Generation Method Research Articles

Image-processing-based model for surface roughness evaluation in titanium based alloys using dual tree complex wavelet transform and radial basis function neural networks.

In this study, we examine the assessment of surface roughness on turned surfaces of Ti 6Al 4V using a computer vision system. We utilize the Dual-Tree Complex Wavelet Transform (DTCWT) to break down the images of the turned surface into sub-images oriented in directions. Three different methods of feature generation have been compared, i.e., the use of Gray-Level Co-Occurrence Matrix (GLCM) and DTCWT-based extraction of second-order statistical features, DTCWT Image fusion, and the use of GLCM for feature extraction, and DTCWT image fusion using Particle Swarm Optimization (PSO) based GLCM features. Principal Component Analysis (PCA) was utilized to identify and select features. The model was developed using a Radial Basis Function Neural Network (RBFNN). Accordingly, six models were designed based on the three feature generation methods, considering all features and features selected using PCA. The RBFNN model, which incorporates DTCWT Image fusion and utilizes PSO with PCA features, achieved a training data prediction accuracy of 100% and a test data prediction accuracy of 99.13%.

Study on Path Planning in Cotton Fields Based on Prior Navigation Information

Aiming at the operation scenario of existing crop coverage and the need for precise row alignment, the sowing prior navigation information of cotton fields in Xinjiang was used as the basis for the study of path planning for subsequent operations to improve the planning quality and operation accuracy. Firstly, the characteristics of typical turnaround methods were analyzed, the turnaround strategy for dividing planning units was proposed, and the horizontal and vertical operation connection methods were put forward. Secondly, the obstacle avoidance strategies were determined according to the traits of obstacles. The circular arc–linear and cubic spline curve obstacle avoidance path generation methods were proposed. Considering the dual attributes of walking and the operation of agricultural machinery, four kinds of operation semantic points were embedded into the path. Finally, path generation software was designed. The simulation and field test results indicated that the operation coverage ratio CR ≥ 98.21% positively correlated with the plot area and the operation distance ratio DR ≥ 86.89% when non-essential reversing and obstacles were ignored. CR and DR were negatively correlated with the number of obstacles when considering obstacles. When considering non-essential reversing, the full coverage of operating rows could be achieved, but DR would be reduced correspondingly.

BEGAN: Boltzmann-Reweighted Data Augmentation for Enhanced GAN-Based Molecule Design in Insect Pheromone Receptors.

Identifying small molecules that bind strongly to target proteins in rational molecular design is crucial. Machine learning techniques, such as generative adversarial networks (GAN), are now essential tools for generating such molecules. In this study, we present an enhanced method for molecule generation using objective-reinforced GANs. Specifically, we introduce BEGAN (Boltzmann-enhanced GAN), a novel approach that adjusts molecule occurrence frequencies during training based on the Boltzmann distribution exp(-ΔU/τ), where ΔU represents the estimated binding free energy derived from docking algorithms and τ is a temperature-related scaling hyperparameter. This Boltzmann reweighting process shifts the generation process toward molecules with higher binding affinities, allowing the GAN to explore molecular spaces with superior binding properties. The reweighting process can also be refined through multiple iterations without altering the overall distribution shape. To validate our approach, we apply it to the design of sex pheromone analogs targeting Spodoptera frugiperda pheromone receptor SfruOR16, illustrating that the Boltzmann reweighting significantly increases the likelihood of generating promising sex pheromone analogs with improved binding affinities to SfruOR16, further supported by atomistic molecular dynamics simulations. Furthermore, we conduct a comprehensive investigation into parameter dependencies and propose a reasonable range for the hyperparameter τ. Our method offers a promising approach for optimizing molecular generation for enhanced protein binding, potentially increasing the efficiency of molecular discovery pipelines.

Black-Box Adversarial Attacks Against SQL Injection Detection Model

Structured Query Language (SQL) injection attacks represent a substantial threat to the security of web applications, making the development of effective detection techniques crucial. These techniques have evolved from traditional signature-based techniques to more advanced techniques based on machine learning models. Machine learning detection models are often vulnerable to adversarial examples. Adversarial examples are deliberately crafted inputs designed to deceive models into making incorrect predictions by subtly altering the original dataset in ways that are typically imperceptible to humans. To train and test these machine learning models, datasets comprising both malicious and normal data are indispensable. However, the lack of sufficient and balanced datasets presents a significant challenge, particularly for models intended to detect SQL injection attacks. Most network traffic datasets exhibit a substantial class imbalance, with a disproportionate amount of normal traffic compared to malicious traffic, making it difficult to train effective and reliable detection models. This study addressed the shortcomings of current SQL injection detection techniques and proposed a conditional tabular generative adversarial network adversarial attack method. We evaluated the effectiveness of the generated adversarial SQL injection examples using qualitative and quantitative methods, measuring their ability to evade the detection model. A conventional neural network algorithm detection model was built and tested, and the generated adversarial examples successfully bypassed the detection model at a rate of up to 6%. The evidence demonstrated that the conditional tabular generative adversarial network successfully captures the statistical properties of real data and generates synthetic data that accurately represents the real data. This method is also expected to address the problem of insufficient and imbalanced SQL injection datasets, which could aid in training various machine learning models beyond the one used in our study.

Impact of Partitioning Temperature Parameters during the Quenching and Partitioning Process on Medium Carbon Steels with Two Different Alloying Strategies

Quenching and partitioning (Q&P) heat treatments have recently gained attention as promising methods for the third generation of advanced high‐strength steels, particularly in industrial applications like automotive. This study investigates the microstructural evolution during Q&P in two medium‐carbon high‐silicon and aluminum‐alloyed steels, exploring potential additional phase transformations controlling the final structure. The choice to focus on Si and Al‐ medium‐carbon steel is linked to the lower cost of these elements compared to commonly alloying elements like Ti, Cr, Mo, and V, while still achieving high mechanical properties through Q&P. The Q&P process is analyzed by varying the volume fraction of primary martensite (M1) at 0.25, 0.50, and 0.75, with partitioning temperatures ranging from 350 to 550 °C for 30 min. At 350 °C, a significant volume fraction of stabilized austenite (up to 0.3) is observed. However, concurrent reactions such as nanostructured bainite and martensite formation lead to deviations from the theoretical constrained carbon equilibrium (CCE). At higher temperatures (450–550 °C), tempering reactions, including cementite precipitation and pearlite formation, reduced the austenite final fraction. The study highlights that heat treatment design, particularly partitioning temperature, must be tailored to the specific steel composition due to the varying effects of Si and Al.

Generation mechanism and anti-friction effect evaluation of continuous micro-groove texture machined by ball-end milling process

Abstract The tool path in ball-end milling process is flexible and controllable considering the structure characteristics and moving way of the cutter. This paper aims to reveal the generation method of continuous surface micro-groove texture by using ball-end milling process and evaluate the corresponding anti-friction performance by combining simulation and friction experiments. Results indicate that continuous micro-grooved surface texture can be obtained by combining small feed per tooth and large radial cutting depth. Radial depth of cut and tool radius affect aspect ratio of micro-groove. The bearing capacity of the oil film shows an increasing and then decreasing trend with the increase of radial depth of cut and orientation angle. Radial depth of cut increased from 0.1 mm to 0.3 mm, resulting in a maximum 4.87% increase in the bearing capacity of oil film because of the enhanced wedge effect. However, further increase in radial depth of cut leads to increased vortex, weakening the wedge effect. The oil film bearing capacity is reduced so that the friction coefficient increases by a maximum of 510%. Moreover, as the orientation angle increases from 10° to 20°, the bearing capacity of the oil film increases by 22.7%. The minimum friction coefficient currently is 0.0215. However, when the micro-groove orientation angle is further increased, the countercurrent effect is enhanced and the dynamic pressure effect is weakened, which finally leads to a decrease in the bearing capacity of the oil film. The friction coefficient has increased by 30.2% maximum. This study is of great significance which provides a new method for guiding the friction reduction on the die surface.

The Generative Generic-Field Design Method Based on Design Cognition and Knowledge Reasoning

Large language model (LLM) and Crowd Intelligent Innovation (CII) are reshaping the field of engineering design and becoming a new design context. Generative generic-field design can solve more general design problems innovatively by integrating multi-domain design knowledge. However, there is a lack of knowledge representation and design process model in line with the design cognition of the new context. It is urgent to develop generative generic-field design methods to improve the feasibility, innovation, and empathy of design results. This study proposes a method based on design cognition and knowledge reasoning. Firstly, through the problem formulation, a generative universal domain design framework and knowledge base are constructed. Secondly, the knowledge-based discrete physical structure set generation method and system architecture generation method are proposed. Finally, the application tool Intelligent Design Assistant (IDA) is developed, verified, and discussed through an engineering design case. According to the design results and discussion, the design scheme is feasible and reflects empathy for the fuzzy original design requirements. Therefore, the method proposed in this paper is an effective technical scheme of generative generic-field engineering design in line with the design cognition in the new context.

Optically tuned soliton dynamics in Bose-Einstein condensates within dark traps

Abstract This study investigates the formation and dynamics of solitons in Bose–Einstein condensates (BECs) within dark traps generated by two crossed Laguerre–Gaussian (LG) beams with varying azimuthal indices ℓ . As the index ℓ increases, the potential transitions from a harmonic trap when ℓ = 1 to a square-well potential for larger values of ℓ . This transition allows us to study a range of soliton dynamics under different confinement conditions while maintaining the same BEC volume. Through the derivation of the Gross-Pitaevskii equation (GPE) and under these specific conditions in both one-dimensional and two-dimensional configurations, we explore the dynamics of solitons across multiple scenarios. The study examines two primary methods for solitons generation: the temporal modulation of the scattering length and the implementation of an initial potential barrier that is subsequently removed. The results indicate that the trap shape plays a critical role in the generation and interaction dynamics of solitons. In harmonic traps, solitons exhibit a behavior different from those observed in anharmonic traps, where the dynamics is significantly influenced by the azimuthal index of the trap. The ability to control soliton dynamics in BECs holds significant promise for applications in quantum technologies, precision sensing, and the exploration of fundamental quantum phenomena.

Applications and Challenges of AI in PCB X-ray Inspection: A Comprehensive Study

As printed circuit boards (PCBs) continue to evolve in complexity and miniaturization, the demand for robust and efficient inspection techniques has become paramount in ensuring the quality and reliability of electronic devices. The application of machine learning and deep learning techniques has revolutionized PCB inspection in recent years, enabling the ability to automate and improve numerous elements of the process. In this paper, a comprehensive analysis is performed on the applications and challenges of artificial intelligence (AI), encompassing techniques of deep learning and machine learning, in the domain of PCB X-ray scrutiny. The main focus of this research centers around defect detection, identification of components and layers, deep learning algorithms for image reconstruction, as well as the identification of defects and features in advanced packaging. This study examines the current cutting-edge advancements in each of these areas, closely examining the existing methodologies and technologies employed. Furthermore, it delves into the limitations and challenges inherent in PCB X-ray inspection, such as the unavailability of data, computational demands, and the interpretability of models. In addition, this article offers prospective insights and presents promising avenues like application of generative adversarial networks and deep learning reconstruction methods for future exploration.

BDD-GAN: a brocade image generation method via dual discriminator and feature fusion

BDD-GAN: a brocade image generation method via dual discriminator and feature fusion

Spectroscopy-guided discovery of three-dimensional structures of disordered materials with diffusion models

Abstract Spectroscopy techniques such as x-ray absorption near edge structure (XANES) provide valuable insights into the atomic structures of materials, yet the inverse prediction of precise structures from spectroscopic data remains a formidable challenge. In this study, we introduce a framework that combines generative artificial intelligence models with XANES spectroscopy to predict three-dimensional atomic structures of disordered systems, using amorphous carbon (a-C) as a model system. In this work, we introduce a new framework based on the diffusion model, a recent generative machine learning method, to predict 3D structures of disordered materials from a target property. For demonstration, we apply the model to identify the atomic structures of a-C as a representative material system from the target XANES spectra. We show that conditional generation guided by XANES spectra reproduces key features of the target structures. Furthermore, we show that our model can steer the generative process to tailor atomic arrangements for a specific XANES spectrum. Finally, our generative model exhibits a remarkable scale-agnostic property, thereby enabling generation of realistic, large-scale structures through learning from a small-scale dataset (i.e. with small unit cells). Our work represents a significant stride in bridging the gap between materials characterization and atomic structure determination; in addition, it can be leveraged for materials discovery in exploring various material properties as targeted.

Defending Against Deep Learning-Based Traffic Fingerprinting Attacks With Adversarial Examples

In an increasingly digital and interconnected world, online anonymity and privacy are paramount issues for Internet users. To address this, tools like The Onion Router (Tor) offer anonymous and private communication by routing traffic through multiple relays with multiple layers of encryption. However, traffic fingerprinting attacks have threatened anonymity and privacy. In response, the community has proposed additional defenses for Tor, but fingerprinting techniques that utilize deep neural networkss (DNNs) have undermined many of these defenses. The latest defenses that are both lightweight and robust against DNNs use adversarial examples, but these defenses require either the full traffic trace beforehand or a database of pre-computed adversarial examples. We propose Prism , a defense against fingerprinting attacks that utilizes adversarial examples with neither prior access to the full traffic trace nor a database. We describe a novel method of adversarial example generation as input is learned over time. Prism injects these adversarial examples into the Tor traffic stream to prevent DNNs from accurately classifying both websites and videos that a user is viewing, even if the DNN is hardened by adversarial training. We also show that the Tor network could implement Prism entirely on relays under certain conditions, extending the defense to users who may run Tor on devices without graphics processing units.

Machine learning based segmentation of delamination patterns from sparse ultrasound data of barely visible impact damage in composites

Composite laminates, known for their strength and flexibility, are widely used but can delaminate under certain barely visible impact. Ultrasound Transmission (UT) scans can detect such failures, but the noise in the scans makes automation of delamination morphology challenging. Machine learning could solve this, but it requires substantial training data. Furthermore, given the considerable time and cost of conducting impact tests on composite laminates, such data remains sparse. This study overcomes data sparsity by enlarging UT scan dataset using image augmentation techniques and synthetic data generation methods. A combination of rotation and elastic deformation of real UT images produced augmented data. Synthetic data was generated by mimicking the statistical variations in real data, including delamination length and rotation per ply, and adding gradients to match the original image noise. A U-Net based machine learning segmentation approach was utilized to capture local and global information through a series of convolutions to apply per-pixel classification. Systematic experimentation was conducted to study the influence of adding augmented and synthetic data on segmentation accuracy. The results show that the use of augmented data increases accuracy significantly. With the addition of synthetic data to the augmented data, there is a marginal improvement in accuracy. However, feature edge accuracy is significantly improved. The preliminary results presented here show the promising use of machine learning techniques for complex, non-destructive inspection methods.

Hybrid Sampling-Based Path Planning for Mobile Manipulators Performing Pick and Place Tasks in Narrow Spaces

A mobile manipulator is capable of traversing a vast area while performing manipulation tasks in confined spaces. However, the high degree of freedom presents a challenge for path planning. In this paper, a hybrid sampling-based path planning method is proposed for mobile manipulators performing pick and place tasks in confined spaces. This method employs a random sampling approach, yet differs from the traditional RRT method. Firstly, a sampling-based configuration generation method for mobile manipulators is proposed, with the objective of generating a valid, collision-free configuration with the end-effector at the desired pose. A path for the end-effector corresponding to the goal configuration is then planned using the RRT method. Secondly, an area-restricted approach that samples in the vicinity of the previous configuration is introduced to generate the next valid configuration. Subsequently, a cost computation rule is devised to identify the optimal subsequent configuration utilizing the trajectory of the end-effector as a guiding principle. Finally, the obtained path is smoothed. Simulations demonstrate that the proposed hybrid sample-based method is an effective solution to the path planning problem for mobile manipulators performing pick and place tasks in narrow spaces.

In silico formulation optimization and particle engineering of pharmaceutical products using a generative artificial intelligence structure synthesis method

Pharmaceutical drug dosage forms are critical for ensuring the effective and safe delivery of active pharmaceutical ingredients to patients. However, traditional formulation development often relies on extensive lab and animal experimentation, which can be time-consuming and costly. This manuscript presents a generative artificial intelligence method that creates digital versions of drug products from images of exemplar products. This approach employs an image generator guided by critical quality attributes, such as particle size and drug loading, to create realistic digital product variations that can be analyzed and optimized digitally. This paper shows how this method was validated through two case studies: one for the determination of the amount of material that will create a percolating network in an oral tablet product and another for the optimization of drug distribution in a long-acting HIV inhibitor implant. The results demonstrate that the generative AI method accurately predicts a percolation threshold of 4.2% weight of microcrystalline cellulose and generates implant formulations with controlled drug loading and particle size distributions. Comparisons with real samples reveal that the synthesized structures exhibit comparable particle size distributions and transport properties in release media.

Multifunctional Organometallic Compounds: An Interesting Luminescent NLO‐Active Alkynylplatinum (II) Complex

AbstractReaction of 4‐methylphenylacetylide with the known chlorido platinum(II) complex bearing the cyclometalated 1,3‐bis(pyridin‐2‐yl)‐4,6‐difluoro‐benzene ligand afforded a novel alkynyl platinum(II) complex, 1, that was fully characterized by elemental analysis, NMR and UV‐visible spectroscopies, by photoluminescence measurements, and by computational modelling. Its structure was determined by X‐ray diffraction studies on a single crystal whereas its second‐order nonlinear optical (NLO) properties were determined in solution by the Electric‐Field Induced Second Harmonic generation method. The novel multifunctional complex 1 is characterized by good luminescence properties, like the related chlorido Pt(II) complex, but by a much higher second‐order NLO response.

A Scene Graph Generation Method for Historical District Street-view Imagery: A Case Study in Beijing, China

A Scene Graph Generation Method for Historical District Street-view Imagery: A Case Study in Beijing, China

Artificial intelligence and machine learning at various stages and scales of process systems engineering

AbstractWe review the utility and application of artificial intelligence (AI) and machine learning (ML) at various process scales in this work, from molecules and reactions to materials to processes, plants, and supply chains; furthermore, we highlight whether the application is at the design or operational stage of the process. In particular, we focus on the distinct representational frameworks employed at the various scales and the physics (equivariance, additivity, injectivity, connectivity, hierarchy, and heterogeneity) they capture. We also review AI techniques and frameworks important in process systems, including hybrid AI modelling, human‐AI collaborations, and generative AI techniques. In hybrid AI models, we emphasize the importance of hyperparameter tuning, especially in the case of physics‐informed regularization. We highlight the importance of studying human‐AI interactions, especially in the context of automation, and distinguish the features of human‐complements‐AI systems from those of AI‐complements‐human systems. Of particular importance in the AI‐complements‐human framework are model explanations, including rule‐based explanation, explanation‐by‐example, explanation‐by‐simplification, visualization, and feature relevance. Generative AI methods are becoming increasingly relevant in process systems engineering, especially in contexts that do not belong to ‘big data’, primarily due to the lack of high quality labelled data. We highlight the use of generative AI methods including generative adversarial networks, graph neural networks, and large language models/transformers along with non‐traditional process data (images, audio, and text).

Mechanical properties and damage plastic model for cement-soil mortar: Laboratory tests and numerical analysis

Cement-soil mortar is a composite material that provides an efficient and cost-effective solution for a wide range of construction applications. This study analyzes the mechanical properties of cement-soil mortar through experimental investigation and explores the application and parameter calibration of the Concrete Damage Plasticity (CDP) model in the finite element simulation of cement-soil mortar. Additionally, an innovative Initial Defect Generation (IDG) method is proposed to enhance the accuracy in simulation of failure mode. The research findings provide a simulation framework that balances simplicity and accuracy for cement-soil mortar. Uniaxial compressive tests are first conducted on cement-soil mortar specimens with water contents ranging from 40 % to 70 %, and cement-to-soil proportions from 30 % to 300 %. Based on the experimental results, the regression relationships correlating unconfined compressive strength (UCS) with elastic modulus, peak strain and stress-strain curves are established. Then, the simplified equations for calculating CDP model parameters from the UCS of cement-soil mortar are further proposed following damage mechanics. A parameter table summarizing the calculation methods for all relevant CDP parameters is provided to streamline the model calibration process. Simulations incorporating the simplified calibration method and IDG method successfully replicated the stress-strain responses and failure modes observed in uniaxial compressive strength tests of cement-soil specimens with varied strength. The results demonstrate the reliability and broad applicability of this simulation framework in predicting the mechanical performance of cement-soil mortar.

Pred-AHCP: Robust Feature Selection-Enabled Sequence-Specific Prediction of Anti-Hepatitis C Peptides via Machine Learning.

Every year, an estimated 1.5 million people worldwide contract Hepatitis C, a significant contributor to liver problems. Although many studies have explored machine learning's potential to predict antiviral peptides, very few have addressed the problem of predicting peptides against specific viruses such as Hepatitis C. In this study, we demonstrate the application and fine-tuning of machine learning (ML) algorithms to predict peptides that are effective against Hepatitis C virus (HCV). We developed a fine-tuned and explainable ML model that harnesses the amino acid sequence of a peptide to predict its anti-hepatitis C potential. Specifically, features were computed based on sequence and physicochemical properties. The feature selection was performed using a combined strategy of mutual information and variance inflation factor. This facilitated the removal of redundant and multicollinear features, enhancing the model's generalizability in predicting anti-hepatitis C peptides (AHCPs). The model using the random forest algorithm produced the best performance with an accuracy of about 92%. The feature analysis highlights that the distributions of hydrophobicity, polarizability, coil-forming residues, frequency of glycine residues and the existence of dipeptide motifs VL, LV, and CC emerged as the key predictors for identifying AHCPs targeting different components of HCV. The developed model can be accessed through the Pred-AHCP web server, provided at http://tinyurl.com/web-Pred-AHCP. This resource facilitates the prediction and re-engineering of AHCPs for designing peptide-based therapeutics while also proposing an exploration of similar strategies for designing peptide inhibitors effective against other viruses. The developed ML model can also be used for validating peptide sequences generated using generative artificial intelligence methods for further optimization.