In this paper,a stand-alone photovoltaic(PV)system based on a Double Ended Forward Converter(DEFC)is presented.The proposed converter is specified for 48 V,100Wapplications as most of the equipment used in telecommuni...In this paper,a stand-alone photovoltaic(PV)system based on a Double Ended Forward Converter(DEFC)is presented.The proposed converter is specified for 48 V,100Wapplications as most of the equipment used in telecommunication and aircraft fall in this range.The literature has limited potential application of DEFCin PV systems.The research work deals with an in-depth study of DEFCand proposes an improvedDEFCfor PV applications with battery backup.Besides,a bi-directional dc-dc converter for the battery is integrated to track theMaximumPower Point(MPP)of the PV generator.The converter is examined under variable irradiance and load conditions,and the analytical analysis of boundary conditions are implemented.The converter’s architecture also ensures built-in I-V curve tracing for the identification of MPP of PV generator.It offers low voltage stresses across switches and avoids sinking power supply and core resetting circuits.The topology’s behavior is analyzed based onMPP achievement and maintaining output under different conditions of battery backup availability,environmental,and load conditions.The PV system architecture is designed and analyzed theoretically and verified with simulations on the PSIM software.展开更多
Double parallel forward neural network （DPFNN） model is a mixture structure of single-layer perception and single-hidden-layer forward neural network （SLFN）. In this paper, by making use of the idea of online sequ...Double parallel forward neural network （DPFNN） model is a mixture structure of single-layer perception and single-hidden-layer forward neural network （SLFN）. In this paper, by making use of the idea of online sequential extreme learning machine （OS-ELM） on DPFNN, we derive the online sequential double parallel extreme learning machine algorithm （OS-DPELM）. Compared to other similar algorithms, our algorithms can achieve approximate learning performance with fewer numbers of hidden units, as well as the parameters to be determined. The experimental results show that the proposed algorithm has good generalization performance for real world classification problems, and thus can be a necessary and beneficial complement to OS-ELM.展开更多
文摘In this paper,a stand-alone photovoltaic(PV)system based on a Double Ended Forward Converter(DEFC)is presented.The proposed converter is specified for 48 V,100Wapplications as most of the equipment used in telecommunication and aircraft fall in this range.The literature has limited potential application of DEFCin PV systems.The research work deals with an in-depth study of DEFCand proposes an improvedDEFCfor PV applications with battery backup.Besides,a bi-directional dc-dc converter for the battery is integrated to track theMaximumPower Point(MPP)of the PV generator.The converter is examined under variable irradiance and load conditions,and the analytical analysis of boundary conditions are implemented.The converter’s architecture also ensures built-in I-V curve tracing for the identification of MPP of PV generator.It offers low voltage stresses across switches and avoids sinking power supply and core resetting circuits.The topology’s behavior is analyzed based onMPP achievement and maintaining output under different conditions of battery backup availability,environmental,and load conditions.The PV system architecture is designed and analyzed theoretically and verified with simulations on the PSIM software.
基金Supported by the National Natural Science Foundation of China(Grant Nos.1140107661473328+1 种基金1117136761473059)
文摘Double parallel forward neural network （DPFNN） model is a mixture structure of single-layer perception and single-hidden-layer forward neural network （SLFN）. In this paper, by making use of the idea of online sequential extreme learning machine （OS-ELM） on DPFNN, we derive the online sequential double parallel extreme learning machine algorithm （OS-DPELM）. Compared to other similar algorithms, our algorithms can achieve approximate learning performance with fewer numbers of hidden units, as well as the parameters to be determined. The experimental results show that the proposed algorithm has good generalization performance for real world classification problems, and thus can be a necessary and beneficial complement to OS-ELM.
