A Study On Barn Ventilation Intelligent Decision Based On Information Fusion

Grain is not only the material of human survival, but also a country’s strategic resources which concerns social harmony and rapid and healthy development of economy. In the process of grain storage, timely and effective ventilation is an important measure to prevent mildew of grain and ensure the safety of grain. Faced the complex grain situation information, how to develop a scientific and accurate ventilation strategy to avoid phenomenon such as ineffective ventilation, excessive ventilation and harmful ventilation, has become a focus problem which needs to overcome in grain storage industry.For the defects and deficiencies of the traditional ventilation decision-making methods based on artificial experience, this paper introduces information fusion technology into barn ventilation decisions to improve and innovate existing ventilation mode decision method using information fusion technology, and proposes two methods which about barn ventilation intelligent decision for the whole use process of barn. The main contents are as follows:1. This paper studies existing intelligent ventilation control systems and information fusion techniques, analyses the realization and deficiencies of the existing intelligent ventilation control system, and build barn ventilation decision information fusion model based on information fusion.2. The system studies the related theory model of information fusion and intelligent ventilation control system, then aiming at the shortcomings of the existing ventilation decision method, this paper improves original decision-making methods using the D-S evidence theory, and proposes a ventilation decision method based on multiple D- S fusion model. In the process of fusion decision-making on a variety of grain condition information, the improved method firstly group the obtained evidence according to certain rules, and makes a higher level of fusion decision-making for the results of each group, to improve accuracy of decision-making results, and to reduce the risk of ventilation decisions.3. The system studies the basic theory of BP neural network algorithm, makes a detailed analysis and derivation of BP, and introduces the genetic algorithm(GA) into BP neural network to optimize its weights and thresholds. This paper designs a ventilation decision method based on GA- BP neural network to achieve intelligent decisions of barn ventilation.Experimental results demonstrate the effectiveness of that based on multiple D-S fusion algorithm and GA-BP neural network model of decision-making in the ventilation proposed in this paper, which achieve the coverage for whole process of the granary use from early to long-term. Barn built in the early, due to the lack of historical data, ventilation decision-making method based on multiple D-S fusion model is used to provide decision support for barn ventilation, the accumulation of grain situation data is available for neural network training until after a period of time, and then a better performance of ventilation decision-making method based on GA- BP neural network provides more stable decision support for barn ventilation.