Automatic MTPA Tracking Using Online Simplex Algorithm for IPMSM Drives in Vehicle Applications

Interior permanent magnet synchronous machines provide a high torque-to-weight ratio which makes them the first choice in hybrid or fully electric vehicle propulsion systems. The design of the field-oriented control is a complex task if one tries to operate the machine with maximum efficiency in the whole torque-speed-range. As previous works have shown it is possible to optimize the stator current references in advance to provide the desired torque with minimal losses. However, when taking magnetic saturation and temperature variations into account, very precise information about the machine is indispensable. In this work an approach to automatically track the maximum-torque-per-ampere trajectory is presented. When the output torque is estimated correctly, the optimal current angle can be found online by using a newly adapted Nelder-Mead simplex algorithm. This procedure is faster than existing signal injection methods and doesn´´t need permanent current and torque oscillations. Using the algorithm, an energy efficient control can be ensured without extended knowledge even for highly exploited machines with strong magnetic saturation influence.