Greenhouse environment has the characteristics such as multiple-input, multiple-output, nonlinearity, and difficulties of establishing its accurate mathematical model. Therefore, in order to make the greenhouse environment control more effective and accurate, an expert system of variable universe fuzzy control based on BP neural network has been proposed. In this method an appropriate universe is selected according to the error and the changing rate of error to overcome the effect for greenhouse control when the error is too large or too small. The paper introduces the design of greenhouse environment control system based on variable universe fuzzy control in details, and presents the main control parameters.