Explainable AI for building energy retrofitting under data scarcity

• URL: http://arxiv.org/abs/2504.06055v1
• Date: Tue, 08 Apr 2025 14:00:08 GMT
• Title: Explainable AI for building energy retrofitting under data scarcity
• Authors: Panagiota Rempi, Sotiris Pelekis, Alexandros Menelaos Tzortzis, Evangelos Karakolis, Christos Ntanos, Dimitris Askounis,
• Abstract summary: This study presents an Artificial Intelligence (AI) and Machine Learning (ML)-based framework to recommend energy efficiency measures for residential buildings.<n>Using Latvia as a case study, the methodology addresses challenges associated with limited datasets, class imbalance and data scarcity.<n>The evaluation of the approach shows that it notably overcomes data limitations, achieving improvements up to 54% in precision, recall and F1 score.
• Score: 40.14307808809578
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Enhancing energy efficiency in residential buildings is a crucial step toward mitigating climate change and reducing greenhouse gas emissions. Retrofitting existing buildings, which account for a significant portion of energy consumption, is critical particularly in regions with outdated and inefficient building stocks. This study presents an Artificial Intelligence (AI) and Machine Learning (ML)-based framework to recommend energy efficiency measures for residential buildings, leveraging accessible building characteristics to achieve energy class targets. Using Latvia as a case study, the methodology addresses challenges associated with limited datasets, class imbalance and data scarcity. The proposed approach integrates Conditional Tabular Generative Adversarial Networks (CTGAN) to generate synthetic data, enriching and balancing the dataset. A Multi-Layer Perceptron (MLP) model serves as the predictive model performing multi-label classification to predict appropriate retrofit strategies. Explainable Artificial Intelligence (XAI), specifically SHapley Additive exPlanations (SHAP), ensures transparency and trust by identifying key features that influence recommendations and guiding feature engineering choices for improved reliability and performance. The evaluation of the approach shows that it notably overcomes data limitations, achieving improvements up to 54% in precision, recall and F1 score. Although this study focuses on Latvia, the methodology is adaptable to other regions, underscoring the potential of AI in reducing the complexity and cost of building energy retrofitting overcoming data limitations. By facilitating decision-making processes and promoting stakeholders engagement, this work supports the global transition toward sustainable energy use in the residential building sector.

Related papers

• DSMoE: Matrix-Partitioned Experts with Dynamic Routing for Computation-Efficient Dense LLMs [70.91804882618243]
  This paper proposes DSMoE, a novel approach that achieves sparsification by partitioning pre-trained FFN layers into computational blocks.<n>We implement adaptive expert routing using sigmoid activation and straight-through estimators, enabling tokens to flexibly access different aspects of model knowledge.<n>Experiments on LLaMA models demonstrate that under equivalent computational constraints, DSMoE achieves superior performance compared to existing pruning and MoE approaches.
  arXiv  Detail & Related papers  (2025-02-18T02:37:26Z)
• AI4EF: Artificial Intelligence for Energy Efficiency in the Building Sector [38.48549968280562]
  AI4EF is an advanced user-centric tool designed to support decision-making in building energy retrofitting and efficiency optimization.<n>Leveraging machine learning (ML) and data-driven insights, AI4EF enables stakeholders to model, analyze, and predict energy consumption, retrofit costs, and environmental impacts.
  arXiv  Detail & Related papers  (2024-12-05T10:36:39Z)
• A Gap in Time: The Challenge of Processing Heterogeneous IoT Data in Digitalized Buildings [15.525789412274587]
  This study investigates the diverse dimensions of IoT data heterogeneity in both intra-building and inter-building contexts. The results emphasize the critical need for multi-modal data integration, domain-informed modeling, and automated data engineering pipelines.
  arXiv  Detail & Related papers  (2024-05-23T07:45:48Z)
• Model Failure or Data Corruption? Exploring Inconsistencies in Building Energy Ratings with Self-Supervised Contrastive Learning [4.152601330407538]
  Building Energy Rating (BER) stands as a pivotal metric, enabling building owners, policymakers, and urban planners to understand the energy-saving potential through improving building energy efficiency. Yet, the BER assessment process is vulnerable to missing and inaccurate measurements. We introduce ttCLEAR, a data-driven approach designed to scrutinize the inconsistencies in BER assessments through self-supervised contrastive learning.
  arXiv  Detail & Related papers  (2024-04-14T17:07:11Z)
• Empowering Aggregators with Practical Data-Driven Tools: Harnessing Aggregated and Disaggregated Flexibility for Demand Response [37.69303106863453]
  This study explores the interaction between aggregators and building occupants in activating flexibility through Demand Response (DR) programs.<n>It introduces a methodology of optimizing aggregated flexibility provision strategies in environments with limited data.
  arXiv  Detail & Related papers  (2024-01-19T14:43:04Z)
• Filling the Missing: Exploring Generative AI for Enhanced Federated Learning over Heterogeneous Mobile Edge Devices [72.61177465035031]
  We propose a generative AI-empowered federated learning to address these challenges by leveraging the idea of FIlling the MIssing (FIMI) portion of local data. Experiment results demonstrate that FIMI can save up to 50% of the device-side energy to achieve the target global test accuracy.
  arXiv  Detail & Related papers  (2023-10-21T12:07:04Z)
• Towards Energy-Aware Federated Traffic Prediction for Cellular Networks [2.360352205004026]
  We propose a novel sustainability indicator that allows assessing the feasibility of machine learning (ML) models. We evaluate state-of-the-art deep learning (DL) architectures in a federated scenario using real-world measurements from base station (BS) sites in the area of Barcelona, Spain. Our findings indicate that larger ML models achieve marginally improved performance but have a significant environmental impact in terms of carbon footprint.
  arXiv  Detail & Related papers  (2023-09-19T14:28:09Z)
• Efficiency Pentathlon: A Standardized Arena for Efficiency Evaluation [82.85015548989223]
  Pentathlon is a benchmark for holistic and realistic evaluation of model efficiency. Pentathlon focuses on inference, which accounts for a majority of the compute in a model's lifecycle. It incorporates a suite of metrics that target different aspects of efficiency, including latency, throughput, memory overhead, and energy consumption.
  arXiv  Detail & Related papers  (2023-07-19T01:05:33Z)
• Low Emission Building Control with Zero-Shot Reinforcement Learning [70.70479436076238]
  Control via Reinforcement Learning (RL) has been shown to significantly improve building energy efficiency. We show it is possible to obtain emission-reducing policies without a priori--a paradigm we call zero-shot building control.
  arXiv  Detail & Related papers  (2022-08-12T17:13:25Z)
• Data-Centric Green AI: An Exploratory Empirical Study [6.4265933507484]
  We investigate the impact of data-centric approaches on AI energy efficiency. Our results show evidence that, by exclusively conducting modifications on datasets, energy consumption can be drastically reduced. Our results call for a research agenda that focuses on data-centric techniques to further enable and democratize Green AI.
  arXiv  Detail & Related papers  (2022-04-06T12:22:43Z)
• Counterfactual Explanations as Interventions in Latent Space [62.997667081978825]
  Counterfactual explanations aim to provide to end users a set of features that need to be changed in order to achieve a desired outcome. Current approaches rarely take into account the feasibility of actions needed to achieve the proposed explanations. We present Counterfactual Explanations as Interventions in Latent Space (CEILS), a methodology to generate counterfactual explanations.
  arXiv  Detail & Related papers  (2021-06-14T20:48:48Z)
• AI Chiller: An Open IoT Cloud Based Machine Learning Framework for the Energy Saving of Building HVAC System via Big Data Analytics on the Fusion of BMS and Environmental Data [12.681421165031576]
  Energy saving and carbon emission reduction in buildings is one of the key measures in combating climate change. The optimization of chiller system power consumption had been extensively studied in the mechanical engineering and building service domains. With the advance of big data and AI, the adoption of machine learning into the optimization problems becomes popular.
  arXiv  Detail & Related papers  (2020-10-09T09:51:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.