基于WA-ABC-WLSSVR的南美白对虾工厂化育苗溶解氧预测模型

发布时间：2018-01-29 11:22

本文关键词： 溶解氧预测加权最小二乘支持向量回归机人工蜂群算法小波分析参数组合优化　出处：《武汉大学学报(工学版)》2017年04期 　论文类型：期刊论文

【摘要】：为解决传统方法水质预测精度低、鲁棒性差等问题,提出了基于小波分析(WA)、人工蜂群(ABC)优化加权最小二乘支持向量回归机(WLSSVR)的工厂化育苗溶解氧组合预测模型(WA-ABC-WLSSVR模型).该模型采用小波分析对原始非平稳溶解氧时间序列数据进行多尺度特征提取,通过加权最小二乘支持向量回归机对不同尺度下的溶解氧数据子序列分别建模,利用改进人工蜂群优化算法(ABC)对各分量序列WLSSVR模型参数进行组合优化,最后叠加各尺度下的预测结果.运用该模型对工厂化育苗溶解氧进行预测,并与BPNN、标准LSSVR、WAACO-LSSVR、WA-PSO-LSSVR等模型对比分析,结果表明,该溶解氧预测模型具有较高的预测精度和泛化能力.
[Abstract]:In order to solve the problems of low precision and poor robustness of traditional water quality prediction method, wavelet analysis based on WAA is proposed. Optimization of weighted least Squares support Vector regression (WLSSVR) by artificial Bee Colony (ABC) A factory model for predicting dissolved oxygen (do) in plantlet breeding is presented in this paper, which is based on WA-ABC-WLSSVR model. The model uses wavelet analysis to extract the original non-stationary dissolved oxygen time series data with multi-scale feature extraction. The subsequences of dissolved oxygen data at different scales were modeled by weighted least squares support vector regression machine. The improved artificial bee colony optimization algorithm (ABC) is used to optimize the parameters of each component sequence WLSSVR model. The model was used to predict the dissolved oxygen (do) in industrial seedling production and compared with BPNN, standard LSS VRV WAACO-LSSVR. The results of WA-PSO-LSSVR and other models show that the model has high prediction accuracy and generalization ability.
【作者单位】：仲恺农业工程学院信息科学与技术学院;中航飞机研发中心飞控液压所;中航飞机汉中零组件制造有限公司;中国农业大学中欧农业信息技术研究中心;
【基金】：国家自然科学基金项目(编号:61471133,61473331) 国家科技支撑计划项目(编号:2012BAD35B07) 广东省科技计划项目(编号:2013B090500127,2013B021600014,2015A070709015,2015A020209171,2016A040402043) 广东省自然基金项目(编号:S2013010014629,2014A030307049) 广东海洋大学创新强校工程项目(编号:GDOU2014050227)
【分类号】：S968.22;TP18
【正文快照】： 南美白对虾工厂化育苗是在封闭水体中进行集约化育苗的一种新模式,其苗种是对虾养殖的基础,虾苗质量好坏直接决定着养殖收益的高低[1-2].而育苗水质是虾苗栖息场所,直接决定虾苗的成活率和健康生长,尤其对水质参数溶解氧要求更为严格.我国渔业水质标准GB1607-89对育苗水体溶解

【相似文献】