Embedded Files
USING STANDARD OFFLINE AI TOOLS LIKE TENSORFLOW TO IDENTIFY 11 DIFFERENT TYPES OF APPLIANCES FROM THEIR ELECTRIC SIGNATURES i.e SPECTROGRAMS.
USING STANDARD OFFLINE AI TOOLS LIKE TENSORFLOW TO IDENTIFY 11 DIFFERENT TYPES OF APPLIANCES FROM THEIR ELECTRIC SIGNATURES i.e SPECTROGRAMS.
EIA’s IEO2017 presents an assessment of long-term world energy markets. IEO2017 energy consumption projections are provided for 16 regions of the world divided according to OECD and non-OECD membership.
EIA’s IEO2017 presents an assessment of long-term world energy markets. IEO2017 energy consumption projections are provided for 16 regions of the world divided according to OECD and non-OECD membership.
The capstone project was a challenge to design a fully automated analyzing arrangement within a single offline script. The automation should even consist un-zipping of the image as well as excel data and direct application. The model was build using multiple neural network techniques including Conv2D (32,63,128), Flatten NN and Dense NN.
The capstone project was a challenge to design a fully automated analyzing arrangement within a single offline script. The automation should even consist un-zipping of the image as well as excel data and direct application. The model was build using multiple neural network techniques including Conv2D (32,63,128), Flatten NN and Dense NN.
The plug load dataset contained current and voltage measurements sampled at 30 kHz from 11 different appliance types present in more than 60 households in Pittsburgh, Pennsylvania. Plus load refers to the energy used by products that are powered utilizing an ordinary AC plug.
The plug load dataset contained current and voltage measurements sampled at 30 kHz from 11 different appliance types present in more than 60 households in Pittsburgh, Pennsylvania. Plus load refers to the energy used by products that are powered utilizing an ordinary AC plug.
For each appliance, plug load measurements were post-processed to extract a three-second-long window of measurements of current and voltage. For some observations, the window contains both the startup transient state (turning the appliance on) as well as the steady-state operation (once the appliance is running).
For each appliance, plug load measurements were post-processed to extract a three-second-long window of measurements of current and voltage. For some observations, the window contains both the startup transient state (turning the appliance on) as well as the steady-state operation (once the appliance is running).
For others, the window only contains the steady-state operation. The observations were then transformed into two spectrograms, one for current, and one for voltage.
For others, the window only contains the steady-state operation. The observations were then transformed into two spectrograms, one for current, and one for voltage.
MAIN COMPONENTS OF THE PROJECT
MAIN COMPONENTS OF THE PROJECT
Neural Network
Neural Network
Conv2D (32,63,128), Flatten NN and Dense NN
Spectrogram
Spectrogram
Quantified by current and voltage measurements
RESULT AND CERTIFICATION OF THE PROJECT
RESULT AND CERTIFICATION OF THE PROJECT
Page updated
Google Sites
Report abuse