Convolutional Neural Network-Based Machine Vision for Non-Destructive Detection of Flooding in Packed Columns.

In chemical processes, packed columns are frequently employed in various unit operations. However, the flow rates of gas and liquid in these columns are often constrained by the risk of flooding. To ensure the safe and efficient operation of packed columns, it is crucial to detect flooding in real time. Conventional flooding monitoring methods rely heavily on manual visual inspections or indirect information from process variables, which limit the real-time accuracy of results. To address this challenge, we proposed a convolutional neural network (CNN)-based machine vision approach for non-destructive detection of flooding in packed columns. Real-time images of the packed column were captured using a digital camera and analyzed with a CNN model, which was been trained on a dataset of recorded images to identify flooding. The proposed approach was compared with deep belief networks and an integrated approach of principal component analysis and support vector machines. The feasibility and advantages of the proposed method were demonstrated through experiments on a real packed column. The results showed that the proposed method provides a real-time pre-alarm approach for detecting flooding, enabling process engineers to quickly respond to potential flooding events.

Peripheral groups of polyhedral oligomeric silsesquioxane (POSS) core-based dendrimers: a crucial factor for higher-level supra-architecture building.

The role of peripheral groups (PGs) on dendrimers in the spontaneous higher-level organization of hierarchically assembled nanofibers was investigated in a series of POSS-based dendritic gelators (POSS-Lys-X, X: -Boc, -Cbz, -Fmoc, etc.). We demonstrate that the PGs not only affect the gelation ability in solutions, but also the construction of orderly entangled fibrous supramolecular networks, e.g., "loofah-like" networks. Attributed to the PGs (especially the -Boc group) causing a lower cooperative assembly, the steady state with the lowest potential energy of gelators can be easily achieved by the higher ordering of nanofiber entanglement into superstructures. The -Boc group-containing dendrimers show low molar enthalpy and molar entropy of gelation, which help the construction of unique three-dimensional (3D) "loofah-like" superstructures. In contrast, the high cooperative assembly of the dendrimer (-Cbz as the PG) promotes the gelator into a higher enthalpy gelation process, with a constructed normal fibrous network. Hence, the PGs of POSS-based dendrimers act as the crucial factor in controlling the hierarchical self-assembly via a thermodynamics approach. This research presents new perspectives to explicate the relationships between PGs of dendrimers, supra-architectures and gel performances, which further guide the design of functional supramolecular materials via controllable self-assembly.

Optimizing the Calibration Error of Refraction Angles in Ultrasonic Angle Beam Testing.

Ultrasonic testing is a useful approach for quantifying the flaws in mechanical components. The height of the flaws in ultrasonic angle beam testing is closely related to the calibration value of the probe refraction angle. In order to reduce the calibration error, some ignored data during the traditional calibration process are reanalyzed and fused to determine the refraction angle. Both arithmetical measurement fusion method and weighted measurement fusion method are applied and compared. Monte Carlo simulation is used to estimate the probability distribution of the refraction angle and obtain the optimal refraction angle weights. Experiments were carried out to verify the results of Monte Carlo simulation. The applicability of data fusion on refraction angles is investigated. It was found in the study that the data fused with the refraction angle is helpful for measuring the height of flaws.

Chirality on dendrimers: "roll booster" of the molecule-level self-sorting assembly in two-component supramolecular gel system.

Chirality-induced recognition-promoted molecule-level self-sorting assembly in a two-component supramolecular gel system constructed by two polyhedral oligomeric silsesquioxane (POSS) core-based dendritic gelators with similar geometry and peripheral functionality, which would normally form a homogeneous co-assembled network.

Soft Sensing of Silicon Content via Bagging Local Semi-Supervised Models.

The silicon content in industrial blast furnaces is difficult to measure directly online. Traditional soft sensors do not efficiently utilize useful information hidden in process variables. In this work, bagging local semi-supervised models (BLSM) for online silicon content prediction are proposed. They integrate the bagging strategy, the just-in-time-learning manner, and the semi-supervised extreme learning machine into a unified soft sensing framework. With the online semi-supervised learning method, the valuable information hidden in unlabeled data can be explored and absorbed into the prediction model. The application results to an industrial blast furnace show that BLSM has better prediction performance compared with other supervised soft sensors.

Microstructural Evolution and Mechanical Evaluation of a Laser-Induced Composite Coating on a Ni-Based Superalloy during Thermal Exposure.

A Ni-17Mo-7Cr-based superalloy was laser surface-modified to improve its tribological properties. Si particles were employed as coating materials. Si melted on the surface of the alloy during the process, triggering the formation of Mo6Ni6C carbides and Ni-Si intermetallics. A defect-free coating obtained was mostly made up of primary Mo6Ni6C and γ-Ni31Si12, as well as a eutectic structure of ß1-Ni3Si and α-Ni-based solid solution (α-Ni (s.s)). The volume fraction of hard reinforcements (Mo6Ni6C, γ-Ni31Si12, and ß1-Ni3Si) reached up to 85% in the coating. High-temperature microstructural stability of the coating was investigated by aging the coating at 1073 K for 240-480 h, to reveal its microstructural evolution. In addition, the mechanical performance of the coating was investigated. The nanoscale elastic modulus and hardness of Mo6Ni6C, γ-Ni31Si12, and α-Ni (s.s) were characterized using the nanoindentation tests. The nanoscratch tests were performed to measure the local wear resistance of the coating. Lastly, the Vickers hardness distribution across the cross-section of the coating before and after thermal exposure was compared. The work performed provides basic information understanding the microstructural evolution and mechanical performance of laser-induced coatings on Ni-based superalloys.

Microstructure and Mechanical Properties of 34CrMo4 Steel for Gas Cylinders Formed by Hot Drawing and Flow Forming.

An integral manufacturing process with hot drawing and cold flow forming was proposed for large-diameter seamless steel gas cylinders. The main purpose of this study was to find out the effects of the manufacturing process on the microstructure and mechanical properties of gas cylinders made of 34CrMo4 steel. Two preformed cylinders were produced by hot drawing. One cylinder was then further manufactured by cold flow forming. The experiments were carried out using three types of material sample, namely, base material (BM), hot drawing cylinder (HD), and cold flow-formed cylinder (CF). Tensile and impact tests were performed to examine the mechanical properties of the cylinders in longitudinal and transverse directions. Microstructure evolution was analyzed by scanning electron microscopy (SEM) and electron backscatter diffraction (EBSD) to reveal the relation between the mechanical properties and the microstructure of the material. It is found that the mechanical properties of the 34CrMo4 steel gas cylinders were significantly improved after hot drawing and flow forming plus a designed heat treatment, compared with the base material. The observations of microstructure features such as grain size, subgrain boundaries, and residual strain support the increase in mechanical properties due to the proposed manufacturing process.

Characterization of SiC Ceramic Joints Brazed Using Au⁻Ni⁻Pd⁻Ti High-Temperature Filler Alloy.

In this work, (Au79Ni17Pd4)96Ti4 (wt.%) filler alloy was designed and employed to join SiC ceramics. The effects of brazing temperature and soaking time on the microstructure and fracture morphology of joints were investigated. The results show that the joint obtained can be described as SiC/reaction layer/braze/reaction layer/SiC. The reaction layer was composed of TiC and Au (Si, Ti). The wettability of the filler alloy toward the SiC ceramics was analyzed. The braze zone was mainly constituted by Pd2Si, Ni2Si, and Au (Ni, Si). A large number of nano-sized TiC particles were distributed within the Au (Ni, Si) layer. The formation mechanism of the braze containing different phases was discussed. The brazing temperature and soaking time had a significant effect on the reaction layer at the SiC/braze interface and TiC particles within the Au (Ni, Si) layer, while they showed a negligible effect on the Pd2Si and Ni2Si within the braze. The inherent reason was also clarified in detail. The joint fractography indicated that a good bonding was achieved between the filler alloy and SiC, while joint fracture was primarily induced by the thermal stresses residing after the brazing cycle.

Just-in-Time Correntropy Soft Sensor with Noisy Data for Industrial Silicon Content Prediction.

Development of accurate data-driven quality prediction models for industrial blast furnaces encounters several challenges mainly because the collected data are nonlinear, non-Gaussian, and uneven distributed. A just-in-time correntropy-based local soft sensing approach is presented to predict the silicon content in this work. Without cumbersome efforts for outlier detection, a correntropy support vector regression (CSVR) modeling framework is proposed to deal with the soft sensor development and outlier detection simultaneously. Moreover, with a continuous updating database and a clustering strategy, a just-in-time CSVR (JCSVR) method is developed. Consequently, more accurate prediction and efficient implementations of JCSVR can be achieved. Better prediction performance of JCSVR is validated on the online silicon content prediction, compared with traditional soft sensors.

Treatment of gaseous alpha-pinene by a combined system containing photo oxidation and aerobic biotrickling filtration.

Biofiltration of hydrophobic and/or recalcitrant volatile pollutants is intrinsically limited. In the present study, a combined ultraviolet-biotrickling filter (UV-BTF) was developed to improve the removal of these compounds, and a single BTF as the control was operated under the same conditions. The experimental results showed that the UV-BTF provided higher removal efficiencies than the single BTF at an inlet concentration range of 600-1500 mg m(-3) under shorter residence times. The maximum elimination capacities (ECs) obtained were 94.2 mg m(-3)h(-1) and 44 mg m(-3)h(-1) in the combined UV-BTF and single BTF, respectively. The mass ratio of carbon dioxide produced to α-pinene removed in the UV-BTF was approximately 2.74, which was much higher than that of the single BTF (1.99). Polymerase chain reaction-denaturing gradient gel electrophoresis (PCR-DGGE) analysis indicated that there was more complicated microbial community in the UV-BTF than that in the single BTF. In addition, we investigated the effect of starvation or stagnation on re-acclimation and removal performance from an engineering standpoint. The results showed that the combined UV-BTF could deal with fluctuating conditions or periods without any flow (air or liquid) supply much better than the single BTF.