The results show that the prediction model,which is established by the method,has high fitting accuracy on historical samples and good testing effect as well,and can be applied to forecast operation.

An empirical Model is built concerning the overflow yield and all the structure and operating parameters employing nonlinear regression. This model is proved to have a fairly high fitting accuracy and can be used in predicting theoverflow yield of cyclone with a relatively high accuracy.

Adaptability analysis shows that the model has high fitting accuracy and good adaptability. Application example is also given to show the advantages of the model and the method.

The real example of the sports result forecasting shows based on grey correlation variable weight combined forecasting has high fitting accuracy than ordinary single model.

First,a group of initial fuzzy rules are produced using self organizing algorithm to cluster learning data. Then error back propagation algorithm is introduced to adjust the network parameters. The simulation results demonstrate that the fuzzy neural network has the advantages of simple structure and the high fitting accuracy.

The function includes five special examples (Gaussian-Cauchy function etc.), satisfies eight requests in application, such as very high fitting accuracy, better flexibility and generality etc.

Two associated solution models, ASMA and ASMB were Presented in this paper. Both models can be applied not only to those solutions with both components exhibiting self-association, but also to partially miscible systems and alcohol aqueous solutions. Their applicability can be also extended to multicomponent systems. 13 binary isobaric VLE data (including partially miscible systems) were fitted by these two models. Their fitting accuracies were compared with those obtained by applying three local composition...

Two associated solution models, ASMA and ASMB were Presented in this paper. Both models can be applied not only to those solutions with both components exhibiting self-association, but also to partially miscible systems and alcohol aqueous solutions. Their applicability can be also extended to multicomponent systems. 13 binary isobaric VLE data (including partially miscible systems) were fitted by these two models. Their fitting accuracies were compared with those obtained by applying three local composition equations (wilson,FHW and UNIQUAC). It has been proved that ASMA model shows the highest fitting accuracy.

A process for filtering the polar coordinates is proposed in this article by combining both the traditional processes, namely the average of figures and the harmonic analysis. It allows us to separate the three principal components (i. e. those of long term, Chandierian period and annual period)of the polar coordinates with a shorter series of data and a higher fitting accuracy. The procedure is as follows: Let the values of the polar coordinates at epoches t (t=0, 1, …, 22, in unit of 0.05 yr. )be D_1,...

A process for filtering the polar coordinates is proposed in this article by combining both the traditional processes, namely the average of figures and the harmonic analysis. It allows us to separate the three principal components (i. e. those of long term, Chandierian period and annual period)of the polar coordinates with a shorter series of data and a higher fitting accuracy. The procedure is as follows: Let the values of the polar coordinates at epoches t (t=0, 1, …, 22, in unit of 0.05 yr. )be D_1, make sums as the equations (4) and (5). The secular term S can be found from the equations (7) or(12), then J_1 and L_1 from the equations (8) and(13). Having deduced the equations(10) and(15)by the least square mzthod, the amplitude and the phase of Chand-lerian component G, P_c and those of annual component A, P_a are obtained respectively. Furthermore, a discussion is given on the amelioration of smoothness of the secular trend of each components by extending the datum-period, and a relative criterion is set up for estimation and comparison of the smoothness.

A process for filtering the polar coordinates is proposed in this article by combining both the traditional processes, namely the average of figures and the harmonic analysis. It allows us to separate the three principal components (i. e. those of long term, Chandlerian period and annual period) of the polar coordinates with a shorter series of data and a higher fitting accuracy. The procedure is as follows: Let the values of the poiar coordinates at epoches t (t=o, 1, …, 22, in unit of 0.05 yr.) be D_(?),...

A process for filtering the polar coordinates is proposed in this article by combining both the traditional processes, namely the average of figures and the harmonic analysis. It allows us to separate the three principal components (i. e. those of long term, Chandlerian period and annual period) of the polar coordinates with a shorter series of data and a higher fitting accuracy. The procedure is as follows: Let the values of the poiar coordinates at epoches t (t=o, 1, …, 22, in unit of 0.05 yr.) be D_(?), make sums as the equations (4) and (5). The secular term S can be found from the equations (7) or (12), then J_t and L_(?) from the equations (8) and (13). Having deduced the equations (10) and (15) by the least square method, the amplitude and the phase of Chandlerian component C, P_c and those of annual component A, P_c are obtained respectively. Furthermore, a discussion is given on the amelioration of smoothness of the secular trend of each components by extending the datum-period, and a relative criterion is set up for estimation and comparison of the smoothness.