View non-flash version
en-USJuly 2012 en-US en-USwww.sname.org/sname/mt the design process will enable design oces and shipyards to respond intelligently to requirements and deliver an acoustically quiet vessel. These tools should also be capable of performing trade-o studies and of choos -ing the optimal noise control approaches. Normally most treatments have adverse impacts on space, weight, cost, or opera -tion of the vessel. Selection of optimal noise controls will reduce these eects and result in acceptable acoustic environments. If ignored during the design stage, the cost of correcting acoustic problems can be orders of magnitude higher than the expense of acoustic engineering studies and selec -tion of the proper abatement approaches. To facilitate this process, acoustic predic -tion tools are needed to enable convenient analysis of the shipboard noise environment by naval architects and at the same time be used by experienced acoustical engineers to optimize treatments. The most promising approach is through the use of statistical energy anal -ysis (SEA) modeling techniques. ese have been applied to shipboard problems with varying degrees of success. The advan -tage of SEA is its ability to consider, in one framework, a large array of structural ele -ments. With this approach, one can possibly predict noise levels in a multitude of com -partments with contributions transmitted over both the airborne and structureborne paths from every significant acoustic source?including machinery and hydro- acoustic, for example, propulsors. However, shipboard noise and vibra -tion measurements show that the physical size of a source, its location on the struc -ture, and its orientation relative to receiver space or structure have an impact on accu -racy of the prediction. ese factors are not normally considered in a pure SEA model. Furthermore, the near field of a finite- sized source needs to be described in any prediction process. Other issues are the complicated shape of machinery room and other shipboard compartments, noise induced by propulsors and thrusters, and other sources specic for ships, for exam -ple, HVAC and uid piping systems. Along the same lines, various vibra -tion criteria?the ABS Guide for Passenger Comfort on Ships and the International Standard Organization?s (ISO) ISO-6954? have been developed to protect crew and passengers from high vibration levels, to prevent structural fatigue due to cyclic vibration, or to prevent high ship vibration from adversely impacting machinery or equipment. Current nite element analy -sis (FEA) software packages can generally be applied to ship structures to assess the expected vibrations. FEA codes are already widely used for stress, fatigue, and other structural design parameters. Extending these mod -els for use in assessing structural vibration requires careful consideration, as does creating new models for the sole purpose of vibration analysis. In most cases, the vibration analysis has to address crew/ passenger habitability vibration lim -its, structural vibration/stress limits, and machinery/equipment environmental (base) excitation, with analyses being per -formed using natural frequency resonance avoidance and forced response methods. Model sizes can range from localized mod -els of masts and equipment foundations to larger sections of a vessel encompassing multiple living spaces, all the way up to full ship models. e frequency range of interest is typically from 1 Hz to 100 Hz. Noise prediction The most recent SEA algorithms devel -oped specically for the marine eld have shown that they can be used to predict and control noise. is approach facilitates the prediction process and increases the accu -racy over that achieved by existing ?cook books? or commercial SEA software not specically geared toward the unique fea -tures of ship construction. ese algorithms account for the both the universal and unique acoustic fea -tures of marine vessels. A modified or hybrid prototype program for habitability noise would consist of a hybrid SEA mod -eling approach, an acoustic architectural engineering model, and empirical formu -las and databases that have been veried by shipboard tests. e algorithms would need to consider the acoustic impact of many factors that are specic to marine vessels, including water loading of sur -faces; finite size and orientation of mechanical sources in small compart -ments; and hydrodynamic sources like propellers, thrusters, turbulent layer exci -tation, and splash noise. e algorithms would need to be applicable to all types and sizes of vessels, including monohulls, SWATHs, hydrofoils, and catamarans, and to construction materials such as metals and composites. ese later components?the hull and construction materials?represent a com -mon starting point for creating a CAD tool that accurately addresses habitabil -ity noise predictions. Furthermore, noise control treatments can be integrated into the model to determine their impact on the overall noise level. e program uses a versatile graphic user interface, which enables various users, from ship designers to acoustical experts, to feel comfortable operating the program. Treatments can easily be applied to the model and ?what if? case studies can quickly be investigated to nd an optimal treatment taking into account such non- acoustical factors as cost, weight, and space. e result will be lower noise levels on ships, which will improve human factors issues by reducing fatigue in crewmembers and increasing productivity. Typical ship topology and element components, gen -erally the size of compartment decks and bulkheads, are shown in Figure 1. ese elements are hundreds of times larger than the typical elements used in a nite ele -ment model and can quickly be developed with the proper CAD modeling techniques. e advantage of SEA is its ability to consider, in one framework, a large array of structural elements.