The significant advance of power electronics in today’s market is calling for high-performance power conversion systems and MEMS devices that can operate reliably in harsh environments, such as high working temperature. Silicon-carbide (SiC) power electronic devices are featured by the high junction temperature, low power losses, and excellent thermal stability, and thus are attractive to converters and MEMS devices applied in a high-temperature environment. This paper conducts an overview of high-temperature power electronics, with a focus on high-temperature converters and MEMS devices. The critical components, namely SiC power devices and modules, gate drives, and passive components, are introduced and comparatively analyzed regarding composition material, physical structure, and packaging technology. Then, the research and development directions of SiC-based high-temperature converters in the fields of motor drives, rectifier units, DC–DC converters are discussed, as well as MEMS devices. Finally, the existing technical challenges facing high-temperature power electronics are identified, including gate drives, current measurement, parameters matching between each component, and packaging technology.

Photovoltaic (PV) output is susceptible to meteorological factors, resulting in intermittency and randomness of power generation. Accurate prediction of PV power output can not only reduce the im-pact of PV power generation on the grid but also provide a reference for grid dispatching. Therefore, this paper proposes an LSTM-attention-embedding model based on Bayesian optimization to predict the day-ahead PV power output. The statistical features at multiple time scales, combined features, time features and wind speed categorical features are explored for PV related meteorological factors. A deep learning model is constructed based on an LSTM block and an embedding block with the connection of a merge layer. The LSTM block is used to memorize and attend the historical information, and the embedding block is used to encode the categorical features. Then, an output block is used to output the prediction results, and a residual connection is also included in the model to mitigate the gradient transfer. Bayesian optimization is used to select the optimal combined features. The effectiveness of the proposed model is verified on two actual PV power plants in one area of China. The comparative experimental results show that the performance of the proposed model has been significantly improved compared to LSTM neural networks, BPNN, SVR model and persistence model.

This paper proposes a cost-effective way to design and operate fuel cell hybrid electric trucks (FCHETs) where a chance-constrained optimization is formulated. The aim of the introduced problem is to minimize a summation of component cost and operational cost with consideration of fuel cell (FC) degradation and cycle life of energy buffer. We propose to decompose the problem into two sub-problems that are solved by sequential convex programming. The delivered power satisfies a cumulative distribution function of the wheel power demand, while the truck can still traverse driving cycles with a similar speed and travel time without delivering unnecessarily high power. This allows to downsize powertrain components, including electric machine, FC and energy buffer. A case study considering different energy buffer technologies, including supercapacitor (SC), lithium-ion battery (LiB), and lithium-ion capacitor (LiC) is investigated in a set of trucking applications, i.e. urban delivery, regional delivery, construction, and long-haul. Results show that the power rating of the electric machine is drastically reduced when the delivered power is satisfied in a probabilistic sense. Moreover, the configuration with LiB as the energy buffer has the lowest expense but the truck with LiC can carry more payload.

This paper deals with the conception and the achievement of a hybrid power source using a fuel cell combined with a battery or a supercapacitor. In which, the fuel cell supplies the main power to the drive system; while the battery or the supercapacitor is used as an auxiliary power source. This gives the benefit that the regenerative energy is stored in battery or supercapacitor during the deceleration and it is transferred back to the drive system during the acceleration when compared to electric vehicles solely powered by a fuel cell. Different energy storage devices, such as fuel cell, battery, and supercapacitor are compared., and then several structures of fuel cell-based electric vehicles are analyzed in the paper. Following that a conventional topology based on fuel cell and battery using a DC/DC converter with the connection between the fuel cell and the inverter, and a floating voltage topology powered by fuel cell and supercapacitor without any DC/DC converters are chosen for the simulation analysis. Simulation results show that power variations of the fuel cell in floating voltage topology can be smoother, and its rated power is downsized, which can extend the fuel cell lifetime and take full advantages of the fuel cell and the supercapacitor.