Engku Ahmad Azrulhisham

Model Predictive Control (MPC) performance depends on prediction and control horizons, particularly for experimentally identified inverse-response processes with noise, actuator constraints, and plant–model mismatch. This study evaluates an offline reinforcement learning (RL)-assisted supervisory framework for horizon selection in constrained MPC of a membrane-filtration process. Pump-speed-command and pump-line-flow data identified a second-order state-space model using Numerical Algorithms for Subspace State-Space System Identification (N4SID), achieving 92.16% estimation fit and 90.90% independent one-step-ahead validation fit, although long-horizon fidelity remained limited. The model was embedded in target-tracking MPC with input and input-move constraints, while RL selected only the horizons. Nominally, the representative RL-selected pair reduced the prespecified composite score from 2.5700 to 2.4562 (4.43%) with zero simulated quadratic-programming failures. Exhaustive evaluation of 146 feasible pairs identified 1.2214, confirming that RL was not globally optimal. Under a 120-iteration budget, Bayesian optimisation reached the global optimum in 20/30 runs versus 3/30 for epsilon-greedy. Extending the domain to 208 pairs reduced the score to 1.1838, indicating sensitivity to the lower prediction-horizon bound. Across 11 model-based robustness scenarios, RL-tuned MPC achieved an average 3.13% score reduction versus baseline, with individual-metric trade-offs. Overall, horizon selection improved the MPC trade-off, but results remained dependent on the search domain and identified-model environment.

Wave energy harvesters are still not as mature as other types of renewable energy harvesting devices, particularly when it comes to commercialization, mass production, and grid integration, even though ocean waves around the world are known to contain high and dense quantities of energy. However, recent studies and optimizations suggested that the point absorber wave energy harvester may be a viable option. This paper provides a comprehensive comparison hydrodynamic damping analysis of the heaving motion on one body and two body point absorber wave energy harvesters by using Computational Fluid Dynamic (CFD) software. A simulation run was proposed for the analysis performance of damping of the point absorber devices to low wave height (0.7m-2.5m) and period (0.9s-2.2s). There are two types of point absorbers: one-body point absorber and two body point absorber. Both point absorber body are examined by simulation in terms of its heaving dynamic motion frequency in calm, medium and strong wave condition. This analysis highlights the extensive research being done to bring point absorbers closer to technical maturity, paving the way for commercialization and mass production for low wave characteristics. The results showed two body point absorber Response Amplitude Operator (RAO) is improved to absorbed low wave height compared to one body by RAO efficiency 17% and device efficiency 25%. The two-body point absorber performed better in all wave condition.

Oscillating Water Column (OWC) is a technology that harvests wave energy to generate electricity. It works by harnessing the wave-induced pressure oscillation in the opening mouth of OWC. The oscillation of the pressurised air is extracted using a turbine placed at the top of the OWC structure. Although OWC has been studied for many years, designing an optimum air duct for such devices is still an open research area. The air duct plays a crucial role in the energy conversion process, as it is responsible for directing the airflow towards the turbine, and its design can greatly affect the performance and efficiency of the system. The article aims to study how the air duct design impacts the power output form. Moreover, the result will be compared with the conventional air duct design. The OWC performance is analysed using Computational Fluid Dynamic (CFD) software to identify which shape of the air duct has high efficiency. Autodesk Inventor is used to design the OWC devices modified from the basis design. Then these designs are imported into CFD software to simulate and obtain the energy extraction result. The CFD software is chosen to simulate the hydrodynamics problem. In this study, the oval design shows significant improvement in the energy extraction performance compared to the basis design. The oval design managed to produce 9.81, 10.02, and 11.66 J/kg of mechanical energy compared to 9.81, 9.82, and 10.04J/kg as prompted by the basis air duct design at the wave height of 0.2505, 0.75, and 2.25m, respectively. The result shows that the non-edge produces higher air pressure than the sharp edge of the air duct shape. Thus, the oval air duct design is more efficient, and it may increase the performance of the OWC wave energy converter device at a low wave height of 0.75 m.

Small hydro system is one of the important sources of renewable energy and it has been recognized worldwide as clean energy sources. Small hydropower generation system uses the potential energy in flowing water to produce electricity is often questionable due to inconsistent and intermittent of power generated. Potential analysis of small hydro system which is mainly dependent on the availability of water requires the knowledge of water flow or stream flow distribution. This paper presented the possibility of applying Pearson system for stream flow availability distribution approximation in the small hydro system. By considering the stochastic nature of stream flow, the Pearson parametric distribution approximation was computed based on the significant characteristic of Pearson system applying direct correlation between the first four statistical moments of the distribution. The advantage of applying various statistical moments in small hydro potential analysis will have the ability to analyze the variation shapes of stream flow distribution.

This paper presents the propose application of GSM wireless datalogger at Small Hydro Power Generation Station. Normally, renewable energy power generation system requires the application of data acquisition. Data acquisition of power, voltage, current, sedimentation, water flowrate and quantity of water are common data needed to evaluate performance. Thus, influence the planning electrical energy supply to the grid TNB station and provide clarification to ecological events. Intake of Small Hydro Station is located inaccessible area that is 2.07km heights and 3.44km distance by route from powerhouse. There is no active GSM line provider at the area. Therefore, advance communication technologies such as radio frequency device offers new possibilities of efficient data transferring. They are more reliable and cheaper than conventional method. Datalogger DT80M has the ability to send data using GSM line provider from remote area. Route distance is approximately 100km between powerhouse Small Hydro station to Host or Master Station. Power and water flowrate, will be monitor using Labview and save into database server. Random raw data derive into mean, standard deviation, skewness and kurtosis will be use to simulate performance model of density function.

In view of increasing pressures of shortened development cycles and desire to save costs, inverse power law scaling has been devised for operative life estimation based on accelerated laboratory test data extrapolation. Derived from the Coffin-Manson's model of fatigue life, this analytical solution however is not capable in addressing probabilistic nature of fatigue. This paper is aimed to provide further insight into the operative life estimation of carbon steel stub axle using parametric models derived from integration of inverse power law stress-life model and probabilistic fatigue life distributions. Maximum likelihood estimation is applied in deriving the parameters of the parametric models and PSN curves are obtained from reliability function of accelerated laboratory test data. PSN curves at operative life as well as any life region can be estimated based on inverse power law acceleration factors. The procedures presented in this study can be applied in estimating the probabilistic fatigue life curves at an accelerated pace resulting in shortened development cycles.