Renewable power interface based rural telecom

Received Sep 6, 2018 Revised Nov 19, 2018 Accepted Feb 15, 2019 To power remote area telecom where grid supplied electricity is unreliable, an alternative exists in the form of renewable energy sources. These sources mainly rely on the weather condition of a particular area for their source of power. To overcome such problem, hybridization of energy sources are prefered. Hybridization of solar-wind systems use a separate converter for each source thus leading to a complex, bulky and less efficient system. This paper presents a single renewable power interface which allows solar and wind sources to supply the load separately or simultaneously depending on the availability of the energy sources. It uses cuk and sepic converter topologies to form cuk-seic interfece. These systems also use battery backup to store excess energy and to make best use of their operating characteristics. Performance comparison between a cuk-sepic renewable power interface with and without MPPT using MATLAB/SIMULINK is presented. A detailed componentwise analysis is also done to calculate best efficient interface. It is found that the cuk-sepic interface with MPPT is efficient. A lab level model for cuk-sepic interface is developed and tested. The experimental results proved the effective operation of a single renewable power cuk-sepic interface with MPPT.


INTRODUCTION
India is the world's second largest telecom subscriber base of 700 million and it may come to 800 million in coming years. 24x7 power supplies are required to operate the telecom equipment. However due to lack of stable grid power in rural or remote areas, telecom operators face a big challenge. So, telecom equipment are running on diesel generator (DG) or batteries during demand period which leads to a threat of global warming and CO2 emissions. This increases the focus on powering telecom towers through renewable energy sources like solar, wind, tidal etc. Yousef Mohmoud et al. [1], and Pradhan et al. [2], have discussed the photovoltaic modelling and simulation of solar arrays. Nath [3] and Bharani Kumar [4] have reported the modelling and simulation of wind turbine. According to both the authors, the components in solar and wind systems are modular which makes the design and installation very feasible and practical.
However the difficulties like intermittency and instability etc., associated with the renewable energy sources, combined utilization of renewable sources with proper battery backup are preferred. Ersan Kabalci [5], Xiong Liu et al. [6], Salameh et al. [7], and Billington [8] suggested a viable independent power industry that serves the public interest by providing cost effective power supply through the environment responsible and efficient development of the available solar/wind/battery energy sources. When a source is unavailable or insufficient in meeting the load demands, the other energy source can compensate for the difference. When a source produces excess energy than load demand, it can be stored in battery. When neither the wind nor the solar systems are producing, power can be supplied by the batteries. So, to Providing a good match between these two energy sources are also a complicated task. Hence DC-DC converters are used whenever change of dc power from one voltage level to another voltage level is required. As reported by Neng Zhang et al., Akar et al., Shahar Banu et al., Narendra Nath et al., [9][10][11][12] the interface systems for renewable energy sources were constructed using the topology which consists of multiple energy inputs connected to a respective DC-DC converter with its own controller and storage element. Each DC-DC converter is connected to a common dc link. The common dc link is where the energy from each input is combined and then distributed to the loads through another DC-DC converter for each load. In a system of this nature [13][14][15], each DC-DC converter communicates with each other in order to control the power transferred from each input to its respective output. With this type of topology if one were to add multiple input modules, the overall system would become too large for its particular application where size and weight are an issue.
According to [16][17][18][19][20]  It is more advantageous to use integrated converters with multiple ports instead of several individual converters as it results in less component count, lower cost. It can deliver power to the load either simultaneously or individually based on the battery utilization. So this paper proposes a single cuk-sepic integrated converter based hybrid solar-wind renewable energy sources powering rural telecom. As reported in the literature [21], the use of a maximum power point tracking technique (MPPT) algorithm is also necessary to extract as much power as possible from the solar and wind when its irradiation and speed changes. Existing literarure [22] also dealt about PI controller, MPPT algorithm for multiple renewable energy sources with separate DC-DC converters. This paper proposes a MPPT algorithm for hybrid solarwind system with a single cuk-sepic integrated DC-DC converter. Most common method to achieve MPPT in solar and wind systems is perturb & observe (P&O) algorithm.In this paper, the perturbation and observe (P&O) method is used for both the solar and wind energy sources.
This paper is organized as follows. The first section is a short overview of the literature. In the second section block diagram, system description and simulation results of single cuk-sepic interface based solar-wind hybrid system powering rural telecom with MPPT and without MPPT are presented. Third section deals about the simulation results and efficiency comparison. Fourth section contains the hardware description and results. Finally, the findings of the investigations are highlighted in the conclusion.

BLOCK DIAGRAM
The block diagram of the cuk-sepic interface without and with MPPT for powering rural telecom is shown in Figure 1. It contains two generating power plants, a solar system and a wind system. AC power obtained from wind source is rectified into dc power using controlled rectifier circuit and connected to the interface circuit. Solar power is directly connected to the interface circuit. Both the solar and wind systems are incorporated together using a single renewable power interface circuit. The load is supplied mainly from wind system during winter time whereas solar system is supplied during summer period. The output of cuksepic interface is used for charging the battery that supplements power to the telecom load.

. Solar system
Solar cell is the basic element of a solar photovoltaic system. A solar cell is a semiconductor diode whose p-n junction is open to the light. When sunlight strikes the solar cell junction, free electrons and holes are generated and current is delivered to the load when it is short circuited. A grouping of solar cells forms a solar panel. To obtain large output voltage solar panels are formed by connecting solar cells in series and to achieve large output current cells are connected in parallel. Figure 2 shows the equivalent circuit of a solar cell. The basic equation that describes the I-V characteristics of the solar cell is given as, Where Vt : Thermal voltage Rs &Rp : Equivalent series and parallel resistances The incident light current of solar cell depends on solar irradiation. It can be expressed as, Where Is,n : Incident light current at nominal conditions(25οC and 1000W/m2) ΔT-T-Tn : (difference between actual and nominal temperature) G and Gn : actual and nominal irradiation respectively Where Eg : band gap energy of the semiconductor Ion : Nominal saturation current Based on the above mathematical equations Simulink modelling of solar module is done and Figure  3 shows the IV characteristics of solar system. From Figure 3, it is clear that the open circuit coltage is 80V and the short circuit current is 19A. So, the total solar power considered for the analysis is 1.5kW. Figure 4 gives solar current and solar voltage and solar power of the same.

Wind system
Wind system is one of the promising, environment friendly, abundant available sources of an alternative energy. The wind system can consist of a wind turbine coupled with a permanent magnet synchronous generator (PMSG) and a rectifier circuit or a wind turbine coupled to a dc machine to get dc voltage. Wind turbine converts the kinetic energy of wind in to mechanical energy. The magnitude of this converted mechanical energy depends on the air density and the wind velocity. The wind power (Pm) that is developed by the turbine is given by the equation, Where Cp : Performance coefficient of turbine ρ : Air density (kg/m3) A : Area of turbine blades (m3) W : Wind velocity (m/sec) λ : Tip speed ratio of the rotor blade tip speed to wind speed β : Blade pitch angle (deg) The coefficient Cp is the fraction of kinetic energy which is converted by wind turbine into mechanical energy. It is related to the tip speed ratio (λ). Wind turbine output torque (Tm) can be calculated using equation, Using the above equations a Simulink model is implemented and Figures 5 and 6 shows generator speed vs mechanical output power (pu) characteristics of the wind system and wind voltage, current, power obtained from simulation.

P &O algorithm
Modeling of perturb and observe (P & O) maximum power point algorithm is discussed here. Renewable energy sources are a form of unstable energy. So, it is necessary to track the maximum power point (MPP) continuously in order to maximize the power output from the renewable energy sources. There are several MPP tracking methods in the literature, such as Perturb and Observation (P &O) algorithm, Incremental Conductance (INC) algorithm, neural network control, fuzzy logic control and digital signal processor based implementation. This work uses Perturb and Observe (P and O) algorithm are due to its least cost implementation.
In P & O algorithm, a small perturbation is introduced in every iteration to alter the duty cycle in order to force the operating point to move near the maximum power. This algorithm compares the power of the previous step with the power of the new step to determine the next perturbation. If the power increases due to the added perturbation then the perturbation will remain in the same direction. If power decreases due to the added perturbation then the perturbation reverses the direction. When the steady-state is reached, the algorithm oscillates around the peak point. In order to maintain small variation, the perturbation size is kept very small. The working of P&O algorithm is illustrated in the Figure 6(a). Comparison of solar output power with and without MPPT, wind output power with and without MPPT is shown in Figure 6(b) and 6(c) respectively. From the figures, it is explicit that the cuk-sepic converter with MPPT produces more renewable power.

Design of cuk-sepic interface topology
The interface is operated in continuous conduction mode of operation, the average voltage of the converter can be calculated as, The topology of cuk-sepic interface topology without and with MPPT is shown in Figure 7a and 7b respectively. It is also the combination of cuk and sepic converters. In order to reduce component and increase efficiency, components of cuk and sepic converters are arranged in such a way that the common elements are shared between themselves.  Figure 7(a) and 7(b), cuk converter is used to process power from solar panel, sepic converter is used to process the power from wind system, both the outputs are connected to a common dc link which is used to power the load, therefore it is suitable for systems with low voltage rating like telecom. It allows both simultaneous and individual operation of solar and wind sources. Both the circuits are simulated with the parameters, L1 of 158 μH, L2 of 84.16 μH, L3 of 108.6 μH, C1 of 2906 μF, C2 of 1132 μF and C3 of 1200 μF resistive load of 1.92Ω and the switching frequency of 20 kHz for its investigation. It is desired to obtain -48V and 23 A from an inputs of 80V and 50V. Battery power can be used as backup power which can compensate load demand when needed. The input and output voltage relationship of the cuk-sepic interface topology is derived by applying volt-second balance on inductors L1, L2 and L3. Inductor L1 gives, On inductor L2, By rearranging the equations (14), (15) and (16), the output voltage expression is obtained Where, V1= Vs=Solar voltage V2= Vw=Wind voltage Vc1, Vc2 and Vc3= Voltage across Capacitors C1, C2 and C3 d1= dpv=duty cycle of switch S1 d2= dw=duty cycle of switch S2

OPERATING MODES
Depending on the conducting states of power switches S1 and S2, there are four different operating modes. They are (i) Mode-1 (S1 ON, S2 ON) , (ii) Mode-2 (S1-ON, S2 OFF) (iii) Mode-3 ( S1-OFF, S2 ON) and (iv) Mode-4 (Both S1, S2-OFF). The voltage and current waveforms of cuk-sepic interface topology with MPPT is given in Figure 8. There are four different modes of operation of cuk-sepic interface with MPPT depending upon the conduction states of the switches S1 and S2.The working principle of different modes of the same are given below.
(i) Mode-1: Figure 9(a) shows the current direction of inductors (IL) when both switches S1and S2 are conducting. The diodes D1, D2 are reverse biased. The power delivered from solar charges the inductor L1 and the wind power charges inductor L3. Energy stored in inductors L2, and C3 is transferred to C2, battery and load.
(ii) Mode-2: In this mode switch S1 conducts, switch S2 is in OFF position. The inductor current directions are shown in Figure 9(b). The diode D1 is in reverse biased and D2 is in forward biased. The power delivered from solar charges inductor L1, energy stored in inductors L2 is transferred to the capacitor C1 also charges C2. Energy stored in L3 is transferred to C2 and C3 which charges the battery and load.
(iii) Mode-3: In this mode swich S1 is in OFF position and S2 is in OFF position. The diodes D1 is forward biased and D2 is reverse biased as shown in Figure 9(c). Energy stored in inductor L1 is transferred to (iv) Mode-4: Figure 9(d) shows the equivalent circuit for this mode. In this mode both the switches S1 and S2 are not conducting. So the diodes D1 and D2 are forward biased. Energy stored in inductor L1 is transferred to capacitor C1. Energy in stored in inductor L2 and L3 are transferred to C2 and C3 which feeds the load and charges the battery. Figure 9(a). Mode-1 S1, S2-ON Figure 9(b). Mode-2 S1-ON, S2-OFF Figure 9(c). Mode-3 S1-OFF S2-ON Figure 9(d). Mode-4 S1, S2-OFF

DISCUSSION ON SIMULATION RESULTS
In order to predict the performance of the cuk-sepic interface topology a detailed MATLAB/SIMULINK based simulation is carried out on both with and without MPPT. The output voltage is regulated at -48V and current at 23A in both the cases. Figure 10 shows the voltage and current through the loadin both the cases. It is observed that the voltage across both the topologies are measured as -47.8V and the current through both the topologies are measured as -23A. Figure 10. Voltage across and current through the load Solar, wind power with MPPT and without MPPT are shown in Figure 11(a) and 11(b). Figure 11(a) exhibits that the topology with MPPT is having increased solar power. From Figure 11 Table 1 shows the comparison of efficiency of both the cases. Efficiencies of the cuk-sepic interface without and with MPPT are found to be 89% and 93%. Thus it can be concluded that cuk-sepic interface with MPPT is more efficient.

HARDWARE IMPLEMENTATION AND DISCUSSION
Hardware implementation of the cuk-sepic interface topology with MPPT is done with the two sources each of 12V. It consists of power supply circuit, gate pulse generation circuit and cuk-sepic interface topology. They are explained in brief below. The design of cuk-sepic interface is given below: 12 Volt DC power supply, 2 Amps (Source 1), 12 Volt DC power supply, 5 Amps (Source 2), Optocoupler circuit with 15 Volt power supply, PIC18F2550 with 5 Volt power supply, Inductor L1-2mH, Inductor L2-3mH, Inductor L3 -2mH, Capacitor C1-2200 µF, 100 Volt, Capacitor C2-1200 µF, 100 Volt, MOSFET-IRF840. Figure 12a and 12b Shows the design of cuk-sepic interface and gate pulse generation circuit in which the Mosfet switch is triggered by the pulses generated by the PIC controller. The pulses are made to trigger the switch through the optocoupler circuit. The hardware results of the cuk-sepic interface recorded in DSO are given here. Figure 13 shows the output voltage of 12V for an input of source voltage 12V, 2A and 12V, 5A.

CONCLUSION
A hybrid wind and solar energy system with single cuk-sepic converter topology is used. This topology allows the two sources to supply the load separately or simultaneously depending on the availability of the energy sources. The output voltage obtained from the hybrid system is the sum of the inputs of the cuk-sepic converters. This system uses less number of components and improved efficiency. This paper also presents the comparison of efficiencies of cuk-sepic interface topology with and without MPPTcontroller. The use of MPPT controller further improves the efficiency than the basic one without MPPT controller. From the analysis it has been found that the efficiency of cuk-sepic interface topology with MPPT has higher efficiency and lesser losses. Solar and wind power are also plotted. Switching losses and conduction losses of cuk-sepic with MPPT is comparatively reduced. So, cuk-sepic converter with MPPT is found to be very attractive in terms of efficiency, component size, and ripples.