CAJ | 학술논문

针对电力电缆健康状态检测及其数据采集过程中由于漏报、误报而造成电力电缆数据的真实性受损的情况，进行了数据修正技术的研究。研究采用线性自回归方法，给出传感器数据流的预测模型，并提出了一种预测模型自动调整策略，以便在预测误差超过预先设定的阈值时自动调整预测模型。实验结果表明，该研究提高了电力电缆感知数据的质量，降低了光纤检测的缺陷发生率。
침대전력전람건강상태검측급기수거채집과정중유우루보、오보이조성전력전람수거적진실성수손적정황，진행료수거수정기술적연구。연구채용선성자회귀방법，급출전감기수거류적예측모형，병제출료일충예측모형자동조정책략，이편재예측오차초과예선설정적역치시자동조정예측모형。실험결과표명，해연구제고료전력전람감지수거적질량，강저료광섬검측적결함발생솔。
In view of the damaged authentication of the electric power cable data in the process of power cable state detection and data acquisition arising from false negatives or negatives,this paper conducts a research on the data correction technology.In this research,it gives a prediction model for the sensor data stream by using the linear regression method and presents an auto-matic adj usting strategy for automatically adj usting the model when the prediction error exceeds the preset threshold value, thus improving the quality of power cable sensing data and reducing the incidence rate of defects in optical fiber detection.