World Energy Outlook 2025
关于本报告
国际能源署的旗舰出版物World Energy Outlook（《世界能源展望》，WEO）拥有全球最为权威的能源分析和预测。该出版物每年发布一次新版本，以反映最新的能源数据、技术与市场趋势、政策动态等，广泛探讨能源未来图景，并深入剖析前述数据及信息对能源安全、可获得性与排放状况的深远影响。
WEO采用情景分析法对整个能源系统进行评估，着力揭示未来可能面临的关键抉择、后果与变数。WEO既包含基于不同的现行政策相关假设所构建的探索性情景，也描绘了全面实现能源与排放目标的规范性路径。多情景分析方法生动演示了关键变量（包括全球各国政府的能源政策）将如何影响能源系统的发展路径。
在今年新一期的WEO发布之际，全球能源政策与市场正值深刻转型，地缘政治也日趋紧张。对于能源安全、可负担性与可持续性这三大关切问题的“最优解”，各国政府正在形成不同结论。WEO一如既往地向读者提供无与伦比的洞见，帮助其深刻理解不同能源政策与投资选择所可能带来的结果。关键矿产供应安全是今年WEO的一个重要主题。
Explore this report with AI Tips for using the IEA WEO GPT
Understand the limitations: This agent has been trained on the content of the IEA's World Energy Outlook 2025 and World Energy Outlook 2025 Free Dataset to make it easier to explore and understand the analysis it contains. The agent has access to a general knowledge base for training purposes, but it cannot access additional sources materials or online resources. For detailed or official interpretations, please refer to the full report or contact the International Energy Agency.
Be clear and specific: Provide detailed and specific instructions to get the most accurate and relevant responses. For example, instead of asking "Tell me about energy trends," try, "Summarize the key findings on renewable energy capacity from the World Energy Outlook 2025."
Ask one question at a time: To ensure clarity and focus, ask one question at a time. This helps the agent provide a more precise and helpful response. If the response isn’t what you expected, rephrase or refine your query to better guide the agent.
Provide context when needed: If your question relates to a specific section or chapter of the World Energy Outlook 2025, mention it in your prompt. For example, "Explain the projections for global oil demand in Chapter 3 of the World Energy Outlook 2025."
Specify desired format: If you need the response in a particular format such as a list or summary, mention it in your prompt. For example, "List the key points from the executive summary of the World Energy Outlook 2025."
Clarify scenario, region and timeframe: When asking for data, be specific about the scenario and region you are interested in. For example, "Provide data on electricity generation from renewables by decade in the Stated Policies Scenario for Europe."
Use comparative analysis: Take advantage of the agent’s capability to compare data across scenarios, regions, or timeframes. For example: "Compare solar power generation in Asia under the Stated Policies Scenario versus the Current Policies Scenario for 2035."
Use natural language: Feel free to ask questions or make requests in a conversational manner. The agent is designed to understand and respond to natural language inputs. Use follow-up questions to dive deeper into the topics. For example, after asking, “What are the key trends in natural gas demand,” you may ask, “How do these trends differ across scenarios in Europe?” If you’d like to start a conversation anew simply say, “Start over”.
Be patient: Complex queries may take a moment to process. Please be patient while the agent generates a response.
Global Energy and Climate Model
Over the past four years, the IEA has worked to develop a new integrated modelling framework: the IEA’s Global Energy and Climate (GEC) Model. This model is now the principal tool used to generate detailed sector-by-sector and region-by-region long-term scenarios across IEA's publications, including the World Energy Outlook series and Energy Technology Perspectives series.
In 2021, the IEA adopted for the first time a new hybrid modelling approach relying on the strengths of both models. The integrated framework of the IEA’s Global Energy and Climate Model (GEC Model) is now the principal tool used to generate detailed sector-by-sector and region-by-region long-term scenarios across IEA publications, including the World Energy Outlook series and Energy Technology Perspectives series.
Download the GEC Model Methodology document for an in-depth description of the overall approach and features of the model, and download the GEC Model Key Input dataset for selected key input data.