针对提升机减速器提出了一种决策层融合结构模型应用在齿轮故障诊断中。对各个传感器的信息进行神经网络属性判决,再将各神经网络的属性判决进行决策层融合,此种方法具有较强的灵活性,当传感器数目增加时,只需增加神经网络个数就可。对以上信息融合故障诊断方法经过试验研究,证明了方法应用在提升机减速器诊断系统中,减小了错判和漏判,提高了系统的可靠性,保证了提升机安全运行。 Aiming at the reducer of elevator, a decision-level fusion structure model is put forward for the gear fault diagnosis. The sensor information of each neural network attribute decision, and then determine the attributes of each neural network decision-making fusion, this method has a strong flexibility, as the number of sensors increases, simply increase the number of neural networks can be. After the above information fusion fault diagnosis method has been tested, it proves that the method is applied to the diagnosis system of hoist gear reducer, which reduces the misjudgment and misjudgment, improves the reliability of the system and ensures the safe operation of the hoist.