Keywords

1 Introduction

Due to the rapid exhaustion of fossil fuels and their effects like environmental hazards, the power systems are now upgrading toward green power generation [1, 2]. The widespread increase of renewable energy sources permits attaining more safe and viable energy prospects to overcome the challenge of growing energy consumption and utilization. The trend of renewable energy power generation is now heading toward generation of power from hybrid renewable energy resources, i.e., solar energy and wind energy.

According to Ministry of New and Renewable Energy (MNRE), the total Installed Grid Interactive Renewable Power Capacity in India as of July 31, 2016, is 44,783.33 MW and by 2022 it is targeted to rise around 175,000.00 MW. The major contributors in renewable energy generations are wind and solar energy [3,4,5].

Wind and solar renewable energy resources are intermittent in nature. A system comprising of wind and solar combined together as hybrid system can be designed to increase the power generation [6]. As these two resources are naturally available in nature, so they are also dependent upon climatic conditions. Wind power output varies with wind speed, and solar power is dependent upon solar irradiance. Hence, the power output keeps fluctuating depending upon the resource available. Therefore to extract maximum power from the renewable energy resources, maximum power tracking algorithms are implemented.

There are variety of MPPT techniques to improve the efficiency of renewable energy systems like perturb and observe [7,8,9], incremental conductance [10], current sweep, constant voltage, distributed MPPT, hill climbing, slide control method, ripple correlation control, DC link capacitor droop control, gauss newton technique, adaptive P&O, curve fitting technique, parasitic capacitance technique, linearization-based MPPT technique, dP/dV or dP/dI feedback control. Nowadays, intelligent MPPT control techniques are opening broad ways in research fields. Fuzzy logic, neural networks, genetic algorithms, evolutionary algorithms, expert systems, swarm intelligence systems like ant colony system, particle swarm optimization, artificial bee colony system, cuckoo search algorithms are some of the new techniques which are evolving very fast for MPPT systems.

2 Hybrid Renewable Energy System

HRES incorporates different renewable energy resources. In this paper, HRES is formed by integrating solar and wind energy conversion system. Both systems are coupled to the DC bus. To obtain maximum power, MPPT control is applied at the converter. DC is converted to AC by inverter which supplies AC to the grid (Fig. 1).

Fig. 1
figure 1

Block diagram of HRES

Full size image

3 Maximum Power Point Tracking

3.1 PV System

The purpose of using MPPT is to make sure that in environmental situation like solar irradiance and temperature changes PV modules are able to supply maximum power. El-Khozondar [11] shows the typical maximum power point characteristic for a PV system. A characteristic curve has been plotted between voltage and current obtained from the PV cell, and the point of maxima of the curve is the point at which the PV cell should operate in order to generate maximum power (Fig. 2; Table 1).

Fig. 2
figure 2

MPPT characteristic of a PV cell

Full size image
Table 1 MPPT techniques for PV system
Full size table

3.2 Wind Energy System

The power output of wind turbine is proportional to the cubic function of the turbine speed. With the changes in wind speed, the wind energy systems are able to supply maximum power. Wei [12] has shown the wind turbine characteristic for wind turbine shaft speed to turbine power and has also obtained the optimal power curve at different wind speed. At the point where turbine power is maximum with respect to wind speed is maximum power point for wind turbine systems (Fig. 3).

Fig. 3
figure 3

MPPT characteristic of a wind turbine

Full size image

4 Results

Sundareswaran [13] shows a comparison of variation of power, voltage, and current between P&O, PSO, and firefly algorithm for a PV system. From the graph, it can be clearly understood that the tracking speed of firefly algorithm is the fastest of all and has highest accuracy. Also, the steady state oscillations are fastest in firefly algorithm (Fig. 4).

Fig. 4
figure 4

Comparison of variation of power, voltage, and current between P&O, PSO, and firefly algorithm for a PV system [13]

Full size image

Abdullah [14] has shown implementation of swarm intelligence technique, i.e., particle swarm optimization to control the duty cycle of the boost converter. The converter provides the interface between wind turbine and the load by extracting maximum power from the wind. As shown from the result, the proposed PSO method has proven to be accurate and efficient to conventional hill climb search method (Fig. 5).

Fig. 5
figure 5

PSO-based and HCS MPPT algorithms simulation results of a the duty cycle and b the power coefficient [14]

Full size image

5 Conclusion

The classical optimization techniques for MPPT have numerous limitations on solving research models or mathematical models. The solution mechanisms of classical approaches are dependent on the type of function, i.e., linear or nonlinear and variable types, i.e., real or integer. The classical methods cannot be applied to problems involving different variable types and different objective functions. The conventional methods strictly follow sequential computations, produce precise answers, and require separate memory address for storing data. Hence, the above several limitations led to use of soft computing-based approaches. The soft computing strategies involve intelligent computational steps, and hence, computational time required is less.

Fuzzy systems and neural network systems relate the human way of thinking and interpretation. The genetic algorithms are related to the biological process in which the systems improve with time. The algorithms which are based on swarm intelligence techniques like ant colony optimization, particle swarm optimization, artificial bee colony optimization, cuckoo search algorithm imitate the intelligent behavior involved in animals, birds, insects, and microorganism. The particle swarm optimization is a special category of firefly algorithm. Many of the research papers have analyzed that firefly algorithm is superior in solving complex optimization problems. The firefly algorithm is more powerful in finding global optimum value in very less computing time. Due to some significant features, the cuckoo search algorithm provides faster convergence speed since it is based on Levy’s flight equation.

The mentioned significance of the new optimization approaches can be used for maximum power point tracking systems in wind and solar energy systems. The cuckoo search algorithm and firefly algorithms can easily handle the shading conditions by their intelligent control approaches.

Apart from these research approaches, a combination of neuro-fuzzy, fuzzy genetic, and neurogenetic systems has also evolved to open new gateways for research. Each technique has their own merits and provides an efficient solution to the problems on different domains. This combination makes the system based on self-learning and decision making.