AI-Based Railway Vehicle Condition Monitoring System
Real-time condition monitoring and AI-based failure prediction analysis system for railway vehicle components Supports preventive maintenance through sensor data visualization and early anomaly detection.
Route/Device Train Information: Train formation status by route (Gyeongbu Line, Honam Line, etc.) and individual device (main transformer, traction motor, etc.) status visualization
Maintenance Management: Diagnostic and action code registration for devices with detected failures, maintenance history tracking
AI Training/Model Configuration: Hyperparameter settings for anomaly diagnosis and prediction AI models, training data selection, model deployment
Statistics/History Lookup: Period-based analysis statistics, maintenance statistics, system history lookup, and inter-device data comparison
Key Achievements
Drag-and-Drop Dashboard: React Grid Layout-based customization system for freely placing 8 widget types on 6x3 grid with layout save/restore functionality
Train Formation Visualization Component: SVG-based 4/6-car train formation graphics as React components with intuitive color-coded status display (normal/warning/danger)
Hierarchical Device Selection UI: 4-level hierarchy navigation (Route→Formation→Car→Device) using Tree component with multi-select (up to 5 devices) for data comparison
Context + Zustand Hybrid State Management: Local state management with Context API for dashboard edit mode, Zustand stores for global data queries, optimizing performance
Responsive Widget Pagination: Dynamic widget height calculation for automatic table row count adjustment, maintaining optimal data display in fullscreen mode
Mantine Form + Zod Integrated Validation: Real-time validation and type safety for complex forms like AI training configuration using Zod schemas