Search results

This study aims to understand the current production scenario emphasizing the significance of green manufacturing in achieving economic and environmental sustainability goals to fulfil future needs; to determine the viability of particular strategies and actions performed to increase the process efficiency of electrical discharge machining; and to uphold the values of sustainability in the nonconventional manufacturing sector and to identify future works in this regard.

A thorough analysis of numerous experimental studies and findings is conducted. This prominent nontraditional machining process’s potential machinability and sustainability challenges are discussed, along with the current research to alleviate them. The focus is placed on modifications to the dielectric fluid, choosing affordable substitutes and treating consumable tool electrodes.

Trans-esterified vegetable oils, which are biodegradable and can be used as a substitute for conventional dielectric fluids, provide pollution-free machining with enhanced surface finish and material removal rates. Modifying the dielectric fluid with specific nanomaterials could increase the machining rate and demonstrate a decrease in machining flaws such as micropores, globules and microcracks. Tool electrodes subjected to cryogenic treatment have shown reduced tool metal consumption and downtime for the setup.

The findings suggested eco-friendly machining techniques and optimized control settings that reduce energy consumption, lowering operating expenses and carbon footprints. Using eco-friendly dielectrics, including vegetable oils or biodegradable dielectric fluids, might lessen the adverse effects of the electrical discharge machine operations on the environment. Adopting sustainable practices might enhance a business’s reputation with the public, shareholders and clients because sustainability is becoming increasingly significant across various industries.

A detailed general review of green nontraditional electrical discharge machining process is provided, from high-quality indexed journals. The findings and results contemplated in this review paper can lead the research community to collectively apply it in sustainable techniques to enhance machinability and reduce environmental effects.

Vision transformers (ViT) detectors excel in processing natural images. However, when processing remote sensing images (RSIs), ViT methods generally exhibit inferior accuracy compared to approaches based on convolutional neural networks (CNNs). Recently, researchers have proposed various structural optimization strategies to enhance the performance of ViT detectors, but the progress has been insignificant. We contend that the frequent scarcity of RSI samples is the primary cause of this problem, and model modifications alone cannot solve it.

To address this, we introduce a faster RCNN-based approach, termed QAGA-Net, which significantly enhances the performance of ViT detectors in RSI recognition. Initially, we propose a novel quantitative augmentation learning (QAL) strategy to address the sparse data distribution in RSIs. This strategy is integrated as the QAL module, a plug-and-play component active exclusively during the model’s training phase. Subsequently, we enhanced the feature pyramid network (FPN) by introducing two efficient modules: a global attention (GA) module to model long-range feature dependencies and enhance multi-scale information fusion, and an efficient pooling (EP) module to optimize the model’s capability to understand both high and low frequency information. Importantly, QAGA-Net has a compact model size and achieves a balance between computational efficiency and accuracy.

We verified the performance of QAGA-Net by using two different efficient ViT models as the detector’s backbone. Extensive experiments on the NWPU-10 and DIOR20 datasets demonstrate that QAGA-Net achieves superior accuracy compared to 23 other ViT or CNN methods in the literature. Specifically, QAGA-Net shows an increase in mAP by 2.1% or 2.6% on the challenging DIOR20 dataset when compared to the top-ranked CNN or ViT detectors, respectively.

This paper highlights the impact of sparse data distribution on ViT detection performance. To address this, we introduce a fundamentally data-driven approach: the QAL module. Additionally, we introduced two efficient modules to enhance the performance of FPN. More importantly, our strategy has the potential to collaborate with other ViT detectors, as the proposed method does not require any structural modifications to the ViT backbone.