Explore the vast range of titles available.

Selecting the optimal parameters for wet steam accumulator of steam-powered catapult is an important task, due to launching safety. There is no literature on the topic of the parameters optimization for wet steam accumulator of steam-powered catapult. The genetic algorithm (GA) was used to determine the optimal wet steam accumulator in this article. The sink-off-the-bow (SOB), angle of attack and rate of climb were used to create the objective function. The multi-objective optimization can be converted to single-objective optimization, which is subject to angle of attack and rate of climb. Moreover, the simulation model of the steam catapult system was built by creating a thermodynamics model of steam-powered catapult, a mathematical model of traction release device, a statics model of tensioning, a statics model of full takeoff power, a mathematical model of catapult force build-up with holdback, a model of release, a dynamics model of power stroke, a dynamics model of free deck run and a dynamics model of fly away. Finally, the optimal combination of the wet steam accumulator was obtained via numerical simulation. The GA method can effectively find the optimal parameters of wet steam accumulator, and its optimized parameters can increase the safety of catapult launch process.

Magnetorheological fluids belong to a class of smart materials, whose rheological characteristics such as yield stress, viscosity, etc. change in the presence of applied magnetic field. In this paper, multi-response optimization of MR fluid constituents is obtained. For this, 18 samples of MR fluids are prepared using L-18 Orthogonal Array. These samples are experimentally tested on the in-house developed and fabricated electromagnet setup. The setup has been validated using a reference fluid. Yield stress of MR fluid mainly depends on the volume fraction of the iron particles and type of carrier fluid used in its preparation. An optimal combination of these input parameters with mineral oil as a carrier fluid and Fe 300 mesh (32% by volume) as an iron particle, oleic acid (0.5% by volume), and tetra-methyl-ammonium-hydroxide (0.7% by volume) has given the largest numerical values of on-state yield stress and viscosity of the MR fluid sample as 48.197 kPa and 573.0944 kPa-s, respectively, within the range of the input parameters. The yield stress of optimized MR fluid is higher than Lord MRF-122-EG fluid. Furthermore, a confirmation test on the optimized MR fluid sample has been carried out and the response parameters thus obtained are found to be matching quite well (with error less than 1%) with the statistically obtained values. This high value of the yield stress and viscosity can be used more effectively in the formulation and designing of MR devices requiring larger variation in the damping.

This paper focuses on parameter optimization for the actually manufactured test vehicle. This method achieves high-precision, rapid computation of vehicle dynamic performance while fully preserving the strongly coupled nonlinear dynamic properties of the system. Firstly, by employing twin modeling technology, the model accurately reflects the physical dynamic characteristics of the actual vehicle, enabling us to determine how much improvement the optimized vehicle dynamic response will exhibit compared to the current state. Next, a mathematical model for multi-parameter, multi-objective collaborative optimization is constructed using big data search, and key parameters significantly influencing vehicle dynamics are identified through Sobol sensitivity analysis for dynamic optimization. Finally, an improved multi-start parallel simulated annealing algorithm is proposed to enhance the computational efficiency and reliability of the optimization results. The results demonstrate significant improvement in the dynamic performance of the experimental vehicle, validating the effectiveness of the proposed method. This approach overcomes the limitations of traditional linearization treatments, providing a new perspective for dynamic optimization of complex coupled systems and demonstrating significant engineering application value in the field of rail transportation.

REINVENT 4 is a modern open-source generative AI framework for the design of small molecules. The software utilizes recurrent neural networks and transformer architectures to drive molecule generation. These generators are seamlessly embedded within the general machine learning optimization algorithms, transfer learning, reinforcement learning and curriculum learning. REINVENT 4 enables and facilitates de novo design, R-group replacement, library design, linker design, scaffold hopping and molecule optimization. This contribution gives an overview of the software and describes its design. Algorithms and their applications are discussed in detail. REINVENT 4 is a command line tool which reads a user configuration in either TOML or JSON format. The aim of this release is to provide reference implementations for some of the most common algorithms in AI based molecule generation. An additional goal with the release is to create a framework for education and future innovation in AI based molecular design. The software is available from https://github.com/MolecularAI/REINVENT4 and released under the permissive Apache 2.0 license. Scientific contribution . The software provides an open–source reference implementation for generative molecular design where the software is also being used in production to support in–house drug discovery projects. The publication of the most common machine learning algorithms in one code and full documentation thereof will increase transparency of AI and foster innovation, collaboration and education.

This article proposes a ‘dynamic’ artificial bee colony (D-ABC) algorithm for solving optimizing problems. It overcomes the poor performance of artificial bee colony (ABC) algorithm, when applied to multi-parameters optimization. A dynamic ‘activity’ factor is introduced to D-ABC algorithm to speed up convergence and improve the quality of solution. This D-ABC algorithm is employed for multi-parameters optimization of support vector machine (SVM)-based soft-margin classifier. Parameter optimization is significant to improve classification performance of SVM-based classifier. Classification accuracy is defined as the objection function, and the many parameters, including ‘kernel parameter’, ‘cost factor’, etc., form a solution vector to be optimized. Experiments demonstrate that D-ABC algorithm has better performance than traditional methods for this optimizing problem, and better parameters of SVM are obtained which lead to higher classification accuracy.

Computer-aided drug design has advanced rapidly in recent years, and multiple instances of in silico designed molecules advancing to the clinic have demonstrated the contribution of this field to medicine. Properly designed and implemented platforms can drastically reduce drug development timelines and costs. While such efforts were initially focused primarily on target affinity/activity, it is now appreciated that other parameters are equally important in the successful development of a drug and its progression to the clinic, including pharmacokinetic properties as well as absorption, distribution, metabolic, excretion and toxicological (ADMET) properties. In the last decade, several programs have been developed that incorporate these properties into the drug design and optimization process and to varying degrees, allowing for multi-parameter optimization. Here, we introduce the Artificial Intelligence-driven Drug Design (AIDD) platform, which automates the drug design process by integrating high-throughput physiologically-based pharmacokinetic simulations (powered by GastroPlus) and ADMET predictions (powered by ADMET Predictor) with an advanced evolutionary algorithm that is quite different than current generative models. AIDD uses these and other estimates in iteratively performing multi-objective optimizations to produce novel molecules that are active and lead-like. Here we describe the AIDD workflow and details of the methodologies involved therein. We use a dataset of triazolopyrimidine inhibitors of the dihydroorotate dehydrogenase from Plasmodium falciparum to illustrate how AIDD generates novel sets of molecules.

The integration of compliant mechanism and parallel mechanism provide an effective solution for medical micromanipulation, especially for the scenarios where high precision and high dexterity are required. The development of spatial compliant parallel mechanism (CPM) takes advantages of the features for both compliant mechanism and parallel mechanism to generate greater comprehensive performances. In this research, a novel three degrees-of-freedom CPM is designed and analyzed. Since performance characterization is one important factor that greatly affects the application potential, the performance indices including stiffness, dexterity, manipulability and workspace are mapped respectively. The finite element analysis is conducted to prove the feasibility of the proposed design. The multi-parameters improvement is implemented to demonstrate a way how to optimize the performance of the compliant parallel mechanism based on a generic method.

Although a multi-stage hydraulically fractured horizontal well in a shale reservoir initially produces gas at a high production rate, this production rate declines rapidly within a short period and the cumulative gas production is only a small fraction (20–30%) of the estimated gas in place. In order to maximize the gas recovery rate (GRR), this study proposes a multi-parameter optimization model for a typical multi-stage hydraulically fractured shale gas horizontal well. This is achieved by combining the response surface methodology (RSM) for the optimization of objective function with a fully coupled hydro-mechanical FEC-DPM for forward computation. The objective function is constructed with seven uncertain parameters ranging from matrix to hydraulic fracture. These parameters are optimized to achieve the GRR maximization in short-term and long-term gas productions, respectively. The key influential factors among these parameters are identified. It is established that the gas recovery rate can be enhanced by 10% in the short-term production and by 60% in the long-term production if the optimized parameters are used. Therefore, combining hydraulic fracturing with an auxiliary method to enhance the gas diffusion in matrix may be an effective alternative method for the economic development of shale gas.

Axial flux motors have attracted significant attention in recent years due to their advantages such as shorter axial length and high torque density. However, the optimization of axial flux motors is an extremely time-consuming process. To reduce the computational time required for motor optimization, this study employed a magnetic charge model to establish a coreless axial flux motor model and analyzed the advantages of this approach. This method is applicable to coreless axial flux motor optimizations with surface-mounted rotors and concentrated windings. Parameter optimization was subsequently performed based on the theoretical model. In terms of seeking optimal solutions, the torque obtained through the magnetic charge method (MCM) reached 99.67% of the finite element method (FEM) results. Finally, a prototype was fabricated, and a test platform was constructed based on the optimization results. The experimental torque showed a 4% deviation from simulations, validating the accuracy of the optimization.

In drug discovery, identifying molecules with desired pharmacological properties remains challenging, as conventional methods often rely on exhaustive trial-and-error and limited exploration of chemical space. Here, we present STELLA, a metaheuristics-based generative molecular design framework that combines an evolutionary algorithm for fragment-based chemical space exploration with a clustering-based conformational space annealing method for efficient multi-parameter optimization. Additionally, it leverages deep learning models for accurate prediction of pharmacological properties. Our case study, which focuses on docking score and quantitative estimate of drug-likeness as primary objectives, demonstrates that STELLA generates 217% more hit candidates with 161% more unique scaffolds and achieves more advanced Pareto fronts compared to REINVENT 4. In performance evaluations optimizing 16 properties simultaneously for MolFinder, REINVENT 4, and STELLA, STELLA consistently outperforms the control methods by achieving better average objective scores and exploring a broader region of chemical space. The results highlight STELLA’s superior performance in both efficient exploration of chemical space and multi-parameter optimization, indicating that STELLA is a powerful tool for de novo molecular design.