Design of BFA-Optimized Fuzzy Electronic Load Controller for Micro Hydro Power Plants

Abstract

With the volatile oil prices and the need to reduce the emission of greenhouse gases, renewable energy resources are a very attractive alternative. Kenya has good potential for micro-hydroelectric power generation. However, only a small proportion of this potential has been harnessed. Further, most rural areas in Kenya have limited access to electric power. Unfortunately, most of the rural communities that are privileged to be connected to the grid have unreliable and expensive grid power. To meet their energy needs, some of the communities have installed and operate pico or Micro Hydro Power Plants (MHPPs) based on the build-own-operate (BOO) model. In most cases, the communities contribute both money and labour to build the plant; the design has to be simple and the cost kept low. After completion, the plants are operated by the community members who have limited technical knowledge and skills. Generally, the tariffs charged for energy supplied by the MHPP are low and sometimes, can not cover the cost of major maintenance work. In any power generation plant, a speed governor is required for regulating the frequency of supply. However, this is the single most expensive component in a MHPP and most of the community owned MHPP can hardly afford it and thus, have no means of continuously regulating frequency. Hence, frequency is controlled manually by controlling the flow of water to the turbine when there is a change in frequency using manually operated water flow control valve. An Electronic Load Controller (ELC) is an electronic device used to control the output power and hence the frequency of a MHPP system, by maintaining a near constant load on the generator. The ELC diverts the power that is not consumed by the consumer load to a damper load which are normally resistors, heating elements or batteries. In this work, consideration was given to the design and implementation of an ELC to control the frequency of a synxv chronous generator output by diverting excess power to a damper load, hence replacing the governor. Magnetostrictive Amorphous Wire (MAW) was used as frequency sensor. The signal from the MAWis a low signal in the order of millivolts. An amplifier and a signal conditioning circuit were designed to convert the analog signal from the MAW to a digital signal that is fed to the Arduino microcontroller. In addition, a Fuzzy-PI ELC was designed and simulated for a MHPP and optimal PI gains were determined by using Bacterial Foraging Algorithm (BFA) to optimise fuzzy logic membership functions. These optimal PI gains were used as reference gains to the Arduino microcontroller so as to generate a Pulse Width Modulation (PWM) signal to control the firing angle of the switching circuit of the ELC. Finally, the designed ELC was tested in a laboratory experimental setup. The results obtained clearly show the effectiveness of the ELC to dump excess power to the damper load when the consumer load changes, and maintain the supply frequency between 49.5 to 50.5 Hz. Also, from the results, it was observed that the MAW sensor was able to measure the frequency effectively with high accuracy when compared to the frequency measurement from the tachometer. This is a key finding since the application of MAW sensor for frequency measurement in ELCs has not been reported in literature. The designed controller is cheaper and yet able to control the frequency very effectively when compared to other ELCs and speed governors that are currently on the market. When the designed ELC is implemented in a typical MHPP, the effects of frequency variation on electrical equipment will be averted. This will promote economic growth especially in the remote communities of Kenya. As a result, the life of the people living in the rural communities will be improved.