This study develops and evaluates an AI-driven stacked hybrid machine learning model for predicting the total power output of wave energy converters (WECs) across four Australian coastal locations: Adelaide, Perth, Sydney, and Tasmania. This research enhances prediction accuracy through advanced ensemble learning techniques while addressing spatial variability in wave energy processes. The dataset comprises spatial coordinates and power output readings from 16 fully submerged WECs per location, capturing the variability of wave energy across different coastal regions. Data preprocessing included missing value imputation, duplicate removal, and spatial feature transformation via Euclidean distance calculation. Principal component analysis (PCA) was employed to reduce dimensionality while preserving critical features influencing power generation. To develop an accurate prediction model, we employed a stacking ensemble approach using XGBoost, LightGBM, and CatBoost as base learners, optimized via Optuna hyperparameter tuning with 10-fold cross-validation. A Ridge regression meta-learner combined the outputs of these models, leveraging their complementary strengths to enhance predictive performance. Experimental results demonstrate that the hybrid model consistently outperforms individual models, enhancing predictive accuracy across all locations. Sydney exhibited the highest accuracy (RMSE = 9089.58 W, R² = 0.8576), while Tasmania posed the greatest challenge (RMSE = 45,032.37 W, R² = 0.8378). The ensemble approach mitigated overfitting and improved generalization by leveraging the complementary strengths of XGBoost, LightGBM, and CatBoost. By leveraging AI-driven ensemble learning, this study provides a scalable and reliable framework for wave energy forecasting, facilitating more efficient grid integration and resource planning in renewable energy systems.

In this article, an optimal size and smart charging abilities of an electric vehicle charging station (EVCS) assuming the quality of service (QoS) using a hybrid technique are proposed. The proposed method is the combination of the Golden Eagle Optimizer (GEO) and the Radial Basis Function Neural Network (RBFNN), and therefore, it is known as the GEO-RBFNN method. The major aim of this work is “to solve the dimensional problem of charging stations (CSs) using an optimization framework that lessens the investment costs of charging station operators, and also, quality of service (QoS) is achieved by its electric vehicle owners.” The effect of smart scheduling is analyzed through the loading capacities of the scheduling charging station. The parameters of the incoming charging tasks are approximated using analytical probability distributions and handle the size of the issue depending on the optimization technique. By using the proposed strategy, the electric vehicle begins to charge at regular intervals. The electric vehicle starts charging, but once it starts the charging, it can be no longer interrupted. The scalability of the optimal sizing approach is better with the aid of the proposed GEO-RBFNN strategy. The proposed approach is carried out on the MATLAB site, and its efficiency is analyzed using other existing approaches. In the GEO-RBFNN technique, the simulation time achieves 14.8 seconds at 100 counts of iteration. At 250 counts of iteration, the simulation time achieves 29.2 seconds. At 500 counts of iteration, the simulation time achieves 69.9 seconds. At 1000 counts of iteration, the simulation time achieves 77.1 seconds.

This manuscript proposes a hybrid strategy for power flow management under a smart grid system connected to a hybrid renewable energy system (HRES). The HRES is the combination of photovoltaic, wind turbine, fuel cell and energy storage system or battery. The proposed technique is the execution of the Integrated Harris Hawk Optimization algorithm (IH²OA). Here, Harris Hawk Optimization is the consolidation of crossover and mutation function, hence it is called IH²OA. Moreover, IH²OA is enhancing the voltage source inverter of control signals under the variation of power replace among source and load side. The numerous characteristics are made up of necessary grid active with reactive power variations created under the view of available power source. The IH²OA method ensures the location of control signals using parallel implementation as opposed to active with reactive power variation. In the IH²OA method, the control scheme-based optimizes the power controller under the power flow variation. The proposed system performed on MATLAB/Simulink platform and the efficiency are compared with diverse existing techniques.

This paper proposes a high-gain self-lift quadratic boost converter (SL-QBC) topology tailored for efficient marine propulsion drive systems. The presented converter architecture incorporates a self-lift circuit embedded within a quadratic boost stage to achieve significant voltage gain while maintaining low component stress and high conversion efficiency. Specifically, the converter is capable of boosting a low input voltage of 45 V to a high output voltage of 473 V. The high-voltage output is then fed to the marine propulsion system through a three-phase, three-level neutral point clamped (NPC) inverter, which drives a three-phase induction motor. Designed to interface low-voltage DC sources such as batteries, photovoltaic panels, or fuel cells with high-power marine propulsion motors, the proposed topology offers a compact and cost-effective solution for electric and hybrid marine applications. The self-lift mechanism enhances voltage boosting capability by leveraging additional inductive and capacitive energy transfer paths. The topology ensures continuous current operation, reduced voltage ripple, and improved dynamic response, all of which are critical for smooth and reliable marine drive performance. Simulation results validate the converter’s effectiveness, confirming its potential as a robust and efficient power conditioning stage in next-generation marine energy systems.