Input measurements, output binary classification, D+ elements, and PowerFactory source file of the test systems

Inertia Estimation of a Power System Area Based on Iterative Equation Error System Identification

“…The fault is cleared by the disconnection of the affected transformer 100 ms later. This event leads to a topology change that affects the elements (5,20) and (5,34) of D + [73] and therefore affects the rotor speed estimation of the generator at bus 34.…”

“…In Figure 4.4 the results of the topology processing at each iteration are reported. As previously discussed, the proposed TI method employs a binary classification, and the results are reported in this format [73]. As can be noticed, the proposed method provides accurate TI along the simulation, correctly detecting that the topology is not changing during the first part of the simulation (output "0000") and detecting the opening of one of the two parallel transformers between buses 20 and 34 (output "1010").…”

“…Before the clearing of the fault, the elements of the D + matrix that affect the rotor speed estimation of generator 30 are (1,2), (1,30) and (1,40) [73], i.e., the relevant frequency measurements to conduct the estimation are those of buses 2, 30 and 40. However, once branch 2-30 is disconnected, the frequency measurements from bus 2 must no longer be taken into account.…”

“…For example, if the number of topology configurations is between 9 and 16, then a 4-digit binary number must be used. The topology configurations considered for the centralized and decentralized case studies in the following sections can be found in [73]. In this work, all the hidden and output layer neurons employ the sigmoid activation function:…”

“…All synchronous generator connection points are equipped with PMUs that measure the bus voltage and the branch current; altogether, a set of 23 PMU measurements is used to feed the DNN. A detailed account of the power system model and the set of measurements can be found in [73].…”