Total and fractional bedload transport in alpine streams approximated by different surrogate measurement systems

<p>Knowledge about bedload transport in rivers is of high importance for many hydraulic engineering applications, in particular related to flood protection measures. Passive acoustic surrogate measurement techniques provide useful continuous estimates of bedload transport in terms of total mass, as well as for different grain-size classes.</p><p>We compare different surrogate measurement systems regarding their performance in quantifying total and fractional bedload transport in three alpine streams. The investigated measurement systems are the well-established Swiss plate geophone (SPG), an equivalent system in which the geophone sensor is replaced by an accelerometer sensor, and the miniplate accelerometer (MPA) system. The latter is a more recent device and consists of four small square metal plates embedded in elastomere elements. While the signal recorded with the SPG is known to be proportional to the transported bedload mass, we find that the MPA-signal shows a non-linear dependency. In addition, the MPA reacts more sensitively to small grain size classes than the other two systems, indicating a possible alternative to improve the quantification of bedload transport consisting of those classes.</p><p>Based on the raw signal recorded with the SPG and the MPA in a flume experiment, we test the ability of different empirical models to predict the known weight of the impacting particle. We show that it is possible to identify the particle weight with high accuracy with relatively simple models using data of either of the two measurement systems. One remaining challenge is to account for the site-to-site variability in the (amount of) signal caused by the combination of differing numbers of plates in the measurement setup and the lateral transmission of the signal across multiple plates, especially for the SPG system.</p>

Point-of-Care Viscosity Surrogate Measurement Through Utilization of Smartphone Sensors and Custom 3D Printed Design

This project sought to develop a method to provide a clinically meaningful, surrogate measure for viscosity that will help analyze complex biofluids. Goals for this project included precise measurements that differentiate a wide variety of standard viscosities, table-top level of size, and ease-of-use. The design utilized a custom 3D-printed analog of a cone and plate viscometer with an attachment for a smartphone to provide gyroscopic data. The device is currently in the stages of final validation and will ultimately be tested in a 40-patient clinical trial intended to assess efficacy of mucolytic therapy in mechanically ventilated patients.

Improve Aggressive Driver Recognition Using Collision Surrogate Measurement and Imbalanced Class Boosting

Real-time recognition of risky driving behavior and aggressive drivers is a promising research domain, thanks to powerful machine learning algorithms and the big data provided by in-vehicle and roadside sensors. However, since the occurrence of aggressive drivers in real traffic is infrequent, most machine learning algorithms treat each sample equally and prone to better predict normal drivers rather than aggressive drivers, which is our real interest. This paper aims to test the advantage of imbalanced class boosting algorithms in aggressive driver recognition using vehicle trajectory data. First, a surrogate measurement of collision risk, called Average Crash Risk (ACR), is proposed to calculate a vehicle’s crash risk. Second, the driver’s driving aggressiveness is determined by his/her ACR with three anomaly detection methods. Third, we train classification models to identify aggressive drivers using partial trajectory data. Three imbalanced class boosting algorithms, SMOTEBoost, RUSBoost, and CUSBoost, are compared with cost-sensitive AdaBoost and cost-sensitive XGBoost. Additionally, we try two resampling techniques with AdaBoost and XGBoost. Among all algorithms tested, CUSBoost achieves the highest or the second-highest Area Under Precision-Recall Curve (AUPRC) in most datasets. We find the discrete Fourier coefficients of gap as the key feature to identify aggressive drivers.