Research And Implementation Of Adaptive Cooperative Current-sharing Strategies For Charging System Of Energy Storage Type Light Rail

:As a new kind of electric traction light rail transport, energy storage type light rail adopts supercapacitors as power supply. With this energy storage technology, it does not need to construct traction power grid and recover the energy of regenerative braking. Energy storage type light rail is charged rapidly with the charging system when it parks on the station. Therefore, the charging system should support the large enough output power to shorten the charging process. An effective method to solve this problem is connecting numbers of chargers in parallel to increase system capacity. However, the disadvantages in parallel charging system are the charging current imbalance between the chargers, and the poor anti-interference performance. In this dissertation, a cooperative technique to solve the charging current imbalance problem is proposed and an adaptive cooperative current-sharing strategy is presented to resistance the disturbance.First of all, according to the structure and characteristics of the parallel charging system, the model of the chargers is presented by using the state-space averaging approach. Considering the capacitance of supercapacitor changing when terminal voltage difference, a Three Branches model for supercapacitor is established. The charging and discharging characteristic of supercapacitor is described accurately by the model. An experiment method is adopted to ensure the parameters of the model. A desired current trajectory is designed by comparing to the common charging modes.Secondly, in order to solve the imbalance problem in charging system, a distributed cooperative current-sharing strategy is presented. According to the local states of neighbouring chargers, a cooperative technique to enable the cooperative errors of distributed chargers convergent to zeros and achieve tracking to the desired trajectory. A Lyapunov method demonstrates the stability of the control law under the control gain select range. The distributed cooperative current-sharing strategy is verified by simulations in different control gains.Thirdly, considering the disturbance of charging system, on the basis of cooperative current-sharing strategy, an adaptive cooperative current-sharing strategy is proposed. The adaptive protocol consists of a linear protocol and a nonlinear control term with neural network at each charger. The neural network turning law is designed to update the weights matrix. A Lyapunov technique is presented for designing a robust adaptive synchronization control protocol. The simulations results demonstrate the anti-interference performance of adaptive cooperative current-sharing strategy is more excellent than the former ones.Finally, based on the overall structure of the charging system of energy storage type light rail, the thesis introduces system’s hardware platform and software architecture. The distributed cooperative current-sharing strategy has already been tested in the charging system. Practical application proves that the proposed scheme is entirely effective and feasible.