Sensorless real-time implementation-based FS-MPCC and deadbeat predictive control with delay and dead-time compensation of PMSM using MRAS and T‐S fuzzy speed controller

In practice, inverter dead time and digital signal processing delay are influential factors in classic predictive control performance. In this paper, delay compensation in finite-set model predictive current control (FS-MPCC) and inverter dead-time compensation in deadbeat predictive control are considered to improve permanent magnet synchronous machine control steady-state error, current and torque ripples, reduce the computational burden, correct inverter dead time-induced deviation, and reduce the sensitivity of parameter uncertainties. Deadbeat control uses a space vector modulation (SWM) block and converts the controller output voltage into duty cycles imposed on the inverter ensuring a fixed inverter switching frequency as opposed to FS-MPCC, which uses a finite set of switching states with a variable switching frequency (without modulation). The second approach looks at how a Takagi–Sugeno fuzzy logic speed controller (TS-FLC) can be applied to operate an intelligent system, offering a great current reference, which is crucial for the design of the FS-MPCC cost function and the deadbeat inner-loop control. A speed estimation observer based on a model reference adaptive system (MRAS) is suggested. By eliminating the encoder or speed sensor, the observer improves system reliability, boosts control performance, and reduces costs. The chosen FS-MPCC and deadbeat controller will be built and used in the laboratory utilizing DSpace.1104. The experimental results comparison shows that both FS-MPCC and deadbeat can be well applied to the PMSM driving system with good speed tracking performance. However, the FS-MPCC can achieve less harmonics in the stator currents and shows advantages in smaller ripples in the mechanical torque.