Junhui Huang

Mr. Junhui Huang


Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 September 2022 in Applied Sciences
Verified
Journal Article
Open access
PDF file
JHJunhui Huang
Junhui Huang
MAMohammed Algahtani
Mohammed Algahtani
Sakdirat KaewunruenSakdirat Kaewunruen
Sakdirat Kaewunruen
A primary energy consumption and CO2 emission source stems from buildings and infrastructures due to rapid urbanisation and social development. An accurate method to forecast energy consumption in a building is thus critically needed to enable successful management of adaptive energy consumption and ease the level of CO2 emission. However, energy forecasting for buildings, especially residential buildings, has several challenges, such as significant variations in energy usage patterns due to unpredicted demands of the residences and some intricate factors, which can randomly affect the patterns. Traditional forecasting approaches require a tremendous number of inputs needed for building physic models and variations often exist between as-built and as-designed buildings in reality. Most recent studies have adopted only ambient weather conditions, building components, and the occupant’s behaviours. As a result, in order to take into account the complexity of factors that can affect the building energy model development and its computation, we develop advanced machine learning models driven by the inherent electricity consumption pattern associated with the day and time. In this study, we demonstrate benchmarking results derived from three different machine learning algorithms, namely SVR, XGBoost, and LSTM, trained by using 1-year datasets with sub-hourly (30 min) temporal granularity to determine the outperformed predictor. Ultimately, the machine learning model robustness and performance on a basis of the coefficient of variation (CV) obtained by the SVR is benchmarked across XGBoost and LSTM trained by the same datasets containing attributes related to the building type, data size, and temporal granularity. The insight stemming from this study indicates that the suitable choice of the machine learning models for building energy forecasts largely depends on the natural characteristics of building energy data. Hyperparameter tuning or mathematical modification within an algorithm may not be sufficient to attain the most accurate machine learning model for building energy forecast.

15 January 2023 in Energies
Verified
Journal Article
Open access
PDF file
JHJunhui Huang
Junhui Huang
Sakdirat KaewunruenSakdirat Kaewunruen
Sakdirat Kaewunruen
Most of the Artificial Intelligence (AI) models currently used in energy forecasting are traditional and deterministic. Recently, a novel deep learning paradigm, called ‘transformer’, has been developed, which adopts the mechanism of self-attention. Transformers are designed to better process and predict sequential data sets (i.e., historical time records) as well as to track any relationship in the sequential data. So far, a few transformer-based applications have been established, but no industry-scale application exists to build energy forecasts. Accordingly, this study is the world’s first to establish a transformer-based model to estimate the energy consumption of a real-scale university library and benchmark with a baseline model (Support Vector Regression) SVR. With a large dataset from 1 September 2017 to 13 November 2021 with 30 min granularity, the results using four historical electricity readings to estimate one future reading demonstrate that the SVR (an R2 of 0.92) presents superior performance than the transformer-based model (an R2 of 0.82). Across the sensitivity analysis, the SVR model is more sensitive to the input close to the output. These findings provide new insights into the research area of energy forecasting in either a specific building or a building cluster in a city. The influences of the number of inputs and outputs related to the transformer-based model will be investigated in the future.

13 October 2022 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Sakdirat KaewunruenSakdirat Kaewunruen
Sakdirat Kaewunruen
JSJessada Sresakoolchai
Jessada Sresakoolchai
JHJunhui Huang
Junhui Huang
YZYingyu Zhu
Yingyu Zhu
Chayut NgamkhanongChayut Ngamkhanong
Chayut Ngamkhanong
ARAlex M. Remennikov
Alex M. Remennikov
The state-of-the-art design methods for railway prestressed concrete sleepers are currently based on the quasi-static stresses resulting from a simplification of dynamic wheel loads, and subsequently the quasi-static responses of concrete sleepers. This method has been widely used in practices to overcome the complexity of dynamic analysis and testing. A single load factor (or called dynamic impact factor) for a partial safety-factored design (or k factors for the test criteria) is commonly used to crudely account for dynamic train–track interactions over different levels of track irregularities. The dynamic impact factors for either design or testing are usually obtained from either (i) railway infrastructure managers (i.e., in EN 13230), or (ii) prescribed standardised factors (i.e., AS 1085.14, AREMA Chapter 30, JSA—JIS E 1201). The existing design concepts for prestressed concrete sleepers using either (i) an allowable stress design or (ii) the limit state design method require many iterations for calculations and optimisations. The design process to achieve optimal products suitable for track, operational, and environmental parameters is, thus, very time-consuming. On this ground, this study investigates the potential capability of machine learning (ML) to learn from large amounts of design data sets and then to facilitate the design and capacity prediction of railway prestressed concrete sleepers. Three ML algorithms are developed, namely deep learning, Bayesian Neural Network, and random forest. Through a combination of hand-calculated design data, industry design data, and experimental investigations in compliance with EN 13230, over 3000 sets of design data have been collected. These data sets are used to assimilate a comprehensive database for machine learning. Four indicators, namely mean squared error (MSE), root-mean-square error (RMSE), mean absolute error (MAE), and R2 are used to benchmark the accuracy and precision of machine learning models. Our results reveal that the random forest algorithm offers the best performance. The values of MSE, RMSE, MAE, and R2 are 0.54, 0.74, 0.25, and 0.99, respectively. Note that the Bayesian neural network also performs very well. In contrast, the deep learning algorithm performs worse than the others. The insight demonstrates machine learning’s capability to aid in the design of railway prestressed concrete sleepers, to satisfy both serviceability and ultimate limit states

9 January 2022 in Sustainability
Verified
Journal Article
Open access
PDF file
JHJunhui Huang
Junhui Huang
Sakdirat KaewunruenSakdirat Kaewunruen
Sakdirat Kaewunruen
JNJingzhiyuan Ning
Jingzhiyuan Ning
To encourage more active activities that have the potential to significantly reduce the risk of people’s health, we aim to develop an AI-based mobile app to identify four gym activities accurately: ascending, cycling, elliptical, and running. To save computational cost, the present study deals with the dilemma of the performance provided by only a phone-based accelerometer since a wide range of activity recognition projects used more than one sensor. To attain this goal, we derived 1200 min of on-body data from 10 subjects using their phone-based accelerometers. Subsequently, three subtasks have been performed to optimize the performances of the K-nearest neighbors (KNN), Support Vector Machine (SVM), Shallow Neural Network (SNN), and Deep Neural Network (DNN): (1) During the process of the raw data converted to a 38-handcrafted feature dataset, different window sizes are used, and a comparative analysis is conducted to identify the optimal one; (2) principal component analysis (PCA) is adopted to extract the most dominant information from the 38-feature dataset described to a simpler and smaller size representation providing the benefit of ease of interpreting leading to high accuracy for the models; (3) with the optimal window size and the transformed dataset, the hyper-parameters of each model are tuned to optimal inferring that DNN outperforms the rest three with a testing accuracy of 0.974. This development can be further implemented in Apps Store to enhance public usage so that active physical human activities can be promoted to enhance good health and wellbeing in accordance with United Nation’s sustainable development goals.

19 April 2024 in Discover Applied Sciences
Verified
Journal Article
Open access
PDF file
Sakdirat KaewunruenSakdirat Kaewunruen
Sakdirat Kaewunruen
AAAbdullah Abimbola Adesope
Abdullah Abimbola Adesope
JHJunhui Huang
Junhui Huang
Ruilin YouRuilin You
Ruilin You
DLDan Li
Dan Li
Railway concrete sleepers are key safety-critical components in ballasted railway tracks. Due to frequent high-intensity impact loadings from train-track interaction over irregularities together with hostile environmental conditions, complicated characteristics of various crack patterns can incur on railway concrete sleepers, which will decrease their durability and service life overtime. Early warning of those cracks can help railway engineers to plan and schedule for renewal and maintenance timely and effectively. This study thus explores the artificial intelligence application of YOLOv5OBB (YOLOv5 with Oriented Bounding Box output) in the identification and classification of cracks in railway sleepers into three distinct types: longitudinal, transverse, and inclined, based on their specific crack angles, which have not been investigated in the past. The identification of crack angles is the novelty of this study. Recognising the various types of cracks is critical, given their varying causes and degrees of severity. Current corrective maintenance methods pose considerable safety risks to workers and exhibit low efficiency, underscoring the need for a more autonomous and efficient solution. This study marks a significant stride towards revolutionising railway maintenance, evidenced by an impressive mAP (Mean Average Precision) of 0.72 for crack detection and a 92% accuracy rate for angle detection. These promising results substantiate our study's potential to pioneer advancements in railway infrastructure maintenance.

4 June 2023 in Applied Sciences
Verified
Journal Article
Open access
PDF file
JYJingyuan Yang
Jingyuan Yang
HDHuayu Duan
Huayu Duan
LLLinxiao Li
Linxiao Li
ESEdward Stewart
Edward Stewart
JHJunhui Huang
Junhui Huang
Roger DixonRoger Dixon
Roger Dixon
Defective droppers pose a significant threat to the performance of the contact between the train pantograph and railway catenary. In this paper, the impact of damaged droppers on the performance of pantograph–catenary interaction behaviour is analysed, and the impact of varying degrees of damage to each dropper is labelled. To improve the classification accuracy when both the damage degree and position are considered, a model integrating multiple 1D CNNs is proposed. Approaches including randomly searching the optimal hyper-parameters and K-fold cross-validation are used to prevent overfitting and to ensure model performance regardless of the training data subset selected. Compared with a conventional 1D CNN, the classification performance of the integrated method is demonstrated using the metrics accuracy, F1-score, precision and recall. It is concluded that, through the use of the integrated 1D CNN, damaged droppers can be detected and localised based on the pantograph–catenary contact force. Hence, intelligent catenary inspection can be enhanced.