2 APPLICATIONS OF PARALLEL-PROCESSING IN POWER-SYSTEM COMPUTATION

This paper discusses performance improvements achieved in two power sy stem software modules through the use of parallel processing technique s. The first software module, EVARISTE, outputs a voltage stability in dicator for various power system situations. This module was designed for extended real-rime use and is therefore required to give guarantee d response times. The second module, MEXICO, assesses power system rel iability and operating costs by simulating a large number of contingen cies for generation and transmission equipment. This module, used for power system planning purposes, uses a Monte-Carlo method to build the various system states, and makes heavy demands on CPU time for runnin g simulations. Like many power system computation packages, both softw are modules are well-suited to coarse-grain parallel processing. The f irst module was parallelized on a distributed-memory machine and the s econd on a shared-memory machine. In this paper, we start by a descrip tion of the parallelization process used in these two cases, then go o n to give details on the performance levels achieved, discussing aspec ts of programming, parameter selection (number of situations processed , number of processors), and machine characteristics (limitations due to interprocessor communications network, for instance).