Public C-MAPSS turbofan engine degradation dataset provided by NASA

The C-MAPSS dataset was produced using a model-based simulation program called Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) provided by NASA, which can accurately simulate the degradation scenarios of the turbofan engines under various degradation patterns . The C-MAPSS turbofan engine dataset contains 4 sub-datasets (Datasets #1 to #4) that represent an increasing level of complexity due to the effects of degradation patterns and operating conditions, as shown in Table 2. Each sub-dataset is composed of multivariate temporal data obtained from 24 sensors, and contains one training set and one test set. The training set includes complete run-to-failure data of multiple aero-engines, while the test set includes a number of instances with incomplete data that end prior to failure. The 24 sensors consist of three operation condition monitoring sensors and 21performance monitoring sensors. Each engine starts with different degrees of initial wear and manufacturing variation that is unknown and considered to be healthy. In addition, the actual RUL values of the test engines are also revealed for datasets #1 to #4,

C-MAPSS数据集由美国国家航空航天局（NASA）提供的商用模块化航空推进系统仿真（Commercial Modular Aero-Propulsion System Simulation，C-MAPSS）这一基于模型的仿真程序生成，该程序可精准复现涡扇发动机在多种退化模式下的性能退化场景。C-MAPSS涡扇发动机数据集包含4个子数据集（数据集#1至#4），如表2所示，这些子数据集因退化模式与运行工况的影响，复杂度依次递增。各子数据集均由24个传感器采集的多变量时序数据构成，且分别包含一套训练集与一套测试集。其中，训练集涵盖多台航空发动机从初始健康状态直至故障停机的完整运行至失效数据，测试集则包含多组在发动机发生故障前就已终止采集的不完整时序数据。该24个传感器可划分为3台运行工况监测传感器与21台性能监测传感器。每台发动机均以未知程度的初始磨损与制造偏差启动，该初始状态被判定为健康初始状态。此外，数据集#1至#4的测试集发动机的实际剩余使用寿命（Remaining Useful Life，RUL）值也已公开。

搜集汇总

背景与挑战

背景概述

该数据集是由NASA提供的C-MAPSS涡轮风扇发动机退化数据集，包含4个子数据集，每个子数据集由24个传感器的多变量时间序列数据组成，用于模拟不同退化模式和操作条件下的发动机退化场景。训练集包含完整的运行至故障数据，测试集则包含未达到故障的不完整数据，适用于发动机健康监测和剩余使用寿命预测研究。