GB/T 43945-2024 in English

This standard systematically regulates the application of statistical energy analysis (SEA) in ship cabin noise forecasting for the first time, and is suitable for the prediction of medium and high frequency noise in living cabins, working cabins, etc. The technical framework includes three core modules:

1. Energy balance equation construction (Formula 1)
2. Subsystem parameters calculation (modal density/loss factor)
3. Acoustic-vibration coupling modeling method

1. Model construction specifications

Chapter 5 of the standard stipulates in detail:

• Subsystem division principle: The minimum size of the board subsystem should ensure that the number of modes within the analysis bandwidth is>5
• Special processing requirements: The underwater part needs to consider the fluid load effect (Clause 5.3.5)
• Typical error avoidance: The cabin interior sound absorption coefficient must be included in the sound cavity subsystem (Clause 8.2.9)

2. Parameter acquisition path

Noise optimization of a cruise ship’s residential area

The application of this standard process revealed that:

1. The main deckdining room 125Hz exceeded the standard by 4.2dB(A)
2. The root cause was that the vibration of the main engine was transmitted through the base-deck path
3. The standard was met after adding damping dressing (η increased from 0.2% to 1.5%)

Verification data: The deviation between the actual measured value and the predicted result was <1.5dB(A)

Compared with the international standard ISO 20283-5, this standard has the following innovations:

• Added sandwich panel modal density calculation formulas (7)-(9)
• Refine the T-type/cross-type connection transfer efficiency algorithm (Figure 3-5)
• Appendix B provides a typical equipment source parameter database

Data management

Establish a ship noise parameter database, focusing on the following accumulations:

• Measured values of loss factors of different ship structures
• Typical cabin sound absorption coefficient curves
• Equipment vibration source characteristic spectrum

Technical integration

Suggested combination:

1. SEA and FEM hybrid modeling (low frequency band)
2. VR technology for noise visualization
3. Machine learning to optimize parameter sensitivity