Health Monitoring of IGBTs with a Rule-Based Sub-safety Recognition Model Using Neural Networks

IGBTs are used everywhere ranging from aerospace, to transportation systems to the grid but it’s the most fragile device in power electronics. So it’s very critical to evaluate the health state and take advanced and active maintenance measures to avoid the accidents. This paper develops a rule-based sub-safety recognition model using neural networks to evaluate the degradation degree of the IGBTs and determine the health state. The model was validated with two groups of experimental data.

Download Full-text

Transformer-based artificial neural networks for the conversion between chemical notations

Scientific Reports ◽

10.1038/s41598-021-94082-y ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Lev Krasnov ◽

Ivan Khokhlov ◽

Maxim V. Fedorov ◽

Sergey Sosnin

Keyword(s):

Neural Networks ◽

Artificial Neural Networks ◽

Rapid Development ◽

Performance Level ◽

Rule Based ◽

Development Costs ◽

Overall Performance ◽

Artificial Neural

AbstractWe developed a Transformer-based artificial neural approach to translate between SMILES and IUPAC chemical notations: Struct2IUPAC and IUPAC2Struct. The overall performance level of our model is comparable to the rule-based solutions. We proved that the accuracy and speed of computations as well as the robustness of the model allow to use it in production. Our showcase demonstrates that a neural-based solution can facilitate rapid development keeping the required level of accuracy. We believe that our findings will inspire other developers to reduce development costs by replacing complex rule-based solutions with neural-based ones.

Download Full-text

Selection of Materials and Sensors for Health Monitoring of Composite Structures

MRS Proceedings ◽

10.1557/proc-785-d11.1 ◽

2003 ◽

Vol 785 ◽

Cited By ~ 5

Author(s):

Seth S. Kessler ◽

S. Mark Spearing

Keyword(s):

Health Monitoring ◽

Lamb Wave ◽

Composite Structures ◽

Structural Damage ◽

Performance Metrics ◽

Full Range ◽

Transportation Systems ◽

Health Monitoring Systems ◽

The Relationship ◽

Selection Of

ABSTRACTEmbedded structural health monitoring systems are envisioned to be an important component of future transportation systems. One of the key challenges in designing an SHM system is the choice of sensors, and a sensor layout, which can detect unambiguously relevant structural damage. This paper focuses on the relationship between sensors, the materials of which they are made, and their ability to detect structural damage. Sensor selection maps have been produced which plot the capabilities of the full range of available sensor types vs. the key performance metrics (power consumption, resolution, range, sensor size, coverage). This exercise resulted in the identification of piezoceramic Lamb wave transducers as the sensor of choice. Experimental results are presented for the detailed selection of piezoceramic materials to be used as Lamb wave transducers.

Download Full-text

Extraction of organic chemistry grammar from unsupervised learning of chemical reactions

Science Advances ◽

10.1126/sciadv.abe4166 ◽

2021 ◽

Vol 7 (15) ◽

pp. eabe4166

Author(s):

Philippe Schwaller ◽

Benjamin Hoover ◽

Jean-Louis Reymond ◽

Hendrik Strobelt ◽

Teodoro Laino

Keyword(s):

Organic Chemistry ◽

Neural Networks ◽

Chemical Synthesis ◽

Unsupervised Learning ◽

Chemical Reactions ◽

Data Driven ◽

Experimental Task ◽

Rule Based ◽

Atom Mapping ◽

Mapping Information

Humans use different domain languages to represent, explore, and communicate scientific concepts. During the last few hundred years, chemists compiled the language of chemical synthesis inferring a series of “reaction rules” from knowing how atoms rearrange during a chemical transformation, a process called atom-mapping. Atom-mapping is a laborious experimental task and, when tackled with computational methods, requires continuous annotation of chemical reactions and the extension of logically consistent directives. Here, we demonstrate that Transformer Neural Networks learn atom-mapping information between products and reactants without supervision or human labeling. Using the Transformer attention weights, we build a chemically agnostic, attention-guided reaction mapper and extract coherent chemical grammar from unannotated sets of reactions. Our method shows remarkable performance in terms of accuracy and speed, even for strongly imbalanced and chemically complex reactions with nontrivial atom-mapping. It provides the missing link between data-driven and rule-based approaches for numerous chemical reaction tasks.

Download Full-text

Identifying nonclassicality from experimental data using artificial neural networks

Physical Review Research ◽

10.1103/physrevresearch.3.023229 ◽

2021 ◽

Vol 3 (2) ◽

Author(s):

Valentin Gebhart ◽

Martin Bohmann ◽

Karsten Weiher ◽

Nicola Biagi ◽

Alessandro Zavatta ◽

...

Keyword(s):

Experimental Data ◽

Neural Networks ◽

Artificial Neural Networks ◽

Artificial Neural

Download Full-text

Accurate Modeling of Complex Antitoxin Effect of Quercetin Based on Neural Networks

International Journal of Bifurcation and Chaos ◽

10.1142/s0218127419500135 ◽

2019 ◽

Vol 29 (01) ◽

pp. 1950013 ◽

Cited By ~ 1

Author(s):

Changju Yang ◽

Entaz Bahar ◽

Hyonok Yoon ◽

Hyongsuk Kim

Keyword(s):

Experimental Data ◽

Neural Networks ◽

Artificial Neural Networks ◽

Secondary Metabolites ◽

Hela Cell ◽

Protective Effect ◽

Nonlinear Modeling ◽

Inflammatory Activity ◽

High Complexity ◽

Anti Oxidant

A nonlinear modeling of the protective effect of Quercetin (QCT) against various Mycotoxins (MTXs) has a high complexity and is conducted using artificial neural networks (ANNs). QCT is known to possess strong anti-oxidant, anti-inflammatory activity that can prevent many diseases. MTXs are highly toxic secondary metabolites that are capable of causing disease and death in humans and animals. The protective model of QCT against various MTXs (Citrinin, Patulin and Zearalenol) on HeLa cell is built accurately via learning of sparsely measured experimental data by the ANNs. It has shown that the neuro-model can predict the nonlinear protective effect of QCT against MTX-induced cytotoxicity for the measurement of percentage of inhibition of MTXs.

Download Full-text