自適應共振理論發展至今，陸續有許多版本被提出，其中的Digital ART版本(DART)改良了之前ART的缺點，DART 使用距離為基礎的不相似標準，能同時處理連續及整數變數，接受動態關聯的輸入；然而DART祗能用於分類，不能用於建立模式。因此我們提出兩個以DART為基礎的類神經網路模型 (DART + RBFN及DART + GRNN) ，能從大量的歷史數據中利用DART找出具代表性的數據點為訓練樣本集，及測試樣本集，再以RBFN或GRNN建立模型
經由一些簡單數學題目包括一維(Single Input Single output, SISO)的兩個函數(高斯曲線函數、三角曲線函數)及二維(Multi Input Single Output, MISO)的兩個函數(Himmeblau function、Peaks function)來驗証我們的模型可行性後，發現我們提出的模型中的DART+RBFN的效果較DART+GRNN來得好。因此將DART + RBFN應用至建立一真實PE製程的MI預測模型，我們發現此一模型可以關聯不同操作條件的穩態之MI 、也可以預測過渡狀態之MI變化，以及預測不同時段相同操作條件之MI。

Many versions of Adaptive Resonance Theory (ART) have been developed. One of these is Distance based ART (DART), which can deal with both continuous and integer inputs, employs a dissimilarity based vigilance measure, and accept dynamically correlated data. However, DART performs only the clustering step. To include a model building step, we proposed two neural networks based on DART --- DART+RBFN and DART+GRNN. Using DART, a representative training set and a test set can be mined from a large set of data. These data can be used to build a radial basis function network (RBFN) or a generalized regression network (GRNN).
DART+RBFN and DART+GRNN are tested using two SISO functions --- Gaussian curve function and trigonometric curve function, and two MISO functions --- Himmeblau function and Peaks function. We found that our method works well but DART+RBFN is better than DART+GRNN due to its superior modeling ability. Therefore DART+RBFN model was applied to construct an empirical model for Melt Index (MI) of a PE plant. We found that the model can be used to correlate and predict MI different steady operations as well as the changes of MI when there are grade transitions.

[1] McCulloch W. S. and Pitts W.,”A logical calculus of the ideas immanent in nervous activity,” Bulletin of Mathematical Biophysics, Vol. 5, pp 115-133 (1943)
[2] Broomhead D. S. and Lowe D.,”Multivariable Functional Interpolation and Adaptive Networks,” Complex Systems, Vol. 2, pp 321-355 (1988)
[3] Park J., and Sandberg I. W., “Universal Approximation Using Radial Basis Function Network,” Neural Computation, Vol. 3, pp 246-257 (1991)
[4] Park J., and Sandberg I. W., “Approximation and Radial Basis Function Network,” Neural Computation, Vol. 5, 305-316 (1993)
[5] Karayiannis N. B., and Mi G. W., “Growing Radial Basis Neural Networks: Merging Supervised and Unsupervised Learning with Network Growth Techniques,” IEEE Transactions on Neural Networks, Vol. 8(6), pp 1492-1506 (1997)
[6] Specht D. F., “A General Regression Neural Network,” IEEE Transactions on Neural Networks, Vol. 2, pp 568 -576 (1991)
[7] Pazen E., “On estimation of a probability density function and mode,” Ann. Math. Statist. , Vol. 33, pp 1065-1076 (1962)
[8] Carpenter, G. A. and Grossberg S., “A Massively Parallel Architecture for A Self-organizing Neural Pattern Recognition Machine,” Computer Vis. Graphics Image Process. 37, 54 (1987)
[9] Carpenter, G. A. and Grossberg S., “ART2: Self-organization of Stable Category Recognition Codes for Analog Input Patterns,” Appl. Opt. 26, 4919 (1987)
[10] Carpenter, G. A. and Grossberg S., “ART3: Hierarchical Search Using Chemical Transmitters in Self-Organizing Pattern Recognition Architectures,” Neural Networks, Vol. 3, pp 129-152, (1990)
[11] Carpenter, G. A., Grossberg S., and Rosen D. B., “Fuzzy ART: Fast Stable Learning and Categorization of Analog Patterns by an Adaptive Resonance System, ” Neural Networks, Vol. 4, pp 759-771 (1991)
[12] 陳丁碩, “自適應共振理論在製程失誤偵測與診斷之應用,” 國立清華大學博士論文，中華民國八十九年
[13] Chen S., Cowan C. F. N., Grant P. M., “Orthogonal Least Squares Learning Algorithm for Radial Basis Function Networks,” IEEE Transaction on Neural Networks. Vol. 2, pp 302-309 (1991)
[14] Berthold M. R. and Diamond J., “Boosting the performance of RBF networks with dynamic decay adjustment,” Advances in Neural Inform. Processing Syst. 3, Lippmann R. P. et al., Eds. San Mateo, CA: Morgan Kaufmann (1991)
[15] Yu S. and Guan L., “Feature Selection Using General Regression Neural Networks for the Automatic Detection of Clustered Microcalcifications,” Proceedings of the 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing, Mar 15-Mar 19 1999 v 2 Phoenix, AZ, USA
[16] Patton J. B., “Brushless DC Motor Control Using a General Regression Neural Network,” Proceedings of the 1995 IEEE 21st International Conference on Industrial Electronics, Control, and Instrumentation. Part 2 (of 2) Nov 6-10 1995 v2 1995 Orlando, FL, USA
[17] Chen J., Wong D. S. H., Jang S. S., Yang S. L., “Product and Process Development Using Artificial Neural-Network Model and Information Analysis,” AICHE J., Vol. 44, pp 876-887 (1998)
[18] Rodriguez Ferdinand, “Principles of Polymer Systems,” Taylor and Francis Washington, D.C (1996)
[19] McAuley K. B., MacGregor J. F., “On-line inference of polymer properties in an industrial polyethylene reactor,” AIChE J. Vol. 37, pp 825-835, (1991)
[20] Ohshima M., Tanigaki M., “ Quality control polymer process,” J. of Process Control, Vol. 10, pp.135-148 (2000)