High-speed permanent-magnet (PM) machines are promising for many industrial applications. The main design challenge of high-speed PM machines comes from the PM rotor design due to the low strength and high thermal sensitivity of PM materials. In this paper, the mechanical constraints and speed limits for a high-speed rotor topology are derived with analytical equations. Thereafter, a case study is done to design an 80 kW, 80,000 rpm PM machine. Analysis results show that the high rotor eddy-current loss is one of the key factors contributing to the overheat of the PMs. Hence, a hybrid sleeve topology, which contains an inner titanium cylinder and an outer cylinder made of carbon fiber composite (CFC), is proposed to suppress the rotor eddy-current losses. The optimal dimensions of the hybrid sleeve are obtained by finite element analysis (FEA). Thereafter, a comparison among the proposed and two other sleeve topologies, namely, the copper clad sleeve and grooved sleeve, is done from the point of view of their rotor eddy-current losses, rotor temperature rise, mechanical stress and critical speed. The results indicate that the hybrid sleeve gains an advantage over the other two topologies. For the designed case, the rotor eddy-current losses and temperature rise are decreased by 66.4% and 73 °C, respectively, by using the hybrid sleeve. Furthermore, a reduction of 14.5% for the rotor maximum stress is obtained with the hybrid sleeve, while the rotor's first critical speed has little reduction.

The accurate estimation of mechanical parameters is essential for the control optimization and condition monitoring of permanent-magnet synchronous motor (PMSM) drives. In this article, an online parallel estimation scheme is proposed based on a network of three interconnected extended sliding-mode observers. With a two-step mechanism derived from the developed network, this scheme can simultaneously achieve the accurate real-time observation of three mechanical parameters, including the viscous friction coefficient, the moment of inertia, and the load torque. The first step utilizes the presented network to precisely acquire the Coulomb friction coefficient by offering only one speed. With aiding from this estimated coefficient, the second step attains the high-precision online identification of the three concerned parameters by compensating the Coulomb friction torque to counteract its adverse effect. These two steps are implemented utilizing the identical network, thus, furthest avoiding increasing the estimation complexity. Besides, any additional knowledge other than the rotor speed and electromagnetic torque is no longer required for guaranteeing the asymptotic convergence of the estimation errors. Sufficient evaluations from simulations and experiments confirm the validity of the proposed scheme.

Acquiring the accurate knowledge of nonlinear friction and load torque is of great interest for optimizing the control behavior of permanent-magnet synchronous motor drives. In this work, a friction-and-load adaptive identification scheme based on a parallel-observer-based network with model compensation (POBN-MC) is presented. The developed network possesses a parallel structure consisting of the designed two novel observers, which involve a gain-adaptation super-twisting load torque observer and a variable-learning-rate Adaline inertia observer. A nonempirical friction model is proposed to capture friction, forming the model compensation part that is exploited for correcting the torque input of the network. With a two-step mechanism derived from the POBN-MC, the proposed scheme attains the online adaptive identification of the friction and load torque in a manner that integrates both accuracy and simplicity. In the first step, an explicit mapping relationship between the nonlinear friction torque and the rotor speed is determined with the speed response triggered by the natural deceleration. The second step accomplishes the online observation with regard to friction-and-load information matching the real-time operating conditions. Sufficient theoretical analyses, as well as the validations of numerous simulations and experiments, are presented to support the suggested scheme.

Accurate and real-time acquisition of vehicle state parameters is key to improving the performance of vehicle control systems. To improve the accuracy of state parameter estimation for distributed drive electric vehicles, an unscented Kalman filter (UKF) algorithm combined with the Huber method is proposed. In this paper, we introduce the nonlinear modified Dugoff tire model, build a nonlinear three-degrees-of-freedom time-varying parametric vehicle dynamics model, and extend the vehicle mass, the height of the center of gravity, and the yaw moment of inertia, which are significantly influenced by the driving state, into the vehicle state vector. The vehicle state parameter observer was designed using an unscented Kalman filter framework. The Huber cost function was introduced to correct the measured noise and state covariance in real-time to improve the robustness of the observer. The simulation verification of a double-lane change and straight-line driving conditions at constant speed was carried out using the Simulink/Carsim platform. The results show that observation using the Huber-based robust unscented Kalman filter (HRUKF) more realistically reflects the vehicle state in real-time, effectively suppresses the influence of abnormal error and noise, and obtains high observation accuracy.

In surface-mounted permanent magnet synchronous motor (SPMSM) drives, the distinguished orthogonal-principle-based method is presently popular for handling simultaneous inertia and disturbance torque estimation due to its simplicity and ease of implementation. Nevertheless, it is facing challenges posed by the rotor flux-linkage mismatch, zero-derivative constraint, and operating-condition limitation. To cope with these issues, this article presents an alternative scheme, which explores a stable simultaneous inertia and disturbance torque estimation with the real-time flux-linkage correction. First, an Adaline-based estimator is developed to acquire the disturbance torque. With an input-adaptive learning rate designed by the Lyapunov theory, this estimator can prevent potential convergence failure. Then, an extended sliding-mode observer for inertia estimation is proposed regarding the inertia as a new system state. By using the identified disturbance torque to devise an adaptive feedback gain, this observer guarantees the asymptotic error convergence. Finally, to counteract the flux-linkage mismatch, a flux-linkage observer with a time-varying learning rate is constructed utilizing the Adaline technique. Moreover, it attains robust convergence and only has one to-be-adjusted parameter. The effectiveness of our scheme is validated by abundant simulations and real-time experiments.

The distributed drive arrangement form has better potential for cooperative control of dynamics, but this drive arrangement form increases the parameter acquisition workload of the control system and increases the difficulty of vehicle control accordingly. In order to observe the vehicle motion state accurately and in real-time, while reducing the effect of uncertainty in noise statistical information, the vehicle state observer is designed based on interacting multiple model theory with square root cubature Kalman filter (IMM-SCKF). The IMM-SCKF algorithm sub-model considers different state noise and measurement noise, and the introduction of the square root filter reduces the complexity of the algorithm while ensuring accuracy and real-time performance. To estimate the vehicle longitudinal, lateral, and yaw motion states, the algorithm uses a three degree of freedom (3-DOF) vehicle dynamics model and a nonlinear brush tire model, which is then validated in a Carsim-Simulink co-simulation platform for multiple operating conditions. The results show that the IMM-SCKF algorithm’s fusion output results can effectively follow the sub-model with smaller output errors, and that the IMM-SCKF algorithm’s results are superior to the traditional SCKF algorithm’s results.