Topology information based decision trees to predict dynamic transfer limits and their sensitivities for Hydro-Quebec's network

Abstract

This paper presents a method, using data mining techniques, to find dynamic power transfer limits and sensitivity of the limits due to element variations for the Hydro-Quebec power system based on topology information. The paper illustrates a systematic way to automatically generate a tremendous amount of cases to represent a wide range of parameter variations and compute their corresponding transfer limits based on time-domain dynamic simulations. These transfer limits are used to determine the transfer limit sensitivity for each parameter. Data mining techniques are used to build regression trees on this huge database, generated by the supercomputer at IREQ (the Hydro-Quebec Research institute), to find the relationship between either the transfer limits or the Δlimits and parameter variations. The benefit of this method is its ability to determine limits and/or Δlimits without requiring time-domain simulations in future studies. Rapid access to these limits and the sensitivities of the status of specific elements can potentially be foreseen to aid planning engineers in the planning and execution steps of limit calculations.

Probabilistic Framework for Online Identification of Dynamic Behavior of Power Systems With Renewable Generation

“…The core of the framework is a Decision Forest (DF), consisting of several DTs that perform the online identification of the power system dynamic signature. It has been observed that changes in the network topology can have a significant impact on the performance of DTs [14], [15]. To address this issue, the DF concept is proposed in this paper, consisting of several DTs that are trained and used for different specific network topologies.…”

“…In addition to uncertainties associated with RES operation the impact of topology changes, which may be more frequent in systems with RES, may lead to difficulties in assessing system post fault behavior. The impact of topology changes on the prediction of system stability using DTs was investigated in [14], [15]. Using a number of regression trees [14] or retraining [15] is proposed to deal with topology changes.…”

“…The impact of topology changes on the prediction of system stability using DTs was investigated in [14], [15]. Using a number of regression trees [14] or retraining [15] is proposed to deal with topology changes. In both these cases the aim was to identify the system stability limits only.…”

Improving process representation in conceptual hydrological model calibration using climate simulations

“…Calibration of HSAMI is performed with the hydrological model parameter optimization software EAUptim developed at Hydro‐Québec's Research Institute (IREQ). It runs on IREQ's high‐performance computer CASIR [see Huang et al ., , for details]. EAUptim is a flexible system which establishes the link between a hydrological model and an optimization procedure.…”

Probabilistic Framework for Transient Stability Assessment of Power Systems With High Penetration of Renewable Generation

“…The other important aspect for the overall TSA, addressed in this paper, is the impact of network topology changes on system transient behaviour. This issue has been addressed to a certain extent in the past in systems with conventional generation in [19], [20] and [5] but not in systems with RES where this effect could be more pronounced.…”