We propose a thermal scheduler that learns temperature patterns in edge devices. It reallocates tasks before heat thresholds are reached, improving longevity and energy efficiency in distributed smart environments.

Chen Zhen Liang, Liu Cheng Hao, Xu Hao Min, Gao Tian Jie, Zhang Ming Rui

This paper presents a green cloud inference model. It routes tasks to data centers based on carbon intensity, minimizing emissions while preserving performance in global AI services.

Laura Whitmore, Thomas Greene, Emily Patterson, Benjamin Clarke, James Russell

This research introduces an eco-friendly agri-IoT system. It uses solar prediction for communication scheduling and edge AI to monitor crop conditions, enabling off-grid sustainability in smart farming initiatives.

Chen Rong Liang, Gao Rui Wen, Liu Min Hao, Xu Wei Long, Zhang Ming Tao

We propose a battery-free grid monitoring solution. Powered by ambient energy, it uses on-device AI to track electrical parameters, ensuring sustainable and low-maintenance rural electrification.

Vikas Sharma, Jean Claude Bernard, Priya Meenakshi Subramanian, Carlos Javier Morales

This paper presents a low-energy AI solution for disaster alerts. It detects environmental anomalies using edge analytics, enabling real-time warnings with minimal energy cost in isolated areas.

Chen Liang Wen, Xu Hao Bo, Gao Ming Fang, Liu Xiu Tao, Zhang Lin Jie

We introduce a sustainable AI inference selector. It evaluates models based on carbon cost and accuracy tradeoffs, enabling responsible deployment of machine learning systems in enterprise-scale production environments.

Michael Foster, Sarah Hudson, Emily Clarke, Rachel Bennett, Nathaniel Brooks

This paper introduces an energy-aware federated learning model. It predicts renewable availability and adapts device involvement dynamically, reducing carbon emissions while maintaining performance in distributed edge AI applications for sustainable intelligent systems.

Ramesh Kulkarni, Jean Claude Laurent, Priya Shalini Desai, Carlos Manuel Gutierrez, Fatima Noor Al-Saadi

This research presents a forest monitoring platform using solar-harvesting sensors. It employs lightweight AI to predict fire risks, enabling real-time alerts in remote regions while ensuring long-term deployment without battery replacement.

Chen Ming Zhe, Liu Bo Sheng, Xu Hao Lin, Zhang Rui Cheng, Gao Fang Liang