论文部分内容阅读
针对小波神经网络在故障诊断中的缺点和不足,主要分析小波神经网络与粒子群、遗传算法、模糊逻辑相结合使用的方法和优势。用粒子群优化的小波神经网络,其收敛速度和精度有很大提高;遗传算法具有全局优化的特点,遗传小波神经网络有效地解决了小波神经网络训练速度慢和易陷入局部极值等问题;基于模糊逻辑而提出的模糊小波神经网络具有高精度的逼近能力和很强的泛化能力。
In view of the shortcomings and deficiencies of wavelet neural network in fault diagnosis, this paper mainly analyzes the methods and advantages of using wavelet neural network in combination with particle swarm optimization, genetic algorithm and fuzzy logic. Particle swarm optimization of wavelet neural network, the convergence speed and accuracy have greatly improved; genetic algorithm has the characteristics of global optimization, genetic wavelet neural network effectively solves the wavelet neural network training speed is slow and easy to fall into the local extreme; The fuzzy wavelet neural network based on fuzzy logic has high precision approximation ability and strong generalization ability.