Determination of gadolinium by PGNAA with a D-T neutron generator and optimization algorithm

Over 2300 years ago, the discovery of tourmaline led to the understanding of pyroelectric properties, which opened new doors to various applications of pyroelectric crystal, such as neutron and x-ray generation, energy harvesting, mass spectrometry, high-voltage sources, and more. In the last two decades, researchers have carried out extensive research and development to select components and materials and innovate the design and construction of the pyroelectric neutron generator (PNG). This manuscript investigates the process and history of the PNG’s development. It explains the physics governing pyroelectric crystals and the method of producing neutrons in a comprehensive and straightforward manner. Although PNGs have a lower yield and shorter lifetime compared to other neutron generators, they are still significant for research purposes due to their lack of need for an external high-voltage power supply, lower cost, smaller size, and safety. The main objective of this manuscript is to bring more attention to the research and development of PNGs. In recent years, new methods have been introduced that reduce the amount of neutron flux required for various applications. This has raised hope for the progress of commercial and industrial use of PNGs in the near future. The manuscript mentions some research cases that represent the future perspective of PNG development. Furthermore, the challenges faced by PNGs can be handled more efficiently with the utilization of generative learning algorithms and improvements in the components/mechanisms used for PNG design.

An optimization algorithm has been developed for the first time for application to International Criticality Safety Benchmark Evaluation Project (ICSBEP) subcritical neutron multiplication inference benchmark experiments. The optimization algorithm is a genetic algorithm for nuclear data evaluation adjustments, specifically applied to subcritical benchmark measurements. The algorithm has been tested and yields improvement in (C-E)/E values of subcritical benchmark observables of interest. In this work, the genetic algorithm is applied to improvement of fission neutron multiplicity distribution parameters using several subcritical neutron multiplication inference benchmarks; specifically a series of reflected 4.5 kg α-phase spherical plutonium benchmarks. The algorithm results suggest changing the mean (ν¯) and standard deviation (σ) of the number of neutrons emitted by 240Pu in spontaneous fission from 2.1510 to 2.1460 and from 1.1510 to 1.1395, respectively. In addition, the standard deviation of the number of neutrons emitted by 239Pu in induced fission should remain unchanged at 1.1400. These changes are all within 1 standard deviation.

This paper presents a new algorithm, called steady-state and generational evolutionary algorithm, which combines the fast and steadily tracking ability of steady-state algorithms and good diversity preservation of generational algorithms, for handling dynamic multiobjective optimization. Unlike most existing approaches for dynamic multiobjective optimization, the proposed algorithm detects environmental changes and responds to them in a steady-state manner. If a change is detected, it reuses a portion of outdated solutions with good distribution and relocates a number of solutions close to the new Pareto front based on the information collected from previous environments and the new environment. This way, the algorithm can quickly adapt to changing environments and thus is expected to provide a good tracking ability. The proposed algorithm is tested on a number of bi- and three-objective benchmark problems with different dynamic characteristics and difficulties. Experimental results show that the proposed algorithm is very competitive for dynamic multiobjective optimization in comparison with state-of-the-art methods.

Generative design of main landing gear for a remote-controlled aircraft

Fused Filament Fabrication is currently among the most commonly used Additive Manufacturing technologies but is highly reliant on temporary support structures during production. Implementing a generative optimization algorithm for support-free Fused Filament Fabrication could streamline the manufacturing process in terms of labor, time- and material use. Despite the current relevancy of Additive Manufacturing, there is a lack of research on structural optimization customized for support-free Fused Filament Fabrication. This research applies a generative optimization algorithm consisting of a multi-objective evolutionary algorithm and a local search algorithm to generate and optimize rigid-jointed 3D truss structures. The results show the capacity to generate and optimize support-free rigid-jointed truss structures with promising solutions to a multi-objective optimization task. This paper suggests that support-free structural optimization algorithms can impact how we design robotic bodies and parts in the future.

Clustering is the task that has been used in numerous applications including digital image analysis and processing. Image clustering refers to the problem of segmenting image for different purposes which leads to various clustering criteria. Finding the optimal clusters represented by their centers is a hard optimization problem and it is one of the main research focuses on clustering methods. In this paper we proposed a novel generative adversarial optimization algorithm for finding the optimal cluster centers while using standard and advance clustering criteria. The proposed method was tested on seven benchmark images and results were compared with the artificial bee colony, particle swarm optimization and genetic algorithm. Based on the obtained results, the generative adversarial optimization algorithm founded better cluster centers for image clustering compared to named methods from the literature.

To answer safety authority requirements and to optimise the management of radioactive waste produced in retrieval and decommissioning activities, which contains a large variety of matrix materials, the accuracy of neutron measurement techniques has to be continuously improved. Active neutron measurements such as the Differential Die-Away (DDA) technique involving pulsed neutron generator as the neutron source, are widely applied to determine the fissile content of waste packages. Unfortunately, the main drawback of such techniques is coming from the lack of knowledge of the waste matrix composition. Thus, the matrix effect correction for the DDA measurement is an essential improvement in the field of fissile material content determination. Different solutions have been developed to compensate the effect of the matrix on the neutron measurement interpretation for a long time. In Low-Level radioactive Waste (LLW) packages examination, the most widely used methods are based on neutron flux monitoring using small <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> He proportional counters added inside the detection device and associated to the "Matrix Interrogation Source" (MIS) measurement. This technique was originally developed for passive neutron measurement. It needs a specific measurement step which can be operated with the neutron generator or, most of the time, with an external isotopic neutron source such as <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">252</sup> Cf located as closed as possible to the waste drum. This step represents a limiting factor for the examination management and duration. In this context, this paper describes a new approach developed with the goal of increasing the accuracy of the matrix effect correction and reducing the measurement time. This is a major objective in the Non Destructive Assay (NDA) especially to enhance industrial process efficiency of large number of waste packages inspection. It deals with an innovative matrix correction method for radioactive waste embedded in a large variety of matrices regarding the density range (0.07 - 0.9 g.cm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-3</sup> ) as well as the composition (wood representative of hydrogenized matrix, PVC, iron, etc.). The implementation of this method is based on the analysis of the raw signal with an optimisation algorithm called the simulated annealing algorithm. This algorithm needs a reference data base of Multi-Channel Scaling (MCS) spectra, to fit the raw signal. The construction of the MCS library involves a learning phase to define and acquire the DDA signals as representative as possible of the real measurement conditions. This database has been provided by a set of active signals from experimental matrices (mock-up waste drums of 118 litres) recorded in a specific device dedicated to neutron measurement research and development of the Nuclear Measurement Laboratory of CEA-Cadarache, called PROMETHEE 6. This equipment has been designed to reach an empty cavity detection efficiency of 25%. It is equipped with a pulsed (D-T) neutron generator which can reach an average neutron emission rate up to 2.4 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">9</sup> ns <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-1</sup> with a pulse duration of 200 μs. This high technology performance allows achieving very low detection limits with the classical DDA measurement of fissile matter located in light waste matrices (close to 30 μg of <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">239</sup> Pu with an active total measurement time of 900 s). The simulated annealing algorithm is applied to make use of the effect of the matrices on the total active signal of DDA measurement. Furthermore, as this algorithm is directly applied to the raw active signal, it is very useful when active background contributions can not be easily estimated and removed. Most of the cases tested during this work which represents the feasibility phase of the method, are within a 4% agreement interval with the expected experimental value. Moreover, one can notice that without any compensation of the matrix effect, the classical DDA prompt neutron signal analysis may induce an underestimation of more than a factor of 200 on the fissile mass determination for the cases tested in this study. The unexpected so good agreement is a very promising result for the method knowing that the compositions of the mock-up drums are quite representative of the most frequently encountered matrices in LLW packages. This work is the first step of a more general thought carried out to increase the relevance of the whole treatment of DDA measurements from innovative electronic tools (specific fast charge amplifiers, list mode data card system...) up to optimised home made algorithms developed for the post-treatment of the measurements recorded by the list mode data card system.

As precursor to a more complex 3-D design optimization under uncertainty, this 2-D study adds fidelity to a previous tractor-trailer base-flap design optimization study by including more design variables with more relaxed constraints, and the following uncertain effects on drag coefficient (Cd) prediction: wind speed and direction, truck speed and elevation, turbulence model parameters, steady Reynolds-Averaged Navier-Stokes (RANS) approximation, and numerical approximation. To down-select models for the 3-D study, as well as to provide greater confidence that a global minimum Cd is found, this study also compares efficiency and accuracy among 2 optimization algorithms (Evolutionary Algorithm (EA) and DIRECT), a dozen response surface approximations or surrogate models, and 9 surrogate-based global optimization algorithms. This study uses the DAKOTA optimization and uncertainty quantification framework to interface the RANS flow solver (Cobalt), grid generator (Gridgen), and optimization algorithm. The computational model is a simplified full-scale Class-8 tractor-trailer with flow at highway speed (ReW = 5.2x10 6 ). Compared with the no-flaps result, the optimized design yields 36.3% reduction in the windaveraged Cd (Cd,wind-avg). For the optimized design in this study, we estimate total predictive uncertainty at 11% of this uncertainty comes from model form (computation vs. experiment) and numerical approximation (due in this case to significant flow unsteadiness). This total predictive uncertainty for the optimized design is also presented in the form of a probability box, which may be used to decide if and where to focus resources to improve the model and reduce uncertainty. The best design from this study only reduces Cd,wind-avg by 3.1% over a straight flap with 18° inward deflection, which casts doubt whether the small gain is worth the greater manufacturing complexity and expense of a curved flap. The Linear Gaussian Process (GP) surrogate model best fits the data set generated by EA and DIRECT.

Computer-assisted design of small molecules has experienced a resurgence in academic and industrial interest due to the widespread use of data-driven techniques such as deep generative models. While the ability to generate molecules that fulfil required chemical properties is encouraging, the use of deep learning models requires significant, if not prohibitive, amounts of data and computational power. At the same time, open-sourcing of more traditional techniques such as graph-based genetic algorithms for molecular optimisation [Jensen, Chem. Sci., 2019, 12, 3567–3572] has shown that simple and training-free algorithms can be efficient and robust alternatives. Further research alleviated the common genetic algorithm issue of evolutionary stagnation by enforcing molecular diversity during optimisation [Van den Abeele, Chem. Sci., 2020, 42, 11485–11491]. The crucial lesson distilled from the simultaneous development of deep generative models and advanced genetic algorithms has been the importance of chemical space exploration [Aspuru-Guzik, Chem. Sci., 2021, 12, 7079–7090]. For single-objective optimisation problems, chemical space exploration had to be discovered as a useable resource but in multi-objective optimisation problems, an exploration of trade-offs between conflicting objectives is inherently present. In this paper we provide state-of-the-art and open-source implementations of two generations of graph-based non-dominated sorting genetic algorithms (NSGA-II, NSGA-III) for molecular multi-objective optimisation. We provide the results of a series of benchmarks for the inverse design of small molecule drugs for both the NSGA-II and NSGA-III algorithms. In addition, we introduce the dominated hypervolume and extended fingerprint based internal similarity as novel metrics for these benchmarks. By design, NSGA-II, and NSGA-III outperform a single optimisation method baseline in terms of dominated hypervolume, but remarkably our results show they do so without relying on a greater internal chemical diversity.

Design of a dynamic threshold generator for [formula omitted]-tuned control loops

The rapid advancement of generative models and optimization algorithms has opened new opportunities for intelligent garden design systems. However, existing approaches often lack the ability to integrate aesthetic, functional and ecological considerations simultaneously and fail to adapt dynamically to complex design constraints. To address these challenges, this study proposes a dynamic garden design decision-making framework that integrates StyleGAN, Convolutional Neural Networks (CNNs) and the Soft Actor-Critic (SAC) reinforcement learning algorithm. The model combines StyleGAN’s capability to generate diverse and high-quality garden layouts, CNN’s feature extraction for spatial and visual representation and SAC’s optimization for balancing multiple objectives, including ecological benefit, resource efficiency and aesthetic quality. Experimental results on the Places Dataset and AID Dataset demonstrate that the proposed model significantly outperforms baseline methods in image quality, design diversity and optimization effectiveness. Specifically, it achieves superior performance in generating realistic and context-aware garden layouts while maintaining flexibility to adapt to environmental constraints. This research provides an adaptive framework for garden design and offers a foundation for extending dynamic design systems to fields like urban planning and interior design, promoting multi-objective optimization in diverse contexts.

Classification and development of the deployable structures is an ongoing process that started at the end of 20th century and is getting more and more attention throughout 21st. With the development of the technology, constructive systems and materials, these categorizations changed – adding new typologies and excluding certain ones. This work is giving a critical review of the work done previously and on the change of the categories. The special interest is given to the pantographs (or scissor structures) and the Zeigler’s dome as the form of their application. It is noticeable that after its introduction in 1977, the dome was a part of the initial classification, but with the time it lost its place. The reason for this is the introduction of more efficient scissor dome structures. However, perhaps with the use of data-driven design, this dome can be optimized and become relevant again. The second part of the paper is dedicated to the development of the structural optimization algorithm for panto-graph structures and its application on the example of Zeigler’s dome. Besides the direct analysis, the final part includes the generative optimization algorithm which could help to a decision-maker in the early stages of the design to understand and select the options for the structure.

The efficiency of universal electric motors that are widely used in home appliances can be improved by optimizing the geometry of the rotor and the stator. Expert designers traditionally approach this task by iteratively evaluating candidate designs and improving them according to their experience. However, the existence of reliable numerical simulators and powerful stochastic optimization techniques make it possible to automate the design procedure. We present a comparative study of six stochastic optimization algorithms in designing optimal rotor and stator geometries of a universal electric motor where the primary objective is to minimize the motor power losses. We compare three methods from the domain of evolutionary computation, generational evolutionary algorithm, steady-state evolutionary algorithm and differential evolution, two particle-based methods, particle-swarm optimization and electromagnetism-like algorithm, and a recently proposed multilevel ant stigmergy algorithm. By comparing their performance, the most efficient method for solving the problem is identified and an explanation of its success is offered.

The use of Generative AI, Optimization Algorithms, and IoT Technologies in Enhancing Pilgrims’ Journey

This research aims to examine the potential of generative and optimization algorithms in the early stage of a school building design in Tabriz to achieve better IEQ. It also investigates the compatibility of the evolutionary optimization tools combined with a parametric model in stimulating building comfort performance in achieving an optimized design. This process includes four steps: defining the parametric building model, defining its material and construction properties, stimulation of thermal and visual comfort and carbon dioxide concentration, optimization, and choosing the best result. The adaptive PMV model is used for thermal comfort, imageless daylight glare probability is used for visual comfort, and a CO2 concentration is used for IAQ assessment. It was found that the performance of the options introduced by the algorithm is more appropriate than the design prototype. However, the results show that the samples are acceptable in carbon dioxide concentration. What needs further investigation is thermal and visual comfort. Among the studied variables on IEQ performance, the WWR ratio of the southern wall had the most significant impact. Based on the optimization results, thermal comfort changed in the range of 10%, visual comfort in the range of 30%, and CO2 concentration in the range of 0.19%.