Abstract :
Before it is possible to put a group of motors into service, it is necessary to provide means for generating and transmitting the energy which probably will be required to operate them. To select the proper and economical capacity of such generating and transmitting equipment, requires that some reasonably correct estimate be made of the probable amount and probable time distribution of the combined load due to the motors when in service. When the motors all operate in a definite sequence on a definite time schedule the problem is simple. But, in the great majority of cases, such as tool, hoist, elevator or traction drive, the motors comprising the group do not in general operate in any such connected sequence. They operate more or less at random. It is usual to estimate the probable load under such conditions of random, or approximately random, operation by guess work controlled only by experience and by comparison with load records obtained from similar installations already in operation. In general this method of estimating from records is not satisfactory. In the first place such records, if sufficient for the purpose, are expensive to obtain and are not always available. In the second place, they are generally records of effects and not of causes. Whereas, as is shown in the paper, if the causes are known, the effects can be foretold with surprising accuracy.
