PV Wind Battery Based DC Microgrid with Neural Network MPPT

PV Wind Battery Based DC Microgrid with Neural Network MPPT

𝐈𝐧𝐭𝐫𝐨𝐝𝐮𝐜𝐭𝐢𝐨𝐧

The PV Wind Battery Based DC Microgrid with Neural Network MPPT is a renewable-energy-based DC power system that combines solar PV, wind energy, battery energy storage, DC–DC converters, and Neural Network Maximum Power Point Tracking (MPPT).

PV Wind Battery Based DC Microgrid with Neural Network MPPT in MATLAB

₹10,000.00₹5,000.00

Buy Now

The main purpose of the system is to:

• Extract maximum available power from the solar PV array.

• Extract maximum available power from the wind energy conversion system.

• Maintain the common DC bus at approximately 400 V.

• Supply continuous power to the DC load.

• Store surplus renewable energy in the battery.

• Discharge the battery when PV and wind generation are insufficient.

• Study system performance under changing solar irradiance and wind speed.

The model demonstrates how renewable sources and battery storage can work together to provide stable DC power under dynamic operating conditions.

𝐒𝐲𝐬𝐭𝐞𝐦 𝐎𝐯𝐞𝐫𝐯𝐢𝐞𝐰

The DC microgrid consists of three main energy sections:

• Solar PV energy conversion system

• Wind energy conversion system

• Battery energy storage system

All three sections are connected to a common 400 V DC bus, which supplies the DC load.

Main System Components

Section	Main Components	Main Function
Solar PV	PV array, boost converter, Neural Network MPPT	Solar power extraction and voltage boosting
Wind System	Wind turbine, PMSG, diode rectifier, boost converter	Wind energy conversion and DC power generation
Wind MPPT	Neural Network MPPT and PWM generator	Maximum power extraction from wind
PV MPPT	Neural Network MPPT and PWM generator	Maximum power extraction from PV
Battery	Battery and bidirectional DC–DC converter	Charging and discharging
DC Bus	Common DC link	Power exchange between sources, battery, and load
DC Load	Resistive DC load	Consumes power from the microgrid
Controller	DC-bus voltage controller	Maintains DC-link voltage near 400 V

𝐒𝐨𝐥𝐚𝐫 𝐏𝐕 𝐒𝐲𝐬𝐭𝐞𝐦

The solar PV array is one of the main renewable energy sources in the DC microgrid.

PV Module Parameters

Parameter	Value
Single PV module power	250 W
Voltage at maximum power point	30.7 V
Current at maximum power point	8.15 A
Series-connected modules	8
Parallel strings	3
Approximate PV array maximum power	6 kW
Standard irradiance	1000 W/m²
Standard temperature	25°C
Approximate PV operating voltage	245 V
DC-bus voltage	400 V

With eight modules connected in series and three parallel strings, the PV array can generate approximately 6 kW under standard test conditions.

𝐏𝐕 𝐍𝐞𝐮𝐫𝐚𝐥 𝐍𝐞𝐭𝐰𝐨𝐫𝐤 𝐌𝐏𝐏𝐓

The PV boost converter is controlled using a Neural Network MPPT controller.

The Neural Network receives:

• PV voltage

• PV current

The controller generates:

• Duty cycle

The duty-cycle command is sent to the PWM generator, which produces the switching pulse for the PV boost converter.

This converter performs two important functions:

• Extracts the maximum available power from the PV array.

• Steps the PV voltage up to the required DC-bus level.

𝐖𝐢𝐧𝐝 𝐄𝐧𝐞𝐫𝐠𝐲 𝐂𝐨𝐧𝐯𝐞𝐫𝐬𝐢𝐨𝐧 𝐒𝐲𝐬𝐭𝐞𝐦

The wind energy section consists of:

• Wind turbine

• Permanent Magnet Synchronous Generator (PMSG)

• Diode rectifier

• Boost converter

• Neural Network MPPT

• PWM generator

Wind System Details

Parameter	Value
Wind turbine rated power	6.5 kW
Rated operating wind speed discussed	12 m/s
Generator	PMSG
AC-to-DC conversion	Diode rectifier
DC–DC stage	Boost converter
MPPT technique	Neural Network MPPT

The mechanical energy produced by the wind turbine drives the PMSG. The generator output is then converted from AC to DC using the diode rectifier.

𝐖𝐢𝐧𝐝 𝐍𝐞𝐮𝐫𝐚𝐥 𝐍𝐞𝐭𝐰𝐨𝐫𝐤 𝐌𝐏𝐏𝐓

The wind-side Neural Network MPPT receives two measured quantities:

• Rectifier output voltage

• Rectifier output current

Based on these inputs, the trained Neural Network produces the required duty cycle.

The PWM generator converts this duty-cycle command into switching pulses for the wind-side boost converter.

Therefore, the converter can:

• Extract the maximum available wind power.

• Increase the rectified DC voltage.

• Transfer wind-generated power to the common DC bus.

𝐁𝐚𝐭𝐭𝐞𝐫𝐲 𝐄𝐧𝐞𝐫𝐠𝐲 𝐒𝐭𝐨𝐫𝐚𝐠𝐞 𝐒𝐲𝐬𝐭𝐞𝐦

The battery provides energy balancing between renewable generation and load demand.

Battery Parameters

Parameter	Value
Battery rated voltage	240 V
Battery capacity	48 Ah
Observed battery operating voltage	Approximately 260 V
Converter	Bidirectional DC–DC converter
DC-bus reference	400 V
Power-flow capability	Charging and discharging

The bidirectional converter is important because the battery must support power flow in both directions.

𝐁𝐚𝐭𝐭𝐞𝐫𝐲 𝐂𝐡𝐚𝐫𝐠𝐢𝐧𝐠 𝐌𝐨𝐝𝐞

When the combined PV and wind generation is greater than the load requirement:

• The load receives the required power.

• Excess renewable power is transferred to the battery.

• Battery current becomes negative according to the measurement convention used in the model.

• Battery SOC gradually increases.

For example, during high renewable generation, the observed battery current is approximately:

−35 A

This indicates that the battery is charging.

𝐁𝐚𝐭𝐭𝐞𝐫𝐲 𝐃𝐢𝐬𝐜𝐡𝐚𝐫𝐠𝐢𝐧𝐠 𝐌𝐨𝐝𝐞

When PV and wind generation become lower than the DC load requirement:

• Renewable sources supply whatever power is available.

• The battery supplies the remaining power.

• Battery current changes from negative to positive.

• The battery operates in discharging mode.

• The DC load continues receiving power.

This allows the system to maintain load supply even during low renewable energy availability.

𝐁𝐢𝐝𝐢𝐫𝐞𝐜𝐭𝐢𝐨𝐧𝐚𝐥 𝐃𝐂–𝐃𝐂 𝐂𝐨𝐧𝐯𝐞𝐫𝐭𝐞𝐫 𝐂𝐨𝐧𝐭𝐫𝐨𝐥

The bidirectional converter is controlled using a DC-bus voltage control strategy.

The controller continuously measures the actual DC-bus voltage and compares it with the reference value of:

400 V

Depending on the voltage deviation, the controller determines the converter switching command.

The control structure includes:

• DC-bus voltage measurement

• DC-bus reference

• Voltage-error calculation

• PI controller

• PWM generator

• Bidirectional converter switches

The main objective is to keep the DC-bus voltage close to 400 V even when PV generation, wind generation, or battery power changes.

𝐃𝐂 𝐋𝐨𝐚𝐝

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

Load Parameters

Parameter	Value
DC-bus voltage	400 V
DC load power	Approximately 2000 W
DC load current	Approximately 5 A
Load behavior	Nearly constant power during the simulation

One of the important observations in the simulation is that the load continues receiving nearly 2 kW despite variations in solar irradiance and wind speed.

𝐖𝐨𝐫𝐤𝐢𝐧𝐠 𝐏𝐫𝐨𝐜𝐞𝐬𝐬

The complete operating process of the DC microgrid can be summarized as follows:

1. The PV array converts solar energy into DC electrical power.

2. PV voltage and current are measured.

3. The PV Neural Network MPPT generates the required duty cycle.

4. The PV boost converter extracts maximum power and transfers it to the DC bus.

5. The wind turbine drives the PMSG.

6. PMSG output is converted to DC using the diode rectifier.

7. Rectifier voltage and current are provided to the wind Neural Network MPPT.

8. The wind-side boost converter transfers the extracted wind power to the common DC bus.

9. The DC load continuously receives power from the DC bus.

10. The battery converter checks the DC-bus voltage.

11. Excess renewable power charges the battery.

12. Insufficient renewable power causes the battery to discharge.

13. The control system keeps the DC bus close to 400 V.

𝐂𝐨𝐧𝐭𝐫𝐨𝐥 𝐒𝐭𝐫𝐚𝐭𝐞𝐠𝐲

The proposed simulation uses two main control concepts.

Neural Network MPPT Control

Used for:

• Solar PV system

• Wind energy conversion system

Inputs:

• Voltage

• Current

Output:

• Converter duty cycle

Purpose:

• Maximum power extraction

• Improved utilization of renewable energy

• Converter switching control

DC-Bus Voltage Control

Used for:

• Battery bidirectional converter

Input:

• Actual DC-bus voltage

Reference:

• 400 V

Purpose:

• Maintain stable DC-bus voltage.

• Control battery charging and discharging.

• Balance renewable generation and load demand.

𝐍𝐞𝐮𝐫𝐚𝐥 𝐍𝐞𝐭𝐰𝐨𝐫𝐤 𝐓𝐫𝐚𝐢𝐧𝐢𝐧𝐠

Separate training data are used for the PV and wind MPPT controllers.

The training process uses suitable input and output datasets representing the operating behavior of each renewable source.

Neural Network Data Structure

Controller	Inputs	Output
PV Neural Network MPPT	PV voltage and PV current	Duty cycle
Wind Neural Network MPPT	Rectifier voltage and rectifier current	Duty cycle

After training, the Neural Network is incorporated into the simulation to generate converter duty-cycle commands during changing environmental conditions.

𝐃𝐲𝐧𝐚𝐦𝐢𝐜 𝐓𝐞𝐬𝐭 𝐂𝐨𝐧𝐝𝐢𝐭𝐢𝐨𝐧𝐬

To evaluate system performance, both solar irradiance and wind speed are varied during simulation.

Solar Irradiance Variation

The discussed operating levels include:

Irradiance	Approximate PV Power Observed
1000 W/m²	Around 6 kW
700 W/m²	Around 4 kW
400 W/m²	Around 2 kW
100 W/m²	Very low PV power

The PV power decreases as irradiance decreases.

Wind-Speed Variation

The wind-speed operating points discussed include:

Operating Condition	Wind Speed
High wind condition	12 m/s
Intermediate condition	10 m/s
Low wind condition	8 m/s

At the higher wind condition, more power is obtained from the wind energy system.

As wind speed decreases, wind-generated power also decreases.

𝐒𝐢𝐦𝐮𝐥𝐚𝐭𝐢𝐨𝐧 𝐑𝐞𝐬𝐮𝐥𝐭𝐬

The simulation monitors several important electrical variables.

Signals Observed

• PV voltage

• PV current

• PV power

• Wind-side rectifier power

• Wind boost-converter power

• DC-bus voltage

• DC load current

• DC load power

• Battery voltage

• Battery current

• Battery state of charge

𝐇𝐢𝐠𝐡 𝐑𝐞𝐧𝐞𝐰𝐚𝐛𝐥𝐞 𝐏𝐨𝐰𝐞𝐫 𝐂𝐨𝐧𝐝𝐢𝐭𝐢𝐨𝐧

At approximately 1000 W/m² solar irradiance:

• PV power is approximately 6 kW.

• PV voltage is around 245 V.

• PV current is approximately 24 A.

• Wind boost-converter output power is around 5.3 kW under the described operating condition.

• DC bus remains close to 400 V.

• DC load consumes approximately 2 kW.

• Load current is around 5 A.

• Battery current is approximately −35 A.

• Battery SOC increases.

This shows that renewable generation exceeds load demand, so the surplus power is stored in the battery.

𝐏𝐕 𝐈𝐫𝐫𝐚𝐝𝐢𝐚𝐧𝐜𝐞 𝐑𝐞𝐝𝐮𝐜𝐭𝐢𝐨𝐧

When solar irradiance reduces to approximately 700 W/m²:

• PV power reduces to around 4 kW.

• Battery charging current decreases.

• Battery current is approximately −30 A under the described condition.

• The load continues receiving the required power.

This shows the battery automatically adapts to changes in available renewable generation.

𝐋𝐨𝐰𝐞𝐫 𝐏𝐕 𝐏𝐨𝐰𝐞𝐫 𝐂𝐨𝐧𝐝𝐢𝐭𝐢𝐨𝐧

At approximately 400 W/m²:

• PV power reduces to around 2 kW.

• Battery charging current further decreases.

• Battery current approaches approximately −10 A under the discussed operating condition.

• DC load power remains around 2 kW.

The battery absorbs less energy because less surplus power is available.

𝐑𝐞𝐝𝐮𝐜𝐞𝐝 𝐖𝐢𝐧𝐝 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐨𝐧

When wind speed decreases:

• Wind turbine output power decreases.

• Wind-side boost-converter power decreases.

• Less surplus energy is available to charge the battery.

• Battery charging current gradually decreases.

An operating value of approximately 2.2 kW is observed from the wind conversion stage during one of the reduced-wind conditions described in the simulation.

𝐋𝐨𝐰 𝐏𝐕 𝐚𝐧𝐝 𝐖𝐢𝐧𝐝 𝐏𝐨𝐰𝐞𝐫

Under very low PV irradiance and low wind speed:

• PV output becomes very small.

• Wind generation also becomes insufficient.

• Renewable generation cannot meet the DC load demand.

• Battery current changes from negative to positive.

• The battery changes from charging to discharging mode.

• The battery supplies the power deficit.

• DC load power remains supported.

This demonstrates effective energy balancing between renewable sources, battery storage, and the load.

𝐒𝐢𝐦𝐮𝐥𝐚𝐭𝐢𝐨𝐧 𝐏𝐚𝐫𝐚𝐦𝐞𝐭𝐞𝐫 𝐒𝐮𝐦𝐦𝐚𝐫𝐲

Parameter	Value
PV module rating	250 W
PV module Vmpp	30.7 V
PV module Impp	8.15 A
PV modules in series	8
PV parallel strings	3
PV maximum power	Approximately 6 kW
PV-side voltage	Approximately 245 V
Wind turbine rating	6.5 kW
Wind-speed conditions	12, 10 and 8 m/s
Generator	PMSG
Battery rated voltage	240 V
Battery capacity	48 Ah
Battery operating voltage observed	Around 260 V
DC-bus reference	400 V
DC load power	Approximately 2 kW
DC load current	Approximately 5 A
MPPT method	Neural Network MPPT
PV converter	Boost converter
Wind converter	Boost converter
Battery converter	Bidirectional DC–DC converter

𝐊𝐞𝐲 𝐅𝐞𝐚𝐭𝐮𝐫𝐞𝐬

• Hybrid PV–wind renewable generation

• Neural Network MPPT for PV

• Neural Network MPPT for wind

• PMSG-based wind energy conversion system

• Independent boost converters for PV and wind

• 400 V common DC bus

• Battery energy storage integration

• Bidirectional battery power flow

• Automatic battery charging and discharging

• DC-bus voltage regulation

• Dynamic solar irradiance testing

• Dynamic wind-speed testing

• Continuous DC load supply

• Renewable power balancing

• Battery SOC monitoring

• PV, wind, battery, load, and DC-bus measurements

𝐀𝐩𝐩𝐥𝐢𝐜𝐚𝐭𝐢𝐨𝐧𝐬

This type of DC microgrid architecture can be useful for studying:

• Renewable-energy-based DC microgrids

• Solar and wind hybrid power systems

• Battery-integrated renewable systems

• Intelligent MPPT techniques

• Neural Network control methods

• DC energy management

• Renewable power balancing

• Battery charging and discharging strategies

• DC-bus voltage regulation

• Hybrid renewable energy systems

• Standalone DC power networks

• Research on intelligent renewable-energy control

𝐖𝐡𝐲 𝐔𝐬𝐞 𝐍𝐞𝐮𝐫𝐚𝐥 𝐍𝐞𝐭𝐰𝐨𝐫𝐤 𝐌𝐏𝐏𝐓?

The Neural Network MPPT approach provides an intelligent method for determining the converter duty cycle using measured voltage and current information.

In this simulation, it is especially useful because:

• PV power changes with solar irradiance.

• Wind power changes with wind speed.

• Both renewable sources operate dynamically.

• The MPPT controllers must continuously adjust converter operation.

• The extracted renewable power must be coordinated with battery storage and load demand.

Using trained Neural Networks allows the controller to map measured operating conditions to suitable converter control commands.

𝐌𝐚𝐢𝐧 𝐏𝐨𝐰𝐞𝐫 𝐅𝐥𝐨𝐰 𝐌𝐨𝐝𝐞𝐬

Operating Condition	PV/Wind Power	Battery Operation	Load Supply
High renewable generation	Greater than load demand	Charging	Supplied continuously
Moderate renewable generation	Near load demand	Reduced charging	Supplied continuously
Low renewable generation	Lower than load demand	Discharging	Supplied continuously
Rapid irradiance/wind variation	Variable	Automatically adjusts	Maintained through DC bus

𝐌𝐚𝐣𝐨𝐫 𝐒𝐢𝐦𝐮𝐥𝐚𝐭𝐢𝐨𝐧 𝐎𝐛𝐬𝐞𝐫𝐯𝐚𝐭𝐢𝐨𝐧𝐬

The simulation demonstrates several important behaviors:

• PV power follows changes in solar irradiance.

• Wind power follows changes in wind speed.

• Neural Network MPPT controls both boost converters.

• Battery charging level decreases when renewable generation decreases.

• Battery switches to discharge mode when renewable generation becomes insufficient.

• Battery SOC increases during charging operation.

• The DC bus remains close to the required 400 V.

• The approximately 2 kW DC load continues receiving power during changing renewable conditions.

• The battery acts as the main balancing source between generation and load demand.

𝐂𝐨𝐧𝐜𝐥𝐮𝐬𝐢𝐨𝐧

The PV Wind Battery Based DC Microgrid with Neural Network MPPT demonstrates the coordinated operation of solar PV, wind energy, and battery storage in a common DC network.

The PV and wind systems use Neural Network MPPT controllers to extract available renewable power, while the battery is interfaced through a bidirectional DC–DC converter to balance power within the microgrid.

Under high renewable generation, surplus power charges the battery. As solar irradiance and wind speed decrease, battery charging reduces. When renewable generation becomes insufficient, the battery automatically changes to discharging mode and supports the DC load.

The simulation therefore provides a clear platform for students, researchers, and engineers to understand hybrid renewable energy integration, intelligent MPPT control, battery energy management, bidirectional power flow, and DC-bus voltage regulation in a PV–wind–battery DC microgrid.