The capability to describe the dyna mic waterway environment and corresponding vessel response using time-domain simulation models has increased significantly in recent years. Provided that comprehensive data on the navigation channel, and sophisticated mathematical models are available, numerical investigations of various navigational problems in waterways and the impact of channel/vessel modifications can be carried out effectively and at low coat. However, such investigations require an adequate modeling capability verified with full scale trial data.
There are basically three alternatives available to evaluate the hydrodynamic forces acting on the vessel:
(1) Directly from model testing including measurements of the resistance/propulsion hydrodynamic forces and ship maneuvering forces (coefficients) acting on the vessel in a maneuver. Any model test results suffer from “scale effects”, primarily of the viscous nature because modeling of the viscous phenomenon cannot be accomplished in typical model tests. Scale effects can be minimized by conducting model tests with large models and recalculation of the viscous forces for the prototype vessel. This is a critical item in ship maneuvering performance assessments because the governing hydrodynamic forces acting on a maneuvering ship are of viscous nature.
(2) Indirectly from various semi-empirical approaches typically synthesizing experimental, statistical and analytical methods for a range of ship hull configurations and ship operational conditions. For certain ship types such as tankers or full-sized cargo vessels which have been extensively tested in the past, accuracy of the predictions are comparable to the accuracy of the model test results. In regard to push-tow operations only very limited Information is available.
(3) Indirectly, from full-scale trials (of ten expensive) by applying various techniques for system identification to evaluate resistance/propulsion and maneuvering forces at full scale Reynolds number. The process of statistical system identification (SI) developed for specific engineering applications typically involves inputting a disturbance to the system, and matching output to input in order to identify the makeup of the system. In the case of vessel motions this corresponds to calling for specific rudder/propeller commands, measuring the vessel response and identifying the hydrodynamic coefficients in surge-sway-yaw maneuvering equations using primarily probabilistic estimation methods. The success of the identification, depends heavily on the quality of the measurements (and therefore , instrumentation, its accuracy, weather, and fairing process techniques utilized) and on the number of the unknown parameters to be determined.
The river tows are extensively used in inland waterway systems and consist of a push towboat and some mix of the standard river barges. The hydrodynamics of the typical river tow flotilla is complex and reliable test data are rare.
In recent years the Army Corps of Engineers and U.S.C.G. has sponsored several fundamental studies carried out by Tracor Hydronautics, Inc. (THI) In support of an integrated approach in ship-channel design and navigation problems, see References 1 to 4. The principal goal of these studies has been experimental/analytical prediction data base for an assessment of maneuvering performance of a typical river tow flotilla in a restricted waterway channel. The investigation (1) included extensive model tests of a 15-barge tow in both deep and shallow waters, development of a method for predicting tow hydrodynamic coefficients, simulations of tow behavior as a function of channel dimensions, and effects of channel alterations on navigation performance. Model test results, analyses, and maneuvering predictions for 6 and 25 barge river tows are given in References 2 and 3. Identification of the hydrodynamic coefficients of the 15 barge tow based on the recently developed identification techniques and full-scale trials of tows on Lake Pontchartrain in New Orleans, are presented in Reference 4. To compliment recently conducted model testing of a 6 barge tow, THI has also identified the maneuvering coefficients of a three long by two wide barge assembly from the 1985 trial data, Reference 4. The principal results of these studies are presented in this paper.
