D. K. M. Kufoalor

Embedded Model Predictive Control on a PLC using a primal-dual first-order method for a subsea separation process

The results of a PLC implementation of embedded Model Predictive Control (MPC) for an industrial problem are presented in this paper. The embedded MPC developed is based on the linear MPC module in SEPTIC (Statoil Estimation and Prediction Tool for Identification and Control), and it combines custom ANSI C code generation with problem size reduction methods, embedded real-time considerations, and a primal-dual first-order method that provides a fast and light QP solver obtained from the FiOrdOs code generator toolbox. Since the primal-dual first-order method proposed in this paper is new in the control community, an extensive comparison study with other state-of-the-art first-order methods is conducted to underline its potential. The embedded MPC was implemented on the ABB AC500 PLC, and its performance was tested using hardware-in-the-loop simulation of Statoils newly patented subsea compact separation process. A warm-start variant of the proposed first-order method outperforms a tailored interior-point method by a factor of 4 while occupying 40% less memory.

Automatic deployment of industrial embedded model predictive control using qpOASES

Different high-speed quadratic programming (QP) solvers are incorporated into an ANSI C code generation framework for embedded Model Predictive Control (MPC). The controllers developed are based on step response (linear) models and design configurations obtained from SEPTIC, Statoils software tool for MPC applications. In order to achieve high online computational efficiency, offline computations/preparations are made at the code generation stage, and appropriate problem data are used in the QP solvers. We discuss implementation aspects arising when running an embedded MPC controller on an industrial PLC and present results of hardware-in-the-loop simulation tests for two challenging industrial applications. The results indicate that the online active-set strategy as implemented in the software package qpOASES exhibits superior performance compared to both a tailored interior-point method and a primal-dual first-order method for the step response class of models considered in this paper.

Reconfigurable fault tolerant flight control based on Nonlinear Model Predictive Control

Constrained Nonlinear Model Predictive Control (NMPC) is shown to have potentials for reconfigurable fault tolerant control of highly nonlinear, intrinsically unstable, high performance aircraft. Results on fault tolerance of NMPC autopilots were obtained for an F-16 fighter aircraft model, without the implementation of any prestabilizing controllers. It has been shown that NMPC has inherent fault detection capabilities due to its effective utilization of feedback and its internal model predictions. Actuator (control surface) faults, including extreme cases of total actuator failure are examined as test cases for the NMPC reconfigurable fault tolerant control scheme developed in this work. The NMPC autopilots implementation and simulations were done using the ACADO nonlinear optimization solver.

Efficient quadratic programming frameworks for industrial embedded model predictive control

Model Predictive Control (MPC) has proven to be successful in numerous advanced high-level process control applications based on software implementation in PC/server technology. However, for challenging applications offshore and subsea, ultra-reliable industrial embedded hardware, such as Programmable Logic Controllers (PLCs), are usually more suitable. The limited computational resources in low power industrial embedded devices, in combination with increased demands for computational speed and system reliability, motivate the development of new solution methods and software tools that enable the use of MPC in industrial embedded realtime applications. This tutorial aims at presenting different approaches for both formulating and efficiently solving industrial MPC problems. A brief review of existing methods will be covered, and new results that expand the computational toolbox for solving step response MPC problems will be presented. The new results include a novel MPC scheme that incorporates step response data in a traditional manner, while facilitating the use of block factorization in efficient QP solution methods. The new MPC scheme introduces a stage-wise prediction formulation that enables the use of both tailored Riccati recursion and condensing algorithms that can be embedded into an interior-point method. Implementation aspects necessary for high performance on embedded platforms are discussed, and results on a PLC are presented. Efficient MPC implementations for a subsea compact separation process application, using fast QP solvers such as HPMPC, qpOASES, FiOrdOs, and CVXGEN, will be used in the discussions.

Efficient implementation of step response models for embedded Model Predictive Control

Abstract This paper proposes efficient step response model implementation strategies that lead to accurate control and high computational performance in an embedded Model Predictive Control (MPC) scheme. Different implementations of the step response prediction model are examined, and inherent properties that directly affect control performance in the presence of disturbances are discussed. Model errors that are inconsistent with bias updates (i.e. the model of unknown disturbances commonly used in step response MPC) are identified, and it is shown that the bias updates may worsen the effect of the errors in some cases. Particular attention is paid to the robustness of the prediction models to small truncation errors and errors in the input or measured disturbance history. Several implementation aspects that are crucial for embedded targets with limited resources are discussed. The findings are illustrated by simple simulation examples and an industrial case-study involving hardware-in-the-loop simulation of a subsea compact separation process.

Scalability of QP solvers for embedded model predictive control applied to a subsea petroleum production system

The performance of two different Quadratic Programming (QP) solvers for embedded Model Predictive Control (MPC), FiOrdOs and qpOASES, is evaluated for a relevant case study from the petroleum industry. Embedded MPC for the considered system is implemented on a PLC (Programmable Logic Controller) using both solvers. The focus is on the computation time and memory requirements of the solvers as the dimensions of the control problem increase. The results show that qpOASES has a superior performance for small systems with respect to computation time. However, qpOASES has a near cubic growth in computation time with respect to the number of system variables, while FiOrdOs only has a near linear growth. FiOrdOs may thus be faster for larger systems. FiOrdOs has a smaller memory footprint than qpOASES for small systems; however, the program size grows faster than with qpOASES, and for the largest system configuration, the program sizes were almost identical. For even larger systems, qpOASES may have smaller program memory requirements than FiOrdOs, though qpOASES requires more data memory for all problem sizes.