A novel adaptive sampling based methodology for feasible region identification of compute intensive models using artificial neural network

Identification of feasible region of operations in multivariate processes is a problem of interest in several fields. This is particularly challenging when the process model is black-box in nature and/or is computationally expensive, as analytical solutions are not available and the number of possible model evaluations is limited. An efficient methodology is required to identify samples where the model is evaluated for developing a computationally efficient surrogate model. In this work, an artificial neural network based surrogate model is proposed which is integrated with a statistical-based approach (Jack-knifing) to estimate the variance of the surrogate model prediction. This allows implementation of an adaptive sampling approach where new samples are identified close to the feasible region boundary or in regions of high prediction uncertainty. The proposed approach performs better than a previously published kriging based method for different dimensionality case studies.     

Predicting the mechanical properties of Cu–Al2O3 nanocomposites using machine learning and finite element simulation of indentation experiments

Micromechanics model, finite element (FE) simulation of microindentation and machine learning were deployed to predict the mechanical properties of Cu–Al₂O₃ nanocomposites. The micromechanical model was developed based on the rule of mixture and grain and grain boundary sizes evolution to predict the elastic modulus of the produced nanocomposites. Then, a FE model was developed to simulate the microindentation test. The input for the FE model was the elastic modulus that was computed using the micromechanics model and wide range of yield and tangent stresses values. Finally, the output load-displacement response from the FE model, the elastic modulus, the yield and tangent strengths used for the FE simulations, and the residual indentation depth were used to train the machine learning model (Random vector functional link network) for the prediction of the yield and tangent stresses of the produced nanocomposites. Cu–Al₂O₃ nanocomposites with different Al₂O₃ concentration were manufactured using insitu chemical method to validate the proposed model. After training the model, the microindentation experimental load-displacement curve for Cu–Al₂O₃ nanocomposites was fed to the machine learning model and the mechanical properties were obtained. The obtained mechanical properties were in very good agreement with the experimental ones achieving 0.99 coefficient of determination R2 for the yield strength.     

梁结构振动支承约束反力控制

支承或连接构件对梁结构的动力学性能有至关重要影响,必须保证其在振动过程中不发生破坏或者失效。通过合理设计和布局附加弹性支承可以实现对这些重要连接构件所承受约束反力的控制。应用微分变换法推导含附加支承的梁结构支承约束反力及其对于附加支承位置和刚度的灵敏度表达式,并通过优化设计附加支承位置和刚度实现具有弹性约束端的简支梁结构各支承约束反力的平衡,可提高结构的动力学性能。     

Two Methods,One Goal: Structural Differences between Cocrystallization and Crystal Soaking to Discover Ligand Binding Poses

In lead optimization, protein crystallography is an indispensable tool to analyze drug binding. Binding modes and non-covalent interaction inventories are essential to design follow-up synthesis candidates. Two protocols are commonly applied to produce protein–ligand complexes: cocrystallization and soaking. Because of its time and cost effectiveness, soaking is the more popular method. Taking eight ligand hinge binders of protein kinase A, we demonstrate that cocrystallization is superior. Particularly for flexible proteins, such as kinases, and larger ligands cocrystallization captures more reliable the correct binding pose and induced protein adaptations. The geometrical discrepancies between soaking and cocrystallization appear smaller for fragment-sized ligands. For larger flexible ligands that trigger conformational changes of the protein, soaking can be misleading and underestimates the number of possible polar interactions due to inadequate, highly impaired positions of protein amino-acid side and main chain atoms. Thus, if applicable cocrystallization should be the gold standard to study protein–ligand complexes.     

A customized low-rank prior model for structured cartoon–texture image decomposition

The mathematical characterization of the texture component plays an instrumental role in image decomposition. In this paper, we are concerned with a low-rank texture prior based cartoon–texture image decomposition model, which utilizes a total variation norm and a global nuclear norm to characterize the cartoon and texture components, respectively. It is promising that our decomposition model is not only extremely simple, but also works perfectly for globally well-patterned images in the sense that the model can recover cleaner texture (or details) than the other novel models. Moreover, such a model can be easily reformulated as a separable convex optimization problem, thereby enjoying a splitting nature so that we can employ a partially parallel splitting method (PPSM) to solve it efficiently. A series of numerical experiments on image restoration demonstrate that PPSM can recover slightly higher quality images than some existing algorithms in terms of taking less iterations or computing time in many cases.     

浅渍法发酵豇豆的工艺优化

以具有降解亚硝酸盐功能的植物乳杆菌（Lactobacillus plantarum）SD-7和具有优良抗氧化能力的植物乳杆菌（Lactobacillus plantarum）FM-LP-9为复合发酵剂（1∶1），采用浅渍法发酵豇豆。以硬度和感官评分为考察指标，通过单因素试验及响应面试验研究复合发酵剂接种量、发酵温度和发酵时间对浅渍法发酵豇豆品质的影响。结果表明，最佳发酵工艺条件为发酵温度25 ℃，接种量5%，发酵时间125 h，在此优化条件下，得到的浅渍法发酵豇豆硬度为47.31 N，感官评分为90.98分，香气浓郁、口感脆嫩。     

Pattern design and optimization of yarn-dyed plaid fabric using isolation niche genetic algorithm and rough set theory

Under the circumstance of perceptual consumption, it is still challenging to grasp consumer's emotions and demands due to the large search space, diversified preferences, and easy fatigue of consumers. To reduce user fatigue and enlarge search space, a novel method was presented to design and optimize the pattern of yarn-dyed plaid fabric using the isolation niche genetic algorithm and rough set theory. Each pattern was encoded as a chromosome based on the real number code. The population was initialized and evolved using INGA to maintain the diversity. The rough set theory was adopted as the fitness function of isolation niche genetic algorithm to extract the consumer's demands. After multiple evolutions, a large set of practical patterns of the yarn-dyed plaid fabric are obtained. Experiments were carried out by 24 testers of different ages and genders. The results prove that the proposed method based on the isolation niche genetic algorithm and rough set theory is feasible and effective, supplying references to the designer.     

An automated treatment planning strategy for highly noncoplanar radiotherapy arc trajectories

Radiation therapy is a technology-driven cancer treatment modality that has experienced significant advances over the last decades, due to multidisciplinary contributions that include engineering and computing. Recent technological developments allow the use of noncoplanar volumetric modulated arc therapy (VMAT), one of the most recent photon treatment techniques, in clinical practice. In this work, an automated noncoplanar arc trajectory optimization framework designed in two modular phases is presented. First, a noncoplanar beam angle optimization algorithm is used to obtain a set of noncoplanar irradiation directions. Then, anchored in these directions, an optimization strategy is proposed to compute an optimal arc trajectory. The computational experiments considered a pool of twelve difficult head-and-neck tumor cases. It was possible to observe that, for some of these cases, the optimized noncoplanar arc trajectories led to significant treatment planning quality improvements, when compared with coplanar VMAT treatment plans. Although these experiments were done in a research environment treatment planning software (matRad), the conclusions can be of interest for a clinical setting: automated procedures can simplify the current treatment workflow, produce high-quality treatment plans, making better use of human resources and allowing for unbiased comparisons between different treatment techniques.     

Towards artificial intelligence at scale in the chemical industry

In the Industry 4.0 era, the chemical industry is embracing broad adoption of artificial intelligence (AI) and machine learning (ML) methods. This article provides a holistic view of how the industry is transforming digitally towards AI at scale. First, a historical perspective on how the industry used AI to aid humans in better decision-making is shown. Then state-of-the-art AI research addressing industrial needs on reliability and safety, process optimization, supply chain, material discovery, and reaction engineering is highlighted. Finally, a vision of the plant of the future is illustrated with critical components of AI-ready culture, model life cycle management, and renewed role of humans in chemical manufacturing.     

Coverage optimization algorithm based on VF-CS in mobile sensor network

A random placement of large-scale sensor network in the outdoor environment often causes low coverage.An area coverage optimization algorithm of mobile sensor network (MSN) based on virtual force perturbation and Cuckoo search (VF-CS) was proposed.Firstly,the virtual force of the sensor nodes within the Thiessen polygon was analyzed based on the partitioning of Voronoi diagram of the monitoring area.Secondly,the force of polygon vertices and neighbor nodes was taken as the perturbation factor for updating the node’s location of the Cuckoo search (CS).Finally,the VF-CS guided the node to move so as to achieve the optimal coverage.The simulation results demonstrate that the proposed algorithm has higher coverage and shorter average moving distance of nodes than the Voronoi diagram based algorithms in literatures.