Introducing
Your new presentation assistant.
Refine, enhance, and tailor your content, source relevant images, and edit visuals quicker than ever before.
Trending searches
Predict temperature of an experimental Electric Motor
Motor is a prototype, tested only on a bench
Anonymized real Data
White noise to simulate
real-driving cycles
Target
2Hz
Sensors fall offline for long time:
this Model can replace NA or simple interpolations
high temperatures are more
important
Sensors are expensive and can get damaged:
this Model can replace them
Autos can adjust the cooling of their motor without a specific sensor
Testing different kind of
Recurrent Neural Networks
Testing new trends in Literature
Seq2Val training to shuffle data
and speed-up learning process
Using Mean Squared Error as Loss-function in a first trial
Using a weighted version of MSE to improve prediction on high temperatures
Simplest approach: something in the past
is reused in the present
Improved version: the Model
learns what to remember
A-sort-of generalized version
Features are channels of the same
time series
Information is extracted and then
flows through a feed-forward NN
Each Model is optimized through
Auto-ML - GP with EI
After Auto-ML, models are evaluated to find the most fitting one
Ideal is with high performance
and low parameters
Using the Weighted MSE
Using the MSE