Romero Pérez, Julio ArielMiguel-Escrig, OscarSánchez Moreno, JoséDormido Bencomo, Sebastián2024-09-092024-09-092024Julio-Ariel Romero-Pérez, Oscar Miguel-Escrig, José Sánchez-Moreno, Sebastián Dormido-Bencomo, Improving the relay feedback identification by using a gain-changing non-linearity. This work was supported by the research project UJI-B2021-45/21I596 from Universitat Jaume I, Spain, and by the State Research Agency under projects PID2020-112658RBI00 and PID2022-1398187OB-I00. IFAC-PapersOnLine, Volume 58, Issue 7, 2024, Pages 418-423, ISSN 2405-8963, https://doi.org/10.1016/j.ifacol.2024.08.098. (https://www.sciencedirect.com/science/article/pii/S2405896324008188)2405-8963https://doi.org/10.1016/j.ifacol.2024.08.098https://hdl.handle.net/20.500.14468/23666This paper proposes the use of the gain-changing non-linearity to improve the estimation of the ultimate point, which is used in many auto-tuning algorithm of PID controllers. The experiments for identification are similar to the relay feedback, therefore some well known advantages of this kind of experiments are maintained while the estimation error is significantly reduced. The identification method is evaluated with a batch of system dynamics, providing suitable results in all cases, which proves the validity of the approach.eninfo:eu-repo/semantics/openAccess12 Matemáticas::1203 Ciencia de los ordenadores ::1203.17 InformáticaartículoIdentification for controlrelay feedback experimentauto-tunePID