Auto-evaluation on urban road traffic atmospheric pollution based on gray class and fuzzy ISODATA

The advantages of both grey clustering method and fuzzy ISODATA method were analyzed and colligated. First, the decision-making evaluation results of uncertainty system with small sample were acquired with grey clustering method. Then, applying the fuzzy ISODATA model to learn and revise the results of above grey clustering, an optimal fuzzy classification could be produced through iterative operation. A cluster center matrix also could be obtained which could be used to achieve automatic decision-making evaluation on new samples. The model would be verified with classification coefficient and average fuzzy entropy. Considering the characteristics of urban road traffic and according to the national ambient air quality standards, the experimental sample data of city road traffic pollutants were normalized. The evaluation matrix of air pollution condition of road traffic was got and the credibility of the results of evaluation model would be further enhanced. The entropy theory was used to verify the system model. Concluding remarks summarized the work and analyzed further studies.