Design of Photoelectric Sensor with Adaptive Filter Based on Fuzzy RBF Neural Network

In the process of photoelectric sensor signal transmission, environmental fluctuations and temperature variations introduce noise. It cause signal distortion and return waveform distortion. This phenomenon poses significant obstacles to accurate signal acquisition and subsequent processing. To remove this obstacle, this paper proposes an approach for filtering design, which is based on the fuzzy radial basis function (RBF). The combination of fuzzy logic and the RBF neural network occurs within the fuzzy RBF framework, enabling it to effectively approximate the objective function. The improved algorithm achieves the separation of useful signals by approximating the interference signals, thereby fulfilling the purpose of filtering and restoring the true waveform. To confirm the effectiveness of the proposed method, this study simulates the filtering design based on the fuzzy RBF algorithm and compares its effect with that of the traditional Butterworth filter. The comparison simulation result demonstrate that the proposed filter can effectively filter out noise signals and exhibits significant advantages over the Butterworth filtering method.