Researchers from the University of Zagreb, including Josip Kir Hromatko, Šandor Ileš, Branimir Škugor, and Joško Deur, have developed a new method for optimizing energy efficiency in electric vehicles (EVs) equipped with multiple motors. Their work, published in the journal IEEE Transactions on Vehicular Technology, focuses on minimizing battery energy consumption through strategic torque allocation.
Electric vehicles with multiple motors offer flexibility in meeting driver torque demands. The researchers aimed to harness this flexibility to improve energy efficiency. They approached this problem by approximating electric motor losses using higher-order polynomials with specific properties. By imposing monotonicity and positivity constraints on these polynomial models using sum of squares programming, they ensured a well-behaved optimization landscape. This methodology is robust against noisy or sparse data and maintains the computational efficiency of polynomial function approximation.
The torque allocation problem was formulated as a constrained nonlinear optimization problem. The researchers found that, in the nominal case, the first-order necessary conditions for optimality could be used to obtain a global solution. This approach allows for real-time optimization and seamless integration with other vehicle control systems.
To evaluate the performance of their method, the researchers tested it on several certification driving cycles against a grid search-based benchmark. The results showed a modest influence on electric energy consumption, demonstrating the potential of this method for practical applications in the energy sector. By optimizing torque allocation, this research could contribute to improving the energy efficiency of electric vehicles, ultimately reducing their environmental impact and operating costs.
This article is based on research available at arXiv.