Data-Driven Trajectory Prediction of Grid Power Frequency Based on Neural Models

Abstract

Frequency in power systems is a real-time information that shows the balance between generation and demand. Good system frequency observation is vital for system security and protection. This paper analyses the system frequency response following disturbances and proposes a data-driven approach for predicting it by using machine learning techniques like Nonlinear Auto-regressive (NAR) Neural Networks (NN) and Long Short Term Memory (LSTM) networks from simulated and measured Phasor Measurement Unit (PMU) data. The proposed method uses a horizon-window that reconstructs the frequency input time-series data in order to predict the frequency features such as Nadir. Simulated scenarios are based on the gradual inertia reduction by including non-synchronous generation into the Nordic 32 test system, whereas the PMU collected data is taken from different locations in the Nordic Power System (NPS). Several horizon-windows are experimented in order to observe an adequate margin of prediction. Scenarios considering noisy signals are also evaluated in order to provide a robustness index of predictability. Results show the proper performance of the method and the adequate level of prediction based on the Root Mean Squared Error (RMSE) index.

Setting and Testing of the Out-of-Step Protection at Mongolian Transmission System

“…In this paper, the model of the SEL-421-5A is used. Three measurement units ("M-I/U"," M-Iab/Ibc/Ica" and" Meas RMS seq" block) are fed by this CT and this VT [3]. The OOS relay consists of the OOS and the power swing detection logic [3].…”

“…Three measurement units ("M-I/U"," M-Iab/Ibc/Ica" and" Meas RMS seq" block) are fed by this CT and this VT [3]. The OOS relay consists of the OOS and the power swing detection logic [3]. Two polygonal zones are defining the power swing detection area ("Zone 6" and "Zone 7" block), one minimum current activation threshold ("I supervision" block), one power swing and OOS detection element ("Out of Step" block) and one-timer associated to the OOS trip signal ("OS Time Delay" block).…”

“…Power systems operating in pseudo-steady-state conditions run typically close to their nominal frequency, and all the electrical variables are inside the operational limits [1,2]. However, some operating conditions can push the electromechanical variables outside acceptable limits and endanger the secure and economic operation of the electrical power system [3]. When the power system is subject to a severe disturbance, the system is not able to reach a new operating equilibrium point.…”

On Short Circuit of Grid-Forming Converters Controllers: A glance of the Dynamic Behaviour

“…Finally, the customer is no longer a passive party in the power system, the integration of embedded generation (EG), the massive deployment of electric vehicles (EV) and electrical energy storage (EES), all those technologies and the changes in the customer behaviour make the costumer a prosumer, an entity who both consumers and produces electricity. There is a common element at the heart of these changes, and that is the massive deployment of power electronic converter interfaced technologies [2], [3]. Modern generation and storage technologies take advantages of the power electronic converter (PEC) to deliver more controllable electricity and the time to interface renewable resources and energy storage.…”

Investigation of Inertia Response and Rate of Change of Frequency in Low Rotational Inertial Scenario of Synchronous Dominated System

“…Hence, this indicator plays an essential relationship in the stability and resilience of the grid [6]. Moreover, the use of RoCoF is not only crucial for the synthetic inertia control [7], but for other applications such as the inertia estimation [8][9][10], the RoCoF sharing for enhancing primary controllers [11], Wide-Area frequency measurements [12], load shading schemes [13], frequency stability margins [14], and possibly, to determine the inertia placement [15].…”