Scribd
Upload
123 views24 pages

For 3rd Review

The document discusses the use of machine learning algorithms to predict groundwater levels, addressing issues caused by climate change and population growth. It outlines objectives such as converting groundwater flow models to machine learning models and providing recommendations to farmers regarding groundwater levels and rainfall. The document also includes a literature survey of various studies and models, accuracy metrics, and the methodology for data collection and model implementation.

Uploaded by

karthick
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
123 views24 pages

For 3rd Review

The document discusses the use of machine learning algorithms to predict groundwater levels, addressing issues caused by climate change and population growth. It outlines objectives such as converting groundwater flow models to machine learning models and providing recommendations to farmers regarding groundwater levels and rainfall. The document also includes a literature survey of various studies and models, accuracy metrics, and the methodology for data collection and model implementation.

Uploaded by

karthick
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 24

Ground water level prediction

using machine learning


algorithm
BY HARSHINI PRIYAA E GUIDED BY
20MCS002 Dr K. R.
BASKARAN
(PROFESSOR CSE)
Problem Statement
 Changes in weather and increase in population has resulted in
higher obstruction rates as caused very concerning changes on both
discharge rates and groundwater level over country. Being reliable
source of good quality water, continual overuse of this water will
cause permanent or temporary decline in available ground water
OBJECTIVE
 To convert the ground water flow models to machine learning models
 Torecommend the farmer about the ground water level and rainfall for
the particular cultivating year in advance
 To sustain the ground water level to prevent the water scarcity
INTRODUCTION
 Groundwater is a highly valuable resource.
 Measurement and analysis of groundwater level is needed for maintaining
groundwater availability.
 The accurate prediction of groundwater levels is essential for sustainable
utilization and management of vital groundwater resources
 For management of groundwater level a model is required which can
predict the groundwater level in future with the current available
information
LITERATURE SURVEY
Year DESCRIPTION
Paper title

Forecasting of 2017 Paper states that the capability of neural networks


Groundwater Level using models in modelling of daily groundwater level using
Artificial Neural Network rainfall, temperature, stream flow and river water level
by incorporating river data as inputs as well as the past value of the
recharge and river bank groundwater level
infiltration

Prediction of ground 2020 Paper introduces various machine learning algorithm


water level based on before which we have collected weather information of
machine learning both monsoon and non-monsoon then checked soil
parameter after that for predicting transient groundwater
levels in groundwater system under variable pumping
and weather.
Year DESCRIPTION
Paper title

Groundwater level 2018 Monthly total rainfall, monthly mean temperature and
prediction using artificial monthly groundwater level data of Reyhanli region
Neural network and M5 were used for the groundwater level prediction.
tree models Artificial Neural Networks (ANN) and M5tree (M5T)
models were investigated in order to improve the
methods to estimate the groundwater level.

Groundwater Level 2016 The potential of neural computing techniques for


Forecasting in A Shallow forecasting ground-water levels is investigated by
Aquifer Using Artificial developing ANN models for a shallow aquifer of Central
Neural Network Approach Godavari Delta System in India. Different ANN models
were developed that forecasts the water level at two
observation wells up to 6 months in advance
Year DESCRIPTION
Paper title

Artificial Intelligence hybrid 2020 Deep learning structure was conducted to estimate
deep learning model for groundwater level. To this end, a Multi-Layer Perceptron
ground water level (MLP) combining an Adam optimization is exerted. The
prediction using MLP-APAM Root Mean Squared Error (RMSE), Mean Absolute Error
(MAE), Mean Squared Error (MSE), and the coefficient of
determination (R2) were used to evaluate the accuracy of
the simulated groundwater level.

Enhancement of ground 2020 The MLP-WA model was the best model among which the
water level prediction input data were just the average of the region temperature
using an integrated and three temporal delays of 3, 6, and 9 months were
machine learning model considered, the models based on the three simultaneous
optimized by whale temperature inputs with mentioned delays had higher
algorithm performance as compared to the inputs just belonging to
temperature input.
PAPER TITLE Year DESCRIPTION
Groundwater 2020 This paper uses a regression-based approach to predict full
prediction using groundwater images based on sequences of monthly
machine learning groundwater maps of the southern part of the African
tools continent using the Gravity Recovery and Climate Experiment
(GRACE) dataset . Five machine-learning techniques are
implemented on the GRACE dataset and compared to predict
pixels in future frames of the dataset

Machine Learning 2017 New approach in the context of two aquifer systems
algorithm for supporting agricultural production in the United States: the
modelling ground High Plains aquifer (HPA) and the Mississippi River Valley
water level changes in alluvial aquifer (MRVA). select input data sets by using a
agricultural regions combination of mutual information, genetic algorithms, and
lag analysis, and then use the selected data sets in a
Multilayer Perceptron network architecture to simulate
seasonal groundwater level change
ACCURACY
MODEL Mean square Mean absolute R Correlation
error error coefficient
ANN[4] 2.26 1.71 0.893
M5T[4] 2.05 1.55 0.892
FNN[5] 0.1346 0.1058 0.697
SVM[8] 0.76
LR[1] 0.62
RNN[5] 0.825
MODEL DEVELOPMENT
TARGET TIME
SERIES(Y)
-GWL TESTING(20%) Validation

DATASET Data
TRAINING (80%) Preprocessing

INPUT VARIABLE Finally


-RAINFALL Matrix Convert validated
-TEMPERATURE data is
-SPRING DISCHARGE displayed in
-GROUND WATER webpage
USUAGE Machine
learning
algorithm feed
MODULES

 Data collection
 Data Pre Processing
 Training and Testing phase
 Implementation and Visualization
Data AQUSITION
 GROUND WATER LEVEL AND USUAGE DATASETS

 Ground water usage datasets consist of amount of water and using


for domestic or industrial purposes , monthly discharge rates ,
collection of rainfall in that region
 The net ground water availability for future and stages of ground
water extraction acts as the independent variable for prediction
process with 685 records .
SAMPLE DATA
Data Pre-processing
 The two data pre-processing techniques where used, they are
 Data Cleaning: Data cleaning is the process of preparing
raw data for analysis by removing bad data, organizing
the raw data, and filling in the null values. Pandas library were
used to remove the null values
 Data Normalization : Normalization is used to scale the
data of an attribute so that it falls in a smaller range.
SAMPLE DATA
Model Implementation

Bayesian Ridge Algorithm:

 Bayesian regression allows a natural mechanism to


survive insufficient data or poorly distributed data by
formulating linear regression using probability distributors
rather than point estimates.
 The output or response ‘y’
Data collection
FLOW OF MODULES

Data Preprocessing
Bayessian
Ridge Validation

Model
SVM
Implementation
Implementation and
Visualization
MLP
Graph of Bayesian ridge algorithm
Result Analysis

 The result of Bayesian algorithm is displayed in graph


with x-axis as the name of the district and y axis as the
ground water availability .
 Final result will be the comparing the three algorithm
and selecting the best fit algorithm to predict the
ground water level and educate farmers using that.
TIMELINE
Improvising the
Literature Data Training and model and
Survey Data Collection extraction(load) testing Implementation

AUG SEP OCT NOV DEC JAN FEB MAR APR MAY
REFERENCES
 [1] Haritha Kagita , Malathi V,Thiyagarajan “Prediction of ground water level
based on machine learning” International Research Journal of Engineering and
Technology(IRJET) 2020
 [2] Mohd Khairul Nizar Shamsuddin, Faradiella Mohd Kusin,Wan Nor Azmin
Sulaiman,Mohammad Firuz Ramli,Mohamad Faizal Tajul Baharuddin “Forecasting
of Groundwater Level using Artificial Neural Network by incorporating river
recharge and river bank infiltration” MATEC Web of Conferences 103, 04007
(2017)
 [3] Andreas Wunsch, Tanja Liesch, and Stefan Broda “Groundwater Level
Forecasting with Artificial Neural Networks: AComparison of LSTM, CNN and
NARX” 2020
 [4] Y. Z. KAYA, F. ÜNEŞ, M. DEMIRCI, B. TAŞAR, H. VARÇİN “Groundwater level
prediction using Artificial Neural Network and M5 tree models” 2018
REFERENCES
 [5] Pallavi Porte, Rajendra Kumar Isaac, Kipoo Kiran Singh Mahilang Khilendra
Sonboier and Pankaj Minj “Groundwater Level Prediction Using Artificial Neural
Network Model” International Journal of Current Microbiology and Applied
Sciences ISSN: 2319-7706 Volume 7 Number 02 (2018)
 [6] Pejman Zarafshan, Saman Javadi, Abbas Roozbahani, Seyed Mehdi
Hashemy, Payam “Artificial Intelligence Hybrid Deep Learning Model for
Groundwater Level Prediction Using MLP-ADAM” 2020
• [7] Fatemeh Barzegari Banadkooki, Mohammad Ehteram, Ali Najah Ahmed
“Enhancement of Groundwater-Level Prediction Using an Integrated Machine
Learning Model Optimized by Whale Algorithm” springer 2020
• [8] S. Sahoo,T. A. Russo,J. Elliott,I. Foster,“Machine Learning algorithm for
modelling ground water level changes in agricultural regions“ AGU Volume 53
issue 5 (2017)
THANK YOU

You might also like