I. Introduction
Electricity is used extensively in sewage treatment plants (STPs). Recently, rising electricity costs and stricter environmental regulations have compelled the municipality to reduce electricity consumption in the STP. RE technologies such as pico-hydropower systems can convert STPs into RE energy producers by reclaiming electricity via a water turbine. According to the literature, the sewage treatment facility has a continuous water flow at the final effluent that can generate electricity via a small hydropower system [1], [2]. The estimated volume of wastewater generated by municipal and industrial sectors is 2.9 billion cubic metres per year, according to the National Water Services Commission's Malaysia Sewage Industries guideline 2009 [3]. As is well known, many controller methods are available for use in process control today, ranging from the most basic conventional PID controller with the basic system mathematical model to the most advanced Self-Tuning Fuzzy PID hybrid (STFPID) or PID Neural hybrid where their system is represented by the Auto-regressive with Exogenous Input (ARX) model structure (obtained through MATLAB System Identification Toolbox). There hasn't been a lot of research done specifically for PHPGS controllers in STP. For example, H Beltran simply used conventional power electronics and a microcontroller to control the PHPGS in STP [4]. K.-J. Chae just mentioned the use of a standard programmable logic controller to adjust the turbine blade angle, which allows the highest water level in the forebay tank to be maintained by regulating the flow passage in accordance with the incoming flow rate [5]. However, much research has been conducted for PHPGS controller in general application. Shailendra Kumar Rai's research proposed an Electronic Load Controller (ELC) with synchronous generator, three-phase controlled bridge rectifier, and IGBT as chopper converter, modelled in MATLAB, Simulink, for controlling the frequency of a micro hydro plant under varying load conditions. The controller was modelled in this paper, and the results were analysed after simulation [6]. Suhas Vilas Kamble's research paper addressed a comprehensive literature survey on load frequency control in micro hydro power plants and presented a design of a fuzzy logic-based controller that would manage nearly the entire operation of the MHP [7]. A paper by Rojesh Dahal discussed the development of a load controller using a proportional type controller to improve the performance of droop-based load controllers in interconnected micro hydro power plants [8]. The paper by RuiXiang Fan discusses the design of a fuzzy PID controller that uses a two-dimensional fuzzy controller. The area control error and regional control deviation change are fed into this controller, and the output is a PID controller parameter variation KP, KD, and KI. When there is an input deviation caused by a load disturbance or an offset of system parameters, the controller will revise the PID parameter online using fuzzy control rules and fuzzy reasoning to meet the different demand of regulating parameters. This will control the output and limit load disturbance and offset of system parameters, allowing the interconnected grid's Automatic Generation Control system to provide a smooth dynamic response [9]. Gilfred Allen M. Madrigal implemented the Fuzzy controller using the Arduino or maintaining the required frequency of 60 Hz within 10% tolerance for Hybrid Off-Grid Pico Hydropower System in his research work. The flow of water can be controlled automatically using a servo motor and a frequency controller [10]. Supriyanto Praptodiyono has also designed and prototyped the ELC for the Pico Hydropower Plant (PHP), which has been tested on a laboratory scale. The frequency sensor and gating system are the most important components of an ELC. The first component is a Zero Crossing Detector, which detects the frequency of the generator [11]. Furthermore, Mohd Mustafa's research has focused on the battery management system in order to improve the reliability of the used solar system. The goal of this research is to create a hybrid power production system that combines energy battery storage PV-wave renewables and an effective power control method to meet load requirements. The results show that despite variations in produced hybrid power and required load power, a controller can maintain a constant voltage DC-link [12]. As based on the above-mentioned research papers, we cannot deny that the Conventional PID controllers suffer from the limitation of tuning parameters for giving optimum control of frequency of PHPGS. This research paper focus on the development of new intelligent control method for PHPGS in Malaysia's STP which are not yet implemented. Next, Renuka S. Rasal used MATLAB to model and simulate a grid-connected hydroelectric power plant in his research. Nonlinear models of hydro turbines, synchronous generators, exciters, and PID governors are used for this purpose. The simulation was based on data collected from the Jayakwadi hydropower plant in Maharashtra, India, which has a capacity of 12MW and a net head of 37m. The results of a three-phase to ground fault simulation at the Synchronous Generator terminal are also presented and discussed. For the dynamic analysis of the system, a simulation for three phase to ground fault at the synchronous generator (SG) terminal was performed. After the fault is cleared, the simulation results show that the steady state condition is restored. However, using a governor system to control the incoming flow rates level into the turbine will complicate and increase the cost of the overall system. Furthermore, as the mechanical governing system is used, the response of the PID governor to changes in voltage and frequency is slower and less accurate. Furthermore, when the system dynamics or disturbance characteristics change, the fixed conventional PID governor controller is unable to modify its parameters compared to the hybrid controller such as coordinated PID-Fuzzy controller. The use of synchronous generators on this research works will also consume additional power for its excitation system, increasing the circuit complexity [13].