Model Complex Research Articles

Advanced stability analysis of a fractional delay differential system with stochastic phenomena using spectral collocation method.

In recent years, there has been a growing interest in incorporating fractional calculus into stochastic delay systems due to its ability to model complex phenomena with uncertainties and memory effects. The fractional stochastic delay differential equations are conventional in modeling such complex dynamical systems around various applied fields. The present study addresses a novel spectral approach to demonstrate the stability behavior and numerical solution of the systems characterized by stochasticity along with fractional derivatives and time delay. By bridging the gap between fractional calculus, stochastic processes, and spectral analysis, this work contributes to the field of fractional dynamics and enriches the toolbox of analytical tools available for investigating the stability of systems with delays and uncertainties. To illustrate the practical implications and validate the theoretical findings of our approach, some numerical simulations are presented.

A Review on the Current Trends in Crop Yield Forecasting Using Artificial Intelligence

In the face of escalating global food demands and the increasing unpredictability of climate conditions, the importance of precise crop yield forecasting has never been more critical. This paper provides a comprehensive review of the current trends in leveraging Artificial Intelligence (AI) to enhance crop yield predictions, which is pivotal for strategic agricultural planning and ensuring food security. Our review covers a range of AI methodologies, including machine learning, deep learning, and hybrid models, that have been employed to predict crop yields with increasing accuracy. Recent advancements have demonstrated that machine learning techniques, such as support vector machines and random forests, are effective in modeling complex agricultural data sets with a notable degree of precision. However, deep learning approaches, including convolutional and recurrent neural networks, have started to outperform traditional machine learning models, owing to their ability to process large-scale spatial-temporal data from remote sensing and IoT-based agricultural sensors. We also explore the emergence of hybrid AI models that combine the strengths of both machine learning and deep learning technologies, providing enhanced accuracy and robustness in yield prediction under varying climatic conditions. Additionally, this review discusses the integration of AI with geographic information systems (GIS) and remote sensing technologies, which has significantly improved the spatial resolution of yield predictions. We highlight several key challenges that remain, such as data scarcity, the need for model generalization, and the integration of socioeconomic factors into yield prediction models. In conclusion, AI presents transformative potential for crop yield forecasting. By harnessing cutting-edge AI technologies and addressing existing challenges, significant strides can be made towards more sustainable and efficient agricultural practices. This paper aims to inspire continued research and innovation in this critical field.

Predictive Analysis of Indonesian Stock Market Prices Using Deep Learning: An Application of Diffusion Variational Autoencoders

This study introduces the application of the Diffusion Variational Autoencoder (D-VAE), a deep learning technique, for predicting stock prices in the Indonesian stock market. With the challenges presented by market volatility and complex data distributions, D-VAE is explored for its capability to encapsulate uncertainty and model complex distributions. This study is significant as it explores the potential of D-VAE in the context of the Indonesian stock market, which has not been widely studied before. Historical stock data from Yahoo Finance was collected over one year and preprocessed for training and validation of the model. The model is trained with an architecture designed to allow tuning of the latent space, utilizing ReLU and linear activation functions for the encoder and decoder. The model's performance is evaluated using the Mean Squared Error (MSE), Mean Absolute Error (MAE), and R-squared metrics, yielding results that highlight the model's capability to enhance the accuracy of stock price predictions. By leveraging machine learning techniques in stock price prediction models, this study underscores the significant contribution such approaches can make to informed and successful investment decisions underpinned by robust data.

Special issue on modelling complex environmental data

Special issue on modelling complex environmental data

Performance-Based Hierarchical Fault-Tolerant Control for Closed-Loop Systems With Multiplicative Faults: A Data-Driven Design Method.

Fault-tolerant control (FTC) is vital for the safety and reliability of automatic systems. Most of the existing FTC methods are developed for open-loop systems subject to additive faults, regardless of the widely present control loops and multiplicative faults within systems. In this article, a performance-based FTC strategy is proposed for the closed-loop systems with multiplicative faults. Considering the high efforts in modeling complex systems, the proposed FTC strategy is realized in the data-driven context. Specifically, a nominal feedback-feedforward controller is first established for the fault-free systems. By selecting the system stability and reference tracking behavior as the key performance indices, two performance evaluators are constructed to detect and classify the occurred multiplicative faults based on the fault-induced effects on the system performance. Then, with the aid of the coprime factorization technique, the multiplicative faults, in the form of additive perturbations to the system coprime factors, are estimated utilizing the closed-loop process data. Furthermore, based on the fault knowledge, a hierarchical fault-tolerant tracking controller is developed according to the levels of system performance degradations, where the functional controller parameters are reconfigured with different priorities. Finally, case studies are provided to validate the effectiveness of the proposed method.

TP-GCL: graph contrastive learning from the tensor perspective

Graph Neural Networks (GNNs) have demonstrated significant potential as powerful tools for handling graph data in various fields. However, traditional GNNs often encounter limitations in information capture and generalization when dealing with complex and high-order graph structures. Concurrently, the sparse labeling phenomenon in graph data poses challenges in practical applications. To address these issues, we propose a novel graph contrastive learning method, TP-GCL, based on a tensor perspective. The objective is to overcome the limitations of traditional GNNs in modeling complex structures and addressing the issue of sparse labels. Firstly, we transform ordinary graphs into hypergraphs through clique expansion and employ high-order adjacency tensors to represent hypergraphs, aiming to comprehensively capture their complex structural information. Secondly, we introduce a contrastive learning framework, using the original graph as the anchor, to further explore the differences and similarities between the anchor graph and the tensorized hypergraph. This process effectively extracts crucial structural features from graph data. Experimental results demonstrate that TP-GCL achieves significant performance improvements compared to baseline methods across multiple public datasets, particularly showcasing enhanced generalization capabilities and effectiveness in handling complex graph structures and sparse labeled data.

A New Approach for Multi-View Models’ Composition using Probes Event

The paper presents the development and application of the VUML Probe profile, an extension of the VUML (View-based UML) approach to modeling complex software systems. It focuses on improving behavioral modeling by introducing probes for observing and monitoring events. In the introduction, the importance of separation of concerns in the management of large software systems is emphasized, with the introduction of view-based modeling and the VUML profile. The application context section presents the VUML analysis and design process, illustrated by a case study of managing an automotive repair shop. Probes are introduced as a modeling concept for event detection and control, with basic categories and methods for projection, derivation, and composition. The VUML Probe profile is presented, integrating probe stereotypes into the VUML meta-model, with conformance rules to maintain semantic consistency. The application of probes in the VUML process is demonstrated, in particular, to ensure the autonomous evolution of model-views. Abstract probes are defined during the composition phase, and then used in view models. Finally, related work and avenues for future research are discussed, including language enhancements, integration with aspect-oriented modeling, and tool development. In summary, the paper offers a comprehensive framework for integrating event observation mechanisms into the VUML approach, aimed at improving the modeling and management of complex software systems.

Cost prediction for water reuse equipment using interpretable machine learning models

During the initial stages of engineering projects management, the accurate estimation of equipment costs plays a crucial role in determining project approval decisions. However, there is a significant research gap regarding the estimation of investment costs for water reuse equipment. Additionally, cost predictions for such construction projects often suffer from limitations, including low accuracy, limited generalizability, and inefficiency. Advanced machine learning (ML) methods, renowned for their ability to model complex decision-making processes, offer powerful solutions. In this study, four traditional models and four ensemble models were employed to predict the cost of water reuse equipment. The results demonstrated that the ensemble models exhibited significantly superior predictive performance to that of traditional models with the three boosting ensemble models achieving the best performance (traditional models: training R2 = 29.77 %–94.85 %, testing R2 = 30.62 %–71.72 %; boosting ensemble models: training R2 = 97.42 %, testing R2 = 82.16 %–93.79 %). Furthermore, this study simplified the features of predictive models and identified the key variables that influence the cost of water reuse equipment using Shapley additive explanations (SHAP) method. The retrained ensemble models re-constructed based on the selected variables achieved significant predictive performance, with the Gradient Boosting Decision Tree (GBDT) outperforming the other models (training set R2 = 97.37 %, testing set R2 = 93.86 %). The water quantity, inflow conductivity, outflow conductivity, and recovery rate emerged as critical factors affecting the cost of water reuse equipment. Overall, the methods proposed in this study can enhance the versatility of cost prediction processes in environmental engineering scenarios, particularly those concerning the construction costs of water treatment equipment.

First-principles insights into the mechanism of CO2hydrogenation reactions by Fe-PNP pincer complex.

Using the state of the art theoretical methods, we have provided a comprehensive mechanistic understanding of the CO2 hydrogenation into HCOOH, H2CO, andCH3OH by 2,6-bis(diisopropylphosphinomethyl)pyridine (PNP)-ligated Fe pincer complex, featuring one CO and two H as co-ligands. For the computational investigation,a verified structural model containing methyl groups in place of the experimental isopropyl groups was used. Three catalytic conversions involving hydrogenation of CO2into formic acid (HCOOH), HCOOH into formaldehyde and methanol were studied indifferent solvent medium. Our modelled complex appears to be a viable base-free catalyst for the conversion of CO2 into HCOOH and HCOOH into H2CO, based on the freeenergy profiles, which show apparent activation energy barriers of 16.28 kcal/mol and23.63 kcal/mol for the CO2 to HCOOH and HCOOH to H2CO conversion, respectively. However, the computed results show that, due to the huge energy span of H2CO toCH3OH conversion, complete hydrogenation of CO2 into methanol could not occur under moderate conditions. Morpholine co-catalyst, which can lower the hydrogenationbarrier by taking part in a simultaneous H-atom donation-acceptance process, couldhave assisted in completing this step.

An alternative experimental configuration to generate wrench zone above a viscous layer

Analogue modelling of wrench tectonics typically utilizes a rigid basement with a velocity discontinuity under a brittle or brittle-viscous cover, such as in Riedel experiments, which confines fault localization in the overlaying model. However, such a set-up is hardly compatible with modeling brittle-ductile systems such as the upper and lower crust or a brittle sedimentary cover overlying a viscous evaporitic layer. To achieve a more realistic experimental approach, Bruno Vendeville designed an alternative experimental set-up decoupling the basement from the brittle overburden with a viscous layer in which the basement is not involved. In this configuration, strike-slip movement is driven laterally rather than from the base up, facilitated by “weak zones” that preferentially localize the deformation during shortening and enable sliding between compartments. This original approach provides greater flexibility for modeling complex strike-slip settings, allowing for more freedom for strike-slip structures to form and evolve through time.Although the experiments described in this work were conducted in the late 1990s, the co-authors have chosen to revisit and adapt this earlier work for this Special Issue to underscore Bruno’s influence on another aspect of salt tectonics and his pioneering foresight in the field of analogue modelling.

Artificial Neural Network Modeling in the Presence of Uncertainty for Predicting Hydrogenation Degree in Continuous Nitrile Butadiene Rubber Processing

The transition from batch to continuous production in the catalytic hydrogenation of nitrile butadiene rubber (NBR) into hydrogenated NBR (HNBR) marks a significant advance for applications under demanding conditions. This study introduces a continuous process utilizing a static mixer (SM) reactor, which notably achieves a hydrogenation conversion rate exceeding 97%. We thoroughly review a mechanistic model of the SM reactor to elucidate the internal dynamics governing the hydrogenation process and address the inherent uncertainties in key parameters such as the Peclet number (Pe), dimensionless time (θτ), reaction coefficient (R), and flow rate coefficient (q). A comprehensive dataset generated from varied parameter values serves as the basis for training an artificial neural network (ANN), which is then compared against traditional models including linear regression, decision tree, and random forest in terms of efficacy. Our results clearly demonstrate the ANN’s superiority in predicting the degree of hydrogenation, achieving the lowest root mean squared error (RMSE) of 3.69 compared to 21.90 for linear regression, 4.94 for decision tree, and 7.51 for random forest. The ANN’s robust capability for modeling complex nonlinear relationships and dynamics significantly enhances decision-making, planning, and optimization of the reactor, reducing computational demands and operational costs. In other words, this approach allows users to rely on a single ML-based model instead of multiple mechanistic models for reflecting the effects of possible uncertainties. Additionally, a feature importance study validates the critical impact of time and element number on the hydrogenation process, further supporting the ANN’s predictive accuracy. These findings underscore the potential of ML-based models in streamlining and enhancing the efficiency of chemical production processes.

Harvesting Earth's heat: A deep learning Odyssey for reservoir characterization and sustainable geothermal energy management

Characterizing and predicting reservoir behavior pose significant challenges in geothermal reservoir engineering. For sustainable geothermal field management, the potential occurrence of thermal breakthroughs in producers during the injection of cold water necessitates a profound understanding of how production is influenced by the injection philosophy. The fractured nature of geothermal reservoirs adds complexity and nonlinearity to the relationship between production and injection wells. In this study, we explore alternative models to simulate reservoir behavior as substitutes for full reservoir simulations. Utilizing deep learning algorithms, we investigate two different architectures, namely the standard Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU). These models map injection flow rates at the injectors to tracer concentration data at the producers through layers, nodes, and activation functions. Training these models on a synthetic geothermal reservoir, Notably, the GRU architecture consistently demonstrates superior predictive capabilities across various scenarios for all three producers. Specifically, in configurations involving a target feed, the GRU model yields higher R2 values compared to its LSTM counterpart, indicating its effectiveness in capturing inter-well relationships. Furthermore, when considering all producers collectively, the GRU model exhibits a trend of lower test errors and higher R2 values, reaffirming its proficiency in modeling complex reservoir dynamics. These findings underscore the significance of GRU as a preferred choice for accurate reservoir behavior prediction. In addition to the aforementioned insights, our study contributes novel approaches to reservoir behavior prediction in the field of geothermal reservoir engineering. By leveraging deep learning algorithms, specifically the GRU and LSTM architectures, we introduce innovative methodologies for simulating reservoir behavior as alternatives to full reservoir simulations. Through comprehensive analysis, we demonstrate the superior predictive capabilities of the GRU model, particularly in capturing inter-well relationships and modeling complex reservoir dynamics.

Advances in CALPHAD Methodology for Modeling Hydrides: A Comprehensive Review

AbstractHydrides enable handling hydrogen at low pressure and near room temperature, offering higher volumetric densities than compressed or liquid hydrogen and enhancing safety. The CALPHAD method, rooted in the principles of thermodynamics, offers a systematic approach for predicting phase equilibria and thermodynamic properties in multicomponent materials. This comprehensive review paper aims to provide a detailed overview of the application of the CALPHAD method in the realm of metallic and complex hydrides. After an introduction to the fundamental thermodynamic aspects of hydrides, key elements of applying the CALPHAD method to model metal-hydrogen systems and complex hydrides are discussed. Subsequently, recent publications are reviewed, highlighting key findings and recent progresses in the field. Finally, the challenges that must be overcome to achieve further progress in this area are explored.

A Unified Asymmetric Knowledge Distillation Framework for Image Classification

Knowledge distillation is a model compression technique that transfers knowledge learned by teacher networks to student networks. Existing knowledge distillation methods greatly expand the forms of knowledge, but also make the distillation models complex and symmetric. However, few studies have explored the commonalities among these methods. In this study, we propose a concise distillation framework to unify these methods and a method to construct asymmetric knowledge distillation under the framework. Asymmetric distillation aims to enable differentiated knowledge transfers for different distillation objects. We designed a multi-stage shallow-wide branch bifurcation method to distill different knowledge representations and a grouping ensemble strategy to supervise the network to teach and learn selectively. Consequently, we conducted experiments using image classification benchmarks to verify the proposed method. Experimental results show that our implementation can achieve considerable improvements over existing methods, demonstrating the effectiveness of the method and the potential of the framework.

Single file motion of robot swarms

We present experimental results on the single file motion of a group of robots interacting with each other through position sensors. We successfully replicate the fundamental diagram typical of these systems, with a transition from free flow to congested traffic as the density of the system increases. In the latter scenario we also observe the characteristic stop-and-go waves. The unique advantages of this novel system, such as experimental stability and repeatability, allow for extended experimental runs, facilitating a comprehensive statistical analysis of the global dynamics. Above a certain density, we observe a divergence of the average jam duration and the average number of robots involved in it. This discovery enables us to precisely identify another transition: from congested intermittent flow (for intermediate densities) to a totally congested scenario for high densities. Beyond this finding, the present work demonstrates the suitability of robot swarms to model complex behaviors in many particle systems. Published by the American Physical Society 2024

Advanced fusion of MTM-LSTM and MLP models for time series forecasting: An application for forecasting the solar radiation

Accurate time series forecasting has become increasingly important across various domains such as finance, energy, and medicine. This study introduces an innovative hybrid model that leverages the power of neural networks, precisely Many To Many LSTM (MTM LSTM) and Multilayer Perceptron (MLP), to improve time series forecasting accuracy. In this new combination, we trained network MTM LSTM to approximate the target at each step, and finally, we used network MLP to combine these approximations. To perform the evaluation, we made a forecast for the amount of solar energy radiation in the city of Mashhad, Iran. The experiment results concerning MSE and MAE showed that the proposed method with five lags outperforms the standard models.We hypothesize that MTM LSTM can effectively capture solar radiation's intricate temporal dependencies and nonlinearity. At the same time, MLP can enhance function approximation by modeling complex interactions, resulting in improved forecast accuracy. By employing the hybrid MTM LSTM and MLP model, we achieved improved accuracy in predicting solar energy radiation, which has significant implications for the renewable energy sector and its energy management and planning applications. This research advances time series forecasting techniques, highlighting the effectiveness of combining neural networks to address complex and dynamic patterns in time-dependent data.Overall, our findings underscore the potential and efficacy of the proposed hybrid model as a robust tool for accurate time series forecasting in various domains, supporting effective decision-making and planning processes.

M2WISH: An easy and efficient protocol for whole‐mount mRNAin situ hybridization that allows 3D cell resolution of gene expression in Arabidopsis thaliana

SUMMARYGene expression analysis is essential for understanding the mechanisms involved in plant development. Here, we developed M2WISH, a protocol based on MicroWave treatment for Wholemount mRNA In Situ Hybridization in Arabidopsis. By permeabilizing tissues without damaging cellular organization this protocol results in high and homogeneous hybridization yields that enable systematic analysis of gene expression dynamics. Moreover, when combined with cellular histochemical staining, M2WISH successfully provides a cellular resolution of gene expression. Thus, we demonstrate the robustness of M2WISH with 10 genes on roots, aerial meristems, leaves, and embryos in the seed. We applied M2WISH to study the spatial dynamics of WUSCHEL (WUS) and CLAVATA3 (CLV3) expression during in vitro meristematic conversion of roots into shoot apical meristems. Thus, we showed that shoot apical meristems could arise from two different types of root structures that differed by their CLV3 gene expression patterns. We constructed 3D cellular representations of WUS and CLV3 gene co‐expression pattern and stressed the variability inherent to meristem conversion. Thus, this protocol generates a large amount of data on the localization of gene expression, which can be used to model complex systems.

Updating Nonlinear Stochastic Dynamics of an Uncertain Nozzle Model Using Probabilistic Learning With Partial Observability and Incomplete Dataset

Abstract This article introduces a methodology for updating the nonlinear stochastic dynamics of a nozzle with uncertain computational model. The approach focuses on a high-dimensional nonlinear computational model constrained by a small target dataset. Challenges include the large number of degrees-of-freedom, geometric nonlinearities, material uncertainties, stochastic external loads, underobservability, and high computational costs. A detailed dynamic analysis of the nozzle is presented. An updated statistical surrogate model relating the observations of interest to the control parameters is constructed. Despite small training and target datasets and partial observability, the study successfully applies probabilistic learning on manifolds (PLoM) to address these challenges. PLoM captures geometric nonlinear effects and uncertainty propagation, improving conditional mean statistics compared to training data. The conditional confidence region demonstrates the ability of the methodology to accurately represent both observed and unobserved output variables, contributing to advancements in modeling complex systems.

Listen as you wish: Fusion of audio and text for cross-modal event detection in smart cities

In the era of smart cities, the advent of the Internet of Things technology has catalyzed the proliferation of multimodal sensor data, presenting new challenges in cross-modal event detection, particularly in audio event detection via textual queries. This paper focuses on the novel task of text-to-audio grounding (TAG), aiming to precisely localize sound segments that correspond to events described in textual queries within an untrimmed audio. This challenging new task requires multi-modal (acoustic and linguistic) information fusion as well as the reasoning for the cross-modal semantic matching between the given audio and textual query. Unlike conventional methods that often overlook the nuanced interactions between and within modalities, we introduce the Cross-modal Graph Interaction (CGI) model. This innovative approach leverages a language graph to model complex semantic relationships between query words, enhancing the understanding of textual queries. Additionally, a cross-modal attention mechanism generates snippet-specific query representations, facilitating fine-grained semantic matching between audio segments and textual descriptions. A cross-gating module further refines this process by emphasizing relevant features across modalities and suppressing irrelevant information, optimizing multimodal information fusion. Our comprehensive evaluation on the Audiogrounding benchmark dataset not only demonstrates the CGI model’s superior performance over existing methods, but also underscores the significance of sophisticated multimodal interaction in improving the efficacy of TAG in smart cities.

Dishevelled2 activates WGEF via its interaction with a unique internal peptide motif of the GEF

The Wnt-planar cell polarity (Wnt-PCP) pathway is crucial in establishing cell polarity during development and tissue homoeostasis. This pathway is found to be dysregulated in many pathological conditions, including cancer and autoimmune disorders. The central event in Wnt-PCP pathway is the activation of Weak-similarity guanine nucleotide exchange factor (WGEF) by the adapter protein Dishevelled (Dvl). The PDZ domain of Dishevelled2 (Dvl2PDZ) binds and activates WGEF by releasing it from its autoinhibitory state. However, the actual Dvl2PDZ binding site of WGEF and the consequent activation mechanism of the GEF have remained elusive. Using biochemical and molecular dynamics studies, we show that a unique “internal-PDZ binding motif” (IPM) of WGEF mediates the WGEF-Dvl2PDZ interaction to activate the GEF. The residues at P2, P0, P-2 and P-3 positions of IPM play an important role in stabilizing the WGEFpep-Dvl2PDZ interaction. Furthermore, MD simulations of modelled Dvl2PDZ-WGEFIPM peptide complexes suggest that WGEF-Dvl2PDZ interaction may differ from the reported Dvl2PDZ-IPM interactions. Additionally, the apo structure of human Dvl2PDZ shows conformational dynamics different from its IPM peptide bound state, suggesting an induced fit mechanism for the Dvl2PDZ-peptide interaction. The current study provides a model for Dvl2 induced activation of WGEF.