4.1 Review

Transfer learning for smart buildings: A critical review of algorithms, applications, and future perspectives

相关参考文献

注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article Engineering, Electrical & Electronic

The key role of aggregators in the energy transition under the latest European regulatory framework

Selina Kerscher et al.

Summary: Future energy systems rely on the expansion and management of low-carbon technologies to achieve climate goals. Studies have shown issues with voltage violation and congestion from the integration of high-penetration intermittent renewable energy sources and electric vehicles in distribution networks. Aggregators play a key role in energy market liberalisation and the integration of distributed energy resources in the latest European framework.

INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS (2022)

Review Computer Science, Artificial Intelligence

Transfer learning in demand response: A review of algorithms for data-efficient modelling and control

Thijs Peirelinck et al.

Summary: Automated decision support tools utilizing machine learning and optimization algorithms have become crucial in addressing the increased electricity demand and variability associated with renewable energy sources. Recent advances have shown that transfer learning can significantly improve performance in this context.

ENERGY AND AI (2022)

Article Thermodynamics

Development of an ANN-based building energy model for information-poor buildings using transfer learning

Ao Li et al.

Summary: This study aims to develop a data-driven building energy prediction model through transfer learning with limited training data. The research findings suggest that building usage and industry are crucial factors influencing the effectiveness of transfer learning. Transfer learning can effectively enhance the accuracy of BPNN-based building energy models for information-poor buildings.

BUILDING SIMULATION (2021)

Article Energy & Fuels

Data-driven district energy management with surrogate models and deep reinforcement learning

Giuseppe Pinto et al.

Summary: The study aims to explore the benefits of coordinated energy management in a cluster of buildings using a data-driven approach with LSTM Neural Networks and DRL. Results show that the DRL algorithm is able to significantly reduce electricity costs for the cluster while maintaining indoor comfort conditions.

APPLIED ENERGY (2021)

Article Energy & Fuels

Optimal Scheduling of Microgrid Based on Deep Deterministic Policy Gradient and Transfer Learning

Luqin Fan et al.

Summary: This paper proposes a microgrid optimal scheduling method based on reinforcement learning and transfer learning, aiming to provide flexible and efficient optimal scheduling strategy through the accumulation of scheduling knowledge and utilization of transfer learning.

ENERGIES (2021)

Article Thermodynamics

A hybrid deep transfer learning strategy for short term cross-building energy prediction

Xi Fang et al.

Summary: This study proposes a novel hybrid deep transfer learning strategy for short term cross-building energy prediction, utilizing LSTM and DANN models to enhance energy prediction performance. Experimental results demonstrate significant improvement in building energy prediction compared to other models.

ENERGY (2021)

Article Construction & Building Technology

A general multi-source ensemble transfer learning framework integrate of LSTM-DANN and similarity metric for building energy prediction

Xi Fang et al.

Summary: The study proposes a multi-source ensemble transfer learning framework to enhance building energy prediction performance. Experimental results show that multi-source transfer learning models can improve the prediction performance of target building power consumption compared to single-source models.

ENERGY AND BUILDINGS (2021)

Article Construction & Building Technology

Transfer learning-based strategies for fault diagnosis in building energy systems

Jiangyan Liu et al.

Summary: This study proposed a transfer-learning based methodology for fault diagnosis in building chillers and validated the value of applying transfer learning to FDD in building energy systems, especially when the experimental data available for model development are limited.

ENERGY AND BUILDINGS (2021)

Article Engineering, Electrical & Electronic

A Comprehensive Survey on Transfer Learning

Fuzhen Zhuang et al.

Summary: Transfer learning aims to improve the performance of target learners by transferring knowledge from related source domains, reducing the reliance on target-domain data. This survey aims to systematize and summarize existing research studies in order to help readers understand the current status and ideas in the area of transfer learning.

PROCEEDINGS OF THE IEEE (2021)

Review Green & Sustainable Science & Technology

Data-driven predictive control for unlocking building energy flexibility: A review

Anjukan Kathirgamanathan et al.

Summary: Managing supply and demand in the electricity grid becomes more challenging with the increasing penetration of variable renewable energy sources. Buildings are expected to have an expanding role in the future smart grid through better grid integration and predictive control. Data-driven predictive control, coupled with the Internet of Things, holds promise for scalable and transferrable approaches in grid integration of buildings.

RENEWABLE & SUSTAINABLE ENERGY REVIEWS (2021)

Review Energy & Fuels

A review of machine learning in building load prediction

Liang Zhang et al.

Summary: This paper reviews the application of machine learning techniques in building load prediction, covering tasks T, performance P, and experience E aspects. It discusses the modeling algorithms that improve performance, as well as data engineering for modeling, and concludes with identified gaps and future trends in machine learning application for building energy systems.

APPLIED ENERGY (2021)

Article Thermodynamics

A general transfer learning-based framework for thermal load prediction in regional energy system

Yakai Lu et al.

Summary: This paper presents a transfer learning-based framework for predicting thermal load with limited data. By selecting the optimum source prediction task and utilizing a model-based transfer learning method, the framework achieves the best prediction stability and reduces prediction errors.

ENERGY (2021)

Review Energy & Fuels

Artificial intelligence based anomaly detection of energy consumption in buildings: A review, current trends and new perspectives

Yassine Himeur et al.

Summary: This paper provides a comprehensive review of existing anomaly detection frameworks for building energy consumption based on artificial intelligence, introducing a comprehensive taxonomy and discussing unresolved challenges and future research directions.

APPLIED ENERGY (2021)

Article Energy & Fuels

A framework for estimating the energy-saving potential of occupant behaviour improvement

Zhiyuan He et al.

Summary: This article proposes a holistic survey-and-simulation-based framework for estimating the energy-saving potential of occupant behavior improvement. Through a case study in Singapore, it was found that improving occupant behavior can significantly reduce building energy consumption.

APPLIED ENERGY (2021)

Article Construction & Building Technology

Transfer learning for thermal comfort prediction in multiple cities

Nan Gao et al.

Summary: The HVAC system is crucial in buildings, accounting for up to 40% of energy usage, and maintaining thermal comfort is important for energy efficiency and well-being. Recent advancements in data-driven thermal comfort models have shown better performance than traditional methods, contributing to more accurate predictions despite challenges in data availability.

BUILDING AND ENVIRONMENT (2021)

Article Energy & Fuels

Comparative Evaluation of Non-Intrusive Load Monitoring Methods Using Relevant Features and Transfer Learning

Sarra Houidi et al.

Summary: The paper focuses on an event-based NILM pipeline, with a particular emphasis on the recognition step of HEAs and the selection of relevant and understandable features to discriminate distinct HEAs efficiently. The study introduces a new annotated dataset and investigates the benefits of different feature selection techniques and classification algorithms in improving HEA identification, as well as exploring transfer learning results for the selected features. The best methods investigated in the study outperform previous state-of-the-art results, achieving a maximum recognition accuracy above 99% on the PLAID evaluation dataset.

ENERGIES (2021)

Article Construction & Building Technology

Reinforcement learning of occupant behavior model for cross-building transfer learning to various HVAC control systems

Zhipeng Deng et al.

Summary: This study developed a policy-based reinforcement learning model to predict occupant behavior in adjusting thermostat and clothing level, achieving accurate predictions in different types of buildings through transfer learning. The model was trained using Q-learning in MATLAB and successfully transferred behavior knowledge through different HVAC control systems.

ENERGY AND BUILDINGS (2021)

Article Construction & Building Technology

From time series to image analysis: A transfer learning approach for night setback identification of district heating substations

Fan Zhang et al.

Summary: This study introduces a novel approach using convolutional neural networks to classify night setbacks in district heating substations, aiming to improve energy efficiency. By converting original time series data into heatmap images, the problem is shifted from conventional time series analysis to image classification, enhancing the model's performance and flexibility.

JOURNAL OF BUILDING ENGINEERING (2021)

Article Construction & Building Technology

Transfer learning based methodology for migration and application of fault detection and diagnosis between building chillers for improving energy efficiency

Xu Zhu et al.

Summary: This paper proposes a generic framework for transferring prior knowledge from a source chiller to build a diagnostic model for a new target chiller. The method involves standardizing heterogeneous data, applying domain adaption transfer learning to overcome domain shift, and utilizing domain adversarial neural network to generate the diagnostic model for the target chiller. Results show that the transferred diagnostic model yields decent performance, outperforming conventional machine learning models.

BUILDING AND ENVIRONMENT (2021)

Article Energy & Fuels

A Procedure for Automating Energy Analyses in the BIM Context Exploiting Artificial Neural Networks and Transfer Learning Technique

Mikhail Demianenko et al.

Summary: Building Information Modelling benefits decision-making by testing digital twins for energy consumption and sustainability. Challenges include considering numerous elements and parameters for energy balance adjustment.

ENERGIES (2021)

Article Thermodynamics

Coordinated energy management for a cluster of buildings through deep reinforcement learning

Giuseppe Pinto et al.

Summary: The study explores the use of Deep Reinforcement Learning for coordinated energy management in a cluster of buildings to achieve higher energy flexibility, leading to reduced overall costs and peak demand.

ENERGY (2021)

Article Engineering, Electrical & Electronic

Deep Transfer Learning for Industrial Automation: A Review and Discussion of New Techniques for Data-Driven Machine Learning

Benjamin Maschler et al.

Summary: In recent years, deep learning has significantly enhanced the capabilities of intelligent technical systems, especially in the industrial automation sector, where new data-driven approaches have led to more automation than ever before.

IEEE INDUSTRIAL ELECTRONICS MAGAZINE (2021)

Article Multidisciplinary Sciences

A synthetic building operation dataset

Han Li et al.

Summary: This study presents a synthetic building operation dataset that can be used in various applications, created through annual simulations and historical weather data, including energy benchmarking, model calibration, and policy development.

SCIENTIFIC DATA (2021)

Article Engineering, Multidisciplinary

A combined deep learning application for short term load forecasting

Ilyas Ozer et al.

Summary: Accurate prediction of buildings' load demand is crucial for improving power system reliability and integrating renewable energy sources. Sensor-based electricity prediction is common, but often requires large amounts of data for training machine learning models. The limited historical data of new buildings and meters poses a challenge for sensor-based approaches.

ALEXANDRIA ENGINEERING JOURNAL (2021)

Proceedings Paper Automation & Control Systems

Deep Learning Explicit Differentiable Predictive Control Laws for Buildings

Jan Drgona et al.

Summary: DPC method can learn constrained control laws for unknown nonlinear systems without expert supervision. By optimizing the neural control law and the system model, the parameters of the control law are optimized.

IFAC PAPERSONLINE (2021)

Article Energy & Fuels

Transfer learning applied to DRL-Based heat pump control to leverage microgrid energy efficiency

Paulo Lissa et al.

Summary: This study investigates the application of transfer learning to deep reinforcement learning-based heat pump control in order to improve energy efficiency in a microgrid. The experiments show that the proposed algorithm achieved up to 10% savings after transfer learning was applied, contributing to load-shifting and reducing the learning time by more than a factor of 5.

SMART ENERGY (2021)

Article Computer Science, Artificial Intelligence

Transfer Learning in Smart Environments

Amin Anjomshoaa et al.

Summary: This paper explores the potential of communication and transfer learning between smart environments to enable seamless and automatic transfer of services and machine learning models between intelligent systems.

MACHINE LEARNING AND KNOWLEDGE EXTRACTION (2021)

Article Computer Science, Information Systems

Individualized Short-Term Electric Load Forecasting With Deep Neural Network Based Transfer Learning and Meta Learning

Eunjung Lee et al.

Summary: It is generally believed that individualized models are the best way to predict electric load, but traditional methods tend to favor one-for-all models. This study utilizes transfer learning and meta learning, successfully integrated into deep neural networks, to form high-performance individualized models using individual data in just a few days.

IEEE ACCESS (2021)

Article Computer Science, Hardware & Architecture

Study on Data Transfer in Meteorological Forecast of Small and Medium-Sized Cities and Its Application in Zhaoqing City

Tianwen Huang et al.

COMPUTER JOURNAL (2020)

Article Engineering, Electrical & Electronic

Transfer Learning for Non-Intrusive Load Monitoring

Michele D'Incecco et al.

IEEE TRANSACTIONS ON SMART GRID (2020)

Article Engineering, Electrical & Electronic

Forecasting customers' response to incentives during peak periods: A transfer learning approach

Long Cai et al.

INTERNATIONAL TRANSACTIONS ON ELECTRICAL ENERGY SYSTEMS (2020)

Article Construction & Building Technology

State-of-the-art on research and applications of machine learning in the building life cycle

Tianzhen Hong et al.

ENERGY AND BUILDINGS (2020)

Article Green & Sustainable Science & Technology

Transfer learning with deep neural networks for model predictive control of HVAC and natural ventilation in smart buildings

Yujiao Chen et al.

JOURNAL OF CLEANER PRODUCTION (2020)

Article Construction & Building Technology

A bi-directional missing data imputation scheme based on LSTM and transfer learning for building energy data

Jun Ma et al.

ENERGY AND BUILDINGS (2020)

Review Automation & Control Systems

All you need to know about model predictive control for buildings

Jan Drgona et al.

ANNUAL REVIEWS IN CONTROL (2020)

Review Green & Sustainable Science & Technology

Artificial intelligence and machine learning approaches to energy demand-side response: A systematic review

Ioannis Antonopoulos et al.

RENEWABLE & SUSTAINABLE ENERGY REVIEWS (2020)

Article Multidisciplinary Sciences

Early decarbonisation of the European energy system pays off

Marta Victoria et al.

NATURE COMMUNICATIONS (2020)

Proceedings Paper Computer Science, Artificial Intelligence

ACCURATE DETECTION OF HISTORICAL BUILDINGS USING AERIAL PHOTOGRAPHS AND DEEP TRANSFER LEARNING

Yongzhu Xiong et al.

IGARSS 2020 - 2020 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (2020)

Review Green & Sustainable Science & Technology

Blockchain technology in the energy sector: A systematic review of challenges and opportunities

Merlinda Andoni et al.

RENEWABLE & SUSTAINABLE ENERGY REVIEWS (2019)

Article Computer Science, Artificial Intelligence

A Survey of User-Centred Approaches for Smart Home Transfer Learning and New User Home Automation Adaptation

S. M. Murad Ali et al.

APPLIED ARTIFICIAL INTELLIGENCE (2019)

Article Engineering, Electrical & Electronic

Non-Intrusive Load Monitoring by Voltage-Current Trajectory Enabled Transfer Learning

Yanchi Liu et al.

IEEE TRANSACTIONS ON SMART GRID (2019)

Proceedings Paper Computer Science, Theory & Methods

Transfer Learning for Leisure Centre Energy Consumption Prediction

Paul Banda et al.

COMPUTATIONAL SCIENCE - ICCS 2019, PT I (2019)

Article Computer Science, Theory & Methods

Using transfer learning for smart building management system

Bens Pardamean et al.

JOURNAL OF BIG DATA (2019)

Article Computer Science, Information Systems

Similarity-Based Chained Transfer Learning for Energy Forecasting With Big Data

Yifang Tian et al.

IEEE ACCESS (2019)

Review Thermodynamics

A review of fault detection and diagnostics methods for building systems

Woohyun Kim et al.

SCIENCE AND TECHNOLOGY FOR THE BUILT ENVIRONMENT (2018)

Article Computer Science, Artificial Intelligence

Deep-learning neural-network architectures and methods: Using component based models in building-design energy prediction

Sundaravelpandian Singaravel et al.

ADVANCED ENGINEERING INFORMATICS (2018)

Article Energy & Fuels

Approximate model predictive building control via machine learning

Jan Drgona et al.

APPLIED ENERGY (2018)

Article Construction & Building Technology

Transfer learning with seasonal and trend adjustment for cross-building energy forecasting

Mauro Ribeiro et al.

ENERGY AND BUILDINGS (2018)

Review Green & Sustainable Science & Technology

A review of data-driven building energy consumption prediction studies

Kadir Amasyali et al.

RENEWABLE & SUSTAINABLE ENERGY REVIEWS (2018)

Article Computer Science, Information Systems

A Scalable Room Occupancy Prediction with Transferable Time Series Decomposition of CO2 Sensor Data

Irvan B. Arief-Ang et al.

ACM TRANSACTIONS ON SENSOR NETWORKS (2018)

Article Construction & Building Technology

Generalized online transfer learning for climate control in residential buildings

Thomas Grubinger et al.

ENERGY AND BUILDINGS (2017)

Article Green & Sustainable Science & Technology

Smart meter deployment in Europe: A comparative case study on the impacts of national policy schemes

Shan Zhou et al.

JOURNAL OF CLEANER PRODUCTION (2017)

Review Green & Sustainable Science & Technology

Data science for building energy management: A review

Miguel Molina-Solana et al.

RENEWABLE & SUSTAINABLE ENERGY REVIEWS (2017)

Article Computer Science, Artificial Intelligence

A Feature-Based Knowledge Transfer Framework for Cross-Environment Activity Recognition Toward Smart Home Applications

Yi-Ting Chiang et al.

IEEE TRANSACTIONS ON HUMAN-MACHINE SYSTEMS (2017)

Article Engineering, Electrical & Electronic

Cooperative MPC-Based Energy Management for Networked Microgrids

Alessandra Parisio et al.

IEEE TRANSACTIONS ON SMART GRID (2017)

Article Construction & Building Technology

Unsupervised energy prediction in a Smart Grid context using reinforcement cross-building transfer learning

Elena Mocanu et al.

ENERGY AND BUILDINGS (2016)

Review Green & Sustainable Science & Technology

Energy savings by energy management systems: A review

Dasheng Lee et al.

RENEWABLE & SUSTAINABLE ENERGY REVIEWS (2016)

Review Computer Science, Information Systems

Load Profiling and Its Application to Demand Response: A Review

Yi Wang et al.

TSINGHUA SCIENCE AND TECHNOLOGY (2015)

Article Computer Science, Artificial Intelligence

A Survey on Transfer Learning

Sinno Jialin Pan et al.

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING (2010)