Physics-guided machine learning from simulated data with different physical parameters

Knowledge and Information Systems - Physics-based models are widely used to study dynamical systems in a variety of scientific and engineering problems. However, these models are necessarily...     

Facile synthesis and characterization of MnO<Subscript>2</Subscript> nanomaterials as supercapacitor electrode materials

MnO₂ nanomaterials are synthesized via calcinations in air at various temperatures. Amorphous MnO₂ masses appear between 100 and 300 °C and nanorods form above 400 °C. Transmission and scanning electron microscopy are used to observe the geometries of each material, with further structural analyses conducted using X-ray photoelectron spectroscopy, X-ray diffraction, and BET method. The electrochemical properties are investigated through galvanostatic charge/discharge cycling, electrochemical impedance spectra, and cyclic voltammetry within a three-electrode test cell filled with 1 mol L^?1 Na₂SO₄ solution. The slightly asymmetric galvanostatic cycling curves suggest that the reversibility of the Faradaic reactions are imperfect, requiring a larger time to charge than discharge. The specific capacitances of each sample are calculated and trends are identified, proving that the samples synthesized at higher temperatures exhibit poorer electrochemical behaviors. The highest calculated specific capacitance is 175 F g^?1 by the sample calcinated at 400 °C. However, the lower temperature samples exhibit more favorable geometric properties and higher overall average specific capacitances. For future research, it is suggested that surface modifications such as a carbon coating could be used in conjunction with the MnO₂ nanorods to reach the electrochemical properties required by contemporary industrial applications.     

Thermodynamic properties and phase equilibria for Pt-Rh alloys

The activity of rhodium in solid Pt-Rh alloys is measured in the temperature range from 900 to 1300 K using the solid-state cell

The effect of load weight on balance control during lateral box transfers

Few studies have endeavoured to measure balance control during manual material handling. This study examined the effects of load weight during a stationary manual material handling task. In total, 36 healthy participants completed 180° lateral transfer tasks of a loaded (5% of body weight) and an unloaded box. The projection of the centre of mass onto the base of support, as measured via a passive-marker 3-D motion analysis system, was used to quantify balance control. Muscle activities of lower extremity muscles were also measured. When moving the loaded box, individuals ventured ≥ 1 cm closer to the edges of the base of support and increased centre of mass movement up to 14%. In addition, muscle electromyographic activity on both sides of the shank increased. In summary, during loaded configurations, vulnerability to loss of balance was increased and individuals appeared to adapt by increasing co-contraction of the shank muscles suggesting increased ankle stiffness. STATEMENT OF RELEVANCE: Industries requiring manual material handling have a particularly high rate of injuries due to falls. This study suggests that larger load weights during lateral material handling tasks adversely affect balance control and may create a vulnerability to imbalance throughout the entire manoeuvre.     

Evolving trees for the retrieval of mass spectrometry-based bacteria fingerprints

In this paper, we investigate the application of Evolving Trees (ET) for the analysis of mass spectrometric data of bacteria. Evolving Trees are extensions of self-organizing maps (SOMs) developed for hierarchical classification systems. Therefore, they are well suited for taxonomic problems such as the identification of bacteria. Here, we focus on three topics, an appropriate pre-processing and encoding of the spectra, an adequate data model by means of a hierarchical Evolving Tree and an interpretable visualization. First, the high dimensionality of the data is reduced by a compact representation. Here, we employ sparse coding, specifically tailored for the processing of mass spectra. In the second step, the topographic information which is expected in the fingerprints is used for advanced tree evaluation and analysis. We adapted the original topographic product for SOMs for ET to achieve a judgment of topography. Additionally we transferred the concept of U-matrix for evaluation of the separability of SOMs to their analog in ET. We demonstrate these extensions for two mass spectrometric data sets of bacteria fingerprints and show their classification and evaluation capabilities in comparison to state of the art techniques.     

Adaptive locks: Combining transactions and locks for efficient concurrency

Transactional memory is being advanced as an alternative to traditional lock-based synchronization for concurrent programming. Transactional memory simplifies the programming model and maximizes concurrency. At the same time, transactions can suffer from interference that causes them to often abort, from heavy overheads for memory accesses, and from expressiveness limitations (e.g., for I/O operations). In this paper we propose an adaptive locking technique that dynamically observes whether a critical section would be best executed transactionally or while holding a mutex lock. The critical new elements of our approach include the adaptivity logic and cost–benefit analysis, a low-overhead implementation of statistics collection and adaptive locking in a full C compiler, and an exposition of the effects on the programming model. In experiments with both micro and macrobenchmarks we found adaptive locks to consistently match or outperform the better of the two component mechanisms (mutexes or transactions). Compared to either mechanism alone, adaptive locks often provide 3-to-10x speedups. Additionally, adaptive locks simplify the programming model by reducing the need for fine-grained locking: with adaptive locks, the programmer can specify coarse-grained locking annotations and often achieve fine-grained locking performance due to the transactional memory mechanisms.     

Composable continuous-space programs for robotic swarms

Programmability is an increasingly important barrier to the deployment of multi-robot systems, as no prior approach allows routine composition and reuse of general aggregate behaviors. The Proto spatial computing language, however, already provides this sort of aggregate behavior programming for non-mobile systems using an abstraction of the network as a continuous-space-filling device. We extend this abstraction to mobile systems and show that Proto can be applied to multi-robot systems with an actuator that turns a vector field into device motion. Proto programs operate on fields of values over an abstract device called the amorphous medium and can be joined together using functional composition. These programs are then automatically transformed for execution by individual devices, producing an approximation of the specified continuous-space behavior. We are thus able to build up a library of simple swarm behaviors, and to compose them together into highly succinct programs that predictably produce the desired complex swarm behaviors, as demonstrated in simulation and on a group of 40 iRobot SwarmBots.