| Greenhouse plays an important role in the realization of crop’s off-season production,the solution of annual production of vegetables and the balanced listing of vegetables.However,at present,greenhouse control in China is still dominated by traditional control mode,which is subjective,lacking of scientific decision-making,serious waste of water,fertilizer and electricity resources,and can not meet the production demand of modern greenhouse.It has become the key link to improve the social and economic benefits of the greenhouse to carry out the research on the precise control of the intelligent greenhouse and realize the intelligent control of the greenhouse.It has important research value and practical significance to realize the high yield,energy saving and low consumption of the modern greenhouse.This paper focuses on the research of multi-sensor data fusion,environmental factor prediction and control decision-making aiming at energy saving in the intelligent greenhouse environmental control system,aiming to provide solutions and core support technologies for the realization of reliable,real-time and accurate greenhouse intensive control.The specific research contents and conclusions are as follows:Firstly,aiming at the problems of large error,multiple conflicts and redundancy in the multi-node data collection of greenhouse environmental information,a multi-sensor data fusion algorithm based on wavelet denoising and adaptive weighting is proposed to preprocess the collected data.The results show that the method can effectively reduce the noise and redundancy in the original data,get the data fusion value with small variance,improve the accuracy of the measurement data and reduce the data transmission volume;at the same time,the method has good robustness,can achieve the reliability and consistency description of greenhouse environmental information,and provide reliable data for the subsequent greenhouse environmental modeling.Secondly,in view of the lag of equipment control in Greenhouse,a forecast model of greenhouse microclimate based on fuzzy neural network is proposed.According to the change of outdoor environment and the switch of greenhouse equipment as interference item,the forecast of environmental factors such as temperature,humidity,illumination and so on in greenhouse can be realized.The results of model validation show that the correlation of temperature,humidity and illuminance model is more than 95%,which has a good prediction effect and can provide an effective basis for subsequent greenhouse control decision.Finally,the decision-making of greenhouse control is studied for the optimization of greenhouse energy consumption.Taking the fan,spray system and heating system as the main energy consumption equipment,taking the lowest energy consumption as the optimization goal,combined with the greenhouse microclimate prediction model,using genetic algorithm to achieve the optimal solution of the objective function.The simulation results show that compared with the traditional control mode,the algorithm can save energy and reduce consumption significantly,which proves the feasibility and effectiveness of the algorithm. |
