A Data Augmentation Method for Multi-energy Parks Considering Different Scenario Factors

As an important physical carrier of the Energy Internet, multi-energy parks are currently mostly demonstration- based, with few samples and limited access to user-side measurement data. In the planning stage of the park, it is an urgent problem to study how to generate enough energy assessment data to meet the planning requirements from, for example, a small amount of cooling, heating and electricity load data. In this paper, a data augmentation method for multi-energy parks is proposed to expand the sample size by considering factors such as different administrative districts, functional areas, seasons and weather conditions, which is based on path distance of Dynamic Time Warping (DTW) as a metric between time series samples. In addition, the feasibility of the proposed method is demonstrated by a case study, which lays a solid data foundation for the planning, design and evaluation of regional multi-energy parks.