost effective solutions. The benefits of multifunctional team review are illustrated with a case study example. © 2016 American Institute of Chemical Engineers Process Saf Prog 36: 257–263, 2017

Layer of Protection Analysis – Quantifying human performance in initiating events and independent protection layers

In this paper tuning procedures are outlined for the infinite horizon model predictive control (MPC). The infinite horizon formulation takes advantage of recent extensions to the theory of MPC, which allows the use of infinite horizons in the presence of hard constraints. The model predictive control structure which will be considered here is a constrained regulator in series with a stable observer. The solution makes use of H/sub /spl infin// loopshaping and results on the normalized left coprime factorization (NLCF) optimal control problem for discrete time systems.

This paper presents an infinite horizon model predictive control (MPC) scheme for constrained linear parameter-varying systems. We assume that the time-varying parameter can be measured online and exploited for feedback. The proposed method is based on a parameter-dependent control law which is obtained via the repeated solution of a convex optimization problem involving linear matrix inequalities (LMIs). Closed-loop stability is guaranteed by the feasibility of the LMIs at initial time. Compared to existing algorithms with static linear control law and more restrictive LMI conditions, the proposed scheme reduces conservatism and improves performance, which is confirmed by a simulation example.

Iterative learning control (ILC) is often used to eliminate repetitive disturbances and improve the tracking performance in the industrial processes. Due to its disability of reacting against real-time disturbances, the integration of model predictive control (MPC) and ILC constitutes the model predictive iterative learning control (MPILC) to improve its ability of rejecting non-repetitive disturbances. MPILC is suitable for linear models which has poor performance for highly nonlinear systems. A nonlinear MPILC (NMPILC) global controller based on T-S model is put forward in this paper. The global controller is composed of a set of local MPILC controllers designed applying the local linear models of the T-S model. The convergence property has been demonstrated. A PWR nuclear power plant is adopted to illustrate the performance of the NMPILC.

Discussion on: “Improved MPC Design based on Saturating Control Laws”

For the purpose of solving problems in water treatment system of the power station boiler from perspective of chemical process safety, a new method which uses the LOPA method in risk assessment for anticorrosion and scale inhibition properties of boiler system is proposed. First of all, a new annotation on the basic principle and calculation steps of LOPA is made in this paper. And then the consequence of the hazard scenario is determined by calculating the probability of accidents, such as putting poor quality water treated by the water treatment agents as initial event. Finally, specific implementable recommendations are provided by comparing the risk levels of independent protective layers system and ordinary system. The results show that the method that LOPA used in risk assessment for anticorrosion and scale inhibition properties of boiler is feasible, measures that recommended can reduce the risk; LOPA can further enrich HAZOP analysis results when introduced into boiler capacity link; layers of protection analysis, risk assessment together with humanistic management could be a new and effective security management model used by energy enterprise.

The paper indicates advantages (and disadvantages) of layer of protection analysis (LOPA) over other methods to support its paramount position among different process safety analysis (PSA) methods in use today. Its simplicity (using order-of-magnitude estimates for all elements that constitute an accident scenario), universality (easy adaptation to particular needs), directness to indicate the effect of risk reduction measures as well as to assess the total level of risk to be compared with the company risk tolerance criteria are strong arguments for wider application of LOPA. However, the method may have some limitations (disadvantages) in comparison to other methods and therefore different extended approaches of LOPA have been developed to overcome the problem. Those approaches include an inclusion of an expert system into accident scenario identification (ExSysLOPA), new approach to deal with uncertainty connected with input data (fuzzy LOPA) as well as for explosion at workplace risk assessment (ExLOPA). Some of these combined techniques will be presented here and they support and extend the applications of layer of protection analysis, especially for safety assurance assessment of risk-based determination that is used in oil and gas process industries.

Integration of model predictive control and backstepping approach and its stability analysis