Study On Key Technology Of Greenhouse Intelligent Controlling Method Based On FPGA

Facility agriculture is a new-style agricultural production technology which has developed rapidly along with the breakthrough of the agricultural environment project technology in recent decade. Greenhouse is the typical representative of facility agriculture .How to achieve the highly effective monitoring and low cost running of greenhouse is a significant problem that needs to tackle with. In view of problems in current intelligent control algorithms of greenhouse monitoring systems such as: needing PC machine involvement which will increase the cost and hard to popularize, the off-line systems which are lack of intelligence, scientific characteristic and accuracy, this paper aims to explore and propose a new intelligent controlling method of scientific and accurate merit ,and integrate it with FPGA. The major research contents and conclusions of this paper are as follows:(1) Proposed a method using FPGA to integrate the complex intelligent controlling algorithm in the greenhouse controlling for the first time. This method can reflect the expert knowledge without connecting to the huge expert systems on PC machines, and meanwhile decrease the costs.(2) Proposed a new greenhouse intelligent controlling method based on RBF net and FALCON net, which can be easily realized in hardware. The greenhouse environment model based on RBF net has perfect effect of fitting to the expert knowledge, can make the controlling more scientific and real-time. The greenhouse controlling method based on FALCON can solve the problems of non-linearity, changeable, highly parameter coupling and long lag of the greenhouse environment, and make the controlling more accurate and have good robustness.(3) Researched and realized the floating ALU based on FPGA that is needed in this research. Simulation results show that the format of the floating data is scientific,floating ALU have good precision and take less resources.(4) In view of the insufficient of custom methods as inquiry form method and Taylor series method, proposed using CORDIC algorithm to realize the computing of the exponential function of ex .This method takes less resource, more effective and high accurate.(5) Researched and realized the neurons of RBF and FALCON based on state machine on FPGA. Results show that this method can make the program have good extension and can be reused easily.