Improved Sensorless DPC With Constant Switching Frequency for Non-Ideal Grid
Abinash Rath, S. Gopalakrishna, Monalisa Pattnaik · IEEE Transactions on Consumer Electronics · 2024
In non-ideal grids, DPC methods involving ac voltage sensors have high accuracy, proving their tremendous adaptability. However, the sensorless methods continue to struggle in the same condition as they are unable to locate the source voltage vector correctly. Unfortunately, there have not been any studies on this particular problem, apart from only one article that employs the model predictive control (MPC) for switching. Traditionally, MPC suffers from variable switching frequency and high algorithm complexity. Hence, this work employs a single-stage SOGI-based estimation method compared to conventional multi-stage control systems for the sequence components of the virtual flux vector, resulting in reduced computational complexity. On the other hand, to ensure constant switching frequency, dead-beat predictive control integrated space vector modulation (SVM) is used. Moreover, to improve system accuracy, a comprehensive mathematical model is employed, eliminating any approximations of modulating signals. Additionally, this work also incorporates real-time estimations of supply frequency and filter inductance to safeguard the controller against performance degradation caused by any drift in their value. Comprehensive simulations and experimental tests were conducted on a laboratory-scale PWM rectifier system, conclusively proving the effectiveness of the proposed controller.