Based on analyzing the limitations of the commonly used back-propagation neural network （BPNN）, a wavelet neural network （WNN） is adopted as the nonlinear river channel flood forecasting method replacing the BPNN....Based on analyzing the limitations of the commonly used back-propagation neural network （BPNN）, a wavelet neural network （WNN） is adopted as the nonlinear river channel flood forecasting method replacing the BPNN. The WNN has the characteristics of fast convergence and improved capability of nonlinear approximation. For the purpose of adapting the timevarying characteristics of flood routing, the WNN is coupled with an AR real-time correction model. The AR model is utilized to calculate the forecast error. The coefficients of the AR real-time correction model are dynamically updated by an adaptive fading factor recursive least square（RLS） method. The application of the flood forecasting method in the cross section of Xijiang River at Gaoyao shows its effectiveness.展开更多
In this paper,an application of a nonlinear predictive controller based on a self recurrent wavelet network (SRWN) model for a direct internal reforming solid oxide fuel cell (DIR-SOFC) is presented. As operating temp...In this paper,an application of a nonlinear predictive controller based on a self recurrent wavelet network (SRWN) model for a direct internal reforming solid oxide fuel cell (DIR-SOFC) is presented. As operating temperature and fuel utilization are two important parameters,the SOFC is identified using an SRWN with inlet fuel flow rate,inlet air flow rate and current as inputs,and temperature and fuel utilization as outputs. To improve the operating performance of the DIR-SOFC and guarantee proper operating conditions,the nonlinear predictive control is implemented using the off-line trained and on-line modified SRWN model,to manipulate the inlet flow rates to keep the temperature and the fuel utilization at desired levels. Simulation results show satisfactory predictive accuracy of the SRWN model,and demonstrate the excellence of the SRWN-based predictive controller for the DIR-SOFC.展开更多
基金The National Natural Science Foundation of China(No.50479017).
文摘Based on analyzing the limitations of the commonly used back-propagation neural network （BPNN）, a wavelet neural network （WNN） is adopted as the nonlinear river channel flood forecasting method replacing the BPNN. The WNN has the characteristics of fast convergence and improved capability of nonlinear approximation. For the purpose of adapting the timevarying characteristics of flood routing, the WNN is coupled with an AR real-time correction model. The AR model is utilized to calculate the forecast error. The coefficients of the AR real-time correction model are dynamically updated by an adaptive fading factor recursive least square（RLS） method. The application of the flood forecasting method in the cross section of Xijiang River at Gaoyao shows its effectiveness.
基金supported by the National High-Tech Research and Devel-opment Program (863) of China (No. 2006AA05Z148)the Shanghai Municipal Natural Science Foundation, China (No. 08ZR1409800)
文摘In this paper,an application of a nonlinear predictive controller based on a self recurrent wavelet network (SRWN) model for a direct internal reforming solid oxide fuel cell (DIR-SOFC) is presented. As operating temperature and fuel utilization are two important parameters,the SOFC is identified using an SRWN with inlet fuel flow rate,inlet air flow rate and current as inputs,and temperature and fuel utilization as outputs. To improve the operating performance of the DIR-SOFC and guarantee proper operating conditions,the nonlinear predictive control is implemented using the off-line trained and on-line modified SRWN model,to manipulate the inlet flow rates to keep the temperature and the fuel utilization at desired levels. Simulation results show satisfactory predictive accuracy of the SRWN model,and demonstrate the excellence of the SRWN-based predictive controller for the DIR-SOFC.
