The lack of a load power prediction function in conventional multi-wheel electric drive vehicles leads to lags in control action. To address this issue, we developed a real-time energy management strategy with improved load power prediction accuracy. Based on an assessment of the overall vehicle structure, a mathematical model for each power source was established using theoretical analysis and data fitting. A method for the joint prediction of non-stationary load power combining Kalman filter and Markov chain forecasting methods was established, and a multi-objective optimization function was constructed under the nonlinear model predictive control framework. To enable real-time optimal control command, a sequential quadratic programming method in the finite time domain was applied. Finally, the multi-power source was optimized and coordinated. Multi-road driving experiments were carried out using a hardware-in-the-loop simulation platform. Comparisons of energy management control strategies with and without power prediction revealed that applying the former enhances the predictability of future load power, significantly optimizes the coordinated control of multiple power sources, improves vehicle fuel economy, and stabilizes bus voltage and battery state of charge. Moreover, it has specific reference significance in engineering application scenarios under conventional model predictive control.INDEX TERMS energy management strategy, integrated power system, load power prediction, model predictive control, multi-wheel electric drive vehicle
This paper proposes an artificial neural network for monitoring and detecting the eccentric error of synchronous reluctance motors. Firstly, a 15 kW synchronous reluctance motor is introduced and took as a case study to investigate the effects of eccentric rotor. Then, the equivalent magnetic circuits of the studied motor are analyzed and developed, in cases of dynamic eccentric rotor and static eccentric rotor condition, respectively. After that, the analytical equations of the studied motor are derived, in terms of its air-gap flux density, electromagnetic torque, and electromagnetic force, followed by the electromagnetic finite element analyses. Then, the modal analyses of the stator and the whole motor are performed, respectively, to explore the natural frequency and the modal shape of the motor, by which the further vibrational analysis is possible to be conducted. The vibration level of the housing is furtherly studied to investigate its relationship with the rotor eccentricity, which is validated by the prototype test. Furthermore, an artificial neural network, which has 3 layers, is proposed. By taking the air-gap flux density, the electromagnetic force, and the vibrational level as inputs, and taking the eccentric distance as output, the proposed neural network is trained till the error smaller than 5%. Therefore, this neural network is obtaining the input parameters of the tested motor, based on which it is automatically monitoring and reporting the eccentric error to the upper-level control center.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.