[1]李敬民,陈斯,唐海晨,等.基于P-ISSA-GRU模型的养殖水体溶解氧含量预测[J].江苏农业学报,2025,(09):1781-1790.[doi:doi:10.3969/j.issn.1000-4440.2025.09.013]
　LI Jingmin,CHEN Si,TANG Haichen,et al.Prediction of dissolved oxygen content in aquaculture water based on P-ISSA-GRU model[J].,2025,(09):1781-1790.[doi:doi:10.3969/j.issn.1000-4440.2025.09.013]
点击复制

基于P-ISSA-GRU模型的养殖水体溶解氧含量预测()

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2025年09期
页码:
1781-1790
栏目:
农业信息工程
出版日期:
2025-09-30

文章信息/Info

Title:
Prediction of dissolved oxygen content in aquaculture water based on P-ISSA-GRU model
作者:
李敬民; 陈斯; 唐海晨; 杨增汪
(江苏师范大学物理与电子工程学院,江苏徐州221116)
Author(s):
LI Jingmin; CHEN Si; TANG Haichen; YANG Zengwang
(School of Physics and Electronic Engineering, Jiangsu Normal University, Xuzhou 221116, China)
关键词:
溶解氧含量预测; 皮尔逊相关系数; 改进的麻雀搜索算法（ISSA）; 门控循环单元（GRU）
Keywords:
prediction of dissolved oxygen content; Pearson correlation coefficient; improved sparrow search algorithm (ISSA); gated recurrent unit (GRU)
分类号:
TP391
DOI:
doi:10.3969/j.issn.1000-4440.2025.09.013
文献标志码:
A
摘要:
为了解决养殖水体溶解氧(DO)含量预测精度低的难题,本研究提出了一种基于改进的麻雀搜索算法(ISSA)优化门控循环单元(GRU)的养殖水体溶解氧含量预测模型(P-ISSA-GRU)。通过皮尔逊(Pearson)相关系数法确定水质中各种因子与溶解氧含量的相关系数,选取强关联因子为模型输入特征；通过引入Tent混沌映射改进种群初始化,自适应动态权重因子ω动态改变权重系数以及高斯扰动(GP)改进最优位置更新,增强了麻雀搜索算法(SSA)在寻找全局最优解和局部最优解的能力,加快了其收敛速度；通过ISSA优化GRU网络,进行模型参数的优化搜索,构建了非线性溶解氧含量预测模型(P-ISSA-GRU)。试验结果表明,P-ISSA-GRU模型与其他5个常用的模型相比显示出更好的预测效果,均方误差(MSE)为0.152 (mg/L)2、平均绝对误差(MAE)为0.311 mg/L、均方根误差(RMSE)为0.390 mg/L、决定系数（R2）为0.984。因此,本研究建立的P-ISSA-GRU模型与传统模型相比在一定程度上提高了对养殖水体溶解氧含量的预测性能。
Abstract:
In order to solve the problem of low prediction accuracy of dissolved oxygen (DO) content in aquaculture water, this study proposed a gated recurrent unit (GRU) prediction model based on an improved sparrow search algorithm (ISSA). The correlation coefficients of various factors in water and dissolved oxygen content were determined by Pearson correlation coefficient method, and the strong correlation factors were selected as the input features of the model. By introducing Tent chaotic mapping to improve population initialization, adaptive dynamic weight factor ω to dynamically change the weight coefficient, Gaussian perturbation (GP) to improve the optimal location updating, the ability of sparrow search algorithm (SSA) to find global and local optimal solutions was enhanced, and its convergence speed was increased. A nonlinear dissolved oxygen content prediction model (P-ISSA-GRU) was constructed by optimizing the GRU network using ISSA for parameter search. The experimental results showed that the P-ISSA-GRU model exhibited superior prediction performance compared to five other commonly used models. The mean square error (MSE)was 0.152 (mg/L)2, the mean absolute error (MAE)was 0.311 mg/L, the root mean square error(RMSE) was 0.390 mg/L, and the coefficient of determination (R2) was 0.984. Therefore, compared to traditional models, the P-ISSA-GRU model developed in this study demonstrates improved predictive performance for dissolved oxygen content in aquaculture water.

参考文献/References:

[1]XIAO R C, WEI Y G, AN D, et al. A review on the research status and development trend of equipment in water treatment processes of recirculating aquaculture systems[J]. Reviews in Aquaculture,2019,11(3):863-895.
[2]吴静,李振波,朱玲,等. 融合ARIMA模型和GAWNN的溶解氧含量预测方法[J]. 农业机械学报,2017,48(1):205-210,204.
[3]CAO X K, LIU Y R, WANG J P, et al. Prediction of dissolved oxygen in pond culture water based on K-means clustering and gated recurrent unit neural network[J]. Aquacultural Engineering,2020,91:102122.
[4]徐龙琴,李乾川,刘双印,等. 基于集合经验模态分解和人工蜂群算法的工厂化养殖pH值预测[J]. 农业工程学报,2016,32(3):202-209.
[5]陈英义,程倩倩,方晓敏,等. 主成分分析和长短时记忆神经网络预测水产养殖水体溶解氧[J]. 农业工程学报,2018,34(17):183-191.
[6]HUAN J, LI H, LI M B, et al. Prediction of dissolved oxygen in aquaculture based on gradient boosting decision tree and long short-term memory network:a study of Chang Zhou fishery demonstration base,China[J]. Computers and Electronics in Agriculture,2020,175:105530.
[7]NEAL C, NEAL M, WICKHAM H, et al. The water quality of a tributary of the Thames,the Pang,southern England[J]. Science of the Total Environment,2000,251:459-475.
[8]徐岩,程姝,薛艳静. 基于故障暂态电流Pearson相关系数的直流配电网保护[J]. 华北电力大学学报(自然科学版),2021,48(4):11-19.
[9]LIU S Y, XU L Q, LI D L, et al. Prediction of dissolved oxygen content in river crab culture based on least squares support vector regression optimized by improved particle swarm optimization[J]. Computers and Electronics in Agriculture,2013,95:82-91.
[10]YANG T, QIU W, MA Y, et al. Fuzzy model-based predictive control of dissolved oxygen in activated sludge processes[J]. Neurocomputing,2014,136:88-95.
[11]施珮,袁永明,张红燕,等. GRNN和Elman神经网络在水体溶解氧预测中的应用[J]. 江苏农业科学,2017,45(23):217-221.
[12]刘双印,徐龙琴,李道亮,等. 基于蚁群优化最小二乘支持向量回归机的河蟹养殖溶解氧预测模型[J]. 农业工程学报,2012,28(23):167-175.
[13]任妮,鲍彤,刘杨,等. 基于粒子群优化算法和长短时记忆神经网络的蟹塘溶解氧预测[J]. 江苏农业学报,2021,37(2):426-434.
[14]宦娟,刘星桥. 基于K-means聚类和ELM神经网络的养殖水质溶解氧预测[J]. 农业工程学报,2016,32(17):174-181.
[15]TA X X, WEI Y G. Research on a dissolved oxygen prediction method for recirculating aquaculture systems based on a convolution neural network[J]. Computers and Electronics in Agriculture,2018,145:302-310.
[16]XU H H, DENG Y. Dependent evidence combination based on shearman coefficient and Pearson coefficient[J]. IEEE Access,2017,6:11634-11640.
[17]孙龙清,吴雨寒,孙希蓓,等. 基于IBAS和LSTM网络的池塘水溶解氧含量预测[J]. 农业机械学报,2021,52(1):252-260.
[18]曾一川,曾会国,袁伟皓,等. 长江口入海通道水质综合分析与模型预测[J]. 环境工程,2022,40(5):95-102,108.
[19]王佳钰,郝思鹏,李森文,等. 基于ES-GRU-LSTM的风电场群功率预测[J]. 计算技术与自动化,2022,41(3):37-41.
[20]黄春桃,范东平,卢集富,等. 基于深度学习模型的广州市大气PM_(2.5)和PM_(10)浓度预测[J]. 环境工程,2021,39(12):135-140.
[21]王海瑞,鲜于建川. 改进麻雀搜索算法在分布式电源配置中的应用[J]. 计算机工程与应用,2021,57(20):245-252.
[22]马晓宁,李笑含. 基于Tent混沌映射的可复制的鲸鱼算法[J]. 计算机仿真,2022,39(8):363-368.

备注/Memo

备注/Memo:
收稿日期：2024-11-26基金项目：国家自然科学基金项目(61975070)；徐州市重点研发计划项目(KC21087)作者简介：李敬民(1997-),男,江苏邳州人,硕士研究生,主要研究方向为机器学习、智能水产养殖。(E-mail)2694760723@qq.com通讯作者：陈斯,(E-mail)chensism@126.com
更新日期/Last Update: 2025-10-27