Scholarlink

<p>As global transportation sector carbon emissions continue to rise, optimizing carbon emission efficiency without compromising service quality becomes a critical challenge. This study introduces an innovative deep learning model, GS-Adam-LSTM, combining Long Short-Term Memory (LSTM) networks with the Adam optimization algorithm, enhanced through Grid Search (GS), to predict transportation carbon emission efficiency in China. Results demonstrate that GS-Adam-LSTM significantly improves prediction accuracy and reveals a trend in China’s transportation carbon emission efficiency over the next two decades: a gradual decline followed by a sharp decrease. Key factors contributing to variations in efficiency include economic development, regional differences, infrastructure levels, and policy focus. Multi-scenario analysis identifies the optimal strategy for improving carbon emission efficiency as a combination of high labor input, high capital investment, and low resource consumption. Conversely, low labor input with high capital investment and low resource consumption is a suboptimal path leading to increased emissions. Based on these findings, the study proposes policy recommendations to improve efficiency, providing a scientific foundation for governments and industry leaders to design effective long-term carbon reduction strategies.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Transportation carbon emission prediction based on improved long and short-term memory deep learning from the perspective of carbon emission efficiency

  • 期刊论文

  • 发表日期：

    2026年4月25日

查看全文
  • Lin Liu,
  • Yuming Liu,
  • Rong Zhao,
  • Xizhen Xu

摘要

As global transportation sector carbon emissions continue to rise, optimizing carbon emission efficiency without compromising service quality becomes a critical challenge. This study introduces an innovative deep learning model, GS-Adam-LSTM, combining Long Short-Term Memory (LSTM) networks with the Adam optimization algorithm, enhanced through Grid Search (GS), to predict transportation carbon emission efficiency in China. Results demonstrate that GS-Adam-LSTM significantly improves prediction accuracy and reveals a trend in China’s transportation carbon emission efficiency over the next two decades: a gradual decline followed by a sharp decrease. Key factors contributing to variations in efficiency include economic development, regional differences, infrastructure levels, and policy focus. Multi-scenario analysis identifies the optimal strategy for improving carbon emission efficiency as a combination of high labor input, high capital investment, and low resource consumption. Conversely, low labor input with high capital investment and low resource consumption is a suboptimal path leading to increased emissions. Based on these findings, the study proposes policy recommendations to improve efficiency, providing a scientific foundation for governments and industry leaders to design effective long-term carbon reduction strategies.