23671 research outputs found
Sort by
Increasing the efficiency of mechanically exfoliating 1T-CrTe2 flakes using the "Easy-Peel" process
No full textPresented to the 24th Undergraduate Research and Creative Activity Forum (URCAF) held in Woolsey Hall, Wichita State University, April 25, 2025.1T-CrTe2 is a room-temperature van der Waals ferromagnet with a Curie temperature (Tc) ? 300 K [1]. This, along with its perpendicular magnetic easy axis for ultrathin thicknesses (< 10 nm) [1], makes 1T-CrTe2 applicable in high-density data storage devices and spintronics. Mechanical exfoliation of 1T-CrTe2 has been shown to be the best method to create large, thin flakes from bulk material. However, producing thin flakes large enough for application is difficult as residual K from KCrTe2 hampers the separation of Te-Cr-Te layers. We present current progress in the development of the "Easy-Peel" method to remove residual K from 1T-CrTe2. Stochiometric amounts of K, Cr, and Te were first sealed under argon in a quartz tube and heated. The resultant KCrTe2 was then stirred in acetonitrile with 0.2g I2 and separated using vacuum filtration. The 1T-CrTe2 was then subjected to additional I2 treatment using the "Easy-Peel" method. Bulk pieces were then mechanically exfoliated using clean tape. The exfoliate was then transferred to a charged glass slide and imaged using optical microscopy. These images were processed using ImageJ image processing software, and each flake's area, length, and breadth were determined. Current statistics demonstrate roughly a 2 times increase in area and breadth between "Non-Easy-Peel" and "Easy-Peel" 1T-CrTe2 prepared under ambient conditions but little difference when prepared under inert conditions. This implies that "Easy-Peel" 1T-CrTe2 is less brittle under ambient conditions, which is consistent with the idea that residual K has been removed.
[1] X. Zhang, et al., Nat Comm.12, 2492 (2021)
Dynamic weight adjusting deep Q-Networks for real-time environmental adaptation
No full textClick on the DOI link to access this article at the publishers website (may not be free).Deep Reinforcement Learning has shown excellent performance in generating efficient solutions for complex tasks. However, its efficacy is often limited by static training modes and heavy reliance on vast data from stable environments. To address these shortcomings, this study explores integrating dynamic weight adjustments into Deep Q-Networks (DQN) to enhance their adaptability. We implement these adjustments by modifying the sampling probabilities in the experience replay to make the model focus more on pivotal transitions as indicated by real-time environmental feedback and performance metrics. We design a novel Interactive Dynamic Evaluation Method (IDEM) for DQN that successfully navigates dynamic environments by prioritizing significant transitions based on environmental feedback and learning progress. Additionally, when faced with rapid changes in environmental conditions, IDEM-DQN shows improved performance compared to baseline methods. Our results indicate that under circumstances requiring rapid adaptation, IDEM-DQN can more effectively generalize and stabilize learning. Extensive experiments across various settings confirm that IDEM-DQN outperforms standard DQN models, particularly in environments characterized by frequent and unpredictable changes. © 2024 IEEE.National Science Foundation, NSF, (2348485, 2426339); National Science Foundation, NSFThe authors Jinghan Zhang, Xinhao Zhang and Kunpeng Liu are supported by NSF 2348485 and NSF 2426339
Nursing: Class of 1979
No full textPersonal and not-profit use only. Contact [email protected] if you have any questions.On photo: left to right - top row: Lawanda Aarnes, Diane Ankenbrandt, Kathleen Arnold, Mary Arnold, Janelle Baizer, Dianne Banka, Brenda Bania, Alice Barsamian, Pat Beckmeyer, Diana Beeler, Tally Bell, Jeanette Biggoose, Janice Bowser, Beth Brackett, Barbara Brown, Unknown (Possibly Beverly Burell)Second row (left to right): Donna Chambers, Mary Chaney, Charles Colbert, Debra Combs, Carol Cook, Candace Cramer, Steven Dees, Arvene DemareeThird row (left to right): Caroline L. Dettbarn, LeAnn Deweese, Connie Dolan, David Dornhoffer, Cris Droege, Janice Eakes, Kim Ekart, Janis Farha, unknown, unknownFourth row (left to right): Michael Green, Stacy Haddock, Ina Hasley, Janet Headrick, Sharon Helm, Lon Hiebert, Carla Hohmann, Kim Huebert, Beckey Johnson, Sheri Johnson, Laura Judilla, Vickie Kelly, Laura Kennedy, Nancy Knackendoffel, Beverly Kolman, UnknownFifth row (left to right): Bonnie Krenning, Stan Lentz, Alyce Love, David Lusk, Carol Manning, Cynthia Masters, Eileen McCarthy, Tamara McNeil, Kathy Mehle, Linda Meili, Leslie Menchetti, Terry Miller, Dana Morton, Judy Navickas, Marcia Navinsky, Mary NevensSixth row (left to right): Deborah O'Connor, Dena Orozco, Kate Orpin, Shirley Orr, Robin O'Rourke, Teresa Parmely, Penny Popp, Roger Prouse, Sandra Qamar, Bonnie Rasmussen, Sandra Ratliff, Mike Reedy, Sharon Reichenberger, Kathy Renek, Patty Richenburg, UnknownSeventh row (left to right): Janice Ridder, Joyce Ridder, Julie Robertson, Kathleen Roesener, Mike Rogers, Sidney Rowe, Guyna Russell, Susan Scheetz, Jo Ellen Schneider, Deborah Schuler, Ruth Shults, Becky Silva, Alan Smith, Lynette Sonntag, Sharon Steinbach, unknownBottom row (left to right): Vicky Swann, Betsy Talbot, Rowena Thoma, Shirley Thompson, Debra Topham, Trudy Unruh, Sharrell Weber, Cindy Wegener, Diane Weilert, Michael Wellemeyer, Rebecca Williams, Shari Williams, Marc Williamson, Rebecca YoungDigitized by University Libraries' Technical Services Institutional Repository & Digitization group
Microfinance and depression: Help-Seeking as a pathway among Mozambican women
No full textThis is an open access article under the CC BY license.Village Savings and Loan Groups (VSLGs) are organized groups that create opportunities for participants to save assets, one of the fast-growing community-based financial capability (FC) approaches to promote economic stability. The FC approach also positively impacts participants' social support and further improves their mental health. Few studies examined the association between VSLG participation and depression status. To fill the knowledge gap, the study examined this linkage using program data from Central Mozambique. The study applied a quasi-experimental design and sampled female VSLG participants and non-participants. Structural Equations Modeling analysis showed that women's VSLG participation ( β = −2.21, p < .001) and help-seeking behaviors ( β = −0.37, p < .001) are inversely correlated with their depression symptom scores. VSLG participation was also positively related to help-seeking behaviors ( β = 1.65, p < .001). Community-based financial capability interventions have the potential to improve women's mental health. Implications for social work practice and research are discussed. © © 2025. Published by Elsevier Ltd
Development of robust machine learning models for predicting flexural strengths of fiber-reinforced polymeric composites
No full textThis article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).Fiber-reinforced composites are widely used in engineering applications due to their excellent physical and chemical properties. However, evaluating their flexural properties using conventional experimental techniques is time-consuming, costly, and limited by material and fabrication variations. This study investigates the potential of machine learning (ML) techniques to predict the flexural properties of fiber-reinforced composites accurately and efficiently. Five ML algorithms—Light gradient boosting regressor (LGBR), Extra tree regressor (ETR), Decision tree regressor (DTR), Histogram-based gradient boosting regressor (HGBR), and Adaptive boosting regressor (ABR)—were employed to predict the flexural strengths using both experimental data generated in-house and data collected from open literature. Including heterogeneous data from both sources enhances the robustness and generalizability of the developed models. The results demonstrate that the extra trees regressor (ETR) achieves excellent accuracy when applied to the heterogeneous dataset, with a coefficient of determination (R2) value of 0.94, MAE of 31.97, and RMSE of 47.64, outperforming the other three models. Furthermore, the in-house experimental data yields even higher prediction accuracy, with the best-performing model achieving an impressive R2 value of 0.99, MAE of 9.53, and RMSE of 13.15. The high prediction accuracy achieved, despite the slight variability in data obtained from the literature, highlights the potential use of ML techniques to streamline the development process and reduce the reliance on extensive experimental testing. These robust models take into consideration important composite production parameters to provide design engineers and research scientists with versatile and efficient tools for the prediction of flexural properties of fiber-reinforced composites and related materials for various industries, including aerospace, defense, energy, biomedical and automotive. © 2025 The Author
Performance and security challenges in next-gen networks: SDN using FFNN as a DDoS mitigation solution
No full textClick on the DOI link to access this article at the publishers website (may not be free).In recent years, there has been a proliferation in the development of next-generation networks, specifically SDN (software defined networks) and IoT (the Internet of Things). However, this proliferation has also led to detrimental growth in cyber terrorism, particularly in the form of various types of DDoS attacks. This research paper evaluates the efficacy of FFNN based back propagation algorithm and Gradient Descent (GD by comparing their performance. We performed a deep analysis of this algorithm and proposed the performance evolution of DDoS attack detection in data set, such as CICIDS2019. We analyze the training, testing, and validation processes of these algorithms using Matrix Laboratory (MATLAB) R2020B. We compare these methods based on various performance parameters like accuracy, precision, recall, F-1 score, and time complexity analysis. We use the big O notation to analyze the time complexity of the optimization algorithm. The GD method provides a linear big O notation, n (o), which is superior to that of other algorithms. A detailed analysis of performance analysis is discussed in the result analysis of the proposed work. A detailed performance analysis is discussed in result section. The proposed assistance aims to identify an improved optimization algorithm for distinguish and mitigating DDoS attacks in the SDN network. © 2025 IEEE
The grounds of Zhuangzi’s hostility to Confucian self-cultivation
No full textClick on the DOI link to access this article at the publishers website (may not be free).The vehemence and stridency of Zhuangzi’s hostility to the Confucian program of self-cultivation cannot be explained simply in terms of his belief that it is misguided or mistaken. The antagonism clearly is grounded in a belief that the program is deeply pernicious and that it must be eradicated. It is not clear, however, precisely why he believes the program is so dangerous. This paper addresses that puzzle. We argue that there are two distinct, albeit closely related, grounds for Zhuangzi’s hostility. He believes: first, that the Confucian program of self-cultivation destroys natural virtue and, second, it makes it impossible for the individual to achieve enlightenment—i.e. to become one with the Way. © 2025 Informa UK Limited, trading as Taylor & Francis Group
Developing and evaluating VR helmet displays for astronauts with HITL and unreal engine
No full textPoster and abstract presented at the FYRE in STEM Showcase, 2025.Research project completed at the Department of Aerospace Engineering and College of Innovation and Design.Virtual Reality (AR) and other immersive technologies have emerged as powerful tools for enhancing astronauts’ performance during space missions. VR system aims to aid situational awareness, decision-making, and operational effectiveness assistance by immersing users in mission-like environments. However, current VR applications often lack usability design and feedback, thus creating a gap in cognitive support and user adaptability. This project explores those gaps by designing, developing, and evaluating user interfaces (UI) into a simulated pressurized rover display.
Using Unreal Engine, we developed a VR prototype interface that presents mission data, environmental conditions, and procedural guidance in a space environment. Unreal engine was selected for its ability to render a high-quality simulation. It can be used to give real-time feedback, interactive support, and closely mimic cognitive and visual demands in space. The use of Unreal Engine can facilitate human-centered testing by capturing a detailed analysis of the user behavior, task performance, and attention distribution.
To test the usability and effectiveness of the interface, we implemented Human-in-the-Loop (HITL) testing with three participants. During the testing, users assess mission-related tasks using the prototype and provide feedback through structured questions and open-minded responses. The key figures are task comprehension, interaction with UI elements, and perceiving through cognitive information. The goal is to determine how well the interface supported real-time decision-making without overwhelming the user.
The results reveal several usability challenges. Those challenges include confusion of the map behavior, nonfunctional zoom feature, and cognitive overload due to excessive abbreviations and unclear menu logic. The participants appreciated the immersive display and checklist clarity but requested a major improvement of alignment between the map and rover behavior. These findings help demonstrate that while VR offers a promising platform for astronaut interface design and testing, there are still many iterations to be made. Our contributions include the development of VR UI, conducting HITL testing, and analyzing the user feedback are the key factors of improving many future missions training scenarios
AI-driven innovations in 3D printing: Optimization, automation, and intelligent control
No full textClick on the DOI link to access this review at the publishers website (may not be free).By greatly increasing automation, accuracy, and flexibility at every step of the additive manufacturing process, from design and production to quality assurance, artificial intelligence (AI) is revolutionizing the 3D printing industry. The integration of AI algorithms into 3D printing systems enables real-time optimization of print parameters, accurate prediction of material behavior, and early defect detection using computer vision and sensor data. Machine learning (ML) techniques further streamline the design-to-production pipeline by generating complex geometries, automating slicing processes, and enabling adaptive, self-correcting control during printing--functions that align directly with the principles of Industry 4.0/5.0, where cyber-physical integration, autonomous decision-making, and human--machine collaboration drive intelligent manufacturing systems. Along with improving operational effectiveness and product uniformity, this potent combination of AI and 3D printing also propels the creation of intelligent manufacturing systems that are capable of self-learning. This confluence has the potential to completely transform sectors including consumer products, healthcare, construction, and aerospace as it develops. This comprehensive review explores how AI enhances the capabilities of 3D printing, with a focus on process optimization, defect detection, and intelligent control mechanisms. Moreover, unresolved challenges are highlighted--including data scarcity, limited generalizability across printers and materials, certification barriers in safety-critical domains, computational costs, and the need for explainable AI