際際滷

際際滷Share a Scribd company logo

Data Analysis in Engineering

Download as PPTX, PDF
A
Ahmed Rezk

This document discusses fast data analysis in automotive engineering. It outlines the typical workflow for data analysis, including accessing data, pre-processing data, performing analysis, creating models, and deploying models. It identifies some common challenges, such as working with large or messy data files in different formats, dealing with missing or abnormal data values, and ensuring user experience. It provides recommendations for faster data analysis, including automating the analysis workflow, using a unified platform that allows easy interfacing with test benches and flexible deployment options, and reducing the amount of coding required.

1 of 4
Download to read offline
Fast Data Analysis in Automotive Engineering. Dr. Ahmed Rezk 1
Data Analysis. Data analysis work flow. Access Data Pre-process Data Data Analysis Model Model Deployment Data file Strain sensor Temp. sensor Sound sensor Press. sensor Working with messy data Data Reduction Efficiency quality Cost Optimisation 0.064 0.066 0.068 0.07 0.072 0.074 0.076 0.078 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 MassflowRate[kg/s] Time[ s ] Test Data Pred Data Prediction Scripting for similar use Validation - Platform with statistical capability is required - User interface software is preferable, for fast processing Coding required System Data analysis model Other Model or Hardware Interface Regression 2
Challenges of Data Analysis Access data file Pre-process Data Model Model Deployment 1- User experience. a- Developing the model. b- Code generation. 2- Model interface. 1- Big data file(s) i.e. DBF. 2- Different files format. 3- Resampling. 4- Missing data, zeros and spikes. Unified Platform 1D Model GT-Power Ricardo wave LMS-AMESIM 3D Model ANSYS COMSOL Hardware Control/Map Automation .dll .CPP PLC HDL Fast Data Analysis: 1- Defined your needs and the stage of the workflow. 2- Platform benchmarking. Faster Data Analysis: 1- Data analysis automation. 2- Platform. a- Unified platform. b- Easy to interface with test bench. c- Versatile deployment capabilities. d- Less coding time. 3
Questions 4References: https://uk.mathworks.com/videos/data-analytics-with-matlab-99066.html

More Related Content

Data Analysis in Engineering

  • 1. Fast Data Analysis in Automotive Engineering. Dr. Ahmed Rezk 1
  • 2. Data Analysis. Data analysis work flow. Access Data Pre-process Data Data Analysis Model Model Deployment Data file Strain sensor Temp. sensor Sound sensor Press. sensor Working with messy data Data Reduction Efficiency quality Cost Optimisation 0.064 0.066 0.068 0.07 0.072 0.074 0.076 0.078 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 MassflowRate[kg/s] Time[ s ] Test Data Pred Data Prediction Scripting for similar use Validation - Platform with statistical capability is required - User interface software is preferable, for fast processing Coding required System Data analysis model Other Model or Hardware Interface Regression 2
  • 3. Challenges of Data Analysis Access data file Pre-process Data Model Model Deployment 1- User experience. a- Developing the model. b- Code generation. 2- Model interface. 1- Big data file(s) i.e. DBF. 2- Different files format. 3- Resampling. 4- Missing data, zeros and spikes. Unified Platform 1D Model GT-Power Ricardo wave LMS-AMESIM 3D Model ANSYS COMSOL Hardware Control/Map Automation .dll .CPP PLC HDL Fast Data Analysis: 1- Defined your needs and the stage of the workflow. 2- Platform benchmarking. Faster Data Analysis: 1- Data analysis automation. 2- Platform. a- Unified platform. b- Easy to interface with test bench. c- Versatile deployment capabilities. d- Less coding time. 3