Search results

The conventional design method relies on a priori knowledge, which limits the rapid and efficient development of electronic packaging structures. The purpose of this study is to propose a hybrid method of data-driven inverse design, which couples adaptive surrogate model technology with optimization algorithm to to enable an efficient and accurate inverse design of electronic packaging structures.

The multisurrogate accumulative local error-based ensemble forward prediction model is proposed to predict the performance properties of the packaging structure. As the forward prediction model is adaptive, it can identify respond to sensitive regions of design space and sample more design points in those regions, getting the trade-off between accuracy and computation resources. In addition, the forward prediction model uses the average ensemble method to mitigate the accuracy degradation caused by poor individual surrogate performance. The Particle Swarm Optimization algorithm is then coupled with the forward prediction model for the inverse design of the electronic packaging structure.

Benchmark testing demonstrated the superior approximate performance of the proposed ensemble model. Two engineering cases have shown that using the proposed method for inverse design has significant computational savings while ensuring design accuracy. In addition, the proposed method is capable of outputting multiple structure parameters according to the expected performance and can design the packaging structure based on its extreme performance.

Because of its data-driven nature, the inverse design method proposed also has potential applications in other scientific fields related to optimization and inverse design.

Surface mount technology (SMT) is widely used and plays an important role in electronic equipment. The purpose of this paper is to reveal the effects of interface cracks on the fatigue life of SMT solder joint under service load and to provide some valuable reference information for improving service reliability of SMT packages.

A 3D geometric model of SMT package is established. The mechanical properties of SMT solder joint under thermal cycling load and random vibration load were solved by 3D finite element analysis. The fatigue life of SMT solder joint under different loads can be calculated by using the modified Coffin–Manson model and high-cycle fatigue model.

The results revealed that cracks at different locations and propagation directions have different effect on the fatigue life of the SMT solder joint. From the location of the cracks, Crack 1 has the most significant impact on the thermal fatigue life of the solder joint. Under the same thermal cycling conditions, its life has decreased by 46.98%, followed by Crack 2, Crack 4 and Crack 3. On the other hand, under the same random vibration load, Crack 4 has the most significant impact on the solder joint fatigue life, reducing its life by 81.39%, followed by Crack 1, Crack 3 and Crack 2. From the crack propagation direction, with the increase of crack depth, the thermal fatigue life of the SMT solder joint decreases sharply at first and then continues to decline almost linearly. The random vibration fatigue life of the solder joint decreases continuously with the increase of crack depth. From the crack depth of 0.01 mm to 0.05 mm, the random vibration fatigue life decreases by 86.75%. When the crack width increases, the thermal and random vibration fatigue life of the solder joint decreases almost linearly.

This paper investigates the effects of interface cracks on the fatigue life and provides useful information on the reliability of SMT packages.

With the rapid development of short videos in China, the public has become accustomed to using short videos to express their opinions. This paper aims to solve problems such as how to represent the features of different modalities and achieve effective cross-modal feature fusion when analyzing the multi-modal sentiment of Chinese short videos (CSVs).

This paper aims to propose a sentiment analysis model MSCNN-CPL-CAFF using multi-scale convolutional neural network and cross attention fusion mechanism to analyze the CSVs. The audio-visual and textual data of CSVs themed on “COVID-19, catering industry” are collected from CSV platform Douyin first, and then a comparative analysis is conducted with advanced baseline models.

The sample number of the weak negative and neutral sentiment is the largest, and the sample number of the positive and weak positive sentiment is relatively small, accounting for only about 11% of the total samples. The MSCNN-CPL-CAFF model has achieved the Acc-2, Acc-3 and F1 score of 85.01%, 74.16 and 84.84%, respectively, which outperforms the highest value of baseline methods in accuracy and achieves competitive computation speed.

This research offers some implications regarding the impact of COVID-19 on catering industry in China by focusing on multi-modal sentiment of CSVs. The methodology can be utilized to analyze the opinions of the general public on social media platform and to categorize them accordingly.

This paper presents a novel deep-learning multimodal sentiment analysis model, which provides a new perspective for public opinion research on the short video platform.

At present, one of the key equipment in pillar industries is a large rotating machinery. Conducting regular health monitoring is important for ensuring safe operation of the large rotating machinery. Because vibrations sensors play an important role in the workings of the rotating machinery, measuring its vibration signal is an important task in health monitoring. This paper aims to present these.

In this work, the contact vibration sensor and the non-contact vibration sensor have been discussed. These sensors consist of two types: the electric vibration sensor and the optical fiber vibration sensor. Their applications in the large rotating machinery for the purpose of health monitoring are summarized, and their advantages and disadvantages are also presented.

Compared with the electric vibration sensor, the optical fiber vibration sensor of large rotating machinery has unique advantages in health monitoring, such as provision of immunity against electromagnetic interference, requirement of less insulation and provision of long-distance signal transmission.

Both contact vibration sensor and non-contact vibration sensor have been discussed. Among them, the electric vibration sensor and the optical fiber vibration sensor are compared. Future research direction of the vibration sensors is presented.

The purpose of this paper is to measure the systemic importance of industry in the world economic system under the system-wide event – the crisis of 2008-2009, by viewing this system as a weighted directed network of interconnected industries.

First, the authors investigate this crisis at three different levels based on network-related indicators: the “macro” global level, the “meso” country level, and the “micro” industry level. This investigation not only provides evidence for the systemic influence, that is, systemic risk, of the crisis, but also reveals the contagion mechanism of the crisis, which supports the stress testing. Second, the authors use a network-related business intelligence algorithm, the combined hyperlink-induced topic search (HITS) algorithm, to measure the contribution of a given individual industry to the overall risk of the economic system or, in other words, the systemic importance of the individual industry.

The HITS algorithm considers both the market information and the interconnectedness of the industries. Based on the stress testing, the performance of the combined HITS is compared with the purely market-based systemic risk measurement. The results show that the combined HITS outperforms the baseline in finding the top N systemically important industries.

The combined HITS algorithm provides a novel network-based perspective of systemic risk measurement.

Measuring the systemic importance based on the combined HITS algorithm can help managers and regulators design effective risk management policies. In this respect, the work initiates a research direction of studying the systemic risk in a business system based on a network-related business intelligence algorithm because the business system can be viewed as an interconnected network.

The purpose of this paper is to predict quantitative level of stray current leaking to the buried metallic structure by establishing convolution neural network (CNN) model.

First, corrosion experimental system of buried metallic structure is established. The research object of this paper is the polarization potential within 110 min, CNN model is used to predict the quantitative level of stray current leakage using the data from corrosion experimental system further. Finally, results are compared with the method using BP neural network.

Results show that the CNN model has better predictive effect and shorter prediction time than the BP model, the accuracy of which is 82.5507 per cent, and the prediction time is shortened by more than 10 times.

The established model can be used to forecast the level of stray current leakage in the subway system effectively, which provides a new theoretical basis for evaluating the stray current corrosion hazard of buried metallic structure.

How to extract useful information from a very large volume of literature is a great challenge for librarians. Topic modeling technique, which is a machine learning algorithm to uncover latent thematic structures from large collections of documents, is a widespread approach in literature analysis, especially with the rapid growth of academic literature. In this paper, a comparison of topic modeling based literature analysis has been done using full texts and abstracts of articles.

The authors conduct a comparison study of topic modeling on full-text paper and corresponding abstract to assess the influence of the different types of documents been used as input for topic modeling. In particular, the authors use the large volumes of COVID-19 research literature as a case study for topic modeling based literature analysis. The authors illustrate the research topics, research trends and topic similarity of COVID-19 research by using Latent Dirichlet allocation (LDA) and topic visualization method.

The authors found 14 research topics for COVID-19 research. The authors also found that the topic similarity between using full-text paper and corresponding abstract is higher when more documents are analyzed.

First, this study contributes to the literature analysis approach. The comparison study can help us understand the influence of the different types of documents on the results of topic modeling analysis. Second, the authors present an overview of COVID-19 research by summarizing 14 research topics for it. This automated literature analysis can help specialists in the health and medical domain or other people to quickly grasp the structured morphology of the current studies for COVID-19.

Between 1860 and 1945, the Chinese port city of Tianjin was the site of up to nine foreign-controlled concessions, functioning side by side. Rogaski defined it as a ‘hyper-colony’, a term which reflects Tianjin's socio-political intricacies and the multiple colonial discourses of power and space. This essay focuses on the transformation of the Tianjin cityscape during the last 150 years, and aims at connecting the hyper-colonial socio-spatial forms with the processes of post-colonial identity construction. Tianjin is currently undergoing a massive renovation program: its transmogrifying cityscape unveils multiple layers of ‘globalizing’ spatialities and temporalities, throwing into relief processes of power and capital accumulation, which operate via the urban regeneration's experiment. This study uses an ‘interconnected history’ approach and traces the interweaving ‘worlding’ nodes of today's Tianjin back to the global connections established in the city during the hyper-colonial period. What emerges is Tianjin's simultaneous tendency towards ‘world-class-ness’ and ‘China-class-ness’.

Digital resources such as smart applications reviews and online feedback information are important sources to seek customers’ feedback and input. This paper aims to help government entities gain insights on the needs and expectations of their customers. Towards this end, we propose an aspect-based sentiment analysis hybrid approach that integrates domain lexicons and rules to analyse the entities smart apps reviews. The proposed model aims to extract the important aspects from the reviews and classify the corresponding sentiments. This approach adopts language processing techniques, rules, and lexicons to address several sentiment analysis challenges, and produce summarized results. According to the reported results, the aspect extraction accuracy improves significantly when the implicit aspects are considered. Also, the integrated classification model outperforms the lexicon-based baseline and the other rules combinations by 5% in terms of Accuracy on average. Also, when using the same dataset, the proposed approach outperforms machine learning approaches that uses support vector machine (SVM). However, using these lexicons and rules as input features to the SVM model has achieved higher accuracy than other SVM models.