To enhance the operational efficiency and fault detection accuracy of hydroelectric units, this paper proposes a parallel convolutional neural network model that integrates the Gramian angular summation field (GASF) with an Improved coati optimization algorithm–parallel convolutional neural network (ICOA-PCNN). Additionally, to further improve the model’s accuracy in fault identification, a multi-head self-attention mechanism (MSA) and support vector machine (SVM) are introduced for a secondary optimization of the model. Initially, the GASF technique converts one-dimensional time series signals into two-dimensional images, and a COA-CNN dual-branch model is established for feature extraction. To address the issues of uneven population distribution and susceptibility to local optima in the COA algorithm, various optimization strategies are implemented to improve its global search capability. Experimental results indicate that the accuracy of this model reaches 100%, significantly surpassing other nonoptimized models. This research provides a valuable addition to fault diagnosis technology for modern hydroelectric units.

The vortex rope in the draft tube is considered as the major contributor to pressure pulsation at partial load (PL) conditions, which causes the hydro unit to operate unstably. Based on the prototype Francis turbine HLA551-LJ-43 in the laboratory, J-grooves are designed on its conical section in this paper. We used numerical simulation to study the effect of the J-grooves on vortex suppression and energy dissipation in the draft tube. Four typical operating conditions were chosen to analyze the vortex suppression; the corresponding flow ratios Q* are 100%, 82%, 69%, and 53%, respectively. Entropy production theory is used to calculate the energy losses and assess the effect of the J-groove on energy dissipation under part-load conditions. By comparing entropy production, circumferential and axial velocity components, swirl intensity, pressure pulsation, and vortex distribution in a draft tube with and without J-grooves at different operating conditions, it can be concluded that the entropy production on the wall containing a conical section with J-grooves is obviously smaller than that without J-grooves, the effects of J-grooves on reducing circumferential velocity component V_u, pressure pulsation, and weakening vortex intensity and vortex rope in the conical section are obvious, especially at part load and deep part-load operating conditions. Using J-grooves shows better performance on vortex control and energy dissipation in the draft tube of a Francis turbine at partial load conditions.