A multi-stage Smart Energy Management System under multiple uncertainties: A data mining approach

Highlights

Introducing a new multi-stage SEMS architecture for optimal energy management in MGs considering various resources of uncertainties.

Performing various tasks such as data acquisition/mining/refinement, pattern recognition, learning parameters and offline/online decision making.

Some data mining algorithms have been applied to reduce the huge amount of raw data, recognize patterns for analysis and learn the given parameters.

For handling of uncertainties, using a stochastic scheduling approach, which includes the mean and variance of energy cost, is applied in the optimization process.

Abstract

Smart Energy Management Systems (SEMS) have become indispensable in Micro-Grid (MG) infrastructure for saving energy usage costs and system control considering the time-varying parameters. In this paper, a new multi-stage SEMS architecture is proposed for optimal energy management in MGs considering various resource uncertainties. The proposed SEMS performs various tasks such as data acquisition/mining/refinement, pattern recognition, learning parameters and offline/online decision making. To meet the energy consumption suitably, the multi-objective SEMS operates in multi-stage scheduling problem, i.e. day-ahead, hour-ahead, and real-time markets. Moreover, some data mining algorithms have been applied to reduce the huge amount of raw data, to recognize patterns for analysis, and to learn the given parameters. From the stochastic point of view, the proposed architecture also takes into account the uncertainties of weather conditions, energy consumption and the spot market price in the risk analysis. To handle these uncertainties, a stochastic scheduling approach which includes the mean and variance of energy cost is considered in the optimization process. The simulation results illustrate the efficiency of the proposed SEMS in different case studies.

Keywords

  • Multi-stage energy management;
  • Data mining algorithm;
  • Multi-objective optimization;
  • Offline and online decision making

Be the first to comment

Leave a Reply

Your email address will not be published.


*