【摘要】 为可靠掌握船舶电气设备状态，保证设备的运行安全，提出多源信息融合的船舶电气设备状态识别方法。采用时间序列模型检测并修正船舶电气设备多源历史数据中的连续异常数据和独立异常数据；基于联合卡尔曼滤波算法融合修正后的电气设备多源历史数据，依据融合后的多源数据训练谱聚类和深度神经网络，构建船舶电气设备状态识别网络模型，结合电气设备的实时运行数据，识别船舶电气设备状态。测试结果显示，该方法能够确定数据中的连续异常数据和独立异常数据，并且完成所有异常数据的修正，保证数据的完整性；离散度结果均在0.016以下；能够完成电气设备正常状态、异常状态以及紧急状态的识别，最小均方根误差值均在0.004 4以下，识别效果良好。更多还原

【Abstract】 To reliably grasp the status of ship electrical equipment and ensure its safe operation, a method for identifying the status of ship electrical equipment based on multi-source information fusion is proposed. Using a time series model to detect and correct continuous abnormal data and independent abnormal data in the multi-source historical data of ship electrical equipment; Based on the fusion of corrected multi-source historical data of electrical equipment using the joint Kalman filtering algorithm, a ship electrical equipment status recognition network model is constructed by training spectral clustering and deep neural networks based on the fused multi-source data. Combined with real-time operating data of electrical equipment, the ship electrical equipment status is identified. The test results show that this method can determine the continuous abnormal data and independent abnormal data in the data, and complete the correction of all abnormal data to ensure the integrity of the data; The dispersion results are all below 0.016; Capable of identifying normal, abnormal, and emergency states of electrical equipment, with a minimum root mean square error value below 0.004 4, indicating good recognition performance.更多还原