Temperature Anomaly Detection by Integrating Local Contrast and Global Contrast

In view of the characteristics of real-time temperature management in cold chain logistics, the paper discusses the technology of RFID and outlier mining. First, the paper puts forward the real-time temperature control system of RFID cold chain logistics, and finds that RFID is more suitable for real-time monitoring of temperature. Then the method of data stream mining is discussed, and it is found that the outlier mining method is more suitable for real-time processing of RFID cold chain data. On this basis, a distributed outlier mining algorithm, QOD, is proposed as the core algorithm of real-time temperature control system, combined with the definition of outliers and the temperature control characteristics of cold chain, which shows the effectiveness of the algorithm. According to its limitations, the pruning strategy is used to demonstrate the neighborhood pruning strategy in detail, and the pruning strategy is used to optimize the QOD algorithm, which improves the algorithm speed. Finally, experiments prove that the performance of the optimized QOD algorithm is improved. Compared with the related algorithm, the analysis shows that the QOD method has some advantages in effectiveness, accuracy and fast response. Finally, the future development direction of RFID cold chain temperature control is prospected.