Engineering topics

Jain, Milan

Publications and source records attributed to Jain, Milan.

15 records

DOE OSTI2024

Analyzing inference workloads for spatiotemporal modeling

Ensuring power grid resiliency, forecasting climate conditions, and optimization of transportation infrastructure are some of the many application areas where data is collected in both space and time. Spatiotemporal modeling is about modeling those patterns for forecasting future trends and carrying out critical decision-making by leveraging machine learning/deep learning. Once trained offline, field deployment of trained models for near real-time inference could be challenging because performance can vary significantly depending on the environment, available compute resources and tolerance to ambiguity in results. Users deploying spatiotemporal models for solving complex problems can benefit from analytical studies considering a plethora of system adaptations to understand the associated performance-quality trade-offs. To facilitate the co-design of next-generation hardware architectures for field deployment of trained models, it is critical to characterize the workloads of these deep learning (DL) applications during inference and assess their computational patterns at different levels of the execution stack. In this paper, we develop several variants of deep learning applications that use spatiotemporal data from dynamical systems. We study the associated computational patterns for inference workloads at different levels, considering relevant models (Long short-term Memory, Convolutional Neural Network and Spatio-Temporal Graph Convolution Network), DL frameworks (Tensorflow and PyTorch), precision (FP16, FP32, AMP, INT16 and INT8), inference runtime (ONNX and AI Template), post-training quantization (TensorRT) and platforms (Nvidia DGX A100 and Sambanova SN10 RDU). Overall, our findings indicate that although there is potential in mixed-precision models and post-training quantization for spatiotemporal modeling, extracting efficiency from contemporary GPU systems might be challenging. Instead, co-designing custom accelerators by leveraging optimized High Level Synthesis frameworks (such as SODA High-Level Synthesizer for customized FPGA/ASIC targets) can make workload-specific adjustments to enhance the efficiency.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Projected income data under different shared socioeconomic pathways for Washington state

Abstract High-resolution income projections under different Shared Socioeconomic Pathways (SSPs) are essential for the climate change research communities to devise climate change adaptation and mitigation strategies. To generate income projections for Washington state, we obtain state-level GDP per capita projections and convert them into projected annual household income. The resulting state-level income projections are subsequently downscaled to the census block-level based on the Longitudinal Origin-Destination Employment Statistics (LODES) dataset. For accuracy assessment, we downscale historical income data from state- level to block- and block group-level and compare the downscaled results against the actual income data from LODES. County-level accuracy assessment is also conducted based on American Community Survey. The results demonstrate a good agreement (Average R 2 of 0.67, 0.8, and 0.99 for block-, block group-, and county-level, respectively) between the downscaled income data and the reference data, thereby validating the methodology employed. Our approach is applicable to other states for income projections, which can be utilized by a broader audience, including those involved in demographic analysis, economic research, and urban planning.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

DOE OSTI2023

Artificial Intelligence for improved facilities operation in the FNAL LINAC

The energy consumption in accelerator structures during beam downtimes is a significant fraction of the overall energy budget. Accurate prediction of downtime duration could inform actions to reduce this energy consumption. The LCAPE project started in 2020 and develops artificial intelligence to improve operations in the FNAL control room by reducing the time to identify the cause of a beam outage, improving the reproducibility of labeling it, predicting their duration and forecasting their occurrence. We present our solution for incorporating information from ~2.5k monitored devices in near-real time to distinguish between dozens of different causes of down time. We discuss the performance of different techniques for modeling the state of health of the facility and we compare unsupervised clustering techniques to distinguish between different causes of down time.

43 PARTICLE ACCELERATORS↗

DOE OSTI2023

Proxy Applications for Converged Workloads: DMC LDRD Initiative

Modern scientific applications are complicated and require coordination of several components. Proxy application driven software-hardware co-design plays a vital role in driving innovation among the developments of applications, software infrastructure and hardware architecture. Proxy applications are self-contained and simplified codes that are intended to model the performance-critical computations within applications. Applications executing on modern High Performance Computing (HPC) systems are susceptible to network congestion, insufficient memory bandwidth within and across compute nodes, and inadvertent loss of performance due to bugs and unoptimized programming models. Modern numerical simulations and machine learning models play a critical role in studying physical phenomenon under myriad uncertainties. Such applications often exhibit irregular computation and memory accesses at specific regions of the application code, which can contribute to various performance bottlenecks at scale. To mitigate such issues and prepare the next generation hardware for a variety of computation and data movement contingencies, a well-known practice is to consider "proxy" applications as representative motifs for various classes of scientific applications. While there is disagreement in the HPC community on the mechanisms of construction of the proxy applications, there is a strong consensus on their positive impact in co-design. Proxy Applications for Converged Workloads (PACER) is about facilitating software-hardware co-design through proxy applications with the goal of improving the performance of converged science workflows on heterogeneous systems.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

A Machine Learning Framework to Deconstruct the Primary Drivers for Electricity Market Price Events

As the electricity grid is moving towards a 100% Renewable Energy Source Bulk Power Grid, the overall operations of the power system operations and electricity markets are changing. The electricity markets are not only dispatching resources economically but also taking into account various controllable actions like renewable curtailment, transmission congestion mitigation, and energy storage optimization to make sure the grid is operating reliably. As a result, price formations in electricity markets have become quite complex. Traditional root cause analysis and statistical approaches are rendered inapplicable to analyze and infer the main drivers behind price formation in the modern grid and markets with variable renewable energy (VRE). In this paper, we propose a machine learning analysis framework to deconstruct some primary drivers for price formation in modern electricity markets with high renewable energy and the outcomes can be utilized for various critical aspects of market design, renewable dispatch and curtailment, operations, and cyber-security applications. The framework can be applied to any ISO or market data and in this paper it is applied to open-source publicly available datasets from California Independent System Operator (CAISO) and ISO New England.

machine learning (ML), electricity markets, Renewa↗

DOE OSTI2023

Areal interpolation of population projections consistent with different SSPs from 1-km resolution to block level based on USA Structures dataset

Population data are normally collected at various census administrative levels, and areal interpolation of population is often required to transform population to the desired spatial resolution. Building footprint datasets, such as Microsoft building footprints, have proven to be useful in estimating population distribution and can therefore be used for areal interpolation of population. In addition to Microsoft building footprints, the recently released USA Structures dataset provides additional information such as building type and building height for some regions, which may provide valuable information for a better depiction of population distribution and improved population areal interpolation accuracy. In this study, we have conducted areal interpolation of population projections consistent with three different Shared Socioeconomic Pathways (SSP2, SSP3, and SSP5) from 1-km grid cells to block level in Washington state for every ten years from 2020 to 2040 based on USA Structures. We assessed USA Structures-based population downscaling accuracy using U.S. decennial survey data in 2020 under three different downscaling schemes, including population downscaling from census tracts to block groups, from census tracts to blocks, and from block groups to blocks. The resulting accuracies were compared with those based on Microsoft building footprints. The comparison showed that USA Structures achieved higher accuracies across different population density regions and areas with different urbanization extent within our study area.

99 GENERAL AND MISCELLANEOUS↗

DOE OSTI2022

Sparse Deep Neural Network Inference using different Programming Models

Sparse deep neural networks have gained increasing attention recently in achieving speedups on inference with reduced memory footprints. However, the real-world applications are little shown with specialized optimizations, yet a wide variety of DNN tasks remain dense without exploiting the advantages of sparsity in networks. Recent work presented by MIT/IEEE/Amazon GraphChallenge has demonstrated significant speedups and various techniques. Still, we find that there is limited investigation of the impact of various Python and C\slash C++ based programming models to explore new opportunities in general cases. In this work, we provide performance evaluation through different programming models using CuPy, cuSPARSE, and OpenMP to discuss the advantages and disadvantages of our sparse implementations on single-GPU and multiple GPUs of NVIDIA DGX-A100 40GB/80GB platforms.

machine learning, HPC↗

DOE OSTI2022

WISP: Watching grid Infrastructure Stealthily through Proxies (Final Technical Report)

The complex interdependencies of cyber systems (sensors and communications), physical grids and associated electricity market operations make protecting electric power grids a significant challenge. The energy sector is constantly under new, targeted, advanced and dangerous cyber-attacks that have the potential to result in the loss of human life. These threats are further exacerbated by our need to modernize the grid. One focus of cyber security research in smart grids is the securing of the SCADA system through advanced intrusion detection systems (IDS) and bad data detection algorithms in state estimation. These methods either require full knowledge of the system topology and parameters or fail to understand the physical behaviors under attack. WISP (Watching grid Infrastructure Stealthily through Proxies) is designed to provide additional protection to the power grid using only publicly available data. In particular, WISP exploits the spatio-temporal nature of the real time locational marginal prices (LMPs), in conjunction with other information such as bids, weather, outages and load data to analyze anomalous power pricing behaviors and then correlate those observations to localize regions of interest and identify potential cyber events. WISP is non-intrusive as the tool is deployed as a service in the Cloud or on premise and provides reliable information to system operators for enhanced situational awareness, without impeding energy delivery functions. The WISP technology comprises three modules: the data-driven anomaly detection core, the vulnerability and risk analysis and the root cause analysis. The data-driven anomaly detection core performs the tasks of feature selection, anomaly detection and attack region localization. The vulnerability and risk analysis module provides system level information of the vulnerable variables and times, assisting the operators in selecting monitoring and protection nodes. The root cause analysis module takes the detection results and identifies potential operational conditions that contribute to the detected anomalies. In Phase I, we have demonstrated the feasibility and effectiveness of WISP. We developed a realistic electricity market simulator capable of generating normal and attack market data under various operational conditions. We developed a series of cyber-attack detection and analysis algorithms and evaluated them under multiple data sources. Finally, we integrated all modules into an end-to-end software, providing functions for data management, data analytics and visualization. Specifically, we have achieved: (i) real-time data acceptance from external utility interfaces with >99% acceptance rate; (ii) high performance anomaly detection algorithms with >98% detection accuracy and <0.1% false alarm rate; and (iii) ultra-low computing delay <50 milliseconds. Additionally, our team developed algorithms to identify the vulnerable variables in electricity market operations and root cause analysis functions to identify major contributors to the price spikes. These ancillary modules are necessary when deploying WISP in real world industry environment. In Phase II, we have demonstrated the effectiveness of WISP software on realistic largescale power systems. We performed red team testing for the Phase I WISP software and identified software vulnerabilities and implemented corresponding mitigation solutions. We adapted the electricity market simulator for the Texas synthetic 2000-bus system and generated datasets for the false data injection attacks. We created database and visualization interfaces for the Texas system and the ISO New England system. We performed software optimization in terms of operation efficiency, computing speed and detection accuracy. Finally, we tested the software on the Texas system and the ISO New England system and evaluated the detection performance. Overall, we achieved above 89% detection rate, below 3% false alarm rate and below 37 seconds of end-to-end detection delay.

24 POWER TRANSMISSION AND DISTRIBUTION↗

DOE OSTI2022

The L-CAPE Project at FNAL

The controls system at FNAL records data asynchronously from several thousand Linac devices at their respective cadences, ranging from 15Hz down to once per minute. In case of downtimes, current operations are mostly reactive, investigating the cause of an outage and labeling it after the fact. However, as one of the most upstream systems at the FNAL accelerator complex, the Linac’s foreknowledge of an impending downtime as well as its duration could prompt downstream systems to go into standby, potentially leading to energy savings. The goals of the Linac Condition Anomaly Prediction of Emergence (L-CAPE) project that started in late 2020 are (1) to apply data-analytic methods to improve the information that is available to operators in the control room, and (2) to use machine learning to automate the labeling of outage types as they occur and discover patterns in the data that could lead to the prediction of outages. We present an overview of the challenges in dealing with time-series data from 2000+ devices, our approach to developing an ML-based automated outage labeling system, and the status of augmenting operations by identifying the most likely devices predicting an outage.

43 PARTICLE ACCELERATORS↗

DOE OSTI2021

Transfer-Learnt Energy Models for Predicting Electricity Consumption in Buildings with Limited and Sparse Field Data

Modeling energy consumption is critical for energy-efficient utilization of the electric appliances in a building, smart grid programs (like demand-response), and many other smart home applications. State-of-the-art energy modeling techniques either rely on theoretical models, or extensive instrumentation of the building envelope to gather ``big" data to train a deep neural network. While theoretical models are often limited by their estimation accuracy, it is not always feasible to gather a significant amount of field data. In this paper, we explore transfer learning-based strategies to train much more accurate model for energy estimation when using a sparse field data. We transferred knowledge, in the form of data and parameters, from the simulation framework to the field data. We evaluated the efficacy of our approach on field data collected from six commercial buildings and our results indicate that transfer learning-based models trained over one month data can perform comparative (and in some cases better) than the state-of-the-art machine learning and deep learning solutions.

Jain, Milan↗

DOE OSTI2021

Occupancy-Driven Stochastic Decision Framework for Ranking Commercial Building Loads

For effective integration of building operations into the evolving demand response programs of the power grid, real-time decisions concerning the use of building appliances for grid services must excel on multiple criteria, ranging from the added value to occupants' comfort to the quality of the grid services. In this paper, we present a data-driven stochastic decision-support framework to dynamically rank load control alternatives in a commercial building, addressing the needs of multiple decision criteria (e.g. occupant comfort, grid service quality) under uncertainties in occupancy patterns. We adopt a stochastic multi-criteria decision algorithm recently applied to prioritize residential on/off loads, and extend it to i) consider complex load control decisions (e.g. dimming of lights, changing zone temperature set-points) in a commercial building; and ii) systematically integrate zonal occupancy patterns to better identify short-term (and time-varying) opportunities for grid service participation. We evaluate the performance of the proposed framework for curtailment of air-conditioning, lighting, and plug-loads in a multi-zone commercial office building for a range of design choices. With the help of a prototype system that integrates an interactive \textit{Data Analytics and Visualization} frontend we demonstrate a way for the building operators to monitor and change in real-time the available flexibility in energy consumption and to develop trust in the decision recommendations by interpreting the rationale behind the ranking.

Jain, Milan↗

DOE OSTI2021

transBEAM

Toolkit to implement and explore transfer learning-based strategies to address unsatisfactory accuracy of models for estimating building energy consumption when available field data for training is sparse or of unacceptable quality

Jain, Milan↗

DOE OSTI2020

Exploration of Domain Aware Machine Learning for Grid Analytics: Transfer-Learnt Energy Models to Assist Buildings Control with Sparse Field Data

Buildings are a primary consumer of energy in the United States and are also increasingly being perceived as providers of grid services such as load shifting, shedding and modulation. High fidelity models of building energy consumption are needed to set appropriate baselines for measurement and verification (M&V) of controllers designed for energy efficient operation of buildings and to enable buildings to provide grid services via. participation in demand response programs. State-of-the-art building energy modeling techniques either rely on Physics based models, or extensive instrumentation of the building envelope to gather “big” data to train machine learning based models such as deep neural networks. While Physics based models are often limited by their accuracy, it is not always feasible to gather a significant amount of field data required to train machine learning based models with sufficient accuracy. In this paper, we explore the use of transfer learning-based strategies to address unsatisfactory accuracy of models for estimating building energy consumption when available field data for training is sparse or of unacceptable quality. In particular, we transfer knowledge in the form of data and parameters, from Physics based simulation frameworks to the field to improve the model accuracy, thus resulting in a Physics-informed Machine Learning framework. We evaluated the efficacy of our approach on field data collected from six commercial buildings and our results indicate that the proposed transfer learning based models provide comparative (and in some cases better) accuracy than state-of-the-art machine learning and deep learning solutions, with just one month of field data.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2019

Beyond Control: Enabling Smart Thermostats for Leakage Detection

Smart thermostats, with multiple sensory abilities, are becoming pervasive and ubiquitous, in both residential and commercial buildings. By analyzing occupants’ behavior, adjusting set temperature automatically, and adapting to temporal and spatial changes in the atmosphere, smart thermostats can maximize both - energy savings and user comfort. In this paper, we study smart thermostats for refrigerant leakage detection. Retail outlets, such as milk-booths and quick service restaurants set up cold-rooms to store perishable items. In each room, a refrigeration unit (akin to air-conditioners) is used to maintain a suitable temperature for the stored products. Often, refrigerant leaks through the coils (or valves) of the refrigeration unit which slowly diminishes the cooling capacity of the refrigeration unit while allowing it to be functional. Such leaks waste significant energy, risk occupants’ health, and impact the quality of stored perishable products. While store managers usually fail to sense the early symptoms of such leaks, current techniques to report refrigerant leakage are often not scalable. We propose Greina - to continuously monitor the readily available ambient information from the thermostat and timely report such leaks. We evaluate our approach on 74 outlets of a retail enterprise and results indicate that Greina can report the leakage a week in advance when compared to manual reporting

Jain, Milan↗

DOE OSTI2017

Portable+: A Ubiquitous And Smart Way Towards Comfortable Energy Savings

An air conditioner (AC) consumes a significant proportion of the total household power consumption. Primarily used in developing countries, decentralised AC has an inbuilt thermostat to cool the room to a temperature, manually set by the users. However, residents are incapable of specifying their goal through these thermostats - maximise their comfort or save AC energy. State-of-the-art portable thermostats emulate AC remotes and assist occupants in remotely changing the thermostat temperature, through their smartphones. We propose extending such thermostats to portable+ by adding a Comfort-Energy Trade-off (CET) knob, realised through an optimisation framework to allow users to balance their comfort and the savings without worrying about the right set temperature. Analysis based on real data, collected from a controlled experiment (across two rooms for two weeks) and an in-situ deployment (across five rooms for three months), indicates that portable+ thermostats can reduce residents’ discomfort by 23% (CET selection for maximal comfort) and save 26% energy when CET is set for maximising savings.

Jain, Milan↗