基于GA-BP算法的含油污泥处理 工艺参数优化

在含油污泥进行资源化处理过程中,针对处理目标受多个因素影响的实际,为了解决工艺之间的耦合问题,采用正交实验的方法来解决,并把主要参数作为优化对象,把含油污泥的脱水率作为评价目标,通过采用GA-BP算法对含油污泥耦合工艺正交实验参数进行了线性与非线性分析.在采用遗传算法优化神经网络的权值和阈值的基础上,用优化后的权值和阈值对测试样本和训练样本进行了预测.预测结果表明,预测误差都有明显减小,分别由0.34211减少到0.031549和0.15476减少到0.040682,可见耦合参数趋向于非线性优化.

Parameters optimization of oil sludge treatment process based on GA-BP algorithm

In order to solve the coupling problems in the oil sludge treatment for process, for the actual process effecting by a number of factors, the orthogonal experiment method was used to solve the problems during oil sludge treatment for resources. The main parameters were taken as the optimization objects and taken the oil sludge dewatering rate as the evaluation objective by using GA-BP algorithm to analyze the orthogonal experiment parameters of the oily sludge coupling process from linear and nonlinear. The optimized weights and thresholds were used to predict the test samples and training samples on the basis of genetic algorithm to optimize the neural network weights and thresholds. The prediction results showed that the prediction errors were significantly reduced,from 0.34211 to 0.031549 and from 0.15476 to 0.040682, respectively, showing that the coupling parameters tended to nonlinear optimization.