李德鹏等：一种基于鲁棒随机向量函数链接网络的磨矿粒度集成建模方法 .77。 [6]Wang X H,Gui W H.Wang Y L,et al.Prediction modeling for [13]Ditterrich TG.Machine learning research:four current diree- particle size of grinding circuit of mixture kemels SVM.Comput tion.Artif Intell Mag,1997,4:97 Eng Appl,2010,46(12):207 [14]Kearns MJ,Valiant LG.Learning Boolean Formulae or Finite (王新华，桂卫华，王雅琳，等.混合核函数支持向量机的磨 Automata is as Hard as Factoring.Cambridge:Harvard Universi- 矿粒度预测模型.计算机工程与应用.2010,46(12)：207) ty,Center for Research in Computing Technology,Aiken Compu- [7]Qiao J H,Chai T Y.Soft measurement model and its application tation Laboratory,1988 in raw meal calcination process.J Process Control,2012,22(1): [15]Schapire R E.The strength of weak leamability.Mach Learn, 344 1990,5(2):197 [8]Igelnik B,Pao Y H.Stochastic choice of basis functions in adap- [16]Schwenk H,Bengio Y.Boosting neural networks.Neural Com- tive function approximation and the functional-link net.IEEE pt,2000,12(8):1869 Trans Neural Nete,1995,6(6):1320 [17]Martinez-Munioz C.Suarez A.Out-of-bag estimation of the opti- [9]Huang G B,Chen Y Q,Babri HA.Classification ability of single mal sample size in bagging.Pattern Recognit,2010,43 (1): hidden layer feedforward neural networks.IEEE Trans Neural 143 Ne0,2000,11(3):799 [18]Dai W,Chai T Y,Yang S X.Data-driven optimization control [10]Zhang S Y,Bao Y P,Zhang C J,et al.Prediction model of alu- for safety operation of hematite grinding process.IEEE Trans Ind minum consumption with BP neural networks in IF steel produc. Electron,2015,62(5):2930 tion.Chin J Eng,2017,39(4):511 [19]Dai W,Liu Q.Chai T Y.Particle size estimate of grinding (张思源，包燕平，张超杰，等.BP神经网络F钢铝耗的预 processes using random vector functional link networks with im- 测模型.工程科学学报，2017,39(4)：511) proved robustness.Neurocomputing,2015,169:361 [11]Pao Y H,Phillips S M,Sobajic D J.Neural-net computing and [20]Rao C R.Generalized Inverse of Matrices and its Applications. the intelligent control of systems.Int J Control,1992.56(2): New York:Wiley,1971 263 [21]Wang D H,Alhamdoosh M.Evolutionary extreme learning ma- [12]Scardapane S,Wang D H.Randomness in neural networks:An chine ensembles with size control.Neurocomputing,2013.102: overview.WIREs Data Min Knowl Dise,2017,7(2):el200 98李德鹏等: 一种基于鲁棒随机向量函数链接网络的磨矿粒度集成建模方法 [6] Wang X H, Gui W H, Wang Y L, et al. Prediction modeling for particle size of grinding circuit of mixture kernels SVM. Comput Eng Appl, 2010, 46(12): 207 (王新华, 桂卫华, 王雅琳, 等. 混合核函数支持向量机的磨 矿粒度预测模型. 计算机工程与应用, 2010, 46(12): 207) [7] Qiao J H, Chai T Y. Soft measurement model and its application in raw meal calcination process. J Process Control, 2012, 22(1): 344 [8] Igelnik B, Pao Y H. Stochastic choice of basis functions in adap鄄 tive function approximation and the functional鄄link net. IEEE Trans Neural Netw, 1995, 6(6): 1320 [9] Huang G B, Chen Y Q, Babri H A. Classification ability of single hidden layer feedforward neural networks. IEEE Trans Neural Netw, 2000, 11(3): 799 [10] Zhang S Y, Bao Y P, Zhang C J, et al. Prediction model of alu鄄 minum consumption with BP neural networks in IF steel produc鄄 tion. Chin J Eng, 2017, 39(4): 511 (张思源, 包燕平, 张超杰, 等. BP 神经网络 IF 钢铝耗的预 测模型. 工程科学学报, 2017, 39(4): 511) [11] Pao Y H, Phillips S M, Sobajic D J. Neural鄄net computing and the intelligent control of systems. Int J Control, 1992, 56(2): 263 [12] Scardapane S, Wang D H. Randomness in neural networks: An overview. WIREs Data Min Knowl Disc, 2017, 7(2): e1200 [13] Ditterrich T G. Machine learning research: four current direc鄄 tion. Artif Intell Mag, 1997, 4: 97 [14] Kearns M J, Valiant L G. Learning Boolean Formulae or Finite Automata is as Hard as Factoring. Cambridge: Harvard Universi鄄 ty, Center for Research in Computing Technology, Aiken Compu鄄 tation Laboratory, 1988 [15] Schapire R E. The strength of weak learnability. Mach Learn, 1990, 5(2): 197 [16] Schwenk H, Bengio Y. Boosting neural networks. Neural Com鄄 put, 2000, 12(8): 1869 [17] Mart侏nez鄄Mu觡oz G, Su觃rez A. Out鄄of鄄bag estimation of the opti鄄 mal sample size in bagging. Pattern Recognit, 2010, 43 ( 1 ): 143 [18] Dai W, Chai T Y, Yang S X. Data鄄driven optimization control for safety operation of hematite grinding process. IEEE Trans Ind Electron, 2015, 62(5): 2930 [19] Dai W, Liu Q, Chai T Y. Particle size estimate of grinding processes using random vector functional link networks with im鄄 proved robustness. Neurocomputing, 2015, 169: 361 [20 ] Rao C R. Generalized Inverse of Matrices and its Applications. New York: Wiley, 1971 [21] Wang D H, Alhamdoosh M. Evolutionary extreme learning ma鄄 chine ensembles with size control. Neurocomputing, 2013, 102: 98 ·77·