An official website of the United States government
Cone penetrometers (CPTs) are commonly used for characterising the soil properties of centrifuge models; CPT data is useful for interpretation and quality control. This paper describes the development and design of a new robust CPT device for centrifuge testing. The new device consists of a 6mm cone, an outer sleeve, and an inner rod that transmits cone tip forces to a load cell above the ground surface. The design eliminates the need for a custom submerged strain gauge bridge near the tip, significantly reducing cost.A direct comparison was performed between this CPT device and another similar device developed at the University of Cambridge. CPT’s were manufactured using the new design and then shipped to eight different centrifuge facilities, for quality control of similar experiments performed for LEAP (Liquefaction Experiments and Analysis Projects). All the centrifuge tests simulated a 4 m deep deposit of soil, all consisting of Ottawa F-65 sand with relative densities ranging between about 45 to 80%. The results obtained have been extremely valuable as an independent assessment of the density calculated from mass and volume measurements at different laboratories.
more »
« less
A data-driven approach to optimize the design configuration of multi-sleeve cone penetrometer probe attachments
This study uses a data-driven approach to address the complexities associated with research focused multi-sleeve Cone Penetration Test (CPT) devices, particularly focusing on the multi-friction attachment (MFA) and multi-piezo-friction attachment (MPFA) CPT devices. Hindered by time-consuming assembly and susceptibility to sensor stream losses due to extensive electronic components, these advanced devices demand optimization to transform from research devices to practice-suitable devices. This study aims at optimizing the design of the multi-sleeve CPT devices using machine learning, with soil type classification performance as the primary metric for device configuration effectiveness. The research scope centers not on using machine learning for soil classification but on refining the design of multi-sleeve CPT devices. A two-phase data-driven approach is adopted, testing various feature combinations across eight machine learning models. The first phase involves identifying the most suitable model for the dataset, followed by a refinement of features to balance sensor number minimization and soil classification accuracy. The result is a proposed configuration for a multi-sleeve CPT device, simplifying the original design while maintaining robustness, thereby enhancing cost-efficiency and operational effectiveness in geotechnical practice. This work sheds light on how the integration of machine learning can guide the design optimization of geotechnical instruments.
more »
« less
- Award ID(s):
- 1826118
- PAR ID:
- 10521959
- Publisher / Repository:
- Elsevier
- Date Published:
- Journal Name:
- Computers and Geotechnics
- Edition / Version:
- 169
- Volume:
- 169
- Issue:
- C
- ISSN:
- 0266-352X
- Page Range / eLocation ID:
- 106248
- Subject(s) / Keyword(s):
- machine learning cone penetration probe multi-sleeve attachments device optimization feature selection random forest
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
This Work-in-Progress Research paper focuses on digital game-based learning (DGBL), which refers to the use of a virtual environment to support students’ learning. In this exploratory study, we examine how students engage with GeoExplorer, a digital game-based learning environment that simulates Cone-Penetration Testing (CPT), an on-site test used in geotechnical engineering to investigate soil properties that students typically don’t have access to. In GeoExplorer’s CPT activity, students participate in a virtual internship in which they examine several sites with varied types of soil. This paper investigates DGBL environments by leveraging Self-Determination Theory (SDT) to ask the following research questions: (1) How do "freedom" and autonomy within GeoExplorer encourage students’ new emergent learning strategies? and (2) How do emergent learning strategies in GeoExplorer support students’ confidence as they self-guide their learning? Ten open-ended semi-structured interviews were performed with civil engineering students from three U.S.-based institutions. The data are analyzed using narrative analysis and a grounded theory approach. Our preliminary findings indicate that, while GeoExplorer is intended as a complement to in-person learning, it serves both as a complement and supplement to the online learning that helps to engage students during the pandemic. Students share that a felt sense of "freedom" within GeoExplorer encourages them to engage in different emergent learning strategies, such as repetition and trial and error. Students also describe that these emergent learning strategies promote knowledge retention and understanding, and further support their confidence in performing CPT. Our preliminary findings provide opportunities for students to practice autonomy and develop competency – two out of three basic psychological needs in SDT – in their educational processes.more » « less
-
Abstract The incorporation of high‐performance optoelectronic devices into photonic neuromorphic processors can substantially accelerate computationally intensive matrix multiplication operations in machine learning (ML) algorithms. However, the conventional designs of individual devices and system are largely disconnected, and the system optimization is limited to the manual exploration of a small design space. Here, a device‐system end‐to‐end design methodology is reported to optimize a free‐space optical general matrix multiplication (GEMM) hardware accelerator by engineering a spatially reconfigurable array made from chalcogenide phase change materials. With a highly parallelized integrated hardware emulator with experimental information, the design of unit device to directly optimize GEMM calculation accuracy is achieved by exploring a large parameter space through reinforcement learning algorithms, including deep Q‐learning neural network, Bayesian optimization, and their cascaded approach. The algorithm‐generated physical quantities show a clear correlation between system performance metrics and device specifications. Furthermore, physics‐aware training approaches are employed to deploy optimized hardware to the tasks of image classification, materials discovery, and a closed‐loop design of optical ML accelerators. The demonstrated framework offers insights into the end‐to‐end and co‐design of optoelectronic devices and systems with reduced human supervision and domain knowledge barriers.more » « less
-
Abstract Photonic device development (PDD) has achieved remarkable success in designing and implementing new devices for controlling light across various wavelengths, scales, and applications, including telecommunications, imaging, sensing, and quantum information processing. PDD is an iterative, five-step process that consists of: (i) deriving device behavior from design parameters, (ii) simulating device performance, (iii) finding the optimal candidate designs from simulations, (iv) fabricating the optimal device, and (v) measuring device performance. Classically, all these steps involve Bayesian optimization, material science, control theory, and direct physics-driven numerical methods. However, many of these techniques are computationally intractable, monetarily costly, or difficult to implement at scale. In addition, PDD suffers from large optimization landscapes, uncertainties in structural or optical characterization, and difficulties in implementing robust fabrication processes. However, the advent of machine learning over the past decade has provided novel, data-driven strategies for tackling these challenges, including surrogate estimators for speeding up computations, generative modeling for noisy measurement modeling and data augmentation, reinforcement learning for fabrication, and active learning for experimental physical discovery. In this review, we present a comprehensive perspective on these methods to enable machine-learning-assisted PDD (ML-PDD) for efficient design optimization with powerful generative models, fast simulation and characterization modeling under noisy measurements, and reinforcement learning for fabrication. This review will provide researchers from diverse backgrounds with valuable insights into this emerging topic, fostering interdisciplinary efforts to accelerate the development of complex photonic devices and systems.more » « less
-
This paper investigates how negative emotions arising from anxious attachment can be effectively mitigated through the design of interactions and how design-supported attachment stabilization can contribute to well-being. Anxious attachment is a type of attachment that is prominent in individuals who are highly dependent on others and have lower self-esteem in interpersonal relationships (e.g., “I need constant reassurance from my partner to feel secure.”). To illustrate how interactions can be systematically designed to alleviate anxious attachment, the paper presents the development of a self- administered interactive device that supports users in the contexts of underachievement, self- depreciation, and future worries. The development process was informed by a diary study with individuals with anxious attachment that explored daily coping strategies, and a design workshop with design professionals that generated a set of design strategies. A ten-day field evaluation study showed the device’s effectiveness in alleviating negative emotions associated with anxious attachment through three aspects: facilitating explicit emotion awareness, encouraging positive self- perception, and introspecting the problems at hand. These positive effects were more pronounced when experiencing moderate and low intense negative emotions. This paper discusses implications for design practice with future research directions.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1016/j.compgeo.2024.106248
