Open Access
Review
Table 1
Summary of the related works based on the data-driven model.
| Work source | Key findings | Performance |
|---|---|---|
| Mavromatidis et al. (2014) | For the estimation of the daylight factor, a polynomial-based regression model was proposed. | Prediction of daylight factor with root mean square levels 0.99 by a set of independent variables |
| Borile et al. (2017) | Light sensors on luminaires and a workplace reference point were used to collect experimental data. | Adjustable dimming levels; Minimized energy use |
| Le et al. (2014) | Developed using a Support Vector Machine (SVM)-based Machine Learning (ML) method. | Thermal discomfort reduction through blind control with SVM technique |
| Caicedo and Pandharipande (2016) | Support Vector Regression (SVR)-based energy consumption model | Energy savings |
| Xiong and Tzempelikos (2016) | Model-based control (MBC) techniques for lighting and shading controls were developed; Minimum sensor inputs (irradiation, indoor/outdoor illuminance) were considered. | Satisfies the visual comfort criteria; Maximize daylight utilization, Minimize lighting energy use. DGP values remained below 0.35; lighting energy use was reduced |
| Sadeghi et al. (2017) | Predict electric light dimming and roller shade lowering/raising actions. The intermediate operational states of electric lighting and shading systems were predicted with Bayesian regression models. | Data on environmental parameters, individual characteristics and human attributes governing human-shading and – electric lighting interactions |
| Gunay et al. (2017) | “Discrete-time Markov logistic regression models from the light-on and blind-closing behaviours in recursion” were developed. | Modest electricity energy reductions of 0.9 kWh/m2 only with adaptive lighting controls and 1.2 kWh/m2 with both adaptive lighting and blind controls |
| Sanjeev Kumar et al. (2020b) | Real-time data driven predictive models for thermal and visual comfort. Luminaire dimming and ensemble learning based window blind position prediction; Daylight glare assessment models using SVM, decision tree, ensemble tree, and ANN | The model consumes 17% less power than the uncontrolled system and 15% less power than the baseline system |
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.