猪舍氨气与二氧化碳浓度变化时序预测模型优化  

Optimal Prediction Model for Gas Concentrations of NH3 and CO2 Time-series in Pig House

作　　者：谢秋菊[1] 马超凡 王圣超 包军[2,3] 刘洪贵 于海明[1] XIE Qiuju;MA Chaofan;WANG Shengchao;BAO Jun;LIU Honggui;YU Haiming(College of Electrical and Information,Northeast Agricultural University,Harbin 150030,China;College of Animal Science and Technology,Northeast Agricultural University,Harbin 150030,China;Key Laboratory of Swine Facilities Engineering,Ministry of Agriculture and Rural Affairs,Harbin 150030,China;Engineering Research Center of Pig Intelligent Breeding and Farming in Northeast Cold Region,Ministry of Education,Harbin150030,China)

机构地区：[1]东北农业大学电气与信息学院,哈尔滨150030 [2]东北农业大学动物科技学院,哈尔滨150030 [3]农业农村部生猪养殖设施工程重点实验室,哈尔滨150030 [4]教育部北方寒区智能化繁育与养殖工程研究中心,哈尔滨150030

出　　处：《农业机械学报》2023年第7期381-391,共11页Transactions of the Chinese Society for Agricultural Machinery

基　　金：国家自然科学基金面上项目(32072787);东北农业大学东农学者计划项目(19YJXG02);黑龙江省博士后资助项目(LBH-Q21070)。

摘　　要：NH3质量浓度和CO_(2)质量浓度是猪舍环境精准控制的重要指标。由于畜禽舍气体浓度具有时变性、非线性耦合等特点,目前有害气体浓度预测模型存在预测精度低的问题。提出了基于门控制循环单元(Gated recurrent unit,GRU)、改进麻雀搜索算法(Improved sparrow search algorithm,ISSA)并融合差分整合移动平均自回归模型(Autoregressive integrated moving average model,ARIMA)的有害气体浓度时序数据预测模型ISSA-GRU-ARIMA。首先构建了GRU气体浓度时序预测模型,然后通过引入Tent混沌序列、混沌扰动和高斯变异增强ISSA算法的局部寻优能力,实现GRU模型超参数优化;然后利用统计学习ARIMA方法提取优化后的ISSA-GRU模型预测残差的线性特征,最终达到提升模型预测精度的目的。以采集的52 d猪舍环境的1248组数据对模型进行训练和测试。结果表明,ISSA-GRU-ARIMA模型NH3质量浓度预测的均方根误差(RMSE)、平均绝对百分比误差(MAPE)和决定系数R2分别为0.263 mg/m^(3)、8.171%和0.928,CO_(2)质量浓度预测的分别为55.361 mg/m^(3)、4.633%和0.985。本文构建的ISSA-GRU-ARIMA模型具有较高的预测精度,可为猪舍有害气体浓度精准控制提供科学依据。Concentrations of ammonia and carbon dioxide are important indicators for indoor environment control in pig house.Due to the time-varying and nonlinear coupling characteristics of gas concentration,the prediction accuracy of pig house environment prediction models is still relatively low.Aiming to achieve the precision control for gases concentration in pig house,a time-series data prediction model named ISSA-GRU-ARIMA for harmful gas concentrations was proposed based on gated recurrent unit(GRU),improved sparrow search algorithm(ISSA)fused with autoregressive integrated moving average model(ARIMA).Firstly,a GRU gas concentration time series prediction model was constructed,and Tent chaotic sequence,chaotic disturbance and Gaussian mutation were introduced to enhance the local optimization ability of ISSA algorithm and optimize the hyperparameters of GRU model;then the statistical learning ARIMA method was used to extract the linear features of the optimized ISSA-GRU model’s prediction residuals in order to improve the prediction accuracy of the model.A dataset with 1248 environment data that collected for 52d was used for model training and testing.It was shown that the RMSE,MAPE andR2of ISSA-GRU-ARIMA model for ammonia concentration prediction were 0.263mg/m^(3),8.171%and 0.928,respectively,and those for carbon dioxide concentration prediction were 55.361mg/m^(3),4.633%and 0.985,respectively.The constructed ISSA-GRU-ARIMA had high predictive performance,it can provide scientific basis for accurate control of harmful gases in pig house.

关 键 词：猪舍空气质量 环境控制 时序预测 残差 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]