Experimental Evaluation Research Articles

Integrating Experimental and Computational Analyses of Yeast Protein Profiles for Optimizing the Production of High-Quality Microbial Proteins.

Microbial proteins represent a promising solution to address the escalating global demand for protein, particularly in regions with limited arable land. Yeasts, such as Saccharomyces cerevisiae, are robust and safe protein-producing strains. However, the utilization of non-conventional yeast strains for microbial protein production has been hindered, partly due to a lack of comprehensive understanding of protein production traits. In this study, we conducted experimental analyses focusing on the growth, protein content, and amino acid composition of nine yeast strains, including one S. cerevisiae strain, three Yarrowia lipolytica strains, and five Pichia spp. strains. We identified that, though Y. lipolytica and Pichia spp. strains consumed glucose at a slower rate compared to S. cerevisiae, Pichia spp. strains showed a higher cellular protein content, and Y. lipolytica strains showed a higher glucose-to-biomass/protein yield and methionine content. We further applied computational approaches to explain that metabolism economy was the main underlying factor for the limited amount of scarce/carbon-inefficient amino acids (such as methionine) within yeast cell proteins. We additionally verified that the specialized metabolism was a key reason for the high methionine content in Y. lipolytica strains, and proposed Y. lipolytica strain as a potential producer of high-quality single-cell protein rich in scarce amino acids. Through experimental evaluation, we identified Pichia jadinii CICC 1258 as a potential strain for high-quality protein production under unfavorable pH/temperature conditions. Our work suggests a promising avenue for optimizing microbial protein production, identifying the factors influencing amino acid composition, and paving the way for the use of unconventional yeast strains to meet the growing protein demands.

Experimental evaluation of a multi-installment scheduling strategy based on divisible load paradigm for SAR image reconstruction on a distributed computing infrastructure

Radar loads, especially Synthetic Aperture Radar (SAR) image reconstruction loads use a large volume of data collected from satellites to create a high-resolution image of the earth. To design near-real-time applications that utilise SAR data, speeding up the image reconstruction algorithm is imperative. This can be achieved by deploying a set of distributed computing infrastructures connected through a network. Scheduling such complex and large divisible loads on a distributed platform can be designed using the Divisible Load Theory (DLT) framework. We performed distributed SAR image reconstruction experiments using the SLURM library on a cloud virtual machine network using two scheduling strategies, namely the Multi-Installment Scheduling with Result Retrieval (MIS-RR) strategy and the traditional EQual-partitioning Strategy (EQS). The DLT model proposed in the MIS-RR strategy is incorporated to make the load divisible. Based on the experimental results and performance analysis carried out using different pixel lengths, pulse set sizes, and the number of virtual machines, we observe that the time performance of MIS-RR is much superior to that of EQS. Hence the MIS-RR strategy is of practical significance in reducing the overall processing time, and cost, and in improving the utilisation of the compute infrastructure. Furthermore, we note that the DLT-based theoretical analysis of MIS-RR coincides well with the experimental data, demonstrating the relevance of DLT in the real world.

Discrete-time graph neural networks for transaction prediction in Web3 social platforms

In Web3 social platforms, i.e. social web applications that rely on blockchain technology to support their functionalities, interactions among users are usually multimodal, from common social interactions such as following, liking, or posting, to specific relations given by crypto-token transfers facilitated by the blockchain. In this dynamic and intertwined networked context, modeled as a financial network, our main goals are (i) to predict whether a pair of users will be involved in a financial transaction, i.e. the transaction prediction task, even using textual information produced by users, and (ii) to verify whether performances may be enhanced by textual content. To address the above issues, we compared current snapshot-based temporal graph learning methods and developed T3GNN, a solution based on state-of-the-art temporal graph neural networks’ design, which integrates fine-tuned sentence embeddings and a simple yet effective graph-augmentation strategy for representing content, and historical negative sampling. We evaluated models in a Web3 context by leveraging a novel high-resolution temporal dataset, collected from one of the most used Web3 social platforms, which spans more than one year of financial interactions as well as published textual content. The experimental evaluation has shown that T3GNN consistently achieved the best performance over time and for most of the snapshots. Furthermore, through an extensive analysis of the performance of our model, we show that, despite the graph structure being crucial for making predictions, textual content contains useful information for forecasting transactions, highlighting an interplay between users’ interests and economic relationships in Web3 platforms. Finally, the evaluation has also highlighted the importance of adopting sampling methods alternative to random negative sampling when dealing with prediction tasks on temporal networks.

Ischemic stroke infarct segmentation model based on depthwise separable convolution for multimodal magnetic resonance imaging

Magnetic resonance imaging (MRI) plays a crucial role in the diagnosis of ischemic stroke. Accurate segmentation of the infarct is of great significance for selecting intervention treatment methods and evaluating the prognosis of patients. To address the issue of poor segmentation accuracy of existing methods for multiscale stroke lesions, a novel encoder-decoder architecture network based on depthwise separable convolution is proposed. Firstly, this network replaces the convolutional layer modules of the U-Net with redesigned depthwise separable convolution modules. Secondly, an modified Atrous spatial pyramid pooling (MASPP) is introduced to enlarge the receptive field and enhance the extraction of multiscale features. Thirdly, an attention gate (AG) structure is incorporated at the skip connections of the network to further enhance the segmentation accuracy of multiscale targets. Finally, Experimental evaluations are conducted using the ischemic stroke lesion segmentation 2022 challenge (ISLES2022) dataset. The proposed algorithm in this paper achieves Dice similarity coefficient (DSC), Hausdorff distance (HD), sensitivity (SEN), and precision (PRE) scores of 0.816 5, 3.668 1, 0.889 2, and 0.894 6, respectively, outperforming other mainstream segmentation algorithms. The experimental results demonstrate that the method in this paper effectively improves the segmentation of infarct lesions, and is expected to provide a reliable support for clinical diagnosis and treatment.

In-depth research on azole corrosion inhibitors - effect of N atoms position and quantity on copper corrosion inhibition performance in alkaline environments: Experimental evaluation and theoretical analysis

Copper's superior electrical properties make it ideal for multi-layer interconnections in integrated circuits. Chemical mechanical polishing (CMP) is the main method for achieving copper wiring flatness. However, copper is susceptible to corrosion in glycine-based alkaline polishing solutions. This study examines benzimidazole, indazole, and benzotriazole as corrosion inhibitors for copper in an alkaline environment (pH=9), focusing on the impact of nitrogen atom arrangement on inhibition efficiency. Through electrochemical, contact angle, static corrosion, and surface characterization tests, it's demonstrated that these inhibitors protect copper surfaces effectively. The adsorption of these inhibitors follows the Langmuir equation with a mixed adsorption mode. X-ray photoelectron spectroscopy (XPS), UV–vis spectroscopy tests, and density functional theory calculations elucidate the adsorption configuration and corrosion inhibition mechanism. The scope of this research extends beyond merely analyzing the performance of the three corrosion inhibitors; more importantly, it has yielded novel insights that are broadly applicable to the field of corrosion inhibitors. The findings are as follows: 1. The coupled structure of dinitrogen and trinitrogen facilitates charge accumulation, manifesting as the formation of additional Cu-N bonds, which prevents desorption; 2. The multi-nitrogen coupled structures provide sites for the formation of intermolecular hydrogen bonds, which is essential for the establishment of an orderly and dense passivation layer. It is expected that this study will provide valuable insights into the development and design of new and efficient copper corrosion inhibitors.

Development and implementation of the course entitled "Virtual Simulation Experiment of Recombinant Human Erythropoietin Manufacturing Process"

Considering the issues present in traditional learning methods of manufacturing process for biotechnology majors, this paper presents the development and implementation process of the course entitled "Virtual Simulation Experiment of Recombinant Human Erythropoiesis Manufacturing Process". The experiment combines modern biological manufacturing technology and three-dimensional information technology, with recombinant human erythropoiesis drug serving as the focal point. This paper elaborates on the teaching concepts, objectives, contents, implementation methods, experimental procedures, interactive steps, and assessment criteria used in the experiment. Through innovative experimental scheme design, teaching methodologies, and evaluation systems, this course aims to cultivate students' analytical and problem-solving skills in the field of biopharmaceutical engineering, while also broadening students' perspective and expanding their vision.

Experimental evaluation of geotextile in-filled sand cushioning system against rock fall impact using 2D digital image correlation technique

Rockfall on slopes always poses potential threat to the safety of people and infrastructures. Though, constructions of rockfall protection shelters were considered an appropriate solution, still there is a lack of understanding on rock mass-structure-sand cushion interaction which was mainly due to the impact and rebound characteristics of the detached falling rock mass. Considering rockfall induced impact response, geotextile in-filled sand bags have been developed in this study as a confined energy dissipation member against rock fall conditions. Using scaled down model structure, the performance of geotextile in-filled sand cushioning system against impact forces were evaluated and compared. Additionally, the study specially used 2D Digital Image Correlation technique for displacement, strain estimation and force calculation measurement on the structure and geotextile in-filled sandbags due to rock fall impact. The observations proved that, provision of geotextile in-filled sand bags over the structure showed 99% energy dissipation against impact load and can be an alternative to conventional sand cushioning layer with minimum additional load on the rock fall protection system. Also, 2D DIC monitoring found reliable in impact application studies for estimating force and displacement measurement.

AI-Driven Real-Time Classification of ECG Signals for Cardiac Monitoring Using i-AlexNet Architecture.

The healthcare industry has evolved with the advent of artificial intelligence (AI), which uses advanced computational methods and algorithms, leading to quicker inspection, forecasting, evaluation and treatment. In the context of healthcare, artificial intelligence (AI) uses sophisticated computational methods to evaluate, decipher and draw conclusions from patient data. AI has the potential to revolutionize the healthcare industry in several ways, including better managerial effectiveness, individualized treatment regimens and diagnostic improvements. In this research, the ECG signals are preprocessed for noise elimination and heartbeat segmentation. Multi-feature extraction is employed to extract features from preprocessed data, and an optimization technique is used to choose the most feasible features. The i-AlexNet classifier, which is an improved version of the AlexNet model, is used to classify between normal and anomalous signals. For experimental evaluation, the proposed approach is applied to PTB and MIT_BIH databases, and it is observed that the suggested method achieves a higher accuracy of 98.8% compared to other works in the literature.

Experimental evaluation and optimization of parameters affecting delamination, geometrical tolerance and surface roughness in ultrasonic drilling of 3D-Printed PLA thermoplastic

Ultrasonic drilling is a suitable process to enhance the generated surfaces by additive manufacturing. In this study, polylactic acid was selected as the workpiece. The examination parameters were thrust force, delamination, geometrical tolerance, chip adhesion, hole wall morphology and surface roughness. It was explained that the harmonic movement of drill bit in ultrasonic drilling reduced thrust force, delamination, circularity, cylindricality and surface roughness up to 14.5%, 3.7%, 44%, 38%, 20% respectively and removed chip adhesion. Furthermore, number of end-mill flutes was examined and observed that 4-flutes compared to 2-flutes induced reduction in thrust force, delamination, circularity, cylindricality and surface roughness up to 15.2%, 2%, 7.5%, 18.9%, 12.5% respectively. Besides, analysis of variance was established to determine the significant parameters. Finally, non-dominated sorting genetic algorithm-II technique was implemented in order to carry out multi-response optimization.

A defect detection method for industrial aluminum sheet surface based on improved YOLOv8 algorithm

In industrial aluminum sheet surface defect detection, false detection, missed detection, and low efficiency are prevalent challenges. Therefore, this paper introduces an improved YOLOv8 algorithm to address these issues. Specifically, the C2f-DSConv module incorporated enhances the network’s feature extraction capabilities, and a small target detection layer (160 × 160) improves the recognition of small targets. Besides, the DyHead dynamic detection head augments target representation, and MPDIoU replaces the regression loss function to refine detection accuracy. The improved algorithm is named YOLOv8n-DSDM, with experimental evaluations on an industrial aluminum sheet surface defect dataset demonstrating its effectiveness. YOLOv8n-DSDM achieves an average mean average precision (mAP50%) of 94.7%, demonstrating a 3.5% improvement over the original YOLOv8n. With a single-frame detection time of 2.5 ms and a parameter count of 3.77 M, YOLOv8n-DSDM meets the real-time detection requirements for industrial applications.

Experimental Evidence of Efficient Phononic-Based Vibration Isolators for Mechanical Applications

Even though the design of vibration isolators is well-established for many engineering applications, their efficiency in wide and multiple frequency ranges is still a challenge. In these cases, the use of Phononic-Based Vibration Isolators (PBVIs) may be advantageous as they present different Attenuation Regions (ARs) in which the elastic waves are strongly attenuated. Therefore, the present paper is devoted to the experimental evaluation, in terms of force transmissibility, of different types of supporting devices tested on a load mass and a motor of a Hermetic Compressor (HC). Those devices are the original Helical Coil Spring (HS) that equips the HC, the PBVI, and the Combined Structure (CS) which is composed of a PBVI combined in series with the HS. Results evidentiate the capability of the CSs to isolate vibrations, where the PBVI contributes with its ARs, thus operating as a “filter” in specific frequency ranges, while the HSs maintain the flexibility of the CSs, which is advantageous for impact-loads and/or transient-case scenarios. Hence, the capability, relevance and impact that these PBVIs present for force transmissibility reduction applications is highlighted here, which should capture the attention of and motivate the industry, e.g., producers of isolation systems, since it has wide-ranging engineering applications.

Experimental evaluation of radiation shielding characteristics of borate-based-glass system reinforced with titanium oxide

In this work, the attenuation capability of new Borate-based-glass system reinforced with titanium oxide containing PbO, CaO, TiO2 and B2O3 with different percentages was determined. Four different glass samples were synthesized by experimental technique using high pure germanium detector at different gamma ray sources using narrow beam method. The experimental method was verified by XCOM software. The values of the two methods were close and the relative deviation was satisfactory (less than 5 %) for all gamma energies (0.060, 0.662, 1.173 and 1.333 MeV). The highest linear attenuation coefficient (LAC) values achieved at low radiation energy of 0.0595 MeV. The Highest LAC values were 13.64, 13.97, 14.32 and 14.67 cm−1 for samples PCBT-1, PCBT-2, PCBT-3 and PCBT-4 in the same order, respectively. Conversely, the lowest LAC values in the examined energy range were found at 1.333 MeV. The lowest LAC values were 0.259, 0.266, 0.272 and 0.279 cm−1 for the aforementioned glasses, respectively. At 0.662 MeV, the transmission factor (TF) values decreases as 64.18, 63.54, 62.88 and 62.19 % as TiO2 content increases as 5, 10, 15 and 20 mol % in the same order, respectively. The decreases in TF values was linked to an observable increase in the radiation protection efficiency (RPE). The increase in RPE with the increase of TiO2 content confirms that the addition of TiO2 leads to an improvement in the radiation shielding performance for the prepared glasses.

Design and experimental evaluation of a novel flow-mode magnetorheological damper without accumulator

Researchers in the field of vibration control have shown increasing interest in magneto-rheological dampers (MRDs) in recent years. Conventional flow-mode MRDs typically employ a gas chamber as an accumulator to accommodate volume changes and promote fluid communication. However, this approach introduces manufacturing complexities and raises production costs. To overcome these challenges, we propose a novel configuration for flow-mode MRDs that replaces the accumulator with a structural constraint. This modification leads to a more compact and cost-effective MRD solution suitable for engineering applications. This paper presents an introduction, followed by the configuration and design of the novel MRD for a case study involving a vehicle suspension system. To enhance output performance, we optimize the significant geometry of the damper using the finite element method (FEM), taking into account the damping force, off-state force, and inductive time constant of the damper. Based on the optimal simulation results, we provide a detailed design of the optimized flow-mode MRD without an accumulator for prototype fabrication. To assess the practical performance of the proposed MRD prototype, we conducted experiments on a test rig and engaged in comprehensive discussions based on the obtained results.

Effect of lithium-containing inorganic phosphate additives in stabilization of carbonate-based electrolyte for 5 V LiNi0.5Mn1.5O4-based lithium-ion batteries

High-voltage lithium-ion batteries (LIBs) with LiNi0.5Mn1.5O4 (LNMO) cathode are considered promising energy sources due to their high energy density. However, enhancing the stability of the electrode/electrolyte interface remains critical for the practical application of LIBs. In this work, lithium-containing inorganic phosphates (LiPO3, Li3PO4, and LiH2PO4) were used as multifunctional inorganic additives to improve the stability of the electrode/electrolyte interface in 5.0 V, LNMO-based LIBs with carbonate-based electrolytes. Based on theoretical calculations and experimental evaluations of the electrochemical performance, LiPO3 was selected as the optimal additive for further study in LNMO/Li and LNMO/Li4Ti5O12 cells. The LNMO/Li half-cells with LiPO3 delivered an initial discharge capacity of 108.8 mAh∙g−1 and retained a capacity of 94.9 % after 500 cycles at 5C, compared to 75.1 mAh∙g−1 and 79.4 % for the standard electrolyte. The increased cycling stability was ascribed to the ability of the multifunctional, film-forming LiPO3 additive to form stable interfacial films on both the cathode and anode, thus, inhibiting side reactions between the electrodes and electrolyte. Moreover, as an HF scavenger, LiPO3 reduced the HF content in the electrolyte to ensure the structural integrity of LNMO and inhibit the dissolution of transition metal ions. In the LNMO/Li4Ti5O12 full cells, the addition of LiPO3 effectively enhanced the cycling performance, and the capacity retention after 400 cycles at 5C was 79.1 %, compared to only 12.3 % for the standard electrolyte. Considering its simplicity, low cost, and feasibility, the addition of LiPO3 as a multifunctional electrolyte additive showed excellent prospects for commercial applications.

Monocular depth estimation using multi-dimensional dynamic convolution

To address the issue of inaccurate prediction results and low accuracy at the object edges in existing self-supervised monocular depth estimation algorithms, this study proposes a self-supervised monocular depth estimation algorithm based on dynamic convolution (OE-Depth). The algorithm employs a multi-dimensional dynamic convolution feature extraction network to acquire more comprehensive feature representations, thereby enhancing the predictive capability. Additionally, the algorithm is optimized by integrating a triplet loss term and employing metric learning techniques to refine the networks performance at the object edges. Experimental evaluations conducted on the KITTI dataset validate the effectiveness of the proposed algorithm, demonstrating an optimization of the error class index by nearly 10%. Notably, the most stringent criterion < 1.25 achieves an accuracy of 0.908 for depth prediction.

Research and Application of the Median Filtering Method in Enhancing the Imperceptibility of Perturbations in Adversarial Examples

In the field of object detection, the adversarial attack method based on generative adversarial network efficiently generates adversarial examples, thereby significantly reducing time costs. However, this approach overlooks the imperceptibility of perturbations in adversarial examples, resulting in poor visual performance and insufficient invisibility of the generated adversarial examples. To further enhance the imperceptibility of perturbations in adversarial examples, a method utilizing median filtering is proposed to address these generated perturbations. Experimental evaluations were conducted on the Pascal VOC dataset. The results demonstrate that, compared to the original image, there is an increase of at least 17.2% in the structural similarity index (SSIM) for generated adversarial examples. Additionally, the peak signal-to-noise ratio (PSNR) increases by at least 27.5%, while learned perceptual image patch similarity (LPIPS) decreases by at least 84.6%. These findings indicate that the perturbations in generated adversarial examples are more difficult to detect, with significantly improved imperceptibility and closer resemblance to the original image without compromising their high aggressiveness.

Graph transformer embedded deep learning for short-term passenger flow prediction in urban rail transit systems: A multi-gate mixture-of-experts model

Urban rail transit (URT) plays a crucial role in mitigating urban traffic congestion by offering faster and higher-quality travel services. Short-term passenger flow predictions have practical significance for metro management and operation. However, the complex spatiotemporal characteristics and the relationship between entry and exit passenger flows make it challenging to detect the dynamic evolution patterns. This study proposes a Spatio-Temporal Graph Transformer (STGT) under the multi-task learning framework, utilizing Graph Transformer network and gated residual units to select and aggregate features. To account for the correlation between entry and exit passenger flow prediction tasks, the STGT model integrates a Multi-gate Mixture-of-Experts (MMoE) approach, which combines different expert networks for diverse input and explicitly learns to model passenger flow relationships in various scenarios. Metro-related characteristics such as weather conditions, train operation characteristics, and accessibility of nearby bus stops are incorporated to enhance prediction accuracy. Experimental evaluations are conducted using real-world historical passenger travel records from the Beijing subway. The results demonstrate the superior robustness and advantages of the STGT-MMoE model over basic and advanced benchmarks for passenger flow prediction tasks. These findings provide compelling evidence to address the challenges of short-term inflow and outflow prediction in urban rail transit systems.

Finite time non-singular terminal synergistic control of hydraulic servo system considering time-varying disturbance

This paper introduces a novel control strategy combining a hybrid nonlinear disturbance observer with a finite-time synergistic control approach for hydraulic servo systems, focusing on compensating time-varying disturbances and suppresses chattering. Recognizing the critical challenges posed by these disturbances and the chattering phenomenon in hydraulic servo systems, our work offers significant advancements in enhancing system robustness and control precision. Through meticulous classification and handling of matched and unmatched disturbances, the study unveils a dual-channel finite-time disturbance observer capable of simultaneously addressing both disturbance types with high accuracy, particularly focusing on the often overlooked and challenging to observe friction effects. Key contributions include the pioneering integration of cooperative control theory with sliding mode control, employing fractional-order terms to minimize chattering, thereby improving system stability. The experimental evaluation underlines the proposed controller’s superior performance over traditional control strategies, including PID, TSMC, and SMC, under both low and high-frequency operational conditions. Notably, the proposed FTNTSC controller demonstrates exceptional robustness and effectiveness, with significantly lower performance degradation when transitioning from low to high-frequency conditions compared to its counterparts.

Experimental evaluation and design of fillet welds under in-plane eccentric loads

The principal objective of this investigation is to furnish equations to compute the design strength of fillet welds under in-plane eccentric loads suitable for load and resistance factor design (LRFD) applications. Over the years, experimental and analytical techniques have been used to arrive at safe estimates for the strength of such welds in common practical applications. The conventional elastic design (ED) method is very conservative and is not compatible with the LRFD design philosophy. In addition, the ED method neglects the weld group's ductility and strain compatibility. On the other hand, the instantaneous center of rotation (IC) method considers the above parameters but requires complex calculations. In the present study, a method for designing fillet welds under eccentric loading is introduced called the plastic design (PD) method. This method considers the inelastic properties of the welds in a simple manner while providing an accurate prediction of the weld group's capacity. The performance of the proposed PD method was compared with the IC method for several cases. The ultimate loads obtained were mostly within 10% difference on the conservative side and only marginally different for a few. Additionally, the fillet weld group's resistance under in-plane eccentric loads was evaluated experimentally using 12C-shaped specimens with the load applied at different angles of 0, 45, and 75°. The test results also confirmed the efficacy of the proposed PD method in evaluating the design strength of such weld groups.

An efficient ensemble explainable AI (XAI) approach for morphed face detection

Numerous deep neural convolutional architectures have been proposed in literature for face Morphing Attack Detection (MADs) to prevent such attacks and lessen the risks associated with them. Although, deep learning models achieved optimal results in terms of performance, it is difficult to understand and analyze these networks since they are black box/opaque in nature. As a consequence, incorrect judgments may be made. There is, however, a dearth of literature that explains decision-making methods of black box deep learning models for biometric Presentation Attack Detection (PADs) or MADs that can aid the biometric community to have trust in deep learning-based biometric systems for identification and authentication in various security applications such as border control, criminal database establishment etc. In this work, we present a novel visual explanation approach named Ensemble XAI integrating Saliency maps, Class Activation Maps (CAM) and Gradient-CAM (Grad-CAM) to provide a more comprehensive visual explanation for a deep learning prognostic model (EfficientNet-B1) that we have employed to predict whether the input presented to a biometric authentication system is morphed or genuine. The experimentations have been performed on three publicly available datasets namely Face Research Lab London (FRLL) dataset, Wide Multi-Channel Presentation Attack (WMCA) dataset, and Makeup Induced Face Spoofing (MIFS) dataset. The experimental evaluations affirms that the resultant visual explanations highlight more fine-grained details of image features/areas focused by EfficientNet-B1 to reach decisions along with appropriate reasoning.