Irrigation Prediction of Greenhouse Flower under Water Stress and Acoustic Emission Based on BP Neural Network
Based on acoustic emission signal, water shortage of flower in greenhouse is detected to judge whether the irrigation should be needed. The control of the irrigating amount is difficult. So it is important to research on irrigation for flower growth and watersaving. In this paper, the irrigating system based on BP arithmetic was established. Acoustic emission signal, temperature, transpiration and illumination were served as input parameters, and predicted value of soil water storage was served as output parameter. The irrigating amount can be predicted. Experimental results showed that this system had lesser errors. The relative errors between detection value and prediction value were about 5％. In order to improve the predicting precision, it is recommended to use BP arithmetic and the one based genetic algorithm.
acoustic emission waterstress BP arithmetic irrigating amount prediction
Zhaobo Huang Ling Shi Tianhui Zhang Bing Kong
Faculty of Information Science and Engineering Yunnan University Kunming 650001,China Faculty of Engineering and Technology Yunnan Agriculture University Kunming 650201,China
国际会议
太原
英文
453-457
2011-02-26（万方平台首次上网日期，不代表论文的发表时间）