Elsevier

Electric Power Systems Research

Volume 155, February 2018, Pages 93-103
Electric Power Systems Research

Intelligent control strategy in the islanded network of a solar PV microgrid

https://doi.org/10.1016/j.epsr.2017.10.006Get rights and content

Highlights

  • Proposed reverse droop and virtual impedance control strategy for the inverter of DG sources (PV units and battery storage), ensured equal power sharing among DG sources and regulation of frequency level in MG network.

  • Proposed power management control strategy, ensured better power balance and co-ordinated control operation of DG sources in MG network.

  • Implemented PSO-based intelligent control strategy in MG power system, ensures the optimized power flow and minimized power variation from the DG sources, and minimized frequency variation as per the assigned control limit during the transient conditions of the islanded MG network

Abstract

In the islanded mode of MG network, due to the lack of inertia and under performance of PI based inverter control, variation in power and frequency level can be expected more during the transient conditions as compared to the grid connected mode. Thus, in this study, a Particle Swarm Optimization (PSO) algorithm with the proposed cost function was implemented in the Grid-support grid-forming (GsGfm) type Voltage Source Inverter (VSI) of the DG sources (solar PV and battery unit) for the islanded photovoltaic (PV) based MG model (built in Matlab-Simulink software environment) to reduce the variation in power and frequency level. The islanded MG model was also implemented with a reverse droop based on virtual impedance control strategy for the VSI and power management control strategy to ensure better power sharing among DG sources with regulation of frequency level and power balance with co-ordinated control operation of DG sources. Implementation of PSO based intelligent control strategy for the inverters of DG units, showed substantial improvement in the performance of controller in terms of reduced settling time and overshoot in power level along with the significant reduction of the power variation and frequency variation during the transient conditions (change in load and change in solar irradiance of PV units). Also proposed PSO control strategy ensured the frequency level to be within the acceptable operating range (Australian network standard) in the islanded MG network.

Introduction

MG provides secure and reliable energy supply to the facilities of critical load and promotes energy independence for the community. MG can be operated in either grid connected or islanded mode of operation [1], [2]. In either mode of operation, it is required to maintain power balance and regulation of frequency level in the MG network [3]. As compared to the grid connected mode of MG operation, maintaining power balance and regulation of frequency level is a big challenge in the islanded mode [4], [5]. A coordinated control operation with equal sharing of power among DG sources and power balance with regulation of frequency level in islanded MG network can be ensured by means of implementing appropriate power sharing and power management control strategies in the MG power system [3]. Similarly as compared to the grid connected mode, variation in power and frequency level can be expected more in the islanded MG network. This is due to the lack of inertia and under performance of PI based controller of DG sources during the transient conditions in the islanded MG network [3]. In such a scenario, implementation of PSO based intelligent control strategy can be considered for the MG model to minimize the variation in power and frequency level during the transient conditions in the islanded MG network. PSO control technique is widely used for MG control applications and most of the research works [6], [7], [8], [9], [10] have been focused on optimizing control parameters of the inverter controller to get an optimized power flow, regulation of voltage and frequency level, improving dynamic response and stability of the MG system. Also some of the research works have been done by using multi objective PSO (MOPSO) [11] and advanced or modified PSO (MPSO) algorithm [12] applied for solving the problem constraints in relation to the energy and operation management of MG.

In most of the research works [13], [14], [15], [16], [17], for power sharing among the DG sources in low voltage (LV) MG network, the reverse droop with virtual impedance control was implemented with either grid feeding (P–Q or current control) or grid forming type (voltage control) VSI of DG sources. However in this study, a reverse droop with virtual impedance (resistor) control has been implemented with GsGfm type VSI for the DG sources in MG model (built in the MATLAB-SIMULINK), since this type of VSI is more flexible and has the ability to be operated in either mode of MG operation without modifying its control configuration. Another novelty in this work is that the PSO algorithm with proposed cost function was also implemented to optimize the power flow and power variation from the DG units and minimize the frequency variation during the transient conditions. The PQ factors in the islanded MG network has been analyzed with the transient conditions like change in load and change in solar irradiance level of the PV units, since both the scenarios are the important factors that could be frequently encountered in a real time MG network and they could cause negative impact over the PQ factors in the islanded MG network.

This paper is organized as follows; Section 2 describes the detail of MG model configuration, control configuration of the VSI, proposed power management control strategies for the MG model, and detail of allowable operating frequency range as per the Network standard, Section 3 outlines the method of optimization which includes the formulation of cost function, setting parameters and process steps of the proposed PSO algorithm, Section 4 demonstrates the optimization analysis, results and discussion, and Section 5 concludes the outcome of this study.

Section snippets

MG model configuration

The configuration of a typical AC MG network model which is operating at voltage and frequency level of 400 V and 50 Hz is shown in Fig. 1. The specification details of each element which are used in the proposed MG model is given in Table 1, Table 2 respectively.

Method of optimization

The PSO algorithm with the assigned cost function was implemented through MATLAB-m-file programme codes in the MATLAB-SIMULINK software environment. The PSO algorithm process was executed through the MATLAB command codes and control gains of the inverter controller in each DG unit was optimized during several run of the MG model simulation.

The PSO control algorithm, applied to the inverter control of DG unit is shown in Fig. 5. Factors like simulation period (per iteration), number of

Optimization, analysis and results discussion

This section includes the detail of results and discussion before and after the optimization for the proposed MG model analysis in island mode with different stage of load change conditions and sudden change in the solar irradiance for the PV units. The analysis results includes the performance factors (overshoot in power level during initial state of simulation, steady state response during the initial state and different stage of load conditions) of the DG inverter controller, optimized power

Conclusion

In this study, a PV based MG model was developed in MATLAB-SIMULINK software environment with implementation of power management and power sharing control strategies. By implementing proposed control strategies, an equal power sharing and better coordinated operation between PV units and battery storage, power balance and regulation of frequency level were ensured during the different stages of load change conditions in the islanded MG network. As compared to the results of PI based inverter

References (27)

  • A. Arzani et al.

    Development of optimal PI controllers for a grid-tied photovoltaic inverter

    IEEE Symposium Series on Computational Intelligence

    (2015)
  • J. Radosavljević et al.

    Energy and operation management of a microgrid using particle swarm optimization

    Eng. Optim.

    (2016)
  • S. Bhattacharya et al.

    Efficient power sharing approach for photovoltaic generation based microgrids

    IET Renew. Power Gener.

    (2016)
  • Cited by (50)

    • A comprehensive review of power quality mitigation in the scenario of solar PV integration into utility grid

      2023, e-Prime - Advances in Electrical Engineering, Electronics and Energy
    • Optimal energy management of distributed generation in micro-grid to control the voltage and frequency based on PSO-adaptive virtual impedance method

      2022, Electric Power Systems Research
      Citation Excerpt :

      The PSO algorithm has many advantages over other meta-heuristic optimization methods that distinguish this algorithm from other algorithms. Features of the PSO algorithm include [6, 7]: The PSO algorithm is a population-based algorithm.

    View all citing articles on Scopus
    View full text