Research on aggregative index number method in water quality assessment based on SVM

On the basis of analysis of water quality assessment methods, an aggregative index number model in water quality evaluation is put forward in this paper. There are totally six indexes selected as evaluation parameters in this model, including DO, potassium permanganate, BOD5, NH3-N, total quantity of phosphorus and total quantity of nitrogen. Besides, this model fully considers their differences existed in different categories of water quality and pollutant contents of the same category water quality. And this established model overcomes its drawbacks of inadequate for the single factor model in water quality assessment. Further, combined with its advantages of Support vector machine (SVM) in solving small sample with, nonlinear and high dimension problems, an aggregative index number method in water quality assessment based on SVM is established. The test results show that the method proposed in this paper has a very high correct ratio of evaluation, and has a good engineering application prospect.