Articles

Burning fossil fuels like coal, oil, and natural gas releases significant CO 2 , driving carbon emissions in electricity production. This contributes to global warming, leading to climate change, extreme weather, rising sea levels, and harm to ecosystems and human health. Achieving zero emissions requires transitioning to low-carbon energy sources. This study uses various Machine Learning (ML) models for predicting low carbon electricity generation additionally eXpalinalbe Artificial Intelligence (XAI) to elucidate how ML model works and suggest some important factor contributing to the models outcome. Among several ML models RF outperforms other by achieving MSE of 2.782, RMSE of 2.782, MAE of 1.443 and R 2 of 99.3%. Also GBM and XGB performed closely as RF. When applying XAI tools like SHAP and Shapash to RF it reveals some important factors such as Electricity from renewables (TWh), Eectricity from fussil fuels (TWh) Renewable energy share in the total final energy consumption (%), and Electricity from nuclear. Which are crucial for future energy planning and policy decisions. Future study can use a more extensive data set, investigate economic aspects, and include real-time data to enhance predictive model performance.

Silk production in Bangladesh's rural economy, which is the practice of raising silkworms, has long been an essential part of the country’s rural economy and this has contributed a lot to the national economy as well as improved people’s lives. Traditional sericulture however faces problems such as pest infestations, inconsistent yields of mulberry leaves, wasteful use of resources and effects of climatic change despite the country having favorable climate that supports silk farming and availability of a skilled labor force. These challenges can be addressed through recent technological innovations including artificial intelligence technology and drone surveillance for instance. Drones fitted with cameras combined with GPS have ability to control fully grown mulberry farms whereas artificial intelligence system analyze data on optimal resource management in terms of pests’ control among others. The objective is to establish how drones and AI can affect Bangladeshi sericulture farming field primarily in Jessore and Rajshahi – two most important sericulture areas in the country. For two crop cycles namely July-October 2022 and March-June 2023, research used drones along with TensorFlow for AI integration where it recorded aspects such as leaf yield, pest loads, input application rates among others. Technologies integrations had resulted to increased mulberry leaf yield by approximately 18% from 2500Kg/ha to 2950Kg/ha.