Abstract
The reasons for requiring an accurate prediction of control system performance under actual operating conditions are discussed. It is shown that a measure of the integral square error in response to a stationary time series disturbance meets most requirements.
The analytical approach to calculating integral square error is described and a specific example is given.
Apparatus for generating the disturbance signal and measuring the integral square error is described. It is suitable for use with systems having a dominant time constant between one second and several hours. Experimental results for a pressure control system are compared with the theoretically predicted results.
The disturbance generator can also be used for generating pseudo-random binary sequences.
Get full access to this article
View all access options for this article.
