Grey Modeling Technologies Based On Three Kinds Of Special Sequences And Their Application

This paper selects mechanism of grey models based on special sequences as a main line, grey models based on special sequences as main researching objects, analysis of city innovation ability as actual background and different features of data as a starting point. This paper combines grey modeling technology to deal with data in order to get a more reliable result with a higher precision. The content is as follows:(1) Aiming at the problem that the precision of descending sequence based on GM(1,1) model is not high, this paper constructs GM(1,1) model based on inverse zeroing starting point operator and take precision into consideration using translation, proving that this method will not magnify reduction error. This model realizes simulating descending sequence in the same direction, eliminating the unreasonable model error when simulating descending sequence with GM(1,1) model. This paper shows the effectiveness of the model based on the consumption of fuel oil in the recent years in Jiangsu province.(2)Aiming at the problem that multiple factors may contain time-lag, this paper simulates the data in the time-lag value according to interpolation method, construct grey time-lag relation degree and determine time-lag coefficient based on maximum grey time-lag relation. And then this paper builds up GM(1,N) model including time-lag coefficient, increasing the ability people know the timelag system, extending the application of GM(1,N) model. This paper uses grey time-lag relation degree to determine the time-lag coefficient between the input of R&D funds and the high-tech industry output value from 2005 to 2012 in Jiangsu province, and then predict the high-tech industry output value according to the model, showing the effectiveness of the model.(3)Aiming at the problem that the clustering result of the variable weight grey clustering doesn’t correspond with the reality in current papers, this paper analyses the reason and constructs a new kind of whitenization weight function, proves that the new whitenization weight function satisfies normative and doesn’t exist triplet crossing, gives the calculation flow of the new method of grey variable weight clustering. At last, the paper takes the level of innovation ability of cities in 2012 in Jiangsu province, showing the effectiveness and operation of the method, providing useful revelation for economic development and improving competitiveness in the future in Jiangsu province.(4) In order to popularize grey theory, this paper makes a GUI based on MATLAB, including common models and models in this paper. People can analyze based on the GUI.