Solar PV Battery Charging with ANN and P&O MPPT in MATLAB

Solar PV Battery Charging with ANN and P&O MPPT in MATLAB

𝐈𝐧𝐭𝐫𝐨𝐝𝐮𝐜𝐭𝐢𝐨𝐧

Solar PV battery charging with ANN and P&O MPPT in MATLAB/Simulink demonstrates two different Maximum Power Point Tracking techniques for extracting solar energy and charging a battery through a DC–DC Zeta converter.

Solar PV battery charging with ANN and P&O MPPT In MATLAB

₹10,000.00₹5,000.00

Buy Now

The model compares an Artificial Neural Network (ANN) MPPT controller with a conventional Perturb and Observe (P&O) MPPT controller under changing solar irradiation conditions. The main objective is to observe PV power tracking, battery charging behavior, transient response, and MPPT performance.

SEO Meta Description: MATLAB/Simulink model of a solar PV battery charger using ANN and P&O MPPT with a Zeta converter. Compare PV power tracking, battery charging, and dynamic performance under changing irradiation.

SEO Keywords: Solar PV battery charging MATLAB, ANN MPPT MATLAB, P&O MPPT Simulink, solar battery charger MATLAB, Zeta converter MATLAB, neural network MPPT, photovoltaic MPPT comparison, battery charging Simulink, solar charger controller

Solar PV output varies mainly with:

• Solar irradiation

• Cell temperature

• Operating voltage

• Connected load and battery condition

To obtain the available maximum power from the PV array, an MPPT controller is required.

This MATLAB/Simulink model implements and compares:

• ANN-based MPPT

• P&O-based MPPT

Both controllers operate the same solar PV battery charging system so that their responses can be compared under identical operating conditions.

𝐒𝐲𝐬𝐭𝐞𝐦 𝐎𝐯𝐞𝐫𝐯𝐢𝐞𝐰

The complete system contains:

• Solar PV array

• DC–DC Zeta converter

• ANN MPPT controller

• P&O MPPT controller

• Battery charging control

• Battery SOC protection logic

• Load

• PV measurement

• Battery measurement

• Load measurement

The PV array supplies power to the load and simultaneously provides the remaining available power for battery charging.

Main System Parameters

Parameter	Value
Single PV module power	250 W
PV voltage at maximum power point	39.9 V
PV current at maximum power point	8.1 A
Modules connected in series per string	4
Parallel strings	2
Maximum PV array power	Approximately 2000 W
Standard irradiation	1000 W/m²
Cell temperature	25°C
Battery voltage during simulation	Approximately 48 V
Battery capacity	200 Ah
Load power	Approximately 500 W
Irradiation change time	5 s
Reduced irradiation	500 W/m²

𝐏𝐕 𝐀𝐫𝐫𝐚𝐲 𝐂𝐨𝐧𝐟𝐢𝐠𝐮𝐫𝐚𝐭𝐢𝐨𝐧

The solar array is formed using multiple PV modules.

The configuration used in the simulation is:

• 4 PV modules in series in each string

• 2 parallel strings

• Total installed PV capacity of approximately 2 kW

Under:

• 1000 W/m² irradiation

• 25°C cell temperature

the PV system can generate close to its rated maximum power.

When irradiation is reduced to 500 W/m², the available PV power decreases accordingly.

𝐙𝐞𝐭𝐚 𝐂𝐨𝐧𝐯𝐞𝐫𝐭𝐞𝐫

A DC–DC Zeta converter is used between the PV array and the battery/load side.

A major advantage of this converter for the demonstrated system is its ability to support:

• Step-up operation

• Step-down operation

• Controlled PV operating point

• Battery charging

• MPPT-based power transfer

The converter switching signal is generated according to the duty cycle provided by the MPPT and battery charging controller.

𝐀𝐍𝐍 𝐌𝐏𝐏𝐓 𝐂𝐨𝐧𝐭𝐫𝐨𝐥𝐥𝐞𝐫

The ANN-based MPPT controller uses an Artificial Neural Network to determine the required PV operating reference.

ANN Inputs

The neural network receives:

• Solar irradiation

• Temperature

Based on these inputs, the ANN estimates the voltage corresponding to the required maximum-power operating region.

ANN MPPT Process

The control process can be summarized as:

1. Irradiation and temperature are measured.

2. These signals are supplied to the ANN model.

3. ANN generates the required PV voltage reference.

4. The reference is compared with the measured PV voltage.

5. The voltage error is processed by a controller.

6. A required converter duty cycle is generated.

7. The PWM generator produces switching pulses.

8. The Zeta converter adjusts the PV operating point.

9. Maximum available PV power is transferred to the battery and load.

𝐁𝐚𝐭𝐭𝐞𝐫𝐲 𝐂𝐡𝐚𝐫𝐠𝐢𝐧𝐠 𝐏𝐫𝐨𝐭𝐞𝐜𝐭𝐢𝐨𝐧

The MPPT duty cycle is not supplied continuously without considering battery condition.

The charging logic monitors:

• Battery State of Charge

• Battery voltage

• Fully charged battery voltage limit

The converter switching signal is allowed when the battery is within the permitted charging range.

Charging Logic

Battery Condition	Charging Action
SOC below 100% and voltage below full-charge limit	Charging enabled
SOC reaches or exceeds 100%	Charging disabled
Battery voltage reaches full-charge limit	Charging disabled

This control helps prevent continued charging after the battery reaches its permitted charging condition.

𝐏&𝐎 𝐌𝐏𝐏𝐓 𝐂𝐨𝐧𝐭𝐫𝐨𝐥𝐥𝐞𝐫

The second model uses conventional Perturb and Observe MPPT.

Unlike the ANN controller, the P&O algorithm determines the operating direction by continuously monitoring changes in:

• PV voltage

• PV current

• PV power

The controller makes a small change in the converter duty cycle and observes the resulting PV power.

Depending on whether the operating condition moves toward or away from the maximum power point, the controller changes the direction of the next duty-cycle adjustment.

𝐖𝐨𝐫𝐤𝐢𝐧𝐠 𝐏𝐫𝐨𝐜𝐞𝐬𝐬

The same operating conditions are applied to both ANN and P&O models.

Initial Condition

For the first 5 seconds:

• Irradiation = 1000 W/m²

• Temperature = 25°C

The PV array operates near its high-power condition.

Irradiation Change

After 5 seconds:

• Irradiation changes from 1000 W/m² to 500 W/m²

This step change is used to analyze:

• MPPT tracking response

• PV voltage response

• PV current response

• PV power response

• Battery charging current

• Battery SOC

• Load power

• Transient performance

𝐌𝐞𝐚𝐬𝐮𝐫𝐞𝐝 𝐒𝐢𝐠𝐧𝐚𝐥𝐬

Several measurements are available in the Simulink model.

PV Measurements

• PV voltage

• PV current

• PV power

Battery Measurements

• Battery voltage

• Battery current

• Battery SOC

Load Measurements

• Load voltage

• Load current

• Load power

These measurements make it easier to compare the behavior of both MPPT controllers.

𝐒𝐢𝐦𝐮𝐥𝐚𝐭𝐢𝐨𝐧 𝐑𝐞𝐬𝐮𝐥𝐭𝐬

ANN MPPT Performance

At approximately 1000 W/m², the PV array produces close to:

• PV power: 2000 W

When the irradiation changes to 500 W/m², the PV output falls to approximately:

• PV power: 1000 W

The ANN controller provides a relatively smooth transition between the two irradiation levels.

PV Power Comparison

Time/Condition	Irradiation	Approx. PV Power
Initial operating condition	1000 W/m²	2000 W
After 5 seconds	500 W/m²	1000 W

This demonstrates the effect of solar irradiation on available photovoltaic power.

𝐁𝐚𝐭𝐭𝐞𝐫𝐲 𝐂𝐡𝐚𝐫𝐠𝐢𝐧𝐠 𝐑𝐞𝐬𝐩𝐨𝐧𝐬𝐞

The load consumes approximately 500 W.

Therefore, when PV generation is higher than the load demand, the remaining PV energy is available for battery charging.

During high irradiation, the demonstrated battery charging current is approximately:

• −30 A

After irradiation decreases to 500 W/m², the charging current reduces to approximately:

• −10 A

The negative current direction represents the battery charging convention used in the simulation.

Battery Response

Operating Condition	Approx. Battery Current
PV irradiation at 1000 W/m²	−30 A
PV irradiation at 500 W/m²	−10 A

The SOC gradually increases according to the available charging current.

A higher charging current produces a faster SOC increase, while lower PV generation results in a smaller charging rate.

𝐋𝐨𝐚𝐝 𝐏𝐞𝐫𝐟𝐨𝐫𝐦𝐚𝐧𝐜𝐞

The load power is maintained at approximately:

500 W

Even when solar irradiation changes, the model demonstrates controlled energy distribution between:

• PV array

• Load

• Battery

When PV power is higher than the required load power, the additional available energy is used for battery charging.

𝐀𝐍𝐍 𝐯𝐬 𝐏&𝐎 𝐌𝐏𝐏𝐓

The most important part of the simulation is the comparison between ANN and P&O MPPT.

Performance Aspect	ANN MPPT	P&O MPPT
Input information	Irradiation and temperature	PV voltage and current
Tracking approach	Neural-network-based prediction	Perturbation and observation
PV voltage response	Smoother	More oscillatory
PV current response	Smoother	More transient variation
Irradiation-change response	Faster and smoother	Takes time to settle
Oscillation near operating point	Lower in demonstrated response	More noticeable
Sudden irradiation change	Better transient behavior	Larger temporary deviation
Battery charging response	Smooth	Greater transient variation

𝐏&𝐎 𝐑𝐞𝐬𝐩𝐨𝐧𝐬𝐞 𝐃𝐮𝐫𝐢𝐧𝐠 𝐈𝐫𝐫𝐚𝐝𝐢𝐚𝐭𝐢𝐨𝐧 𝐂𝐡𝐚𝐧𝐠𝐞

When irradiation changes from 1000 W/m² to 500 W/m², the P&O controller shows a more noticeable transient response.

The simulation demonstrates:

• Temporary reduction in PV power

• Larger variation in PV voltage

• Larger variation in PV current

• Additional settling time

• Oscillations before reaching the new operating point

During the transition, PV power temporarily drops significantly before recovering toward the new maximum power level.

This behavior occurs because the P&O controller has to repeatedly perturb the operating point to identify the new maximum-power region.

𝐀𝐍𝐍 𝐑𝐞𝐬𝐩𝐨𝐧𝐬𝐞 𝐃𝐮𝐫𝐢𝐧𝐠 𝐈𝐫𝐫𝐚𝐝𝐢𝐚𝐭𝐢𝐨𝐧 𝐂𝐡𝐚𝐧𝐠𝐞

The ANN controller provides a noticeably smoother response in the demonstrated simulation.

When irradiation changes:

• PV voltage changes smoothly

• PV current transitions with lower oscillation

• PV power reaches the new operating region quickly

• Battery charging current changes smoothly

• Transient disturbance is reduced

This makes the ANN MPPT controller attractive for solar battery charging applications where fast changes in environmental conditions may occur.

𝐊𝐞𝐲 𝐅𝐞𝐚𝐭𝐮𝐫𝐞𝐬

• 2 kW solar PV array

• ANN-based MPPT control

• Conventional P&O MPPT control

• Direct comparison of two MPPT methods

• Zeta DC–DC converter

• Battery charging control

• Battery SOC monitoring

• Battery overcharging protection logic

• Irradiation variation from 1000 to 500 W/m²

• PV voltage, current, and power monitoring

• Battery voltage, current, and SOC monitoring

• Load voltage, current, and power monitoring

• Dynamic MPPT performance analysis

• Battery charging response comparison

𝐖𝐡𝐲 𝐔𝐬𝐞 𝐀𝐍𝐍 𝐟𝐨𝐫 𝐌𝐏𝐏𝐓?

ANN-based MPPT can be useful because a trained neural network can establish the nonlinear relationship between environmental conditions and the required PV operating point.

In this simulation, the ANN uses:

• Irradiation

• Temperature

to predict the required operating voltage.

The demonstrated results show advantages such as:

• Faster dynamic response

• Reduced oscillation

• Smooth PV power tracking

• Better behavior during irradiation variation

• Smooth battery charging transition

𝐀𝐩𝐩𝐥𝐢𝐜𝐚𝐭𝐢𝐨𝐧𝐬

This MATLAB/Simulink model is useful for understanding:

• Solar PV battery charging systems

• ANN-based MPPT control

• P&O MPPT implementation

• Neural-network-based renewable energy control

• DC–DC converter control

• PV energy management

• Battery charging systems

• Solar energy storage

• MPPT performance comparison

• Renewable energy control research

• Intelligent power electronic converters

𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 𝐎𝐮𝐭𝐜𝐨𝐦𝐞𝐬

By studying this model, students, researchers, and engineers can understand:

• How a PV array responds to irradiation changes

• How ANN can be applied to MPPT control

• How P&O searches for the maximum power point

• How the duty cycle controls the Zeta converter

• How battery SOC can be included in charging logic

• How battery charging current changes with available solar power

• How ANN and P&O responses can be compared using simulation waveforms

• Why transient response is important in MPPT systems

𝐂𝐨𝐧𝐜𝐥𝐮𝐬𝐢𝐨𝐧

The Solar PV Battery Charging with ANN and P&O MPPT in MATLAB/Simulink model demonstrates two different approaches for maximum power extraction and battery charging.

Under the same changing irradiation conditions:

• Both controllers track the available solar power.

• PV output changes from approximately 2000 W to 1000 W when irradiation changes from 1000 W/m² to 500 W/m².

• Battery charging current changes according to the available PV power.

• P&O MPPT shows greater oscillation and transient deviation during irradiation changes.

• The ANN MPPT provides a smoother and faster response in the demonstrated simulation.

Overall, the model provides a clear comparison of intelligent ANN-based MPPT and conventional P&O MPPT for solar PV battery charging using a Zeta converter.