In order to coordinate multiple power sources appropriately, and avoid the transients and rapid changes of power demand, a power management strategy based on an adaptive droop control, which is combined with a multimode strategy and an equivalent consumption minimization strategy, is proposed for a large-scale and high-power hybrid tramway. According to the hybrid system model of tramway developed with commercially available devices, the proposed power management strategy is evaluated with a real driving cycle of tramway. The results obtained from RT-LAB platform testify that the proposed strategy is able to manage and coordinate multiple power sources based on their natural characteristics, maintain dc bus voltage stabilization, improve the efficiency of overall tramway, and alleviate the stress on the hybrid power system.

For low energy consumption and operational stability, Multimode Equivalent Energy Consumption, an optimal energy management and control method, is proposed in this paper. This method effectively solves the problem of energy distribution between the primary power — a high-power fuel cell and an auxiliary power — a supercapacitor, in complex operating conditions. In all four states of the hybrid electric tram: traction, coasting, braking and station parking, the optimal energy distribution method, with the highest energy efficiency of the tram, is obtained through the whole system modeling and the equivalent energy consumption calculation. In order to verify the responding speed, this method gets the analytical solution and is tested on the RT-LAB hardware-in-the-loop simulation platform. The results show that compared to the Power Following Method, the new method can reduce the overall tram energy consumption by 2.5% in complex operating conditions. What's more, when this method is successfully applied to the world's first commercial fuel cell/supercapacitor LF-LRV hybrid tram, its energy consumption and fuel cell system (FCS) output fluctuation rate are effectively reduced according to the operational data, which indicates a good chance of extending the life of the fuel cell system as well.

Aimed at the input-independence and output-parallel dual-active-bridge (DAB) dc-dc converters applied in power electronic traction transformer (PETT), this paper proposes a simple power estimation scheme with triple phase-shift control (PES-TPS) to achieve the transmission power balance, dynamic performance improvement, and efficiency optimization, simultaneously. Through executing current stress optimization and introducing power estimation based on operational parameters, the efficiency of converters can be further improved and the power balance among DAB cells can be achieved over the whole power range without adding extra sensors to sample the output/inductor current of each cell. In addition, the dynamic performance of the adopted converters can be improved by compensating power loss online. Especially, dynamic response of the output voltage can be enhanced under input voltage and load fluctuation conditions. Finally, a comprehensive experimental comparison of current stress optimization with dual phase-shift (CSO-DPS), current stress optimization with triple phase-shift (CSO-TPS), and the proposed PES-TPS control is conducted in the DAB converters experimental prototype with three cells. The experimental results have confirmed the excellent performance of the proposed PES-TPS control and the correctness of theoretical analysis.

This paper studies bipartite tracking problem of nonlinear multiagent systems over signed directed graphs. Each following agent is modeled by a higher-order nonlinear system in strict-feedback form with unknown dynamics and hysteresis input uncertainty. Both distributed state feedback and output feedback control laws are proposed to achieve bipartite tracking confined by the prescribed performance bounds. The proposed approximation-free distributed controllers only utilize error variables incorporating with performance bound functions, which lead to a low-complexity control algorithm. Moreover, the proposed control laws guarantee that all signals of the closed-loop system are uniformly ultimately bounded.

Fault diagnosis of high-speed train (HST) bogie is essential in guaranteeing the normal daily operation of HST. In prior works, feature extraction from multi-sensor vibration signals mainly relies on signal processing methods, which is independent of the classification process. Based on convolutional neural networks (CNNs), this paper presents a novel fault diagnosis system using the residual-squeeze net (RSNet), which is directly applicable to raw data (time sequences) and does not require any signal transformation or post-processing. In this network, information fusion is achieved by using the convolutional layer. More specifically, via the squeeze operation, an optimal combination of channels is learnt by training the network. Experimental results obtained by using SIMPACK simulation data demonstrate the effectiveness of the proposed approach in both complete failure case and single failure case, with diagnosis accuracy near 100%. The proposed approach also shows good performance in identifying the locations of faulty components. Comparisons between RSNet and competitive methods shows the advantages of RSNet for fault classification.

As an important part of high-speed train (HST), the mechanical performance of bogies imposes a direct impact on the safety and reliability of HST. It is a fact that, regardless of the potential mechanical performance degradation status, most existing fault diagnosis methods focus only on the identification of bogie fault types. However, for application scenarios such as auxiliary maintenance, identifying the performance degradation of bogie is critical in determining a particular maintenance strategy. In this article, by considering the intrinsic link between fault type and performance degradation of bogie, a novel multiple convolutional recurrent neural network (M-CRNN) that consists of two CRNN frameworks is proposed for simultaneous diagnosis of fault type and performance degradation state. Specifically, the CRNN framework 1 is designed to detect the fault types of the bogie. Meanwhile, CRNN framework 2, which is formed by CRNN Framework 1 and an RNN module, is adopted to further extract the features of fault performance degradation. It is worth highlighting that M-CRNN extends the structure of traditional neural networks and makes full use of the temporal correlation of performance degradation and model fault types. The effectiveness of the proposed M-CRNN algorithm is tested via the HST model CRH380A at different running speeds, including 160, 200, and 220 km/h. The overall accuracy of M-CRNN, i.e., the product of the accuracies for identifying the fault types and evaluating the fault performance degradation, is beyond 94.6% in all cases. This clearly demonstrates the potential applicability of the proposed method for multiple fault diagnosis tasks of HST bogie system.