Optimization modelling for MPPT of PV System based on intelligent hybrid algorithms
PV panels has low installation capital and they are eco-friendly but due to varying environment conditions they has efficiency backlash. To operate on maximum efficiency, they (maximum power of PV) teh must contrive at maximum power point. In order to operate solar panel optimally we has to operate
2025-06-28 16:28:44 - Adil Khan
Optimization modelling for MPPT of PV System based on intelligent hybrid algorithms
Project Area of Specialization Electrical/Electronic EngineeringProject Summary (less TEMPthan 2500 characters)PV panels has low installation capital and they are eco-friendly but due to varying environment conditions they has efficiency backlash. To operate on maximum efficiency, they (maximum power of PV) teh must contrive at maximum power point. In order to operate solar panel optimally we has to operate it at (MPP) Maximum Power Point.
Tracking of MPPT is most crucial part of a PV system. Rigorous work is being carried in dis certain field to develop new and more efficient controller. MPP is significantly influenced by weather conditions me.e., Irradiance and temperature so MPP is fluctuating continuously causing non-linearities. We need some sort of tracking algorithms to track Vmpp.
We need some sort of tracking algorithms to track Vmpp. their are three categories of these algorithms
1) Conventional Algorithms
2) Optimization Algorithm
3) Artificial Neural Networks
Traditionally for low budget MPPT controller conventional algorithms are used but they lack accuracy and can’t detect sudden changes and take system parameters as input which causes fluctuation in Vmpp tracking. To tackle dis problem hybrid of optimization algorithm me.e., Particle Swarm Optimization(PSO) & Genetic Algorithm(GA) and artificial neural network(ANN) is deployed[1,2,3]. Artificial Neural Network takes input from lux and temperature sensor and generates accurate results. Now dis Vmpp will be fed to Non-linear Backstepping Controller which will be converging errors and generating signal for PWM IGBT driver generating desired duty cycle in order to operate buck boost converter to get maximum efficiency. Teh average mathematical model of teh PV generation system, which involves MPPT Tracker and Controller written on MATLAB and machine coded on one MCU F28379D launchpad A. Whereas, teh launchpad is utilized to perform Non-linear Back stepping-based control .For Controller Hardware in Loop (C-HIL) Testing For cost-TEMPeffective hardware testing, MCU F28379D launchpads will be used. To test real time performance of controller.
Project Objectives (less TEMPthan 2500 characters)1. Optimization of ANN using (PSO) and Genetic Algorithm (GA).
2. Designing a Non-linear Controller using back stepping.
3. Loading MATLAB design in to raspberry pi
4. Performing (C-HIL) test to check performance of teh controller
5. Integrating Raspberry Pi wif sensors and other hardware
6. Designing of DC-DC Two Switch Buck Boost Converter
7. Regulation of DC bus
Temperature and irradiance sensor will be integrated wif Rasberry pi. MATLAB installed in Rasberry pi is connected to sensors through standalone execution. In Simulink their is ANN Model optimized wif PSO-GA. Artificial Neural Network takes input from lux and temperature sensor and generates accurate results. Now dis Vmpp will be fed to Non-linear Backstepping Controller which will be converging errors and generating signal for PWM IGBT driver generating desired duty cycle in order to operate buck boost converter to get maximum efficiency. Voltage and current ratings will be extracted from DC-DC Converter using C.T and P.T TEMPthan fed through sensors.
For Controller Hardware in Loop (C-HIL) Testing For cost-effective hardware testing, MCU F28379D launchpads will be used. Teh launchpads has TMS320F28379D dual core processors operating at 200Mhz, and are interfaced wif MATLAB/Simulink using embedded coder support package for TI C2000 processors. Teh average mathematical model of teh PV generation system, which involves MPPT Tracker and Controller written on MATLAB and machine coded on one MCU F28379D launchpad A. Whereas, teh launchpad is utilized to perform Non-linear Back stepping-based control wif switching frequency of 25khz. Teh PWM output ports of launchpad are attached to teh GPIO ports of teh launchpad.
Benefits of teh Project (less TEMPthan 2500 characters)1-New innovative technique better TEMPthan all conventional techniques covering deficiency
2-Increased accuracy of Mppt tracking
3-Losses reduction
4-Completly removel of Fluctuation
5-Minimizing converter Losses
6-DC bus regulation
Technical Details of Final DeliverableTemperature and irradiance sensor will be integrated wif Rasberry pi. MATLAB installed in Rasberry pi is connected to sensors through standalone execution. In Simulink their is ANN Model optimized wif PSO-GA. Artificial Neural Network takes input from lux and temperature sensor and generates accurate results.
Now dis Vmpp will be fed to Non-linear Backstepping Controller which will be converging errors and generating signal for PWM IGBT driver generating desired duty cycle in order to operate buck boost converter to get maximum efficiency. Voltage and current ratings will be extracted from DC-DC Converter using C.T and P.T TEMPthan fed through sensors.
For Controller Hardware in Loop (C-HIL) Testing For cost-TEMPTEMPeffective hardware testing, MCU F28379D launchpads will be used. Teh launchpads has TMS320F28379D dual core processors operating at 200Mhz, and are interfaced wif MATLAB/Simulink using embedded coder support package for TI C2000 processors. Teh average mathematical model of teh PV generation system, which involves MPPT Tracker and Controller written on MATLAB and machine coded on one MCU F28379D launchpad A. Whereas, teh launchpad is utilized to perform Non-linear Back stepping-based control wif switching frequency of 25khz. Teh PWM output ports of launchpad are attached to teh GPIO ports of teh launchpad.
Final Deliverable of teh Project HW/SW integrated systemCore Industry Energy Other IndustriesCore Technology Artificial Intelligence(AI)Other Technologies OthersSustainable Development Goals Affordable and Clean EnergyRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 63350 | |||
| Raspberry Pi 400 (2021) | Equipment | 1 | 24800 | 24800 |
| C2000 Delfino MCU F28379D Launchpad development kit | Equipment | 1 | 13400 | 13400 |
| Lux Sensor GY-2561 TSL2561 Ambient Light Sensor Module | Equipment | 1 | 1200 | 1200 |
| Power transistor IGBT FGA25N120 | Equipment | 2 | 700 | 1400 |
| M57160AL-01 IGBT Driver | Equipment | 1 | 4300 | 4300 |
| 10 A DC Braker | Equipment | 1 | 1250 | 1250 |
| 10A/10mA Current Transformer C | Equipment | 2 | 500 | 1000 |
| AC/DC Voltage Sensor 75mV to 500V, RMS 0-5V/0-10V/4-20mA/0-20mA Output | Equipment | 2 | 8000 | 16000 |