六相永磁同步电机具有缺相运行能力，因此必须对其高阻连接状态作出精准预判，以确保对故障线路实施有效切断，防止系统扰动引起保护误动作，并为容错控制提供可靠判据。基于矢量空间解耦方法建立了六相永磁同步电机完全解耦的数学模型，并建立其控制系统模型。采集正常状态与高阻连接状态下的电机信号，通过小波包分解提取其能量距特征，输入前向反馈神经网络进行离线训练，最后将其应用于剧烈变化工况下，在线感知高阻连接状态的发展态势。基于Matlab进行仿真，结果表明所提策略能够有效识别高阻连接状态，灵敏感知其发展态势，并在高阻故障发生前发出预警信号，同时对剧烈变化工况具有一定鲁棒性。

Six-phase permanent magnet synchronous motors have the ability of phase-deficient operation, thus precise prediction of their high resistance connection state must be made to ensure effective disconnection for faulty lines, prevent protection misoperation caused by system disturbances, and provide reliable criteria for fault-tolerant control. A mathematical model for complete decoupling of six-phase permanent magnet synchronous motor is established based on vector space decomposition, and its control system model is established. Motor signals in normal state and high resistance connection state are collected, and their energy distance features are extracted by wavelet packet decomposition, input to the back propagation neural network for offline training, and finally applied to sense development situation of high resistance connection state online under drastic conditions. Simulations are carried out based on Matlab, and the results show that the proposed strategy can effectively identify high resistance connection state, sensitively sense its development situation, send warning signals before the high resistance faults occur, and have certain robustness to drastic conditions.

国家电网有限公司总部管理科技项目资助（5209KZ220002）；国家自然科学基金(52077135)