Method and system for collecting and transmitting pollution source real-time monitoring data

The invention discloses a pollution source real-time monitoring data acquisition and transmission method and system. The transmission method comprises the following steps: identifying and diagnosing abnormal data in a data set based on an isolated forest algorithm; and the LSTM neural network is adopted to repair the abnormal data. Using an isolated forest method to complete the recognition and diagnosis of the automatic monitoring equipment on abnormal data; repairing and marking abnormal data by using an LSTM (Long Short Term Memory) neural network; and the transmission system realizes real-time accurate transmission of the system by using an RTP + RTCP transmission mechanism. 本发明公开了一种用于污染源实时监控数据采集传输方法和系统，所述传输方法包括基于孤立森林算法对数据集中的异常数据进行识别和诊断；采用LSTM神经网络对异常数据进行修复。利用孤立森林方法，完成自动监测设备对异常数据的识别和诊断；利用LSTM神经网络完成异常数据的修复和标记；所述传输系统利用RTP+RTCP的传输机制实现系统的实时精准传输。