# Real-time data fetching approach for performance evaluation of a DFIG wind power generation system using an IoT-enabled wind emulator

## TL;DR

This paper introduces an IoT-based system for real-time performance evaluation of wind power systems using a wind emulator, enabling accurate and scalable testing.

## Contribution

The novel contribution is an IoT-enabled wind emulator with low-latency real-time synchronization and cloud-based predictive analysis for DFIG wind systems.

## Key findings

- The system achieved 87% model accuracy (MAPE) between theoretical and emulator power outputs.
- Health index reliability was 95%, and grid power factor reached near-unity at 0.999.
- The setup demonstrated low latency of 180 ms for real-time synchronization with global weather data.

## Abstract

The increased integration of wind energy into the power system network requires advanced testing and performance evaluation methods to ensure reliability and efficiency. This paper presents an IoT-based real-time data collection method for analyzing the performance of the Wind Power Generation System (WPGS) using an intelligent IoT-enabled wind emulator system. The proposed system uses IoT to gather comprehensive real-time wind data, which is processed by a digitally controlled emulator to accurately assess wind turbine responses under different wind conditions. The advantage of this system is that IoT and cloud-based data analytics enable predictive analysis of the WPGS, helping evaluate its behavior and performance under various real-time scenarios. The approach is tested with a wind emulator setup consisting of a 1 kW Doubly-Fed Induction Generator (DFIG) connected to a Brushless DC (BLDC) motor, where the wind turbine model is developed on the VEE Pro platform, integrating an IoT-NodeRed, cloud API, and FPGA controller to simulate real-world wind conditions. Unlike conventional systems, the proposed architecture achieves real-time synchronization between global weather data and emulator control with low latency of 180 ms. Experimental results indicate 87% model accuracy Mean Absolute Percentage Error (MAPE) between theoretical and emulator power outputs, 95% health index reliability, and near-unity grid power factor of 0.999. This study provides a cost-effective, scalable, and adaptable solution for real-time wind energy analysis, supporting ongoing research and development.