AI in HVAC fault detection and diagnosis: A systematic review

Energy Reviews, Journal Year: 2024, Volume and Issue: 3(2), P. 100071 - 100071

Published: Feb. 9, 2024

Recent studies show that artificial intelligence (AI), such as machine learning and deep learning, models can be adopted have advantages in fault detection diagnosis for building energy systems. This paper aims to conduct a comprehensive systematic literature review on (FDD) methods heating, ventilation, air conditioning (HVAC) covers the period from 2013 2023 identify analyze existing research this field. Our work concentrates explicitly synthesizing AI-based FDD techniques, particularly summarizing these offering classification. First, we discuss challenges while developing HVAC Next, classify into three categories: those based traditional hybrid AI models. Additionally, also examine physical model-based compare them with methods. The analysis concludes FDD, despite its higher accuracy reduced reliance expert knowledge, has garnered considerable interest compared physics-based However, it still encounters difficulties dynamic time-varying environments achieving resolution. Addressing is essential facilitate widespread adoption of HVAC.

Language: Английский

---

Towards scalable physically consistent neural networks: An application to data-driven multi-zone thermal building models

Applied Energy, Journal Year: 2023, Volume and Issue: 340, P. 121071 - 121071

Published: April 5, 2023

With more and data being collected, data-driven modeling methods have been gaining in popularity recent years. While physically sound, classical gray-box models are often cumbersome to identify scale, their accuracy might be hindered by limited expressiveness. On the other hand, black-box methods, typically relying on Neural Networks (NNs) nowadays, achieve impressive performance, even at deriving statistical patterns from data. However, they remain completely oblivious underlying physical laws, which may lead potentially catastrophic failures if decisions for real-world systems based them. Physically Consistent (PCNNs) were recently developed address these aforementioned issues, ensuring consistency while still leveraging NNs attain state-of-the-art accuracy, applied zone temperature modeling. In this work, we scale PCNNs model dynamics of buildings with several connected thermal zones propose a thorough comparison methods. More precisely, design three distinct PCNN extensions different levels information sharing between modeled zones, thereby exemplifying modularity flexibility architecture, formally prove consistency. presented case study, shown outperforming NN-based despite constrained structure. Our investigations furthermore provide clear illustration achieving seemingly good performance remaining physics-agnostic, can misleading practice. comes cost computational complexity, hand show improvements 17–35% compared all consistent paving way scalable performance.

Language: Английский

---

Energy flexibility quantification of a tropical net-zero office building using physically consistent neural network-based model predictive control

Advances in Applied Energy, Journal Year: 2024, Volume and Issue: 14, P. 100167 - 100167

Published: Feb. 24, 2024

Building energy flexibility plays a critical role in demand-side management for reducing utility costs building owners and sustainable, reliable, smart grids. Realizing tropical regions requires solar photovoltaics storage systems. However, quantifying the of buildings utilizing such technologies has yet to be explored, robust control sequence is needed this scenario. Hence, work presents case study evaluate controls operations net-zero office Singapore. The utilizes data-driven quantification workflow employs novel model predictive (MPC) framework based on physically consistent neural network (PCNN) optimize flexibility. To best our knowledge, first instance that PCNN applied mathematical MPC setting, stability system formally proved. Three scenarios are evaluated compared: default regulated flat tariff, real-time pricing mechanism, an on-site battery (BESS). Our findings indicate incorporating into could more beneficial leverage decisions than flat-rate approach. Moreover, adding BESS PV generation improved self-sufficiency self-consumption by 17% 20%, respectively. This integration also addresses mismatch issues within framework, thus ensuring reliable local supply. Future research can proposed PCNN-MPC different types.

Language: Английский

---

Application of physics-informed neural networks in fault diagnosis and fault-tolerant control design for electric vehicles: A review

Measurement, Journal Year: 2025, Volume and Issue: unknown, P. 116728 - 116728

Published: Jan. 1, 2025

Language: Английский

---

A review on full-, zero-, and partial-knowledge based predictive models for industrial applications

Information Fusion, Journal Year: 2025, Volume and Issue: unknown, P. 102996 - 102996

Published: Feb. 1, 2025

Language: Английский

---

An AI framework integrating physics-informed neural network with predictive control for energy-efficient food production in the built environment

Applied Energy, Journal Year: 2023, Volume and Issue: 348, P. 121450 - 121450

Published: July 3, 2023

Language: Английский

---

A district-scale spatial distribution evaluation method of rooftop solar energy potential based on deep learning

Solar Energy, Journal Year: 2023, Volume and Issue: 268, P. 112282 - 112282

Published: Dec. 27, 2023

Language: Английский

---

Probabilistic machine learning for enhanced chiller sequencing: A risk-based control strategy

Energy and Built Environment, Journal Year: 2024, Volume and Issue: unknown

Published: March 1, 2024

Multiple-chiller systems are widely adopted in large buildings due to their high flexibility and efficiency providing cooling capacity. A reliable robust chiller sequencing control strategy is crucial ensure the energy stability of multiple-chiller systems. However, conventional strategies usually based on real-time measured load without considering changes following hours. Conventional rule-based may result unnecessary switching off, leading waste impairing system stability. Therefore, this study proposes a that utilizes probabilistic predictions. 1h-ahead prediction form normal distribution made using natural gradient boosting (NGBoost). Compared machine learning algorithms, NGBoost can predict not only future but also uncertainty predicted load, which enables handle uncertainties associated with data/measurements adequately. novel risk-based developed The data experiment shows proposed significantly improve reliability plant by reducing total number up 43.6%.

Language: Английский

---

A Future Direction of Machine Learning for Building Energy Management: Interpretable Models

Energies, Journal Year: 2024, Volume and Issue: 17(3), P. 700 - 700

Published: Feb. 1, 2024

Machine learning (ML) algorithms are now part of everyday life, as many technological devices use these algorithms. The spectrum uses is wide, but it evident that ML represents a revolution may change almost every human activity. However, for all innovations, comes with challenges. One the most critical challenges providing users an understanding how models’ output related to input data. This called “interpretability”, and focused on explaining what feature influences model’s output. Some have simple easy-to-understand relationship between output, while other models “black boxes” return without giving user information influenced it. lack this knowledge creates truthfulness issue when inspected by human, especially operator not data scientist. Building Construction sector starting face innovation, its scientific community working define best practices models. work intended developing deep analysis determine interpretable could be among promising future technologies energy management in built environments.

Language: Английский

---

A review on hybrid physics and data-driven modeling methods applied in air source heat pump systems for energy efficiency improvement

Renewable and Sustainable Energy Reviews, Journal Year: 2024, Volume and Issue: 204, P. 114804 - 114804

Published: Aug. 14, 2024

Language: Английский

---