Table 1.

Summary of steady-state methods.

Steady-State Methods	Features	Advantages	Shortcomings
Power Change [6,12,13,26,30]	Steady State Variation of Real and Reactive Power, ΔP, ΔQ	High-Power Residential Loads can easily be identified, Low-sampling rate requirement,	Low power appliances overlap in P-Q plane, Poor performance in recognizing Type-II, III and Type-IV loads.
Time and Frequency Domain Characteristics of VI Waveforms [18,32–38]	Higher order Steady-State Harmonics, Irms, Iavg,Ipeak, Vrms, Power factor	Device classes can easily be categorized into resistive, inductive and electronic loads	High sampling rate requirement, Low accuracy for Type-III loads, overlapping features for consumer electronics of Type-I and II category, unable to distinguish between overlapping activation events
V-I Trajectory [39,40]	Shape features of V-I trajectory : asymmetry, looping direction, area, curvature of mean line, self-intersection, slope of middle, segment, area of segments and peak of middle segment	Detail taxonomy of electrical appliances can be formed due to distinctive V-I curves	Sensitive to multi-load operation scenario, computationally intensive, smaller loads have no distinct trajectory patterns
Steady-State Voltage Noise [11,41]	EMI signatures	Motor-based appliances are easily distinguishable as they generate synchronous voltage noise, Detection of simultaneous activation events, Consumer appliances equipped with SMPS can be recognized with high accuracy	Sensitive to wiring architecture, EMI signatures overlap, Not all appliances are equipped with SMPS