This paper examines the complex dependence between peak district heating demand and outdoor temperature. Our aim is to provide the probability law of heat demand given extreme weather conditions, and derive useful implications for the management and production of thermal energy. We propose a copula-based approach and consider the case of the city of Bozen-Bolzano. The analysed data concern daily maxima heat demand observed from January 2014 to November 2017 and the corresponding outdoor temperature. We model the univariate marginal behaviour of the time series of heat demand and temperature with autoregressive integrated moving average models. Next, we investigate the dependence between the residuals’ time series through several copula models. The selected copula exhibits heavy-tailed and symmetric dependence. When taking into account the conditional behaviour of heat demand given extreme climatic events, the latter strongly affects the former, and we find a high probability of thermal energy demand reaching its peak.

本文研究了高峰地区供热需求与室外温度之间的复杂关系。我们的目标是在极端天气条件下提供热量需求的概率定律，并对热能的管理和生产产生有益的启示。我们提出一种基于copula的方法，并考虑Bozen-Bolzano市的情况。分析的数据涉及2014年1月至2017年11月观察到的每日最大热量需求以及相应的室外温度。我们使用自回归综合移动平均模型对热需求和温度时间序列的单变量边际行为进行建模。接下来，我们通过几种关联模型研究残差的时间序列之间的依赖性。选定的系动词表现出重尾和对称性。