Adaptive Model Research Articles

Dynamics of stand density and self-thinning in Chinese fir plantations: theoretical insights and empirical validation

IntroductionStand density management is essential for adaptive silviculture, thinning decisions, growth modeling, and yield prediction in forestry, particularly for plantations. Despite extensive research on self-thinning rules and the maximum size-density law, significant gaps remain in the biophysical understanding and validation of the relationships among key stand variables and parameters.MethodsThis study theoretically explored and validated the relationship between maximum size-density and two key metrics: average diameter at breast height (D) and tree height (H). We used time-series data from a 30-year clear-cut, fully stocked Chinese fir plantation, a fast-growing commercial species in China, for validation.ResultsA growth balance status for fully stocked stands was proposed, wherein prior to self-thinning, the growth rate of the stand basal area (G) aligns with that of the average tree height (H), expressed as G'/(G−b0)=H'/H and approaching a constant slope, b1. Generalized maximum size-density and stand density index (SDI) equations were developed: N1.0=A×D−2 and SDI=A⋅D−2 with A=4×(b0+b1H)/π, differing from traditional equations. Additionally, a generalized self-thinning equation, v=kHqN1.0−1 or w=c1HqN1.0−1, was introduced, indicating that in fully stocked stands, tree volume or biomass depends on both tree height and tree count.DiscussionThese findings advance understanding of the maximum size-density law and self-thinning boundary, providing refined tools for managing stand density in Chinese fir plantations.

Adaptive hybrid prediction model for adapting to data distribution shifts in machining quality prediction

Abstract The accuracy of data-driven intelligent prediction for machining quality relies on the training samples. However, in actual applications, the continuous operation of machining equipment leads to gradual distribution shifts between the process data and the training samples for modeling. The shifts result in a degradation in the performance of predictive model, previous studies have often overlooked this issue. To tackle with the intricate problem, this research proposes a real-time model optimization approach. Firstly, a method for detecting machining data distribution shifts based on the two-sample Kolmogorov–Smirnov test is proposed. Then, an adaptive hybrid prediction model (AHPM) capable of real-time optimization is developed. This model consists of a deep neural network (DNN) and a broad learning system (BLS). DNN plays a primary role in prediction within the hybrid model with excellent generalization capability. BLS quickly completes optimization prior to DNN with its unique parameter update mechanism to compensate for prediction loss. Experimental results indicate that AHPM achieves the shortest optimization time while maintaining high accuracy, with post-optimization error reduction rates for mean squared error, mean absolute error, and mean absolute percentage error all exceeding 10%. In the test of application to actual machining cases, accuracy improved by 8.88% compared to traditional methods without optimization.

Artificial Intelligence and/or Machine Learning Algorithms in Microalgae Bioprocesses

This review examines the increasing application of artificial intelligence (AI) and/or machine learning (ML) in microalgae processes, focusing on their ability to improve production efficiency, yield, and process control. AI/ML technologies are used in various aspects of microalgae processes, such as real-time monitoring, species identification, the optimization of growth conditions, harvesting, and the purification of bioproducts. Commonly employed ML algorithms, including the support vector machine (SVM), genetic algorithm (GA), decision tree (DT), random forest (RF), artificial neural network (ANN), and deep learning (DL), each have unique strengths but also present challenges, such as computational demands, overfitting, and transparency. Despite these hurdles, AI/ML technologies have shown significant improvements in system performance, scalability, and resource efficiency, as well as in cutting costs, minimizing downtime, and reducing environmental impact. However, broader implementations face obstacles, including data availability, model complexity, scalability issues, cybersecurity threats, and regulatory challenges. To address these issues, solutions, such as the use of simulation-based data, modular system designs, and adaptive learning models, have been proposed. This review contributes to the literature by offering a thorough analysis of the practical applications, obstacles, and benefits of AI/ML in microalgae processes, offering critical insights into this fast-evolving field.

Global reliability sensitivity analysis of cradle-type double rotary table

The rotary table is a crucial component of the machine tool, and its accuracy and reliability have an important impact on the machining accuracy and machining efficiency of the machine tool. In this paper, a global reliability sensitivity analysis method of the rotary table is proposed considering the random uncertainty for material parameters of each component. Firstly, the mechanical model of the rotary table is established based on the finite element method. Subsequently, the functional relationship between the parameters of each component and the deformation of the rotary table is reconstructed using the adaptive Kriging surrogate model. Finally, a global reliability sensitivity analysis of the rotary table is performed based on the Bayesian formulation. The results show that the proposed global reliability sensitivity analysis method of the rotary table demonstrates satisfactory efficiency and can provide theoretical guidance for the design and optimization of the rotary table.

Efficient Federated Recommender System with Adaptive Model Pruning and Momentum-based Batch Adjustment

With the development of 5G communication and smart devices, the prosperity of online content has boosted the research of Recommender System (RS). Due to the data scarcity problem, researchers employ knowledge transfer techniques to improve the accuracy of RS. Data sharing or data augmentation are promising methods for such a problem, but the data suffers from privacy leakage during sharing. Thus, Federated Learning (FL) has been adopted to collaboratively train recommender models while preserving data privacy. Federated Recommender System (FRS) combines FL and RS to provide distributed recommendation services to the users. However, the existing FRS suffers from massive communication and system heterogeneity, where the necessity of model transmission and the diversity of the clients bring significant communication overhead to the system. In this paper, we propose Efficient Federated Recommender System with Adaptive Model Pruning and Momentum-based Batch Adjustment ( eFRSA 2 ) to reduce the communication overhead of FRS. eFRSA 2 contains two modules. Adaptive Model Pruning utilizes magnitude pruning to reduce the communication volume and adaptively modifies the compression ratios of different clients to maintain the model accuracy. Momentum-based Batch Adjustment adjusts the local training batch number by a similar method of gradient descent with momentum to align the local computation time of the clients and reduce the communication overhead. The experimental results demonstrate that eFRSA 2 can reduce up to 90% communication volume and mitigate the system heterogeneity by over 75%, demonstrating the priority of eFRSA 2 in training efficiency. Source code can be found at https://github.com/shhjwu5/eFRSA2.

Resonance-State Temperature Compensation Method for Ultrasonic Resonance Wind Speed and Direction Sensors

To achieve high-precision wind speed and direction measurements in complex environments, a resonance-state temperature compensation method is proposed based on an ultrasonic resonance principle. This method effectively addresses the issue of sound velocity compensation errors caused by the temperature difference between the internal and external environments when using an internal temperature sensor for temperature compensation. By utilizing an adaptive resonance-state tracking model, the resonance frequency shift issues under varying conditions such as altitude, pressure, and temperature are mitigated. This approach ensures that the resonance frequency is strongly correlated with temperature, enabling temperature compensation through resonance frequency alone, without the need for a temperature sensor. The experimental results indicate that the resonance frequency variation rate with temperature for the resonance-state temperature-compensated ultrasonic resonance wind speed and direction sensor is approximately 0.08 kHz/°C. The wind speed measurement accuracy is ±0.3 m/s (≤15 m/s)/±2.3% (15 m/s~50 m/s), which is superior to the measurement accuracy of traditional ultrasonic wind speed and direction sensors (±0.5 m/s (≤15 m/s)/±4% (15 m/s~50 m/s)). The consistency of wind speed measurement is ≤±0.3%, representing an improvement of approximately 3% compared to ultrasonic resonance wind speed and direction sensors without resonance-state temperature compensation.

Adaptive Two-Degree-of-Freedom Robust Gain-Scheduling Control Strategy

This study introduces a novel tracking control strategy tailored to aeroengines, which are highly nonlinear and characterized by significant uncertainty. The proposed method entails a robust extended Kalman filter (REKF) enhanced by a forgetting factor for improved performance. An accompanying augmented, mixed onboard adaptive model based on the REKF precisely estimates and manages engine performance degradation. This advanced model effectively counters the degradation term in the perturbation block of the engine’s uncertain model. Using this strategic approach, a robust gain-scheduling controller was constructed and was found to outperform its predecessors, marking a notable advancement in control system design. Controlling twin rotor multi-input, multi-output (MIMO) systems is a highly complex process due to model uncertainties and unpredictable external disturbances. To address these challenges, we constructed an adaptive two-degree-of-freedom robust gain-scheduling controller (ATDF-RGSC) using a mixed sensitivity approach. Rigorous performance analysis confirms that this controller offers enhanced robustness, faster tracking, and more precise disturbance attenuation compared to other methods. These advanced control strategies successfully manage uncertainties and disturbances, improving performance metrics in both simulated and experimental scenarios. The proposed method may significantly enhance the safety and reliability of aeroengines and MIMO systems in practical applications.

Managing operational resilience during the implementation of digital transformation in healthcare organisational practices.

The aim of this study is to investigate ways in which healthcare organisations can successfully maintain operational resilience within intricate and varied engagements during digital transformation processes. The present research applied cultural-historical activity theory as the theoretical framework and the ethnographic account as an approach and strategy to interpret and understand theoperational resilience of digital transformation tools in daily practices. Fieldwork was based on the research technique of shadowing, whereby the researcher closely accompanied the participants to record their conduct, activities and exchanges. Research results propose that effective operational resilience management in the implementation of digital transformation projects is based on (1) identifying and interpreting internal contradictions in everyday interactions as opportunities for capability developments; (2) navigating through multiple sites in fast and improvised movements, which derives in distributed and emergent practices; (3) interplaying between dyadic interactions and networked dependencies, which is achieved through the articulation of varied interests and (4)implementing novel intermediary tools, roles and regulations that facilitate the reduction of disturbances. The propositions of the present study indicate that the management of operational resilience extends beyond conventional adaptive and socio-technical models in healthcare services. The study emphasises the significance of expressing and converting differing interests into mutual advantages. It additionally demonstrates the intricacy of this obstacle, as it entails navigating through uncertain information, concealed interpretations and conflicting interests.

Bridging the gap: opportunities, challenges, and future directions for teleradiology in rural healthcare

Teleradiology, the electronic transmission of radiological images and data for diagnostic purposes, holds significant promise for improving healthcare access in rural and underserved regions. This review explores the opportunities and challenges of implementing teleradiology in rural settings, where limited access to radiology services exacerbates healthcare disparities. Teleradiology offers multiple benefits, including improved access to timely diagnostics, enhanced decision-making, and cost savings for rural patients who would otherwise need to travel to urban centers for radiology services. It has shown positive outcomes globally, with examples in India, Mali, the Democratic Republic of Congo, and Brazil, demonstrating its feasibility and potential to reduce geographic health inequities. However, implementing teleradiology in these settings faces several challenges, such as technological limitations, workforce shortages, regulatory and legal barriers, data security concerns, and cultural acceptance issues. Solutions and recommendations include investments in infrastructure, workforce training, supportive regulatory frameworks, secure data management, and adaptable telehealth models that consider socioeconomic factors in rural areas. By addressing these challenges, stakeholders can enhance the impact of teleradiology in rural healthcare systems, ultimately contributing to reduced disparities and improved healthcare outcomes. This review provides a comprehensive overview of the current landscape and suggests actionable pathways for advancing teleradiology in rural settings worldwide.

KEY STAGES AND INNOVATIVE APPROACHES IN RISK MANAGEMENT UNDER UNCERTAINTY

The purpose of the article is to analyze in detail the process of risk management at enterprises, in particular, its key stages. In addition, the article aims at studying the current instruments of bankruptcy risk management which can be implemented in the modern economy. The relevance of the article is stipulated by the need to improve approaches to risk management in today's dynamic business environment, where enterprises are constantly facing financial and operational threats. Of particular importance is bankruptcy risk management, since economic instability and high level of competition require effective mechanisms to protect financial stability. The following methods are used in the study: the method of analysis and synthesis - for a structured consideration of the stages of the risk management process, identifying the relationships between them and forming a holistic picture of management actions at each stage; the method of logical generalization - for formulating general conclusions and recommendations on the stages and tools of risk management based on the analysis of literature and practical examples. The result of the study is to determine the essential characteristics of the main stages of the enterprise risk management process and the relevant tools for managing the bankruptcy risk. This includes: a structured description of the key stages of risk management – from risk identification and assessment to monitoring and control, which ensures a systematic approach to responding to threats; analysis of current tools for managing bankruptcy risk – identification of tools that facilitate early detection of risks and support the financial stability of the enterprise. The practical value of the article lies in providing enterprises with a clear description of the risk management process, which includes the main stages from identification to monitoring and control, allowing companies to take a more systematic approach to risk management and reducing the likelihood of negative financial consequences. The use of the current tools for managing bankruptcy risk described in the article provides an opportunity to quickly identify early signs of financial instability, which helps to maintain their financial stability. In addition, the proposed recommendations can be integrated into the enterprise management system, creating a flexible and adaptive model to maintain stable development in the face of economic uncertainty.

Optimizing pharmaceutical supply chain management through AI-driven predictive analytics: A conceptual framework

The pharmaceutical supply chain is a complex, multi-layered system that faces unique challenges, including fluctuating demand, stringent regulatory requirements, and logistical constraints. This paper explores the role of AI-driven predictive analytics in optimizing pharmaceutical supply chain management, presenting a conceptual framework that enables companies to leverage advanced data analytics for improved decision-making, risk management, and operational efficiency. Key AI techniques, such as machine learning, data mining, and predictive demand forecasting, are discussed as tools for addressing critical supply chain issues, including inventory optimization, demand variability, and regulatory compliance. The proposed framework details essential components—data collection, processing, integration with existing systems, and real-time decision-making—and outlines critical success factors for effective implementation. By offering insights into the transformative potential of predictive analytics, this paper provides actionable recommendations for pharmaceutical companies seeking to build resilient and responsive supply chains. Future research directions are also proposed, emphasizing the need for adaptable predictive models and the ethical implications of AI applications in pharmaceuticals. Keywords: Pharmaceutical Supply Chain, AI-driven Predictive Analytics, Machine Learning, Demand Forecasting, Inventory Optimization, Regulatory Compliance.

Crowd Flow Prediction: An Integrated Approach Using Dynamic Spatial–Temporal Adaptive Modeling for Pattern Flow Relationships

ABSTRACTPredicting crowd flows in smart cities poses a significant challenge for the intelligent transportation system (ITS). Traffic management and behavioral analysis are crucial and have garnered considerable attention from researchers. However, accurately and timely predicting crowd flow is difficult due to various complex factors, including dependencies on recent crowd flow and neighboring regions. Existing studies often focus on spatial–temporal dependencies but neglect to model the relationship between crowd flow in distant areas. In our study, we observe that the daily flow of each region remains relatively consistent, and certain regions, despite being far apart, exhibit similar flow patterns, indicating a strong correlation between them. In this paper, we proposed a novel Multiscale Adaptive Graph‐Gated Network (MSAGGN) model. The main components of MSAGGN can be divided into three major parts: (1) To capture the parallel periodic learning architecture through a layer‐wise gated mechanism, a layer‐wise functional approach is employed to modify gated mechanism, establishing parallel skip periodic connections to effectively manage temporal and external factor information at each time interval; (2) a graph convolutional‐based adaptive mechanism that effectively captures crowd flow traffic data by considering dynamic spatial–temporal correlations; and (3) we proposed a novel intelligent channel encoder (ICE). The task of this block is to capture citywide spatial–temporal correlation along external factors to preserve correlation for distant regions with external elements. To integrate spatio‐temporal flexibility, we introduce the adaptive transformation module. We assessed our model's performance by comparing it with previous state‐of‐the‐art models and conducting experiments using two real‐world datasets for evaluation.

Leveraging Machine Learning for Cybersecurity: Techniques, Challenges, and Future Directions

This study aims to explore the application of machine learning (ML) in enhancing cybersecurity measures, focusing specifically on intrusion detection, malware detection, fraud detection, and anomaly detection systems. A systematic review of 743 scholarly papers was conducted, from which 115 were selected for in-depth analysis. The review process involved evaluating advancements in ML techniques within the context of cybersecurity. The findings reveal that ML-driven systems significantly enhance the automation of security processes, improve the recognition of novel threats, and reduce human error in cyber threat management. However, challenges such as adversarial attacks and the need for high-quality model training pose significant barriers to the broader adoption of ML in cybersecurity. The study discusses the implications of these findings, emphasizing the necessity for developing robust and adaptive ML models that can withstand adversarial threats while improving integration across various cybersecurity applications. The insights gained from this research provide a comprehensive overview of the potential benefits and challenges of implementing ML in cybersecurity, highlighting the need for continuous innovation in threat detection mechanisms. This review acknowledges limitations such as the predominance of literature from Western contexts, potentially overlooking insights from other regions. Furthermore, the complexity of implementing ML systems in dynamic cyber environments remains a critical challenge. Future research should concentrate on refining ML algorithms to enhance resilience against adversarial threats, exploring the integration of emerging technologies for improved cybersecurity, and addressing gaps in the existing literature regarding the life cycle of ML models in real-world applications.

Revolutionizing education with AI: The adaptive cognitive enhancement model (ACEM) for personalized cognitive development

Abstract. The integration of artificial intelligence (AI) into education has opened doors to personalized learning experiences. This paper introduces the Adaptive Cognitive Enhancement Model (ACEM), a cutting-edge AI-driven framework designed to personalize cognitive development for students. Leveraging advanced machine learning algorithms and quantitative analysis, ACEM adapts educational content and learning strategies to individual cognitive needs. The model encompasses five key components: cognitive profiling, adaptive learning paths, intelligent feedback, motivational strategies, and longitudinal tracking. Through quantitative analysis and mathematical modeling, the paper demonstrates how ACEM significantly enhances learning outcomes compared to traditional education models. The discussion section provides a detailed exploration of each model component, its architecture, and its role in optimizing personalized cognitive development. Furthermore, challenges such as data privacy, scalability, and model interpretability are examined, alongside potential solutions. The conclusion underscores the transformative potential of ACEM in revolutionizing education.

Adaptive Depth Estimation via SoftHebbLayer in a Hybrid ResNet-UNet Architecture

Abstract. Depth estimation is a key task in computer vision, critical for applications such as autonomous navigation and augmented reality. This paper introduces a novel hybrid neural network that combines ResNet and UNet architectures with a SoftHebbLayer, inspired by Hebbian learning principles, to improve depth estimation from RGB images. The ResNet backbone extracts robust hierarchical features, while the UNet decoder reconstructs fine-grained depth maps. The SoftHebbLayer dynamically adjusts feature connections based on co-activation, enhancing the models adaptability to diverse environments. This approach addresses common challenges in depth estimation, including poor generalization and computational inefficiency. We evaluated the model on the DIODE dataset, achieving strong results in Mean Squared Error (MSE), which is 0.0800 and Root Mean Squared Error (RMSE), which is 0.2805, demonstrating improved accuracy in both indoor and outdoor scenes. While the model excels in precision, further refinement is needed to reduce computational overhead and improve performance in challenging environments. This research paves the way for more efficient, adaptable depth estimation models, with potential applications in mobile robotics and real-time edge computing systems.

Satellite Image Restoration via an Adaptive QWNNM Model

Due to channel noise and random atmospheric turbulence, retrieved satellite images are always distorted and degraded and so require further restoration before use in various applications. The latest quaternion-based weighted nuclear norm minimization (QWNNM) model, which utilizes the idea of low-rank matrix approximation and the quaternion representation of multi-channel satellite images, can achieve image restoration and enhancement. However, the QWNNM model ignores the impact of noise on similarity measurement, lacks the utilization of residual image information, and fixes the number of iterations. In order to address these drawbacks, we propose three adaptive strategies: adaptive noise-resilient block matching, adaptive feedback of residual image, and adaptive iteration stopping criterion in a new adaptive QWNNM model. Both simulation experiments with known noise/blurring and real environment experiments with unknown noise/blurring demonstrated that the effectiveness of adaptive QWNNM models outperformed the original QWNNM model and other state-of-the-art satellite image restoration models in very different technique approaches.

Информационно-аналитическая система детектирования движения объектов на пешеходном переходе

To manage pedestrian and vehicle flows at intersections, systems are implemented that use adaptive traffic light models, adjusting time intervals based on the volume of pedestrians and vehicles. These systems include video cameras that monitor road user movements, enhancing real-time traffic control. This paper introduces an information and analytical system for managing transport and pedestrian flows using the YOLO neural model, which enables object recognition. The system performs several operations: converting the original image to gray scale, applying Gaussian blurring, detecting object boundaries through the Canny filter and fuzzy logic for edge detection, and contour processing, where each identified contour is assigned a unique number. The neural network then compares detected contours to the training sample data, determining whether the object is a person or a vehicle. The paper presents experimental results for these algorithms in object recognition. Modified software and images of intersections with pedestrian crossings captured from street video cameras in Kursk were used in the experiments. The recognition accuracy rate from the experiments was 72.4%.

How does turbulence intensity affect the fatigue life of composite airfoils based on existing CFD and experimental data?

This research investigates the effect of airflow turbulence intensity on the fatigue life of composite airfoil structures in aerospace engineering. Using advanced Computational Fluid Dynamics (CFD) simulations, this study provides a comprehensive analysis of how varying levels of turbulence intensity—ranging from mild (5%) to severe (30%)—impact the mechanical degradation and fatigue performance of carbon-fiber-reinforced polymers (CFRPs) used in airfoil designs. The research quantifies the relationship between turbulence levels and changes in stress distribution, highlighting critical points of fatigue crack initiation and subsequent propagation. The study includes detailed comparisons of CFD results to experimental data to validate the predictive models and discusses how localized stress fluctuations under high-intensity turbulence accelerate the onset of fatigue failure. We present significant findings that demonstrate a non-linear relationship between turbulence intensity and the reduction in fatigue life, suggesting optimal design modifications for enhanced durability. Furthermore, this paper explores current challenges in merging CFD data with real-world fatigue testing and proposes advanced techniques such as adaptive mesh refinement and hybrid simulation models to improve predictive accuracy and computational efficiency in future research.

Forecasting of Menstruation using a Calendar-based Method based on a Web Application

Accurately predicting menstrual dates is valuable for women, yet it poses a significant challenge, particularly for those with irregular cycles. This study introduces a web application that utilises a calendar-based method to predict upcoming menstrual dates while offering additional features such as tracking past periods, logging symptoms, and providing personalised health advice. Using the software development life cycle approach, the critical contribution of this work is the development of a user-friendly tool that not only aids in menstrual tracking but also highlights the limitations of calendar-based predictions, especially for women with irregular cycles. Our findings suggest that while the method provides accurate predictions for women with regular cycles, it is less reliable for those with irregular cycles due to its dependence on a fixed cycle length range. This study underscores the need for more adaptive models to improve prediction accuracy for a broader population.

Model-Assisted Probabilistic Neural Networks for Effective Turbofan Fault Diagnosis

A diagnostic method for gas-path faults of turbofan engines, relying on a Probabilistic Neural Network (PNN) coupled with a thermodynamic model of the engine, is presented. The novel aspect of the method is that its training information is generated dynamically by an accompanying Engine Performance Model. In the proposed approach, the PNN efficiently addresses the first step of a diagnostic process (i.e., detection of the faulty component at the current operating point), while with the aid of an adaptive engine model, the fault is then further isolated and identified. A description of the proposed method and training aspects of the PNN are presented. The method is applied to the case of a mixed-flow turbofan engine to diagnose common gas-path faults in compressors and turbines (i.e., fouling, FOD, erosion, and tip clearance). Its performance is evaluated using realistic fault data that may be acquired at various operating conditions within a flight envelope.