Design and control of a new power conditioning system based on superconducting magnetic energy storage

Superconducting magnetic energy storage (SMES) is characteristic as high power capacity and quick response time, which can be widely applied in power grid to suppress rapid power fluctuation, and improve transient stability. But the traditional power conditioning system (PCS) of SMES, which handle the power transfer between the superconducting magnet and AC power grid, will induce high frequency overvoltage on the magnet, which may lead to the rapid aging of the coil and the decrease of its service life. In this paper, a new circuit scheme of PCS is proposed, and GaN power semiconductor is adopted to increase switching frequency and power density. Then, the model predictive control (MPC) method based on the new scheme is proposed. And the software implementation process for MPC is optimized to reduce the implementation of MPC algorithm time. Finally, simulation and experiment show that the new PCS can not only solve the overvoltage problem of superconducting magnet from the source but also reduce the power tracking error of the superconducting energy storage system. By these software optimization measures, the MPC algorithm can be finished within 10 μs, and the portability and response speed of SMES can be significantly improved to make SMES used in more situations. •The obvious advantages of using GaN-based semiconductors in power regulation systems e are discussed.•A new power conditioning system is proposed to protect the superconducting inductor from high- frequency pulse voltage.•The prediction model of the new power conditioning system is established.•The model prediction control method to realize the tracking control of charge and discharge power instructions is proposed.•The control cycle is shortened so that model prediction control can be applied to occasions with higher operating frequency.