New Control Strategy Boosts Efficiency in Electric Vehicle Motors
In the relentless pursuit of longer range and higher efficiency for electric vehicles (EVs), a groundbreaking study from Shandong University has unveiled a novel control strategy that promises significant improvements in motor performance. This research, spearheaded by doctoral candidate Wang Bingdong under the guidance of Professor Wang Daohan, introduces an innovative approach to maximizing torque output while minimizing copper losses in a specialized type of permanent magnet synchronous motor (PMSM). The findings, published in the prestigious Transactions of China Electrotechnical Society, could pave the way for a new generation of more efficient and powerful EV drivetrains.
The heart of this innovation lies in a unique motor design known as the AC Flux-Regulation Permanent Magnet Synchronous Motor (ACFR-PMSM). Unlike conventional motors found in most EVs today, which rely on a single magnetic field source—either permanent magnets or electrical windings—the ACFR-PMSM employs a sophisticated dual-field architecture. This design features two distinct but interconnected sets of windings: a standard radial winding and an additional axial winding mounted on the rotor’s end. These windings operate independently, each controlled by its own dedicated inverter, creating a system with unparalleled flexibility.
Traditional PMSMs, while highly efficient at low speeds due to their strong permanent magnets, face a critical limitation at high speeds. To prevent the back-electromotive force (back-EMF) from exceeding the battery voltage and damaging the system, engineers must “weaken” the magnetic field. This is typically done by injecting a negative current into the motor’s d-axis, a process known as weak-field control. However, this method comes at a steep cost: it drastically reduces the motor’s ability to produce torque, leading to a narrow constant-power speed range. For an EV driver, this translates to a vehicle that performs well up to a certain speed but then experiences a sharp drop-off in acceleration, limiting its overall utility and efficiency on highways.
The ACFR-PMSM offers a fundamentally different solution to this problem. Its key advantage is what the researchers describe as “magnetic flux co-regulation.” When the motor needs to operate at high speeds, the radial winding can be used to weaken the main magnetic field, just like in a traditional motor. But here’s where the magic happens: instead of simply dissipating the magnetic energy, the ACFR-PMSM uses its independent axial winding to actively compensate. By adjusting the current in the axial winding, the motor can strengthen a secondary magnetic path. This doesn’t just maintain power; it effectively redirects the magnetic flux that would otherwise be lost during weakening. The result is a far broader constant-power operating range, allowing the EV to maintain high performance across a much wider spectrum of speeds without sacrificing efficiency. The researchers aptly liken this process to “turning blockage into flow and transforming leakage into use,” a metaphor that captures the elegant efficiency of the system.
This complex interplay between the radial and axial magnetic fields presents both an opportunity and a challenge. The primary opportunity is the potential for dramatically improved performance. The challenge, however, is controlling it. With two sets of windings, each capable of carrying currents with multiple components (d-axis and q-axis), the number of possible control combinations is vast. Simply applying the classic “id=0” control strategy—a common method for maximizing torque per ampere in simple PMSMs—is insufficient. It fails to harness the full potential of the motor’s dual-field structure and does not optimize for the ultimate goal of EV manufacturers: maximum efficiency.
To unlock the ACFR-PMSM’s true potential, Wang Bingdong and his team developed a revolutionary control algorithm called Maximum Torque Per Copper Loss (MTPCL) control. The objective of MTPCL is deceptively simple: for any given amount of electrical current (which directly correlates to copper loss and heat generation), produce the absolute maximum amount of mechanical torque. Copper losses are a major source of inefficiency in electric motors, converting valuable electrical energy into wasted heat. By minimizing these losses for every unit of torque produced, the MTPCL strategy directly increases the overall efficiency of the drivetrain, which in turn extends the vehicle’s driving range.
Achieving this optimal balance is an extraordinarily complex mathematical problem. The total torque output of the ACFR-PMSM is not just the sum of the torques from the radial and axial parts; it is a dynamic interaction influenced by the cross-coupling of their magnetic fields. Injecting a current in one winding affects the magnetic environment of the other. Therefore, the control strategy must consider the entire system holistically, not as two separate motors bolted together.
To solve this, the Shandong University team employed advanced optimization techniques. They first constructed a comprehensive mathematical model of the ACFR-PMSM, capturing its non-linear, multi-variable nature. This model included detailed equations for voltage, magnetic flux linkage, electromagnetic torque, and motion, all within a dual dq-axis coordinate system that allowed them to analyze the radial and axial components simultaneously. From this model, they derived a formula for the motor’s total copper loss based on the currents flowing through all four winding components (radial-d, radial-q, axial-d, and axial-q).
The core of their MTPCL algorithm is finding the precise combination of these four currents that maximizes torque for a given copper loss. This is a constrained optimization problem, limited by the physical capabilities of the inverters and the thermal limits of the windings. The researchers tackled this using the Lagrange multiplier method, a powerful tool from calculus designed for finding maxima and minima of functions subject to constraints. This method allowed them to derive a set of equations whose solutions represent the “optimal current trajectory”—a map that tells the controller exactly how much current to send to each winding component for any desired torque level.
To ensure the accuracy of their theoretical calculations, the team conducted a rigorous validation process. They implemented a brute-force “current selection exhaustive method,” systematically testing millions of possible current combinations within the operational limits. For each combination, they calculated the resulting torque and copper loss, then identified the ones that offered the best torque-per-loss ratio. The results of this computationally intensive simulation were then compared to the trajectories predicted by the Lagrange multiplier method. The close agreement between the two methods served as a robust verification of the algorithm’s correctness, lending significant credibility to their findings.
With the MTPCL control strategy fully defined, the next step was to test it in the real world. The team did not rely solely on simulations; they took the crucial step of building a physical prototype. Constructing a motor with such a complex radial-axial structure presented significant manufacturing challenges, particularly in ensuring the precise alignment of the stators and the integrity of the rotor. Nevertheless, they successfully fabricated a 200-watt ACFR-PMSM prototype with six poles and 36 slots. To control it, they built two custom-designed inverters, one for the radial winding and one for the axial winding, creating a complete, functional testbed.
This experimental platform was used to conduct a series of rigorous tests comparing the new MTPCL control against the conventional double-id=0 control. The results were compelling. In terms of dynamic performance—how quickly and smoothly the motor responded to changes in speed and load commands—the MTPCL control performed on par with the traditional method. There was no compromise in responsiveness or stability, a critical factor for drivability and safety in a real-world vehicle.
The real difference emerged when examining efficiency. Under various load conditions, from light cruising to full-throttle acceleration, the MTPCL-controlled motor consistently demonstrated lower copper losses. The data showed a clear trend: as the required torque increased, the efficiency advantage of the MTPCL strategy became even more pronounced. At the motor’s rated torque of 2.0 Newton-meters, the reduction in copper loss was a remarkable 12.93%. This figure is not just a laboratory curiosity; it represents a tangible improvement that could translate to several extra miles of range for an EV over a typical driving cycle.
A deep dive into the underlying physics revealed why the MTPL strategy works so well. Under the classic id=0 control, only the q-axis currents are active, producing torque but doing nothing to optimize the internal magnetic fields. In contrast, the MTPCL control strategically uses negative d-axis currents in both the radial and axial windings. This might seem counterintuitive, as a negative d-axis current typically demagnetizes the field and reduces torque. However, because of the motor’s unique coupled structure, this action has a synergistic effect. The weak-field current in the radial winding prepares the main path for high-speed operation, while the weak-field current in the axial winding actually strengthens the coupling, enhancing the contribution of the radial winding. The net result is a significant boost in the torque generated by the radial part, which more than compensates for any small decrease in torque from the axial part. It is a masterclass in systems-level engineering, where a local “loss” leads to a global gain.
This research from Shandong University represents a significant leap forward in electric motor technology. While hybrid-excitation motors have been explored before, often using DC windings for field control, the ACFR-PMSM’s use of an independent AC winding for dynamic flux regulation is a key differentiator. It transforms the auxiliary winding from a passive regulator into an active participant in power production, enabling a level of performance refinement that was previously unattainable.
The implications for the automotive industry are substantial. As EVs continue to evolve, the focus is shifting from simply replacing the internal combustion engine to reimagining the entire drivetrain. Efficiency is paramount, not just for extending range but also for reducing battery size, weight, and cost. The MTPCL control strategy for the ACFR-PMSM offers a clear path to achieving these goals. It demonstrates that by embracing more complex, integrated designs and developing equally sophisticated control algorithms, engineers can squeeze more performance out of every watt-hour of energy stored in the battery.
While the current prototype is a modest 200 watts, the principles are scalable. The fundamental architecture and control logic can be applied to larger, more powerful motors suitable for passenger cars and commercial vehicles. The research team acknowledges that there are still engineering hurdles to overcome, such as managing torque ripple and perfecting the manufacturing process for mass production. However, the successful demonstration of the concept, backed by both simulation and hard experimental data, provides a solid foundation for future development.
In conclusion, the work of Wang Bingdong, Wang Daohan, Wang Xiaoji, Xu Guangsheng, and Wang Xiuhe presents a compelling vision for the future of electric propulsion. By ingeniously combining a novel motor topology with a mathematically rigorous control strategy, they have created a system that is greater than the sum of its parts. Their research, published in the Transactions of China Electrotechnical Society, stands as a testament to the power of innovation in addressing the core challenges of sustainable transportation.
Wang Bingdong, Wang Daohan, Wang Xiaoji, Xu Guangsheng, Wang Xiuhe, School of Electrical Engineering, Shandong University, Transactions of China Electrotechnical Society, DOI: 10.19595/j.cnki.1000-6753.tces.230602