| 基于改进粒子群算法优化BP神经网络的废水处理软测量模型 |
| |
| 引用本文： | 何丹,林来鹏,李小勇,牛国强,易晓辉,黄明智.基于改进粒子群算法优化BP神经网络的废水处理软测量模型[J].华南师范大学学报(自然科学版),2021,53(2):114-120. |
| |
| 作者姓名： | 何丹 林来鹏 李小勇 牛国强 易晓辉 黄明智 |
| |
| 作者单位： | 广东省广业环保产业集团有限公司,广州510030;华南师范大学环境学院∥广东省化学品污染与环境安全重点实验室∥教育部环境理论化学重点实验室,广州510006;华南师范大学环境学院∥广东省化学品污染与环境安全重点实验室∥教育部环境理论化学重点实验室,广州510006;华南师范大学环境学院∥广东省化学品污染与环境安全重点实验室∥教育部环境理论化学重点实验室,广州510006;广东省环境资源利用与保护重点实验室∥中国科学院广州地球化学研究所,广州510640 |
| |
| 基金项目： | 国家自然科学基金项目41977300广东省自然科学基金项目2016A030306033广东省科技计划项目2017B030314057福建省科技计划项目2020I1001广州市民生计划项目202002020055 |
| |
| 摘 要： | 针对废水处理过程BP神经网络软测量模型受处理过程非线性特征影响，存在收敛速度慢、陷入局部极小点等问题，用改进的粒子群算法(PSO)优化BP神经网络，建立废水处理过程中出水化学需氧量(CODeff)与出水固体悬浮物(SSeff)的软测量模型(PSO-BP模型)，并与基于遗传算法-BP神经网络算法的模型(GA-BP模型)及BP模型的预测效果进行对比. 研究结果表明:采用PSO-BP模型预测CODeff时，均方根误差(RMSE)、决定系数(R2)分别为3.995 5、0.640 1，而用于预测SSeff时，RMSE、R2分别为0.650 3、0.681 1；相比BP模型和GA-BP模型，PSO-BP模型对CODeff的预测性能分别提高了4.49%、0.44%，对SSeff的预测性能分别提高了40.11%、24.77%.
|
| 关 键 词： | 废水处理 BP神经网络 改进的粒子群算法 软测量 |
| 收稿时间： | 2020-11-09 |
A Soft Measurement Model of Wastewater Treatment Process Based on BP Neural Network with Improved Particle Swarm Optimization Algorithms |
| |
| Institution: | 1.Guangdong Guangye Environmental Protection Industry Group Co., Ltd, Guangzhou 510030, China2.School of Environment// Guangdong Provincial Key Laboratory of Chemical Pollution and Environmental Safety//MOE Key Laboratory of Theoretical Chemistry of Environment, South China Normal University, Guangzhou 510006, China3.Guangdong Provincial Key Laboratory of Environmental Protection and Resources Utilization//Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou 510640, China |
| |
| Abstract: | Considering that the BP neural network-based soft sensor model of wastewater treatment process is affected by the nonlinear characteristics of the system and has such problems as low convergence speed and local mini-mum, a hybrid soft measurement PSO-BP model for predicting the effluent chemical oxygen demand (CODeff) and the effluent solid suspended matter (SSeff) in the wastewater treatment process is developed based on BP neural network with improved particle swarm optimization algorithms. This model is compared with models based on genetic algorithm-BP neural network (GA-BP model) and BP neural network. The research results show that when using the PSO-BP model to predict CODeff, the root mean square error (RMSE) and the coefficient of determination (R2) are 3.995 5 and 0.640 1, respectively. When it is used to predict SSeff, RMSE and R2 are 0.650 3 and 0.681 1, respectively. Compared with the BP model and the GA-BP model, the prediction performance of the PSO-BP model on CODeff is improved by 4.49% and 0.44% respectively, the prediction performance of SSeff is improved by 40.11% and 24.77% respectively. |
| |
| Keywords: | |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《华南师范大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《华南师范大学学报(自然科学版)》下载免费的PDF全文 |
|
