Prof. S. N. Singh delivered an IEEE PES Distinguished Lecture (DL) on ‘Role of forecasting in power system’ on February 20, 2021. The DL event was organized by IEEE PES Perú. It was like an international webinar where 124 participants took part from 18 different countries and 35 cities around the world. Prof. Dr. Sri Niwas Singh focused on the challenges, accuracy, new techniques, and contribution to the role of forecasting in the power system in his lecture.

Prof. S. N. Singh obtained his M. Tech. and Ph.D. in Electrical Engineering from Indian Institute of Technology Kanpur, in 1989 and 1995. Presently, he is a Professor (HAG), Department of Electrical Engineering, Indian Institute of Technology Kanpur, India. He is the Immediate Past Chairman of IEEE India Council. Dr Singh is a Fellow of IEEE (USA), FIET (UK), FNAE, FIE(I), FIETE, and AvH. He received several awards including the Young Engineer Award 2000 of the Indian National Academy of Engineering (INAE), Khosla Research Award of IIT Roorkee, and Young Engineer Award of CBIP New Delhi (India), 1996. Prof Singh is receipt of the Humboldt Fellowship of Germany (2005, 2007) and Otto-Monsted Fellowship of Denmark (2009-10). Prof Singh became the first Asian to receive the 2013 IEEE Educational Activity Board Meritorious Achievement Award in Continuing Education. He also received the INAE Outstanding Teacher Award 2016 and IEEE R10 region (Asia-Pacific) Outstanding Volunteer Award 2016. Dr. Singh is appointed as IEEE Distinguish Lecturer of PES from 2019 and Industry application Society for 2019-2021. He is also a recipient of the NPSC 2020 Academic Excellence Award.


Role of forecasting in power system presented by Prof. Dr. Sri Niwas Singh


Participants from 18 countries and 35 cities around the world.

Dr. Arias Velásquez, IEEE PES Perú Chapter Chair inaugurated the event by welcoming the participants to the first event of 2021. Dr. Mejía Lara, Vice-chair of IEEE PES Perú introduced the keynote speaker, Prof. Singh.

Prof. S. N Singh presented the problems of determining the future values of a time series from current and past values according to the factors of load demand and its effect through time. With a magistral’s webinar, Prof. Singh introduced the main contribution of factors for wind power generation prediction in time series, with several approaches like AR, ARMA, ARIMA, fractional-ARIMA, GARCH, and so on. He also introduced the correlation analysis with linear dependency on the past historical values and limitations due to limitation between non-linear relations in look-ahead time.

Prof. Singh also developed Assisted Wind Power Forecasting Using Feed-Forward Neural networks (Adaptive Wavelet Neural Network AWNN architecture) with the description of the training algorithms. It is necessary for creating competition among the suppliers in the deregulation of electric power systems compared to Feed Forward Neural Network (FFNN) and Autoregressive model (AR). The training process for nonlinear approximation considered the circular variable wind direction input. It is transformed into a direction difference input as a new strategy since “wind direction information with proper transformation helps in reducing the forecast errors” [1]. The case study was developed on a wind farm. The process is a highly stochastic one and non-stationary for very short-term and medium-term forecasting.

Lastly, Dr. Singh described the WT-ARIMA model as a more appropriate one because of its better-conditioned time-series compared to the ill-conditioned time-series of the ARIMA model. He concluded by explaining the different opportunities provided by new algorithms and approaches. He also explained about the multi-objective repository-based constrained non dominated sorting genetic algorithm with preference order ranking (RCNSGA-PO)-based reconfiguration methodology for daily and hourly reconfiguration by minimizing daily energy loss, Energy Not Supplied (ENS), and Cumulative Current Unbalance Factor (CCUF) [2].

The session concluded with a vote of thanks from Dr. Arias Velásquez.

Resources:

  1. B. Kanna, S.N. Singh, Long term wind power forecast using adaptive wavelet neural network, 2016 IEEE Uttar Pradesh section international conference on Electrical, Computer, and electronics engineering (UPCON), 2016, 671-676.
  2. P. Gangwar, S. Chakrabarti, S.N. Singh, Short-term forecasting-based network reconfiguration for unbalanced distribution system with distributed generators, IEEE Transactions on industrial informatics, 16, 7, 2020, 4378-4389.

Ricardo Arias Velásquez, PhD. (SMIEEE)
IEEE PES PERU – Chapter Chair.


Forecasting case study for March 2021with AWNN and FENN


Next events in IEEE PES PERU 2021