Characteristics of structural loess strength and preliminary framework for joint strength formula

The strength of structural loess consists of the shear strength and tensile strength. In this study, the stress path, the failure envelope of principal stress （ Kf line）, and the strength failure envelope of structurally intact loess and remolded loess were analyzed through three kinds of tests： the tensile strength test, the uniaxial compressive strength test, and the conventional triaxial shear strength test. Then, in order to describe the tensile strength and shear strength of structural loess comprehensively and reasonably, a joint strength formula for structural loess was established. This formula comprehensively considers tensile and shear properties. Studies have shown that the tensile strength exhibits a decreasing trend with increasing water content. When the water content is constant, the tensile strength of the structurally intact soil is greater than that ofremolded soil. In the studies, no loss of the originally cured cohesion in the structurally intact soil samples was observed, given that the soil samples did not experience loading disturbance during the uniaxial compressive strength test, meaning there is a high initial structural strength. The results of the conventional triaxial shear strength test show that the water content is correlated with the strength of the structural loess. When the water content is low, the structural properties are strong, and when the water content is high, the structural properties are weak, which means that the water content and the ambient pressure have significant effects on the stress-strain relationship of structural loess. The established joint strength formula of structural loess effectively avoids overestimating the role of soil tensile strength in the traditional theory of Mohr-Coulomb strength.

Structural form-finding of bending components in buildings by using parametric tools and principal stress lines

This paper aims to provide an efficient and straightforward structural form-finding method for designers to extrapolate component forms during the conceptual stage.The core idea is to optimize the classical method of structural form-finding based on principal stress lines by using parametric tools.The traditional operating process of this method relies excessively on the designer’s engineering experience and lacks precision.Meanwhile,the current optimization work for this method is overly complicated for architects,and limitations in component type and final result exist.Therefore,to facilitate an architect’s conceptual work,the optimization metrics of the method in this paper are set as simplicity,practicality,freedom,and rapid feedback.For that reason,this paper optimizes the method from three aspects:modeling strategy for continuum structures,classification processing of data by using the k-nearest neighbor algorithm,and topological form-finding process based on stress lines.Eventually,it allows architects to create structural texture with formal aesthetics and modify it in real time on the basis of structural analysis results.This paper also explores a comprehensive application strategy with internal force analysis diagramming to form-finding.The finite element analysis tool Karamba3D verifies the structural performance of the form-finding method.The performance is compared with that of the conventional form,and the comparison results show the practicality and potential of the strategy in this paper.

Palmprint Phenotype Feature Extraction and Classification Based on Deep Learning

Palmprints are of long practical and cultural interest.Palmprint principal lines,also called primary palmar lines,are one of the most dominant palmprint features and do not change over the lifespan.The existing methods utilize filters and edge detection operators to get the principal lines from the palm region of interest(ROI),but can not distinguish the principal lines from fine wrinkles.This paper proposes a novel deep-learning architecture to extract palmprint principal lines,which could greatly reduce the influence of fine wrinkles,and classify palmprint phenotypes further from 2D palmprint images.This architecture includes three modules,ROI extraction module(REM)using pre-trained hand key point location model,principal line extraction module(PLEM)using deep edge detection model,and phenotype classifier(PC)based on ResNet34 network.Compared with the current ROI extraction method,our extraction is competitive with a success rate of 95.2%.For principal line extraction,the similarity score between our extracted lines and ground truth palmprint lines achieves 0.813.And the proposed architecture achieves a phenotype classification accuracy of 95.7%based on our self-built palmprint dataset CAS_Palm.