You are on page 1of 24

A

MAJOR PROJECT PRESENTATION


ON
“PREDICTION OF AIR POLLUTION USING ARTIFICIAL INTELLIGENCE: A CASE
STUDY OF DELHI NCT”

(BATCH: 2015-2019)
DEPARTMENT OF CIVIL ENGINEERING
Dr. Akhilesh Das Gupta Institute of Technology and Management
(An Institute affiliated to Guru Gobind Singh Indraprastha University)
New Delhi- 110053
GUIDED BY: -
MS. MANINDER KAUR SUBMITTED BY: -
MS. TARUSHI SINGH
DR. PRATUL ARVIND SHIKHAR MITTAL (07015603415)
SANYAM JAIN (36015603415)
CO-ORDINATOR:- SHUBHAM DHANGWAL (41115603415)
MR. VIKAS KATARIA
INTRODUCTION
•Air pollution occurs when harmful or excessive quantities of substances including gases,
particles, and biological molecules are introduced into Earth's atmosphere.
•It may cause diseases, allergies and even death to humans; it may also cause harm to
other living organisms such as animals and food crops, and may damage the natural or built
environment.
OBJECTIVE
The Primary objectives of the study are:

•Study the Air Pollution Data for various locations in New Delhi.
•To identify patterns of spike in Air Pollution levels w.r.t to various monitored parameters.
•Identify the Metrological factors that correlate with the air pollution levels for the respective
locations.
•Develop a Model for predicting the level for key pollutant suspended particulate matter PM2.5 &
PM10.
Data Collection

Data Cleaning, Preparation

Data Normalisation

Data Reduction, Transformation

Develop NN Model

Validate Model
METHODOLOGY

1. Study Area
Among the 38 monitoring station, there are 11 key location for which the survey has been done.
The key locations are as follows:

ANAND VIHAR
CRRI
DTU
EAST ARZUN NAGAR
IHBAS
ITO
NSIT
MANDIR MARG
RK PURAM
PUNJABI BAGH
SHADIPUR
2. Data Collection
The data for the Project was obtained from Central Pollution Control Board (CPCB) website.
Currently, CPCB track the Air Pollution levels across 23 dimensions (variables). Period of
Data collection: January 1, 2015 to December 31, 2018.

POLLUTANTS:
Nitric Oxide (NO)
Carbon Monoxide (CO)
Suspended Particulate Matter/RPM/PM10
Nitrogen Dioxide (NO2)
Sulphur Dioxide (SO2)
PM 2.5 (DUST PM2.5)
Oxides of Nitrogen (NOx)

METEROLOGICAL PARAMETERS:
Relative Humidity (RH)
Temperature
Wind Speed (Wind speed S)
Vertical Wind speed (Wind speed V)
Wind Direction
Solar Radiation.
3. Materials
3.1 INTERPOLATING OF DATA

In the mathematical field of numerical analysis, interpolation is a method of constructing new


data points within the range of a discrete set of known data points. Through Cubic Spline
method, interpolation is done.
3.2 NORMALISATION OF DATA

It is a technique of organizing the data in the database. Normalization is a systematic


approach of decomposing tables to eliminate data redundancy (repetition) and undesirable
characteristics like Insertion, Update and Deletion . It is a multi-step process that puts data
into tabular form, removing duplicated data from the relation tables.
3.3 Principal Component Analysis & Factor Analysis

Principal component analysis (PCA) is a mathematical procedure that transforms a number of (possibly)
correlated variables into a (smaller) number of uncorrelated variables called principal components. Principal
components analysis is similar to another multivariate procedure called Factor Analysis.
Example of PCA report
3.4 Neural Network Tool

A neural network is a computing model whose layered structure resembles the networked
structure of neurons in the brain, with layers of connected nodes. A neural network can learn from
data—so it can be trained to recognize patterns, classify data, and forecast future events.
STEP 1:MATLAB Coding for PCA and without PCA
STEP 2:Select NARX problem
STEP 3:Provide input (Meteorological parameters) & target (Pollutants)

2019/4/26
STEP 4: Network Architecture configuration
STEP 5: Train the data and obtain RMSE(root mean square error) & R (regression) values
STEP 6: Generate the regression plot
RESULTS
PM 10 WITH & WITHOUT PCA

STATIONS REGRESSION REGRESSION

With PCA Without PCA

Punjabi Bagh 0.90315 0.90377

Mandir Marg 0.85518 0.871

Anand Vihar 0.84699 0.85902

R.K.Puram 0.89618 0.8962


2019/4/26
PM 2.5 WITH & WITHOUT PCA
STATIONS REGRESSION REGRESSION
With PCA Without PCA
DTU 0.94285 0.9435
Punjabi Bagh 0.92577 0.93384
Shadipur 0.94063 0.92289
Mandir Marg 0.91055 0.92676
Anand Vihar 0.88803 0.89321
IHBAS 0.50997 0.58022
R.K.Puram 0.88688 0.89129
NSIT 0.93155 0.92467
CONCLUSION
Neural Network Model developed so far is capable to predict the daily average
concentrations of PM2.5 & PM10 one day in advance.

PM 2.5 WITH & WITHOUT PCA


PM10 WITH & WITHOUT PCA
STATIONS REGRESSION

DTU 0.94285 0.9435


STATIONS REGRESSION
Punjabi Bagh 0.92577 0.93384

Shadipur 0.94063 0.92289 Punjabi Bagh 0.90315 0.90377


Mandir Marg 0.91055 0.92676
Mandir Marg 0.85518 0.871
Anand Vihar 0.88803 0.89321

IHBAS 0.50997 0.58022


Anand Vihar 0.84699 0.85902
R.K.Puram 0.88688 0.89129

NSIT 0.93155 0.92467 R.K.Puram 0.89618 0.8962


REFERENCES
 https://app.cpcbccr.com/ccr/#/caaqm-dashboard-all/caaqm-landing
 Kolehmainen.M, (et.al), INTERCOMPARISION OF AIR QUALITY DATA USING
PRINCIPAL COMPONENT ANALYSIS, AND FORECASTING OF PM10 ANND PM2.5
USING ANN, 2010.
 Niharika, (et. al), A NOVEL AIR QUALITY PREDICTION MODEL USING
ARTIFICIAL NEURAL NETWORKS, 2012.
 Barai.S, (et. al), NEURAL NETWORK MODELS FOR AIR QUALITY PREDICTION: A
COMPARATIVE STUDY, 2015.
 Sekar.C, (et.al), POTENTIAL ASSESSMENT OF NEURAL NETWORK AND
DECISION TREE ALGORITHMS FOR FORECASTING AMBIENT PM2.5 AND CO
CONCENTRATIONS: CASE STUDY, 2015.
 Rahman.P, (et.al), USING NEURAL NETWORKS FOR PREDICTION OF AIR
POLLUTION INDEX IN INDUSTRIAL CITY, 2017
 Li.X, (et.al), LONG SHORT TERM MEMORY NEURAL NETWORK FOR AIR
POLLUTANT CONCENTRATION PREDICTIONS, 2017.
 Franceschi.F, (et.al), DISCOVERING RELATIONSHIPS AND FORECASTING PM10
AND PM2.5 CONCENTRATIONS IN BOGOTÁ, COLOMBIA, USING ARTIFICIAL
NEURAL NETWORKS, PRINCIPAL COMPONENT ANALYSIS, AND K-MEANS
CLUSTERING, 2018.