Multimodal process monitoring based on transition-constrained Gaussian mixture model

Reliable process monitoring is important for ensuring process safety and product quality. A production process is generally characterized by multiple operation modes, and monitoring these multimodal processes is challenging. Most multimodal monitoring methods rely on the assumption that the modes are independent of each other, which may not be appropriate for practical application. This study proposes a transition-constrained Gaussian mixture model method for efficient multimodal process monitoring. This technique can reduce falsely and frequently occurring mode transitions by considering the time series information in the mode identification of historical and online data. This process enables the identified modes to reflect the stability of actual working conditions, improve mode identification accuracy, and enhance monitoring reliability in cases of mode overlap. Case studies on a numerical simulation example and simulation of the penicillin fermentation process are provided to verify the effectiveness of the proposed approach in multimodal process monitoring with mode overlap.     

Exploring the methods on improving CH4 delivery performance to surpass the Advanced Research Project Ageney-Energy target

CH_4 storage associated with adsorbed natural gas(ANG) technology is an issue attracting great concern.Following the Advanced Research Project Agency-Energy(ARPA-E) targeted deliverable capacity of 315 cm~3·cm~(-3)(STP), hundreds of thousands of materials have been experimentally or theoretically evaluated,while the best results still show a 35% gap from the target. Moreover, recent theoretical research reveals that the target is beyond the possibility that real materials can be designed. To get rid of the awkward situation, we make attempts on investigating the CH_4 delivery performance under other operation conditions. Methods of raising the discharge temperature(to infinite high) or elevating the storage pressure(to 25 MPa) have been proved to show limited effectiveness. In this work, it is found that the ARPA-E target can be achieved by using a decreasing storage temperature strategy. By taking 280 Co RE(computation-ready, experimental) COFs(covalent organic frameworks) as ANG materials, when reduce the storage temperature to 190.6 K, the highest deliverable capacity can reach 392 cm~3·cm~(-3)(STP), and16.1% Co RE COFs can surpass the target. The target is also achievable when storage at 220 K. Structure performance relationships study shows strong correlation between deliverable capacity and void fraction. Hence, 120 hypothetical COFs are generated to ascertain the optimum void fraction. In addition,the performance of 2 D-COFs can be greatly enhanced by increasing the interlayer spacings, e.g. CH_4 deliverable capacity(storage at 190.6 K) of ATFG-COF can be improved from 239 to 411 cm~3·cm~(-3)(STP) when interlayer spacing is enlarged to 1.65 nm.     

Electron beam welding of Re and BT5-1 titanium alloy

Based on the binary alloy phase diagram of Re-Ti, the weldability of Re and BT5-1 titanium alloy was analyzed. Using two methods of direct electron beam welding （EBW-D） and intergradafion electron beam welding （EBW-I）, Re and BT5-1 was welded. Experimental results show that the joint figuration of EBW-D between Re and BT5-1 is not fine, and the joint is inclined to brittleness rupture. The joint figuration of EBW-I between Re and BT5-1 is fine. No cracking and other disfigures occur in the intergradation joint. The element distribution of Re, Me, and Ti in the weld metal is progressional diversification.     

氨分解制氢装置工艺设备的技术改造

介绍了氨分解制氢工艺设备的改进.改造氨汽化器,用蒸汽加热液氨替换电加热液氨;用电加热器直接加热氮气,使分子筛加热再生,改造净化系统.     

一个面向微博的情感倾向性分析模型

在微博情感倾向性分析中,一种典型分析方法是先对微博进行主客观分类,再对判定为主观的微博进行褒贬分类,但其问题在于主客观分类错误将直接传导到褒贬分类。针对这一问题,本文提出了一个主客观分类和褒贬分类融合的评估情感倾向性强度的模型。首先使用改进的逻辑回归模型构建主客观分类模型,并结合情感词典构建褒贬分类模型;然后,将二者融合,构建情感倾向性强度模型来选出具有较强情感的微博;最后应用褒贬分类模型判定情感倾向性。该方法在第六届中文倾向性分析评测(COAE2014)的微博观点句识别任务中获得了主要指标Micro_F1值和Macro_F1值的第二名。     

A high propylene productivity over B2O3/SiO2@honeycomb cordierite catalyst for oxidative dehydrogenation of propane

Boron-based metal-free catalysts for oxidative dehydrogenation of propane (ODHP) have drawn great attention in both academia and industry due to their impressive activity and olefin selectivity. Herein, the SiO₂ and B₂O₃ sequentially coated honeycomb cordierite catalyst is designed by a two-step wash-coat method with different B₂O₃ loadings (0.1%-10%) and calcination temperatures (600, 700, 800 ℃). SiO₂ obtained by TEOS hydrolysis acts as a media layer to bridge the cordierite substrate and boron oxide via abundant Si-OH groups. The welldeveloped straight channels of honeycomb cordierite make it possible to carry out the reactor under high gas hourly space velocity (GHSV) and the thin wash-coated B₂O₃ layer can effectively facilitate the pore diffusion on the catalyst. The prepared B₂O₃/SiO₂@HC monolithic catalyst exhibits good catalytic performance at low boron oxide loading and achieves excellent propylene selectivity (86.0%), olefin selectivity (97.6%, propylene and ethylene) and negligible CO₂ (0.1%) at 16.9% propane conversion under high GHSV of 345,600 ml·(g B₂O₃)^-1·h^-1, leading to a high propylene space time yield of 15.7 g C₃H₆·(g B₂O₃)^-1·h^-1 by suppressing the overoxidation. The obtained results strongly indicate that the boron-based monolithic catalyst can be properly fabricated to warrant the high activity and high throughput with its high gas/surface ratio and straight channels.     

Integrated UV-based photo microreactor-distillation technology toward process intensification of continuous ultra-high-purity electronic-grade silicon tetrachloride manufacture

Ultra-high-purity silicon tetrachloride (SiCl₄) is demanded as an electronic-grade chemical to meet the stringent requirements of the rapidly developing semiconductor industry. The high requirement for ultra-high-purity SiCl₄ has created the need for a high-efficient process for reducing energy consumption as well as satisfying product quality. In this paper, a mass of production technology of ultra-high-purity SiCl₄ was successfully developed through chlorination reaction in the ultraviolet (UV)-based photo microreactor coupled with the distillation process. The influences of key operational parameters, including temperature, pressure, UV wavelength and light intensity on the product quality, especially for hydrogen-containing impurities, were quantified by the infrared transmittance of Fourier transform infrared spectroscopy (FT-IR) at 2185 cm^-1 and 2160 cm^-1 indicating that characteristic vibrational modes of Si—H bonds, as well as the operating conditions of distillation were also investigated as key factors for metal impurities removing. The advanced intensification of SiCl₄ manufactured by the integration of photo microreactor and distillation achieves the products with superior specifications higher than the standard commercial products.     

Integrated UV-based photo microreactor-distillation technology toward process intensification of continuous ultra-high-purity electronic-grade silicon tetrachloride manufacture

Ultra-high-purity silicon tetrachloride (SiCl₄) is demanded as an electronic-grade chemical to meet the stringent requirements of the rapidly developing semiconductor industry. The high requirement for ultra-high-purity SiCl₄ has created the need for a high-efficient process for reducing energy consumption as well as satisfying product quality. In this paper, a mass of production technology of ultra-high-purity SiCl₄ was successfully developed through chlorination reaction in the ultraviolet (UV)-based photo microreactor coupled with the distillation process. The influences of key operational parameters, including temperature, pressure, UV wavelength and light intensity on the product quality, especially for hydrogen-containing impurities, were quantified by the infrared transmittance of Fourier transform infrared spectroscopy (FT-IR) at 2185 cm⁻¹ and 2160 cm⁻¹ indicating that characteristic vibrational modes of SiH bonds, as well as the operating conditions of distillation were also investigated as key factors for metal impurities removing. The advanced intensification of SiCl₄ manufactured by the integration of photo microreactor and distillation achieves the products with superior specifications higher than the standard commercial products.     

A high propylene productivity over B2O3/SiO2@honeycomb cordierite catalyst for oxidative dehydrogenation of propane

Boron-based metal-free catalysts for oxidative dehydrogenation of propane (ODHP) have drawn great attention in both academia and industry due to their impressive activity and olefin selectivity. Herein, the SiO₂ and B₂O₃ sequentially coated honeycomb cordierite catalyst is designed by a two-step wash-coat method with different B₂O₃ loadings (0.1%–10%) and calcination temperatures (600, 700, 800 °C). SiO₂ obtained by TEOS hydrolysis acts as a media layer to bridge the cordierite substrate and boron oxide via abundant SiOH groups. The well-developed straight channels of honeycomb cordierite make it possible to carry out the reactor under high gas hourly space velocity (GHSV) and the thin wash-coated B₂O₃ layer can effectively facilitate the pore diffusion on the catalyst. The prepared B₂O₃/SiO₂@HC monolithic catalyst exhibits good catalytic performance at low boron oxide loading and achieves excellent propylene selectivity (86.0%), olefin selectivity (97.6%, propylene and ethylene) and negligible CO₂ (0.1%) at 16.9% propane conversion under high GHSV of 345,600 ml·(g B₂O₃)⁻¹·h⁻¹, leading to a high propylene space time yield of 15.7 g C₃H₆·(g B₂O₃)⁻¹·h⁻¹ by suppressing the overoxidation. The obtained results strongly indicate that the boron-based monolithic catalyst can be properly fabricated to warrant the high activity and high throughput with its high gas/surface ratio and straight channels.