Hybrid learning algorithm for intelligent short-term load forecasting

Kumluca Topallı, Ayça
Short-term load forecasting (STLF) is an important part of the power generation process. For years, it has been achieved by traditional approaches stochastic like time series; but, new methods based on artificial intelligence emerged recently in literature and started to replace the old ones in the industry. In order to follow the latest developments and to have a modern system, it is aimed to make a research on STLF in Turkey, by neural networks. For this purpose, a method is proposed to forecast Turkey̕s total electric load one day in advance. A hybrid learning scheme that combines off-line learning with real-time forecasting is developed to make use of the available past data for adapting the weights and to further adjust these connections according to the changing conditions. It is also suggested to tune the step size iteratively for better accuracy. Since a single neural network model cannot cover all load types, data are clustered due to the differences in their characteristics. Apart from this, special days are extracted from the normal training sets and handled separately. In this way, a solution is proposed for all load types, including working days, weekends and special holidays. For the selection of input parameters, a technique based on principal component analysis is suggested. A traditional ARMA model is constructed for the same data as a benchmark and results are compared. Proposed method gives lower percent errors all the time, especially for holiday loads. The average error for year 2002 data is obtained as 1.60%.


Adaptive controller based on grid impedance estimation for stable operation of grid-connected inverters under weak grid conditions
Temiz, Hakan; Keysan, Ozan; Demirok, Erhan (Institution of Engineering and Technology (IET), 2020-10-01)
An electric grid having high impedance seen from the connection point is considered as a weak grid and it adversely affects the system stability of grid-tied voltage source inverters in renewable power devices. In this study, an adaptive controller is proposed by configuring the settling time of the phase-locked loop based on the estimated grid impedance. Pseudorandom binary sequence injection and Fourier techniques are carried out for grid impedance estimation. Impedance-based stability analysis is perform...
Transmit beamformer design with a PAPR constraint to trade-off between beampattern shape and power efficiency
Çayır, Ömer; Candan, Çağatay (Elsevier BV, 2020-04-01)
This study examines the effect of peak-to-average power ratio (PAPR) constraint on the transmit beamformer design problem with the goal of establishing a trade-off between the power efficiency (maximizing the average transmitted power) and other metrics such as the power level fluctuation in mainlobe, peak-sidelobe level (PSL), etc. Typically, unimodular weights are utilized in transmit beamforming to maximize the average transmitted power. Yet, unimodular weights maximize the power efficiency at the expens...
Altinoz, Okkes Tolga; KOŞALAY, İLHAN; Gezer, Derya (VSB - Technical University of Ostrava, 2020-06-01)
Speed governors have critical importance on hydroelectric power plants, which are adjusted to the rotating speed of hydroelectric generation based on load demand of the grid. The rotating speed is the main factor to balance power generation and load demand. The well-designed controller is needed to control speed governors with high accuracy. A well-defined model is needed to obtain desired control structure. Therefore, in this study, initially, the mathematical model of a hydroelectric power plant is obtain...
Multiobjective Optimization of the DC-DC Stage of a Module-Integrated Inverter Based on an Efficiency Usage Model
Mirjafari, Mehran; Balog, Robert S.; Turan, Raşit (Institute of Electrical and Electronics Engineers (IEEE), 2014-05-01)
As photovoltaic (PV) energy continues to gain market penetration due to the lower cost of PV modules, attention is shifting to the balance of system. In this paper, a new efficiency evaluation technique based on a usage model that is synthesized from high-temporal operational data of a PV module is investigated, along with a multiobjective design optimization of the dc-dc stage of a module-integrated inverter. The analysis shows that the technique provides more accurate results than the California Energy Co...
Fully Integrated Autonomous Interface With Maximum Power Point Tracking for Energy Harvesting TEGs With High Power Capacity
Tabrizi, Hamed Osouli; Jayaweera, Herath M. P. C.; Muhtaroglu, Ali (Institute of Electrical and Electronics Engineers (IEEE), 2020-05-01)
In this article, a novel fully autonomous and integrated power management interface circuit is introduced for energy harvesting using thermoelectric generators (TEGs) to supply power to Internet of Thing nodes. The circuit consists of a self-starting dc & x2013;dc converter based on a dual-phase charge pump with LC-tank oscillator, a digital MPPT unit, and a 1-V LDO regulator. The novel maximum power point tracking (MPPT) algorithm avoids open-circuit state, and accommodates varying input power and ultra-lo...
Citation Formats
A. Kumluca Topallı, “Hybrid learning algorithm for intelligent short-term load forecasting,” Ph.D. - Doctoral Program, Middle East Technical University, 2003.