Abstract:

In this paper, a real time improvement algorithm based on sensitivity information for numerical optimization of chemical plants is developed based on the original receding horizon optimization algorithm to solve the problems in the large-scale and nonlinear operation optimization of chemical plants such as high complexity, difficult convergence and poor real-time performance. While overcoming the uncertainty of optimization, this algorithm uses the sensitivity information of nonlinear programming optimization result to predict on line the approximate optimal solution of operation optimization problem in the next optimization cycle. Meanwhile, the accuracy of the approximate optimal solution is further improved by background computation and off-line correction, so as to ensure that the optimal solution of operation optimization problem can be obtained quickly and accurately. Finally, a typical chemical plant is selected as the simulation case and the optimization computation time is reduced by 64.3% compared with that of the traditional method, which verifies the validity of the real time improvement algorithm.

Key words:

chemical plant')">"> chemical plant, numerical optimization, receding horizon, real time, sensitivity