PMSM vs Induction Motor in MATLAB | Torque Performance and Control Simulation

Video thumbnail Watch simulation overview
matlab projects illustration

Introduction

In the rapidly evolving landscape of electric vehicles (EVs), industrial drives, robotics, and renewable energy systems, selecting the optimal electric motor is critical for achieving high efficiency, superior torque density, dynamic performance, and cost-effectiveness. Two dominant AC motor technologies — Permanent Magnet Synchronous Motor (PMSM) and Induction Motor (IM, also known as Asynchronous Motor) — are frequently compared due to their widespread use in traction and variable-speed applications.

PMSMs offer significant advantages including higher efficiency (up to 95–97.5% vs. 90–93% for IMs), superior power-to-weight ratio, better torque-to-inertia characteristics, no rotor copper losses (eliminating slip-related losses), and excellent low-speed torque production with precise control. These benefits stem from the permanent magnets on the rotor, enabling synchronous operation and high torque density, making PMSMs the preferred choice for modern EVs (e.g., Tesla, Nissan Leaf) and high-performance servo drives. However, they come with higher initial cost due to rare-earth magnets and potential demagnetization risks at high temperatures.

In contrast, Induction Motors remain popular for their robustness, lower cost (no permanent magnets), mature manufacturing, and excellent field-weakening capability for high-speed operation. They are widely used in industrial automation and legacy EV designs but suffer from lower efficiency at part-load, slip-induced losses, and reduced torque density compared to PMSMs.

This project performs a comprehensive torque performance and control simulation comparison of PMSM vs Induction Motor using MATLAB Simulink (Motor Control Blockset and Simscape Electrical). Key aspects analyzed include torque-speed characteristics, efficiency under varying loads, dynamic response to step changes in torque/speed, and control performance using advanced strategies like Field-Oriented Control (FOC / Vector Control) for both motors. Simulations evaluate constant torque and field-weakening regions, maximum torque per ampere (MTPA) for PMSM, and slip compensation in IM. Results highlight PMSM's superior torque response, lower losses, and better efficiency across wide speed ranges, while IM excels in cost-sensitive, high-reliability applications.

This MATLAB-based analysis provides valuable insights for final-year projects, EV drive-train design, motor selection studies, and control algorithm validation, demonstrating why PMSMs are increasingly dominating high-performance applications in 2026.

 

Methodology

The methodology employs a simulation-driven approach in MATLAB/Simulink (R2024b or later recommended) using Motor Control Blockset, Simscape Electrical, and Control System Toolbox to model, control, and compare the two motors under identical operating conditions.

  1. Motor Modeling
    • PMSM Model: Implemented using the PMSM block (Simscape Electrical) or parameterized via Motor Control Blockset. Configurations include Surface-Mounted PMSM (SPMSM) for simplicity and Interior PMSM (IPMSM) for reluctance torque contribution. Key parameters: stator resistance (Rs), d-q axis inductances (Ld, Lq), permanent magnet flux linkage (ΨPM), pole pairs (p), inertia (J), and viscous friction (B).
    • Induction Motor Model: Used the Asynchronous Machine SI Units block with squirrel-cage rotor. Parameters include stator/rotor resistance (Rs, Rr), leakage/stator/rotor inductances (Ls, Lr, Lm), pole pairs, and mechanical parameters.
    • Both motors sized equivalently (e.g., 5–50 kW rating, 400 V, similar base speed) for fair comparison.
  2. Control Strategies Implementation
    • Field-Oriented Control (FOC / Vector Control): Applied to both motors using Motor Control Blockset examples (e.g., mcb_pmsm_foc_qep_f28379d adapted for IM). Includes:
      • Clarke & Park transformations for d-q frame decoupling.
      • Inner current loops (PI controllers for Id, Iq) with anti-windup.
      • Outer speed/torque loop (PI controller).
      • Space Vector PWM (SVPWM) for inverter modulation.
      • For PMSM: MTPA trajectory in constant torque region + flux-weakening (FW) above base speed.
      • For IM: Rotor flux-oriented control with slip estimation/compensation.
    • Alternative: Direct Torque Control (DTC) variant simulated for PMSM to compare torque ripple and response time.
  3. Inverter and Power Electronics
    • Universal three-phase voltage-source inverter (VSI) modeled with ideal switches or detailed IGBT model. DC bus voltage fixed (e.g., 400–600 V). Switching frequency: 10–20 kHz.
  4. Simulation Scenarios
    • Torque-Speed Characteristics: Plotted using mcb.PMSMCharacteristics (for PMSM) and equivalent for IM under current/voltage limits. Includes constant torque region, field-weakening/power region.
    • Dynamic Performance: Step changes in reference torque (e.g., 0 to rated) and speed (e.g., 0 to 3000 rpm), load torque disturbances.
    • Efficiency & Losses Analysis: Computed copper losses, iron losses, inverter losses, and overall system efficiency across operating points.
    • Environmental Variations: Constant load, variable load profiles (e.g., EV drive cycle), temperature effects on magnet flux (PMSM demagnetization risk).
    • Run time: 0–10 seconds with variable-step solver (ode45 or ode23tb) for accuracy.
  5. Performance Metrics & Comparison
    • Torque ripple, rise/settling time, overshoot in torque/speed response.
    • Efficiency map, power factor, current THD.
    • Maximum torque per ampere (MTPA), field-weakening range, base speed, peak power.
    • Visualization: Scope blocks, Dashboard gauges, MATLAB plots (torque vs. speed curves, efficiency contours).
  6. Validation & Extensions
    • Models validated against MathWorks examples and published benchmarks.
    • Parametric sweeps: Vary magnet flux, inductance ratio (Ld/Lq for IPMSM), rotor resistance (IM).
    • Extendable to sensorless control (e.g., sliding mode observer), hardware-in-the-loop (HIL), or real-time code generation.

This Simulink-based methodology enables objective, repeatable comparison of PMSM vs Induction Motor torque performance and control dynamics, supporting informed decisions in EV traction, industrial automation, and renewable drive systems.