The Rust programming language presents a steep learning curve and significant coding challenges, making the automation of issue resolution essential for its broader adoption. Recently, LLM-powered code agents have shown remarkable success in resolving complex software engineering tasks, yet their application to Rust has been limited by the absence of a large-scale, repository-level benchmark. To bridge this gap, we introduce Rust-SWE-bench, a benchmark comprising 500 real-world, repository-level software engineering tasks from 34 diverse and popular Rust repositories. We then perform a comprehensive study on Rust-SWE-bench with four representative agents and four state-of-the-art LLMs to establish a foundational understanding of their capabilities and limitations in the Rust ecosystem. Our extensive study reveals that while ReAct-style agents are promising, i.e., resolving up to 21.2% of issues, they are limited by two primary challenges: comprehending repository-wide code structure and complying with Rust's strict type and trait semantics. We also find that issue reproduction is rather critical for task resolution. Inspired by these findings, we propose RUSTFORGER, a novel agentic approach that integrates an automated test environment setup with a Rust metaprogramming-driven dynamic tracing strategy to facilitate reliable issue reproduction and dynamic analysis. The evaluation shows that RUSTFORGER using Claude-Sonnet-3.7 significantly outperforms all baselines, resolving 28.6% of tasks on Rust-SWE-bench, i.e., a 34.9% improvement over the strongest baseline, and, in aggregate, uniquely solves 46 tasks that no other agent could solve across all adopted advanced LLMs.
Stock price prediction is a critical task in the financial sector due to its profound implications for traders and investors. This paper presents a comparative analysis of machine learning models applied to stock price prediction using historical data from the Nasdaq stock index spanning the years 2015 to 2023. The study introduces an Extreme Gradient Boosting Regression (XGBR) model optimized with three distinct metaheuristic algorithms: Battle Royal Optimization (BRO), Moth Flame Optimization (MFO), and Artificial Bee Colony (ABC). These optimization techniques aim to enhance the model's predictive performance by improving parameter tuning and model generalization. Among the optimized models, the ABC-XGBR demonstrated superior performance due to its strong balance between exploration and exploitation and its effective search capability in high-dimensional feature spaces. The experimental results show significant improvements over the baseline XGBR model, with R² values of 0.9721 for BRO-XGBR, 0.9885 for MFO-XGBR, and 0.9936 for ABC-XGBR. These outcomes underscore the effectiveness of combining machine learning with nature-inspired optimization algorithms to produce more accurate stock price forecasts. This research contributes valuable insights into the practical application of hybrid models for financial forecasting, emphasizing their utility in enhancing predictive accuracy. It also offers decision-makers—such as investors, analysts, and financial institutions—a robust framework for incorporating data-driven strategies into risk assessment and portfolio management. Future work may explore additional datasets, real-time prediction capabilities, and further refinement of optimization algorithms to extend the applicability of these methods to broader financial contexts.
Yellow phosphorus is an important industrial raw material used in pharmaceuticals, food, pesticides, the military, and the chemical industry, significantly impacting the economy. The drying effect is the most important aspect that affects the material properties. Traditional drying methods are often inefficient and energy-intensive, while microwave drying offers unique heating advantages, making it a promising alternative. This paper explores how initial mass, moisture content, and microwave power influence the drying process using microwave technology. The study examined how varying initial mass, moisture content, and microwave power affect the drying characteristics of materials. It calculated the microwave drying efficiency (η) and unit energy consumption (Qs) under different conditions. Results indicated that increasing the initial mass and moisture content enhances the microwave’s drying efficiency and reduces the unit energy consumption. However, as microwave power increased, the microwave drying efficiency gradually decreased, while the unit energy consumption gradually increased. When the initial mass increased from 20 g to 50 g, drying efficiency rose from 6.58% to 13.12%, while unit energy consumption decreased from 34.33 to 17.20 MJ·kg−1. Similarly, increasing initial moisture content increased from 20% to 40% improved efficiency from 12.36% to 19.15%, unit energy consumption decreased from 14.28 to 11.70 MJ·kg−1. The results showed that the maximum microwave drying efficiency (η) reached 21.51% and the minimum unit energy consumption (Qs) was 10.99 MJ·kg−1 at a mass of 50 g, a moisture content of 40%, and a microwave power of 360 W. Furthermore, this efficiency and energy consumption were consistent when the initial moisture content ranged from 20% to 40%. Four thin-layer drying kinetic models were used to fit the relevant experimental data, revealing that the Modified Page model was the most suitable for describing the microwave drying process of the material. Surface diffusion coefficients of water molecules were calculated under different conditions, and activation energy was derived from these coefficients. The maximum diffusion coefficient was 1.29 × 10−10 m2·s−1 for an initial mass of 40 g, 1.53 × 10−10 m2·s−1 for an initial moisture content of 30%, and 1.64 × 10−10 m2·s−1 for a microwave power of 900 W. The activation energy was calculated to be 5.95 W·g−1. Using COMSOL, simulations of the electromagnetic and temperature fields under varying microwave power conditions were conducted. The electric field intensity increased with higher microwave power, rising from 8.13 × 104 V·m−1 at 360 W to 1.15 × 105 V·m−1 at 720 W. In the experimental phase, increased microwave power reduces the time required for drying, and the temperature field distribution aligns with experimental results, effectively describing the drying process under microwave influence. This provides a theoretical basis and technical support for the efficient drying of yellow phosphorus-like materials.
Owing to wave obstruction, the speed of the horizontal wind field decreases as it approaches the sea surface. Consequently, the wind field that increases with height is referred to as a gradient wind field. On the sea surface, albatrosses exploit this gradient wind field to glide efficiently, enabling them to fly thousands of kilometers using wind energy. As an emerging class of bionic flying robots, flapping-wing robots mimic birds’ flying methods. This wind energy utilization behavior holds significant potential for flapping-wing robots, offering a promising solution to address their current endurance limitations. To utilize wind energy effectively, albatrosses rely on their exceptional gliding characteristics. This study replicates the aerodynamic features of albatrosses, including their high aspect ratio and superior lift-to-drag ratio, to enhance the performance of the independently developed flapping-wing robot, USTB-Hawk. Given that the improved flapping-wing robot must transition between flapping flight mode and gliding flight mode, this paper also introduces a mode-switching mechanism. This mechanism, based on a ratchet stop system and a flapping phase detector, ensures stability in the gliding posture during flight experiments. In addition to aerodynamic characteristics, albatrosses primarily exploit wind energy by continuously ascending and descending within the gradient wind field, achieved through efficient planning of their gliding trajectory. To simulate the gliding trajectory of the improved flapping-wing robot, it is essential to determine its aerodynamic characteristics in gliding posture, including lift and drag data across different angles of attack. This study conducts fluid mechanics simulations on the improved flapping-wing robot. In the simulation, the robot’s design is simplified, with some complex structures removed, to reduce computational costs and model complexity. The results indicate that the gliding posture of the flapping-wing robot avoids a stall state within an angle of attack range of −10°– 20°. In addition, the lift generated by the robot is sufficient to counteract gravity at angles of attack between 2.86° and 20°. With the enhanced aerodynamic characteristics identified, this study further investigates the gliding trajectory of the flapping-wing robot by integrating the gradient wind field model with the kinematic model of the gliding posture for various trajectory angles. Considering that the trajectories of different entry angles vary under a constant wind field, this study conducts a detailed analysis of the gliding trajectories corresponding to different initial heading angles for the same track angle. Trajectories with track angles of −30°, 0°, 30°, and 60° are selected for flight experiments in a real wind field. The experimental results reveal that the energy consumption of a gliding flight is significantly lower than that of a flapping flight over the same distance. These findings demonstrate that the flapping-wing robot can effectively utilize wind energy and enhance its endurance by strategically planning its gliding trajectory within the gradient wind field.
Wire arc additive manufacturing (WAAM) is a commonly used additive manufacturing technique for producing magnesium alloy components with complex structures. However, the effect of friction stir processing (FSP) on WAAM magnesium alloy samples has not been properly examined. In this study, a bulk sample of the AZ91 magnesium alloy was prepared via multi-layer and multi-pass WAAM technology, and its microstructure was modified by FSP to improve its mechanical properties. The effect of FSP on the microstructure and mechanical properties of the WAAM sample was investigated. The obtained results revealed that the lack of fusion defects in the lapped area between the adjacent deposited passes were significantly eliminated after FSP, while recrystallization occurred in the stirred zone (SZ) and thermos-mechanically affected zone (TMAZ). Compared with that of the unprocessed sample, the grain size of the sample after FSP became more uniform, and the amount of the β-Mg17Al12 phase decreased significantly. In addition, a short lamellar β-Mg17Al12 phase precipitated within some grains in the TMAZ. The average ultimate tensile strength and elongation of the sample after FSP along the pass overlap direction reached 237.9 MPa and 21.7 %, which were more than 80 % higher than those before FSP (127.1 MPa and 11.6 %, respectively) owing to the more uniform microstructure and defect elimination. The anisotropy of the tensile properties of the WAAM sample was significantly reduced by the introduction of FSP.
Thiago Barradas, Aline Paes, Vânia de Oliveira Neves
The effective execution of tests for REST APIs remains a considerable challenge for development teams, driven by the inherent complexity of distributed systems, the multitude of possible scenarios, and the limited time available for test design. Exhaustive testing of all input combinations is impractical, often resulting in undetected failures, high manual effort, and limited test coverage. To address these issues, we introduce RestTSLLM, an approach that uses Test Specification Language (TSL) in conjunction with Large Language Models (LLMs) to automate the generation of test cases for REST APIs. The approach targets two core challenges: the creation of test scenarios and the definition of appropriate input data. The proposed solution integrates prompt engineering techniques with an automated pipeline to evaluate various LLMs on their ability to generate tests from OpenAPI specifications. The evaluation focused on metrics such as success rate, test coverage, and mutation score, enabling a systematic comparison of model performance. The results indicate that the best-performing LLMs - Claude 3.5 Sonnet (Anthropic), Deepseek R1 (Deepseek), Qwen 2.5 32b (Alibaba), and Sabia 3 (Maritaca) - consistently produced robust and contextually coherent REST API tests. Among them, Claude 3.5 Sonnet outperformed all other models across every metric, emerging in this study as the most suitable model for this task. These findings highlight the potential of LLMs to automate the generation of tests based on API specifications.
Successfully engineering interactive industrial DTs is a complex task, especially when implementing services beyond passive monitoring. We present here an experience report on engineering a safety-critical digital twin (DT) for beer fermentation monitoring, which provides continual sampling and reduces manual sampling time by 91%. We document our systematic methodology and practical solutions for implementing bidirectional DTs in industrial environments. This includes our three-phase engineering approach that transforms a passive monitoring system into an interactive Type 2 DT with real-time control capabilities for pressurized systems operating at seven bar. We contribute details of multi-layered safety protocols, hardware-software integration strategies across Arduino controllers and Unity visualization, and real-time synchronization solutions. We document specific engineering challenges and solutions spanning interdisciplinary integration, demonstrating how our use of the constellation reporting framework facilitates cross-domain collaboration. Key findings include the critical importance of safety-first design, simulation-driven development, and progressive implementation strategies. Our work thus provides actionable guidance for practitioners developing DTs requiring bidirectional control in safety-critical applications.
Model-driven engineering (MDE) is believed to have a significant impact in software quality. However, researchers and practitioners may have a hard time locating consolidated evidence on this impact, as the available information is scattered in several different publications. Our goal is to aggregate consolidated findings on quality in MDE, facilitating the work of researchers and practitioners in learning about the coverage and main findings of existing work as well as identifying relatively unexplored niches of research that need further attention. We performed a tertiary study on quality in MDE, in order to gain a better understanding of its most prominent findings and existing challenges, as reported in the literature. We identified 22 systematic literature reviews and mapping studies and the most relevant quality attributes addressed by each of those studies, in the context of MDE. Maintainability is clearly the most often studied and reported quality attribute impacted by MDE. Eighty out of 83 research questions in the selected secondary studies have a structure that is more often associated with mapping existing research than with answering more concrete research questions (e.g., comparing two alternative MDE approaches with respect to their impact on a specific quality attribute). We briefly outline the main contributions of each of the selected literature reviews. In the collected studies, we observed a broad coverage of software product quality, although frequently accompanied by notes on how much more empirical research is needed to further validate existing claims. Relatively, little attention seems to be devoted to the impact of MDE on the quality in use of products developed using MDE.
Vladimir F. Borisenko, Vladimir A. Sidorov, Andrey E. Sushko
et al.
The technical condition of ball mills, employed in the fine grinding of minerals, ores, coal, cement clinker, and other materials, is dictated by both the operational load and the actual physical state of the equipment. Vibration metrics serve as the most versatile diagnostic parameter for developing an informational profile of equipment in active use. The distinct operational environments of high-powered ball mills with frequency-controlled drive systems – one with an induction motor (FC-IM) and the other with a synchronous motor (FC-SM) – necessitate the development of universal approaches to assessing vibration loading that consider each mill's unique design features and operational modes. This study presents the first analysis of the key interrelated technical characteristics of industrial ball mills, including drum volume, diameter, rotational speed, ball load, total weight, and drive power, enabling a more substantiated approach to selecting technical parameters and operational modes. The installation of a permanent vibration control system on ball mills used for grinding mineral raw materials required the individual determination of technical condition category thresholds for the motor, gear-shaft, and drum. The category thresholds were determined individually for each shaft using statistical classification, under the assumption that coupled components are in a state influenced by the energy potential of damage during staged progression. Standard 'reference' ratios of vibration values across three mutually perpendicular directions were established. Characteristic patterns and sequences of damage progression were identified based on the direct spectra of vibration velocity and acceleration. During the analysis of vibration signal time series, a beat frequency mode was detected, indicating potential damage development within gear elements. Effective informational support for the operational condition of ball mills is achieved through the analysis of overall vibration levels, direct trends in vibration velocity and acceleration, time series of the vibration signal, and both long-term and short-term trend analyses. Vibration velocity trends provide insights into technical condition by assessing operational stability, startup frequency, and maintenance intervals.
Large-scale code datasets have acquired an increasingly central role in software engineering (SE) research. This is the result of (i) the success of the mining software repositories (MSR) community, that pushed the standards of empirical studies in SE; and (ii) the recent advent of deep learning (DL) in software engineering, with models trained and tested on large source code datasets. While there exist some ready-to-use datasets in the literature, researchers often need to build and pre-process their own dataset to meet specific requirements of the study/technique they are working on. This implies a substantial cost in terms of time and computational resources. In this work we present the SEART Data Hub, a web application that allows to easily build and pre-process large-scale datasets featuring code mined from public GitHub repositories. Through a simple web interface, researchers can specify a set of mining criteria (e.g., only collect code from repositories having more than 100 contributors and more than 1,000 commits) as well as specific pre-processing steps they want to perform (e.g., remove duplicates, test code, instances with syntax errors). After submitting the request, the user will receive an email with a download link for the required dataset within a few hours. A video showcasing the SEART Data Hub is available at https://youtu.be/lCgQaA7CYWA.
In this work, we describe the design and architecture of the open-source Quantum Engine Compiler (qe-compiler) currently used in production for IBM Quantum systems. The qe-compiler is built using LLVM's Multi-Level Intermediate Representation (MLIR) framework and includes definitions for several dialects to represent parameterized quantum computation at multiple levels of abstraction. The compiler also provides Python bindings and a diagnostic system. An open-source LALR lexer and parser built using Bison and Flex generates an Abstract Syntax Tree that is translated to a high-level MLIR dialect. An extensible hierarchical target system for modeling the heterogeneous nature of control systems at compilation time is included. Target-based and generic compilation passes are added using a pipeline interface to translate the input down to low-level intermediate representations (including LLVM IR) and can take advantage of LLVM backends and tooling to generate machine executable binaries. The qe-compiler is built to be extensible, maintainable, performant, and scalable to support the future of quantum computing.
High silicon steel with 6.5% silicon content is the best because of its excellent magnetic properties, such as high saturation magnetization, high resistivity, low iron loss and near zero magnetostriction. High silicon steel can greatly save energy, and reduce the weight and size of electrical appliances. This has a very important application prospect for energy and aerospace industry. The high brittleness of high silicon steel makes its production and processing very difficult. For more than 30 years, many steel companies and research institutions around the world have adopted various technical means to study the industrialization of high silicon steel, but they have not been successful . JFE-NKK steel company in Japan has realized the small batch production of high silicon steel by using SiCl4-CVD technology. However, due to the complex process, corrosion and pollution, high cost, its production scale is greatly limited. So far, large-scale production of high silicon steel is still a major challenge in the world. This paper will introduce the experimental results of successfully preparing high silicon steel by Double Glow Plasma Surface Metallurgy Technology. The process is simple and easy without any corrosion or pollution, which may provide a new way for the world to achieve large-scale production of high silicon steel. The large-scale production and wide application of high silicon steel is likely to change the pattern of the world's energy and electric power industry, save a lot of energy for mankind, and create huge economic benefits.
With the rapid developments of marine resource exploitation, mounts of marine engineering equipment are settled on the ocean. When it is not possible to move the damaged equipment into a dry dock, welding operations must be performed in underwater environments. The underwater laser welding/cladding technique is a promising and advanced technique which could be widely applied to the maintenance of the damaged equipment. The present review paper aims to present a critical analysis and engineering overview of the underwater laser welding/cladding technique. First, we elaborated recent advances and key issues of drainage nozzles all over the world. Next, we presented the underwater laser processing and microstructural-mechanical behavior of repaired marine materials. Then, the newly developed powder-feeding based and wire-feeding based underwater laser direct metal deposition techniques were reviewed. The differences between the convection, conduction, and the metallurgical kinetics in the melt pools during underwater laser direct metal deposition and in-air laser direct metal deposition were illustrated. After that, several challenges that need to be overcame to achieve the full potential of the underwater laser welding/cladding technique are proposed. Finally, suggestions for future directions to aid the development of underwater laser welding/cladding technology and underwater metallurgical theory are provided. The present review will not only enrich the knowledge in the underwater repair technology, but also provide important guidance for the potential applications of the technology on the marine engineering.
Understanding the synthesis process of Portland cement's various mineral phases and the phase interactions during hydration is highly critical for the design and optimization of cement-based materials. In this work, by using an optimized solid-state reaction method, three pure clinker phases (alite, belite and ferrite) and two polyphase phases (alite-belite and alite-ferrite-belite) of Portland cement were synthesized. The optimized method is featured with high-quality product yield, short burning time without regrinding or reburning. The factors that influence synthesis results, such as crystallinity/purity of the synthesized pure phases, were discussed based on X-ray powder diffraction (XRD) tests along with Rietveld refinement analysis with an external standard. The hydration behavior of the synthesized phases in monophase, polyphase as well as mixture states was investigated via isothermal calorimetry along with thermogravimetry and XRD. The results indicate that alite and belite have relatively little interactions during hydration, while the hydration rate of ferrite can be strongly inhibited in the presence of alite both in mixture and polyphase states.
Rini Setiati, Fahrurrozi Akbar, Gabriel Prasucipto Karisma
et al.
The Neutron Computed Tomography (NCT) technique was adopted to study oil content within sandstone. In recent times, oil industries have turned to the Enhanced Oil Recovery (EOR) technique that uses chemicals/surfactants to reduce interfacial tension to maximize oil production from mature oil fields. The Berea core played a crucial role in investigating surfactant effectiveness for liberating trapped oil. The presented research is based on laboratory experiments carried out at the National Research and Innovation Agency and at the Trisakti University, Indonesia. The research proceeded in three steps. The first step was observation on the Berea core that is saturated with a low salinity brine of 40,000 ppm and 60,000 ppm and assessment of the total porosity of the sandstone. The second step included injecting oil into the sample and examining the water distribution. The final step was to inject sugarcane bagasse surfactant with 2% surfactant concentration to extract oil from the sandstone, then it was subjected to NCT analysis. The observations show that the pores are evenly distributed in the middle, the oil content accumulates at the edge, almost no oil remains in the middle of the Berea core after water injection, and the remaining oil returns to the main flow after surfactant injection. The results show that oil movement in the Berea core with 40,000 ppm salinity is better than oil movement with 60,000 ppm. The salinity of the brine can affect the mobility of oil droplets in the Berea core, and surfactant injection processes have better performance at lower salinity and are susceptible to higher salinity.
The hydrogen induced cracking (HIC) resistance of four (4) L80 casing steels, with different nominal compositions and processing conditions, was determined using the NACE TM0284 HIC test. Microstructural and inclusion characterization of each steel was undertaken using optical microscopy (OM), scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDS). The morphology (equivalent diameter, aspect ratio and major length) of inclusions over identical analysis regions were measured for each steel. In total, >15,000 inclusions were characterized. Non-linear models correlating inclusion morphology with the measured values of CLR (crack length ratio) and CTR (crack thickness ratio) were developed. The models show that both CLR and CTR are complex functions of the diameter (equivalent) of oxide-based inclusions and the major length of the manganese sulfide (MnS) stringers. The models show that oxides <10 μm in diameter were found to have negligible effect on HIC resistance (CLR) while oxides >20 μm exhibited a significant and escalating negative impact on HIC resistance comparable to long (>50 μm) MnS stringers.
The emerging Flexible Ring Mode(FRM) laser technology has shown great potential for welding copper and its alloys with high reflectivity and has been widely used. The design of the welding trajectory significantly affects the porosity and performance of welded joints. This paper first selects the ellipse and its parameters from the three trajectories by numerical simulation method, then studies the appearance forming and counts the weld porosity through the longitudinal and cross-section developing quality, and finally uses a high-speed camera to monitor the copper Behavior of molten pool and spot oscillation process at the front of the plate. It is found that the trajectory of circular motion has a more stable heating process and has a preheating function for the center of the weld. Compared with linear welding, orbital welding achieves penetration of weld seam and forms better weld seam. Laser confocal tests show that S4(a1b0.5) has the most negligible variation in weld appearance and the best formability. S3(a0.5b0.5) has the lowest porosity, which also explains the mechanism of laser welding trajectory stirring the molten pool and pooling of pores under hydrostatic pressure. The tensile test shows that: S3 has the highest tensile strength (215.35 MPa), 89.79% of the base material. But its strain is only 50% of that of the S4 sample with almost the same tensile strength (215.03 MPa). Therefore, this paper proposes that an elliptical welding trajectory can be adopted to obtain welded joints with better tensile strength and plasticity, with the long axis along the welding direction and the short axis perpendicular to the welding direction. This study provides theoretical guidance for the optimal oscillation mode selection for tunable ring mode laser welding of highly reflective materials.
We describe a method, based on Jennifer Nado's proposal for classification procedures as targets of conceptual engineering, that implements such procedures by prompting a large language model. We apply this method, using data from the Wikidata knowledge graph, to evaluate stipulative definitions related to two paradigmatic conceptual engineering projects: the International Astronomical Union's redefinition of PLANET and Haslanger's ameliorative analysis of WOMAN. Our results show that classification procedures built using our approach can exhibit good classification performance and, through the generation of rationales for their classifications, can contribute to the identification of issues in either the definitions or the data against which they are being evaluated. We consider objections to this method, and discuss implications of this work for three aspects of theory and practice of conceptual engineering: the definition of its targets, empirical methods for their investigation, and their practical roles. The data and code used for our experiments, together with the experimental results, are available in a Github repository.