Whether or not to compensate the oil pressure because of lift crane velocity reasonably, i.e., the velocity compensation was thought to be the key to obtain accurate dynamic weighing about loaders. After the method of dynamic weighing was given, the parameter inferring course of least square support vector machines (LS-SVM) within Bayesian evidence framework was introduced. Then the frame model of velocity compensation based on LS-SVM was given, and the means of velocity compensation and the course of weight computing were introduced in detail. Test results indicate that using LS-SVM within Bayesian evidence framework for solving velocity compensation, a relative measuring error within 1% can be obtained, which verifies that the validity of the method.
Vapnik V N. The nature of statistical learning theory[M]. New York: Springer-Verlag, 1995
Suykens J A K. Nonlinear modeling and support vector machines Conference Record-IEEE Instrumentation and Measurement Technology Conference.Piscataway,NJ:IEEE,2001,1:287-294
Kowk J T. The evidence framework applied to support vector machines[J].IEEE Transactions on Neural Network.2000,11(5):1162-1173
Mackay D J C. Probable networks and plausible predictions-A review of practical Bayesian methods for supervised neural networks[J].Network Computation in Neural Systems.1995,6:469-505
Gestel T V, Suykens J A K, Baestaens D E, et al. Financial time series prediction using least squares support vector machines within the evidence framework [J].IEEE Transactions on Neural Network.2001,12(4):809-820
Van Gestel T, Espinoza M, Baesens B, et al. A Bayesian nonlinear support vector machine error correction model[J].Journal of Forecasting.2006,25(2):77-100
��ΰ,������. װ�ػ����ض�̬��������ѧ������ʵ�ַ���[J].�й���е����,2006,11(22): 2333-2337 Wang Wei, Wang Tianmiao. Kinetics analysis and method of dynamic weighing about wheel loader[J]. China Mechanical Engineering, 2006, 11(22):2333-2337 (in Chinese)