Abstract: The monitoring and early warning of the health status of electric submersible screw pump is crucial in oil and gas production management. The structure of the electric submersible screw pump is different from that of the ground-driven screw pump, leading to different fault types. Thus, the fault diagnosis of electric submersible screw pump is of great significance for improving the operation efficiency of the submersible screw pump and minimizing the damage and production shutdown caused by failure. In this paper, a computation model and prediction method for the health index of electric submersible screw pump wells based on real-time data monitoring are proposed. Specifically, principal component analysis is used to determine the weight of health monitoring parameters of electric submersible screw pump wells. The health degree calculation model is constructed by the ideal state comparison method. The Density-Based Spatial Clustering of Applications with Noise(DBSCAN) algorithm is employed to classify the health degree. On the basis of PCA weight, Long Short-Term Memory(LSTM) neural network is introduced to predict the health index, and the prediction effect of other models is comparatively analyzed. Based on the operation data from oil wells in Huabei Oilfield, the corresponding health index is calculated, which can accurately reflect the equipment failures after verification through field conditions. The equipment status is categorized into healthy, sub-healthy, and fault in terms of 0.7-1, 0.4-0.7, and less than 0.4 to achieve fault warning and alarm. The PCA-LSTM model predicts the health status of multiple wells, presenting superior training speed and prediction effect compared to conventional LSTM and random forest models. The research and field application show that the construction of the health index model exhibits sound characterization accuracy in reflecting the actual health state of equipment, and can effectively guide the quantitative evaluation, fault early warning, operation and maintenance management of the comprehensive operational health state of electric submersible screw pump wells.