Recursive Smooth Variable Structure Filter for Estimation Processes in Direct Power Control Scheme Under Balanced and Unbalanced Power Grid

Abstract

This paper presents a new application of the smooth variable structure filter (SVSF) in the direct power control (DPC) scheme to estimate the instantaneous feedback power and negative-sequence (NS) currents. This filter reduces the number of required sensors from six sensors to only two sensors. Moreover, it reduces the noise effect on the estimated parameters. The paper also develops and utilizes an advanced exponential sliding mode control (SMC) to improve the dynamic and steady-state performance of the control scheme. The DPC scheme is utilized to control the injected power of the distributed generation system to the power grid (distribution system) in a grid-connected mode. The presented concept of the direct power scheme is modified to cover the injected power to the unbalanced power grid as well. In this case, the NS current is estimated and its detrimental impact on the injected power is well mitigated. To prove the meritorious performance of this presented control scheme, the utilized filter is compared to Kalman filter, and virtual-flux technique for power estimation. Moreover, the developed exponential SMC is also compared to the adaptive SMC, power-rate SMC, and proportional-integral controllers. The overall performance of the suggested control scheme along with the presented recursive filter is verified using simulation results for the entire power grid in a grid-connected mode. The applications presented in this paper highlight the advantage of using the newly developed filter, SVSF, to accurately estimate some parameters in power systems and consequently reduce the number of required sensors (measuring devices).