Saman Mostafavi
YOU?
Author Swipe
View article: Modelica Supported Automated Design
Modelica Supported Automated Design Open
We propose a component-based, automated, bottom-up method to system design, using models expressed in the Modelica language. This bottom-up approach is based on a meta-topology that is iteratively refined via optimization. Each topology li…
View article: Sensorium: commissioning abundant sensors with augmented reality and QR codes
Sensorium: commissioning abundant sensors with augmented reality and QR codes Open
In the future, it will be possible to build high-quality models of building interiors based on data from a dense fleet of sensors reporting on air volumes much smaller than a room or zone. To enable such models, we are creating technologie…
View article: Optimal sensor/actuator placement and control of buildings sub-zones
Optimal sensor/actuator placement and control of buildings sub-zones Open
We propose a novel framework to jointly optimize the placement of actuators and sensors, as well as the control laws for optimally controlling sub-zonal, spatiotemporal temperature profiles in a spacious room. Unlike traditional approaches…
View article: Benchmarking model predictive control algorithms in Building Optimization Testing Framework (BOPTEST)
Benchmarking model predictive control algorithms in Building Optimization Testing Framework (BOPTEST) Open
We present a data-driven modeling and control framework for physics-based building emulators. Our approach consists of: (a) Offline training of differentiable surrogate models that accelerate model evaluations, provide cost-effective gradi…
View article: A data-driven method to control buildings sub-zones to spatio-temporal profiles
A data-driven method to control buildings sub-zones to spatio-temporal profiles Open
The paper develops a data-driven approach to optimally control sub-zonal, spatiotemporal temperature profiles in a spacious room. We first calibrate a 2D diffusion Partial Differential Equation (PDE) model for heat transfer with Bluetooth …
View article: Benchmarking Model Predictive Control Algorithms in Building Optimization Testing Framework (BOPTEST)
Benchmarking Model Predictive Control Algorithms in Building Optimization Testing Framework (BOPTEST) Open
We present a data-driven modeling and control framework for physics-based building emulators. Our approach consists of: (a) Offline training of differentiable surrogate models that accelerate model evaluations, provide cost-effective gradi…
View article: Optical monitoring to detect contamination of power grid components
Optical monitoring to detect contamination of power grid components Open
A monitoring system includes an array of optical sensors disposed within a transformer tank. Each optical sensor is configured to have an optical output that changes in response to a temperature within the transformer tank. An analyzer is …
View article: Discovering Symbolic Policy for Building Control using Reinforcement Learning
Discovering Symbolic Policy for Building Control using Reinforcement Learning Open
We propose a learning framework for interpretable HVAC control in buildings using deep reinforcement learning (DRL). Our framework includes a data-driven surrogate environment to emulate building dynamics and a Deep Symbolic Policy for dis…
View article: Nonlinear Moving Horizon Estimation and Model Predictive Control for Buildings with Unknown HVAC Dynamics
Nonlinear Moving Horizon Estimation and Model Predictive Control for Buildings with Unknown HVAC Dynamics Open
We present a solution for modeling and online identification for heating, ventilation, and air conditioning (HVAC) control in buildings. Our approach comprises: (a) a resistance-capacitance (RC) model based on first order energy balance fo…
View article: Modeling and simulating thermostat behaviors of office occupants
Modeling and simulating thermostat behaviors of office occupants Open
Existing literature postulates thermal behaviors primarily to be a consequence of thermal discomfort. However, some thermal behaviors, such as changing the thermostat settings in an office building, can cause thermal discomfort to other oc…
View article: Nonlinear Moving Horizon Estimation and Model Predictive Control for Buildings with Unknown HVAC Dynamics
Nonlinear Moving Horizon Estimation and Model Predictive Control for Buildings with Unknown HVAC Dynamics Open
We present a solution for modeling and online identification for heating, ventilation, and air conditioning (HVAC) control in buildings. Our approach comprises: (a) a resistance-capacitance (RC) model based on first order energy balance fo…
View article: Model Development for Robust Optimal Control of Building HVAC
Model Development for Robust Optimal Control of Building HVAC Open
Many control schemes have been proposed for improving the operation of variable-air volume HVAC systems. These approaches typically employ some form of simplified state-space model to predict the temperature evolution in each zone. One rea…
View article: Low-Cost Embedded Optical Sensing Systems For Distribution Transformer Monitoring
Low-Cost Embedded Optical Sensing Systems For Distribution Transformer Monitoring Open
There is a mounting need for low-cost monitoring with online sensing technologies to maintain grid reliability and uptime. Here we introduce an innovative low-cost, embedded optical sensing technology initially focused on transformers that…