Multi-objective optimisation for tuning building heating and cooling loads forecasting models

Saleh Seyedzadeh, Farzad Pour Rahimian, Parag Rastogi, Stephen Oliver, Ivan Glesk, Bimal Kumar

Research output: Contribution to conferencePaper

Abstract

Machine learning (ML) has been recognised as a powerful method for modelling building energy consumption. The capability of ML to provide a fast and accurate prediction of energy loads makes it an ideal tool for decision-making tasks related to sustainable design and retrofit planning. However, the accuracy of these ML models is much dependant on the selection of the right hyper-parameters for specific building dataset. This paper proposes a method for optimising ML model for forecasting both heating and cooling loads. The technique employs multi-objective optimisation with evolutionary algorithms to search the space of possible parameters. The proposed approach not only tune one model to precisely predict building energy loads but also accelerates the process of model optimisation. The study utilises a simulated building energy data generated in EnergyPlus to demonstrate the efficiency of the proposed method, and compares the outcomes with the regular ML tuning procedure (i.e. grid search). The optimised model provides a reliable tool for building designers and engineers to explore a large space of the available building materials and technologies.

Conference

Conference36th CIB W78 2019 Conference
Abbreviated titleCIB W78
CountryUnited Kingdom
CityNewcastle
Period18/09/1920/09/19

Fingerprint

Multiobjective optimization
Learning systems
Loads (forces)
Tuning
Cooling
Heating
Dynamic loads
Evolutionary algorithms
Energy utilization
Decision making
Engineers
Planning

Keywords

  • building energy loads
  • building energy prediction
  • machine learning
  • model optimisation

Cite this

Seyedzadeh, S., Pour Rahimian, F., Rastogi, P., Oliver, S., Glesk, I., & Kumar, B. (2019). Multi-objective optimisation for tuning building heating and cooling loads forecasting models. Paper presented at 36th CIB W78 2019 Conference, Newcastle, United Kingdom.
Seyedzadeh, Saleh ; Pour Rahimian, Farzad ; Rastogi, Parag ; Oliver, Stephen ; Glesk, Ivan ; Kumar, Bimal. / Multi-objective optimisation for tuning building heating and cooling loads forecasting models. Paper presented at 36th CIB W78 2019 Conference, Newcastle, United Kingdom.10 p.
@conference{b34e7c361b30447e98082cdd7d6c7654,
title = "Multi-objective optimisation for tuning building heating and cooling loads forecasting models",
abstract = "Machine learning (ML) has been recognised as a powerful method for modelling building energy consumption. The capability of ML to provide a fast and accurate prediction of energy loads makes it an ideal tool for decision-making tasks related to sustainable design and retrofit planning. However, the accuracy of these ML models is much dependant on the selection of the right hyper-parameters for specific building dataset. This paper proposes a method for optimising ML model for forecasting both heating and cooling loads. The technique employs multi-objective optimisation with evolutionary algorithms to search the space of possible parameters. The proposed approach not only tune one model to precisely predict building energy loads but also accelerates the process of model optimisation. The study utilises a simulated building energy data generated in EnergyPlus to demonstrate the efficiency of the proposed method, and compares the outcomes with the regular ML tuning procedure (i.e. grid search). The optimised model provides a reliable tool for building designers and engineers to explore a large space of the available building materials and technologies.",
keywords = "building energy loads, building energy prediction, machine learning, model optimisation",
author = "Saleh Seyedzadeh and {Pour Rahimian}, Farzad and Parag Rastogi and Stephen Oliver and Ivan Glesk and Bimal Kumar",
year = "2019",
month = "9",
day = "18",
language = "English",
note = "36th CIB W78 2019 Conference, CIB W78 ; Conference date: 18-09-2019 Through 20-09-2019",

}

Seyedzadeh, S, Pour Rahimian, F, Rastogi, P, Oliver, S, Glesk, I & Kumar, B 2019, 'Multi-objective optimisation for tuning building heating and cooling loads forecasting models' Paper presented at 36th CIB W78 2019 Conference, Newcastle, United Kingdom, 18/09/19 - 20/09/19, .

Multi-objective optimisation for tuning building heating and cooling loads forecasting models. / Seyedzadeh, Saleh; Pour Rahimian, Farzad; Rastogi, Parag; Oliver, Stephen; Glesk, Ivan; Kumar, Bimal.

2019. Paper presented at 36th CIB W78 2019 Conference, Newcastle, United Kingdom.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Multi-objective optimisation for tuning building heating and cooling loads forecasting models

AU - Seyedzadeh, Saleh

AU - Pour Rahimian, Farzad

AU - Rastogi, Parag

AU - Oliver, Stephen

AU - Glesk, Ivan

AU - Kumar, Bimal

PY - 2019/9/18

Y1 - 2019/9/18

N2 - Machine learning (ML) has been recognised as a powerful method for modelling building energy consumption. The capability of ML to provide a fast and accurate prediction of energy loads makes it an ideal tool for decision-making tasks related to sustainable design and retrofit planning. However, the accuracy of these ML models is much dependant on the selection of the right hyper-parameters for specific building dataset. This paper proposes a method for optimising ML model for forecasting both heating and cooling loads. The technique employs multi-objective optimisation with evolutionary algorithms to search the space of possible parameters. The proposed approach not only tune one model to precisely predict building energy loads but also accelerates the process of model optimisation. The study utilises a simulated building energy data generated in EnergyPlus to demonstrate the efficiency of the proposed method, and compares the outcomes with the regular ML tuning procedure (i.e. grid search). The optimised model provides a reliable tool for building designers and engineers to explore a large space of the available building materials and technologies.

AB - Machine learning (ML) has been recognised as a powerful method for modelling building energy consumption. The capability of ML to provide a fast and accurate prediction of energy loads makes it an ideal tool for decision-making tasks related to sustainable design and retrofit planning. However, the accuracy of these ML models is much dependant on the selection of the right hyper-parameters for specific building dataset. This paper proposes a method for optimising ML model for forecasting both heating and cooling loads. The technique employs multi-objective optimisation with evolutionary algorithms to search the space of possible parameters. The proposed approach not only tune one model to precisely predict building energy loads but also accelerates the process of model optimisation. The study utilises a simulated building energy data generated in EnergyPlus to demonstrate the efficiency of the proposed method, and compares the outcomes with the regular ML tuning procedure (i.e. grid search). The optimised model provides a reliable tool for building designers and engineers to explore a large space of the available building materials and technologies.

KW - building energy loads

KW - building energy prediction

KW - machine learning

KW - model optimisation

UR - http://cibw78.northumbria-eee.co.uk/

M3 - Paper

ER -

Seyedzadeh S, Pour Rahimian F, Rastogi P, Oliver S, Glesk I, Kumar B. Multi-objective optimisation for tuning building heating and cooling loads forecasting models. 2019. Paper presented at 36th CIB W78 2019 Conference, Newcastle, United Kingdom.