2. Domain Knowledge
Artificial Intelligence is the science and engineering of making intelligent
machines, especially intelligent computer programs.
Machine Learning (ML) is a subset of Artificial Intelligence. ML is a science
of designing and applying algorithms that are able to learn things from past
cases..
Machine Learning (ML) models for flood prediction (disaster management
domain) can be beneficial for flood alerts and flood reduction or prevention.
The ability to analyze a large amount of data enables artificial intelligence
systems and machine learning algorithms to anticipate possible natural
disasters before they happen, thus preventing the loss of human lives.
3. Title Knowledge
According to title ,we will focus on providing :
● Adequate forewarning for the area where floods are likely to
occur.
● Alerts to the low lying areas about the release of accurate
quantity of water from the reservoirs and thus evacuation/shifting
of the people can be planned.
● Help the Response forces to deploy their resources accordingly
● Prediction of release of water based on rainfall in catchment area
and dissemination of an information to the affected public through
mobile and other mediums.
4. Abstract
IMD (Indian Meteorology department) is responsible to issue warnings for the
rainfall and CWC (Central Water Commission) keeps a record of water reservoirs,
however there is a lack of collation of data issued from both these departments.
This prevents us from determining the impact/seriousness and due to which there
are times where adequate forewarnings are not provided.
There are several High rainfall areas, low lying areas or flood prone areas.
Currently there are limitations that these areas cannot be alerted before the
critical situation because of the data unavailability or unavailability of simulation
models which can calculate and predict the data.
5. Literature Survey
Estimating Reservoir Release Using Multi-Source Satellite Datasets and Hydrological Modeling Techniques
(PDF) Estimating Reservoir Release Using Multi-Source Satellite Datasets and Hydrological Modeling Techniques (researchgate.net)
Flood Forecasting Using Machine Learning: A Review
Flood Forecasting Using Machine Learning: A Review | IEEE Conference Publication | IEEE Xplore
Flood Prediction Using Machine Learning Models
Flood Prediction Using Machine Learning Models | IEEE Conference Publication | IEEE Xplore
Probabilistic Mapping of August 2018 Flood of Kerala, India, Using Space-Borne Synthetic
Aperture Radar
Probabilistic Mapping of August 2018 Flood of Kerala, India, Using Space-Borne Synthetic Aperture Radar | IEEE Journals & Magazine | IEEE Xplore
6. Existing System
Synthetic aperture radar (SAR) imaging provides weather sensing technique that is
suitable for near-real-time mapping of disasters such as floods.
In this system, SAR data acquired by satellites to investigate a flood event that
affected Kerala in August 2018 is used .
They have applied Bayesian approach to generate probabilistic flood maps, which
contain for each pixel its probability to be flooded rather than binary flood information.
They have observed that no apparent correlation between the spatial distributions of
the flooded areas and the rainfall amounts at the district level of the study area.
The lack of apparent correlation is likely due to two reasons:
first, there is often some delay between the rainfall event and the flooding, especially
for rather large catchments where flood waves need some time to reach floodplains
from higher elevations. Second, rainfall is more abundant at overhead catchments (hills
and mountains), whereas flood occurs further downstream in the floodplains.
7. Proposed System/Architecture
In proposed system ,instead of satellite images ,we will use data of
rainfall in that particular region and data of different dam water level in
the same region.
We will create a dashboard consisting different features mentioned in
the next slide. For prediction ,we will use different algorithms such as
SVM(Support Vector Machine) ,Nonlinear Autoregressive Exogenous Model
(NARX) specially to predict water level changes. ANN (Artificial Neural
Networks) derives historical data for flood prediction and decision
making.
8. Modules And Functionality
1.Alert system : Flood alert to government’s disaster management
department will be sent.
2.Risk level: Intensity of flood will be shown.
3.Notification : Rescue team will get notified for high alert.
4.Prediction : Release of water will be predicted based on rainfall.