In this study, the reliable control for time-varying systems with random actuator faults and probabilistic nonlinearities is investigated. The system under consideration has the following main features: (1) nonlinearities with new characters. The probability information of nonlinearities belonging to different varying bounds is used; (2) its multi-actuators are subject to various possible faults/failures, and failure rates can vary in some measure; and (3) there are uncertainties in the plant model parameters. Covering these features, a comprehensive model is developed for uncertain time-varying delay systems. By employing the Lyapunov functional method, free-weighting matrix method, and the linear matrix inequality technique, we can obtain several delay-distribution-dependent sufficient conditions to ensure the exponentially mean square stability of the system. Those conditions are characterized in terms of linear matrix inequalities, and the reliable controller feedback gain can be solved by the standard numerical software. A simulation example is presented to show the effectiveness and applicability of the results. Copyright © 2012 John Wiley & Sons, Ltd.