An Adaptive Control of Remote Hybrid Microgrid based on the CMPN Algorithm

Abstract

• The paper proposes a decentralized droop-based control technique power flow through the ILC, meanwhile, to keep the dc microgrid voltage and the ac microgrid frequency within their acceptable ranges. • To overcome the continuous operation of ILC, The modified droop control based CMPN algorithm is used to determine the necessity of operation of hybrid microgrid by estimating the status of the ac and dc microgrids by comparing local measurements with their critical values at both sides. • An adaptive filtering algorithm based on continuous mixed P-norm (CMPN) algorithm is utilized for the proposed autonomous control strategy to insure adaptive tuning of the controller parameters. The variable step size feature of the CMPN algorithm results in fast response compared to other techniques such as robust mixed-norm and the least absolute deviation techniques. The robustness against impulsive noise while having fast convergence, compared to other algorithms, render the CMPN algorithm to be an effective technique for online adaptation of the PI controller used for the ILC, especially in an islanded archetype to maintain the stability of the network. • A comparison between the proposed strategy based on both CMPN algorithm and adaptive PI control under different loading conditions are implemented to validate the proposed methodology. This paper proposes a new adaptive communication-less control strategy to enhance the dynamic performance of the interlinking converter (ILC) of a remote hybrid microgrid. The proposed autonomous strategy is based on the continuous mixed P-norm (CMPN) algorithm to simultaneously maintain the frequency of ac-side and the voltage of dc-side of the ILC within their acceptable ranges under various conditions. The variable step nature of the CMPN algorithm and robustness ensure stable operation of the proposed strategy for the ILC. In addition, the power sharing between both sides of the ILC is achieved by a modified ac-dc droop control to avoid using any communication systems between different distributed generators (DG). Moreover, a simple algorithm is used to determine the desired power flow direction based on local measurements of dc voltage and frequency. The proposed CMPN algorithm provides an online adaption of the controller's parameters to avoid the need for fine tuning or optimization techniques under various operating conditions. The dynamic performance of the proposed autonomous control strategy is evaluated using the PSCAD/EMTDC simulation package under different operating modes. The results demonstrate the superior dynamic performance of the proposed strategy based on the CMPN algorithm compared to the adaptive PI (API) controller in terms of fast response and tight regulation of power flow, voltage and frequency under different conditions. Compared to API, it has been found that the proposed controller results in more improvement of the dc-side voltage within the operating frequency range of the ac-side of the ILC.