Table 1 Summary of performance of various methods available in the literature

Hybrid dynamic simulation [15]

Independent of the type, model and complexity of the system

Only applicable to post-fault analysis

Dissipation energy method [16]

Applicable with PMU measurements to determine oscillation energy flow.

Requirement of prior information about the model restricts its use in large interconnected system

Damping torque method [17]

Clear physical meaning, with PMU measurement applicable for implementation in wide area systems

Often requires rotor speed, load angle and electromagnetic torque of generator, which are not directly available

Phasor and energy analysis method [18]

Robust to data quality issues, can detect oscillations from multiple sources

Construction of energy function for a large interconnected power system is quite complex

Oscillation phasors based method [19]

Strong physical meaning as there is no oscillation phasor in case of stationary active power

With the increasing trend of renewable energy integration into the grid, the detection of oscillation source becomes more challenging

Frequency response function method [20], Bayesian Method [21]

Free from load modeling assumptions, robust against measurement noise as well as uncertainties in generator parameters

Detailed dynamic modeling is required, which affects the accuracy in case accurate models are not available

Double Stage Mode Decomposition Method [22]

Independent of the type, model and complexity of the system

Relatively complex and requires high computational time due to signal decomposition in two stages, criteria for selection of most suitable IMF for analysis is missing, consideration of measurement noise is not taken into account

Unknown input observer (UIO) based method [23]

Easier to estimate state variables related to the source

Detailed and accurate system modeling is required