Optimal Iterative Learning Control Based on a Time-Parametrized Linear Time-Varying Model for Batch Processes

Self Cite

Robust model predictive control (MPC) is widely applied to batch processes to handle its intrinsic plant-model mismatch by considering the worst-case scenario and thus leads to conservative performance. Identification methods that make use of system repetitiveness have been invented and combined with robust MPC to reduce conservativeness. In this paper, we show that in addition to the repetitiveness, time-wise continuity in system dynamics can also help to improve model accuracy and accelerate the convergence rate. Specifically, we propose a two-dimensional online identification method in the framework of set membership identification. The method is capable of (1) employing time-wise correlation and cycle-wise repetitiveness in system dynamics; (2) identifying both of the nominal value and the corresponding uncertainty set of the unknown parameters, thus being amenable to robust MPC schemes; and (3) guaranteeing convergence to the true parameters under mild conditions. Numerical examples are provided to illustrate its fast convergence and superiority in improving prediction accuracy and control performance.

“…In this example, we consider a temperature control problem of a fed-batch reactor where a liquid-phase irreversible exothermic chemical reaction A → B is taking place …”

Section: Numerical Examplesmentioning

confidence: 99%

Control-Oriented Two-Dimensional Online System Identification for Batch Processes

Gao

et al. 2021

Self Cite

“…The emulation of biological systems with respect to learning, thinking, and making decisions represents the main feature of making these new strategies more adaptive and less restricted to a limited set of models. Being able to modify control actions in real time to improve plant performance without any model of the system is what makes intelligent control strategies unique. , In this scenario, agent-based control has been used to emulate intelligent control because it provides distributed intelligence, adaptability, and autonomy in the decision-making process. ,− Agent-based controls provide the potential to effectively manage complex and highly coupled systems such as hybrids . The coordination among agents through the use of particle swarm optimization, ant colony optimization, or biologically inspired parallel computing optimization can be used as examples to show how more adaptability and flexibility can be provided in the control actions to accommodate system changes or demand. − Agents can be added or removed as needed to sufficiently account for the complexity of the problem and for the number of objectives that have to be achieved. , Agent-based control is frequently used as a macro-level controller in micro grid applications, , smart grids systems, and manufacturing chains.…”

Section: Introductionmentioning

confidence: 99%

A Novel Multiagent Resource Sharing Algorithm for Control of Advanced Energy Systems

Pezzini

Bryden

Tucker

2021

This paper implements a novel resource sharing control strategy on a fuel cell−gas turbine hybrid power system at the National Energy Technology Laboratory's Hybrid Performance Facility (Hyper). In a fuel cell−gas turbine hybrid power system, the simultaneous interaction of the gas turbine and the fuel cell creates a tightly coupled environment characterized by conflicting dynamics. In this paper, a model-free control approach is applied to solve the tightly coupled control problem posed by this challenging environment. Specifically, this control problem is presented as a resource sharing problem that can be solved using a resource sharing algorithm that is defined based on the distribution construction concept. This algorithm creates computational agents and solves the problem through the distribution and redistribution of shared resources defined as blocks. Two agents were created; the first agent (agent 1) controls the gas turbine speed by adjusting the electric load, and the second agent (agent 2) controls the cathode mass flow through the fuel cell using the cold-air bypass valve. A parametric study was performed over the course of 15 experimental tests for both agents 1 and 2 through an evaluation of the responses based on setpoint changes. The algorithm was shown to have behavior comparable to a previously implemented multiinput multioutput state-space controller, which was designed through a model-based control approach. The resource sharing algorithm was able to find stable performance during run-time operations without any prior system knowledge identification on the power plant and without creating models used in traditional control strategies.

“…In the study of Zhao et al, a two-dimensional (2D) LPV modeling method is proposed by considering the 2D nature of data in the batch process. Xu et al developed a time-parameterized LTV model to represent the batch process, in which model coefficients are parameterized as nonlinear functions of the time index. Furthermore, this idea was extended to build the time-partitioned piecewise affine output error model …”

Section: Introductionmentioning

confidence: 99%

Nonparametric Identification Based on Multi-inherited Gaussian Process Regression for Batch Process

Chen

Zhao

et al. 2020

Self Cite

This study presents a multi-inherited Gaussian process regression (GPR)-based nonparametric identification method for batch process. In the GPR framework, the impulse response of each time point consists of two parts: one is the inheritance part for utilizing the model information of previous time points, in which a Gaussian prior is imposed over the unknown inheritance weight, and the other is the residual impulse response, which is interpreted as a zero-mean Gaussian process. Following the empirical Bayes approach, we derive the joint estimation of the inheritance weight and the residual impulse response. The hyperparameters are determined by maximizing the marginal likelihood. As the correlation of process dynamics at adjacent time points is considered by model inheritance, the proposed method can effectively improve the estimation accuracy. Finally, we demonstrate the superiority of the proposed identification method in two case studies.