ABSTRACT
Integrated circuit (IC) X-ray wire bonding image inspections are crucial for ensuring the quality of packaged products. However, detecting defects in IC chips can be challenging due to the slow defect detection speed and the high energy consumption of the available models. In this paper, we propose a new convolutional neural network (CNN)-based framework for detecting wire bonding defects in IC chip images. This framework incorporates a Spatial Convolution Attention (SCA) module to integrate multi-scale features and assign adaptive weights to each feature source. We also designed a lightweight network, called the Light and Mobile Network (LMNet), using the SCA module to enhance the framework's practicality in the industry. The experimental results demonstrate that the LMNet achieves a satisfactory balance between performance and consumption. Specifically, the network achieved a mean average precision (mAP50) of 99.2, with 1.5 giga floating-point operations (GFLOPs) and 108.7 frames per second (FPS), in wire bonding defect detection.
ABSTRACT
In this work, the collapse ultimate load of a prefabricated reinforced concrete column-steel beam composite frame structure was studied. During the study, the "new RCS beam-to-column joint" was used as the beam-to-column connection in the experimental model. Further, the half-scale fabricated RCS space frame structure (2-story, 1 × 2 bay) was subjected to instantaneous failure experiments twice at the bottom of the side column under various load levels and the 2A column was quickly pulled out by the traction force of the vehicle. The experimental results demonstrated that the method of dismantling the failure column provided a relatively true response to the condition of progressive collapse. The remaining RCS structure was found to be in the elastic stage during various load level tests. Moreover, the displacement time history curve did not have a vibration phenomenon during the first experimentation. The SAP2000 finite element program was used to verify that the test results were similar to that of the numerical simulation results, and it was further explored and found that the collapse ultimate load value was 10.25 times the structure design load value.
ABSTRACT
The precast shear wall behavior in the serviceability and ultimate limit states depends on the shear and shear-flexural behavior of the joints between the precast components or between the precast component and footing. This study presents a series of tests on the shear strength of joints, which were applied to the interface of precast shear walls. The tested parameters included the joint types, the numbers of shear keys, the existence of high strength steel bars inserted at the joints, and the levels of confining stress. The shear capacity, stiffness, and shear transfer mechanisms of these joints were investigated. It could be concluded that the epoxied and high strength reinforcing joints had consistently higher shear strength than that of dry and plain joints. For the specimens with an inclined angle at the end of the keys of less than 60 degrees, the width of the dry joint opening may be excessively large, resulting in large shear slip and the key not shearing-off under confining stress of less than 1.0 MPa. The tested results were compared with AASHTO and other design criteria. Several formulas regarding the joint shear capacities were also proposed according to the specifications and the tested results.
