ANNSA: a hybrid artificial neural network/simulated annealing algorithm for optimal control problems

Department of Chemical Engineering, Indian Institute of Science, Bangalore 560012, India
Chemical Engineering Science (Impact Factor: 2.61). 07/2003; DOI: 10.1016/S0009-2509(03)00168-4

ABSTRACT This paper introduces a numerical technique for solving nonlinear optimal control problems. The universal function approximation capability of a three-layer feedforward neural network has been combined with a simulated annealing algorithm to develop a simple yet efficient hybrid optimisation algorithm to determine optimal control profiles. The applicability of the technique is illustrated by solving various optimal control problems including multivariable nonlinear problems and free final time problems. Results obtained for the different case studies considered agree well with those reported in the literature.

  • [Show abstract] [Hide abstract]
    ABSTRACT: In many industrial processes, especially chemistry and metallurgy industry, the plant is slow for feedback and data test because of complex and varying factors. Considering the multi-objective feature and the complex problem of production stability in optimal control, this paper proposed an optimal control strategy based on genetic programming (GP), used as a multi-step state transferring procedure. The fitness function is computed by multi-step comprehensive evaluation algorithm, which provides a synthetic evaluation of multi-objective in process state based on single objective models. The punishment to process state variance is also introduced for the balance between optimal performance and stability of production. The individuals in GP are constructed as a chain linked by a few relation operators of time sequence for a facilitated evolution in GP with compact individuals. The optimal solution gained by evolution is a multi-step command program of process control, which not only ensures the optimization tendency but also avoids violent process variation by adjusting control parameters step by step. An optimal control system for operation direction is developed based on this strategy for imperial smelting process in Shaoguan. The simulation and application results showed its effectiveness for production objects optimization in complex process control.
    Engineering Applications of Artificial Intelligence 08/2004; 17(5):491-500. · 1.96 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The aim of this paper is to present an approach to dynamic off-line optimization of batch emulsion polymerization reactors using a stochastic optimizer. The control objective is to find the optimal temperature profile that minimizes the final batch time constrained by the final conversion and molecular weight. In this study, we evaluate the applicability of MSIMPSA, a simulated-annealing-based algorithm, to solve the optimal control problem. Two cases are studied: first, a simple case without energy balances and, second, a more realistic case using energy balances and constraints on heat transfer. In addition, an SQP optimizer was applied to perform a local optimization around the best results obtained by MSIMPSA. The following conclusions can be drawn from the results: (i) MSIMPSA can be applied in an easy and straightforward manner (black-box approach) to such optimal control problems. (ii) Even though MSIMPSA is a stochastic algorithm, the best obtained solution is so good that it cannot be further improved by local optimization methods.
    Industrial & Engineering Chemistry Research 10/2004; 43(24). · 2.24 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: In the present work, it was developed an adaptive model predictive control algorithm to control a semi batch pyrolysis reactor. An 8L reactor and a separation system were assembled for this purpose. The reactor temperature control was carried out through a digital control system implemented for this process. The model used to infer about the process was a multilayered neural network completely recursive. To avoid offset problems an adaptive algorithm was applied, performing on-line weights actualization. The neural network was used to explicitly predict the process output (reactor temperature) through a pre-defined prediction horizon. Through optimization, this output vector was used to estimate the process input (heat power supply). A qualitative analysis of the products and the total time of operation for a fast pyrolysis, sustained for ten minutes in the set point temperature, had pointed out, in this case, a superior performance to the proposed controller when compared with the classical feedback controller. Besides, the temperature stabilizes without overshoots and offsets. The developed control algorithm was able to compensate the strong disturbances that occur during the partial discharge of pyrolysis products, due to reactor pressure relief. A performance index based on ISA criteria was used and again, exhibits consistent improvement of the adaptive control over a classical feedback algorithm.

Full-text (2 Sources)

Available from
May 27, 2014