Greenhouse Temperature Control Algorithm Based On The Web Greenhouse Crop Cultivation And Management Expert System

This paper is an important part of "Research and Development on Practical Intelligent Greenhouse for Protected Horticulture" (one key project of Zhejiang province) . By studying intelligent control algorithm of greenhouse and an expert system for plant growing and management based on Web, it can help greenhouse managers to plant crop reasonably in greenhouse, increase the level of greenhouse management and create more benefits of society and economic.The greenhouse system is a more variable and non-line system, which includes the phenomenon of delaying. It is also a half-open system, which key technology is condition control. It is not finished only by hand. So, automation is very necessary. In all, the temperature is an important parameter. What's more, it is very difficult to form a math model for this system because it is affected by many conditions. Thus, it is not controlled to realize accurate control by traditional way and modern means. On the other hand, the expert system is a hotspot of intelligence. And the agricultural expert system has agricultural technology. What's more, it is applied in the Internet. Which will quicken the prevalence of information and quicken the development of agricultural expert system.This paper introduces the PI control of temperature by concluding the status of horticulture and the control algorithm in and abroad. It gives the study on optimizingof PI -Smith compensation control of amending model and single nerve cellcontrol, They increase the stability of system by innovating the PI control. Now, the controller of two inputs and one output is applied in our country. We bring forward the innovational fuzzy controller of two inputs and two outputs and design the course of fuzzy control. The paper also studies the fuzzy PI control to achieve the control of greenhouse temperature by uniting the PI and fuzzy control.In this paper, Smith control of amending model is to settle the time delaying of PI, the single nerve cell optimizing control is to optimize the parameters of PI. Which are simulated by MATLAB. The simulation results show that the Smith control avoids the bad effect of time delaying, the control course is postponed only in time, but the control shape is not changed. The single nerve cell optimizing control saves the time of response, the changing values are becoming little, which increases the stability of system.Otherwise, the paper brings forward fuzzy control to avoid the disadvantage of PI, which can't form an accurate math model. The fuzzy control system adopts two input and two output, which stand for error , error changeable of temperature and heat , fenestration variable. The fuzzy rules are easy, the system is extended easily...