Research on Prediction of Environmental Parameters in the Greenhouse Based on ISSA⁃ELM Model

doi:10.12422/j.issn.1672-6952.2024.04.010

Abstract

Abstract:

Traditional mechanism models of greenhouses are difficult to reflect the real greenhouse environment due to nonlinear, multivariate, and strongly coupled characteristics. In this paper, extreme learning machine (ELM), back propagation (BP) neural network, and support vector machine (SVM) are used to predict and analyze the temperature, humidity, and light intensity of the greenhouse. The results show that the predicted values of ELM model are the most similar to the real?time parameters of greenhouse environment. In order to further improve the prediction accuracy of environmental parameters in the greenhouse, the improved sparrow search algorithm (ISSA) is used to optimize ELM model in this paper. The predicted environmental parameters are in good agreement with the measured data of a greenhouse in Tianjin, which confirms the feasibility of the proposed prediction model for the control of greenhouse environment.

Key words: Environmental parameters, Prediction model, Extreme learning machine, Sparrow search algorithm

摘要：

温室环境系统具有非线性、多变量和强耦合的特点，传统的温室模型难以预测其真实环境。采用极限学习机、BP神经网络和支持向量机三种模型对温室温度、湿度和光照强度进行了预测分析，结果显示极限学习机模型预测值与温室环境实时参数最为相近。为提高温室环境参数的预测精度，采用改进的麻雀搜索算法对极限学习机模型进行优化，预测的环境参数与天津某温室实测数据吻合较好，证实了所提出预测模型用于温室环境调控的可行性。

关键词: 环境参数, 预测模型, 极限学习机, 麻雀搜索算法

CLC Number:

TK01

Yao WANG, Menghang ZHANG, Wei WANG, Jin WANG. Research on Prediction of Environmental Parameters in the Greenhouse Based on ISSA⁃ELM Model[J]. Journal of Liaoning Petrochemical University, 2024, 44(4): 75-81.

王瑶, 张孟航, 王伟, 王进. 基于ISSA⁃ELM模型的温室环境参数预测研究[J]. 辽宁石油化工大学学报, 2024, 44(4): 75-81.

Figures/Tables 11

Fig. 1 The structure of a single hidden layer feed?forward neural network

Fig. 2 Flowchart of the ISSA?ELM prediction model

Table 1 Initialization parameters of SSA?ELM model optimization algorithm

参数	数值	参数	数值
种群规模	20	发现者比例	0.7
进化次数	50	警戒者比例	0.2
预警值	0.6

Table 2 Initialization parameters of ISSA?ELM model optimization algorithm

参数	数值	参数	数值
种群规模	20	预警值	0.6
进化次数	50	发现者比例	0.7
优化变量维度	2	警戒者比例	0.2
权重最大值	3	权重最小值	1

Fig. 3 Summer greenhouse temperature, humidity and light intensity change curve from time to time

Fig. 4 Correlation of greenhouse environmental parameters

Fig. 5 Evaluation results of greenhouse temperature prediction models

Fig. 6 Evaluation results of the greenhouse humidity prediction model

Fig. 7 Evaluation results of the light intensity prediction model

Fig. 8 Prediction results of greenhouse temperature, greenhouse humidity and light intensity

Table 3 Evaluation results of the prediction model

参数	模型	MAE	MAPE/%	MSE	RMSE	R²
温室温度	SSA⁃ELM	1.031	4.16	1.962	1.400	0.890
温室温度	ISSA⁃ELM	1.006	4.13	1.590	1.261	0.917
温室湿度	SSA⁃ELM	4.290	6.50	28.500	5.300	0.831
温室湿度	ISSA⁃ELM	3.880	5.75	25.510	5.050	0.874
光照强度	SSA⁃ELM	1.880×10³	733.00	9.610×10⁶	3.120×10³	0.887
光照强度	ISSA⁃ELM	1.710×10³	380.00	9.240×10⁶	3.040×10³	0.901

References 17

1	ALAOUI M E， CHAHIDI L O， ROUGUI M， et al. Evaluation of CFD and machine learning methods on predicting greenhouse microclimate parameters with the assessment of seasonality impact on machine learning performance［J］. Scientific African， 2023， 19： e01578.
2	李其操，董自健. 基于GA⁃BP神经网络的温室温度预测研究［J］. 智能计算机与应用， 2023， 13（9）： 168⁃171.
	LI Q C， DONG Z J. Greenhouse temperature prediction based on GA⁃BP neural network［J］. Intelligent Computer and Applications， 2023， 13（9）： 168⁃171.
3	杨承磊，兰玉彬，王庆雨，等. 神经网络在温室小气候预测中的应用［J］. 中国农机化学报， 2023， 44（5）： 89⁃99.
	YANG C L， LAN Y B， WANG Q Y， et al. Application of neural network in greenhouse microclimate prediction［J］. Journal of Chinese Agricultural Mechanization， 2023， 44（5）： 89⁃99.
4	高国祥，王仰仁，田文艳，等. 基于实时监测数据的温室墒情预测研究［J］. 节水灌溉， 2020（10）： 34⁃40.
	GAO G X， WANG Y R， TIAN W Y， et al. Research on soil moisture content prediction of greenhouse based on real⁃time monitoring data［J］. Water Saving Irrigation， 2020（10）： 34⁃40.
5	ZHANG Y， XU L H， ZHU X H， et al. Thermal environment model construction of Chinese solar greenhouse based on temperature⁃wave interaction［J］. Energy and Buildings， 2023， 279： 112648.
6	BAGLIVO C， MAZZEO D， PANICO S， et al. Data from a dynamic simulation in a free⁃floating and continuous regime of a solar greenhouse modelled in TRNSYS 17 considering simultaneously different thermal phenomena［J］. Data in Brief， 2020， 33： 106339.
7	SABERIAN A， SAJADIYE S M. The effect of dynamic solar heat load on the greenhouse microclimate using CFD simulation［J］. Renewable Energy， 2019， 138： 722⁃737.
8	HE F， MA C W. Modeling greenhouse air humidity by means of artificial neural network and principal component analysis［J］. Computers and Electronics in Agriculture， 2010， 71（Suppl 1）： S19⁃S23.
9	YU H H， CHEN Y Y， HASSAN S G， et al. Prediction of the temperature in a Chinese solar greenhouse based on LSSVM optimized by improved PSO［J］. Computers and Electronics in Agriculture， 2016， 122： 94⁃102.
10	YANG Y B， ZHAO Y. Prevailing wind direction forecasting for natural ventilation djustment in greenhouses based on LE⁃SVM［J］. Energy Procedia， 2012， 16（Part A）： 252⁃258.
11	DING J T， TU H Y， ZANG Z L， et al. Precise control and prediction of the greenhouse growth environment of Dendrobium candidum［J］. Computers and Electronics in Agriculture， 2018， 151： 453⁃459.
12	邹伟东，张百海，姚分喜，等. 基于改进型极限学习机的日光温室温湿度预测与验证［J］. 农业工程学报， 2015， 31（24）： 194⁃200.
	ZOU W D， ZHANG B H， YAO F X， et al. Verification and forecasting of temperature and humidity in solar greenhouse based on improved extreme learning machine algorithm［J］. Transactions of the Chinese Society of Agricultural Engineering， 2015， 31（24）： 194⁃200.
13	LI B， WANG H L. Multi⁃objective sparrow search algorithm： A novel algorithm for solving complex multi⁃objective optimisation problems［J］. Expert Systems with Applications， 2022， 210： 118414.
14	那新宇，余华鹏，金鑫，等. 基于ISSA的多变量ORVFL网络自适应预测控制［J］. 辽宁石油化工大学学报， 2023， 43（1）： 80⁃88.
	NA X Y， YU H P， JIN X， et al. Multivariable ORVFL network adaptive predictive control based on ISSA［J］. Journal of Liaoning Petrochemical University， 2023， 43（1）： 80⁃88.
15	崔俊勇，李奇安. 改进WLSSVM模型在汽油干点预测中的应用［J］. 辽宁石油化工大学学报， 2023， 43（1）： 67⁃72.
	CUI J Y， LI Q A. Application of improved WLSSVM model in the prediction of gasoline dry point at the top of atmospheric Towe［J］. Journal of Liaoning Petrochemical University， 2023， 43（1）： 67⁃72.
16	郭丽莹，李文娜，郎宪明. SKPCA⁃LSSVM模型在汽油干点预测中的应用［J］.辽宁石油化工大学学报， 2022， 42（3）： 74⁃78.
	GUO L Y， LI W N， LANG X M. Application of SKPCA⁃LSSVM model in gasoline dry point prediction［J］. Journal of Liaoning Petrochemical University， 2022， 42（3）： 74⁃78.
17	MAHMOOD F， GOVINDAN R， BERMAK A， et al. Energy utilization assessment of a semi⁃closed greenhouse using data⁃driven model predictive control［J］. Journal of Cleaner Production， 2021， 324： 129172.

[1]	Xiaodong LU, Wei CHEN, Peijuan SHANG, Yijie WANG, Yongli PANG, Wei XIE, Junjie CAO, Qiyu HUANG. The Influence of Scale on the Strength of Wax Deposits and the Pigging Characteristics of Polyethylene Pipeline [J]. Journal of Liaoning Petrochemical University, 2024, 44(4): 33-43.
[2]	Xinyu Na, Huapeng Yu, Xin Jin, Yue Wang. Multivariable ORVFL Network Adaptive Predictive Control Based on ISSA [J]. Journal of Liaoning Petrochemical University, 2023, 43(1): 80-88.
[3]	Tao Wenhua， Chen Jiao， Gui Yunjin， Kong Pingping. Coking Energy Consumption RBF Prediction Model Improved by Differential Evolution Algorithm [J]. Journal of Liaoning Petrochemical University, 2020, 40(2): 91-96.
[4]	Zhao Zihan. Research Progress on Scaling Prediction and Scaling Prevention Technology in Oilfields [J]. Journal of Liaoning Petrochemical University, 2018, 38(05): 19-23.
[5]	Chang Jiaming,Li Weijun,Shi Chengjiang. Aluminum Foil Sealing Seal Detection Based on Gabor Transformation and Extreme Learning Machine [J]. Journal of Liaoning Petrochemical University, 2017, 37(3): 64-69.
[6]	ZHANG Xiaoling，LIU Feng，LI Junjun. Method of the Lifetime Prediction for Thermo Mechanical Fatigue in NickelBased Superalloy [J]. Journal of Liaoning Petrochemical University, 2013, 33(4): 65-69.