A5072522101- Engineering and Test Systems
- Fault Detection and Control Systems
- Software Testing and Debugging Techniques
- Bayesian Modeling and Causal Inference
- VLSI and Analog Circuit Testing
- Evolutionary Algorithms and Applications
- Metaheuristic Optimization Algorithms Research
- Software System Performance and Reliability
- Software Reliability and Analysis Research
- AI-based Problem Solving and Planning
- Advanced Multi-Objective Optimization Algorithms
- Spectroscopy and Chemometric Analyses
- Software Engineering Research
- Smart Agriculture and AI
- Neural Networks and Applications
- Data Quality and Management
- Risk and Safety Analysis
- Machine Learning and Data Classification
- Remote-Sensing Image Classification
- Wheat and Barley Genetics and Pathology
- Integrated Circuits and Semiconductor Failure Analysis
- Complex Network Analysis Techniques
- Domain Adaptation and Few-Shot Learning
- Machine Fault Diagnosis Techniques
- Remote Sensing in Agriculture
Montana State University
2016-2025
Montana State University System
2015-2017
Johns Hopkins University
1997-2015
Boeing (Australia)
2014
Northrop Grumman (United States)
2014
IEEE Computer Society
2014
Missouri University of Science and Technology
2014
Employment Agency
2010
University College Dublin
2003-2007
Maryland Department of Natural Resources
2002-2006
In this paper, we study the behavior of a day-ahead (DA) retail electrical energy market with price-based demand response from air conditioning (AC) loads through hierarchical multiagent framework, employing machine learning approach. At top level hierarchy, retailer agent buys DA wholesale and sells it to consumers. The goal is maximize its profit by setting optimal prices, considering price-sensitive loads. Upon receiving at lower AC agents employ Q-learning algorithm optimize their...
Recently operators of complex systems such as aircraft, power plants, and networks have been emphasizing the need for on-line health monitoring purposes maximizing operational availability safety. The discipline prognostics management (PHM) is being formalized to address information prediction requirements addressing these needs. Herein, we will explore how standards currently under development within IEEE can be used support PHM applications. Particular emphasis placed on role PHM-related...
Particle Swarm Optimization (PSO) has been shown to perform very well on a wide range of optimization problems. One the drawbacks PSO is that base algorithm assumes continuous variables. In this paper, we present version able optimize over discrete This new algorithm, which call Integer and Categorical (ICPSO), incorporates ideas from Estimation Distribution Algorithms (EDAs) in particles represent probability distributions rather than solution values, update modifies distributions. describe...
Factored evolutionary algorithms (FEAs) are a new class of search-based optimization that have successfully been applied to various problems, such as training neural networks and performing abductive inference in graphical models. An FEA is unique it factors the objective function by creating overlapping subpopulations optimize over subset variables function. In this paper, we give formal definition present empirical results related their performance. One consideration using an determining...
In recent years, the use of remotely sensed and on-ground observations crop fields, in conjunction with machine learning techniques, has led to highly accurate yield estimations. this work, we propose further improve prediction task by using Convolutional Neural Networks (CNNs) given their unique ability exploit spatial information small regions field. We present a novel CNN architecture called Hyper3DNetReg that takes multi-channel input raster and, unlike previous approaches, outputs...
Due to uncertainties in generation and load, optimal decision making electrical energy markets is a complicated challenging task. Participating agents the market have estimate bidding strategies based on incomplete public information private assessment of future state market, maximize their expected profit at different time scales. In this paper, we present an agent-based model address problem short-term strategic conventional companies (GenCos) power pool. Based proposed model, each GenCo...
Abstract The majority of the clinico-pathological variability observed in patients harboring a repeat expansion C9orf72-SMCR8 complex subunit ( C9orf72 ) remains unexplained. This expansion, which represents most common genetic cause frontotemporal lobar degeneration (FTLD) and motor neuron disease (MND), results loss expression generation RNA foci dipeptide (DPR) proteins. protein itself plays role vesicular transport, serving as guanine nucleotide exchange factor that regulates GTPases. To...
A segment of the field precision agriculture is being developed to accurately and quickly map location herbicide-resistant herbicide-susceptible weeds using advanced optics computer algorithms. In our previous paper, we classified kochia [<italic>Bassia scoparia</italic> (L.) Schrad.] ground-based hyperspectral imaging a support vector machine learning algorithm, achieving classification accuracies up 80%. current work, imaged along with marestail (also called horseweed) [<italic>Conyza...
Hyperspectral imaging systems are becoming widely used due to their increasing accessibility and ability provide detailed spectral responses based on hundreds of bands. However, the resulting hyperspectral images (HSIs) come at cost increased storage requirements, computational time process, highly redundant data. Thus, dimensionality reduction techniques necessary decrease number bands while retaining most useful information. Our contribution is two-fold: First, we propose a filter-based...
Accounting for the effects of test uncertainty is a significant problem in and diagnosis, especially within context built-in test. Of interest here, how does one assess level then utilize that assessment to improve diagnostics? One approach, based on measurement science, treat probability false indication [e.g., built-in-test (BIT) alarm or missed detection] as measure uncertainty. Given ability determine such probabilities, Bayesian approach by extension, prognosis suggests itself. In...
Faulty automotive systems significantly degrade the performance and efficiency of vehicles are often major contributors vehicle breakdown; they result in large expenditures for repair maintenance. Therefore, intelligent health-monitoring schemes needed effective fault diagnosis systems. Previously, we developed a data-driven approach using data-reduction technique, coupled with variety classifiers, In this paper, consider problem fusing classifier decisions to reduce diagnostic errors....
In the original publication [...]
Recently, operators of complex systems such as aircraft, power plants, and networks, have been emphasizing the need for online health monitoring purposes maximizing operational availability safety. The discipline prognostics management (PHM) is being formalized to address information prediction requirements addressing these needs. In this paper, we will explore how standards currently under development within IEEE can be used support PHM applications. Particular emphasis placed on role...
Few mechanisms turn field-specific ecological data into management recommendations for crop production with appropriate uncertainty. Precision agriculture is mainly deployed machine efficiencies and soil-based zonal management, the traditional paradigm of small plot research fails to unite agronomic effective under farmers’ unique field constraints. This work assesses use on-farm experiments applied precision technologies open-source gain local knowledge spatiotemporal variability in...
The authors expand on the form of information flow model they introduced previously, (see ibid., vol.8, no.3, p.16-30 (1991)). Compiling requires three algorithms for determining higher-order relationships. One these, algorithm computing logical closure, helps to simplify modeling task. also introduce a hypothetical antitank missile launcher illustrate concepts and computations presented previously.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML"...
This paper explores the use of evolutionary algorithms (EAs) to formulate additional biases for a probabilistic motion planner known as rapidly exploring random tree (RRT) algorithm in environments with changing obstacle locations. An offline EA is utilized produce bias an filled environment prior rearranging obstacles. It demonstrated that finds reflecting original and improves RRT's efficiency during re-planning small number rearrangements. The (RET) introduced hybrid RRT employing online...
While a great deal of research has been directed towards developing neural network architectures for RGB images, there is relative dearth specifically multi-spectral and hyper-spectral imagery. We have adapted recent developments in small efficient convolutional networks (CNNs), to create CNN architecture capable being trained from scratch classify 10 band using much fewer parameters than popular deep architectures, such as the ResNet or DenseNet architectures. show that this provides higher...