Prediction of pipeline wax formation degree and evaluation of pigging effect based on timing drive

According to the field manual experience to determine the blindness of pigging cycle, in order to avoid the occurrence of excessive pigging phenomenon, based on the slow timing of the wax formation process, based on the study of the relationship between production data parameters and the degree of wax formation, the improved K-means algorithm was used to determine the degree of wax formation in different time series, and the data was introduced into the convolutional neural network (CNN). By using the feature extraction and adaptive learning ability of CNN model, the online update of duration of different degrees of wax formation was realized, and the pigging effect evaluation index model was constructed according to the changes of wax formation grade and pigging cycle, so as to realize the early warning of pigging operation. The results show that different pipelines have different grade time and cycle length in the complete paraffin pigging cycle, which reflects the different degree of paraffin pigging caused by the difference of input parameters. The improved K-means algorithm divides the degree of paraffin pigging into 4 grades; The overall average error of this model is 0.781d, which is smaller than 2.025d and 1.225d of the RNN model and the LSTM model. The pigging effect evaluation index of the pipeline to be evaluated is 0.828, indicating that the pigging effect is good. The research results can provide practical reference for the improvement of pipeline integrity management level.