A revised support vector regression (SVR) ensemble model based on boosting algorithm (SVR-Boosting) is presented in this paper for electricity price forecasting in electric power market. In the light of characteristic...A revised support vector regression (SVR) ensemble model based on boosting algorithm (SVR-Boosting) is presented in this paper for electricity price forecasting in electric power market. In the light of characteristics of electricity price sequence, a new triangular-shaped 为oss function is constructed in the training of the forecasting model to inhibit the learning from abnormal data in electricity price sequence. The results from actual data indicate that, compared with the single support vector regression model, the proposed SVR-Boosting ensemble model is able to enhance the stability of the model output remarkably, acquire higher predicting accuracy, and possess comparatively satisfactory generalization capability.展开更多
A predictive current control algorithm for the Buck-Boost DC-DC converter is presented in this paper. The continuous time model of the system is properly introduced, then, by imposing a proper PWM modulation pattern, ...A predictive current control algorithm for the Buck-Boost DC-DC converter is presented in this paper. The continuous time model of the system is properly introduced, then, by imposing a proper PWM modulation pattern, its discrete time model is achieved. This last one is successfully employed in determining the steady state locus of the Buck-Boost converter, both in CCM （continuous conduction mode） and DCM （discontinuous conduction mode）. A novel continuous time equivalent circuit of the converter is introduced too, with the aim of determining a ripple free representation of the state variables of the system, over both transient and steady state operation. Then, a predictive current control algorithm, suitable in both CCM and DCM, is developed and properly checked by means of computer simulations. The corresponding results have highlighted the effectiveness of the proposed modelling and of the predictive control algorithm, both in CCM and DCM.展开更多
基金Sponsored by the National Outstanding Young Investigator Grant (Grant No6970025)the Key Project of National Natural Science Foundation (GrantNo59937150)+2 种基金863 High Tech Development Plan (Grant No2001AA413910)of China and the Key Project of National Natural Science Foundation(Grant No59937150)the Project of National Natural Science Foundation (Grant No60274054)
文摘A revised support vector regression (SVR) ensemble model based on boosting algorithm (SVR-Boosting) is presented in this paper for electricity price forecasting in electric power market. In the light of characteristics of electricity price sequence, a new triangular-shaped 为oss function is constructed in the training of the forecasting model to inhibit the learning from abnormal data in electricity price sequence. The results from actual data indicate that, compared with the single support vector regression model, the proposed SVR-Boosting ensemble model is able to enhance the stability of the model output remarkably, acquire higher predicting accuracy, and possess comparatively satisfactory generalization capability.
文摘A predictive current control algorithm for the Buck-Boost DC-DC converter is presented in this paper. The continuous time model of the system is properly introduced, then, by imposing a proper PWM modulation pattern, its discrete time model is achieved. This last one is successfully employed in determining the steady state locus of the Buck-Boost converter, both in CCM （continuous conduction mode） and DCM （discontinuous conduction mode）. A novel continuous time equivalent circuit of the converter is introduced too, with the aim of determining a ripple free representation of the state variables of the system, over both transient and steady state operation. Then, a predictive current control algorithm, suitable in both CCM and DCM, is developed and properly checked by means of computer simulations. The corresponding results have highlighted the effectiveness of the proposed modelling and of the predictive control algorithm, both in CCM and DCM.
