Robust and secure electricity grids such as the national and European transmission grid are characterised by their stability, guaranteeing that the system is capable of restoring suitable operating conditions in the event of disruptions, including, among other things, frequency oscillations between approximately 0.16 and 0.24 Hz related to said disruptions.
One of the main reasons for which inter-area oscillations must be kept within certain limits is that the system may not be able to compensate for the disruption in order to restore the suitable status, i.e., it does not have sufficient damping capability, which could also lead to the triggering of numerous protections and probable damage to installations.
Calculating these parameters, in order to check the robustness of the system when it comes to covering demand, is highly complex, so the contribution of each of the technologies provided to the generation mix affects this in a different and non-linear way. In addition to the current situation in which new agents and technologies for generating electricity are connected on a daily basis, investing in methods which accurately establish the safest scenarios for the transmission grid in this field is of vital importance.
Red Eléctrica’s System Reliability Department, a company integrated into Redeia which is responsible for transmission and operation of electricity in Spain, in collaboration with Elewit and AIA, has jointly developed a solution to calculate oscillation damping, a project that seeks to use artificial intelligence to calculate the relationship between each of the configurations and states of the transmission grid and damping, not only in Spain but across Europe as well, with the aim of identifying which combinations are safe for the transmission grid in advance.
Considering that highly valuable data for optimised management of the electrical system is generated daily, studying the viability of this tool for predicting and calculating the system’s status has been proposed. So that it not only learns about what affects damping under the supervision of experts, but also provides highly important information to operators about the contribution of grid components connected at any time to the damping.
Once the project has been completed, we can study the viability of implementing a new algorithm for calculating the damping and influence of variables, and its potential implementation in a software tool. So, this will allow manual calculations to be carried out by technicians, as well as offer predictions made by AI, and allow real accumulated errors to be monitored, so that automatic retraining can be implemented or so that assessments can be carried out as to whether further developments are required.