Energy management for data centre microgrids considering co-optimisation of workloads and waste heat

Abstract

The high energy consumption of data centres (DCs) has recently become a significant problem with the rapid development of DCs worldwide. This work studies the daily energy management of a data centre microgrid (DCM). The energy management problem is formulated as a two-stage stochastic mixed-integer linear programing (MILP) model that accounts for workload schedules, cooling resources, uncertainties of onsite renewable generation, and electricity price. An energy management scheme that co-optimises workloads and waste heat is proposed to minimise the operating cost. Furthermore, the potential for waste heat recovery and reuse is also considered. The simulation results illustrate that the proposed model can reduce the operating cost of DCM by the co-optimisation of workloads and waste heat while meeting the temperature requirements for server clusters.