top of page
Writer's pictureLMS Solution

Neural network mppt controlled pv wind battery system

Neural network mppt controlled pv wind battery system


Introduction

The neural network MPPT controlled PV-wind-battery system, which is also known as a hybrid AC and DC microgrid. This system operates in islanded mode and integrates wind energy conversion systems and solar PV systems, controlled by neural network MPPT (Maximum Power Point Tracking) technology.



System Overview

The hybrid microgrid consists of several components:

  • Wind Energy Conversion System (WECS)

  • Solar PV System

  • Battery Storage

  • Islanded Mode Controller



Wind Energy Conversion System (WECS)

The WECS includes a wind turbine rated at 2.9 kW, a Permanent Magnet Synchronous Generator (PMSG), a rectifier, and a boost converter. The rectifier converts AC power from the PMSG into DC power, which is then processed by the boost converter. The boost converter is controlled by a neural network MPPT that receives inputs such as rectifier voltage (V_rc) and current (I_rc).

Solar PV System

The solar PV system consists of a PV panel rated at 2 kW, connected to a boost converter. The PV panel has a series configuration with eight modules. The boost converter is controlled by the neural network MPPT, which receives inputs like PV voltage (V_pv) and current (I_pv).

Battery Storage

The battery storage system uses 22V batteries connected to the DC bus via a bi-directional converter. This converter is controlled to maintain the DC bus voltage at 400V, ensuring a stable power supply to both AC and DC loads.

Loads

The system includes AC and DC loads. The AC loads are connected via a full-bridge inverter and an LCL filter. Initially, a 1000W AC load is connected, with an additional 1400W AC load added after 2 seconds. The DC load is rated at 1000W.

Control Strategy

Neural Network MPPT

The neural network MPPT controls the boost converters in both the WECS and the PV system. It receives inputs:

  • For WECS: Rectifier voltage (V_rc) and current (I_rc)

  • For PV system: PV voltage (V_pv) and current (I_pv)

The neural network processes these inputs to generate duty cycles, which control the boost converters to extract maximum power.

Bi-directional Converter Control

The bi-directional converter is controlled using a voltage control method to maintain the DC bus voltage at 400V. The control system measures the DC load voltage, compares it with the reference voltage (400V), and processes it through a PI controller to generate the duty cycle for the converter.

Simulation Results

The system was simulated under varying conditions to test its performance.

Initial Conditions

  • Irradiance: 1000 W/m²

  • Wind Speed: 12 m/s

The initial results showed the PV voltage around 250V and current around 7A, generating approximately 1950W at 1000 W/m². For the wind system, the boost converter provided around 2.9 kW.

Changing Conditions

  • Wind Speed: Changed to 1.2 m/s after 2 seconds

  • Irradiance: Alternated between 1000 W/m², 500 W/m², and 10 W/m² every 0.3 seconds

With these variations, the system demonstrated effective MPPT control, maintaining optimal power output from both PV and wind systems. The battery charge varied to balance the power, and the DC bus voltage was consistently maintained around 400V.

Load Conditions

  • AC Load: Initially 1000W, increased to 2400W after 2 seconds

  • DC Load: 1000W

The system successfully managed the load changes, with stable voltage and power outputs for both AC and DC loads.

Conclusion

The neural network MPPT controlled PV-wind-battery system efficiently manages power generation and distribution in a hybrid microgrid, even under varying environmental conditions. The control strategies ensure stable operation, maximizing power extraction from renewable sources while maintaining load balance.

0 views0 comments

Comments


bottom of page