A new control method for distortions compensation and power control using microgrid connecting voltage source converters

Abstract

In this paper, a new control method is proposed for microgrid-connected voltage source converters (VSCs), which generally contain voltage and current control loops. The voltage of the point of common coupling (PCC) is analyzed by recursive least squares (RLS) algorithm and its fundamental positive symmetric components are extracted by the voltage control loop. Using the fundamental positive symmetric component, the required active and reactive powers are injected into the main grid via the microgrid-connected VSC without any distortions. This control loop is equipped with two PI controllers for the DC and AC voltage regulation. On the other hand, the same as the PCC voltage assessment, the local load current signal is evaluated by the RLS algorithm in the current control loop, with its harmonics and unbalanced components extracted. Eventually, the reference signal is obtained using these two control loops for generating switching pulses of the microgrid-connected VSC. Accordingly, the VSC acts as a shunt active power filter (APF), and also simultaneously injects the required active and reactive powers to the main grid. The proposed control method has high accuracy and fast dynamic response, and also its performance is independent of system parameters. The effectiveness of the proposed method has been verified by simulation and experiment.

Introduction

A microgrid can be described as a group of interconnected loads and distributed energy resources (DERs) within clearly defined electrical boundaries that acts as a single controllable entity with respect to the grid [1]. A microgrid can be connected and disconnected to/from the grid to enable it to operate in both grid-connected or island modes. One of the typical configurations of microgrids is that the DERs produce their powers in the DC form. Then, after conversion to the AC form through the voltage source converters, it injects them to the local loads or/and main grid. Fig. 1 displays this structure. The main objective of the grid-connected microgrid is transmitting the generated active and reactive powers through VSC, which can be affected by the operating conditions of the AC grid system [2]. Thus, the operating conditions of the grid can affect the performance of the microgrid connecting VSC. To this end, various control strategies for this VSC have been developed during the last decade [3], [4] . The most commonly used control strategies include virtual direct torque control (VDTC), voltage-oriented control (VOC), and direct power control (DPC) [5].

The VDTC strategy was developed as an analogy to the well-known Direct Torque Control used for drives. In this strategy, the converter switching states are appropriately selected by a switching table based on instantaneous errors between the commanded and estimated values of active and reactive powers [6]. The idea is to model the grid as a virtual electrical machine and to estimate the virtual equivalent air–gap flux for control purposes. The estimation is obtained while integrating the measured grid voltage can be used for synchronization purposes and for estimating the power injected into the grid for controlling a microgrid. The idea of VOC is based on instantaneous power theory for decoupling active and reactive powers [7]. This strategy guarantees fast transient response and high static performance via an internal current control loop. Thus, the performance depends upon the current control loop [8]. The VOC implementation is based on the dq transformation. The DPC is a special case of VOC. They both have similar implementation and structures, such as the PWM modulator block and Proportional Integrator (PI) controller [9]. Further, the DPC is based on the instantaneous power theory for decoupling active and reactive powers, and the PLL output angle is used for dq transformation [10]. The studies of these control algorithms have shown that the VDTC have some disadvantages such as requiring high frequency sampling and a high inductance value, and a high THD value in the output current [6]. Moreover, the VOC has complex and heavy calculation burden, and decomposition of active and reactive powers is difficult using this control strategy [11].

DPC and VOC are both based on the transformation between a natural frame abc and synchronous reference frame dq. This strategy promises fast transient response with a high dynamic performance. However, this control method has the same previously mentioned disadvantage of VOC [12], [13]. Regarding power quality, the VOC has the best performance in comparison to other control methods. However, this control method, as with DPC, is based on rotational dq frame calculation, and as a result, it needs a phase locked loop (PLL). While the performance of PLLs is affected by distorted AC grid, the active and reactive powers cannot be separated for an independent control [14], [15]. The Synchronous Reference Frame PLL (SRF-PLL) is the most common type of PLLs. It has good efficiency in normal conditions. However, if the AC grid voltage has significant amounts of harmonics or has unbalanced conditions, SRF-PLL suffers from steady state error in phase tracking, and also shows slow dynamic response in the transient state [16]. To overcome these drawbacks, Delayed Signal Cancellation PLL (DSC-PLL) and Moving Average Filter PLL (MAF-PLL) have been introduced [17], [18]. Both of them have been based on synchronous reference frame theory, and they have also used PI controller for parameters tracking. Thus, they provide a slow dynamic response in the transient state, and their bandwidth is also restricted due to PI parameters. The Second Order Generation Integrator PLL (SOGI-PLL) can improve these drawbacks in non-ideal conditions [19]. On the other hand, due to the low impedance of the microgrid compared to the AC grid, the distortions of local loads tend to flow through the main grid. This can cause some major problems for the main grid such as reduction of the efficiency and capability of transmission lines, etc. Employment of the shunt active power filters or passive filters can resolve these distortions, but they impose heavy costs and complexity [20], [21], [22]. Accordingly, this paper proposes a new control method, which is based on Recursive Least Squares (RLS) algorithm, and involves the voltage and current control loops. The main novelties of this paper are as follows:

A new hierarchical control method has been suggested for a microgrid-connected VSC including primary and secondary control layers. In the primary control layer, the RLS algorithm has been employed for estimating the components of the PCC voltage and local load current. The primary control layer consists of the voltage and current control loops, where in the current control loop the distortions of the local load have been calculated for their compensation. At the same time, the injection of the active and reactive powers has been accomplished using the RLS method in the voltage control loop estimating the positive sequences of the fundamental components of the PCC voltage. Further, the regulation of DC and AC voltages has been accomplished in the secondary control layer.

According to the aforementioned explanations, the proposed control method involves two estimation loops in the primary control layer for the PCC voltage and local loads current. In the PCC voltage estimation loop, positive symmetric sequences of fundamental components are extracted, and then the active and reactive powers are independently controlled and delivered to the main grid by the VSC in accordance with the estimated components. Similarly, in the current estimation loop, the distortions components of the local non-linear load current have been computed. Eventually, these distortions have been compensated by the VSC, which acts as a shunt Active Power Filter (APF). In other words, the microgrid-connected VSC has two roles, delivering active and reactive powers, as well as acting as APF. The proposed control method has many prominent features such as accurate and quick dynamic response under different conditions, independent performance of the system parameters, and high stability. The proposed control method has been validated by MATLAB/SIMULINK simulation, as well as experimental results based on TMS320F2812 DSP.

The rest of this paper is organized as follows. Section 2 introduces the proposed control method. In section 3, the simulation results of the proposed control method have been presented. In this section, the performance of the proposed control method is compared with that of well-known VOC and DPC control methods. Section 4 presents the experimental results using the TMS320F2812 digital signal processor (DSP). Finally, in section 5, the conclusion has been drawn.