Study On Assessment Model Of Effects Of Simulated Acid Rain On Leafy Vegetables

Acid rain (acid rain) generally refers to the precipitation pH value of that is less than 5.6, including rain, hail, fog and other forms of precipitation. The term was proposed the first time in the "air and rain:the beginning of chemical climatology " by Smith (R.A.Smith), the British chemist, in 1872. Since then, acid rain has aroused the concern of scientists around the world. Now, the pollution of acid rain has become one of the first tenth global environmental problems.According to statistics, China has become the third biggest acid rain region following north America and Europe in the world. The acid rain area in China has covered nearly forty percent of land area. Acid rain has different damages with natural ecosystems, human health, constructions and so on, especially for crop physiology, growth and nutritional quality, etc.This discussion group selects leafy vegetables as research objects, which play an important role in the dietary structure of people in China and are sensitive to acid rain, and chooses three varieties of leafy vegetables as representative--cabbage, lettuce, rape, for acid rain simulation experiments. Though the analysis of experimental data analysis, detect acid rain effects on the mechanism of leafy vegetables and extent.In this paper, fuzzy comprehensive evaluation method and BP artificial neural network algorithm are modeled for researching the impact of acid rain on of leafy vegetables, through the establishment of comprehensive evaluation index system and index evaluation level by using principal component analysis to calculate the weight coefficient of each index factor, with standardized experimental data of simulated acid rain used as the evaluation samples for each comprehensive evaluation. The comprehensive evaluation of two methods reflects that the comprehensive evaluation level about the quality of leafy vegetables increases as the acidity decreases, while the sub acidity environment is beneficial for the quality of the combined effects of rape. the fuzzy comprehensive evaluation of acid rain with the two methods has the following characteristics:(1) through accurate means to deal with fuzzy objects, the blurry comprehensive evaluation can make the blurry information presented by the contained information more scientific, reasonable, close to the actual quantitative evaluation; (2) the result of the fuzzy comprehensive evaluation is a vector, not a point value, containing rich information; (3) in this paper, determining the index weight with the analysis of the principal component avoids the disadvantage of strong subjectivity about the fuzzy comprehensive evaluation method determining the index weight; (4)fuzzy comprehensive evaluation has some following disadvantages:there are more indicators; under the weight vector and the constraints of 1, the relative weights of the membership is smaller; the weight vector does not match with the fuzzy matrix R, resulting in a super-blur phenomenon, poor resolution, and it is hard to distinguish a higher degree of membership. The evaluation result has many samples the gap of which between the first largest membership degree and the second one is small.Artificial neural network BP algorithm characteristics are as follows:(1)through training the mapping from input to output, artificial neural network BP algorithm reflects the complex combined effects of acid rain on leafy vegetables in a certain extent, to achieve a complex nonlinear mapping function; (2) the calculation is clear and has a certain extension, the capacity of overview; (3) the training of artificial neural network BP algorithm is learnt on networks with accurate results, but the impact of acid rain on leafy vegetables is an exploration of process. The result is not determined, and the training of network reliability has yet to be verified; (4) artificial neural network BP algorithm require a higher number of training samples. The more training samples, the better learning networks. In this paper, because of using the classification standard value as the training sample, the sample size is only 5, so less number of training samples makes the network not precise enough; (5) in the evaluation results, the comprehensive evaluation results of the cabbage pH5.0 (1) is better than that of pH5.6 (1), which is inconsistent with the facts. It shows that BP network is lack of fault-tolerant capability for data, and the result includes abnormal value.Comprehensively comparing of two methods, each has advantages and disadvantages. But if the output is in the absence of established and training samples is enough, the fuzzy comprehensive evaluation results is more practical and more comprehensive.This paper is the exploratory research about comprehensive evaluation model of the impact of acid rain on leafy vegetables. Because there are no standard about the target selection, grading and weight and no determined test method, even to avoid using subjective methods during the research process, there is still some differences with the fact. These are some problems found in the research of evaluation model about the impact of acid rain on leafy vegetables.