Distillation Method Research Articles

CHEMICAL COMPOSITION OF THE ESSENTIAL OIL OF WORMWOOD (ARTEMISIA ABSINTHIUM L.), GROWING IN THE WESTERN REGION OF AZERBAIJAN

Wormwood (Artemisia absinthium L.) is a widespread plant species in the Western region of Azerbaijan. The dependence of the composition, quantity and growth of various biologically active substances contained in the above-ground parts of wormwood on the vegetation period and climatic conditions was studied. The results of the study of the essential oil obtained by the steam distillation method from wet and dry samples of the aerial parts of bitter wormwood are presented. The yield of essential oil is determined in % in terms of the weight of air-dried raw materials. The study of the yield of essential oil was different months of the year showed that its amount varies depending on the vegetation period (yield of 0.56-0.66% by mass of fresh raw materials; yield of 0.58-0.77% by mass of dry raw materials). The composition of volatile distillates was studied by the chromatography method. In all samples, the amount of camphor component, which belongs to oxygenated terpenoid hydrocarbons, prevailed with a percentage above 35%. 25.44-34.24% phase budding in June for fresh and dry plant material; 25.14-35.21% phase mass flowering in August; it was determined between 25.67-35.20% phase the seed ripening in September.

Generous teacher: Good at distilling knowledge for student learning

Knowledge distillation is a technique that aims to transfer valuable knowledge from a large, well-trained model (the teacher) to a lightweight model (the student), with the primary goal of improving the student's performance on a given task. In recent years, mainstream distillation methods have focused on modifying student learning styles, resulting in less attention being paid to the knowledge provided by the teacher. However, upon re-examining the knowledge transferred by the teacher, we find that it still has untapped potential, which is crucial to bridging the performance gap between teachers and students. Therefore, we study knowledge distillation from the teacher's perspective and introduce a novel teacher knowledge enhancement method termed “Generous Teacher.” The Generous Teacher is a specially trained teacher model that can provide more valuable knowledge for the student model. This is achieved by integrating a standardly trained teacher (Standard Teacher) to assist in the training process of the Generous Teacher. As a result, the Generous Teacher accomplishes the task at hand and assimilates distilled knowledge from the Standard Teacher, effectively adapting to distillation teaching in advance. Specifically, we recognize that non-target class knowledge plays a crucial role in improving the distillation effect for students. To leverage this, we decouple logit outputs and selectively use the Standard Teacher's non-target class knowledge to enhance the Generous Teacher. By setting the temperature as a multiple of the logit standard deviation, we ensure that the additional knowledge absorbed by the Generous Teacher is more suitable for student distillation. Experimental results on standard benchmarks demonstrate that the Generous Teacher surpasses the Standard Teacher in terms of accuracy when applied to standard knowledge distillation. Furthermore, the Generous Teacher can be seamlessly integrated into existing distillation methods, bringing general improvements at a low additional computational cost. The code will be publicly available at https://github.com/EifelTing/Generous-Teacher.

High-throughput computational screening for optimal solvents in methyl 2-hydroxyisobutyrate-water separation

In the semiconductor industry, the efficacy of methyl 2-hydroxyisobutyrate (HBM) as a photoresist thinner is crucial but often compromised by water contamination. Additionally, the manufacturing process tends to form an azeotrope with water, rendering traditional distillation methods inefficient for purification. This study introduces a novel computational screening approach to identify optimal solvents for the liquid-liquid extraction of HBM, offering a sustainable alternative to conventional distillation. Our three-step sequential screening protocol utilizes molecular dynamics simulations to evaluate intermolecular interaction energy, partition coefficient, and environmental, health, and safety (EHS) parameters. Initially, a comprehensive database of 4919 solvents was screened using molecular dynamics simulations to assess interaction energies and partition coefficients. Solvents that met these energy criteria were further evaluated for their EHS profiles. This process identified seven solvents that optimize extraction efficiency while adhering to stringent environmental and safety standards. Our approach not only enhances the purity and performance of HBM in semiconductor manufacturing but also presents a green, cost-effective solution for other industrial chemical processes requiring refined solvent selection.

The benefits and challenges of treating skin with natural oils.

The term natural oil refers to a fixed (non-volatile) oil of animal or plant origin. These types of oils - in contrast to essential (volatile) oils, which are obtained by steam distillation methods of plant matter - are typically obtained from plant seeds and nuts by a mechanical pressing technique or solvent extraction. The natural movement in cosmetics of the 21st century has led to renewed interest in formulating skin care products with botanical ingredients. In this article, we discuss the benefits and caveats of natural oil treatments as moisturizing agents (as occlusives and emollients) as well as their utility in wound healing and treatment of skin diseases. We also address the paradoxical behaviour of natural oils in relation to barrier function and highlight the current state of our knowledge with respect to the use of natural oils in neonatal skin care. Finally, we provide a comparison of natural oils to conventional petroleum-based oils.

Vision-Language Models for Vision Tasks: A Survey.

Most visual recognition studies rely heavily on crowd-labelled data in deep neural networks (DNNs) training, and they usually train a DNN for each single visual recognition task, leading to a laborious and time-consuming visual recognition paradigm. To address the two challenges, Vision-Language Models (VLMs) have been intensively investigated recently, which learns rich vision-language correlation from web-scale image-text pairs that are almost infinitely available on the Internet and enables zero-shot predictions on various visual recognition tasks with a single VLM. This paper provides a systematic review of visual language models for various visual recognition tasks, including: (1) the background that introduces the development of visual recognition paradigms; (2) the foundations of VLM that summarize the widely-adopted network architectures, pre-training objectives, and downstream tasks; (3) the widely-adopted datasets in VLM pre-training and evaluations; (4) the review and categorization of existing VLM pre-training methods, VLM transfer learning methods, and VLM knowledge distillation methods; (5) the benchmarking, analysis and discussion of the reviewed methods; (6) several research challenges and potential research directions that could be pursued in the future VLM studies for visual recognition.

Dynamic control analysis of energy-efficient side-stream extractive distillation process for separating close-boiling mixture

To separate the close-boiling mixture comprising heptane and toluene, this study introduces an environmentally-friendly method of side-stream extractive distillation by utilizing aniline as the separating agent. This approach modifies traditional extractive distillation by adding a side rectifier, converting two-way vapor and liquid communication into one-way liquid transfer using an additional distillation section along with corresponding reboiler. Our present study investigates the dynamic control behavior for a strongly non-ideal system without and with heat integration. It focuses on achieving stable and reliable regulatory control by closely keeping the dual-product composition to their original design standards. Key control loops are pinpointed, with a specific emphasis on those centered around adjusting reboiler heat duty and side-stream flow rate within extractive column C1 to regulate the temperatures of Stages 48 and 36. Furthermore, control effectiveness is enhanced by introducing a feedforward action of reboiler heat duty-to-feed flowrate. By developing a new and robust control architecture, high feed flowrate and composition disturbances can be managed efficiently, solving the issues of oscillation and large transient response.

A Lightweight and High-Precision Passion Fruit YOLO Detection Model for Deployment in Embedded Devices.

In order to shorten detection times and improve average precision in embedded devices, a lightweight and high-accuracy model is proposed to detect passion fruit in complex environments (e.g., with backlighting, occlusion, overlap, sun, cloud, or rain). First, replacing the backbone network of YOLOv5 with a lightweight GhostNet model reduces the number of parameters and computational complexity while improving the detection speed. Second, a new feature branch is added to the backbone network and the feature fusion layer in the neck network is reconstructed to effectively combine the lower- and higher-level features, which improves the accuracy of the model while maintaining its lightweight nature. Finally, a knowledge distillation method is used to transfer knowledge from the more capable teacher model to the less capable student model, significantly improving the detection accuracy. The improved model is denoted as G-YOLO-NK. The average accuracy of the G-YOLO-NK network is 96.00%, which is 1.00% higher than that of the original YOLOv5s model. Furthermore, the model size is 7.14 MB, half that of the original model, and its real-time detection frame rate is 11.25 FPS when implemented on the Jetson Nano. The proposed model is found to outperform state-of-the-art models in terms of average precision and detection performance. The present work provides an effective model for real-time detection of passion fruit in complex orchard scenes, offering valuable technical support for the development of orchard picking robots and greatly improving the intelligence level of orchards.

Cross-modal knowledge distillation for continuous sign language recognition

Continuous Sign Language Recognition (CSLR) is a task which converts a sign language video into a gloss sequence. The existing deep learning based sign language recognition methods usually rely on large-scale training data and rich supervised information. However, current sign language datasets are limited, and they are only annotated at sentence-level rather than frame-level. Inadequate supervision of sign language data poses a serious challenge for sign language recognition, which may result in insufficient training of sign language recognition models. To address above problems, we propose a cross-modal knowledge distillation method for continuous sign language recognition, which contains two teacher models and one student model. One of the teacher models is the Sign2Text dialogue teacher model, which takes a sign language video and a dialogue sentence as input and outputs the sign language recognition result. The other teacher model is the Text2Gloss translation teacher model, which targets to translate a text sentence into a gloss sequence. Both teacher models can provide information-rich soft labels to assist the training of the student model, which is a general sign language recognition model. We conduct extensive experiments on multiple commonly used sign language datasets, i.e., PHOENIX 2014T, CSL-Daily and QSL, the results show that the proposed cross-modal knowledge distillation method can effectively improve the sign language recognition accuracy by transferring multi-modal information from teacher models to the student model. Code is available at https://github.com/glq-1992/cross-modal-knowledge-distillation_new.

Policy Compression for Intelligent Continuous Control on Low-Power Edge Devices.

Interest in deploying deep reinforcement learning (DRL) models on low-power edge devices, such as Autonomous Mobile Robots (AMRs) and Internet of Things (IoT) devices, has seen a significant rise due to the potential of performing real-time inference by eliminating the latency and reliability issues incurred from wireless communication and the privacy benefits of processing data locally. Deploying such energy-intensive models on power-constrained devices is not always feasible, however, which has led to the development of model compression techniques that can reduce the size and computational complexity of DRL policies. Policy distillation, the most popular of these methods, can be used to first lower the number of network parameters by transferring the behavior of a large teacher network to a smaller student model before deploying these students at the edge. This works well with deterministic policies that operate using discrete actions. However, many real-world tasks that are power constrained, such as in the field of robotics, are formulated using continuous action spaces, which are not supported. In this work, we improve the policy distillation method to support the compression of DRL models designed to solve these continuous control tasks, with an emphasis on maintaining the stochastic nature of continuous DRL algorithms. Experiments show that our methods can be used effectively to compress such policies up to 750% while maintaining or even exceeding their teacher's performance by up to 41% in solving two popular continuous control tasks.

Level of Contamination and Alteration of Cuttlefish Fished and Processed in Companies in Côte d’Ivoire

This study was carried out from 2019 to 2021. Its aim was to assess the quality of cuttlefish intended for local and international markets in a company processing fishery products in Côte d'Ivoire. The biochemical analyses (Total Volatile Basic Nitrogen (TVBN)) were based on the distillation method of an extract deproteinised by perchloric acid (HClO<sub>4</sub>). Microbiological analyses (total aerobic mesophilic flora (TAMF), staphylococci (SPP), anaerobic sulphite-reducing bacteria (RAS), faecal coliforms (FC) and salmonella) involved isolation and identification techniques. The various analyses carried out by the National Laboratory for Support to Agricultural Development (LANADA) made it possible to assess the median contamination values and satisfaction percentages (TAMF, FC, SPP), which are respectively 2.2.10<sup>4</sup> CFU/g with 94%; less than 10 CFU/g with 94%; less than 100 CFU/g with 100%. RSA and salmonella were absent in all the products analysed. The temperature of the cuttlefish was high just after processing. The maximum temperature value (26.4°C) was obtained after processing, while the minimum temperature value (1.6°C) was recorded during storage in a positive cold room. However, it is essential to improve manufacturing practices and preservation methods in order to avoid poisoning the population. These data should also encourage fishermen to practise good preservation methods.

Decoupled graph knowledge distillation: A general logits-based method for learning MLPs on graphs

While Graph Neural Networks (GNNs) have demonstrated their effectiveness in processing non-Euclidean structured data, the neighborhood fetching of GNNs is time-consuming and computationally intensive, making them difficult to deploy in low-latency industrial applications. To address the issue, a feasible solution is graph knowledge distillation (KD), which can learn high-performance student Multi-layer Perceptrons (MLPs) to replace GNNs by mimicking the superior output of teacher GNNs. However, state-of-the-art graph knowledge distillation methods are mainly based on distilling deep features from intermediate hidden layers, this leads to the significance of logit layer distillation being greatly overlooked. To provide a novel viewpoint for studying logits-based KD methods, we introduce the idea of decoupling into graph knowledge distillation. Specifically, we first reformulate the classical graph knowledge distillation loss into two parts, i.e., the target class graph distillation (TCGD) loss and the non-target class graph distillation (NCGD) loss. Next, we decouple the negative correlation between GNN’s prediction confidence and NCGD loss, as well as eliminate the fixed weight between TCGD and NCGD. We named this logits-based method Decoupled Graph Knowledge Distillation (DGKD). It can flexibly adjust the weights of TCGD and NCGD for different data samples, thereby improving the prediction accuracy of the student MLP. Extensive experiments conducted on public benchmark datasets show the effectiveness of our method. Additionally, DGKD can be incorporated into any existing graph knowledge distillation framework as a plug-and-play loss function, further improving distillation performance. The code is available at https://github.com/xsk160/DGKD.

Research on Isomorphic Task Transfer Algorithm Based on Knowledge Distillation in Multi-Agent Collaborative Systems.

In response to the increasing number of agents and changing task scenarios in multi-agent collaborative systems, existing collaborative strategies struggle to effectively adapt to new task scenarios. To address this challenge, this paper proposes a knowledge distillation method combined with a domain separation network (DSN-KD). This method leverages the well-performing policy network from a source task as the teacher model, utilizes a domain-separated neural network structure to correct the teacher model's outputs as supervision, and guides the learning of agents in new tasks. The proposed method does not require the pre-design or training of complex state-action mappings, thereby reducing the cost of transfer. Experimental results in scenarios such as UAV surveillance and UAV cooperative target occupation, robot cooperative box pushing, UAV cooperative target strike, and multi-agent cooperative resource recovery in a particle simulation environment demonstrate that the DSN-KD transfer method effectively enhances the learning speed of new task policies and improves the proximity of the policy model to the theoretically optimal policy in practical tasks.

Identification of internal voids in pavement based on improved knowledge distillation technology

Investigating methods for the detection of internal voids within road structures is a critical measure to ensure the safety and integrity of roadway operations. The purpose of this research is to investigate on the identification method of internal voids in pavement based on improved knowledge distillation technology. Ground penetrating radar data in three dimensions were extensively collected to capture the internal voids present within roadways, and this data was subsequently validated through in-situ verification. The echo characteristics of ground penetrating radar for areas with road voids were analyzed, and a dataset containing 1700 images of these internal voids was established. A YOLOv8 model improvement method was proposed, and a model for the detection of internal road voids was constructed based on the improved YOLO v8 framework. To further refine the model's performance, a knowledge distillation method based on multiple guidance from teacher assistants was developed. A stochastic learning approach was integrated, resulting in the establishment of a model optimized by this stochastic learning scheme for the identification of internal road voids. The results demonstrate that the presence of overfitting during the training phase of the void identification model can restrict its performance within a certain domain. The proposed stochastic learning-optimized, multi-teacher assistant guided knowledge distillation model, adeptly harnesses the performance benefits of both the teacher and assistant models by means of knowledge transfer, consequently achieving a significant improvement in the detection of internal road voids.

Shape-position perceptive fusion electronic skin with autonomous learning for gesture interaction

Wearable devices, such as data gloves and electronic skins, can perceive human instructions, behaviors and even emotions by tracking a hand's motion, with the help of knowledge learning. The shape or position single-mode sensor in such devices often lacks comprehensive information to perceive interactive gestures. Meanwhile, the limited computing power of wearable applications restricts the multimode fusion of different sensing data and the deployment of deep learning networks. We propose a perceptive fusion electronic skin (PFES) with a bioinspired hierarchical structure that utilizes the magnetization state of a magnetostrictive alloy film to be sensitive to external strain or magnetic field. Installed at the joints of a hand, the PFES realizes perception of curvature (joint shape) and magnetism (joint position) information by mapping corresponding signals to the two-directional continuous distribution such that the two edges represent the contributions of curvature radius and magnetic field, respectively. By autonomously selecting knowledge closer to the user's hand movement characteristics, the reinforced knowledge distillation method is developed to learn and compress a teacher model for rapid deployment on wearable devices. The PFES integrating the autonomous learning algorithm can fuse curvature-magnetism dual information, ultimately achieving human machine interaction with gesture recognition and haptic feedback for cross-space perception and manipulation.

Physical and Chemical Changes of Seven Selected Herbs Used as Herbal Bath Affected by Different Drying Methods

The effect of oven drying (OD) at 50±5°C, sun drying (SD), and fresh leaves (control) of seven selected herbs used in herbal preparation as herbal baths were evaluated. Herbal baths (HB) involve mixing herbs into water and boiling them or simply immersing them in the mixture during a regular bath. The herbs selected were leaves of torch ginger, greater galangal, pandan, citronella grass, henna, betel leaves, and kaffir lime leaves and fruits. The herbs were planted in a plot at the Institute of Bioscience, Universiti Putra Malaysia, Malaysia. Physical changes such as HB herbs’ colour, aroma, and chemical composition were evaluated. The hydro distillation method was used for the extraction process of HB herbs, where it produced essential oils (EO), essential water (EW), and boiling water (BW), and their chemical composition was determined by gas chromatography-mass spectrometry. As a result, the OD herbs possessed brighter and more attractive colours compared to the SD method, which was dull and pale. The colour of EO was yellow, colourless for EW, and reddish for BW. Additionally, OD herbs possessed 80% and only 50% of the scent strength of the SD herbs from extracting fresh herbs. The fresh and dried HB herbs contained chemical constituents mostly from the terpene group. Herbal preparation developed using all treatments retained an appreciable amount of chemical composition studied, thus having the potential for commercial purposes.

Characterization and Antioxidant Activity Assay of Essential Oil Parts of Rosella (Hibiscus sabdariffa L.)

Rosella (Hibiscus sabdariffa L.) is known as a plant whose calyx contain high levels of antioxidant bioactive compounds and good inhibitory power against free radicals. However, other parts of the rosella plant have not yet been utilized for their antioxidant content and activity. The aim of this research is to compare the profiles of essential oils of parts of the rosella plant (seeds, leaves and flower petals) and their antioxidant activity. This research method begins with the process of extracting essential oils using the water-steam distillation method. Furthermore, the profile of the essential oil is known from the results of yield calculations and characterization using the GC-MS instrument. The antioxidant activity was tested using the DPPH method. The yield of essential oils obtained was 0.0107% for seeds, 0.0087% for leaves and 0.0136% for flower petals. GC-MS characterization shows that the most abundant chemical components contained in the essential oils of seeds, leaves, and flower petals are fatty acid compounds (2-propanoic acid, butanoic acid, hexanoic acid) and esters (neopentyl isobutyrate). The results of the antioxidant activity test showed the IC50 value for each essential oil, namely, seeds: 30.15 μg/ml, leaves: 171.27 μg/ml and flower petals: 126.58 μg/ml. The compound characteristics obtained showed the best results for rosella seed essential oil, this was supported by the fairly high antioxidant activity test results IC50 <50 μg/ml. Keywords: Antioxidant activity, Characterization, Essential oils, IC50, Yield.

Natural deep eutectic solvents-based concurrent approach for qualitative and quantitative enhancement of Rosa damascena essential oil and recovery of phenolics from distilled rose petals

In pursuit of adopting environment friendly methodologies to amplify essential oil (EO) yield of industrially important aromatic plants, enrichment of bioactive content in distillation waste, and to establish the recovery process for the enriched product; this study employed natural deep eutectic solvents (NADESs) as cosolvent in the conventional hydro distillation method using Rosa damascena Mill L (Damask rose) petals as biomass. R. damascena is one of the most important species of Rosaceae family known for producing high-valued rose EO [often termed as “liquid gold”] but with very low yield. Herein, nine NADESs were used as cosolvent along with conventional hydro distillation method which led to 1.7 times EO yield enhancement (retention of EO quality as per ISO guidelines) using NADES-4 [choline chloride: glycerol (1:1)] as cosolvent. Further, to sustainably utilize the residue (otherwise waste) left after hydro distillation, hydrolate (aqueous part left in the flask) and spent (dried solid waste) were analyzed with UHPLC to determine targeted phenolics compounds. All the samples (EO, dried hydrolate, and spent) were also evaluated for in-vitro antioxidant activity using DPPH-radical scavenging assay, which revealed that hydrolate obtained by using NADES-5 [choline chloride: glucose (5:1)] showed the highest activity (IC50 0.85 µg/mL) as compared to ascorbic acid (IC50 5.40 µg/mL). Hence, the present study provides a green protocol for enhancing rose EO yield with phenolics-enriched antioxidant byproducts.

Efficient separation of alcohol ester via extractive distillation and pressure swing distillation based on molecular interaction mechanism analysis

N-amyl alcohol, n-amyl formate, n-propanol and n-propyl formate are all organic solvents with important industrial application value. In industrial production, related alcohol ester azeotropic systems are often generated. Therefore, it is necessary to study the refining process of these substances. The work designed purification processes for two different azeotropic systems of alcohol esters via extractive distillation and pressure swing distillation methods. The required binary interaction parameters were obtained through vapor–liquid equilibrium experiments. Based on experimental data, a suitable entrainer was determined through different methods of analysis. The processes were optimized with the economic cost as the objective. In addition, it evaluated and analyzed different processes from multiple perspectives. This work explores the purification process of alcohol ester systems, providing theoretical guidance for achieving efficient separation of alcohol ester systems.

An improved multi-scale and knowledge distillation method for efficient pedestrian detection in dense scenes

An improved multi-scale and knowledge distillation method for efficient pedestrian detection in dense scenes

Joint Dual Feature Distillation and Gradient Progressive Pruning for BERT compression

The increasing size of pre-trained language models has led to a growing interest in model compression. Pruning and distillation are the primary methods employed to compress these models. Existing pruning and distillation methods are effective in maintaining model accuracy and reducing its size. However, they come with limitations. For instance, pruning is often suboptimal and biased by transforming it into a continuous optimization problem. Distillation relies primarily on one-to-one layer mappings for knowledge transfer, which leads to underutilization of the rich knowledge in teacher. Therefore, we propose a method of joint pruning and distillation for automatic pruning of pre-trained language models. Specifically, we first propose Gradient Progressive Pruning (GPP), which achieves a smooth transition of indicator vector values from real to binary by progressively converging the values of unimportant units’ indicator vectors to zero before the end of the search phase. This effectively overcomes the limitations of traditional pruning methods while supporting compression with higher sparsity. In addition, we propose the Dual Feature Distillation (DFD). DFD adaptively globally fuses teacher features and locally fuses student features, and then uses the dual features of global teacher features and local student features for knowledge distillation. This realizes a “preview-review” mechanism that can better extract useful information from multi-level teacher information and transfer it to student. Comparative experiments on the GLUE benchmark dataset and ablation experiments indicate that our method outperforms other state-of-the-art methods.