English

Journal of Chinese Agricultural Mechanization

Journal of Chinese Agricultural Mechanization ›› 2022, Vol. 43 ›› Issue (12): 67-74.DOI: 10.13733/j.jcam.issn.2095-5553.2022.12.011

Previous Articles     Next Articles

Design and experiment on multiparameter environment monitoring system in aquaponics

Qiu Yujun, Wang Xiaochan, Li Tianpei, Ge Pengbiao   

  • Online:2022-12-15 Published:2022-12-02

鱼菜共生环境中多参数监测系统设计与试验

仇宇俊1,汪小旵2,李天沛2,葛朋彪3   

  1. 1. 南京农业大学人工智能学院,南京市,210031; 2. 南京农业大学工学院,南京市,210031;

    3. 苏州鼎兴斯沃水产养殖设备有限公司,江苏苏州,215011
  • 基金资助:
    江苏省重点研发计划(BE2021362);江苏省现代农机装备与技术示范推广项目(NJ2020—03)

Abstract: Realtime monitoring of key environmental indexes in aquaponics system is of great significance to the water quality control of the whole system. A multiparameter environmental information monitoring system based on GPRS is designed. The system can remotely monitor eleven environmental parameters of water quality information in aquaculture area and vegetable cultivation area, upload the data to the cloud server, and then realize the realtime monitoring, historical data query, remote regulation and other control through PC and mobile terminal. Combined with various environmental information, the composition of ammonia nitrogen and water quality are analyzed. The prediction model of ionic ammonia concentration is established by multiple linear regression. The test results show that the multiparameter monitoring system runs smoothly and the success rate of data acquisition is about 99.53%. The coefficient of determination R2 of the established multiple linear regression equation of ionic ammonia is 0.817, and the MAPE of the prediction results is 468%, which can effectively predict the concentration of ionic ammonia in aquaponics and provide early warning.

Key words: aquaponics, Internet of Things, GPRS, ammonia nitrogen, multiple linear regression

摘要: 实时监测鱼菜共生系统中的关键环境信息对整个系统的水质调控具有重要意义。设计一种基于GPRS的多参数环境信息监测系统,系统可对水产养殖区与蔬菜栽培区中共11项环境参数进行远程监测,并将数据上传至云端服务器,再通过PC端以及移动端实现实时监测、历史数据查询、远程调控等功能,联合多种环境信息对氨氮的组成以及水质状况进行分析,同时将获取的环境数据通过多元线性回归的方法建立离子氨浓度预测模型。试验结果表明,设计的系统运行平稳,数据采集成功率约为99.53%;建立的离子氨多元线性回归方程决定系数R2为0.817,预测结果平均绝对百分比误差MAPE为4.68%,可以有效预测养殖环境的离子氨浓度,实现预警。

关键词: 鱼菜共生, 物联网, GPRS, 氨氮, 多元线性回归

CLC Number: