-
Creating synthetic energy meter data using conditional diffusion and building metadata
Authors:
Chun Fu,
Hussain Kazmi,
Matias Quintana,
Clayton Miller
Abstract:
Advances in machine learning and increased computational power have driven progress in energy-related research. However, limited access to private energy data from buildings hinders traditional regression models relying on historical data. While generative models offer a solution, previous studies have primarily focused on short-term generation periods (e.g., daily profiles) and a limited number o…
▽ More
Advances in machine learning and increased computational power have driven progress in energy-related research. However, limited access to private energy data from buildings hinders traditional regression models relying on historical data. While generative models offer a solution, previous studies have primarily focused on short-term generation periods (e.g., daily profiles) and a limited number of meters. Thus, the study proposes a conditional diffusion model for generating high-quality synthetic energy data using relevant metadata. Using a dataset comprising 1,828 power meters from various buildings and countries, this model is compared with traditional methods like Conditional Generative Adversarial Networks (CGAN) and Conditional Variational Auto-Encoders (CVAE). It explicitly handles long-term annual consumption profiles, harnessing metadata such as location, weather, building, and meter type to produce coherent synthetic data that closely resembles real-world energy consumption patterns. The results demonstrate the proposed diffusion model's superior performance, with a 36% reduction in Frechet Inception Distance (FID) score and a 13% decrease in Kullback-Leibler divergence (KL divergence) compared to the following best method. The proposed method successfully generates high-quality energy data through metadata, and its code will be open-sourced, establishing a foundation for a broader array of energy data generation models in the future.
△ Less
Submitted 30 March, 2024;
originally announced April 2024.
-
Large-scale Training of Foundation Models for Wearable Biosignals
Authors:
Salar Abbaspourazad,
Oussama Elachqar,
Andrew C. Miller,
Saba Emrani,
Udhyakumar Nallasamy,
Ian Shapiro
Abstract:
Tracking biosignals is crucial for monitoring wellness and preempting the development of severe medical conditions. Today, wearable devices can conveniently record various biosignals, creating the opportunity to monitor health status without disruption to one's daily routine. Despite widespread use of wearable devices and existing digital biomarkers, the absence of curated data with annotated medi…
▽ More
Tracking biosignals is crucial for monitoring wellness and preempting the development of severe medical conditions. Today, wearable devices can conveniently record various biosignals, creating the opportunity to monitor health status without disruption to one's daily routine. Despite widespread use of wearable devices and existing digital biomarkers, the absence of curated data with annotated medical labels hinders the development of new biomarkers to measure common health conditions. In fact, medical datasets are usually small in comparison to other domains, which is an obstacle for develo** neural network models for biosignals. To address this challenge, we have employed self-supervised learning using the unlabeled sensor data collected under informed consent from the large longitudinal Apple Heart and Movement Study (AHMS) to train foundation models for two common biosignals: photoplethysmography (PPG) and electrocardiogram (ECG) recorded on Apple Watch. We curated PPG and ECG datasets from AHMS that include data from ~141K participants spanning ~3 years. Our self-supervised learning framework includes participant level positive pair selection, stochastic augmentation module and a regularized contrastive loss optimized with momentum training, and generalizes well to both PPG and ECG modalities. We show that the pre-trained foundation models readily encode information regarding participants' demographics and health conditions. To the best of our knowledge, this is the first study that builds foundation models using large-scale PPG and ECG data collected via wearable consumer devices $\unicode{x2013}$ prior works have commonly used smaller-size datasets collected in clinical and experimental settings. We believe PPG and ECG foundation models can enhance future wearable devices by reducing the reliance on labeled data and hold the potential to help the users improve their health.
△ Less
Submitted 6 March, 2024; v1 submitted 8 December, 2023;
originally announced December 2023.
-
Updated Corpora and Benchmarks for Long-Form Speech Recognition
Authors:
Jennifer Drexler Fox,
Desh Raj,
Natalie Delworth,
Quinn McNamara,
Corey Miller,
Migüel Jetté
Abstract:
The vast majority of ASR research uses corpora in which both the training and test data have been pre-segmented into utterances. In most real-word ASR use-cases, however, test audio is not segmented, leading to a mismatch between inference-time conditions and models trained on segmented utterances. In this paper, we re-release three standard ASR corpora - TED-LIUM 3, Gigapeech, and VoxPopuli-en -…
▽ More
The vast majority of ASR research uses corpora in which both the training and test data have been pre-segmented into utterances. In most real-word ASR use-cases, however, test audio is not segmented, leading to a mismatch between inference-time conditions and models trained on segmented utterances. In this paper, we re-release three standard ASR corpora - TED-LIUM 3, Gigapeech, and VoxPopuli-en - with updated transcription and alignments to enable their use for long-form ASR research. We use these reconstituted corpora to study the train-test mismatch problem for transducers and attention-based encoder-decoders (AEDs), confirming that AEDs are more susceptible to this issue. Finally, we benchmark a simple long-form training for these models, showing its efficacy for model robustness under this domain shift.
△ Less
Submitted 26 September, 2023;
originally announced September 2023.
-
BIM-to-BRICK: Using graph modeling for IoT/BMS and spatial semantic data interoperability within digital data models of buildings
Authors:
Filippo Vittori,
Chuan Fu Tan,
Anna Laura Pisello,
Adrian Chong,
Clayton Miller
Abstract:
The holistic management of a building requires data from heterogeneous sources such as building management systems (BMS), Internet-of-Things (IoT) sensor networks, and building information models. Data interoperability is a key component to eliminate silos of information, and using semantic web technologies like the BRICK schema, an effort to standardize semantic descriptions of the physical, logi…
▽ More
The holistic management of a building requires data from heterogeneous sources such as building management systems (BMS), Internet-of-Things (IoT) sensor networks, and building information models. Data interoperability is a key component to eliminate silos of information, and using semantic web technologies like the BRICK schema, an effort to standardize semantic descriptions of the physical, logical, and virtual assets in buildings and the relationships between them, is a suitable approach. However, current data integration processes can involve significant manual interventions. This paper presents a methodology to automatically collect, assemble, and integrate information from a building information model to a knowledge graph. The resulting application, called BIM-to-BRICK, is run on the SDE4 building located in Singapore. BIM-to-BRICK generated a bidirectional link between a BIM model of 932 instances and experimental data collected for 17 subjects into 458 BRICK objects and 1219 relationships in 17 seconds. The automation of this approach can be compared to traditional manual map** of data types. This scientific innovation incentivizes the convergence of disparate data types and structures in built-environment applications.
△ Less
Submitted 24 July, 2023;
originally announced July 2023.
-
Deep learning-based Segmentation of Rabbit fetal skull with limited and sub-optimal annotations
Authors:
Rajath Soans,
Alexa Gleason,
Tosha Shah,
Corey Miller,
Barbara Robinson,
Kimberly Brannen,
Antong Chen
Abstract:
In this paper, we propose a deep learning-based method to segment the skeletal structures in the micro-CT images of Dutch-Belted rabbit fetuses which can assist in the assessment of drug-induced skeletal abnormalities as a required study in developmental and reproductive toxicology (DART). Our strategy leverages sub-optimal segmentation labels of 22 skull bones from 26 micro-CT volumes and maps th…
▽ More
In this paper, we propose a deep learning-based method to segment the skeletal structures in the micro-CT images of Dutch-Belted rabbit fetuses which can assist in the assessment of drug-induced skeletal abnormalities as a required study in developmental and reproductive toxicology (DART). Our strategy leverages sub-optimal segmentation labels of 22 skull bones from 26 micro-CT volumes and maps them to 250 unlabeled volumes on which a deep CNN-based segmentation model is trained. In the experiments, our model was able to achieve an average Dice Similarity Coefficient (DSC) of 0.89 across all bones on the testing set, and 14 out of the 26 skull bones reached average DSC >0.93. Our next steps are segmenting the whole body followed by develo** a model to classify abnormalities.
△ Less
Submitted 24 May, 2023;
originally announced July 2023.
-
Development and Evaluation of an Online Home Energy Management Strategy for Load Coordination in Smart Homes with Renewable Energy Sources
Authors:
Xiaoling Chen,
Cory Miller,
Mithun Goutham,
Prasad Dev Hanumalagutti,
Rachel Blaser,
Stephanie Stockar
Abstract:
In this paper, a real time implementable load coordination strategy is developed for the optimization of electric demands in a smart home. The strategy minimizes the electricity cost to the home owner, while limiting the disruptions associated with the deferring of flexible power loads. A multi-objective nonlinear mixed integer programming is formulated as a sequential model predictive control, wh…
▽ More
In this paper, a real time implementable load coordination strategy is developed for the optimization of electric demands in a smart home. The strategy minimizes the electricity cost to the home owner, while limiting the disruptions associated with the deferring of flexible power loads. A multi-objective nonlinear mixed integer programming is formulated as a sequential model predictive control, which is then solved using genetic algorithm. The load shifting benefits obtained by deploying an advanced coordination strategy are compared against a baseline controller for various home characteristics, such as location, size and equipment. The simulation study shows that the deployment of the smart home energy management strategy achieves approximately 5% reduction in grid cost compared to a baseline strategy. This is achieved by deferring approximately 50\% of the flexible loads, which is possible due to the use of the stationary energy storage.
△ Less
Submitted 23 April, 2023;
originally announced April 2023.
-
Longitudinal thermal imaging for scalable non-residential HVAC and occupant behaviour characterization
Authors:
Vasantha Ramani,
Miguel Martin,
Pandarasamy Arjunan,
Adrian Chong,
Kameshwar Poolla,
Clayton Miller
Abstract:
This work presents a study on the characterization of the air-conditioning (AC) usage pattern of non-residential buildings from thermal images collected from an urban-scale infrared (IR) observatory. To achieve this first, an image processing scheme, for cleaning and extraction of the temperature time series from the thermal images is implemented. To test the accuracy of the thermal measurements u…
▽ More
This work presents a study on the characterization of the air-conditioning (AC) usage pattern of non-residential buildings from thermal images collected from an urban-scale infrared (IR) observatory. To achieve this first, an image processing scheme, for cleaning and extraction of the temperature time series from the thermal images is implemented. To test the accuracy of the thermal measurements using IR camera, the extracted temperature is compared against the ground truth surface temperature measurements. It is observed that the detrended thermal measurements match well with the ground truth surface temperature measurements. Subsequently, the operational pattern of the water-cooled systems and window AC units are extracted from the analysis of the thermal signature. It is observed that for the water-cooled system, the difference between the rate of change of the window and wall can be used to extract the operational pattern. While, in the case of the window AC units, wavelet transform of the AC unit temperature is used to extract the frequency and time domain information of the AC unit operation. The results of the analysis are compared against the indoor temperature sensors installed in the office spaces of the building. It is realized that the accuracy in the prediction of the operational pattern is highest between 8 pm to 10 am, and it reduces during the day because of solar radiation and high daytime temperature. Subsequently, a characterization study is conducted for eight window/split AC units from the thermal image collected during the nighttime. This forms one of the first studies on the operational behavior of HVAC systems for non-residential buildings using the longitudinal thermal imaging technique. The output from this study can be used to better understand the operational and occupant behavior, without requiring to deploy a large array of sensors in the building space.
△ Less
Submitted 20 March, 2023; v1 submitted 16 November, 2022;
originally announced November 2022.
-
Optimal Phase Design for RIS Channel Estimation
Authors:
Chelsea L. Miller,
Peter J. Smith,
Pawel A. Dmochowski
Abstract:
We develop an optimal version of a prior two-stage channel estimation protocol for RIS-assisted channels. The new design uses a modified DFT matrix (MDFT) for the training phases at the RIS and is shown to minimize the total channel estimation error variance. In conjunction with interpolation (estimating fewer RIS channels), the MDFT approach accelerates channel estimation even when the channel fr…
▽ More
We develop an optimal version of a prior two-stage channel estimation protocol for RIS-assisted channels. The new design uses a modified DFT matrix (MDFT) for the training phases at the RIS and is shown to minimize the total channel estimation error variance. In conjunction with interpolation (estimating fewer RIS channels), the MDFT approach accelerates channel estimation even when the channel from base station to RIS is line-of-sight. In contrast, prior two-stage techniques required a full-rank channel for efficient estimation. We investigate the resulting channel estimation errors by comparing different training phase designs for a variety of propagation conditions using a ray-based channel model. To examine the overall performance, we simulate the spectral efficiency with MRC processing for a single-user RIS-assisted system using an existing optimal design for the RIS transmission phases. Results verify the optimality of MDFT while simulations and analysis show that the performance is more dependent on the user-to-RIS channel correlation and the coarseness of the interpolation used, rather than the training phase design. For example, under a scenario with more highly correlated channels, the procedure accelerates channel estimation by a factor of 16, while the improvement is a factor of 5 in a less correlated case. The overall procedure is extremely robust, with a maximum performance loss of 1.5bits/sec/Hz compared to that with perfect channel state information for the considered channel conditions.
△ Less
Submitted 18 October, 2022;
originally announced October 2022.
-
Elastic buildings: Calibrated district-scale simulation of occupant-flexible campus operation for hybrid work optimization
Authors:
Martín Mosteiro-Romero,
Clayton Miller,
Adrian Chong,
Rudi Stouffs
Abstract:
Before 2020, the way occupants utilized the built environment had been changing slowly towards scenarios in which occupants have more choice and flexibility in where and how they work. The global COVID-19 pandemic accelerated this phenomenon rapidly through lockdowns and hybrid work arrangements. Many occupants and employers are considering kee** some of these flexibility-based strategies due to…
▽ More
Before 2020, the way occupants utilized the built environment had been changing slowly towards scenarios in which occupants have more choice and flexibility in where and how they work. The global COVID-19 pandemic accelerated this phenomenon rapidly through lockdowns and hybrid work arrangements. Many occupants and employers are considering kee** some of these flexibility-based strategies due to their benefits and cost impacts. This paper simulates various scenarios related to the operational technologies and policies of a real-world campus using a district-scale City Energy Analyst (CEA) model that is calibrated with measured energy and occupancy profiles extracted from WiFi data. These scenarios demonstrate the energy impact of ram** building operations up and down more rapidly and effectively to the flex-based work strategies that may solidify. The scenarios show a 4-12% decrease in space cooling demand due to occupant absenteeism if centralized building system operation is in place, but as high as 21-68% if occupancy-driven building controls are implemented. The paper discusses technologies and strategies that are important in this paradigm shift of operations.
△ Less
Submitted 24 April, 2023; v1 submitted 22 August, 2022;
originally announced October 2022.
-
Deep Learning for Classification of Thyroid Nodules on Ultrasound: Validation on an Independent Dataset
Authors:
**gxi Weng,
Benjamin Wildman-Tobriner,
Mateusz Buda,
Jichen Yang,
Lisa M. Ho,
Brian C. Allen,
Wendy L. Ehieli,
Chad M. Miller,
Jikai Zhang,
Maciej A. Mazurowski
Abstract:
Objectives: The purpose is to apply a previously validated deep learning algorithm to a new thyroid nodule ultrasound image dataset and compare its performances with radiologists. Methods: Prior study presented an algorithm which is able to detect thyroid nodules and then make malignancy classifications with two ultrasound images. A multi-task deep convolutional neural network was trained from 127…
▽ More
Objectives: The purpose is to apply a previously validated deep learning algorithm to a new thyroid nodule ultrasound image dataset and compare its performances with radiologists. Methods: Prior study presented an algorithm which is able to detect thyroid nodules and then make malignancy classifications with two ultrasound images. A multi-task deep convolutional neural network was trained from 1278 nodules and originally tested with 99 separate nodules. The results were comparable with that of radiologists. The algorithm was further tested with 378 nodules imaged with ultrasound machines from different manufacturers and product types than the training cases. Four experienced radiologists were requested to evaluate the nodules for comparison with deep learning. Results: The Area Under Curve (AUC) of the deep learning algorithm and four radiologists were calculated with parametric, binormal estimation. For the deep learning algorithm, the AUC was 0.69 (95% CI: 0.64 - 0.75). The AUC of radiologists were 0.63 (95% CI: 0.59 - 0.67), 0.66 (95% CI:0.61 - 0.71), 0.65 (95% CI: 0.60 - 0.70), and 0.63 (95%CI: 0.58 - 0.67). Conclusion: In the new testing dataset, the deep learning algorithm achieved similar performances with all four radiologists. The relative performance difference between the algorithm and the radiologists is not significantly affected by the difference of ultrasound scanner.
△ Less
Submitted 4 May, 2023; v1 submitted 27 July, 2022;
originally announced July 2022.
-
A Semi Empirical Approach to a Physically Based Aging Model for Home Energy Management Systems
Authors:
Cory Miller,
Mithun Goutham,
Xiaoling Chen,
Prasad Dev Hanumalagutti,
Rachel Blaser,
Stephanie Stockar
Abstract:
A growing interest in the study of aging related phenomena in lithium-ion batteries is propelled by the increasing utilization of energy storage systems in electric vehicles and in buildings as stationery energy accumulators paired with renewable energy sources. This paper proposes a mixed-degradation model approach that combines the benefits of a semi-empirical approach with that of a physics-bas…
▽ More
A growing interest in the study of aging related phenomena in lithium-ion batteries is propelled by the increasing utilization of energy storage systems in electric vehicles and in buildings as stationery energy accumulators paired with renewable energy sources. This paper proposes a mixed-degradation model approach that combines the benefits of a semi-empirical approach with that of a physics-based model. This enables easy calibration for different battery chemistries, the ability to extrapolate when necessary, and is computationally efficient enough to be coupled with real-time running control systems. To demonstrate the effectiveness of the proposed approach, the effect of two different control strategies in a smart home energy management system is demonstrated on the aging of a Lithium iron phosphate (LFP) battery.
△ Less
Submitted 13 June, 2022;
originally announced June 2022.
-
Space-Constrained Arrays for Massive MIMO
Authors:
Chelsea L. Miller,
Peter J. Smith,
Pawel A. Dmochowski
Abstract:
We analyse the behaviour of a massive multi-user MIMO (MU-MIMO) system comprising a base station (BS) equipped with one of five different antenna topologies for which the spatial aperture is either unconstrained, or space-constrained. We derive the normalized mean interference (NMI) with a ray-based channel model, as a metric for topology comparison in each of the two cases. Based on the derivatio…
▽ More
We analyse the behaviour of a massive multi-user MIMO (MU-MIMO) system comprising a base station (BS) equipped with one of five different antenna topologies for which the spatial aperture is either unconstrained, or space-constrained. We derive the normalized mean interference (NMI) with a ray-based channel model, as a metric for topology comparison in each of the two cases. Based on the derivation for a horizontal uniform rectangular array (HURA) in [1], we provide closed-form NMI equations for the uniform linear array (ULA) and uniform circular array (UCirA). We then derive the same for a vertical URA (VURA) and uniform cylindrical array (UCylA). Results for the commonly-considered unconstrained case confirm the prior understanding that topologies with wider azimuth footprints aid performance. However, in the space-constrained case performance is dictated by the angular resolution afforded by the topology, particularly in elevation. We confirm the behavioural patterns predicted by the NMI by observing the same patterns in the system SINR with minimum mean-squared error (MMSE) processing.
△ Less
Submitted 26 October, 2020;
originally announced October 2020.
-
Nonlinear Model Predictive Control of Permanent Magnet Synchronous Generators in DC Microgrids
Authors:
Luis Herrera,
Chad Miller,
Bang-Hung Tsao
Abstract:
A new strategy is proposed to control interior permanent magnet generators in dc microgrids interfaced through an active rectifier. The controller design is based on the decomposition of the system dynamics into slow and fast modes using singular perturbation theory. An inner current controller is developed based on output regulation techniques and an outer voltage controller is proposed using Non…
▽ More
A new strategy is proposed to control interior permanent magnet generators in dc microgrids interfaced through an active rectifier. The controller design is based on the decomposition of the system dynamics into slow and fast modes using singular perturbation theory. An inner current controller is developed based on output regulation techniques and an outer voltage controller is proposed using Nonlinear Model Predictive Control (NMPC). The NMPC regulates the dc bus voltage and minimizes the ac side losses. Simulation results are then presented based on realistic conditions for aircraft power systems.
△ Less
Submitted 8 October, 2020;
originally announced October 2020.
-
Favorable Propagation with User Cluster Sharing
Authors:
Chelsea Miller,
Peter J. Smith,
Pawel A. Dmochowski,
Harsh Tataria,
Andreas F. Molisch
Abstract:
We examine the favorable propagation (FP) behavior of a massive multi-user multiple-input-multiple-output (MU-MIMO) system equipped with a uniform linear array (ULA), horizontal uniform rectangular array (HURA) or uniform circular array (UCA) using a ray-based channel model with user cluster sharing. We demonstrate FP for these systems and provide analytical expressions for the mean-squared distan…
▽ More
We examine the favorable propagation (FP) behavior of a massive multi-user multiple-input-multiple-output (MU-MIMO) system equipped with a uniform linear array (ULA), horizontal uniform rectangular array (HURA) or uniform circular array (UCA) using a ray-based channel model with user cluster sharing. We demonstrate FP for these systems and provide analytical expressions for the mean-squared distance (MSD) of the FP metric from its large-system limit for each of the aforementioned topologies. We use these results to examine the detrimental effects of user cluster sharing on FP behavior, and demonstrate the superior performance of the ULA as compared to the UCA and the HURA with equal inter-element spacing. Although cluster sharing has a negative impact on FP for finite arrays, we additionally examine the asymptotic rate of convergence to FP as a function of array size and show that this rate is unchanged with or without user cluster sharing.
△ Less
Submitted 25 June, 2020;
originally announced June 2020.
-
EnergyStar++: Towards more accurate and explanatory building energy benchmarking
Authors:
Pandarasamy Arjunan,
Kameshwar Poolla,
Clayton Miller
Abstract:
Building energy performance benchmarking has been adopted widely in the USA and Canada through the Energy Star Portfolio Manager platform. Building operations and energy management professionals have long used a simple 1-100 score to understand how their building compares to its peers. This single number is easy to use, but is created by inaccurate linear regression (MLR) models. This paper propos…
▽ More
Building energy performance benchmarking has been adopted widely in the USA and Canada through the Energy Star Portfolio Manager platform. Building operations and energy management professionals have long used a simple 1-100 score to understand how their building compares to its peers. This single number is easy to use, but is created by inaccurate linear regression (MLR) models. This paper proposes a methodology that enhances the existing Energy Star calculation method by increasing accuracy and providing additional model output processing to help explain why a building is achieving a certain score. We propose and test two new prediction models: multiple linear regression with feature interactions (MLRi) and gradient boosted trees (GBT). Both models have better average accuracy than the baseline Energy Star models. The third order MLRi and GBT models achieve 4.9% and 24.9% increase in adjusted R2, respectively, and 7.0% and 13.7% decrease in normalized root mean squared error (NRMSE), respectively, on average than MLR models for six building types. Even more importantly, a set of techniques is developed to help determine which factors most influence the score using SHAP values. The SHAP force visualization in particular offers an accessible overview of the aspects of the building that influence the score that non-technical users can readily interpret. This methodology is tested on the 2012 Commercial Building Energy Consumption Survey (CBECS)(1,812 buildings) and public data sets from the energy disclosure programs of New York City (11,131 buildings) and Seattle (2,073 buildings).
△ Less
Submitted 17 June, 2020; v1 submitted 30 October, 2019;
originally announced October 2019.
-
Data-driven Analysis of Regional Capacity Factors in a Large-Scale Power Market: A Perspective from Market Participants
Authors:
Zhongyang Zhao,
Caisheng Wang,
Huaiwei Liao,
Carol J. Miller
Abstract:
A competitive wholesale electricity market consists of thousands of interacting market participants. Driven by the variations of fuel costs, system loads and weathers, these market participants compete actively and behave variously in the power market. Although electricity markets tend to become more transparent, a large amount of market information is still not publicly available to market partic…
▽ More
A competitive wholesale electricity market consists of thousands of interacting market participants. Driven by the variations of fuel costs, system loads and weathers, these market participants compete actively and behave variously in the power market. Although electricity markets tend to become more transparent, a large amount of market information is still not publicly available to market participants. Hence, data-driven analysis based on public data is crucial for market participants to better understand and model large-scale power markets, and ultimately to perform better in power trading. While most of the previous researches related to the large-scale power markets are based on the synthetic networks, a data-driven approach utilizing the real power market data is proposed in this paper. First, the power plants' monthly net generation and capacity data are obtained from U.S. Energy Information Administration (EIA) and aggregated to figure out the monthly regional capacity factors which are used to characterize the market's regional behaviors for market participants. Then, the regional capacity factors are analyzed against the metered system loads and natural gas prices to study the generation behaviors in the power market. The analysis reveals the impacts of regional natural gas prices on capacity factors and the responses of generating behaviors to the system loads. The analysis results present the solid evidence and rational references for market participants to model and validate the large-scale power market in the future.
△ Less
Submitted 31 October, 2019;
originally announced October 2019.
-
Improving Short-Term Electricity Price Forecasting Using Day-Ahead LMP with ARIMA Models
Authors:
Zhongyang Zhao,
Caisheng Wang,
Matthew Nokleby,
Carol Miller
Abstract:
Short-term electricity price forecasting has become important for demand side management and power generation scheduling. Especially as the electricity market becomes more competitive, a more accurate price prediction than the day-ahead locational marginal price (DALMP) published by the independent system operator (ISO) will benefit participants in the market by increasing profit or improving load…
▽ More
Short-term electricity price forecasting has become important for demand side management and power generation scheduling. Especially as the electricity market becomes more competitive, a more accurate price prediction than the day-ahead locational marginal price (DALMP) published by the independent system operator (ISO) will benefit participants in the market by increasing profit or improving load demand scheduling. Hence, the main idea of this paper is to use autoregressive integrated moving average (ARIMA) models to obtain a better LMP prediction than the DALMP by utilizing the published DALMP, historical real-time LMP (RTLMP) and other useful information. First, a set of seasonal ARIMA (SARIMA) models utilizing the DALMP and historical RTLMP are developed and compared with autoregressive moving average (ARMA) models that use the differences between DALMP and RTLMP on their forecasting capability. A generalized autoregressive conditional heteroskedasticity (GARCH) model is implemented to further improve the forecasting by accounting for the price volatility. The models are trained and evaluated using real market data in the Midcontinent Independent System Operator (MISO) region. The evaluation results indicate that the ARMAX-GARCH model, where an exogenous time series indicates weekend days, improves the short-term electricity price prediction accuracy and outperforms the other proposed ARIMA models
△ Less
Submitted 4 January, 2018;
originally announced January 2018.
-
Improvement to the Prediction of Fuel Cost Distributions Using ARIMA Model
Authors:
Zhongyang Zhao,
Chang Fu,
Caisheng Wang,
Carol Miller
Abstract:
Availability of a validated, realistic fuel cost model is a prerequisite to the development and validation of new optimization methods and control tools. This paper uses an autoregressive integrated moving average (ARIMA) model with historical fuel cost data in development of a three-step-ahead fuel cost distribution prediction. First, the data features of Form EIA-923 are explored and the natural…
▽ More
Availability of a validated, realistic fuel cost model is a prerequisite to the development and validation of new optimization methods and control tools. This paper uses an autoregressive integrated moving average (ARIMA) model with historical fuel cost data in development of a three-step-ahead fuel cost distribution prediction. First, the data features of Form EIA-923 are explored and the natural gas fuel costs of Texas generating facilities are used to develop and validate the forecasting algorithm for the Texas example. Furthermore, the spot price associated with the natural gas hub in Texas is utilized to enhance the fuel cost prediction. The forecasted data is fit to a normal distribution and the Kullback-Leibler divergence is employed to evaluate the difference between the real fuel cost distributions and the estimated distributions. The comparative evaluation suggests the proposed forecasting algorithm is effective in general and is worth pursuing further.
△ Less
Submitted 22 February, 2018; v1 submitted 4 January, 2018;
originally announced January 2018.
-
Optimal Control of Conditional Value-at-Risk in Continuous Time
Authors:
Christopher W. Miller,
Insoon Yang
Abstract:
We consider continuous-time stochastic optimal control problems featuring Conditional Value-at-Risk (CVaR) in the objective. The major difficulty in these problems arises from time-inconsistency, which prevents us from directly using dynamic programming. To resolve this challenge, we convert to an equivalent bilevel optimization problem in which the inner optimization problem is standard stochasti…
▽ More
We consider continuous-time stochastic optimal control problems featuring Conditional Value-at-Risk (CVaR) in the objective. The major difficulty in these problems arises from time-inconsistency, which prevents us from directly using dynamic programming. To resolve this challenge, we convert to an equivalent bilevel optimization problem in which the inner optimization problem is standard stochastic control. Furthermore, we provide conditions under which the outer objective function is convex and differentiable. We compute the outer objective's value via a Hamilton-Jacobi-Bellman equation and its gradient via the viscosity solution of a linear parabolic equation, which allows us to perform gradient descent. The significance of this result is that we provide an efficient dynamic programming-based algorithm for optimal control of CVaR without lifting the state-space. To broaden the applicability of the proposed algorithm, we propose convergent approximation schemes in cases where our key assumptions do not hold and characterize relevant suboptimality bounds. In addition, we extend our method to a more general class of risk metrics, which includes mean-variance and median-deviation. We also demonstrate a concrete application to portfolio optimization under CVaR constraints. Our results contribute an efficient framework for solving time-inconsistent CVaR-based sequential optimization.
△ Less
Submitted 22 January, 2017; v1 submitted 15 December, 2015;
originally announced December 2015.
-
Optimal Dynamic Contracts for a Large-Scale Principal-Agent Hierarchy: A Concavity-Preserving Approach
Authors:
Christopher W. Miller,
Insoon Yang
Abstract:
We present a continuous-time contract whereby a top-level player can incentivize a hierarchy of players below him to act in his best interest despite only observing the output of his direct subordinate. This paper extends Sannikov's approach from a situation of asymmetric information between a principal and an agent to one of hierarchical information between several players. We develop an iterativ…
▽ More
We present a continuous-time contract whereby a top-level player can incentivize a hierarchy of players below him to act in his best interest despite only observing the output of his direct subordinate. This paper extends Sannikov's approach from a situation of asymmetric information between a principal and an agent to one of hierarchical information between several players. We develop an iterative algorithm for constructing an incentive compatible contract and define the correct notion of concavity which must be preserved during iteration. We identify conditions under which a dynamic programming construction of an optimal dynamic contract can be reduced to only a one-dimensional state space and one-dimensional control set, independent of the size of the hierarchy. In this sense, our results contribute to the applicability of dynamic programming on dynamic contracts for a large-scale principal-agent hierarchy.
△ Less
Submitted 17 June, 2015;
originally announced June 2015.
