MATLAB Simulation of Neural Network MPPT Controlled PV Wind Battery System

MATLAB Simulation of Neural Network MPPT Controlled PV Wind Battery System

𝐈𝐧𝐭𝐫𝐨𝐝𝐮𝐜𝐭𝐢𝐨𝐧

The MATLAB Simulation of Neural Network MPPT Controlled PV Wind Battery System demonstrates an intelligent renewable energy system designed for islanded hybrid AC and DC microgrid operation.

Neural network mppt controlled pv wind battery system

₹12,000.00₹6,000.00

Buy Now

The system combines:

• ☀️ Solar PV power generation

• 🌬️ Wind Energy Conversion System (WECS)

• 🧠 Neural Network based MPPT control

• 🔋 Battery Energy Storage System

• 🔄 Bidirectional DC–DC converter

• ⚡ Common 400 V DC bus

• 🔌 DC load

• 🔁 Single-phase inverter with LCL filter

• 🏠 Variable AC load

The main objective is to extract maximum available power from both PV and wind sources while maintaining a stable 400 V DC bus under changing solar irradiation, wind speed, and load conditions.

𝐒𝐲𝐬𝐭𝐞𝐦 𝐎𝐯𝐞𝐫𝐯𝐢𝐞𝐰

The complete MATLAB/Simulink model operates as an islanded hybrid renewable microgrid.

Power generated from PV and wind systems is connected to a common DC bus. The battery energy storage system regulates the DC-link voltage, while an inverter supplies the AC load.

System Section	Main Components	Function
Wind System	Wind turbine, PMSG, rectifier, boost converter	Wind power generation
Solar PV	PV array, boost converter	Solar power generation
MPPT Control	Neural Network, PWM generators	Maximum power extraction
Battery System	Battery, bidirectional converter	DC bus regulation
DC Bus	Common DC link	Renewable power integration
DC Load	DC electrical load	Direct DC power consumption
AC System	Full-bridge inverter, LCL filter	DC-to-AC conversion
AC Load	Variable load bank	Islanded AC power consumption

𝐍𝐞𝐮𝐫𝐚𝐥 𝐍𝐞𝐭𝐰𝐨𝐫𝐤 𝐌𝐏𝐏𝐓 𝐂𝐨𝐧𝐭𝐫𝐨𝐥

A major feature of this model is the use of Neural Network MPPT instead of a conventional MPPT technique.

The neural network receives four electrical measurements:

Neural Network Input	Description
Vrec	Wind rectifier output voltage
Irec	Wind rectifier output current
Vpv	PV array voltage
Ipv	PV array current

The neural network uses these measurements to determine suitable converter duty-cycle commands.

𝐖𝐢𝐧𝐝 𝐌𝐏𝐏𝐓

• Vrec and Irec are measured from the wind rectifier.

• The neural network determines the required wind boost converter duty cycle.

• The duty-cycle signal is sent to the PWM generator.

• PWM pulses control the boost converter switching device.

• The converter extracts maximum available wind power.

• Boost converter output is connected to the common DC bus.

𝐏𝐕 𝐌𝐏𝐏𝐓

• PV voltage and current are continuously measured.

• Vpv and Ipv are supplied to the neural network.

• The controller determines the PV boost converter duty cycle.

• The PWM generator converts the command into switching pulses.

• The boost converter operates the PV array near its maximum power point.

This allows intelligent MPPT operation under changing environmental conditions.

𝐖𝐢𝐧𝐝 𝐄𝐧𝐞𝐫𝐠𝐲 𝐂𝐨𝐧𝐯𝐞𝐫𝐬𝐢𝐨𝐧 𝐒𝐲𝐬𝐭𝐞𝐦

The wind generation section contains:

• 🌬️ Wind turbine

• ⚙️ Permanent Magnet Synchronous Generator (PMSG)

• 🔄 AC–DC rectifier

• ⬆️ DC–DC boost converter

• 🧠 Neural Network MPPT

• PWM switching controller

Main Wind System Parameters

Parameter	Value
Wind turbine rated power	Approximately 2.9 kW
Initial wind speed	12 m/s
Wind speed after 2 s	Approximately 1.2 m/s
Initial extracted wind power	Approximately 2.9 kW
DC bus voltage	400 V

The PMSG produces variable AC power according to wind conditions. A rectifier converts this generated AC power into DC before it is processed by the boost converter.

𝐒𝐨𝐥𝐚𝐫 𝐏𝐕 𝐒𝐲𝐬𝐭𝐞𝐦

The solar PV subsystem consists of a 2 kW PV array connected to a boost converter.

PV Array Configuration

Parameter	Value
PV system rated power	2 kW
Individual PV module rating	250 W
Series-connected modules	8
Parallel strings	1
MPPT method	Neural Network MPPT
DC bus voltage	400 V

The PV model also considers standard module parameters such as:

• Open-circuit voltage

• Short-circuit current

• Voltage at maximum power point

• Current at maximum power point

The PV boost converter provides the required voltage conversion while the neural network continuously searches for maximum available PV power.

𝐁𝐚𝐭𝐭𝐞𝐫𝐲 𝐄𝐧𝐞𝐫𝐠𝐲 𝐒𝐭𝐨𝐫𝐚𝐠𝐞

The battery plays an important role in maintaining the stability of the hybrid microgrid.

It is connected to the common DC bus through a bidirectional DC–DC converter.

The battery system can:

• Absorb surplus renewable power.

• Supply power when renewable generation decreases.

• Compensate for sudden load changes.

• Reduce DC-link voltage fluctuations.

• Maintain power balance between generation and loads.

𝐃𝐂 𝐁𝐮𝐬 𝐕𝐨𝐥𝐭𝐚𝐠𝐞 𝐂𝐨𝐧𝐭𝐫𝐨𝐥

The reference DC-link voltage is maintained at:

400 V DC

The voltage controller operates using:

1. DC bus voltage measurement

2. 400 V reference comparison

3. PI voltage controller

4. Duty-cycle generation

5. PWM pulse generation

6. Bidirectional converter switching

When excess renewable energy is available, the battery can operate in charging mode.

When renewable generation is insufficient, the battery can support the DC bus by supplying the required power.

𝐁𝐢𝐝𝐢𝐫𝐞𝐜𝐭𝐢𝐨𝐧𝐚𝐥 𝐃𝐂–𝐃𝐂 𝐂𝐨𝐧𝐯𝐞𝐫𝐭𝐞𝐫

The bidirectional converter creates a controlled power path between the battery and DC bus.

Main Functions

• Battery charging

• Battery discharging

• DC-link voltage regulation

• Renewable power balancing

• Load transient compensation

• Power flow management

The converter is controlled through PWM signals generated from the DC bus voltage controller.

𝐃𝐂 𝐋𝐨𝐚𝐝 𝐎𝐩𝐞𝐫𝐚𝐭𝐢𝐨𝐧

A DC load is directly connected to the common DC bus.

Parameter	Approximate Value
DC bus voltage	400 V
DC load power	1000 W
DC load connection	Common DC bus

Despite changes in PV and wind generation, the simulation shows that the DC bus remains close to its required operating voltage.

𝐈𝐬𝐥𝐚𝐧𝐝𝐞𝐝 𝐈𝐧𝐯𝐞𝐫𝐭𝐞𝐫 𝐒𝐲𝐬𝐭𝐞𝐦

The AC load is supplied from the common DC bus through a full-bridge inverter.

The AC power conversion stage contains:

• Full-bridge inverter

• PWM inverter control

• LCL output filter

• AC voltage and current measurement

• Variable AC load

Because the system operates in islanded mode, the inverter is responsible for supplying the required AC voltage to the local AC load without depending on the utility grid.

𝐀𝐂 𝐋𝐨𝐚𝐝 𝐕𝐚𝐫𝐢𝐚𝐭𝐢𝐨𝐧

A dynamic load condition is introduced to test the ability of the system to maintain stable operation.

Simulation Time	Load 1	Load 2	Total AC Load
0–2 s	1000 W	Disconnected	1000 W
After 2 s	1000 W	1400 W	2400 W

At 2 seconds, an additional 1.4 kW AC load is connected.

Therefore, the AC power demand increases from:

1.0 kW → 2.4 kW

This load step is useful for studying the transient response of the battery, DC bus, and inverter.

𝐃𝐲𝐧𝐚𝐦𝐢𝐜 𝐒𝐨𝐥𝐚𝐫 𝐈𝐫𝐫𝐚𝐝𝐢𝐚𝐧𝐜𝐞

The PV system is tested under rapidly changing irradiation conditions.

Condition	Solar Irradiance
High irradiation	1000 W/m²
Medium irradiation	500 W/m²
Very low irradiation	10 W/m²
Recovery	500 W/m²
High irradiation restored	1000 W/m²

The irradiation is varied approximately every 0.3 seconds.

This test demonstrates the tracking capability of the Neural Network MPPT controller during fast environmental changes.

𝐖𝐢𝐧𝐝 𝐒𝐩𝐞𝐞𝐝 𝐕𝐚𝐫𝐢𝐚𝐭𝐢𝐨𝐧

The wind turbine is also tested dynamically.

Time	Wind Speed
Before 2 s	12 m/s
After 2 s	Approximately 1.2 m/s

The sudden wind-speed reduction changes the available power from the wind generator.

The battery and DC bus controller compensate for the resulting power imbalance.

𝐖𝐨𝐫𝐤𝐢𝐧𝐠 𝐏𝐫𝐨𝐜𝐞𝐬𝐬

The complete energy conversion process can be understood in the following steps:

1. 🌬️ Wind energy drives the PMSG.

2. The PMSG produces variable AC electrical power.

3. The rectifier converts wind-generator AC power into DC.

4. Neural Network MPPT controls the wind boost converter.

5. ☀️ The PV array generates DC power according to irradiation.

6. Neural Network MPPT controls the PV boost converter.

7. Both renewable sources supply the common DC bus.

8. 🔋 The bidirectional battery converter regulates the DC bus.

9. The DC load receives power directly from the DC link.

10. ⚡ The inverter converts DC power into AC power.

11. The LCL filter improves inverter output quality.

12. The AC load is supplied in islanded operation.

𝐂𝐨𝐧𝐭𝐫𝐨𝐥 𝐒𝐭𝐫𝐚𝐭𝐞𝐠𝐲

The model combines three major control functions.

🧠 Neural Network MPPT Control

Used for:

• Wind maximum power extraction

• Solar PV maximum power extraction

• Duty-cycle estimation

• Dynamic environmental adaptation

🔋 DC Bus Voltage Control

Used for:

• Maintaining approximately 400 V

• Controlling battery charging and discharging

• Managing renewable power fluctuations

• Responding to load disturbances

⚡ Islanded Inverter Control

Used for:

• Supplying AC loads

• Maintaining stable inverter voltage

• Producing sinusoidal AC output

• Handling load changes

𝐒𝐢𝐦𝐮𝐥𝐚𝐭𝐢𝐨𝐧 𝐓𝐞𝐬𝐭 𝐂𝐨𝐧𝐝𝐢𝐭𝐢𝐨𝐧𝐬

The simulation includes several disturbances simultaneously.

Test	Initial Condition	Changed Condition
Solar irradiation	1000 W/m²	500, 10, 500 and 1000 W/m²
Wind speed	12 m/s	About 1.2 m/s after 2 s
AC load	1000 W	2400 W after 2 s
DC load	Approximately 1000 W	Maintained
DC bus reference	400 V	400 V
Operating mode	Islanded	Islanded

These operating conditions make the simulation useful for evaluating controller performance under realistic renewable-energy variations.

𝐒𝐢𝐦𝐮𝐥𝐚𝐭𝐢𝐨𝐧 𝐑𝐞𝐬𝐮𝐥𝐭𝐬

☀️ PV Performance

At high solar irradiation, the PV array operates close to its rated power.

Irradiance	Approximate PV Power
1000 W/m²	1.9–2.0 kW
500 W/m²	Around 990 W
10 W/m²	Near 0 W
500 W/m² after recovery	Around 990 W
1000 W/m² after recovery	Around 2 kW

Typical PV electrical quantities observed during high-irradiation operation include:

• PV voltage of approximately 250 V

• PV current of approximately 7 A

• PV output near its rated power

During extremely low irradiation, PV power falls close to zero.

When irradiation increases again, the neural network controller quickly restores maximum power extraction.

𝐖𝐢𝐧𝐝 𝐏𝐨𝐰𝐞𝐫 𝐑𝐞𝐬𝐮𝐥𝐭𝐬

Under the initial high wind-speed condition:

• Wind speed is approximately 12 m/s.

• Wind generation reaches approximately 2.9 kW.

• The neural network controls the boost converter.

• Maximum available wind energy is transferred to the DC bus.

Following the wind-speed change after 2 seconds, wind generation reduces and the battery adjusts its power contribution accordingly.

𝐁𝐚𝐭𝐭𝐞𝐫𝐲 𝐑𝐞𝐬𝐩𝐨𝐧𝐬𝐞

The battery current changes continuously according to renewable generation and load demand.

During high renewable generation:

• Excess energy is transferred to the battery.

• Battery charging current changes with PV output.

• Battery state of charge gradually increases.

In the displayed model convention, negative current values represent battery charging during several operating intervals.

When renewable generation or load conditions change, battery current automatically adjusts to support system power balance.

𝐃𝐂 𝐁𝐮𝐬 𝐑𝐞𝐬𝐮𝐥𝐭

One of the most important simulation results is the DC-link voltage.

DC Bus Reference = 400 V

Even with:

• Rapid irradiation variations

• Wind-speed reduction

• PV power variation

• Wind power variation

• Battery charging changes

• Sudden AC load increase

the DC bus remains regulated close to 400 V after transient disturbances.

This demonstrates the effectiveness of the bidirectional battery converter and voltage-control strategy.

𝐀𝐂 𝐋𝐨𝐚𝐝 𝐑𝐞𝐬𝐮𝐥𝐭𝐬

During the first two seconds:

AC Load ≈ 1 kW

After the additional load is switched on:

AC Load ≈ 2.4 kW

The simulation results show:

• Stable AC load voltage

• Sinusoidal load current

• Correct load-power increase

• Stable inverter operation

• Successful response to the load step

𝐈𝐧𝐯𝐞𝐫𝐭𝐞𝐫 𝐑𝐞𝐬𝐮𝐥𝐭𝐬

The inverter scope demonstrates:

• Sinusoidal inverter output voltage

• Controlled inverter current

• Stable AC supply during islanded operation

• Proper response to changes in AC load

• Continuous operation despite renewable power fluctuations

The LCL filter helps provide a smooth AC output to the connected load.

𝐏𝐨𝐰𝐞𝐫 𝐁𝐚𝐥𝐚𝐧𝐜𝐢𝐧𝐠

The hybrid microgrid continuously balances:

PV Power + Wind Power + Battery Power → DC and AC Loads

The battery becomes particularly important when there is a mismatch between renewable generation and load demand.

For example:

• High renewable power → battery charging

• Reduced PV power → battery compensates

• Reduced wind generation → battery contribution changes

• Increased AC load → battery and sources maintain balance

This enables stable islanded microgrid operation without support from the utility grid.

𝐊𝐞𝐲 𝐅𝐞𝐚𝐭𝐮𝐫𝐞𝐬

• ✅ Neural Network based MPPT for solar PV

• ✅ Neural Network based MPPT for wind generation

• ✅ Four electrical inputs to the intelligent MPPT controller

• ✅ Approximately 2.9 kW wind energy system

• ✅ 2 kW solar PV array

• ✅ 400 V regulated DC bus

• ✅ Bidirectional battery energy storage interface

• ✅ Dynamic battery charging and discharging

• ✅ DC and AC loads in the same microgrid

• ✅ Islanded inverter operation

• ✅ LCL output filtering

• ✅ Dynamic solar irradiation testing

• ✅ Dynamic wind-speed testing

• ✅ Sudden AC load switching

• ✅ Source-load power balancing

• ✅ MATLAB/Simulink based implementation

𝐀𝐝𝐯𝐚𝐧𝐭𝐚𝐠𝐞𝐬 𝐨𝐟 𝐍𝐞𝐮𝐫𝐚𝐥 𝐍𝐞𝐭𝐰𝐨𝐫𝐤 𝐌𝐏𝐏𝐓

Neural Network MPPT provides several advantages for hybrid renewable systems:

• Fast response to changing operating conditions

• Suitable for nonlinear renewable energy characteristics

• Ability to process multiple electrical inputs

• Improved adaptation to irradiation changes

• Effective tracking during wind-speed variations

• Reduced dependence on fixed operating rules

• Suitable for intelligent microgrid control research

• Easy integration with MATLAB/Simulink converter models

𝐀𝐩𝐩𝐥𝐢𝐜𝐚𝐭𝐢𝐨𝐧𝐬

This MATLAB simulation concept can be useful for studying:

• Hybrid PV–wind renewable energy systems

• Islanded AC/DC microgrids

• Neural Network based MPPT

• Battery energy storage control

• Intelligent renewable energy management

• Bidirectional DC–DC converter operation

• DC bus voltage regulation

• Standalone inverter systems

• Renewable power balancing

• Artificial intelligence in power electronics

• Smart energy systems

• Hybrid renewable power conversion

𝐖𝐡𝐨 𝐂𝐚𝐧 𝐔𝐬𝐞 𝐓𝐡𝐢𝐬 𝐌𝐨𝐝𝐞𝐥?

The simulation is suitable for:

• 🎓 Electrical engineering students

• 🔬 Renewable energy researchers

• ⚡ Power electronics engineers

• 🧠 Artificial intelligence researchers

• 🔋 Battery energy storage researchers

• 🌐 Microgrid researchers

• 💻 MATLAB/Simulink learners

• 🌞 Solar and wind energy engineers

It provides a clear platform for understanding how AI-based MPPT, renewable energy sources, energy storage, and inverter systems interact inside an islanded microgrid.

𝐂𝐨𝐧𝐜𝐥𝐮𝐬𝐢𝐨𝐧

The MATLAB Simulation of Neural Network MPPT Controlled PV Wind Battery System demonstrates an intelligent islanded hybrid AC/DC microgrid integrating solar PV, wind generation, battery storage, DC loads, and AC loads.

The Neural Network MPPT controller successfully controls the PV and wind boost converters according to real-time voltage and current measurements. The PV system responds effectively to irradiation variations, while the wind energy conversion system adapts to changing wind-speed conditions.

A bidirectional battery converter maintains the common DC bus close to 400 V and balances the difference between renewable generation and load demand. The inverter supplies the islanded AC load through an LCL filter, including a load increase from 1 kW to 2.4 kW.

Overall, the simulation demonstrates effective maximum power extraction, DC-link voltage regulation, battery energy management, inverter operation, and source-load power balancing under dynamic environmental and loading conditions.