Effective machine learning regression models are useful toolsets for managing and planning energy in PV grid-connected systems. Machine learning regression models, however, have been crucial in the analysis, forecasting, and prediction of numerous parameters that support the efficient management of the production and distribution of green energy. This article proposes multiple regression models for power prediction using the Sharda University PV dataset (2022 Edition). The proposed regression model is inspired by a unique data pre-processing technique for forecasting PV power generation. Performance metrics, namely mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), R2-score, and predicted vs. actual value plots, have been used to compare the performance of the different regression. Simulation results show that the multilayer perceptron regressor outperforms the other algorithms, with an RMSE of 17.870 and an R2 score of 0.9377. Feature importance analysis has been performed to determine the most significant features that influence PV power generation.
With the introduction of 5G technology, wireless communication is expected to undergo revolutionary changes that will allow high-speed connectivity and scalability. Though 5G networks have the potential to be revolutionary, they also pose new challenges in ensuring the security and integrity of data transfer, especially in Non-IP Data Delivery (NIDD) scenarios. The need for robust anomaly detection systems becomes even more critical in this scenario to safeguard IoT and other reliable networks. Anomaly detection has been the subject of much research in network contexts, as it is crucial for identifying hostile activity, system failures, and odd behavior. The growing dependence on technologies, particularly with the advent of 5G and its potential to connect nearly everything, has made it imperative to investigate intelligent and efficient techniques that ensure network availability, secrecy, and integrity. To address the botnet infiltration, DDoS mitigation, and other incursions in 5G networks, a novel lightweight intrusion detection model is proposed for 5G and beyond networks, which uses the 5GNIDD dataset in the experiments. The proposed model is powered by a robust prepossessing model, which uses Gini Importance for feature selection and state-of-the-art classifiers, namely, AdaBoost, Easy Ensemble, GRU, 1D-CNN, LSTM, and hybrid CNN-LSTM for classification. Two different case studies with k-best features are driven in experiments showcasing the effect of the curse of dimensionality on precision. The model has obtained 99.64% accuracy and a 0.9830 precision using 1D-CNN and a hybrid LSTM-CNN model.