Sheikh Izzal Azid

Battery storage systems (BSS) are critical for maintaining stability and flexibility in renewable energy microgrids. However, the generation and load profiles introduce challenges for optimal energy management. This study presents a deep reinforcement learning (DRL) approach to real-time battery scheduling in a grid-connected microgrid. The control problem is modeled as a Markov decision process (MDP), and a Deep Q-Network (DQN) agent is trained to output optimal charging and discharging actions based on system states. Results indicate that the proposed DQN-based EMS achieves superior performance in balancing economic cost, energy self-sufficiency, and battery health compared to conventional rule-based strategy. This study contributes to advancing reinforcement learning-based approaches for real-time battery scheduling in microgrid energy systems.

This paper develops a three-parameter simple fractional-order controller to manage load frequency in a multi-area interconnected multisource power system (PS). The novel suggested controller has only three parameters to tune, like the classical PID structure. The suggested controller is tested in a two-area interconnected PS with thermal, gas, and hydraulic power plants and renewables (wind and solar power). The new controller's effectiveness is proven by comparing its performance to fractional-order proportional-integral controllers and tilted integral derivative controllers from the recent literature. The numerical findings show that the proposed controller considerably enhances system frequency stability under diverse load variations, uncertainties, fundamental limits, propagation delay, and high renewable energy source access.

Grid-forming inverters play a vital role in facilitating the incorporation of distributed renewable energy sources into microgrids by regulating the AC voltage output of inverters. Traditionally, LCL filters are preferred due to their exceptional ability to attenuate harmonics. However, the conventional control approach for grid-forming inverters faces challenges in achieving desired performance due to interactions between control loops and low bandwidth limitations. To overcome this issue, this study proposes using an integral full-state feed controller that is tuned using the linear quadratic regulator approach. Moreover, implementing a full-state feedback controller requires measuring various parameters such as inverter side current, capacitor voltage, grid side current, and point of common coupling voltage. However, the integration of these measurements adds complexity and increases costs in system design. To address this challenge, a proposed solution called the Generalized Extended State Observer eliminates the requirement for current sensors on the inverter side and voltage sensors at PCC. The effectiveness of this control scheme is demonstrated through MATLAB/Simulink simulations.

This article addresses the problem of poor dynamic performance of conventional virtual synchronous generators (VSGs) when switched to grid-connected mode (GCM) due to varying grid conditions. The proposed research involves an analysis of the coupling between active and reactive power and its dependence on grid conditions, leading to the proposal of an adaptive VSG (AVSG) control strategy based on online grid impedance estimation. The Recursive Least Square (RLS) method is proposed for real-time grid impedance estimation, allowing the AVSG to exhibit superior performance compared to conventional fixed-parameter-based controller designs. The effectiveness of the proposed control scheme is demonstrated by using the Simscape electrical library in MATLAB/Simulink. The major findings indicate that the AVSG ensures robust operation under both strong and weak grid conditions, with specified settling time and damping ratio and minimal overshoot. The conclusions reached demonstrates the effectiveness of the AVSG control strategy in improving the dynamic performance of VSGs in GCM, with potential for wider applications in renewable energy systems.

This paper verifies a fractional-order PI controller to stabilize integrating time delay plants. Firstly, the stability region by non-integer order controller namely, PI λ is illustrated using boundary condition. This is achieved using complex root boundary domains. The controller parameters are calculated using the ITAE criterion for integrating plant with time delay. The twin rotor MIMO system (TRMS) as an integrating plant is verified in our study. Another example with large time delay is studied to check the validity of fractional-order integrator. Both simulation and experimental results are discussed and compared by PI λ with classical PI. The result proves the effectiveness of the fractional order controller.

The recent decrease in the price as well as size of semi-conductor logic and due to significant advancements in technologies such as microcontrollers, motors and sensors, the application of quadcopters in several fields has been achieved. However, testing of quadcopter prototypes still has a risk of damage due to faults and unexpected behavior. Hence, a method of testing of quadcopters in simulation mimicking the actual conditions of the real environment in an actual hardware test has been proposed. For this purpose, Gazebo simulator integrated with ROS has been chosen for the simulation of the path of the quadcopter. Moreover, the software Matlab/Simulink has been interfaced with Gazebo in order for the simulation of the quadcopter to be achieved.

This paper describes a new method for dynamical estimation of load disturbance in induction motors by using Nonlinear Unknown Input Observers (NUIO). This estimation is then used to compensate dynamically the load torque in a Field Oriented Control (FOC) induction motor drive to increase its load-rejection capability. The method has been verified both in simulation and experimentally on a experimental rig.

This paper presents a new technology for designing, simulation, implementation and validation of a driverless vehicle system with the use of the Arduino microcontroller, ultrasonic sensors and stepper motors are integrated to Lyapunov based control scheme. The goal is specifically to develop a driverless vehicle system which is able to move from an initial to target location in an obstacle free environment in 2D workspace while taking the shortest route to the destination. Mathematical modeling for the vehicle has also been done in order to find the dynamics of the vehicle which would be used in the software simulation of the vehicle and subsequently verified through experimental design. The new methodology for the prototype and validation are presented with interesting scenarios. The final design and the methodology to reach the final design, prototyping, testing and verification and validation of the output will be presented.

This era has witnessed a pedagogical shift in the area of ICT driven education. While the developed countries have it and use it in every aspect of their lives, the developing countries desperately need the ICT tool for its students. Due to the geographical location of the Pacific Island countries, there is a genuine need to design and develop tailor-made Intelligent Classrooms. Hence, another new initiative of USP is to invest in the Intelligent Classroom would be located in all the USP regional campuses with networking of intelligent sensors. This paper presents the prototype of smart classroom which ultimately scaffold to Intelligent Classroom. The intelligent classrooms would be interfaced such that the distance student/staff would be able to interact actively in real time. Adding live-blog for students during lectures to discuss and share ideas with other the students in the same lecture throughout the region. Through live-blog students in small number class in some campus would not feel isolated. Smart timetable is also linked where students will be able to view lecture times via Mobile app, or even the lecturer can book the classroom using app. Moreover human action recognition and voice to text is also interfaced to ease load on lecturer and convert lecture into subtitles preferred to students.

This era has witnessed a pedagogical shift in the area of ICT driven education. While the developed countries have it and use it in every aspect of their lives, the developing countries desperately need the ICT tool for its students. Due to the geographical location of the Pacific Island countries, there is a genuine need to design and develop tailor-made Intelligent Classrooms. Hence, another new initiative of USP is to invest in the Intelligent Classroom would be located in all the USP regional campuses with networking of intelligent sensors. This paper presents the prototype of smart classroom which ultimately scaffold to Intelligent Classroom. The intelligent classrooms would be interfaced such that the distance student/staff would be able to interact actively in real time. Adding live-blog for students during lectures to discuss and share ideas with other the students in the same lecture throughout the region. Through live-blog students in small number class in some campus would not feel isolated. Smart timetable is also linked where students will be able to view lecture times via Mobile app, or even the lecturer can book the classroom using app. Moreover human action recognition and voice to text is also interfaced to ease load on lecturer and convert lecture into subtitles preferred to students.