Joao Gari da Silva Fonseca

Photovoltaic power production forecasts with support vector regression: A study on the forecast horizon

The objective of this study is to verify how different forecast horizons affect the accuracy of a method to forecast photovoltaic power production using support vector regression and numerically predicted weather variables. One year of power production forecasts were done for a 1 MW photovoltaic power plant in Kitakyushu, Japan. Two forecast horizons were evaluated, up to 2 hours ahead and up to 25 hours ahead. The results showed a variation of the accuracy of the forecasts according to the forecast horizon. The root mean square error for 1 year of up-to-2-hours-ahead forecasts was 0.104 MWh; whereas for the up-to-25-hours-ahead it was 0.118 MWh. The mean absolute error was 0.065 MWh for the up-to-2-hours-ahead forecasts and 0.076 MWh for the up-to-25-hours-ahead forecasts. In percentage terms, the root mean square error and the mean absolute error increased 13% and 17%, respectively, with the increase of the forecast horizon.

Impact of forecast error of photovoltaic power output on demand and supply operation in power systems

Increasing the proportion of power generation from renewable energy sources has become increasingly important in Japan since the nuclear accident caused by the 2011 Tohoku earthquake. Photovoltaic (PV) generation in particular has gained much attention in Japan. In general, the supply and demand of electricity are maintained by economic-load dispatching control (EDC), which refers to regulation of the output of conventional (thermal or hydro) power plants to minimize their operational costs. Using the PV generation forecast in EDC, which includes the unit commitment (UC) of conventional power plants, is essential to maintain the economy and reliability of the power system with large-scale integration of PV generation. In this study, we consider an EDC that determines the UC on the basis of the day-ahead PV generation forecast. The frequency and trend of outages and power surpluses due to the forecast errors of the PV power output are evaluated.

Reduction of PV Reverse Power Flow through the Usage of EV's Battery with Consideration of the Demand and Solar Radiation Forecast

This paper presents an evaluation of the usage of electric vehicle battery as an energy storage device for surplus power generated by Photovoltaic power generation system (PV) installed at home. The main goals are the reduction of the reverse power flow caused by PV power generation, and the mitigation of the dependency on the power system grid to charge the EVs battery. We conducted a study based on the forecast of electricity load and solar radiation. The study shows that by using PV-EV system, yhe amount of power bought from power utility can be reduced while keeping. The study shows that by using PV-EV system, the amount of power bought from power utility can be reduced while keeping the reverse power flow level controlled.

Day-ahead scheduling for supply-demand-storage balancing - model predictive generation with interval prediction of photovoltaics

Large-scale penetration of photovoltaic (PV) power generators and storage batteries is expected into the power system in Japan. To maintain the supply-demand balance with energy storage, the optimal power generation and the charge/discharge power of storage batteries can be determined in a manner of the model predictive control of generators. In view of this, this paper addresses a problem of the day-ahead scheduling for the supply-demand-storage balance with explicit consideration of the model predictive power generation. This scheduling is performed by using demand prediction, whose uncertainty is expressed in terms of interval prediction. Formulating the day-ahead scheduling problem as an interval-valued allocation problem, we give a solution to it by taking an approach based on the monotonicity analysis with respect to the optimal solution. Finally, the efficiency of the proposed method is verified through a numerical simulation, where we use an interval prediction of PV power generation constructed by experimental data.