Download as pdf or txt
Download as pdf or txt
You are on page 1of 2

➢ maximum power point tracking (MPPT) algorithms have been introduced to ensure that the

PV generator operates at or near to the MPP, one of the most widely used MPPT
techniques is Perturb and observe (P&O) and Incremental Conductance (InC) algorithms
because of their simple feedback structure, the ease of implementation and the reduced
number of their parameters . However, these techniques exhibit some limitations during
steady state condition and fast change in the environmental conditions

➢ Nowadays, MPPT techniques based on artificial intelligence such as Fuzzy Logic


Controller (FLC), Artificial Neural Networks (ANN), fuzzy Inference System (ANFIS), and
Genetic Algorithms (GA) are proposed to solve these limitations and improve the steady
and dynamic states

➢ But we are using Artificial neural network technology in our project for maximizing the
efficiency and accuracy during the tracking process under different weather conditions.

To train an ANN for a hybrid microgrid, you need data on various parameters like solar
power generation, wind speed, electricity consumption, and battery status. Install sensors
and meters to measure these values.

As from wind Pannel , outputs current and voltage of rectifier and duty cycle obtain from
P&O algo are feed as input to neural network

Similarly from PV Pannel , output current, voltage and duty cycle for solar power boost
converter are fed as input to neural network

These all inpute and target parameter are collected in neural network . After collection of
large set of these data NN are trained.

When our neural network is completely trained then it generate target value corresponding
to inpute parameter.

In this slide we will talk about how neural network is beneficial than conventional algorithm.

• Adaptability:
Adaptability for Artificial Neural Networks (ANNs) means that they can change and improve their
behaviour based on the data they receive. It's like how people learn from their experiences and
become better at doing something over time.

In a microgrid, adaptability with ANNs means that if the energy production or consumption patterns
change, perhaps due to weather fluctuations or variations in demand, the ANNs can modify their
decisions on how to distribute and manage energy within the microgrid. This adaptability helps
optimize the microgrid's performance and ensure that it operates efficiently under different
circumstances.

• Nonlinear Modelling

Neural networks are highly flexible and can model complex, nonlinear relationships in data.
This can be advantageous when dealing with systems that have nonlinear characteristics,
such as photovoltaic systems with varying environmental conditions.

You might also like