top of page
Search

Fuzzy Logic-Based Energy Management in a Grid-Connected PV-Battery System using MATLAB

🟢 PV Array Configuration and Power Estimation

The PV system consists of 8 panels connected in series per string, and two such strings are used in parallel. Each panel is rated at 250 W, so:

  • Single string output = 8 × 250 = 2000 W

  • Total array output = 2 × 2000 = 4000 W

With each panel delivering around 30.7 V, the series configuration results in approximately 245–246 V DC per string. The system is designed to deliver 400 V at the DC bus.

⚡ Boost Converter Design

The boost converter steps up the panel voltage (~245 V) to 400 V. Key design steps include:

  • Using standard boost converter equations to calculate inductance (L) and capacitance (C)

  • Parameters are selected based on:

    • Input power = 4000 W

    • Output voltage = 400 V

    • Switching frequency = user-defined (e.g., 10 kHz)

The computed L and C values are implemented in the boost converter to maintain voltage regulation and ensure MPPT tracking.

🔋 Bidirectional DC-DC Converter for Battery Integration

A bidirectional DC-DC converter connects the battery (rated at 240 V) to the 400 V DC bus. Its purpose is to:

  • Charge the battery from excess PV power

  • Discharge the battery to support the load or grid when PV is insufficient

Design of the converter again follows classical formulas, considering:

  • Power rating (based on PV and battery capability)

  • Input voltage = 240 V, output voltage = 400 V

  • Operation in both directions (boost and buck modes)

🧰 LCL Filter Design for Grid Interface

To interface with the AC grid, an LCL filter is used to reduce harmonic distortion. Key design inputs include:

  • Power handling = up to 5000 W (PV + battery)

  • AC voltage = 230 V RMS

  • Switching frequency = 7.5 kHz

  • Grid frequency = 50 Hz

From the design equations, inductors (L1 and L2) and filter capacitor (C) values are derived to meet grid standards and reduce THD.

🌞 MPPT Control of the PV Panel

To extract maximum power from the PV panels:

  • Voltage and current from the PV array are continuously measured

  • These inputs are fed into an MPPT algorithm, which generates the duty cycle for the boost converter’s IGBT

  • The MPPT ensures the PV array operates at its maximum power point while maintaining a 400 V DC output

🔋 Voltage Control for Battery Converter

A PI controller is implemented to regulate the battery converter:

  • The actual DC bus voltage is compared with a 400 V reference

  • The resulting error is processed by the PI controller to adjust the duty cycle

  • This maintains stable DC bus voltage regardless of load or source variations

🧠 Fuzzy Logic-Based Energy Management System (EMS)

At the core of the system lies a fuzzy logic controller that manages energy flow between PV, battery, and grid. The controller takes two inputs:

  1. PV Power (scaled 0–1): Actual power divided by 4000 W

  2. Battery SoC (scaled 0–1): State of charge scaled from 0–100%

The fuzzy logic controller defines multiple membership functions:

  • PV: Zero, Small, Medium, Big, Very Big

  • SoC: Zero, Low, Medium, High, Very High

  • Output (Reference Current): Negative High, Negative Medium, Zero, Positive Medium, Positive High

Logic behavior:

  • If PV is low and SoC is low → draw power from the grid

  • If PV and SoC are high → inject power to the grid

  • If balanced → maintain zero power flow

The fuzzy surface maps PV and SoC conditions to appropriate current reference values between −1.3 to +1.3.

🔄 Inverter Control and Current Regulation

The reference current from the EMS is used to generate a sinusoidal signal:

  • Multiplied with a sine wave to produce an AC reference signal

  • Transformed from abc to dq frame

  • Compared with actual grid current (also in dq frame)

  • The error is fed into a current controller, generating control signals for PWM

These PWM pulses control the inverter's switches, allowing bidirectional power flow:

  • From DC bus to AC grid

  • Or from AC grid to the DC bus (during low irradiance or low SoC)

🔁 Dynamic Response to Changing Irradiance Conditions

The system is tested under different irradiance levels:

  • 1000 W/m² → 500 W/m² → 10 W/m² → back to 1000 W/m²

The response observed:

  • During high irradiance, PV meets load demand and charges the battery

  • During low irradiance, battery discharges or power is drawn from the grid

  • State of charge transitions from charging to discharging as needed

  • Grid current and voltage remain sinusoidal and in phase during normal conditions; phase shift appears during grid import

📊 Monitoring and Visualization

Multiple scopes are used in Simulink to monitor:

  • PV voltage, current, and power

  • DC link voltage

  • Battery voltage, current, and SoC

  • Load power and grid power

  • Grid voltage/current at Point of Common Coupling (PCC)

  • Frequency stability (maintained at 50 Hz)

The system effectively maintains all performance parameters under variable conditions.

✅ Conclusion

This MATLAB/Simulink model provides a robust approach to integrating PV and battery systems with the grid. By combining traditional converter control with fuzzy logic-based EMS, the system can:

  • Extract maximum power from PV

  • Manage battery charging/discharging dynamically

  • Maintain grid standards (voltage, frequency, THD)

  • Enable smart and reliable bidirectional power flow

This demonstration offers a practical and scalable foundation for real-world energy management systems using renewable energy sources.

Comments

bottom of page