Home > Research > Publications & Outputs > HVAC-based hierarchical energy management syste...

Electronic data

  • 2020MaPhD

    Final published version, 7.22 MB, PDF document

Text available via DOI:

View graph of relations

HVAC-based hierarchical energy management system for microgrids

Research output: ThesisDoctoral Thesis

Publication date21/10/2020
Number of pages191
Awarding Institution
  • Lancaster University
<mark>Original language</mark>English


With the high penetration of renewable energy into the grid, power fluctuations and supply-demand power mismatch are becoming more prominent, which pose a great challenge for the power system to eliminate negative effects through demand side management (DSM). The flexible load, such as heating, ventilation, air conditioning (HVAC) system, has a great potential to provide demand response services in the electricity grids.

In this thesis, a comprehensive framework based on a forecasting-management optimization approach is proposed to coordinate multiple HVAC systems to deal with uncertainties from renewable energy resources and maximize the energy efficiency. In the forecasting stage, a hybrid model based on Multiple Aggregation Prediction Algorithm with exogenous variables (MAPAx)-Principal Components Analysis (PCA) is proposed to predict changes of local solar radiance, vy using the local observation dataset and real-time meteorological indexes acquired from the weather forecast spot. The forecast result is then compared with the statistical benchmark models and assessed by performance evaluation indexes. In the management stage, a novel distributed algorithm is developed to coordinate power consumption of HVAC systems by varying the compressors’ frequency to maintain the supply-demand balance.
It demonstrates that the cost and capacity of energy storage systems can be curtailed, since HVACs can absorb excessive power generation. More importantly, the method addresses a consensus problem under a switching communication topology by using Lyapunov argument, which relaxes the communication requirement. In the optimization stage, a price-comfort optimization model regarding HVAC’s end users is formulated and a proportional-integral-derivative (PID)-based distributed algorithm
is thus developed to minimize the customer’s total cost, whilst alleviating the global power imbalance. The end users are motivated to participate in energy trade through DSM scheme. Furthermore, the coordination scheme can be extended to accommodate battery energy storage systems (BESSs) and a hybrid BESS-HVAC system with increasing storage capacity is proved as a promising solution to enhance its selfregulation ability in a microgrid. Extensive case studies have been undertaken with the respective control strategies to investigate effectiveness of the algorithms under various scenarios.

The techniques developed in this thesis has helped the partnership company of
this project to develop their smart immersion heaters for the customers with minimum energy cost and maximum photovoltaic efficiency.