Grigore Stamatescu

Grigore Stamatescu

Professor at UPB

Old News

Old(-er) Publications 2019 —— Stamatescu G., Entezari R., Römer K., Saukh O., Deep and Efficient Impact Models for Edge Characterization and Control of Energy Events, 25th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2019, December 4-6, 2019, Tianjin, China. PDF

Popescu D., Stoican F., Stamatescu G., Chenaru O., Ichim L., A Survey of Collaborative UAV-WSN Systems for Efficient Monitoring, Sensors, 2019. PDF

Chitu C., Stamatescu G., Cerpa A., Building Occupancy Estimation using Supervised Learning Techniques, 23rd International Conference on System Theory, Control and Computing, ICSTCC 2019, October 9-11, 2019, Sinaia, Romania. PDF

Nichiforov C., Stamatescu G., Stamatescu I., Fagarasan I., Iliescu S.St., Building Electrical Load Forecasting through Neural Network Models with Exogenous Inputs, 23rd International Conference on System Theory, Control and Computing, ICSTCC 2019, October 9-11, 2019, Sinaia, Romania. PDF

Chitu C., Stamatescu G., Stamatescu I., Sgarciu V., Assessment of Occupancy Estimators for Smart Buildings, 10th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2019, September 18-21, Metz, France. PDF

Nichiforov C., Stamatescu G., Stamatescu I., Arghira N., Fagarasan I., Iliescu S.St., Embedded On-line System for Electrical Energy Measurement and Forecasting in Buildings, 10th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2019, September 18-21, Metz, France. PDF

Popescu D., Stoican F., Ichim L., Stamatescu G., Dragana C., Collaborative UAV-WSN System for Data Acquisition and Processing in Agriculture, 10th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2019, September 18-21, Metz, France. PDF

Calofir V., Nica G., Stamatescu G., Arghira N., Dynamic Nonlinear Modelling of Building Structure Using the Force Analogy Method, 10th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2019, September 18-21, Metz, France. PDF

Stamatescu G., Dragana C., Stamatescu I., Ichim L., Popescu D., IoT-Enabled Distributed Data Processing for Precision Agriculture, 27th Mediterranean Conference on Control and Automation, MED 2019, July 1-4, Akko, Israel. PDF

Nichiforov C., Stamatescu G., Stamatescu I., Făgărăşan I., Evaluation of Sequence-Learning Models for Large-Commercial-Building Load Forecasting. Information 2019, 10, 189. LINK

Stamatescu G., Făgărăşan I., Sachenko A., Sensing and Data-Driven Control for Smart Building and Smart City Systems, Journal of Sensors, vol. 2019, Article ID 4528034, 2019. LINK

Stamatescu G., Stamatescu I., Arghira N., Făgărăşan I., Data-Driven Modelling of Smart Building Ventilation Subsystem, Journal of Sensors, vol. 2019, Article ID 3572019, 2019. LINK

2018

Mihai V., Dragana C., Stamatescu G., Popescu D. and Ichim L., Wireless Sensor Network Architecture based on Fog Computing, 2018 5th International Conference on Control, Decision and Information Technologies (CoDIT), Thessaloniki, 2018, pp. 743-747. PDF

Old(-er) News

Large commercial buildings offer significant economic and environmental incentives for improved energy management under growing urbanization tendencies in smart cities. Current monitoring and automation systems are mostly closed hardware-software solutions with high associated costs. The objective is to develop a new methodology for integration of open source components in the automation of modern buildings. This will bridge wired and wireless communication protocols such as Modbus, BACNet, ZigBee, hardware interfaces and software libraries, structured data representations and learning algorithms for modelling and control.

As the on-board resources of embedded sensing and actuation nodes increase, local extraction of relevant information becomes possible as well as embedded predictive control schemes that use this information. This leads to a flatter automation hierarchy composed of cooperating intelligent periphery and an on-demand cloud layer providing advanced control/optimization as a service. The objective is to develop and validate such approaches in both simulation (MATLAB/Simulink) and small scale pilot deployments (two/three tank system, flexible assembly line).

Modern automation systems increasingly leverage advanced information technologies and protocols for data access and exchange, security, visualisation and reporting. These in turn are being adapted to specific application domains in the industry using significant domain expertise. The objective is to identifiy first the key areas of overlap and differentiation and proposed a targeted approach that models this convergence, also known as IT/OT integration. A demonstrator platform will be built based on COTS technologies and components such as industrial development boards for monitoring and control, in order to validate the feasability of the proposed solutions.

Description: The Information Processing course covers a range of fundamental knowledge on statistics elements, probabilities and data processing, in an engineering context. These include, selectively: statistical indicators, probability distributions, parameter estimation methods and verification of statistical hypotheses and regression methods. Methods for identifying and handling outliers are presented. Rigorous methodologies and best practices for data set processing, classification of data sources and types, including open data sets and public modeling and prediction competitions, preliminary data processing/pre-processing as well as data visualization and presentation techniques towards diverse audiences are discussed. Applications are carried out in interactive Matlab and Python programming environments.

Keywords: applied statistics, probability theory, exploratory data analysis, regression methods, outlier handling, data science