Time-Optimal Current Control of Synchronous Motor Drives

Václav Šmídl, Antonín Glac, Zdeněk PeroutkaActuators 2022 · DOI

TL;DR

This paper presents a novel time-optimal current control strategy for synchronous motor drives that accounts for stator resistance and differing d-q axis inductances, outperforming existing methods by up to four times in high-speed scenarios. This improved controller offers faster, more accurate current tracking, crucial for dynamic applications like robotics.

What Problem Does This Paper Address?

Synchronous motor drives are fundamental in many applications, from industrial automation to electric vehicles. A critical component of their control is the current loop, which needs to be as fast and accurate as possible. However, achieving this is challenging, especially at high speeds or during field weakening, where the available voltage is limited. Existing "time-optimal" controllers often simplify the motor model by neglecting stator resistance and assuming equal inductances in the direct (d) and quadrature (q) axes. While these simplifications make the math easier, they lead to suboptimal performance and can even result in slower responses or increased energy losses in real-world drives. This paper aims to bridge this gap by developing a time-optimal current control strategy that accurately models the synchronous motor, including its resistance and differential inductances, and can be implemented efficiently in real-time.

Methods

The core of this work lies in deriving a time-optimal control (TOC) solution for a general synchronous motor model. The authors leverage Pontryagin's maximum principle to formulate the problem in continuous time. They then derive an explicit formula for the time-optimal current trajectory, which considers stator resistance and distinct d- and q-axis inductances. A key contribution is a simplification of this exact formula that allows for real-time evaluation with low computational cost, making it suitable for implementation on industrial DSPs. This simplified formula is then solved using a bisection-based optimization algorithm to determine the optimal control inputs. For very small time steps or when the system is close to its target, a conventional deadbeat controller is used as a fallback to ensure stability and efficiency.

Datasets

Not discussed in the paper. The evaluation relies on simulations and experimental data from a specific IPMSM drive prototype.

Key Insights and Ideas

  • Resistance and Inductance Matter: The paper strongly emphasizes that neglecting stator resistance and assuming equal inductances significantly degrades the performance of time-optimal controllers, especially at higher speeds.
  • Trade-offs in Simplification: While simplified models are computationally cheaper, they lead to a "poor solution" that is slower and less efficient in real-world scenarios, particularly when resistance effects become significant.
  • Feed-forward Advantage: The proposed time-optimal feed-forward controller can achieve significantly faster transients than traditional feedback-based controllers (like PI) and even standard deadbeat controllers, especially in field-weakening regimes.
  • Universal Building Block: The developed TOC strategy is presented as a modular component that can be integrated with existing feedback controllers and observers, offering a flexible upgrade path for current control systems.
  • Bisection for Real-time: The use of a bisection method to solve the implicit equation for the optimal time is a practical approach for real-time implementation, balancing accuracy with computational feasibility.

Key Results

  • Four Times Faster: The proposed time-optimal controller achieved settling times up to four times faster than conventional PI and deadbeat controllers at a rotor speed of 480 rad/s.
  • Improved Accuracy: The controller demonstrated more stable tracking of required torque and current setpoints, with lower oscillations compared to PI and deadbeat controllers.
  • Robustness to Parameters: Simulations showed that the proposed controller's open-loop trajectory remained accurate even with modified parameters (lower inductance), unlike simplified TOC approaches.
  • Low-Speed Equivalence: At very low speeds, the proposed TOC's performance was comparable to the deadbeat controller, indicating its broad applicability.
  • Computational Efficiency: The simplified formula and bisection algorithm allowed for real-time implementation on an industrial DSP, with computational times comparable to or better than existing methods for certain operations.

Limitations and Future Work

  • Approximation for Costate: The paper acknowledges that an approximation was used for the costate equation in the case of different d-q inductances, and its validity was tested via simulation.
  • State Knowledge Assumption: The TOC controller relies on perfect state knowledge. The authors use a Kalman filter to reconstruct states from measurements, but this introduces potential for error.
  • Computational Cost at Low Speeds: For very low speeds, the hyperbolic functions (sinh, cosh) used in the TOC are more computationally expensive than sine and cosine. The authors suggest switching to a deadbeat controller in this regime if computational cost is a major concern.
  • Future Work: The authors suggest exploring the combination of their proposed feed-forward strategy with various feedback solutions and observers to further enhance performance and investigate all potential benefits of this approach. They also propose further study into the optimal setpoint calculation for field weakening.

Summary updated

Abstract

We are concerned with the problem of fast and accurate tracking of currents in the general synchronous drive. The problem becomes complicated with decreasing available voltage, which is common in high-speed and field weakening regimes. The existing time-optimal controllers rely on a simplified model, ignoring stator resistance and differences in inductances. We derive a solution for the general model considering all parameters and show how the parameters affect the current trajectory. One simplifying assumption had to be made, but we show in simulation that it has a negligible impact on accuracy. The simplification allowed for the design of a feed-forward controller that has a low computational cost and can be easily implemented in realtime. We provide experimental validation of the controller on the developed IPMSM drive prototype of the rated power of 4.5 kW using conventional industrial DSP. The controller is compared to conventional PI and deadbeat solutions, demonstrating that the time-optimal controller can reach the required setpoint four times faster than the competitors at the field weakening regime of the drive. The proposed feed-forward control can be seen as a universal building block that can be combined with existing feedback controllers and observers and thus incorporated into existing control solutions.
BibTeX
@inproceedings{smidl2022time,
  author = {Václav Šmídl and Antonín Glac and Zdeněk Peroutka},
  title = {{Time-Optimal Current Control of Synchronous Motor Drives}},
  booktitle = {Actuators},
  year = {2022},
  doi = {10.3390/act12010015},
}
Figure