Prediction of rockhead using a hybrid N-XGBoost machine learning framework

The spatial information of rockhead is crucial for the design and construction of tunneling or underground excavation. Although the conventional site investigation methods (i.e. borehole drilling) could provide local engineering geological information, the accurate prediction of the rockhead position with limited borehole data is still challenging due to its spatial variation and great uncertainties involved. With the development of computer science, machine learning (ML) has been proved to be a promising way to avoid subjective judgments by human beings and to establish complex relationships with mega data automatically. However, few studies have been reported on the adoption of ML models for the prediction of the rockhead position. In this paper, we proposed a robust probabilistic ML model for predicting the rockhead distribution using the spatial geographic information. The framework of the natural gradient boosting (NGBoost) algorithm combined with the extreme gradient boosting (XGBoost) is used as the basic learner. The XGBoost model was also compared with some other ML models such as the gradient boosting regression tree (GBRT), the light gradient boosting machine (LightGBM), the multivariate linear regression (MLR), the artificial neural network (ANN), and the support vector machine (SVM). The results demonstrate that the XGBoost algorithm, the core algorithm of the probabilistic N-XGBoost model, outperformed the other conventional ML models with a coefficient of determination (R²) of 0.89 and a root mean squared error (RMSE) of 5.8 m for the prediction of rockhead position based on limited borehole data. The probabilistic N-XGBoost model not only achieved a higher prediction accuracy, but also provided a predictive estimation of the uncertainty. Thus, the proposed N-XGBoost probabilistic model has the potential to be used as a reliable and effective ML algorithm for the prediction of rockhead position in rock and geotechnical engineering.     

Modification of pork-skin jelly by enzymatic cross-linking: melting resistance and quality improvement

Improvements of melting resistance and quality by modification of pork-skin jelly through enzymatic cross-linking were studied, and the mechanism of quality improvement was discussed in this work. Gel strength, springiness and chewiness of modified gel increased significantly (P < 0.05). Transglutaminase (TGase) also improved the viscoelasticity, stability and melting resistance of gel system, as proved by rheological analysis. Sensory evaluation showed that increase in texture need to be moderate and the utilisation of 0.6% TGase was the most appropriate for pork-skin jelly. Significant effects of TGase on inducing protein cross-link and aggregation were confirmed by determining rheology during enzyme treatment and cross-linking extent of pork-skin soup. Correlation analysis showed TGase could improve melting temperature and texture by facilitating cross-linking. Covalent interaction based on ε-(γ-glutamyl)-lysine induced by TGase could play the main role in these improvements. This study suggested that TGase could be applied to design gelatin-based food for tailored quality properties through enzymatic cross-linking.     

香肠整体营养价值评价与包装正面标签应用研究进展

为促进我国消费者对香肠的合理消费与饮食健康，引导企业生产营养健康的香肠，利用《中国食物成分表》（第1册第2版）的16 种香肠及其营养素数据，构建富含营养素食物（nutrient-rich foods，NRF9.3）模型，评价香肠整体营养价值，并借鉴瑞典的锁孔（Keyhole）标识和英国的多交通灯信号标签进行相应的包装正面（front of package，FOP）标签应用探索。结果表明：除了儿童肠、火腿肠外，14 种香肠的NRF9.3＜0，每100 kcal的推荐性营养素含量均低于限制性营养素，对人体的营养价值较低，适合采用总结指示体系和特定营养素体系的FOP标签进行标示；因此，居民日常应限制香肠摄入，建议对脂肪、盐含量超标的香肠采用红色交通灯信号标签提醒消费者，对NRF9.3＞0的香肠实施单一的健康选择标识推荐消费者选购。     

Influence of aliphatic polycarbonate middle block on mechanical and microstructural behaviour of triblock copolymers based on poly(l‐lactide) and polycarbonate

Three different ABA‐type triblock copolymers each containing biodegradable aliphatic polycarbonate as the middle block and poly(l ‐lactide) as the outer blocks were synthesized and the influence of the methylene chain length of the aliphatic polycarbonate middle block on various properties of the triblock copolymers was evaluated. Differential scanning calorimetry and wide‐angle X‐ray diffraction studies revealed that the incorporation of the outer blocks reduced the crystallinity of the middle aliphatic polycarbonate block. Variation of methylene chain length of the middle block led to a change in morphology from spherical to cylindrical as evidenced from atomic force microscopy studies. In addition, the mechanical properties of the block copolymers showed semi‐ductile to quasi‐brittle behaviour depending upon the composition of the middle block which was also confirmed using scanning electron microscopy. Dynamic mechanical analysis of the triblock copolymers indicated that storage modulus increased with a decrease in methylene chain length. © 2018 Society of Chemical Industry     

Self-doped N-propansulfonic acid polyaniline-polyethylene terephthalate film used as active sensor element for humidity or gas detection

In this work, we report the fabrication of sensors’ element for humidity or gases, prepared by in situ polymerization of aniline N-propansulfonic acid using ammonium persulfate in acidic medium. The polymer is being used in the form of powder or deposited in multiple layers onto the PET film. Various techniques including Fourier Transform infrared (FTIR), ultraviolet-–visible spectroscopy (UV–vis), X-ray diffraction (XRD) and scanning electron microscopy (SEM) were employed to characterize the as-prepared sensing materials. The film has been tested for humidity influence, where the significant variations in electrical characteristics were observed, suggesting its usefulness for humidity sensors. Also, for different organic and inorganic gases, a relatively low operating temperature and important sensitivity were observed that indicate its applicability as an active element for general gases sensors. © 2019 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2019 , 136, 47743.     

Customers segmentation in eco-friendly hotels using multi-criteria and machine learning techniques

This study aims to investigate the travellers' choice behaviour towards green hotels through existing online travel reviews on TripAdvisor. Accordingly, a method combining segmentation and the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) techniques was developed to segment travellers based on their provided reviews and to prioritize green hotel attributes based on their level of importance in each segment. The data were taken from travellers' online reviews of Malaysian eco-friendly hotels on TripAdvisor. The results showed that the sleep quality was one of the most imporant factors for eco-hotel selection in the majority of segments. The developed method in this study was able to analyse travellers’ reviews and ratings on eco-friendly hotels to identify the future choice behaviour and aid travellers in their decision-making process. The study provides new insights for hotel managers and green policy makers on developing environmental-friendly practices.     

Methylcyclohexane production under fluctuating hydrogen flow rate conditions

Energy storage using liquid organic hydrogen carrier (LOHC) is a long-term method to store renewable energy with high hydrogen energy density. This study investigated a simple and low-cost system to produce methylcyclohexane (MCH) from toluene and hydrogen using fluctuating electric power, and developed its control method. In the current system, hydrogen generated by an alkaline water electrolyzer was directly supplied to hydrogenation reactors, where hydrogen purification equipment such as PSA and TSA is not installed to decrease costs. Hydrogen buffer tanks and compressors are not equipped. In order to enable MCH production using fluctuating electricity, a feed-forward toluene supply control method was developed and introduced to the system. The electrolyzer was operated under triangular waves and power generation patterns of photovoltaic cells and produced hydrogen with fluctuating flow rates up to 7.5 Nm³/h. Consequently, relatively high purity of MCH (more than 90% of MCH mole fraction) was successfully produced. Therefore, the simplified system has enough potential to produce MCH using fluctuating renewable electricity.