Urban travel carbon emission mitigation approach using deep reinforcement learning DOI Creative Commons
Jie Shen, Feng Zheng,

Yuanli Ma

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Nov. 13, 2024

The urbanization process has led to a significant increase in energy consumption and carbon emissions, which can be mitigated through scientific urban planning management. This research proposes bottom-up emission mitigation strategy based on deep reinforcement learning (DRL). Using Ningbo City as case study, multi-source data, including points of interest (POI) data transportation system are utilized, along with varying coefficients for different travel modes, construct comprehensive environment areas. proposed DRL model adopts an Actor-Critic framework, iteratively optimizes the land use configuration building type proportions within matrix achieve goal mitigating emissions. Experimental results demonstrate that this approach exhibits reduction effects scenario. By adjusting discount rate reward function, various optimization strategies obtained, such short-term long-term strategies, achieving reductions 0.47% 0.61%, respectively, notably higher than 0.39% expected if emissions were uniformly distributed across matrix. findings highlight potential DRL-based approaches adaptive data-driven mitigation.

Language: Английский

Urban travel carbon emission mitigation approach using deep reinforcement learning DOI Creative Commons
Jie Shen, Feng Zheng,

Yuanli Ma

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Nov. 13, 2024

The urbanization process has led to a significant increase in energy consumption and carbon emissions, which can be mitigated through scientific urban planning management. This research proposes bottom-up emission mitigation strategy based on deep reinforcement learning (DRL). Using Ningbo City as case study, multi-source data, including points of interest (POI) data transportation system are utilized, along with varying coefficients for different travel modes, construct comprehensive environment areas. proposed DRL model adopts an Actor-Critic framework, iteratively optimizes the land use configuration building type proportions within matrix achieve goal mitigating emissions. Experimental results demonstrate that this approach exhibits reduction effects scenario. By adjusting discount rate reward function, various optimization strategies obtained, such short-term long-term strategies, achieving reductions 0.47% 0.61%, respectively, notably higher than 0.39% expected if emissions were uniformly distributed across matrix. findings highlight potential DRL-based approaches adaptive data-driven mitigation.

Language: Английский

Citations

1