Energy Efficiency through the Implementation of an AI Model to Predict Room Occupancy Based on Thermal Comfort Parameters

Abstract

Room occupancy prediction based on indoor environmental quality may be the breakthrough to ensure energy efficiency and establish an interior ambience tailored to each user. Identifying whether temperature, humidity, lighting, and CO2 levels may be used as efficient predictors of room occupancy accuracy is needed to help designers better utilize the readings and data collected in order to improve interior design, in an effort to better suit users. It also aims to help in energy efficiency and saving in an ever-increasing energy crisis and dangerous levels of climate change. This paper evaluated the accuracy of room occupancy recognition using a dataset with diverse amounts of light, CO2, and humidity. As classification algorithms, K-nearest neighbors (KNN), hybrid Adam optimizer–artificial neural network–back-propagation network (AO–ANN (BP)), and decision trees (DT) were used. Furthermore, this research is based on machine learning interpretability methodologies. Shapley additive explanations (SHAP) improve interpretability by estimating the significance values for each feature for classifiers applied. The results indicate that the KNN performs better than the DT and AO-ANN (BP) classification models have 99.5%. Though the two classifiers are designed to evaluate variations in interpretations, we must ensure that they have accurate detection. The results show that SHAP provides successful implementation following these metrics, with differences detected amongst classifier models that support the assumption that model complexity plays a significant role when predictability is taken into account.

AI explainability and governance in smart energy systems: A review

“…The least research is in AI Behaviour and Governance where much effort is needed in the future. The methods and tools to support trustworthiness (explainability and other AI traits) in AI for energy systems include, among others, visual explanation techniques using Gradient-weighted Class Activation Mapping (Grad-CAM) (Ardito et al, 2022), sequenceto-sequence RNN methods for visual explanation of short-term load forecasting (Gürses-Tran et al, 2022), the Scale-Invariant Feature Transform (SIFT) method (Singstock et al, 2021), post hoc interpretability (Allen and Tkatchenko, 2022), SHapley Additive exPlanation (SHAP) (Pinson et al, 2021;Abdel-Razek et al, 2022;Kruse et al, 2022), interpretable Tiny Neural Networks (TNN) (Longmire and Banuti, 2022), model-agnostic methods (Gürses-Tran et al, 2022), the use of Temporal Fusion Transformer (TFT) method to enhance interpretability (López Santos et al, 2022), the decision tree and Classification and Regression Tree (CART) algorithms for ML explainability (Sun et al, 2021), visual data exploration for the interpretability of fault diagnosis (Landwehr et al, 2022), a partially interpretable method using Long short-term memory (LSTM) and MLP (multilayer perceptron) for short-term load forecasting (Xie et al, 2021), and Local Interpretable Model-Agnostic Explanation (lime) (Tsoka et al, 2022). We expect that many more methods will be developed for XAI in the future.…”

“…It captures different dimensions of "Energy-Efficient Buildings," including the generalizability of IML for estimating building energy consumption and making buildings more energy efficient (Manfren et al, 2022), classification of building energy performance certificates using XAI (Tsoka et al, 2022), and XAI approach for forecasting long-term building energy consumption (Wenninger et al, 2022). Further dimensions include providing smart recommendations based on XAI for evaluating building energy efficiency systems (Himeur et al, 2022), improving the effective use of energy by adopting an IML model to forecast room occupancy (Abdel-Razek et al, 2022), predicting energy consumption in buildings, and machine learning-based models for energy efficiency in buildings.…”

Development of A User-Centered Design Framework for Palliative and Hospice Care Patients for a Better Quality of Life Experience

“…Personalization of the interior space includes personalized thermal comfort, in contrast to a central HVAC facility that may become an irritation rather than a convenience (Abdel-Razek et al, 2022). Patients with terminal illness are usually immuno-compromised and may thus be more sensitive to temperature than others, this is also affected by their metabolic rate, medication and type of illness.…”

Open Issues for Intelligent Connectivity Wireless Sensor Networks (WSNs) and IoT: State of the Art

“…The difficult deployment conditions may expose WSNs to failure [19]. In the network region, sensor nodes should be autonomous [20], [21], as this enables all nodes to communicate wirelessly and alters the topology. As a result, various attacks present a chance for numerous security issues [22], [23].…”