Search

Deep learning has been gradually used in the field of machining condition monitoring. However, at present only single-task prediction can be performed, which results in increased experimental costs, wasted datasets, and repetitive work. In this article, a new multitask learning method based on a deep belief network (DBN) is proposed, which can be used to predict the tool wear condition and part surface quality. The single-task data transmission of the last few hidden layers of the DBN network is improved to multitask parallel data transmission so that the improved DBN can realize multitask learning. The loss function of the multitask learning model is defined as the weighted sum of all single-task loss functions. According to the loss of different tasks in the iteration process, the weight of corresponding tasks can be adjusted automatically. Furthermore, the multitask deep learning method can realize information sharing, suppress overfitting, improve prediction accuracy, and require less computing time. Combined with the abovementioned improvements, a multitask model for tool wear and part surface quality was developed. Experimental verification was performed on a KVC850M three-axis vertical machining center. The results show that the accuracy of the proposed multitask prediction model is 99% for the tool wear prediction and 92.86% for part surface quality prediction.

Maize seeding is greatly affected by improper seed placement and poor planter performance under a no-tillage mechanisation system. To overcome the issue, we explored the impact of separating board and anti-blocking mechanism (0, 1/4, 1/2, 2/3, 3/4, and 1 type) on maize seeding under different forward speeds (3, 5, 7 km/h) and rotational speeds (260, 400, 530, 740 rpm), where the performance metrics included the mass of straw coiled, seeding height, emergence rate, soil mound depth, straw movement, and straw clearance. The study results show that separating board helps to increase forward and side displacements of the straw, which avoids localized accumulation of straw around the antiblocking mechanism. The straw clearance rate of the anti-blocking mechanism with a separating board is greater than that without the separating board. Therefore, the 2/3 type anti-blocking mechanism with a separating board is recommended for maize seeding at a forward speed of 5 km/h and a rotational speed of 400 rpm.

In this paper we share findings from our effort to build practical machine translation (MT) systems capable of translating across over one thousand languages. We describe results in three research domains: (i) Building clean, web-mined datasets for 1500+ languages by leveraging semi-supervised pre-training for language identification and developing data-driven filtering techniques; (ii) Developing practical MT models for under-served languages by leveraging massively multilingual models trained with supervised parallel data for over 100 high-resource languages and monolingual datasets for an additional 1000+ languages; and (iii) Studying the limitations of evaluation metrics for these languages and conducting qualitative analysis of the outputs from our MT models, highlighting several frequent error modes of these types of models. We hope that our work provides useful insights to practitioners working towards building MT systems for currently understudied languages, and highlights research directions that can complement the weaknesses of massively multilingual models in data-sparse settings., Comment: V2: updated with some details from 24-language Google Translate launch in May 2022 V3: spelling corrections, additional acknowledgements

In the process of parts processing, due to the real working conditions and data acquisition equipment, the collected working data of tools are actually limited. Meanwhile, the tool usually works in the normal state, so it is prone to cause the problem of unbalanced data set, which restricts the accuracy of tool condition monitoring. Aiming at this problem, this paper proposes a tool condition monitoring method based on generative adversarial network (GAN) for data augmentation. Specifically, first collect original samples data during processing in different tool conditions, then the collected sample data is input into GAN, and the generator of GAN can generate new samples which has similar distribution with original samples from tool condition signals data, finally the real samples and generated samples are combined to train deep learning network to predict tool conditions. Experimental results show that the proposed method can significantly improve the accuracy of tool condition monitoring. This paper compares and visualizes the impact of the training data set on the classification ability of the deep learning network model. In addition, some traditional methods are used for comparison, and F1 measure is introduced to evaluate the quality of the results. The results show that this method is better than the Adaptive Synthetic Sampling (Adasyn), add-noise, and resampling.

The paper investigates the feasibility of confidence estimation for neural machine translation models operating at the high end of the performance spectrum. As a side product of the data annotation process necessary for building such models we propose sentence level accuracy $SACC$ as a simple, self-explanatory evaluation metric for quality of translation. Experiments on two different annotator pools, one comprised of non-expert (crowd-sourced) and one of expert (professional) translators show that $SACC$ can vary greatly depending on the translation proficiency of the annotators, despite the fact that both pools are about equally reliable according to Krippendorff's alpha metric; the relatively low values of inter-annotator agreement confirm the expectation that sentence-level binary labeling $good$ / $needs\ work$ for translation out of context is very hard. For an English-Spanish translation model operating at $SACC = 0.89$ according to a non-expert annotator pool we can derive a confidence estimate that labels 0.5-0.6 of the $good$ translations in an "in-domain" test set with 0.95 Precision. Switching to an expert annotator pool decreases $SACC$ dramatically: $0.61$ for English-Spanish, measured on the exact same data as above. This forces us to lower the CE model operating point to 0.9 Precision while labeling correctly about 0.20-0.25 of the $good$ translations in the data. We find surprising the extent to which CE depends on the level of proficiency of the annotator pool used for labeling the data. This leads to an important recommendation we wish to make when tackling CE modeling in practice: it is critical to match the end-user expectation for translation quality in the desired domain with the demands of annotators assigning binary quality labels to CE training data.

In this study, NiO/Co3O4 nanocomposites were synthesized by a facile hydrothermal method and exploited as electrode material for electrochemical supercapacitor. The synergistic interaction between NiO and Co3O4 was explored systematically by varying their mass ratio. Interestingly, the electrochemical performances were dependent on the mass ratio of NiO/Co3O4 nanocomposite. NiO/Co3O4 electrode with mass ratio of 1:1 exhibited the highest specific capacitance of 937 F/g at a current density of 1 A/g, due to their distinct microspherical structure with large surface area, which would lead to the more accessible active sites for Faradaic reactions. This investigation provides a useful strategy to develop a new class of electrode materials with the improved electrochemical performance.