The cross experiments indicated that the average matching relative error and the average prediction relative error of this model, was 0.0063% and 0.1210% respectively. The model was applied to predict the concentration of CO_2 in the exit of purifying column.

If the fixed-length pattern was taken as 10 amino acids,the average prediction accuracies of the"1041type" were 83.1% and 79.8% by the 3-cross validation test and jack-knife test,respectively.

The compositions of amino acids and twin amino acids were chosen as the information parameters of the amino acids sequences. The prediction results show that average prediction accuracies of the"822type"for fixed-length pattern with 8 amino acids were 78.1% and 76.7% by the 3-cross validation test and jack-knife test,respectively.

By using MIMO to predict a real example of case the result shows,the error of average prediction is small than 0.37%,the accurate degree raised 2.91 times than moving average method,raised 4.28 times than other analysis method.

Finally,the integrated prediction was taken based on the above two prediction results,whose weights were calculated by its log-logistic probability density and average prediction error was reduced to 21.20%.

An integrated improvement on the force model has been made in the present investigation, including model structure, measuring methods and sub-routine of three most important variables-flow stress, contact length and friction coefficient. A high accuracy force model for cold rolling is constructed with average prediction deviation <5% before ANN treatment.

The experimental results show that the mapping between the colorimetric space and the recipe space can be realized by the multi-layer BP neural networks, and the average prediction error for 64 training samples is less than 1 unit of CIELAB color difference.

By the means of the extra-smoothing average prediction, the dissertation attempts topredict the tendency of climatechanges in the coming 5 or 9 years from 1995. By applyingregression analysis and numericalvalue simulation in Excel and SPSS, the author sums up thefluctuation tendency of the grainyield in Yunnan Province in the past 45 years.

The average prediction accuracies for "non-coding" promoters and "coding" promoters are 86.7% and 82.4%,respectively. The results indicate that our algorithm outperforms most of the existing approaches based on several performance measurements.

Numerical experiments are carried out for day-to-day prediction and temporal average prediction.

Average prediction errors for serum trough concentrations were 48.3% for neural networks and 59.0% for NONMEM.

While inferred Bayesian trees demonstrate low average prediction error rates, there is reason to believe that error rates will be higher for those leaves with few training examples.

Both the method and the continuous data set were tested and tuned to obtain the minimum of a normalized average prediction error (E) during the last millennium using several past millennia as a training data set.

With a genetic algorithm it was possible to construct hypothetical subsite maps (with inhibition constants) that gave further improvements in the average prediction for all saccharides.

The self-tuning technology is a new branch of the mordern control theory. In this paper we try to present the application of the self tuning prediction approach to the power system load forecasting. After a short introduction of its basic principle, a new power system load model is proposed in the paper, as it is the key-point of the forecasting problem. Simulation with a set of real power system ioad data shows that the average prediction error is about 3.24%. It also demonstrates that the approach has...

The self-tuning technology is a new branch of the mordern control theory. In this paper we try to present the application of the self tuning prediction approach to the power system load forecasting. After a short introduction of its basic principle, a new power system load model is proposed in the paper, as it is the key-point of the forecasting problem. Simulation with a set of real power system ioad data shows that the average prediction error is about 3.24%. It also demonstrates that the approach has a number of advantages such as simple algorithm, good convergence, etc.. It is considered that the approach is applicable while further work is needed for its practical application.

The correlations of power consumption (P_g) in aerated agitated vessels were analyzed. It was shown that the aeration number N_q, the most prominent factor of determining P_g, should be amended with the gas recirculation coefficient α,otherwise a prediction error more than 40% would be caused before large cavities were formed. A correlation was obtained for predicting P_g,which was more reliable and could be used under more conditions: For more than 150 data under kinds of conditions from threeliteratures,...

The correlations of power consumption (P_g) in aerated agitated vessels were analyzed. It was shown that the aeration number N_q, the most prominent factor of determining P_g, should be amended with the gas recirculation coefficient α,otherwise a prediction error more than 40% would be caused before large cavities were formed. A correlation was obtained for predicting P_g,which was more reliable and could be used under more conditions: For more than 150 data under kinds of conditions from threeliteratures, the average prediction error from the above correlation was only 9.5%, and the largest error was less than 20%.

The performance of codebooks designed for LSP with GLA and stochastic relaxation(SR) algorithm is investigated, and the features of two algorithms are discussed. The first order moving average prediction residual vectors of the mean removing LSP is used as the training sequence. The SR algorithm applied is a simplified SR algorithm for the decoder perturbation. The traditional conclusion is that SR saves GLA 1 bit per vector index. The experiment shows that the conclusion is right for a small amount...

The performance of codebooks designed for LSP with GLA and stochastic relaxation(SR) algorithm is investigated, and the features of two algorithms are discussed. The first order moving average prediction residual vectors of the mean removing LSP is used as the training sequence. The SR algorithm applied is a simplified SR algorithm for the decoder perturbation. The traditional conclusion is that SR saves GLA 1 bit per vector index. The experiment shows that the conclusion is right for a small amount of training sequence, but the difference in their performances is slight for a large amount of training sequence. The SNR improvement of codebook designed with SR algorithm is little when a LSP quantizer is designed.