Introduction to Grid-Tied PV Systems
A grid-connected photovoltaic system allows solar energy to be harnessed and fed directly into the power grid, providing clean energy to households and industries. However, the power output of PV systems depends heavily on external factors such as sunlight intensity and temperature. MPPT algorithms help optimize the output by continuously adjusting the system's operating point to extract the maximum possible power from the PV panels.
In this particular system, an Artificial Neural Network (ANN)-based MPPT is employed to control the operation of the grid-tied PV system. By using ANN, the system can adapt to varying environmental conditions and effectively manage power generation.
Key Components of the System
The system consists of a 154 MW, 34 kV main grid with a 34 kV to 400 V transformer, and two AC loads (500 kW and 30 kW). The PV array used in the system is rated at 41 kW, with a line-to-line voltage of 400 V, making it suitable for grid connection. Additionally, the system operates at a fixed grid frequency of 50 Hz.
The PV panels in the system are designed to produce a maximum power output of 414 W per panel, with an open-circuit voltage of 85.3 V and a short-circuit current of 6.09 A. These panels are connected in series to form a larger string, contributing to the overall system’s power generation.
PV Panel Details and System Simulation
The PV array operates under varying irradiation and temperature conditions. In the simulation, the irradiation is varied between 0 and 1000 W/m², and the temperature ranges from 15°C to 35°C. These variations help in training the ANN model, which is designed to predict the maximum power point of the system based on the environmental inputs.
Using this data, the system calculates the maximum power voltage and current for the PV panels, which are crucial for the ANN model. The input data for the ANN includes solar irradiation and temperature, while the output data includes the corresponding maximum power voltage.
Training the Neural Network Model
The next step in the process is training the Artificial Neural Network (ANN). The input data (irradiation and temperature) is used to train the neural network to predict the reference voltage for maximum power extraction. Once trained, the ANN model can generate the optimal reference voltage for the inverter.
During training, the system measures how well the predicted data aligns with the actual performance of the PV panels. An R-value of 1 indicates a perfect match, demonstrating that the ANN model is functioning accurately. This trained model is then integrated into the simulation to manage the inverter control effectively.
Controlling the Inverter Using MPPT
The ANN-generated reference voltage is used to control the inverter and extract maximum power from the PV system. The inverter is responsible for converting the DC power generated by the solar panels into AC power that can be fed into the grid.
In the system, real power is sent from the PV system to the grid, while the reactive power is kept at zero. This is accomplished by regulating the current using a controller that ensures only real power is transferred to the grid. By converting AC quantities to DC quantities, the system improves the efficiency of the power conversion process, making it easier to manage the system’s operation.
Managing Power Flow Between the Grid and Loads
The power flow between the PV system, grid, and AC loads is constantly monitored. The system is designed to manage different load conditions by ensuring that the grid supplies power when solar power is insufficient, and vice versa. For example, if the solar power generation drops due to cloud cover or other factors, the grid compensates for the deficit.
The PV system can supply power to different loads depending on their requirements. If the solar panels generate more power than the loads require, the excess energy is sent to the grid. Conversely, if solar power is insufficient, the grid supplies the additional required power.
Impact of Changing Irradiation Conditions
The system also accounts for changes in irradiation. If the irradiation level drops (e.g., from 1000 W/m² to 500 W/m²), the system adjusts to the new conditions, reducing the power output from the PV panels accordingly. In this case, the solar power output decreases, and the grid helps to meet the power demand.
The MPPT algorithm ensures that, even in changing environmental conditions, the system continues to extract the maximum power possible from the PV array. This dynamic adjustment is crucial for maintaining a stable and efficient power supply to the grid.
Conclusion: Efficiency of Neural Network-Based MPPT
In conclusion, the integration of Artificial Neural Networks in MPPT for grid-tied PV systems significantly improves the efficiency of solar energy extraction. By continuously adapting to changes in temperature and irradiation, the system ensures maximum power generation from the PV panels, optimizing the overall energy output. This approach not only enhances the performance of the system but also makes it more resilient to environmental fluctuations, helping to meet energy demands more effectively.
Comentarios