Power quality control based on voltage sag/swell, unbalancing, frequency, THD and power factor using artificial neural network in PV integrated AC microgrid

https://doi.org/10.1016/j.segan.2020.100365Get rights and content

Highlights

  • Power quality (PQ) has been assessed for PV integrated microgrid system.

  • An ANN library has been proposed for non-linear PQ parameters in microgrid.

  • Proposed methodology shows better result while compared with conventional methods.

  • Realistic microgrid structure is tested with the effect of communication delay.

  • Effect of line impedance, demand response and off-nominal conditions are analyzed.

Abstract

In the present scenario of microgrid system, conversion of electrical energy has initiated a challenge to maintain the power quality within a satisfactory range. It can be influenced by the voltage deviation, sag/swell, unbalancing, frequency, total harmonic distortion (THD) and power factor as per nature of local loads and the condition of distributed energy resources (DERs). The relationship between power quality and the set of these variables are non-linear in nature. The existing literature show that the above mentioned parameters are not considered simultaneously for the assessment and controlling of power quality in PV based AC microgrid. To minimize the effect of these variables, a novel artificial neural network (ANN) based control approach has been proposed which can control the power quality as per IEEE/IEC standards. The proposed method has shown fast, smooth and stable operation while the performance of the same is verified with that of the proportional–integral (PI) and fuzzy-PI controllers using Matlab-Simulink software. The small size microgrid model is tested with the effect of line impedance and communication delay for the assessment of power quality parameters. This model is extended to a large size realistic microgrid structure for the feasibility of control methodology. The realistic microgrid structure is verified under the analysis of line impedance, communication delay, demand response and off-nominal conditions. The proposed control methodology is validated in a realistic microgrid structure and simulation results are presented to show the performance of proposed controller under different test conditions to identify an ANN library.

Introduction

As a part of the power quality assessment in robust microgrid operation and control, it is realized that with the combination of different distributed energy resources (DERs) there are huge amount of non-linearity in the system. The microgrid comprises the interconnection of DERs as per their geographical locations which increase the number of variables and equations like solar irradiance level, temperature, humidity, weather condition, wind speed, seasons etc. To make stable and reliable operation of microgrid, the power quality should be maintained to the best satisfaction of consumers with lesser burden on power supplier entities or distributed generators (DGs).

The custom power devices (CPDs) such as active power conditioner (APC), active voltage conditioner (AVC), dynamic voltage restorer (DVR) and distribution static synchronous compensator (DSTATCOM) are beneficial to improve the power quality due to the disturbances like voltage sag/swell, interruptions and harmonics [1]. The applications of CPDs are presented in [2] which showed the wide range survey in power quality control in favor of customers’ satisfaction. The three-phase current and power factor could be maintained using static var compensation (SVC) and active filters with the application of reactive power compensation to improve the power quality [3]. In connection to the power quality improvement, a corrective action can be taken by implementing the unified power quality conditioner (UPQC) which compensates the voltage and current harmonics [4]. The power quality disturbances could be detected and classified using sparse signal decomposition on hybrid dictionaries considering different noise levels [5]. In addition, some other methods/controllers that are used for improving the power quality are summarized herewith. An inductively active filtering method, a photovoltaic (PV) fed UPQC, advanced reactive power compensation and an advanced fuzzy power extraction control methods are discussed in [6], [7], [8], [9] which have shown importance of the power quality control at high as well as low level of power systems to the satisfaction of power suppliers and customers.

The concept of microgrid suggests the integration of distributed generating units such as micro-turbines, wind turbines, PV panels, fuel cells etc. within their geographical peripheral. It is realized that a typical microgrid setting, control, management and coordination need to be configured at different levels of DERs and networks [10]. The overall system functionalities of a microgrid could be framed with the hierarchy of network interface, microgrid control, protection and management [11]. The microgrid control strategy could be characterized as local, secondary, global and central or emergency controls. The droop control strategies could be implemented in three ways such as P/f and Q/V droop control for automatic power sharing, adjustment of power sharing capacity of the inverter and DC–DC converter interfacing for power sharing between the energy sources and load bus [12]. A generalized droop control scheme can be extended based on adaptive neuro-fuzzy inference system (ANFIS) for wide range of load change scenarios [13]. A distributed proportional–integral method could be used for the restoration of microgrid frequency while balancing active power sharing with the combination of a finite-time observer [14]. The voltage and frequency deviations could be restored using secondary control strategy for the power electronic based AC microgrid based on inner and outer level of high bandwidth control [15]. The generalized droop characteristic could be applied to control the system voltage and frequency based on demand response (DR) in an isolated microgrid [16]. In addition, there are some optimization tools for power quality improvement such as photovoltaic fed DSTATCOM based on JAYA optimization, forecasting model for off-grid system based on multi-objective optimization, optimization of power factor using predictive data etc. [17], [18], [19].

The power quality and power energy management reliability of a standalone microgrid could be enhanced through optimization in terms of active damping technique, reduction in the size of LC filter and implementing a strong fuzzy logic supervisor [20]. The power quality could be ensured at common bus of loads by implementing a novel cost-function-based decentralized control of unbalance and harmonics for simultaneous compensation of voltage and current as compared to existing virtual-impedance-based method due to nonlinearities of inverters [21]. The power quality could also be improved using adaptive filtering technique including momentum-based least mean square (MLMS) control for nonlinear loads integrated with wind, solar and utility grid along with the battery energy storage (BES) system [22]. By the optimal utilization of BES, the microgrid voltage and frequency could be restored based on hybrid differential evolution optimization and artificial neural network (DEO-ANN) controller considering wide range of disturbances [23]. The optimal dispatch of power sources could be levelized based on the optimum combination of energy from grid, solar and energy storage system (ESS) [24]. Further, the combination of optimal energy and power flow could provide feasible and optimized operating point for both unbalanced and balanced loads in islanded microgrid [25]. The optimum reconfiguration of islanded microgrid could optimize the usage of resources, possibility of maximum demand and minimizing the switching costs among DERs as a multi-objective optimization problem [26] and also, the optimum power quality service could regulate the power quality as an individual standard in different areas as a part of multi-bus microgrid systems [27]. Like master–slave control method, it has the benefits of simple, better voltage and frequency regulation characteristics and dynamic response in microgrid applications. The improved V/f control strategy, frequency-based control, power quality enhancement modeling and control with communication delay, modified droop control for load sharing and peer to peer integration of microgrids based on master–slave control are discussed in [28], [29], [30], [31], [32], [33].

These control strategies deal with the control of DGs, improvement in power quality and performance, power transfer between main grid and microgrid using microgrid central controller (MGCC) and international standards (IEEE/IEC) of microgrid interfacing with utility grid [34]. The radial basis function neural network (RBFNN) is implemented for better power sharing in microgrid system under the operation of non-linear and unbalanced loads. The power quality could be improved by compensating the harmonics and unbalance in voltage and frequency under the governance of RBFNN model [35]. The maximum power point of renewable energy resources is tracked using adaptive neural network (NN). The adaptive NN controls the power exchange between DGs and utility grid (UG); extracting minimum power from the UG to satisfy the load requirement under different operating conditions [36]. The series active power filter is introduced to compensate the voltage and current harmonics and interruptions [37]. A fuzzy logic controller is used to improve the power quality of PV based three-phase three-wire DSTATCOM by maintaining DC link voltage of voltage source converter (VSC) [38].

In addition, an ANFIS-unified power quality conditioner (ANFIS-UPQC) device is investigated to enhance the power quality of microgrid with the consideration of grid and DER side’s disturbances [39]. Further, the RBFNN is also proposed to track the maximum power point (MPP) in PV system to improve the dynamic response of microgrid under different modes of operation [40]. The voltage and frequency of islanded microgrid are maintained to their nominal values by implementing online tuning of control parameters using ANN approach, subjected to sudden change in load [41]. The power balancing and quality could be maintained and secured by providing the standardized communication network under IEC 61850 [42]. The active and reactive power flow, grid-side power factor and THD could be controlled using double-band hysteresis controller as the power quality of grid integrated renewable energy system [43]. The coordination among active power sharing and voltage regulation is controlled using droop control within the interconnected AC and DC microgrids [44]. The fuzzy droop control strategy could also be implemented to design storage system of DC microgrid [45]. The different storage devices could coordinate to each other with the optimal design of model predictive control (MPC) for load frequency control [46].

It has been realized that in most of the research articles such as [37], [38], [39] and [43], the combination of voltage, frequency, THD and power factor as estimating variables have not been considered simultaneously in microgrid application. It is found from the literature that ANN can be implemented for large set of non-linear equations as compared to conventional approach like Gauss–Seidel and Newton Raphson methods. The major contributions and advantages of the proposed control topology are listed below.

  • The simultaneous effects of power quality parameters such as voltage deviation, THD, frequency and power factor have been assessed and controlled.

  • Small size microgrid has been simulated without/with the effects of line impedance and communication delay on the performance of proposed controller during on-grid and off-grid operations.

  • Existence of the proposed controller has been validated through the extended realistic three-phase AC microgrid structure.

  • The performance of the proposed controller has been investigated with the effect of line impedance on power sharing and losses during on-grid and off-grid conditions for realistic microgrid structure.

  • Demand response has been analyzed and the power quality parameters have been controlled with the effect of communication delay on controller action for a realistic microgrid structure. In addition, the off-nominal frequency condition with additive noise has also been incorporated and assessed for realistic microgrid structure.

  • The proposed ANN-based controller can be used as multi-functional device for large size of DERs and loads as stated in realistic microgrid structure due to its flexibility and stability which has not been addressed in several literature such as conventional droop control, distributed proportional–integral method, optimum power reconfiguration optimization tools, master–slave control, BES coordination control etc.

As the first objective of this paper, the power quality is monitored with the help of number of variables such as voltage deviation, voltage sag/swell, unbalancing, frequency, THD and power factor which have not been explored simultaneously in previous literature. The BES system is incorporated as a backup for critical load if there is a deficit in power supply from DER or utility grid. As the second objective, an ANN-based MGCC has been proposed to enhance the overall performance with fast, smooth and stable operation of microgrid due to large number of variables with non-linear equations. In this study, an ANN based MGCC has been proposed to enhance the power quality of microgrid subjected to the set of large non-linear equations and variables. In addition, an ANN library is identified under significant test conditions in re-training of the feedforward neural network to assess the different kind of dynamics. The third objective is that the proposed controller is also tested and validated in a realistic three-phase AC microgrid structure with the effect of line impedance and communication delay in addition to the demand response analysis.

The Section 2 describes the problem definition of the presented work. The system description along with the microgrid components is analyzed in Section 3. The Section 4 discusses the proposed methodology followed by small size and realistic microgrid structures. The results are discussed for small size and realistic microgrid structures without/with the effect of line impedance, communication delay, demand response and off-nominal conditions in Section 5. The conclusion and future proposition of present research work are discussed in Section 6.

Section snippets

Problem definition

The operation of non-linear loads like power electronic based equipment generates harmonics and leads to poor quality of the power. The custom power devices such as DVR, D-STATCOM, AVR, UPQC etc. are used to maintain the power quality for both power suppliers and consumers. These devices are installed to investigate the performance of microgrid system under the disturbances like voltage sag/swell, THD, power factor, frequency and unbalanced system. In microgrid system, the major problems are to

Description of system components

The proposed microgrid system comprises two PVs, BES, inverter, line impedance, loads and utility grid as shown in Fig. 2. The specifications of microgrid system parameters are presented in Table 2 while designing a small microgrid model as presented in the current section. This small microgrid model of Fig. 2 can be extended to a medium size distribution network in the form of realistic microgrid structure as described in detail in Section 4.1.3. The components of microgrid are discussed below

Proposed methodology

In this study, the power quality is enhanced by controlling gate pulses to the inverter circuit using proposed ANN-based MGCC. A 10 kW, three-phase AC microgrid has been proposed to fulfill the critical (4 kW) and non-critical (6 kW) loads. The majority of fulfilling the demand is given to the critical load especially under islanded condition, if grid supply gets failed or subjected to the maintenance. The entire load is specified with the combination of active and reactive powers, that is, 10

Results and discussion

The proposed approach is simulated and tested in Matlab-simulink for three scenarios as already discussed in previous Section 4.1. The results are discussed for small size microgrid without/with effect of line impedance and realistic microgrid structure in Sections 5.1, 5.2 and 5.3 respectively.

Conclusion

The backpropagation function fitting neural network is implemented to maintain the power quality of three-phase AC microgrid system during grid-connected and islanded modes of operation showing satisfactory results. The proposed ANN based algorithm for controlling the power quality of microgrid is presented in detail so that the results can be reproduced by the readers for further research work. In this paper, the power quality is monitored in reference to the variables such as voltage

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgment

The authors thankfully acknowledge the authority of Thapar Institute of Engineering and Technology, India for offering the infrastructural facility for the present research work.

References (65)

  • GowthamN. et al.

    UPQC: A custom power device for power quality improvement

    Mater. Today: Proc.

    (2018)
  • Senthil KumarA. et al.

    Power quality profile enhancement of utility connected microgrid system using ANFIS-UPQC

    Proc. Technol.

    (2015)
  • M. Farhoodnea, A. Mohamed, H. Shareef, A comparative study on the performance of custom power devices for power quality...
  • B.S. Kumar, S. Praveena, K.R. Kumar, Power quality improvement using custom power Devices (AVC, DVR, APC), in: 2018...
  • LiuY. et al.

    Improvement of power quality by using advanced reactive power compensation

    IEEE Trans. Ind. Appl.

    (2018)
  • ManikandanM.S. et al.

    Detection and classification of power quality disturbances using sparse signal decomposition on hybrid dictionaries

    IEEE Trans. Instrum. Meas.

    (2015)
  • LiY. et al.

    An inductively active filtering method for power-quality improvement of distribution networks with nonlinear loads

    IEEE Trans. Power Deliv.

    (2013)
  • DashS.K. et al.

    Power quality improvement utilizing PV fed unified power quality conditioner based on UV-PI and PR-R controller

    CPSS Trans. Power Electron. Appl.

    (2018)
  • LiuY. et al.

    Improvement of power quality by using advanced reactive power compensation

    IEEE Trans. Ind. Appl.

    (2018)
  • BhattacharjeeC. et al.

    Advanced fuzzy power extraction control of wind energy conversion system for power quality improvement in a grid tied hybrid generation system

    IET Gene. Transm. Distrib.

    (2016)
  • HatziargyriouNikos

    The microgrids concept

  • HatziargyriouNikos

    Microgrids control issues

  • F. Luo, Y.M. Lai, K.H. Loo, C.K. Tse, X. Ruan, A generalized droop-control scheme for decentralized control of...
  • BevraniH. et al.

    An intelligent droop control for simultaneous voltage and frequency regulation in islanded microgrids

    IEEE Trans. Smart Grid

    (2013)
  • LouG. et al.

    Distributed MPC-based secondary voltage control scheme for autonomous droop-controlled microgrids

    IEEE Trans. Sustain. Energy

    (2017)
  • HeydariR. et al.

    High-bandwidth secondary voltage and frequency control of VSC-based AC microgrid

    IEEE Trans. Power Electron.

    (2019)
  • F. Habibi, H. Bevrani, Q. Shafiee, Generalized droop characteristic-based demand response and secondary frequency...
  • MishraS. et al.

    Power quality improvement using photovoltaic fed DSTATCOM based on JAYA optimization

    IEEE Trans. Sustain. Energy

    (2016)
  • VantuchT. et al.

    The power quality forecasting model for off-grid system supported by multiobjective optimization

    IEEE Trans. Ind. Electron.

    (2017)
  • HuangC. et al.

    Economical optimization of grid power factor using predictive data

    IEEE/CAA J. Autom. Sin.

    (2019)
  • TidjaniF.S. et al.

    Optimization of standalone microgrid considering active damping technique and smart power management using fuzzy logic supervisor

    IEEE Trans. Smart Grid

    (2017)
  • LiuJ. et al.

    Cost-function-based microgrid decentralized control of unbalance and harmonics for simultaneous bus voltage compensation and current sharing

    IEEE Trans. Power Electron.

    (2019)
  • ChishtiF. et al.

    Development of wind and solar based AC microgrid with power quality improvement for local nonlinear load using MLMS

    IEEE Trans. Ind. Appl.

    (2019)
  • AlshehriJ. et al.

    Power quality improvement in microgrids under critical disturbances using an intelligent decoupled control strategy based on battery energy storage system

    IEEE Access

    (2019)
  • N. Gupta, et al. Cost optimal control of microgrids having solar power and energy storage, in: 2018 IEEE/PES...
  • E.R. Sanseverino, et al. Optimal power flow for technically feasible energy management systems in islanded microgrids,...
  • AbdelazizM.M.A. et al.

    Optimum reconfiguration of droop-controlled islanded microgrids

    IEEE Trans. Power Syst.

    (2016)
  • L. Meng, J.M. Guerrero, Optimum power quality service in multi-bus microgrid systems, in: 2017 IEEE Power & Energy...
  • WangC. et al.

    Improved V/f control strategy for microgrids based on master–slave control mode

    IET Renew. Power Gener.

    (2016)
  • ButicchiG. et al.

    Analysis of the frequency-based control of a master/slave micro-grid

    IET Renew. Power Gener.

    (2016)
  • MortezaeiA. et al.

    Cooperative control of multi-master–slave islanded microgrid with power quality enhancement based on conservative power theory

    IEEE Trans. Smart Grid

    (2018)
  • A. Alfergani, A. Khalil, Modeling and control of master–slave microgrid with communication delay, in: 2017 8th...
  • Cited by (0)

    View full text