Reliability of vibration predictions Synthesis of predictions and measurements

Conditions of use of this publication The full-text of this report may be used under the condition of a correct and full reference to this publication. Abstract In this report the reliability of vibration predictions in civil engineering is quantified. Emphasis is laid on the vibration predictions for road-and rail traffic and vibrations from building activities such as (sheet)pile driving. Several kinds of prediction techniques were used: expert opinion, very simple empirical models, dedicated models, and FEM. The prediction techniques were applied by four different institutes which are leading in the Dutch practice in vibration prediction: TNO, GeoDelft, Delft University and Holland Railconsult. Predictions were generated for a variety of characteristic situations. The predictions were compared with measurements. Besides the total uncertainty, which can be derived from the difference between prediction and measurement, a breakdown of the uncertainty sources was made. Executive Summary In civil infrastructure where vibrations can arise either during the construction or the exploitation phase it is usual to determine firstly what vibration levels can be expected using a model and then to compare these levels with those supplied in standards or directives. The reliability of these models is presently unknown. Even the question whether a sophisticated model produces more reliable answers than an empirical model cannot be answered. To increase confidence in vibration predictions and to stimulate their use as a basis in decision making processes, it is important that these issues are resolved. Within the framework of Delft Cluster, the project 01.05.02 " reliability vibration predictions and reducing measures " has been set up for this purpose. One of the main goals of this project is to quantify the uncertainty in the vibration levels, which are calculated using mathematical models. The study must also provide insight into which elements of such models contribute dominantly to this uncertainty. With this knowledge it should then be possible to look for means to reduce the uncertainty in the results. The central question in this study concerns the reliability of vibration predictions, using this models. To answer this question, the uncertainty in the predictions has to be analyzed. This uncertainty may result from essentially four sources: 1. incomplete information about the specification of the system; 2. incomplete information about the input and the boundary conditions; 3. simplifications and approximations in the physical modeling; 4. discretizations and approximations in the numerical modeling. Several types of predictions have been analysed:-expert opinion-very …