VOLTAGE REGULATION AND DROOP CONTROL OF ELECTRIC SPRING USING SINGLE PHASE CASCADED H-BRIDGE INVERTER FOR STABILIZING SMART GRID

This paper proposes “Electric Spring” (ES) based on Single Phase three-level Cascaded H-Bridge Inverter to achieve effective demand side management for stabilizing smart grid fed by substantial intermittent renewable energy sources (RES). Considering the most attractive features of multilevel inverter (MLI), an effective structure of Electric Spring is proposed for suppressing voltage fluctuation in power distribution network arising due to RES and maintaining the critical load voltage. Also, the operation of ES in capacitive as well as inductive mode is discussed. Further, the paper describes droop control method for parallel operation of distributed electric spring for stabilization the power grid. An exclusive dynamic performance of the system using electric spring has been tested and demonstrated through detailed MATLAB simulation.


Introduction
Presently, the power is generated according to the load demand. But due to energy crises and the increase in consumer requirement, generated power by traditional method is not sufficient. So to satisfy the load demand, the penetration of intermittent renewable energy sources (RESs) into the power grid has been increased [1], [2]. The intermittent nature of RESs leads to major problems like the power imbalance, voltage instability, voltage fluctuation, etc. [3]. To overcome these power quality issues various types of FACTS devices like capacitor banks, capacitor reactor, Static Compensator (STATCOM), Dynamic Voltage Restorer (DVR) have been proposed [4]. Using their fast and dynamic controller's power quality issues can be effectively mitigated. However, these devices face the problem of cost, control complexity, etc. Therefore to overcome some of these limitations, a multilevel inverter based Single-phase Electric Spring (ES) is proposed in this paper.ES is basically power electronics based circuit and its concept is derived by implementing hook's law in electric realm. ES is different from traditional conventional custom power devices which improves the voltage profile of the power system. ES is applied to the Load Demand Side for stabilizing voltage fluctuation caused by renewable energy sources (RESs) [5], [6]. ES is low cost, autonomous device and although it is small power devices, multiple distributed electric springs can contribute to the robust stabilizing effect. The ES is connected in series with a non-critical load (Znc) (i.e. The ES is connected in series with a non-critical load ( less sensitive to voltage fluctuation) to form a smart load [7]. To obtain voltage regulation across the critical load (Zc) (i.e. voltage sensitive), the ES generates the compensation voltage or electric spring voltage (Va) which is perpendicular to the non-critical load current (Ia) [8].
Up till now electric spring using two-level inverter topology has been anatomized but this paper proposes ES using multilevel inverter (MLI). MLI topology has emerged recently as a most important alternative in stabi-lizing voltage fluctuation caused in power distribution network [9]. Some of the most attractive features of MLI are, it can generate output waveform which is more close to sinusoidal wave hence low distortion and lower dv/dt, they generate smaller common-mode (CM) voltage as compared to the conventional two-level inverter, operates with a lower switching frequency means lower switching loss and high efficiency Different type of MLI based topologies have been proposed with their several modulation and dynamic controls. One of such effective topology is Cascaded H-bridge inverter (CHMIL) [10]. CHMIL has advantages like it needs less number of components compared with diode clamped (DCI) and flying capacitor inverters (FCI) thus cost and weight of the inverter decreases. CHMLI are used to eliminate the bulky transformer required in case of conventional multiphase inverters, clamping diodes required in case of diode-clamped inverters and flying capacitors required in case of flying capacitor inverters but these require a large number of isolated voltages supply for each cell.

A. Hooks law in mechanical regime
The mechanical spring has been a crucial component in mechanical engineering since Physicist Robert Hooke firstly smoked out its scientific principle in the 1660's [11]. The law gives the relationship between force and displacement. According to Hooke's law force (F) is directly proportional to displacement (x) .i.e. when mechanical spring is stretched or compressed; the force it exerts is proportional to its change in displacement. x The potential energy (PE) store d in a mechanical spring is : The mechanical spring is the elastic device which performs the function like: 1) It provides mechanical support; 2) It stores mechanical energy; 3) It damps the mechanical oscillations. Mechanical springs are widely used in many other applications such as toys, watches, vehicles, beds etc.

A. Hooks law in electrical regime
Similar to mechanical spring, ES in electric regime can perform various functions as follow: 1) It provides electric voltage support; 2) It stores electric energy; 3) It helps to damp the electric oscillation. Analogous to equation (1) the basic physical relationship of the ES is shown as follow: Where q is the electric charge stored in the capacitor of capacitance (C ) of the ES; Va is the electric potential f difference across the capacitor. From equation (4) it can be concluded that by controlling the current (I ) flowing a through the capacitor (i.e. indirectly by controlling the charging and discharging states of a capacitor) operating modes of electric spring can be determined as the neutral, the voltage boosting and the voltage suppression mode. Thus, the ES can be depicted as a controlled current source.

Basic Principle and Operating
Modes of Electric Spring  nc Thus inductive and capacitive modes of electric spring are two different modes similar to the mechanical spring that can change the displacement in either direction when mechanical force is applied. Consequently, the power consumption (P ) of the non-critical load can also be 1 controlled, which is given by equation (6). P and P will follow the profile of generated power (P ).

2 in
Thus electric spring is the special type of reactive power compensator (RPC) developed for decentralized installation and operation. It is different from a traditional series RPC which follows "an input feedback and input voltage control" [12]. This is done by regulating the input voltage (V ) and letting the output voltage to fluctuate

A. Proposed configuration of electric spring
The proposed configuration of ES is shown in Fig. 3 consist of a power source to generate voltage V feeding g transmission impedances X and X respectively. In

F1 F2
order to emulate a renewable energy source a disturbance source, capable of injecting variable active/reactive power is connected to power grid which is connected in parallel to non-critical load (Z ) and critical load (Z ). In nc c this analysis the performance of Electric Spring is evaluated by considering non-critical load (Z ) and  Fig. 4. It consists of four switches S -S 1 4 realized using IGBT/diodes and supported by a dc-link. In order to regulate the voltage at the point of common coupling (PCC) the electric spring voltage changes dynamically to regulate the mains voltage (V ) and also s allow the load consumption to follow the power generation. This system is modeled using MATLAB and system parameters are demonstrated in Table 1.

B. Control strategy for electric spring
The operating principle of ES is different from traditional custom power devices. In the case of traditional custom power devices, output feedback control is used i.e., the output of the system is compared with the reference signal and then processed further; but for ES the input feedback control is used in which input signal is compared with its reference voltage. Fig. 5 shows a simple close loop control scheme derived reference (V ). Their difference s_ref is sent to Proportional-Integral (P ) controller whose

A. Inductive (voltage suppression) mode of operation of electric spring
To test the voltage suppression capability of the ES a disturbance source is programmed at t = 0.2 sec to generate a voltage of 245 V (i.e. 20 V higher than nominal value of 225 V) shown in Fig. 6 (a). In order to suppress the increased line voltage back to its nominal reference value the ES is activated at t = 0.4 sec and voltage of electric spring is changed to stabilize the line voltage as shown in Fig. 6 (b). As soon as ES is activated; it is observed that voltage across non-critical load increases as indicated in Fig. 6 (c). Finally it is seen from Fig. 6 (d)

B. Capacative (voltage boosting) mode of operation of electric spring
To illustrate the voltage support capability of ES the disturbance source is programmed at t = 0.2 sec to generate a voltage of 210 V (i.e. 15 V lower than nominal value of 225 V) shown in Fig. 8(a). In order to restore the line voltage back to its nominal reference value the ES is activated at t = 0.4 sec and voltage of the ES is increased from its near-zero value to about 160 V. As soon as ES is activated, it is observed that voltage across non-critical load decreases to 157.7 V as indicated in Fig. 8

(b)-(c).
Finally it is seen from Fig. 8 (d) that the voltage is regulated back to its nominal reference value i.e. 225V. The corresponding instantaneous value of supply voltage, electric spring voltage and critical load voltage and current are demonstrated in Fig. 9 (a) -(b) in which Fig. 9 (b) clearly indicates that current leads the voltage by 90º. Thus successful implementation of ES in capacitive mode is conducted using MATLAB.
The simulation studies are also carried out using twolevel VSI based ES. Fig. 10 directly shows the comparative performance in terms of FFT analysis of compensated critical load voltage for two-level and three-level ES. It is seen that the three-level CHMLI based ES provides much reduced THD of 1.65% as compared two-level VSI based ES which is 2.19% meeting IEEE 519 harmonic standard. This proves the overall improved performance of ES using multilevel inverter as compared with two-level VSI based ES.

Role of Distributed Mli Based Electric Spring
For parallel operation of distributed ES for stabilizing ac power grid, droop control methodology is employed. Fig. 12 shows four ES working as a group to support and stabilize the ac mains voltage levels at their respective location in distribution network. Since along the transmission line these ES's will be distributed and installed at divergent locations and the exact value of ac mains voltage which ES's are supposed to maintain at their respective nominal reference values are location specific. Therefore to generate respective nominal reference value, control used for ES is modified as shown in Fig. 11. This control scheme allows distributed ES's to have automatically adaptable references voltage according to the value of the main voltage level at their specific location.
The ES is assumed to be located at a distance x from the ac   The intermittent nature of renewable power generation has been identified as a key factor that destabilizes power grid. Such instability problem is expected to become more severe as the amount of intermittent renewable power generation increases. It must be stressed that power system stability must be considered for the design and control of grid connected power inverters. In this section, the recent development of the electric spring technology is summarized. The electric spring concept is proposed to dent the problem caused by the intermittent nature of renewable energy sources such as wind and solar power. This concept can be realized with the help of power electronics technology. Therefore, in this paper the simple close loop control for voltage regulation is proposed. This shows that the system performance of the proposed control scheme is better than the conventional controller in terms of mitigating voltage fluctuations at the common coupling point. The reactive power handling capability of and ES is governed by the product of the voltage it injects and the current flowing through it. In order to enhance it, high proportion of non-critical load (NC) to the critical load (C) .i.e. NC: C = 9:1, easily helps to restored fluctuated voltage back to its nominal value. Thus effectiveness of an ES improves with the proportion of non-critical load.
The other major power quality problem of three-phase power imbalance can be solved by this control scheme. For this purpose two close loop control i.e. reactive power controller and active power controller respectively can be used. By modulating the control for nonlinear load, ES can also achieve harmonic compensation, power factor correction and voltage regulation simultaneously. Further, the inverters of ES can also be connected back to back which not only reduces the size of the hardware but also extends the controllability of smart load in active power.

Conclusion
The paper proposes new approach for regulating the mains voltage using MLI based ES for smart grid applications. The implemented Three-level CHMLI based ES for smart grid application effectively regulates the ac mains voltage and reduces the THD content as compared with Two-level VSI based ES. The effectiveness of ES is validated through digital simulation in terms of THD. Lastly simulation results of droop control for Four Electric springs have been implemented in a large-scale distributed pattern in order to make multiple ES act in coordinating manner so as to have robust stabilizing effect.