A methods on Windmills for checking predicting fault in the generator using defined sensors and hence preventing the generator from reaching a complete halt.
This notebook is an implementation of “Pedro Pedrosa Rebouças Filho, Navar M.M. Nascimento, Igor R. Sousa, Cláudio M.S. Medeiros, Victor Hugo C. de Albuquerque, A reliable approach for detection of incipient faults of short-circuits in induction generators using machine learning,”
The aim of this paper was to detect the incipient of faults or short circuit occurring in the induction generator thereby allowing us to pinpoint where the actual error took place.
Now let’s have a look at the necessary imports that are required to run this notebook
As you can see, Seed has also been sent in this notebook. This ensures that even when the weights and biases are being randomly assigned my results can be achieved by you too as the seed value will take care of that.
Let’s have a look at the data that we are having with us
This is the head of the data meaning only 5 rows are being shown
Now let’s have a look at the model we have build.
As you can see the model take very few parameters which is idea for a SBC’s or microchips to work with.
Now, lets have a look at the graphs showing us whether the model which we have build can actually help us or not.
As you can see we have reached an accuracy of more than 80%. This accuracy will help us in detecting and pin pointing the issues in an operating windmill. Hence the time which is wasted when a windmill stops or malfunction will be reduced.
Notebook Link: Here
Credit: vishal yadav
Also Read: Fuel Efficiency Prediction using Deep Learning