Electrical and Computer Engineering
North Carolina A&T State University
Email: mgorjise AT aggies.ncat.edu
Control Systems - Machine Learning - Computational Statistics
BSc:2005, Iran University of Science and Technology, Control Engineering
MSc: 2008, Iran University of Science and Technology, Control Engineering
PhD: 2011-now, North Carolina A&T State University, Electrical Engineering
My Google Scholar webpage: (publications)
- M. Moradi, M.Gorji Sefidmazgi,A. Homaifar, "Delayed and Hidden Variables Interactions in Gene Regulatory Networks", IEEE 14th International Conference on Bioinformatics and Bioengineering (BIBE), 2014
- M. Gorji Sefidmazgi, M. Sayemuzzaman, A., M.K Jha, and S. Liess, “Trend analysis using non-stationary time series clustering based on the finite element method,” Nonlinear Processes in Geophysics 21 (3), 605-615
- M. Moradi, M. Gorji, A. Homaifar and D. KC, "A Novel Evolutionary Artificial Neural Network with the Application to Chaotic Time-Series Forecasting, "Genetic and Evolutionary Computation Conference, GECCO 2014, July 12-16, 2014, Vancouver, BC, Canada.
- M. Gorji Sefidmazgi, M. Moradi Kordmahalleh, A. Homaifar, and A. Karimoddini, “A Finite Element Based Method for Identification of Switched Linear Systems,” in American Control Conference, Oregon, 2014.
- M. Gorji Sefidmazgi, M. Sayemuzzaman, and A. Homaifar, “Non-stationary Time Series Clustering with Application to Climate Systems,” , Third Annual World Conference on Soft Computing, San Antonio, 2014, vol. 312, pp. 55-63.
- M. Gorji Sefidmazgi, M. Sayemuzzaman, A. Homaifar, M. K Jha, and S. Liess, “Analyzing Temperature regime/trends during 1950-2010 in North Carolina,”, The Third International Workshop on Climate Informatics, Boulder, 2013.
Change Detection in Climate:
The trend analysis is useful to better understand the climate change and variability and analyze low frequency variability of climate. Due to non-stationarity, a time-varying model is needed to represent the multi-dimensional climatic time series. Thus, locating the times of significant changes in the time series is important. Also, it is necessary to find connections between results of such a data-based analysis and physical phenomena generate these changes.
Hybrid systems are mixtures of real-time (continuous) dynamics and discrete events. Continuous dynamics may be represented by a continuous-time control system, such as a linear system. Hybrid systems are important in many real-world problems such as mechanical, biological, data networks. They can be used for modeling the real phenomena that exhibit discontinuous behavior. Moreover, hybrid models can be used to approximate continuous phenomena. For instance, a nonlinear dynamical system can be approximated by switching among various linear model. Our current research focused on the identification of hyybrid systems using input/output pairs of data. Also, how a nonlinear system can be modeled by hybrid systems with nonlinear boundries.