Close

Daniel Arnold

Research Scientist

Download Resume

About Me

Welcome to my personal website. I am a research scientist with a background in control systems, optimization, and machine learning, who is passionate about applied research. In addition to studying and applying algorithms from these fields to solve practical real-world problems, I greatly enjoy teaching both fundamental and advanced concepts in these areas. On this site, you'll find some details of my CV, as well as links to some independent projects.

Prior to earning my PhD at Cal, I developed nonlinear and adaptive control algorithms for unmanned underwater vehicles for the U.S. Navy. During my PhD studies, I developed model-free optimal control techniques for managing renewable energy sources in electric grids. Since joining Lawrence Livermore National Laboratory, I've conducted research focusing on the use of optimal control/reinforcement learning techniques for the electric power system. Additionally, I am studying the development of statistical and dynamic models of data center behavior and power consumption to improve data center efficiency, reliability, and to assess impacts on the bulk electricity system.

Google Scholar

Experience

Lawrence Livermore National Laboratory (LLNL)

Lead Power Grid Engineer

I lead research projects at the intersection of electric power systems, machine learning, statistical modeling, optimization, and control theory. My work has been focused on:

  • Developing statistical and dynamic models of hyperscale data center behavior and energy consumption to increase data center efficiency and reduce bulk system impacts
  • Creating co-simulation frameworks for training reinforcement learning algorithms for power system estimation, control, and cybersecruity
  • Developing grid foundational models to enable zero-shot transfer learning across unseen power system topologies

University of California Berkeley

Adjunct Professor, Civil and Environmental Engineering Department

E93 - Energy Engineering Seminar

  • Organized speaker series which brings together industry leaders, researchers, and innovators in the energy sector to share insights on sustainable energy solutions, technological advancements, and policy implications
  • Fall 2026 focus on data center impacts, modeling, and optimization

CEE 295 - Data Science for Energy

  • Introduces data science fundamentals and programming techniques to graduate and upper division students
  • Course content includes: dynamic systems modeling, state estimation, convex optimization, machine learning, and optimal control

CEE 191 - Systems Analysis (Introduction to Optimization)

  • Introduces optimization fundamentals and programming techniques to graduate and upper division undergraduate students
  • Course content includes: linear programming, quadratic programming, mixed integer programming, nonlinear programming, search algorithms, and dynamic programming

Lawrence Berkeley National Lab (LBNL)

Research Scientist

I lead and co-lead several projects at the intersection of control systems/optimization/machine learning and the electric power system. My work has been focused on:

  • Developing optimal control algorithms and software for stochastic optimization of electric power distribution systems
  • Conducting research into black-box model-free optimization algorithms
  • Designing adaptive control schemes for Distributed Energy Resource (DER) cybersecurity

Lawrence Berkeley National Lab (LBNL)

ITRI-Rosenfeld Postdoctoral Fellow

I researched model-free optimal control strategies for managing solar photovoltaic systems. Additionally, I applied semi-supervised learning techniques for inference of properties of electric distribution grids using Phasor Measurement Unit (PMU) data.

Marine Renewable Energy Engineer

Naval Facilities Engineering Service Center (NAVFAC-ESC), U.S. Navy

I researched ocean-based renewable energy technologies both under development or commercially available and prepared technology readiness level assessments for U.S. Navy shore facility commanders.

Research and Development Engineer

Space and Naval Warfare Center (SPAWAR), U.S. Navy

I developed and tested control algorithms for unmanned underwater vehicles for the U.S. Navy

Education

University of California Berkeley

September 2009- December 2015

Doctor of Philosophy in Mechanical Engineering

I developed model-free optimal control algorithms to manage solar photovoltaic generation systems to optimize the behavior of electric power distribution systems. My advisors were Dave Auslander (Mechanical Engineering) and Duncan Callaway (Energy and Resources Group)

University of California San Diego

September 2005 - December 2006

Master of Science in Mechanical Engineering

I researched extremum seeking control approaches for peak seeking mobile robot applications. My advisor was Miroslav Krstic.

University of California San Diego

September 2001 - September 2005

Bachelor of Science in Mechanical Engineering

Elective courses in signal processing, control systems, and robotics. Graduated cum laude.

Projects

from_scratch

A collection of small machine learning projects built completley from scratch. I don't use any tools from advanced ML python packages (sklearn, scipy) except for comparison to the from_scratch results.

View Project

time_series

A collection of time series analysis and forecasting techniques from control systems, signal processing, and machine learning.

View Project

reinforcement_learning

A collection of reinforcement learning problems from OpenAI gym solved with Keras and TensorFlow.

View Project

extremum_seeking

Extremum seeking is a nonlinear adaptive control algorithm that is incredibly useful for model-free optimization. This project is a collection of extremum seeking control implementations on python notebooks.

View Project

Skills

Get in Touch