An innovative heterogeneous transfer learning framework to enhance the scalability of deep reinforcement learning controllers in buildings with integrated energy systems

Deep Reinforcement Learning(DRL)-based control shows enhanced performance in the management of integrated energy systems when compared with Rule-Based Controllers(RBCs),but it still lacks scalability and generalisation due to the necessity of using tailored models for the training process.Transfer Learning(TL)is a potential solution to address this limitation.However,existing TL applications in building control have been mostly tested among buildings with similar features,not addressing the need to scale up advanced control in real-world scenarios with diverse energy systems.This paper assesses the performance of an online heterogeneous TL strategy,comparing it with RBC and offline and online DRL controllers in a simulation setup using EnergyPlus and Python.The study tests the transfer in both transductive and inductive settings of a DRL policy designed to manage a chiller coupled with a Thermal Energy Storage(TES).The control policy is pre-trained on a source building and transferred to various target buildings characterised by an integrated energy system including photovoltaic and battery energy storage systems,different building envelope features,occupancy schedule and boundary conditions(e.g.,weather and price signal).The TL approach incorporates model slicing,imitation learning and fine-tuning to handle diverse state spaces and reward functions between source and target buildings.Results show that the proposed methodology leads to a reduction of 10% in electricity cost and between 10% and 40% in the mean value of the daily average temperature violation rate compared to RBC and online DRL controllers.Moreover,online TL maximises self-sufficiency and self-consumption by 9% and 11% with respect to RBC.Conversely,online TL achieves worse performance compared to offline DRL in either transductive or inductive settings.However,offline Deep Reinforcement Learning(DRL)agents should be trained at least for 15 episodes to reach the same level of performance as the online TL.Therefore,the proposed online TL methodology is effective,completely model-free and it can be directly implemented in real buildings with satisfying performance.

On the applicability of various levels of detail for occupant behavior representation and modeling in building performance simulation

Occupant behavior(OB)is one of the significant sources of uncertainty in building performance simulation.While OB modeling has received increased attention in the past decade,research on the degree of granularity or level of detail(LoD)required for representing occupants is still in the nascent stages.This paper analyzes the modeling and applicability of three LoDs to represent occupants in building performance assessment.A medium-sized prototype office building located in Chicago,Illinois is used as the simulation case study.Ten occupant-centric attributes are adopted to develop the LoDs for OB representation.We first demonstrate the different modeling approaches required for simulating the three fidelity levels.Later,we illustrate the suitability of the developed LoDs in supporting six building performance use cases across different lifecycle stages.This study intends to provide guidance for the building simulation community on appropriate OB representation to support various use cases.

Application and evaluation of a pattern-based building energy model calibration method using public building datasets

Building performance simulation has been adopted to support decision making in the building life cycle.An essential issue is to ensure a building energy simulation model can capture the reality and complexity of buildings and their systems in both the static characteristics and dynamic operations.Building energy model calibration is a technique that takes various types of measured performance data(e.g.,energy use)and tunes key model parameters to match the simulated results with the actual measurements.This study performed an application and evaluation of an automated pattern-based calibration method on commercial building models that were generated based on characteristics of real buildings.A public building dataset that includes high-level building attributes(e.g.,building type,vintage,total floor area,number of stories,zip code)of 111 buildings in San Francisco,California,USA,was used to generate building models in EnergyPlus.Monthly level energy use calibrations were then conducted by comparing building model results against the actual buildings’monthly electricity and natural gas consumption.The results showed 57 out of 111 buildings were successfully calibrated against actual buildings,while the remaining buildings showed opportunities for future calibration improvements.Enhancements to the pattern-based model calibration method are identified to expand its use for:(1)central heating,ventilation and air conditioning(HVAC)systems with chillers,(2)space heating and hot water heating with electricity sources,(3)mixed-use building types,and(4)partially occupied buildings.

Developing occupant archetypes within urban low-income housing: A case study in Mumbai, India

Rapid urbanization pressure and poverty have created a push for affordable housing within the global south.The design of affordable housing can have consequences on the thermal(dis)comfort and behaviour of the occupants,hence requiring an occupant-centric approach to ensure sustainability.This paper investigates occupant behaviour within the urban poor households of Mumbai,India and its impact on their thermal comfort and energy use.This study is a first-of-its-kind attempt to explore the socio-demographic characteristics and energy-related behaviour of low-income occupants within Indian context.Three occupant archetypes,Indifferent Consumers;Considerate Savers;and Conscious Conventionals,were identified from the behavioural and psychographic characteristics gathered through a transverse field survey.A two-step clustering approach was adopted for occupant segmentation that highlighted considerable diversity in occupants’adaptation measures,energy knowledge,energy habits,and their pro-environmental behaviour within similar socio-economic group.Building energy simulation of the representative archetype behaviour estimated up to 37%variations for air-conditioned and up to 8%variation for fan-assisted naturally ventilated housing units during peak summer months.The results from this study establish the significance of occupant factors in shaping energy demand and thermal comfort within low-income housing and pave way for developing occupant-centric building design strategies to serve this marginalized population.The developed low-income occupant archetypes would be useful for architects and energy modelers to generate realistic energy use profiles and improve building performance simulation results.