Flood Judge

From OpenCommons
Revision as of 23:09, January 24, 2023 by Pinfold (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search


Flood Judge
GCTC logo 344x80.png
FloodJudgeS.png
Ross Valley Flood
Team Organizations Civic Foundary
Team Leaders Vijay Sammeta
Participating Municipalities Ross Valley CA
Status Ready for Public Announcement
Document None

Description

Combining low cost / commodity sensors, enterprise IT skillsets, existing / familiar tools with the power of the cloud and artificial intelligence to proactively monitor and alert water levels in creeks during storm activity.

Challenges

Communities around the world suffer from sudden flooding, sometimes with little to no warning. Our solution would use historical data to model potential water level rise at a hyperlocal level and predict associated flooding dates and time. The sensor was inspired by our own community tragedy. In Feb 2017 14,000 residents in the City of San Jose were displaced by a flood for which there was no warning. In the days, weeks and months that followed the lack of real-time data, data in the appropriate format (height as opposed to cubic feet per second) and a notification system all contributed to nearly 100 million dollars in damage to the community we live and work in.

Solutions

TBD

Major Requirements

  • Build the hardware using commodity / off-the-shelf components
  • Write the application to read water levels every 5-15 mins and send the data to the cloud
  • Apply streaming analytics to determine if the current reading is at flood stage
  • Apple machine learning to the predicted rainfall, current water levels and historical modeling for the current location of sensor deployment
  • Build a dashboard for administrators, a subscription page for notifications and real-time events to push mass notifications

Performance Targets

Key Performance Indicators (KPIs) Measurement Methods

Provide automated notifications up to an hour prior to creek flooding (obviously we will try to model as much notice as possible). Due to the hyper local nature of each sensor the pilot will begin this fall to begin collecting data surrounding creek level and projected rainfall between every 5-15 mins. We will be working with the Ross Valley Fire Department determine alerting levels and finalize the KPI’s over the next two months

The dashboard will include projected rainfall, associated hyperlocal creek rise and historical data. Using this combination we will be able to validate if the model is working properly as well as measure the accuracy the rainfall projections. The KPI’s will be a part of the dashboard and available to key stakeholders.

Standards, Replicability, Scalability, and Sustainability

  • The current solution uses Wifi and is completely based around enterprise technology such as off-the-shelf hardware, web services and the data center. It can even be viewed in a simple spreadsheet.
  • The entire goal of the project is to highlight Replicability, Scalability, and Sustainability by extending municipal IT skill sets incrementally as opposed

to introducing new ones that often times are ultimately unsustainable due to costs, attraction & retention of talent or skillsets. Our goal is to move smart cities from proof-of-concept to as common as email in cities by lowering barriers to entry and sustainability.

Cybersecurity and Privacy

TBD

Impacts

This is a life, health, safety and property protection project.

Demonstration/Deployment

We will have our prototype sensors, dashboard, community notification page and alerting solution available for demo. We are prepared to demonstrate various water levels including flood stage and alerting by pour water into a cylinder.