lectric power systems comprise E a nearly infinite n u m b e r of devices, exhibiting dynamic characteristics in a wide range of bandwidths and with significant nonlinear effects. Historically, the n a t u r e of t h e s e devices, t h e robust configuration of the electric power system, and its loading were s u c h that t h e interaction between devices became relatively unimportant in system perform a n c e several s e c o n d s after disturbances. Greater utilization of electric plant, through heavier system loadings, interconnections, and increasing use of controls, has a t times necessitated predictions of s y s t e m perforIn studying long-term effects, a 11 modeling fidelity detail o f stability-type simulations must be preserved, including such important features as ceilings, non linea rities, and override action mance through simulation extending over a time range of tens of seconds to several minutes. The phenomena occurring over this extended time frame has been referred to as long-term dynamics.There are basically two classes of problems involving long-term dynamics. One is the problem of islanding with significant imbalances between load and generation where prime mover action in response t o frequency deviations is significant. The other concerns problems of insufficient damping and/or synchronizing power, and voltage collapse.The lack of damping (oscillatory instability), lack of synchronizing (aperiodic or steady state instability) or voltage collapse problems could occur not only as the immediate consequence of the contingency, but also over the longer term following the contingency through changes in system conditions due to restoration of loads (tap changer or load control effects), capacitor or reactor switching or redistribution of generation through primary or secondary prime mover control.Investigators have traditionally dealt with long-term effects through intelligent use of load flow and dynamic * Power Technologies, Inc simulation, reinitializing system conditions t o r e p r e s e n t t h e s e long-term changes in load or generation with load flow and then testing the system dynamic performance with the use of simulation programs with detailed models of devices such as excitation systems, stabilizers, governors and turbines, HVDC, and SVC controls. In studying long-term effects, therefore, all the modeling fidelity detail of stability-type simulations must b e p r e s e r v e d , including such important features a s ceilings, nonlinearities, a n d override action (e.g., minimum and maximum excitation limiters, or prime mover load runbacks). The phenomena of concern require simulation fidelity up to about 5 Hz.In studies of long-term dynamics, therefore, begin with and preserve the network and equipment model database normally used for shorter (0 to 10 seconds) stability-type simulations. In cases of disturbances involving significant imbalances between generation and load, prime mover models should include long-term effects such as boiler response.Once the proper model is developed, a...
