High-Performance Torque Control of Electrical Excited Synchronous Motors

Energy is a key aspect of maintaining economic growth and daily living. The world’s energy system's heavy dependence on fossil fuels has unfortunately directly threatened human existence due to ascending phenomena like air pollution and climate change. The transportation sector is one of the largest consumers of energy and a major contributor to air pollutants and greenhouse gas (GHG) emissions. This is the reason why transport electrification is a highly researched area nowadays. Nevertheless, the production of electric transport makes use of critical materials, including Rare-Earth Elements (REE), such as ferromagnetic materials, most of which commercial electric motors rely on. REE availability and cost increasingly depend on geopolitical and economic factors. EESMs (also known as wound-field synchronous motors) have recently been garnering attention from the automotive industry as a compelling permanent magnet-free alternative motor. It replaces PMs with an excitation winding placed on the rotor, and the injected current can be regulated using slip rings or wireless power transfer solutions. This way, the magnetization level can be calibrated at any operating point to maximize motor efficiency. Currently, the technical literature reports few control algorithms for EESMs able to guarantee high- performance torque regulation in any operating condition, including deep flux weakening under voltage and current constraints imposed by the whole eDrive chain. This thesis will address the problem and perform advanced elaboration of torque and flux maps of an EESM to support the development of an innovative torque controller characterized by the best dynamic performance of torque regulation in any possible machine’s operating point. Beyond the various torque control algorithms proposed by technical literature, analyzed one by one by means of pros and cons for the specific mission, the choice fell on the recently proposed Flux Polar Control (FPC), which performs torque regulation by directly controlling the stator flux vector in terms of amplitude and load angle, thus involving two control loops. This approach guides through some additional advantages with respect to other solutions (such as CVC, DTC, or DFVC): (i) the performance of the load angle control loop is independent of the machine’s operating point, and implementing a normalization to the actual stator flux amplitude, the gains of the load angle regulator are automatically self-calibrated to retain the same dynamic performance in any possible operating point, (ii) MTPV operation can be straightforwardly performed by simply limiting the reference load angle, thus avoiding time-consuming outer regulators. The two control loops feature the same performance, decoupled voltage margins, and independence of the machine’s operating point. FPC properties potentially lead to the highest dynamic performance of torque regulation compared to other proposed control algorithms, with the only drawback of not automatically preserving linearity of torque regulation. Therefore, one of the challenges of this work is to overcome this drawback by computing a proper load angle look-up table to be interpolated to get the reference load angle, fulfilled with all the needed constraints. The proposed methodology is experimentally validated considering, as a case study, the wound field traction motor adopted by Renault ZOE R135.