The intelligent optimization of GM(1,1) power model and its application in the forecast of traffic accident

In order to improve prediction accuracy of traffic accidents, an optimization GM (1, 1) power model is put forward based on the characteristics of S-type and a single traffic accident peak. Specifically, an intelligent optimization method is proposed to optimist the parameter α using PSO algorithm and the optimization formula of background value in GM (1, 1) power model is given based on the integral mean value theorem as well as the intelligent optimization algorithm of background value is proposed. The number of traffic fatalities during the period 1996-2009 is fitted and predicted using the optimization GM (1, 1) power model, whose results show that the optimization GM (1, 1) power model has higher fitting and prediction accuracy than the grey Verhaulst model.