This study presents the design and optimization of a spin wave demultiplexer using inverse design methods. The optimization process employs a direct binary search (DBS) algorithm to effectively tailor the device's response to different frequencies in magnonic circuits. The optimized structure successfully differentiates spin wave signals at distinct output ports, demonstrating its effectiveness in frequency demultiplexing. To further validate its practical applicability, a frequency detection test was conducted, showing that the device can recognize multiple frequency components at the outputs. The results highlight the potential of this optimized design for magnonic signal processing and multifunctional wave-based computing. Future work will focus on refining the structure to further improve frequency selectivity and enhance the efficiency of magnonic devices.