a Abhiram Mullapudi
Profile Name

Hello! I am Abhiram Mullapudi (అభిరామ్ ముళ్లపూడి). I build digital water systems that integrate machine learning, wireless sensor networks, and physical modeling to improve the resilience of urban water infrastructure. I am currently working at Inframark as a Lead Data Scientist, where I develop scalable machine learning solutions for small and medium-sized water utilities. My interests lie in signal processing, probabilistic methods, and optimization for cyber-physical infrastructure systems. I write about some of these at randomstorms.substack.com. I am based in Washington, DC 🌸. During my free time, I row with the amazing crew at Capital Rowing Club, work on improving my espresso skills, and try to keep my plants alive."

Email  /  CV  /  GitHub  /  Google Scholar  /  LinkedIn

Publications

Bayes

Identification of stormwater control strategies and their associated uncertainties using Bayesian Optimization

Abhiram Mullapudi, Branko Kerkez

In preparation, 2023

poster / paper

Bayesian optimization is an automated data-driven approach for identifying a control strategy that achieves the desired response from the stormwater network.

pystorms

pystorms : Simulation sandbox for the evaluation and design of stormwater control algorithms

Sara P. Rimer, Abhiram Mullapudi, et al.

Environmental Modelling and Software, 2023

code / paper/ webpage

pystorms provides a curated collection of stormwater control scenarios to enable the development and quantitative comparison of stormwater control algorithms.

pyswmm

Improvement of phosphorus removal in bio retention cells using real-time control

Brooke E. Mason, Abhiram Mullapudi, Cyndee L. Gruden, Branko Kerkez

Urban Water Journal, 2022

code/ paper

Real-time control improves the nutrient capture efficiency of bio retention cells, and thus reducing the size of bio retention cells needed for nutrient removal.

pyswmm

StormReactor: An open-source Python package for the integrated modeling of water quality and water balance

Brooke E. Mason, Abhiram Mullapudi, Branko Kerkez

Environmental Modelling and Software, 2021

code/ paper

StormReactor is a python package for updating pollutant concentrations in EPA-SWMM durng simulations.

pyswmm

PySWMM : The Python Interface to Stormwater Management Model (SWMM)

Bryant E. McDonnell, Katherine Ratliff, Michael E. Tryby, Jennifer Jia Xin Wu, Abhiram Mullapudi

Journal of Open Source Software, 2020

code / paper

pyswmm is a python wrapper for interfacing with US EPA's Stormwater Management Model.

deeprl

Deep Reinforcement Learning for the Real Time Control of Stormwater Systems

Abhiram Mullapudi, Matthew J. Lewis, Cyndee L. Gruden, Branko Kerkez

Advances in Water Resources, 2020

code / paper

Reinforcement learning can be used for creating autonomous stormwater systems that can dynamically change their behavior based on the state of the watershed for achieving system scale objectives.

deeprl

rrcf : Implementation of the Robust Random Cut Forest algorithm for anomaly detection on streams

Matt D. Bartos, Abhiram Mullapudi, Sara C. Troutman

Journal of Open Source Software, 2019

code / paper

Anomalies in the streaming data can be detected by estimating the shift in the structure of the random forest caused by the addition of a new data point.

deeprl

Shaping streamflow using a real-time stormwater control network

Abhiram Mullapudi, Matt D. Bartos, Brandon P. Wong, Branko Kerkez

Sensors, 2018

paper

Response of a stormwater network can be precisely shaped with the data from a wireless sensor network.

deeprl

Building a theory for smart stormwater systems

Abhiram Mullapudi, Brandon P. Wong, Branko Kerkez

Environmental Science: Water Research and Technology, 2017

code / paper

By re-imagining physical watersheds as a network of interconnected systems, they can be dynamically reconfigured in real-time to target the removal of specific pollutants.