In this study, an integrated method (abbreviated as MGCMs-SCA-FER) is developed for assessing the impacts of climate change, which incorporates multiple global climate models (MGCMs), stepwise cluster analysis (SCA), and fixed-effects regression (FER) within a general framework. MGCMs-SCA-FER is capable of (i) dealing with the uncertainty in climate change projection caused by heterogeneity of structures and parameters of GCM; (ii) capturing nonlinear relationship between input variables and outputs without assumption of their functions; (iii) identifying interaction of different units and quantifying the effects of climate change on electricity demand. MGCMs-SCA-FER is then applied to Jing-Jin-Ji for assessing the impacts of climate change on single-city and entire-region electricity demands. Results demonstrated that climate change projections and electricity demand predictions varied significantly across different GCMs and RCPs. Results disclose that (i) Jing-Jin-Ji region would experience a warmer climate in the next 80 years of 2021 ~ 2100 (For every decade, temperatures would increase by [0.17, 0.23] °C under RCP4.5 and [0.35, 0.54] °C under RCP8.5); (ii) For 1 °C increase in temperature, annual electricity demand would rise by 4.5%; (iii) electricity intensity has the most significant impact on electricity demand for Jing-Jin-Ji region; (iv) electricity demand would increase under all scenarios, and the electricity demand under RCP8.5 would be higher than that under RCP4.5. From a long-term perspective, analyzing the climate change impacts on electricity demand and making adaptive management strategy are important for the regional sustainability

中文翻译：

---

气候变化影响评估综合方法（MGCMs-SCA-FER）的开发——以京津冀地区为例

在本研究中，开发了一种综合方法（简称 MGCMs-SCA-FER）来评估气候变化的影响，该方法结合了多个全球气候模型 (MGCMs)、逐步聚类分析 (SCA) 和固定效应回归 (FER) ) 在一个总体框架内。MGCMs-SCA-FER 能够 (i) 处理由 GCM 结构和参数的异质性引起的气候变化预测的不确定性；(ii) 在不假设函数的情况下捕捉输入变量和输出之间的非线性关系；(iii) 确定不同单位的相互作用并量化气候变化对电力需求的影响。然后将 MGCMs-SCA-FER 应用于京津冀以评估气候变化对单个城市和整个区域的电力需求的影响。结果表明，气候变化预测和电力需求预测在不同的 GCM 和 RCP 之间存在显着差异。结果表明(i)京津冀地区在2021~2100年的未来80年将经历更温暖的气候（在RCP4.5和[0.35，每十年，温度将增加[0.17，0.23]°C， 0.54] °C 在 RCP8.5 下）；(ii) 温度每升高 1°C，年电力需求将增加 4.5%；(iii) 电力强度对京津冀地区电力需求的影响最为显着；(iv) 所有情景下的电力需求都会增加，RCP8.5 下的电力需求将高于 RCP4.5 下的电力需求。从长远来看，