Tutorial Sessions

The WIECON-ECE Conference offers two tutorials on Sunday 16 December 2018.

Tutorial I: Time Series Data Visualization and Prediction using Python

Tutorial II: Use of Data Mining Tools and Cognitive Intelligence in Improving Operational Decisions of Smart Grid under Competitive Restructured Environment

Tutorial I ia a hand-on session.  Participants need to bring their own notebook. Participants can attend any tutorial and may move between tutorials during the session.  For planning purpose, please fill out the following form https://goo.gl/forms/QmD6Ga2imYHXwvQB2.

Tutorial I: Time Series Data Visualization and Prediction using Python
Instructor: Associate Professor Dr. Chantana Chantrapornchai
                     Dept. of Computer Engineering,  Kasetsart University, Thailand
Date and time:  16 December 2018,  10.30-12.30 
Details: 

  •  Introduction to python and pandas:  – Constructor, index(row/column), operation, formatting, import data from file (CSV, date-time reshape, write dataframe to file.
  •  Data cleansing: – Missing data, resampling methods, interpolation, dropping data
  • Data Visualization: – Matplotlib: line, scatter chart, heatmap, bar
  • Prediction with regressions: – Multivariate and time series

Note:  please bring your laptop and install following software and libraries: 

  • Jupyter Notebook (http://jupyter.org/install) and python2
  • Python Libraries: pandas, numpy, matplotlib, statsmodels, pyramid-arima

 Alternative accesses for ones without Python are online jupyter, jupyter lab and  cocalc.

Tutorial II:  Use of Data Mining Tools and Cognitive Intelligence in Improving Operational Decisions of Smart Grid under Competitive Restructured Environment
Instructor: Assoc Prof. Dr. Surekha Deshmukh
                     Electrical Engg Dept, Dean- Industry Institute interaction,
                     PVG’s College of Engg & Tech, Pune , India
Date and time:  16 December 2018,  13.30-15.30
Details: 

The Restructured Power Sector has opened numerous research opportunities in data mining and analytics of power system with use of cognitive intelligence. The power system is rich with variety of randomly, dynamically varying parameters, in huge volume.  This data is a treasure to understand the pulse of real time power system. With advancement in smart metering, smart sensors, smart instrumentation, it is very much possible to collect huge data with minimum time line, but the real challenge is to analyze the data in respect of utilizing it for improving real time operational decisions. The power system data has characteristics of being random, uncertain and volatile. The cognitive intelligent tool has an ability to understand, adopt, incorporate and generalize these features while analyzing it, which gives improved accuracy in data analytics. K-mean and k-NN data mining, short long NN techniques are best suited for analyzing power system data of smart grid.

The aim of this tutorial is to introduce

  • Significance of variety of power system-data with vide features in operational decisions
  • Data Mining and Analytics in Smart Grid
  • Role of Cognitive Intelligence in power system data analytics
  • Real time applications in domains as short term planning, short term forecasting, economics of power generation, real time markets, power trading, renewable energy

The tutorial will cover development of Intelligent tool for various applications with most important practical points such as smoothening techniques of data, sample-strategy in data analytics, optimum data sets, , appropriate error evaluation technique. The tutorial will also present actual case studies of data analytics in association with Maharashtra State Electricity Transmission Company Ltd, Indian Energy Exchange, Suzlon Wind Energy.

 

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>