Research On Coordinated Management Methods Of Energy And Thermal Comfort For Intelligent Office Buildings

With the further population growth in the world and the rapid increase of purchasing power in emerging economies,the demand for building energy is increasing.Since building energy is mainly derived from traditional fossil fuels,the increasing demand for building energy will cause a series of energy and environmental issues,which also increase the financial burden on building owners.In office buildings,the energy consumption of heating,ventilation,and air conditioning systems(HVAC)account for about 40%of the total energy consumption of buildings,but the thermal comfort satisfaction rate(TCSR)of occupants is still low.The main reason for this phenomenon is that the temperature set-point can not be properly adjusted according to thermal comfort requirements of occupants,which incurs the situation of high energy consumption but low thermal comfort.With the rapid development of information technologies such as the Internet of Things and artificial intelligence,there are new opportunities for achieving the optimal trade-off between thermal comfort and energy consumption by dynamically adjusting temperature set-points.Therefore,it is very necessary to study the joint management of energy and thermal comfort for smart office buildings.Firstly,this thesis investigates a weighted sum optimization problem of HVAC system energy consumption and TCSR of occupant in shared office spaces.Due to the existence of a variety of uncertain system parameters(e.g.,outdoor temperature,preferred temperature of occupants,and occupancy state),temporally coupled constraints,and unexplicit building thermal dynamic models,it is challenging to solve the formulated optimization problem.Therefore,this thesis reformulates the above problems as a Markov decision process(MDP)and proposes an optimal HVAC control algorithm for HVAC systems based on deep Q-network and priority experience replay.Simulation results based on real data show the effectiveness of the proposed algorithm.Secondly,this thesis investigates a problem of optimal tradeoff between HVAC energy consumption and TCSR in shared office spaces.Due to the existence of uncertain system parameters,temporally coupled constraints,unexplicit building thermal dynamic models,and different units related to two objectives,it is challenging to solve the formulated optimization problem.To this end,this thesis reformulates the optimization problem as an MDP and proposes an HVAC control algorithm based on multi-objective double deep Q-network.The proposed algorithm does not need to choose appropriate objective weights beforehand,which can generate a set of strategies.Simulations results show the effectiveness of the proposed algorithm and its advantages in terms of implementing performance tradeoff,finding pareto solutions,and tracking target TCSR.Finally,this thesis makes a brief summary and points out future research directions.