Machine learning-based wind pressure prediction of low-rise non-isolated buildings

Weng, Yanmo; Paal, Stephanie German

doi:10.1016/j.engstruct.2022.114148

Cited by 22 publications

(5 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Li and Hao, (2018) adopted a computational fluid dynamics model prediction method combining a flame acceleration simulator simulation and ANSYS Fluent simulation, demonstrating high accuracy in far-field overpressure predictions. Weng and German Paal, (2022) proposed a new machine learning-based wind pressure prediction model for low-rise non-isolated buildings. As the pressure in the transition ladle during the manufacturing of amorphous alloys could be influenced by numerous factors, (Liu et al, 2022) employed a backpropagation (BP) neural network to ensure the prediction of the transition ladle pressure during the production of amorphous alloys.…”

Section: Open Accessmentioning

confidence: 99%

“…The prediction technique employed in this study is based on machine learning. The existing popular research methods in machine learning include the GA-BP neural network model, Grey neural network model, BP neural network, extreme learning machine, support vector machine, artificial neural network, grid search algorithm, a gradient boosting decision tree, and generative time intervals imputation network (Huang et al, 2019;Tan et al, 2019;Arshad et al, 2021;Kim et al, 2021;Wang et al, 2022a;Liu et al, 2022;Weng and German Paal, 2022). However, the existing prediction methods are only applicable to data features with a single data quantity or evident data characteristic.…”

Section: Open Accessmentioning

confidence: 99%

See 1 more Smart Citation

Multiple regression method for working face mining pressure prediction based on hydraulic support monitoring dataset

Dong

Xie²,

Jiang³

et al. 2023

Front. Earth Sci.

View full text Add to dashboard Cite

Introduction: In the coal mining process, the intense mining pressure is an important factor hindering the safe and efficient production of the working face. In severe cases, it causes deformations in roadways such as roof breakages and rockbursts, and leads to instability. This can result in the roof falling over a large area and the coal wall, thereby inducing dynamic disasters. These aspects have restricted the economic benefits of coal.Methods: In this study, we set four model limitations based on the limited scope of action of the mining pressure itself and the quantitative relationships between mining pressures in different regions. A multiple linear regression model with these limitations is proposed for predicting the mining pressure for preventing roof breakages and rockbursts. Based on a hydraulic support monitoring dataset from a fully mechanized caving face of coal mining, the mining pressure prediction model is trained by using the first 70% of the dataset. And the linear regression coefficient of the model and the predicted value of the mining pressure are obtained. Then, the last 30% of the dataset was used for the validation of the model.Results: The research results show that the constrained multiple linear regression model can achieve remarkable prediction results. According to predictions of tens of thousands of on-site mining pressure datasets, the predicted data and actual pressure data have the same change trend and maintain a low relative error.Discussion: Therefore, after real-time mining pressure monitoring, the system obtains the roof pressure of the fully mechanized mining face. According to the dataset, the proposed prediction model algorithm quickly predicts the roof pressure value of the next mining section and effectively forewarns roof breakages and other accidents.

show abstract

Section: Open Accessmentioning

confidence: 99%

Section: Open Accessmentioning

confidence: 99%

Multiple regression method for working face mining pressure prediction based on hydraulic support monitoring dataset

Dong

Xie²,

Jiang³

et al. 2023

Front. Earth Sci.

View full text Add to dashboard Cite

show abstract

“…The complex convolution mapping relationship between impact load and response, which is hard to solve, can be replaced with well-trained machine learning model. Traditional machine learning models such as support vector machine [24,25] and decision tree [26] have clear principle and fast training speed which need appropriate feature extraction yet. Deep learning [27][28][29] has powerful learning ability and has played an important role in the fields of pattern recognition [30,31] and intelligent driving [32] in recent years.…”

Section: Introductionmentioning

confidence: 99%

Impact load identification and localization method on thin-walled cylinders using machine learning

Guo¹,

Jiang²,

Yang³

et al. 2023

Smart Mater. Struct.

View full text Add to dashboard Cite

In this paper, a novel impact load identification and localization method on actual engineering structures using machine learning is proposed. Three machine learning models, including a Gradient Boosting Decision Tree (GBDT) model based on ensemble learning, a Convolutional Neural Network (CNN) model and a Bidirectional Long Short-Term Memory (BLSTM) model based on deep learning, are trained to directly identify and locate impact loads according to dynamic response. The GBDT model and the CNN model can reversely identify force peak and location of impact loads. The BLSTM model can reconstruct the time history of impact loads. The method is verified on a thin-walled cylinder with obvious nonlinearity. The result shows that the method can accurately identify impact loads and its location. The characteristics of the three models are compared and the influence of structural boundary conditions on the accuracy of identification is discussed. The method proposed in the paper has the potential to be applied to various engineering structures and multiple load types.

show abstract

“…Zheng and Yuan (2021) employ an Artificial Neural Network (ANN) to generate and analyse different building forms. Other methods focus on rapid urban microclimate prediction (He et al, 2021;Weng and German Paal, 2022) and building environmental performance prediction (Mokhtar et al, 2021). These ML methods can learn the underlying distribution and features behind the input data, and subsequently generate appropriate solutions (Sato et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

A “Designer-centric” Framework for Building Massing optimization

Tan,

Ying

2023

eCAADe Proceedings

View full text Add to dashboard Cite

Performance-based architectural design often faces challenges due to the complex interactions between wind, solar data, and building massing. Although massing optimization frameworks help address these challenges, limitations still exist, such as restricting designers' authorship and being time-consuming. To tackle these issues, a new workflow is proposed that combines wind and sun analysis, Evolutionary Algorithm (EA), and machine learning (ML) methods. This framework consists of two main stages: (1) combining designer intention with wind and solar simulations in an EA to generate optimized massing based on real-world buildings, and ( 2) training a machine learning model that remaps the distribution of the designer's intentions and building layouts. This study demonstrates the effectiveness of the proposed workflow in generating efficient and detailed building massing layouts that meet specific requirements. This "designer-centric" approach enables machines to optimize architect's intentions, rather than forcing them to adapt to machine calculations. It will become an efficient, user-friendly human-machine collaborative tool for assisting in future building layout designs.

show abstract

Machine learning-based wind pressure prediction of low-rise non-isolated buildings

Cited by 22 publications

References 36 publications

Multiple regression method for working face mining pressure prediction based on hydraulic support monitoring dataset

Multiple regression method for working face mining pressure prediction based on hydraulic support monitoring dataset

Impact load identification and localization method on thin-walled cylinders using machine learning

A “Designer-centric” Framework for Building Massing optimization

Contact Info

Product

Resources

About