Aiming at the popularization and application of a real-time monitoring parameter acquisition system of the electric submersible pump (ESP) well, this paper proposes a fault diagnosis method of ESP well operation based on the SPC rule and prior knowledge fusion. Based on the study of parameter variation rules of ESP well, the SPC expansion rule model is established; by analyzing the variation of some typical characteristic parameters of ESP well, combined with SPC expansion rules and expert experience, a priori knowledge of fault diagnosis of ESP well is formed, that is, multiparameter fault analysis table and weight factor; the SPC extended rule model and prior knowledge are fused to establish the fault probability model of ESP well, form the fault diagnosis method of ESP well, develop the online fault diagnosis software of ESP well, and deploy it in 425 ESP wells in a block. Taking five types of tubing leakage, pump wear, shaft breakage, gas influence, and pump plugging as examples, the application process of fault diagnosis method is analyzed. The research and application show that compared with other fault diagnosis methods, this method needs a smaller time window and higher diagnosis accuracy. By setting multiple time windows, this diagnosis method is applied to calculate the fault probability of ESP well in real time, solve the real time and accurate identification of 14 sudden faults and gradual faults, and significantly improve the intelligent diagnosis level of production faults of ESP well.

In-well hydraulic fracture monitoring based on joint low-frequency distributed acoustic sensing (LF-DAS)/distributed temperature sensing (DTS) enables the acquisition of optical fiber mechanical strain data, which reflect fracture propagation and rock deformation during hydraulic fracturing. This paper presents an analytical method to interpret the mechanical strain profile measured by in-well LF-DAS/DTS during the fracturing process based on strain transfer theory and the Sneddon solution for fracture propagation. The analytical method is validated by a numerical model that simulates the strain field induced by fracture propagation. The sensitivity of the fiber strain to key factors, such as fracture geometry parameters and gauge length, is analyzed. The results indicate that compressive strain in the formation adjacent to the propagating fracture remains observable from the mechanical strain profile under the low fiber lag parameter condition. The presented method is applied to analyze the mechanical strain profile measured from a fractured horizontal well. Considering the reactivation of the pre-existing fracture, the location of the fractures is identified, and the fractures’ geometric parameters are inverted. This study provides a quantitative evaluation method for fracture geometry characterization based on joint LF-DAS/DTS fracturing monitoring.

Distributed acoustic sensing (DAS) provides crucial high-resolution monitoring for hydraulic fracturing, yet manually interpreting its massive datasets remains inefficient and subjective. Weak off-target fluid-entry responses often carry critical evidence of inter-stage communication, but they are easily masked by dominant perforation signals and background noise in DAS records, limiting real-time fracture diagnostics and quantitative inversion. To overcome this, we propose a Dual-Branch Transformer-CNN Fusion Network (DTCF-Net) to automatically classify DAS responses. Our model simultaneously extracts transient temporal dependencies from raw one-dimensional waveforms using a Transformer branch, and multiscale time-frequency and structural features from two-dimensional representations using an Inception branch. Through feature-level fusion, the network classifies signals into strong, weak or off-target, and background responses. Testing demonstrates that DTCF-Net achieves 97.7% accuracy and a 95.5% macro F1 score. Furthermore, an optimized configuration removing computationally expensive two-dimensional features accelerates processing speed by over eight times (0.89 seconds per window) with minimal accuracy loss. A field application in a Chinese shale oil reservoir confirms the model effectively distinguishes strong active-stage fluid entry from weak, sustained off-target responses that conventional waterfall plots typically mask. These predictions align independently with offset-well low-frequency DAS strain-rate measurements and treatment parameters. DTCF-Net delivers a robust, near-real-time diagnostic tool for evaluating inter-stage communication and fluid distribution, improving the efficiency and reliability of field interpretation in complex unconventional reservoirs.