基于模糊贝叶斯网络的叉装车制动系统故障诊断研究

Research on fault diagnosis of forklift loader braking system based on fuzzy Beyesian network

为解决因缺乏实际数据而无法准确计算叉装车制动系统部件的故障概率问题,提出一种结合模糊集理论和贝叶斯网络的模糊贝叶斯网络故障诊断方法。该方法利用模糊数表达故障发生的可能性,将专家给出的节点故障概率主观语言评判值转换为模糊数,经过解模糊后得到精确值,再利用贝叶斯网络推理进行故障的诊断,提高了贝叶斯网络对模糊信息和不确定信息的处理能力。通过Ge NIe软件对所建立的叉装车制动系统故障诊断模型仿真分析,验证了该方法的有效性。

In order to solve the problem that the diagnosis probability of the forklift loader braking system components is hardly calculated due to lack of actual data,a new diagnosis method of fuzzy Bayesian network combining fuzzy set thoery with Bayesian network is proposed in this paper.This method uses fuzzy numbers to express diagnosis probability,mapping experts' linguistic judgments on nodes' diagnosis probabilities into fuzzy numbers.Fuzzy numbers are de-fuzzied and precise results are obtained.Fault diagnosis is implemented with the inference of Beyesian network on the previous basic.The ability of Beyesian network to deal with fuzzy information and uncertain information is improved with this method.The fault diagnosis model of forklift loader braking system is simulated and analyzed with the software tool of Ge NIe,the results show that this method is valid.