Scribd
Upload
14 views10 pages

Presentation 2

The document discusses air quality sensor calibration including the concepts of bias, accuracy, and linearity. It also discusses best practices for sensor placement and how neural networks can help with tasks like data preprocessing, anomaly detection, and optimizing sensor placement.

Uploaded by

Sped
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
Download as pptx, pdf, or txt
14 views10 pages

Presentation 2

The document discusses air quality sensor calibration including the concepts of bias, accuracy, and linearity. It also discusses best practices for sensor placement and how neural networks can help with tasks like data preprocessing, anomaly detection, and optimizing sensor placement.

Uploaded by

Sped
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1/ 10

Hazza Al Ameri

Mansour Al Nahdi
AIR QUALITY SENSOR CALLIBRATION
• Air quality sensor calibration is the process of correcting or adjusting the sensor
measurements by juxtaposition with a known/reference concentration of the
parameters.

• Calibration is an essential part of quality control of the sensors as it helps identify and
eliminate data drifts.

• Drift is a deviation in the sensor readings due to external influences such as


temperature fluctuations or environmental conditions. It leads to a change in the
measurements over time leading to misinformation.
AIR QUALITY SENSOR CALLIBRATION
• Bias is a persistent systematic error in measurement which always maintains a constant drift from
the actual value. The output might show a higher or lower reading than the true value.

• Accuracy is the level of agreement or the closeness of values reported by the sensor with respect
to the reference measurements/ known values of the target parameter.

• R2 or coefficient of determination value is the measure of sensor performance as compared to a


reference station.

• R2 value of 1 indicates a perfect correlation with the reference-grade monitor.


AIR QUALITY SENSOR CALLIBRATION

Linearity refers to the efficiency.


Framework
PLACEMENT OF AIR QUALITY
SENSOR
•IT IS IDEAL TO PLACE INDOOR SENSORS NEAR THE TYPICAL BREATHING ZONE HEIGHT (1
– 2 METERS). SENSORS SHOULD BE PLACED AWAY FROM AIR POLLUTION SOURCES, LIKE A
TOASTER, AND AIR POLLUTION SINKS, LIKE AIR CLEANERS, TO GET A MORE
REPRESENTATIVE MEASURE OF INDOOR AIR QUALITY. SENSORS SHOULD HAVE FREE AIR
FLOW AND NOT BE PLACED BEHIND FURNITURE OR TUCKED AWAY IN CORNERS.
WINDOWS, DOORS, AND HEATING, VENTILATION, AND AIR CONDITIONING (HVAC) DUCTS
CAN INTRODUCE RAPIDLY CHANGING TEMPERATURE AND RELATIVE HUMIDITY
CONDITIONS, WHICH MAY ADVERSELY IMPACT SOME SENSORS. ADDITIONALLY, AIR
QUALITY CONDITIONS NEAR DOORS, WINDOWS, AND DUCT INLETS OR EXITS MAY BE
OVERLY INFLUENCED BY EXTERNAL SOURCES AND NOT BE REPRESENTATIVE OF
AVERAGE INDOOR CONCENTRATIONS.
NUERAL NETWORK AND ITS BENEFITS TO US
• Data Preprocessing and Feature Engineering:
 Automatic feature extraction from raw sensor data.
 Useful for large, complex datasets.

• Anomaly Detection:
 Recognizes normal patterns.
 Identifies anomalies.
 Identifies potential problems like sudden pollution increasing.

• Data Denoising:
 Neural networks can decrease the noise in your data
 Improves dataset quality.
NUERAL NETWORK AND ITS BENEFITS TO US
• Visualization:
 Neural networks can assist in creating visualizations that make it easier to explain complex
relationships within the data.

• Optimization of Sensor Placement:


 Neural networks can assist in optimizing the placement of sensors by analyzing the locational
dependencies in air quality data.

• Model Validation and Evaluation:


 Neural network can assess our data collection.
 Evaluate the reliability of our collected data.
USING A NEURAL NETWORK TO ANALYZE DATA
• Steps:

1. Define the Problem 7. Train the Model


2. Collect and Prepare Data 8. Evaluate Performance
3. Data Preprocessing 9. Test the Model
4. Choose a Neural Network Architecture 10.Explain the results
5. Build the Neural Network 11.Improve
6. Compile the Model 12.Deploy the Model
NEURAL
NETWORK & https://realpython.com/python-ai-neural-network/#python-ai-starting-t
o-build-your-first-neural-network
PYTHON

You might also like