<p>Plate and shell structures are prone to vibration and substantial sound radiation. Inspired by leaf venation patterns, this study proposes a novel adaptive growth method for stiffener optimization to suppress structural sound radiation. The essence of the method is to optimize stiffener sizing and layout by seeking the optimal geometric parameters of discrete beam elements. First, based on the principle of venation growth, a growth competition criterion and a sensitivity genetic mechanism are proposed to guide stiffeners to grow along the direction of the objective function's gradient descent, achieving an adaptive layout of stiffeners. Next, a mathematical model for minimizing sound power is established based on the high-frequency approximation, with stiffener widths as the design variables, to drive the structural size evolution. The sound power sensitivity is then rigorously derived, and the globally convergent method of moving asymptotes is employed to update design variables, ultimately yielding the optimal stiffener configuration for sound radiation performance. Finally, a multi-frequency optimization example on a thin plate demonstrates that the obtained stiffener layout is clear and achieves significant reductions in sound radiation levels alongside improved computational efficiency.</p>