the fraction of the total time, T, when the power demand is as described. We have defined this time interval to be dt^; the probability that t falls in that time interval is . The calculated probability that the available power generating capacity shall be greater than some specific value depends strongly on the number, power generating capacity and the reliability of the individual generator on-line in the power pool. These numbers are not constant throughout the year but vary from maintenance interval to maintenance interval; i.e., each machine must be taken off-line (not available for use as standby generation) for 20% of the year. Thus, the installed margin must be calculated for each maintenance interval independent of all the others and the results for all the maintenance intervals combined to give the yearly average. The total required installed generating capacity is that which allows th.e appropriate number of machines to be on-line during each maintenance interval and still allows each machine to be off-line for 20% of the year. The problem of calculating the probability that the available power generating capacity shall be equal to or greater than some specific value during a specific maintenance interval for a general set of power pool characteristics is complex. In order to simplify the problem, we have assumed the power pool to be made up of either (a) n identical machines, each with a generating capacity of 1 GWe and a forced outage probability of .05 or (b) n' identical machines with the same characteristics and one or more SPS with generating capacities of 5 GWe and forced outage probability of .05. The forced outage probability for any piece of equipment is obtained from historical data and is really a composite of the forced outage rate (the probability that the unit will fail in a unit of time) and the average time required to repair the unit. The interpretation of this single number is somewhat ambiguous. It
RkJQdWJsaXNoZXIy MTU5NjU0Mg==