Jump to Main Content
A top-down control method of nZEBs for performance optimization at nZEB-cluster-level
- Huang, Pei, Wu, Hunjun, Huang, Gongsheng, Sun, Yongjun
- Energy 2018 v.159 pp. 891-904
- algorithms, batteries, buildings, control methods, economic costs, energy, imports, renewable energy sources
- Nearly zero energy buildings (NZEBs) are considered as a promising solution to the mitigation of the energy problems. A proper control of the energy system operation of the nZEB cluster is essential for improving load matching, reducing grid interaction and reducing energy bills. Existing studies have developed many demand response control methods to adjust the operation of energy systems to improve performances. Most of these studies focus on optimizing performances at individual-nZEB-level while neglecting collaborations (e.g. energy sharing and battery sharing) between nZEBs. Only a few studies consider the collaborations and optimize the system operation at nZEB-cluster-level, yet they cannot take full advantage of nZEB collaborations as optimization is conducted in a bottom-up manner lacking global coordination. This paper, therefore, proposes a top-down control method of nZEBs for optimizing performances at the cluster level. The top-down control method first considers the nZEB cluster as ‘one’ and optimizes its energy system operation using the genetic algorithm (GA), and then it coordinates the operation of every single nZEB inside the cluster using non-linear programming (NLP). The top-down control enables collaborations among nZEBs by coordinating single nZEB's operations. Such collaborations can bring significant performance improvements in different aspects. For instance, in aspect of economic cost, the collaborations can reduce the high-priced energy imports from the grid by sharing the surplus renewable energy with nZEBs which have insufficient energy generations. The proposed top-down control has been compared with a traditional non-collaborative control. The study results show that the top-down control is effective in improving performances at cluster level.