无人直升机尾桨轴承超声信号故障映射模型

无人直升机被广泛应用于军事民用领域中执行高危任务，对其进行健康维护具有重要意义。尾桨轴承是无人直升机尾桨的关键零件，关系到无人直升机的平衡与航向控制。传统基于振动信号的监测诊断方法易受环境噪音干扰，诊断算法也易受噪声混叠影响。为解决以上问题，提出了一种基于超声信号的无人机尾桨轴承故障映射模型。首先，采集轴承不同故障状态下的超声信号。然后，利用自适应噪声完备集合经验模态分解(CEEMDAN)算法将信号分解，对分解后的信号分量计算各类熵值并融合构造特征向量。最后，将特征向量输送到基于粒子群优化算法的最小二乘支持向量机中建立特征向量与故障类型的映射模型，实现故障诊断。该方法在超声信号下对尾桨轴承早期故障诊断具有有效性和敏感性。

Fault Mapping Model of Unmanned Helicopter Tail Rotor Bearing Based on Ultrasound Signal

Unmanned helicopters(UH) are widely applied in military domain and civil domain to execute the high risk mission,it is important to maintain the health of UH.Tail rotor bearing is the key components of a UH and its health status affects the course of the flight and keep the balance.The traditional fault diagnosis methods mainly rely on vibration-based signal analysis.However,the vibration signals are easily contaminated by environment noises and the signal analysis algorithm is also easily influenced by noise interference.To overcome this challenge,a fault mapping model of UH tail rotor bearing based on ultrasound signal is proposed.Firstly,the ultrasound signal is measured and stored with the different health condition.Then,the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is utilized to decompose the ultrasound as the several components which contains the abundant information.And these signals are computed the entropy value to extract the features.Finally,the extracted features are input into the least square support vector machine (LSSVM) based on particle swarm optimization (PSO) to construct the mapping model to realize the fault diagnosis.Experiment analysis verifies the feasibility and effectiveness of the proposed fault mapping model.