Office Building’s Occupancy Prediction Using Extreme Learning Machine Model with Different Optimization Algorithms

Abstract

Increasing energy efficiency requirements lead to lower energy consumption in buildings, but at the same time occupants’ influence on the energy balance of the building during the use phase becomes more crucial. The randomness of the building’s occupancy often leads to the mismatch of the predicted and measured energy demand, also called Energy Performance Gap. Therefore, prediction of occupancy is important both in the design and use phases of the building. The goal of the study is to apply Extreme Learning Machine (ELM) models with different optimisation algorithms – Genetic (GA-ELM) and Simulated Annealing (SA–ELM) for occupancy prediction in an office building based on measured CO2 concentrations. Both models show similar and high accuracy of prediction: R2 – 0.73–0.74 and RMSE – 1.8–1.9 for the whole measured period. Influence of population size, number of neurons, and number of iterations on results accuracy was also analysed and recommendations are given. It was concluded that both methods are suitable for occupancy prediction, but because of different simulation times, SA-ELM is recommended for the Building Management Systems (BMS), where higher speed is required.

Quality Evaluation and Satisfaction Analysis of Online Learning of College Students Based on Artificial Intelligence

“…ELM can randomly generate independent samples w and b before training, and the hidden neuron threshold β can be calculated by determining L and H [19]. To further analyze the in uence of parameters on ELM, the weight proportion curves under di erent parameters were drawn, as shown in Figure 3.…”

Office buildings occupancy analysis and prediction associated with the impact of the COVID-19 pandemic

“…As it was concluded in previous studies ( Bielskus et. al, 2020 ; Motuzienė et al, 2021 ), a sufficient measurement period is 4 weeks. Here periods are much longer, but prediction accuracy is low (R 2 =0.27) for the period of strict quarantine and increases for light quarantine and post-quarantine accordingly up to R 2 =0.50; 0.56.…”

“…Hypothesis, that prediction reliability may be influenced by a drastic decrease of the occupancy of the building during pandemics, is checked on Building A, as it has long-term measurement both for occupancy and for CO 2 . CO 2 is considered the most suitable indoor parameter for occupancy prediction ( Bielskus et al, 2020 ; Motuzienė et al, 2021 ). All the data gathered for this building are processed dividing them into 5 min.…”

“…Different improvements of the algorithm have been proposed, thus enabling its applications also for big data analysis Chen et al (2017) . Hence, the potential of the ELM until now is not sufficiently exploited in the field of the building's occupancy prediction, where both prediction accuracy and speed of the prediction is important as it is related to the control of the building services systems ( Motuzienė et al, 2021 ).…”

“…The paper presented bellow also deals with the occupancy data collection issues and COVID-19 influence on machine learning based occupancy prediction models reliability. Authors of the paper already have studied the applicability of few ELM methods ( Bielskus et al 2020 ; Motuzienė et al, 2021 ) and concluded that ELM–SA (Extreme Learning Machine model combined with Simulated Annealing optimisation algorithm) methods is reliable and fast for the prediction of the occupancy. Therefore, it has a high potential for integration into office buildings’ BMS to increase their efficient and sustainable operation.…”