Contents I. Introduction
Using fossil fuels to generate electrical energy has negatively affected the environment [1]. To remedy this problem, renewable energy sources (RESs) such as wind, solar, hydro, biomass are suggested [2]. Recently, solar energy-based photovoltaic (PV) array has received a great attention especially in remote areas due to many reasons such as: 1)PV can provide clean and safe energy; 2)energy (sunlight) is totally free and abundant; and 3)their operating and maintenance costs are considerably low [3]. Unfortunately, this RES possesses some drawbacks related to its performance during the night and the change in climatic conditions [4]. However, additional elements such as the battery energy storage system (BESS) and the dump load are required to ensure uninterrupted power supply and power balance in standalone system (SS) [5]. Furthermore, two-stage converters are suggested to obtain the maximum of power from the PV array and to regulate the voltage and the frequency at the point of common coupling (PCC). Regarding the maximum power production from the PV array, several maximum power point tracking (MPPT) control techniques have been proposed in the literature to achieve this objective by controlling the dcdc boost converter. According to [6], the conventional control techniques such as incremental conductance (IC) and perturbation and observation (P&O), which are widely used, suffer when the PV array is subjected to rapid irradiation change. On the other side, it has been reported in the literature [7] [12] that the nonlinear controller, which is based on adaptive laws, energy balance, and sliding mode approach (SMC), can ensure system stability and fast dynamic response. Furthermore, it is characterized by its robustness against parameters variations and easy to implement. Recently, it has been reported [13] that the use of linear sliding surface and the appropriate adjustment of the parameters of the linear sliding surface disturb the system stability. To solve these problems and to obtain finite time convergence and higher control precision, terminal SMC (TSMC) is proposed in [14]. This solution itself possesses two disadvantages [15]: 1)singularity point and 2)the requirement of the bound of the uncertainty. Therefore, extra solutions such as nonsingular TSMC (NTSMC) [16] and uncertainty estimator [17] are required. These methods and their extra solutions have been suggested for specific applications such as mobile robot [18] [20] or an aircraft control as well electric vehicle control [21], [22]. But for applications, such as control of PV array [23] or control of electrical machines [24], when the chattering is not an issue, SMC is more suitable because it is less complicated and easy to implement in real time. Generally, to get perfect results in real time using NTSMC or adaptive sliding mode, high switching frequency is required (200kHz) [25], which is impossible when the limit is 10kHz.
