Fuzzy Expert System Based On Greenhouse Environment Control

With the emergence of greenhouse, the agriculture production can get rid of therestriction of the natural climate conditions. It can make plants realize periodic andanti-season production. After adopting the greenhouse, it can solve problems of traditionalagriculture and will allocate resources more reasonable. For the masses of farmers, if theywant to use greenhouse reasonable, then they must be reasonably controlled greenhouseenvironment. But, greenhouse environment is a multivariable, nonlinear and complexsystem. It is very difficult to regulate reasonably. Especially for our country, thegreenhouse control technology is more relatively backward than the abroad. Moreover, ourcountry’s farmers are lack of experience for greenhouse control. And, they could onlyregulate the greenhouse parameters roughly. In addition, the number of agricultural expertsin greenhouse control technology in our country is relatively few. Therefore, it is verymeaningful to this paper to design a fuzzy expert system for greenhouse parametersregulation.In this paper, the design of fuzzy consists of input and output modules, rule basemodule, the fuzzy inference module, the defuzzification module. The working principle ofthe system is, first of all computer receive the real-time data which sensor send to, then thisdata has been fuzzificated. Through the fuzzy reasoning query the corresponding fuzzyrules in the rule base. Finally the execution strategy has been output.Through collect the three parameters (temperature, humidity, light intensity) whichhave the greatest influence on crop growth. The three environment parameters are as theinput quantity. Then these environment parameters have been fuzzification processed. Atthe same time, the actuators have been fuzzification processed. In the rule base module, itneed to establish the fuzzy rules. This part is the core of fuzzy expert system. It has been determined by expert experience and research achievements of the past. Throughfuzzification, fuzzy rules are determined by the fuzzy set of dividing the article number. Inthe fuzzy inference module, it has been determined by the method of maximummembership degree function after the fuzziness of input of the fuzzy rules. In VS2010programming environment, the design of the fuzzy expert system of the greenhouse cropparameters control has been realized. Some common situation of validation has beenproved in this paper. The design of fuzzy expert system in this paper is feasible.