综合传动装置磨损产生的金属颗粒在润滑油液中均匀混合, 导致装置工作环境的恶化并最终导致装置磨损失效事故的发生。 因此, 实现综合传动装置磨损劣化状态的准确监测和视情维护策略的合理制定对提高装置的可靠性与可维护性具有重要意义。 携带着磨损部位与磨损状态信息的油液光谱与综合传动装置寿命的相互关系反映了装置磨损劣化的分布特征, 使实现基于油液光谱数据的装置劣化建模和维护决策成为可能。 现有综合传动装置视情维护研究是通过油液光谱数据趋势分析结合经验阈值实现的, 没有考虑维护成本、 装备可用度等因素的影响。 鉴于此, 提出基于油液光谱数据的综合传动装置视情维护决策方法。 首先, 针对综合传动装置的历史故障油液光谱数据, 考虑装备寿命与各劣化变量间的相互关系及各劣化变量对装备劣化的贡献程度, 采用Weibull比例风险回归建立了装置的工作寿命模型。 然后, 针对综合传动装置训练演习和执行任务两种使用工况, 分别以最少维护成本、 最大可用度为目标建立了装置的维护决策模型。 与传统的综合传动装置维护决策方法相比, 该方法考虑了维护成本因素和装备可用度因素的影响, 能够根据维护目标有效的制定装置最优维护时间, 为装置的视情维护决策提供了一个客观的量化方法。 最后, 通过对Ch系列综合传动装置历史故障油液光谱数据的实例分析证明了该方法的有效性, 它能够实现综合传动装置视情维护策略的合理制定, 也为其他装备的视情维护决策提供了有益的参考。

The mental debris produced by the wear of power-shift steering transmission(PSST), which is uniformly mixed in lubrication oil, leads to the working environment degradation and the PSST failure afterwards. Therefore, it is essential to monitor the PSST degradation degree and formulate the condition-based maintenance(CBM) strategy, which can help improve the reliability and maintainability of the PSST. The oil spectral data contain wear position and wear state information, and its relationship with the PSST life reflects the distribution of the PSST degradation, which makes the oil spectral data-based degradation modeling and maintenance decision become possible. However, the current CBM studies of PSST are implemented by trend analysis of spectral oil data combined with predetermined threshold, without considering the maintenance costs and the equipment availability. In this paper, the CBM decision method of PSST is presented based on spectral oil data. First, considering the relationship between the PSST life and degradation variables and the contribution rate of each degradation variable to PSST degradation, the life model is established based on Weibull proportional hazards regression using the spectral oil data from historical faults. Then, the maintenance decision model of the PSST is further established with the minimum maintenance cost and maximum availability as the maintenance objectives for the training exercise and the execution task, respectively. Compared with the traditional PSST maintenance decision method, the proposed method takes into account the influence of maintenance cost and equipment availability, which provides an objective quantization scheme for CBM decision that can effectively determine the optimal maintenance time of the PSST according to the maintenance objectives. Finally, the effectiveness of the proposed method is verified by a case study using spectral oil datum from historical faults of several Ch series PSST, and the results indicate that the proposed method provides a reasonable formulation of the PSST maintenance strategy. The proposed method also provides a useful reference for other equipment’s maintenance decision.