Adaptive System Research Articles

An adaptive neuro-fuzzy inference system-based irrigation sprinkler system for dry season farming

In recent years, the management of irrigation systems has emerged as one of the most pressing concerns in the agricultural industry, especially in areas that experience dry seasons. In this research, an adaptive neuro-fuzzy inference system (ANFIS)-based irrigation system that uses a hot and cold sprinkler mechanism is presented. The goal of the system is to reduce the amount of water needed for farming and increase crop output during dry seasons. Adaptive control of water release is achieved via the use of MATLAB and the ANFIS model. This is done in response to changes in soil moisture, ambient temperature, and crop water demand. According to the findings, the suggested system performs noticeably better than conventional irrigation methods in terms of both the amount of water used and the number of crops produced.

Morphological and Physiological Changes of Hydroponic Lettuce Grown in Varying Potassium Concentrations and an Adaptive Lighting Control System

We investigated the growth dynamics of hydroponic lettuce (Lactuca sativa) driven by the influence that potassium (K+) has on crop growth. This study aimed to determine whether increased K+ concentrations under different daily light integrals (DLIs) in a hydroponic system will boost growth of greenhouse lettuce. This study was conducted within a controlled glass greenhouse environment with varying DLIs achieved by integrating an adaptive lighting control system over a 16-hour photoperiod. We used three K+ treatments of 200, 400, or 600 mg⋅L−1 K+ and six DLI lighting treatments of 11.1, 12.9, 14.6, 15.9, 16.9, and 17 mol⋅m−2⋅d−1. We found that increasing K+ did not increase shoot dry weight, leaf area, or specific leaf area with increasing DLIs. Although K+ and DLI had an interacting effect on the root dry weight fraction, leaf chlorophyll content, and quantum yield of photosystem II, the K+ treatments did not increase or decrease with increasing DLIs. The influencing factor was DLI, which led to increases in shoot dry weight and leaf area, whereas a decrease in specific leaf area was observed with increasing DLIs. Ultimately, adding supplemental concentrations of K+ did not enhance lettuce growth, nor did these effects show any increase with increasing DLIs.

Deep learning with improved hybrid neuro-turning for predictive control of flux based on experimental DCMD module design of water desalination system

This study explores two scenarios for optimizing the predictive control of flux in water desalination systems: experimental design of direct contact membrane distillation (DCMD) and artificial intelligence (AI) models. Deep learning networks, specifically Long Short-Term Memory (LSTM), and standalone AI models, including hybrid Adaptive Neuro-Fuzzy Inference System (ANFIS) and Artificial Neural Network (ANN), were evaluated. Evaluation criteria, along with 2D graphical visualization, were used to assess model accuracy. ANFIS-C4 emerged as the standout model, achieving perfect predictive skills with 100 % accuracy and low RMSE of 0.0522 in the training phase, and maintaining high performance in testing with 99.73 % accuracy and RMSE of 0.7121. Similarly, ANN-C4 demonstrated near-perfect accuracy with 100 % accuracy and RMSE of 0.0643 in the training phase and also maintained excellent performance in the testing phase. Despite requiring multiple inputs, ANFIS and ANN show remarkably high capability in predictive accuracy. Among LSTM models, LSTM-C3 achieved the highest training DC of 91.35 %, indicating robust performance in capturing training data variability. LSTM-C2 showed superior generalization with a testing DC of 0.8539, despite using only two input parameters. The narrow distribution of predicted values around the median in violin plots for ANFIS-C4 and ANN-C4, along with their low RMSE, validates their superior performance. The exceptional results of ANFIS-C4 and ANN-C4 highlight their potential in water desalination applications, supporting Sustainable Development Goal 6 and EPA objectives for sustainable water management. This research underscores the importance of advanced computational models in enhancing the efficiency and reliability of water desalination systems, contributing to global efforts in sustainable water resource management.

Investigations of BLDC motor speed characteristics via THD under conventional and advanced hybrid controllers

This project investigates brushless direct current (BLDC) motor speed control through total harmonic distortion (THD) analysis, employing proportional integral (PI), fuzzy logic (FLC), adaptive neuro-fuzzy inference system (ANFIS), and an innovative hybrid ANFIS-PD/PI controller. Considering the vital role of BLDC motors in precision-dependent industries like robotics, electric vehicles, and industrial automation, our primary focus is on understanding BLDC motor operation and recognizing THD's significance as a performance metric. Controllers are meticulously implemented in real-time, fine-tuned, and optimized to achieve desired speed characteristics, incorporating considerations like response time, accuracy, and energy efficiency. The project's core involves THD analysis, quantifying harmonic content in the BLDC motor's speed waveform. This facilitates a comprehensive comparative evaluation of controller performance, assessing their capability to maintain speed stability and influence power quality. The discussion covers the merits and limitations of each controller, with a special emphasis on the hybrid ANFIS-PD/PI controller, seamlessly blending ANFIS adaptability with PD/PI control stability. Results illustrate the hybrid controller's excellence in optimizing BLDC motor speed control, demonstrating superior performance in speed accuracy, disturbance rejection, and THD reduction. These findings drive advancements in motor control technology, providing practical guidance for selecting controllers tailored to specific application requirements. Simulation results can be analyzed using MATLAB/Simulink 2018a Software.

The Dark Side of GVC Resilience: A Complex Adaptive System View

The Dark Side of GVC Resilience: A Complex Adaptive System View

The Dark Side of GVC Resilience: A Complex Adaptive System View

The Dark Side of GVC Resilience: A Complex Adaptive System View

The frequency and function of nucleoprotein-specific CD8+ T cells are critical for heterosubtypic immunity against influenza virus infection.

Cytotoxic T lymphocytes (CTLs) are an important component of the adaptive immune system that clears virus-infected cells or tumor cells. Hence, developing next-generation vaccines that induce or recall CTL responses against cancer and infectious diseases is crucial. However, it is not clear if the frequency, function, or both are essential in conferring protection, as in the case of influenza. In this study, we demonstrate that both CTL frequency and function are crucial for providing heterosubtypic immunity to influenza by utilizing an Ad-viral vector expressing a CD8 epitope only to rule out the role of antibodies, single-cell RNA-seq analysis, as well as adoptive transfer experiments. Our findings have implications for developing T cell vaccines against infectious diseases and cancer.

The coevolution of corporate capabilities in intellectual property management and technological innovation

PurposeIn this study, we investigate how capability in managing intellectual property may be treated as a type of “dynamic capability,” and we seek to understand how, when it is linked to the new technology development capability of a complex technological organization, these two types of dynamic capabilities may coevolve.Design/methodology/approachWe conducted a longitudinal empirical case study of Airbus, incorporating an abductive research methodology that required investigating theory and empirical data concurrently and iteratively. The data, ranging over a period of two decades from 2000 to 2021, was collected from four different sources, including interviews, internal company documents, publicly available information and patent data.FindingsOur main findings are that the capabilities and roles of the Intellectual Property function in Airbus and their interaction with the company's Technology function have indeed influenced the overall innovation strategy of the organization, and that three coevolutionary phases may be identified in the interactive development of the two functions.Research limitations/implicationsOur investigation into how new technology development and intellectual property (IP) management capabilities coevolve within complex technological organizations, exemplified by Airbus, provides significant theoretical and practical implications. Theoretically, it enhances understanding of capability co-development dynamics in complex organizations, particularly in strategic IP management. Practically, it suggests aligning IP strategy with overall corporate objectives and optimizing organizational structures to promote collaboration and efficiency across IP and technology teams. This alignment may foster innovation, maximize the value of intellectual assets and strengthen collaborations, positioning organizations for long-term success in competitive landscapes.Originality/valueThis study makes a fresh contribution to the innovation studies literature by showing how if intellectual property management is treated as a core function of a complex technological organization – rather than simply as a vehicle for protecting new inventions and products after the fact or simply as a constituent part of the organization's legal function – it may contribute proactively to the organization's technological innovation performance. We also address the current gap in the academic literature for a clear understanding of the processes by which different function-specific dynamic capabilities may coevolve in a complex organization operating as part of a dynamic and complex adaptive system.

Integrating constructivist principles in an adaptive hybrid learning system for developing social entrepreneurship education among college students

Education in social entrepreneurship for college students is a gradual and evolving process that shapes their career outlook and skill development. This journey involves employing various strategies to engage with and analyze innovative concepts. Adaptive hybrid learning systems adjust course materials dynamically based on individual students' skills and learning progress, drawing from diverse learning theories and instructional interventions. Empowering students to drive positive social change is a central focus, with social entrepreneurs tackling global challenges such as poverty, unemployment, gender inequality, inadequate education, healthcare, and governance. While some students may find traditional classroom settings inconvenient and prefer the flexibility of hybrid learning, others weigh the risks and benefits of classroom instruction differently. Data Envelopment Analysis (DEA) is utilized for empirical analysis, utilizing early-stage social entrepreneurship education as a performance measure and cultural probabilities as variables. The significance of college students' contributions through social enterprises is increasingly recognized, prompting the development of Convolutional Neural Network (CNN) models to forecast and assess career growth trends. The DEA-CNN framework aims to enhance adaptive hybrid learning systems by evaluating social entrepreneurs' knowledge, skills, and competencies. Social entrepreneurship not only addresses environmental concerns but also enriches cultural diversity. Many college students prioritize practical action over mere ideation, particularly in entrepreneurship.

Автоматизированное технологическое обеспечение и повышение эксплуатационных свойств деталей машин

The article outlines two directions of automated engineering support for the operational properties of machine parts (wear resistance, contact stiffness, etc.). The first direction is a traditional one. It's a two-stage provision of the operational properties of machine parts: in the first stage it is the dimensioning of working surfaces quality of the part that determines the required values of operational properties; in the second stage it is technological provision of quality parameters for the working surfaces of machine parts. The second new direction is a single – stage automated engineering support for the current operational properties of machine parts, which has been actively developed over the past 25 years at the Bryansk State Technical University. It is based on the theoretical and experimental dependences of the relationship between the operational properties of machine parts directly with the processing modes of their working surfaces. Various automated systems of scientific research have been developed to obtain experimental dependencies. An example of such an automated system for studying contact stiffness is given. Adaptive control systems used on various machines for high-performance engineering support aimed at obtaining the required quality parameters of the treated surfaces and their operational properties have been developed. When processing new materials and taking into account the absence of theoretical and experimental data, it is possible to use self-learning technological systems. An example of such a system used for a lathe, is given. All these developments contribute to the creation of the machines with artificial intelligence.

Predictive modeling and optimization of Rockwell hardness of additively manufactured PEEK using RSM, ANFIS and RNN integrated with PSO

The purpose of this research is to investigate the correlation between statistical and machine learning techniques and additive manufacturing, with a specific focus on predicting the Rockwell hardness of FDM-printed polyether ether ketone (PEEK) components. These components have a significant impact on various industries, such as aerospace, biomedical, and automobile. The study analyzes the hardness by conducting experimental analysis of four process parameters, including infill density, layer height, printing speed, and infill pattern. The research utilizes Response Surface Methodology (RSM), Adaptive Neuro-Fuzzy Inference System (ANFIS), and Recurrent Neural Network (RNN) to accurately predict the Rockwell hardness of the printed parts, with an average deviation of less than 5% from the experimental value. The study also investigates how hardness varies with FDM process parameters using contour and surface plots. Furthermore, the study utilizes RNN integrated with the Particle Swarm Optimization (PSO) algorithm to optimize Rockwell hardness. This approach achieved a peak Rockwell hardness value of 66.89 RHN under conditions of 80% infill density, 0.1mm layer height, 25 mm sec−1 printing speed, and an octet infill pattern. Microstructural examinations and test results corroborate the findings derived from parametric analysis and optimization efforts.

Adaptive task migration strategy with delay risk control and reinforcement learning for emergency monitoring

The timely and reliable handling of post-disaster emergency monitoring tasks is crucial for effective rescue operations. UAV-assisted edge computing plays a pivotal role in the rapid deployment of such systems. However, challenges persist due to communication and computation resource bottlenecks when dealing with delay-sensitive monitoring tasks. In dynamic post-disaster environments, effective task scheduling and resource allocation decisions directly impact the system’s ability to process tasks. Therefore, this paper proposes an adaptive task migration decision-making system for emergency monitoring tasks in UAV-assisted edge computing. Firstly, We decomposed the optimization objectives based on the task processing workflow, then devised a stepwise delay risk control and resource recovery mechanism based on early discarding. Secondly, by integrating multi-agent reinforcement learning (MARL), optimal strategies for task offloading, UAV queue scheduling, and communication resource allocation are learned to enhance the decision system’s environmental awareness and maximize the successful completion of emergency monitoring tasks. Simulation experiments demonstrate that the algorithm significantly improves the success rate of migration tasks and data processing capacity, thereby validating its convergence and effectiveness.

Machine learning-based prediction of construction and demolition waste generation in developing countries: a case study.

Data is needed for making informed decisions regarding managing waste in the time of construction and demolition phases of buildings. However, data availability is very limited in most developing countries in the area of waste generation. The objective of this study is to employ an artificial intelligence (AI)-based approach to develop a reliable model for forecasting monthly construction and demolition waste (C&DW) generation in the case study of Tehran, Iran. We have trained different prediction models using various AI algorithms, including multilayer perceptron neural network, radial basis function neural network, support vector machines, and adaptive neuro-fuzzy inference system (ANFIS). According to the findings, all employed AI algorithms demonstrated high prediction performance for C&DW forecasting models. The ANFIS model, with R2 = 0.96 and RMSE = 0.04209, was identified as the model that better represented the observed values of C&DW generation. The better efficiency of the ANFIS model could be due to its effective enhancement of neural networks to model subjective variables based on fuzzy logic capabilities. The developed prediction model can be employed as an efficient tool for policy and decision-making for C&DW management by predicting waste quantities in the future.

Performance Evaluation of Benzyl Alcohol Oxidation with tert-Butyl Hydroperoxide to Benzaldehyde Using the Response Surface Methodology, Artificial Neural Network, and Adaptive Neuro-Fuzzy Inference System Model

Performance Evaluation of Benzyl Alcohol Oxidation with <i>tert</i>-Butyl Hydroperoxide to Benzaldehyde Using the Response Surface Methodology, Artificial Neural Network, and Adaptive Neuro-Fuzzy Inference System Model

Optimizing Fish Feeding with FFAUNet Segmentation and Adaptive Fuzzy Inference System

Efficient and optimized fish-feeding practices are crucial for enhancing productivity and sustainability in aquaculture. While many studies have focused on classifying fish-feeding intensity, there is a lack of research on optimizing feeding, necessitating a precise and automated model. This study fills this gap with a hybrid solution for precision aquaculture feeding management involving segmentation and optimization phases. In the segmentation phase, we used the novel feature fusion attention U-Net (FFAUNet) to accurately segment fish-feeding intensity areas. The FFAUNet achieved impressive metrics: a mean intersection over union (mIoU) of 89.39%, a mean precision of 95.07%, a mean recall of 95.08%, a mean pixel accuracy of 95.12%, and an overall accuracy of 95.61%. In the optimization phase, we employed an adaptive neuro-fuzzy inference system (ANFIS) with a particle swarm optimizer (PSO) to optimize feeding. Extracting feeding intensity percentages from the segmented output, the ANFIS with PSO achieved an accuracy of 98.57%, a sensitivity of 99.41%, and a specificity of 99.53%. This model offers fish farmers a robust, automated tool for precise feeding management, reducing feed wastage and improving overall productivity and sustainability in aquaculture.

Analysis of immunoglobulin organization and complexity in mink (Neovison vison)

Mink are susceptible to viruses such as SARS-CoV-2, H1N1 and H9N2, so they are considered a potential animal model for studying human viral infections. Therefore, it is important to study the immune system of mink. Immunoglobulin (Ig) is an important component of humoral immunity and plays an important role in the body's immune defense. In this study, we described the gene loci structure of mink Ig germline by genome comparison, and analysed the mechanism of expression diversity of mink antibody library by 5′RACE and next-generation sequencing (NGS). The results were as follows: the IgH, Igκ and Igλ loci of mink were located on chromosome 13, chromosome 8 and chromosome 3, respectively, and they had 25, 36 and 7 V genes, 3, 5 and 7 J genes and 10 DH genes, respectively. Mink Ig heavy chain preferred the IGHV1, IGHD2 and IGHJ4 subgroups, κ chain mainly use the IGKV1, IGKJ1 and IGHL4 subgroups, and λ chain mainly use the IGLV3 and IGLJ3 subgroups. Linkage diversity analysis revealed that N nucleotide insertion was the main factor affecting the linkage diversity of mink Igs. On the mutation types of mink Ig Somatic Hypermutation (SHM), the high mutation types of heavy chain were mainly G > A, C > T, T > C, A > G, C > A, G > T, A > C, and T > G; the high mutation types of κ chain were G > A and T > C; and the high mutation types of λ chain were G > A and A > G. The objective of this study was to analyse the loci structure and expression diversity of Ig in mink. The results contribute to our comprehension of Ig expression patterns in mink and were valuable for advancing knowledge in mink immunogenetics, exploring the evolution of adaptive immune systems across different species, and conducting comparative genomics research.

Geometry-modulated all organic 3D printed smart PLA fibers for flextension amplified giant mechanical energy harvesting and Machine learning assisted pressure mapping

Piezoelectret sensors offer a promising avenue for harvesting abundant mechanical energy in the environment, providing a sustainable power solution for low-powered electronics. However, effectively harnessing large impact mechanical energy remains a challenge. In this study, we introduce a novel 3D cylindrical self-powered piezoelectret sensor capable of converting irregular axial impact forces into uniform radial pressures, enabling efficient large-scale impact mechanical energy harvesting. The cylindrical piezoelectret sensor, operating via a flextensional mechanism, was fabricated using 3D printed poly(lactic acid) (PLA) fibers as the piezoelectret layer and poly(3,4-ethylenedioxythiophene) (PEDOT)-coated PLA as the electrode layer. Our cylindrical sensor demonstrates the output voltage, and current of ∼ 11.4 V and 7.2 μA, representing remarkable improvement of 250 % and 160 % respectively, compared to flat fiber membrane device. Furthermore, we leverage IoT connectivity and machine learning techniques to utilize the fabricated robotic sensor for pressure mapping applications. Through machine learning algorithms, we achieve a remarkable accuracy of 100 % in character identification based on sensor data. This work highlights the potential of organic robotic sensors for efficient energy harvesting and intelligent pressure mapping applications, paving the way for the development of sustainable and adaptive sensor systems for various real-world scenarios.

Towards dynamic decision-making in government as service organization: insights from systems thinking

Purpose Service organizations operate in an increasingly complex and uncertain context that makes decision-making challenging. Despite well-recognized changes in the operational context of government as service organization, service literature has given surprisingly limited attention to what these changes imply for organizational decision-making. This study aims to face with the lack of fit of decision-making theorizing with the reality, within which most service practitioners operate, in order to foster the relevance of decision-making in service research and properly approach the false assumptions and misguided instructions for action. Design/methodology/approach To rectify the situation, the purpose of this paper is to advance a more holistic understanding of decision-making in government as service organization. The authors do so by reviewing the sparse, though insightful, prior literature on decision-making in service research and identifying four foundational assumptions of decision-making in the service context, that radically differ from the traditional assumptions of decision-making within the wider management literature. Findings The authors contribute to service research by further advancing the emerging dynamic understanding of decision-making by developing eight systems thinking-informed research propositions and a connected research agenda. In doing so, the paper offers the essential ground work that can revitalize the field of service management and equip it for facing the challenges that government as service organization is encountering in the 21st century. Originality/value The formulated eight research propositions demonstrate that decision-making in a government as service organization occurs within complex adaptive systems composed of multiple subsystems and is characterized by a high degree of unpredictability. It is a process influenced by multiple actors part of the system and subsystems, through multiple feedback loops, where the implications of prior decisions inform the future decisions.

Towards sustainable coastal management: a hybrid model for vulnerability and risk assessment

This paper presents the development of a Hybrid Model (HM) integrated with a Bayesian Network (BN) for comprehensive coastal vulnerability and risk assessment, with a focus on Konyaaltı Beach, Antalya, Turkey. The HM incorporates critical environmental parameters such as wind, waves, currents, and sediment transport to simulate conditions at vulnerable coastal areas and perform risk assessments for storm effects, flooding, and erosion. The model includes submodules for predicting coastal storms, quantifying sediment transport rates, assessing tsunami inundation severity, and categorizing storms based on beach typologies. The Adaptive Neuro-Fuzzy Inference System (ANFIS) is utilized for significant wave height predictions, enhancing the model's accuracy. The integration of hydrodynamic modeling, Bayesian networks, and ANFIS offers a robust framework for assessing coastal vulnerability and informing sustainable management practices. The study's results highlight the necessity for integrated risk management strategies, including adaptive infrastructure design, zoning and land use regulations, ecosystem-based management, and continuous monitoring and model refinement to enhance coastal resilience against dynamic environmental forces. This research provides valuable insights for mitigating the impacts of hazards on urban developments, contributing to the advancement of sustainable coastal management.

Pre- and post-surgery: advancements in artificial intelligence and machine learning models for enhancing patient management in infective endocarditis.

Infective endocarditis (IE) is a severe infection of the inner lining of the heart, known as the endocardium. It is characterized by a range of symptoms and has a complicated pattern of occurrence, leading to a significant number of deaths. IE poses significant diagnostic and treatment difficulties. This evaluation examines the utilization of artificial intelligence (AI) and machine learning (ML) models in addressing information extraction (IE) management. It focuses on the most recent advancements and possible applications. Through this paper, we observe that AI/ML can significantly enhance and outperform traditional diagnostic methods leading to more accurate risk stratification, personalized therapies as well and real-time monitoring facilities. For example, early postsurgical mortality prediction models like SYSUPMIE achieved 'very good' area under the curve (AUROC) values exceeding 0.81. Additionally, AI/ML has improved diagnostic accuracy for prosthetic valve endocarditis, with PET-ML models increasing sensitivity from 59% to 72% when integrated into ESC criteria and reaching a high specificity of 83%. Furthermore, inflammatory biomarkers such as IL-15 and CCL4 have been identified as predictive markers, showing 91% accuracy in forecasting mortality, and identifying high-risk patients with specific CRP, IL-15, and CCL4 levels. Even simpler ML models, like Naïve Bayes, demonstrated an excellent accuracy of 92.30% in death rate prediction following valvular surgery for IE patients. Furthermore, this review provides a vital assessment of the advantages and disadvantages of such AI/ML models, such as better-quality decision support approaches like adaptive response systems on one hand, and data privacy threats or ethical concerns on the other hand. In conclusion, Al and ML must continue, through multi-centric and validated research, to advance cardiovascular medicine, and overcome implementation challenges to boost patient outcomes and healthcare delivery.