Fuel cell vehicles represent a groundbreaking form of new energy vehicles, garnering considerable attention and development in recent years. The energy management of fuel cell hybrid electric vehicles has always been a research difficulty. This paper aims to improve the economy and durability of fuel cell and lithium battery hybrid systems, and proposes an innovative online energy management strategy that incorporates road condition recognition. Key features from multiple driving datasets are extracted to construct a classifier based on ensemble learning and decision tree, which can effectively identify road patterns. Then, a fuzzy Q-learning method is proposed for online power allocation. Different action spaces of Q-learning are designed to cater to different road patterns, enabling more effective adaptation to diverse roadway environments. In addition, the reward function is designed to account for hydrogen costs and degradation expenses. Through a series of simulation and hardware tests, the proposed method has been demonstrated to effectively achieve online energy management of fuel cell/battery system under varying road conditions. And a quantitative comparison validates that the proposed method outperforms traditional reinforcement learning algorithms in reducing both fuel consumption and system degradation rates.

An LLC dual active bridge (DAB) bidirectional converter with reduced switch count as a dc–dc transformer (DCX) is proposed in this article. It can achieve voltage regulation, full load range zero voltage switching (ZVS) for eight main switches and wide load range ZVS for four auxiliary switches, with consideration of the switch junction capacitors under different power direction. A part portion of total power is transferred through an auxiliary DAB circuit to achieve voltage regulation and bidirectional power conversion. The main power is transferred through LLC circuit with an easy PWM scheme achieving high conversion efficiency. Different from the existing LLC -DAB DCX structure, the two secondary side transformer outputs are in parallel, and connected to the same full-bridge circuit. In this way, the auxiliary DAB circuit can not only adjust the output voltage, but also can inject current into the secondary side to help achieving ZVS of switches without synchronous rectifier circuit. The converter operation principle is analyzed under different power directions, and ZVS is analyzed with the consideration of all the switch junction capacitors. Based on the analysis, the design consideration is introduced. Finally, a 2-kW prototype with 750 V input and 400 V output is built up to verify the effectiveness of the proposed solution.

Input-series output-parallel (ISOP) connected Dual Active Bridge (DAB) converters allow the use for occasions requiring high input voltage and galvanic isolation with bidirectional power capability. However, when DAB converters are ISOP connected, the traditional analysis without consideration of the dead-time cannot effectively explain the reason why the ISOP converter system is stable under common phase shift control when parameters are mismatched. In this paper, it is firstly revealed that the stability of the ISOP DAB converter under common phase ratio control can be ensured by consideration of the dead-time. Based on the analysis of the DAB power transmission characteristics with consideration of dead time, a new model is derived for the ISOP DAB converter, which can reveal the reason why the converter can operate stably under common phase shift control when the design parameters of the individual module are mismatched. Then based on the derived model, the stability analysis and the parameters design are further performed. Finally, the effectiveness of the working principle and analysis are verified through results from a two-module ISOP experimental prototype.

Due to the disparity of power modules, asymmetry of driving pulses and measurement errors of sensors, dc currents may be injected to grid-connected photovoltaic (PV) inverters. The dc current injection may cause magnetic saturation of the power transformers. To solve this issue, this paper thus proposes an effective current control strategy and compensation method, which does not require any extra sensor and hardware circuit. Firstly, the root-cause of dc current injection is comprehensively analyzed. Subsequently, a proportional-integral-resonant (PIR) controller is proposed to eliminate the dc component caused by disparity of power modules, asymmetry of driving pulses and measurement errors of grid voltage. The injected dc current caused by grid current measurement error is estimated from the line-frequency ripple of the dc-link voltage and then it is suppressed by a feedback compensation controller. In addition, the dc current rejection capability is evaluated and the proposed method is benchmarked with the virtual capacitor-based method. Finally, experimental tests are performed on a 1.2-kW single-phase PV inverter to verify the effectiveness of the proposal.

Combining the auxiliary DAB converter and the bidirectional LLC converter operating at a fixed frequency in an input-series output-parallel (ISOP) structure can achieve bidirectional power regulation and high efficiency in DC-DC transformer (DCX) applications. However, for bidirectional DCX applications, the synchronous rectification (SR) operation becomes challenging since that the instantaneous-transferred power is difficult to obtain, and LLC-DCX will operate in discontinuous current mode (DCM) under light load conditions. In addition, the rectifier-bridge has to be swapped under different power flows. To achieve on-line SR modulation for LLC-DCX under bidirectional power application, a new solution to achieve accurate SR operation, based on instantaneous-transferred power estimation is proposed under full load range. The proposed method uses the added auxiliary DAB converter as a power detector to estimate the power value without input voltage sampling and any auxiliary current sensor or SR detection circuit. Besides, the mathematical derivation of power estimation, DCM operation conditions and the SR conduction period for the rectifier-bridge, are analyzed in depth. Finally, the effectiveness of the proposed method and the correctness of analysis are verified by the simulation and experimental results.