New Control Strategy Boosts Precision and Stability in Electric Vehicle Motors
In the rapidly evolving world of electric mobility, where performance, efficiency, and reliability are paramount, a groundbreaking advancement in motor control technology is capturing the attention of automotive engineers and researchers alike. A recent study introduces a novel control framework designed to significantly enhance the speed tracking accuracy and disturbance rejection capabilities of permanent magnet synchronous motors (PMSMs) used in electric vehicles (EVs). This innovation, developed by a collaborative team of engineers from State Grid Jiangsu Electric Power Co., Ltd. and Zhejiang Normal University, promises to deliver smoother acceleration, improved energy efficiency, and a more responsive driving experience, particularly under challenging real-world conditions.
The research, published in the Control and Information Technology journal, presents a sophisticated control strategy known as the Disturbance-Observer Based Prescribed-Performance Backstepping Controller (DPBC). This method directly addresses two persistent challenges in EV motor control: the disruptive impact of external load disturbances—such as those caused by changing road gradients, wind resistance, or sudden braking—and the need for motors to respond with both lightning-fast precision and rock-solid stability during frequent acceleration and deceleration cycles.
Traditional Proportional-Integral (PI) controllers, which have been the workhorse of motor control for decades, often struggle in these dynamic scenarios. While simple and robust, PI controllers can exhibit significant speed fluctuations, overshoot, and sluggish response when faced with unexpected load changes. This can lead to a less comfortable ride, reduced energy efficiency, and increased wear on the drivetrain. The quest for more advanced control algorithms has led to the exploration of techniques like fuzzy logic, neural networks, and various forms of nonlinear control. Among these, backstepping control has emerged as a powerful theoretical framework for managing the complex, nonlinear dynamics of PMSMs. However, even this advanced method has its own Achilles’ heel: the “differential explosion” problem, where the mathematical complexity of the controller grows exponentially with each step of the design process, making it difficult to implement in real-time systems.
The DPBC strategy, as detailed in the study, is a masterclass in systems engineering, elegantly weaving together three powerful control concepts to overcome these limitations. The first pillar of this approach is the integration of a prescribed performance control framework. This is not merely about achieving a target speed; it is about guaranteeing how the system reaches that target. By defining a “prescribed performance function,” the engineers can set strict, pre-determined boundaries for the speed tracking error—the difference between the actual motor speed and the desired speed. This function ensures that the error converges to a very small value within a specified time, with a guaranteed maximum overshoot and a defined convergence rate. For an EV driver, this translates to a motor that accelerates and decelerates with a predictable, smooth, and jerk-free motion, eliminating the lurching or hunting behavior that can sometimes be felt with conventional systems. This level of control is particularly critical for functions like cruise control and regenerative braking, where consistent and stable performance is essential for both comfort and safety.
The second key innovation lies in the design of a novel disturbance observer. An observer, in control theory, is like a sophisticated sensor that can estimate things that are difficult or impossible to measure directly. In this case, the observer is tasked with estimating the combined effect of all external load disturbances and internal system uncertainties that act upon the motor. The brilliance of this new observer is its use of a “super-twisting algorithm,” a type of second-order sliding mode control. This allows the observer to estimate the disturbance not just accurately, but also in a finite amount of time, which is a significant improvement over many traditional observers that may take an infinite time to converge. The estimated disturbance is then fed directly into the main controller, allowing it to proactively compensate for the disturbance before it can significantly affect the motor’s speed. This is akin to a skilled driver who can anticipate a hill and adjust the throttle accordingly, but done at a speed and precision far beyond human capability. This real-time compensation is what gives the DPBC system its exceptional robustness, allowing the EV to maintain its speed with minimal fluctuation even when climbing a steep incline or encountering a strong headwind.
The third and final component of the DPBC strategy is the solution to the “differential explosion” problem. To implement the backstepping control law, the controller needs to know the rate of change (derivative) of certain internal control signals. Calculating these derivatives analytically can lead to extremely complex and unwieldy equations. The research team ingeniously sidesteps this issue by employing a second-order sliding mode differentiator (SOSMD). This differentiator acts as a high-speed, real-time calculator that can accurately estimate the required derivatives from the available signals. It provides a clean, stable estimate of the signal’s rate of change without the need for complex mathematical operations, making the entire control system much more practical for implementation in the embedded microcontrollers found in modern EVs. This component ensures that the theoretical elegance of the backstepping design can be translated into a reliable and efficient real-world application.
The validation of this new control strategy was conducted through rigorous computer simulations, a standard and essential step in the development of any new control system. The researchers built a detailed model of an EV’s PMSM drive system using MATLAB/Simulink, a powerful simulation environment widely used in the automotive and aerospace industries. They then subjected this model to two distinct driving scenarios to test the limits of their DPBC controller. The first scenario simulated urban driving, with the vehicle constantly accelerating and decelerating—a situation that is notoriously difficult for motor controllers. The second scenario mimicked highway cruising, where the vehicle is expected to maintain a constant speed despite external disturbances.
The results of these simulations were nothing short of impressive. When compared directly against both a traditional PI controller and a standard backstepping (BC) controller, the DPBC system demonstrated a remarkable improvement in performance. The most striking finding was the dramatic increase in speed tracking precision. The data showed that the DPBC controller improved tracking accuracy by more than five times compared to the conventional PI control. This means the motor’s actual speed stayed much closer to the desired speed, with deviations measured in mere fractions of a radian per second. This level of precision is a game-changer, as it directly contributes to a smoother, more refined driving experience and more efficient energy use.
Furthermore, the simulations clearly illustrated the superior anti-interference capability of the DPBC system. Under the same load disturbance, the PI and BC controllers showed significant speed oscillations and “jitter,” while the DPBC-controlled motor maintained a remarkably smooth and stable speed profile. The speed tracking error for the PI and BC controllers frequently exceeded their predefined performance boundaries, indicating a loss of control precision. In stark contrast, the error for the DPBC controller remained tightly bounded within its prescribed limits at all times, a testament to the effectiveness of the integrated performance function and disturbance observer.
The benefits of this advanced control extended beyond just speed. The study also analyzed the motor’s electromagnetic torque, the force that actually turns the wheels. A smooth, stable torque output is critical for a comfortable ride, as torque “ripple” or pulsation can be felt as a vibration or shudder in the vehicle. The simulation results showed that the DPBC method produced significantly less torque ripple compared to both the PI and BC methods. This reduction in torque pulsation means that an EV equipped with this controller would not only accelerate more precisely but also do so in a noticeably smoother and quieter manner, enhancing the overall quality of the driving experience.
The implications of this research extend far beyond a single academic paper. As the automotive industry pushes toward higher levels of automation and performance, the demand for smarter, more responsive, and more efficient powertrains will only grow. The DPBC strategy represents a significant step forward in meeting this demand. Its ability to guarantee high transient and steady-state performance under real-world disturbances makes it a highly attractive solution for next-generation EVs. While the current work is based on simulation, the authors have indicated their intention to build a physical experimental platform to validate the controller’s real-time performance. This next step is crucial for transitioning the technology from the laboratory to the production line.
The success of this research is a product of a strong interdisciplinary collaboration. The team, led by NI Shuangfei from the Changzhou Jintan District Power Supply Branch, brought together expertise from both the power systems and academic research domains. DAI Yuchen from Zhejiang Normal University’s College of Engineering contributed specialized knowledge in advanced control theory. This blend of practical engineering insight and theoretical rigor is often the key to developing truly impactful technologies. Their work is a prime example of how fundamental research in control systems can have a direct and tangible impact on the products that consumers use every day.
This advancement also highlights a broader trend in the automotive sector: the increasing importance of software and control algorithms as differentiating factors. In an era where the basic hardware of EVs—batteries, motors, and power electronics—is becoming more standardized, the “intelligence” of the vehicle’s control systems is where manufacturers can truly set themselves apart. A motor that responds with unparalleled smoothness, efficiency, and resilience to disturbances is a powerful selling point. The DPBC strategy, with its focus on guaranteed performance and robustness, is perfectly aligned with this trend.
In conclusion, the development of the disturbance-observer based prescribed-performance backstepping controller marks a significant milestone in the field of electric vehicle motor control. By seamlessly integrating prescribed performance, a novel finite-time disturbance observer, and a practical solution to the differential explosion problem, the research team has created a control system that outperforms conventional methods by a wide margin. With a proven fivefold improvement in speed tracking accuracy and exceptional resistance to real-world disturbances, this technology has the potential to redefine the standards for EV drivetrain performance. As the world transitions to electric mobility, innovations like this one will be essential in building vehicles that are not only sustainable but also deliver an exceptional, high-performance driving experience.
NI Shuangfei, DAI Yuchen, CAI Qicheng, SUN Zhongyang, Control and Information Technology, doi:10.13889/j.issn.2096-5427.2024.01.003