Operational Electricity Dispatch Based on Direct Normal Irradiance (DNI) and Load Forecasting: Case Study: STTP with TES system

The rise in the use of concentrated solar power (CSP) systems has drawn attention to the fluctuations that affect the grid due to the variability nature of solar resources especially Direct Normal Irradiance (DNI), the main component of these systems. These variations can cause serious damages to the grid causing sometimes black-outs. In terms of technical operations, the chaotic behavior of DNI makes the dispatch in many cases impossible to provide electricity for the right place at the right time when the need is there. This study proposes an operational method to dispatch electricity based on DNI and load forecasting using statistical and machine learning techniques combined with the dispatch tool provided in SAM (System Advisor Model) software. The case study explored is the Solar Tower with the molten salt combined with Thermal Energy Storage (TES) system used in NOOR 3 Ouarzazate, Morocco. The aim of the proposed method is to reduce the effect of fluctuations on the electrical grid by anticipating them and providing an easier and more accurate operational way of scheduling and making dispatch decisions. The results show a considerable increase in the performance of the simulated grid due to the new proposed dispatch method and the machine learning techniques used.

[1]  Carlos F.M. Coimbra,et al.  On the role of lagged exogenous variables and spatio–temporal correlations in improving the accuracy of solar forecasting methods , 2015 .

[2]  Mohammad H. Alomari,et al.  PVPF tool: an automatedWeb application for real-time photovoltaic power forecasting , 2019, International Journal of Electrical and Computer Engineering (IJECE).

[3]  Mario Graff,et al.  Evolutive Design of ARMA and ANN Models for Time Series Forecasting , 2012 .

[4]  Maher Chaabene,et al.  Neuro-fuzzy dynamic model with Kalman filter to forecast irradiance and temperature for solar energy systems , 2008 .

[5]  Pierre Pinson,et al.  Robust optimisation for self-scheduling and bidding strategies of hybrid CSP–fossil power plants , 2015 .

[6]  José Manuel Bravo,et al.  A novel two-model based approach for optimal scheduling in CSP plants , 2016 .

[7]  R. Saidur,et al.  Application of support vector machine models for forecasting solar and wind energy resources: A review , 2018, Journal of Cleaner Production.

[8]  M. Diagne,et al.  Review of solar irradiance forecasting methods and a proposition for small-scale insular grids , 2013 .

[9]  Ding Liu,et al.  Research on the optimal dispatch of wind power consumption based on combined heat and power with thermal energy storage , 2018 .

[10]  E. Cogliani.  The Role of the Direct Normal Irradiance (DNI) Forecasting in the Operation of Solar Concentrating Plants , 2014 .

[11]  Nan Chen,et al.  Solar irradiance forecasting using spatial-temporal covariance structures and time-forward kriging , 2013 .

[12]  Davorka R. Jandrlic,et al.  SVM and SVR-based MHC-binding prediction using a mathematical presentation of peptide sequences , 2016, Comput. Biol. Chem..