Inertial, Governor, and AGC/Economic Dispatch Load Flow Simulations of Loss of Generation Contingencies

The simulation of the transients for loss of generation contingencies can be performed using Transient/Mid Term [1 ] or Long Term Simulation [2] packages. The load flow simulation of loss of generation contingencies usually assigns the mismatch caused by the loss of generation contingency to a swing bus. Programs exist that allow the mismatch to be split among all or some subset of the generators, according to participation factors. The participation factors could be proportional to inertia, generation capacity, AGC participation factors, or economic dispatch participation factors. The transients accompanying loss of generation contingen¬ cies are described in classical papers and books [3, 4, 5]. Although load flow programs have been written that allow inertial or governor distribution of the mismatch, use of such programs in planning only appears in regional planning documents that are not open literature. Moreover, there has never been any effort to determine whether the inertial load flow (inertial distribution of mismatch) captures a snapshot of the actual transient/midterm simulation of that contingency. Moreover, there never has been an effort to model the effect of governor deadband on the governor load flow (governor distribution of mismatch). Thus, there is confusion as to the similarity and differences between the inertial and governor load flow. The transient phenomena associated with the loss of generation contingency is described in this paper. The propagation of the disturbance, the time instant at which the inertial load flow captures a snapshot of the transient, and the time instant at which the governor load flow captures a snapshot of the transient are described. A method of deter¬ mining the subset of generators that participate in governor load flow is described based on this discussion of the propagation and evolution of the transient phenomena asso¬ ciated with loss of generation contingencies. Finally, the application of inertial and governor load flow in planning is discussed. A description of the inertial and governor load flow program, developed for EPRI and residing in the EPRI Soft¬ ware Center, is described. Comparisons of the inertial load flow and governor load flow bus angles with a snapshot of the angles at t = 0.25 seconds and t = 10 seconds from EPRI Midterm Stability simulation is presented for different contingencies on a 49 bus test system. The results not only indicate that the inertial and governor load flows are accurate but also that the inertial and governor load flow capture snapshots of the Midterm Stability simulation at particular instants.