Key Issues Of Agricultural Plant Protection Forecasting And Decision Support Systems Several

Agriculture plant protection system of China is the key system that decides the whole country’s eating problem. This system is very important for agriculture production. Forecast and decision are the core in the plant protection system, which involve the whole production process about production organization and management, pest control and forecasting etc. Agriculture plant protection system is a complex system. It includes many of undefined and uncertain factors that lead to this system showing grey system characteristic. If we can not deal with the factors well in the agriculture plant protection system; it may cause many huge negative results about the income of agriculture system member and even the Chinese provisionment.Establishing process of Agriculture plant protection forecasting system has many uncertain factors and technical difficulties. These problems are the key factors in this system and are the key in the production precision and control process. During this project research, the author participated in the product and plant test in test field organized by the plant protection institute of Yunnan Agricultural University, Kunming plant protection unit, and other eight counties plant protection units in Yunnan province. In the last few years the author got a large number of test data collection and test research, and cooperative studied with a large number of agricultural plant protection experts, as well as collected advices and opinions from the peasant farmers, basic technique personnel, technical personnel, plant protection experts who came from counties, cities and province, and considerately analysed research on the difficult problems in plant protection system. At last the author selected the plant protection forecasting and decision support system as his research direction. At the same time, parts of achievements from this paper have been used in Yunnan province which achieved high academic, social and economic benefits.The paper based on the difficulties from traditional experience and methods in agricultural plant protection system, and actively used modern science and technology advantages, and introduced knowledge management methods and techniques, such as management engineering, computer, information collection, information processing, database and artificial intelligence etc to deeply research on traditional plant protect and forecast system which was primary with manual and experience. Discussing and improving the science of the principle and method, using the method and technology from management engineering and information technology established a series of application models, and developed them by computer. The application models have been tested for a long time.The paper mainly to solve the following problems:1, The prediction of threshold period is a widespread problem in plant protection system, and people always uses the experience model to judge the traditional threshold period, thus the universality of the prediction system is very poor. This paper deals with the traditional experience model scientifically, and put forward the self-adaptive problem of the weeding threshold period by weeding threshold period prediction as an example. The solution of the parameters dynamic self-adaptive method thoroughly changes the difficult problems of the poor self-adaptive of small sample experience model in different crop growth environment.2, Using fuzzy evaluation method establishes the superiority evaluation model of weeds in the field and the herbicides effect evaluation model, and provides the method for advantage factors determine and effect evaluation in plant protection decision support system.3, Using vector synthesis method to establish the disease transmission model, solves the problem that spread trend is difficult to predict in traditional plant protection forecast system, and provides a method to the automatic prediction of the factors trend for the spread of the disease.4, This paper quantitatively processes the most important factor of the assessment on agricultural expert, but it’s usually keep changeless in traditional method. It puts forward the concept of dynamic adjustment degree, and establishes the model of support degree dynamic adjustment, that provides a set of methods for the automatic operation of experts to participate the plant protection decision support system and the objective description the importance of experts in system.5, Density and accuracy of the data acquisition is the key to precision agriculture, which is the key to decision. Generally speaking, the higher density acquisition gets the more accurate prediction, however the higher collection density the cost of collection is more expensive, finally you can’t afford to it. This paper puts forward two sets of low-cost acquisition system structure scheme and implementation technologies, and thoroughly resolves field data collection problem, which lay the foundation for the implementation of precision agriculture.Although these theories and methods, are based on traditional management engineering ideas and methods, have certain innovation in solving agricultural forecast and decision. They provide a new train and method for the precision of agricultural production and scientific management, change the model of mainly empirical decision in agricultural production a certain degree, and have some demonstration effects on forecast and decision support system in agricultural production protection field. The paper’s data acquisition experienced several years. Although workload is huge, and technical difficulty is higher, also have the more achievements.