论文部分内容阅读
复杂系统或过程参数优化问题往往采用建模发现其潜在规律,再通过优化方法利用该规律获取最佳工艺参数。而建模误差的存在,往往使优化解与实际最佳工艺参数存在差距,难以获得理想性能。为此提出一种基于误差补偿模型的优化决策方法,通过分析并选取影响建模误差的因素,构建误差补偿模型,修正模型,提高决策性能。首先,从数据挖掘角度建立复杂工艺近似模型,并分析影响建模误差的主要因素;其次,以训练误差为导师信号,利用BP网络建立影响因素与建模误差之间的函数关系,确定误差补偿函数;最后,将近似模型与补偿函数叠加作为最终的工艺模型。数学仿真与电路系统优化实验结果表明:误差补偿后,仿真模型得到的优化函数最优值相对误差降低9.63%,而电路系统中决策参数的超调量下降2.17%。可见,补偿模型优化参数控制效果优于近似模型,验证了所提方法对于提高工艺参数优化决策性能有效性。
Complex system or process parameter optimization problems are often found using modeling the underlying law, and then use the law to optimize the optimal process parameters. However, the existence of modeling error often makes the optimization solution and the actual optimum process parameters have a gap, it is difficult to obtain the desired performance. Therefore, an optimization decision method based on the error compensation model is proposed. By analyzing and selecting the factors that affect the modeling error, the error compensation model is constructed and the model is modified to improve the decision performance. First of all, to establish the approximate model of complex process from the perspective of data mining, and to analyze the main factors that affect the modeling error. Secondly, using the training error as the supervisor signal, using BP network to establish the functional relationship between the influencing factors and the modeling error, Function; Finally, the approximate model and the compensation function superimposed as the final process model. The experimental results of mathematical simulation and circuit system optimization show that the relative error of optimal value of optimization function decreases by 9.63% after error compensation, while the overshoot of decision parameters in circuit system decreases by 2.17%. It can be seen that the control effect of the compensation model optimization parameters is better than the approximate model, which verifies the effectiveness of the proposed method in improving the performance of process optimization parameters.