Method and device for monitoring abnormal operation of air cooling unit

The invention provides an air cooling unit operation abnormity monitoring method and device, and relates to the technical field of air cooling units, and the method comprises the steps: training an ELM neural network based on the corresponding relation between the back pressure and the operation frequency, the environment temperature, the environment wind speed and the unit load in historical operation data, so as to obtain a back pressure prediction model; inputting the real-time operation data into a back pressure prediction model to obtain predicted back pressure; based on the corresponding target operation data when the difference value between the predicted back pressure and the real-time back pressure in the real-time operation data is greater than or equal to a preset threshold value, establishing an abnormal threshold value matrix; and determining the position of the abnormal air cooling fan according to a difference value matrix between the abnormal threshold value matrix and a frequency matrix estab