P. Teja Kuruganti

Non-Intrusive Energy Disaggregation Using Non-Negative Matrix Factorization With Sum-to-k Constraint

Energy disaggregation or non-intrusive load monitoring addresses the issue of extracting device-level energy consumption information by monitoring the aggregated signal at one single measurement point without installing meters on each individual device. Energy disaggregation can be formulated as a source separation problem, where the aggregated signal is expressed as linear combination of basis vectors in a matrix factorization framework. In this paper, an approach based on Sum-to-k constrained non-negative matrix factorization (S2K-NMF) is proposed. By imposing the sum-to-k constraint and the non-negative constraint, S2K-NMF is able to effectively extract perceptually meaningful sources from complex mixtures. The strength of the proposed algorithm is demonstrated through two sets of experiments: Energy disaggregation in a residential smart home; and heating, ventilating, and air conditioning components energy monitoring in an industrial building testbed maintained at the Oak Ridge National Laboratory. Extensive experimental results demonstrate the superior performance of S2K-NMF as compared to state-of-the-art decomposition-based disaggregation algorithms.

Security of Classic PN-Spreading Codes for Hybrid DS/FH Spread-Spectrum Systems

Hybrid direct sequence/frequency hopping (DS/FH) spread-spectrum communication systems have recently received considerable interest in commercial applications in addition to their use in military communications because they accommodate high data rates with high link integrity, even in the presence of significant multipath effects and interfering signals. The security of hybrid DS/FH systems strongly depends on the choice of PN-spreading code employed. In this paper, we examine the security, in terms of unicity distance, of linear maximal-length, Gold, and Kasami PN-spreading codes for DS, FH, and hybrid DS/FH spread-spectrum systems without additional encryption methods. The unicity distance is a measure of the minimum amount of ciphertext required by an eavesdropper to uniquely determine the specific key used in a cryptosystem and hence break the cipher. Numerical results are presented to compare the security of the considered PN-spreading codes under known-ciphertext attacks.

PHY and MAC layer design of Hybrid Spread Spectrum based smart meter network

The selection of the appropriate communication technology for different smart grid applications has drawn a great attention in the recent past. In this paper, we propose a Hybrid Spread Spectrum (HSS) based Advanced smart Metering Infrastructure (AMI) that reduces the overhead and latency in data transfer when compared to the use of 3G/4G technologies for smart meter data collection. We present a preliminary PHY and MAC layer design of a HSS based AMI network and evaluate their performance using matlab and NS2 simulations.

Application of optimal production control theory for home energy management in a micro grid

We consider the optimal stochastic control problem for home energy systems with solar and energy storage devices when the demand is realized from the grid. The demand is subject to Brownian motions with both drift and variance parameters modulated by a continuous-time Markov chain that represents the regime of electricity price. We model the systems as pure stochastic differential equation models, and then we follow the completing square technique to solve the stochastic home energy management problem. The effectiveness of the efficiency of the proposed approach is validated through a simulation example. For practical situations with constraints consistent to those studied here, our results imply the proposed framework could reduce the electricity cost from short-term purchase in peak hour market.

Very short-term photovoltaic power forecasting using uncertain basis function

Solar photovoltaics (PV), one of the most promising and rapidly developing renewable energy technologies, has evolved towards becoming a main renewable electricity source. It is termed variable energy resources since solar irradiance is intermittent in nature. This variability is a critical factor when predicting the available energy of solar sources. Capital and operational costs associated with solar PV implementation are highly affected when inaccurate predictions are carried out. This paper presents a new forecasting model for solar PV by utilizing historical inter-minute data to outline a short-term probabilistic model of solar. The proposed methodology employs a probabilistic approach to predict short-term solar PV power based on uncertain basis functions. The PV forecasting model is applied to power generation from a 13.5 kW rooftop PV panel installed on the Distributed Energy, Communications, and Controls (DECC) laboratory at Oak Ridge National Laboratory. The results are compared with standard time series approach, which have shown a substantial improvement in the prediction accuracy of the total solar energy produced.

Home energy management based on optimal production control scheduling with unknown regime switching

We propose a novel home energy management framework to intelligently schedule the distributed energy storage (DES) for the cost reduction of customers in this paper. The proposed optimal production control technique determines the action policy (e.g., charging or discharging) and the power allocation policy of the DES to provide DES power at proper time with lower price than that of the utility grid, resulting in the reduction of the long term financial cost. Specifically, we first formulate the optimal decision problem for home energy systems with solar and energy storage devices, when the demand, renewable energy, electricity purchase from grid are all subject to Brownian motions. Both drift and variance parameters are modulated by a continuous-time Markov chain that represents the regime of electricity price. In particular, we set up a mean-variance problem where the cost function is both the running cost of diesel generator and deviation from the target State of Charge (SOC) of batteries. We assume the regime information follows a Hidden Markov Model (HMM), and then estimate the state by change of measure based on the Girsanovs theorem. Finally, the problem boils down to solving a stochastic differential equation (SDE), which we provide both the explicit and numerical solutions to this specific SDE. An example is provided to illustrate the effectiveness of our proposed approach. Moreover, we compare it with the traditional Model Predictive Control (MPC) technique, and show it outperforms MPC.

SMAC: A Soft MAC to Reduce Control Overhead and Latency in CDMA-Based AMI Networks

The use of state-of-the-art 3G cellular CDMA technologies in a utility owned AMI network results in a large amount of control traffic relative to data traffic, increases the average packet delay and hence are not an appropriate choice for smart grid distribution applications. Like the CDG, we consider a utility owned cellular like CDMA network for smart grid distribution applications and classify the distribution smart grid data as scheduled data and random data. Also, we propose SMAC protocol, which changes its mode of operation based on the type of the data being collected to reduce the data collection latency and control overhead when compared to 3G cellular CDMA2000 MAC. The reduction in the data collection latency and control overhead aids in increasing the number of smart meters served by a base station within the periodic data collection interval, which further reduces the number of base stations needed by a utility or reduces the bandwidth needed to collect data from all the smart meters. The reduction in the number of base stations and/or the reduction in the data transmission bandwidth reduces the CAPital EXpenditure (CAPEX) and OPerational EXpenditure (OPEX) of the AMI network. The proposed SMAC protocol is analyzed using markov chain, analytical expressions for average throughput and average packet delay are derived, and simulation results are also provided to verify the analysis.

Statistical analysis of solar PV power frequency spectrum for optimal employment of building loads

In this paper, a statistical analysis of the frequency spectrum of solar photovoltaic (PV) power output is conducted. This analysis quantifies the frequency content that can be used for purposes such as developing optimal employment of building loads and distributed energy resources. One year of solar PV power output data was collected and analyzed using one-second resolution to find ideal bounds and levels for the different frequency components. The annual, seasonal, and monthly statistics of the PV frequency content are computed and illustrated in boxplot format. To examine the compatibility of building loads for PV consumption, a spectral analysis of building loads such as Heating, Ventilation and AirConditioning (HVAC) units and water heaters was performed. This defined the bandwidth over which these devices can operate. Results show that nearly all of the PV output (about 98%) is contained within frequencies lower than 1 mHz (equivalent to ∼15 min), which is compatible for consumption with local building loads such as HVAC units and water heaters. Medium frequencies in the range of ∼15 min to ∼1 min are likely to be suitable for consumption by fan equipment of variable air volume HVAC systems that have time constants in the range of few seconds to few minutes. This study indicates that most of the PV generation can be consumed by building loads with the help of proper control strategies, thereby reducing impact on the grid and the size of storage systems.

Real-time urban population monitoring using pervasive sensor network

It is estimated that 50% of the global population lives in urban areas occupying just 0.4% of the Earths surface. Understanding urban activity constitutes monitoring population density and its changes over time, in urban environments. Currently, there are limited mechanisms to non-intrusively monitor population density in real-time. The pervasive use of cellular phones in urban areas is one such mechanism that provides a unique opportunity to study population density by monitoring the mobility patterns in near real-time. Cellular carriers such as AT&T harvest such data through their cell towers; however, this data is proprietary and the carriers restrict access, due to privacy concerns. In this work, we propose a system that passively senses the population density and infers mobility patterns in an urban area by monitoring power spectral density in cellular frequency bands using periodic beacons from each cellphone without knowing who and where they are located. A wireless sensor network platform is being developed to perform spectral monitoring along with environmental measurements. Algorithms are developed to generate real-time fine-resolution population estimates.

OTRA-THS MAC to reduce Power Outage Data Collection Latency in a smart meter network

The deployment of advanced metering infrastructure by electric utilities poses unique communication challenges, particularly as the number of meters per aggregator increases. When there is a power outage, a smart meter tries to report it instantaneously to the electric utility. In a densely populated residential/industrial locality, it is possible that a large number of smart meters simultaneously try to get access to the communication network to report the power outage. If the number of smart meters is very high on the order of tens of thousands (metropolitan areas), the power outage data flooding can lead to Random Access CHannel (RACH) congestion. Several utilities are considering the use of cellular network for smart meter communications. In 3G/4G cellular networks, RACH congestion not only leads to collisions, retransmissions and increased RACH delays, but also has the potential to disrupt the dedicated traffic flow by increasing the interference levels (3G CDMA). In order to overcome this problem, in this paper we propose a Time Hierarchical Scheme (THS) that reduces the intensity of power outage data flooding and power outage reporting delay by 6/7th, and 17/18th when compared to their respective values without THS. Also, we propose an Optimum Transmission Rate Adaptive (OTRA) MAC to optimize the latency in power outage data collection. The analysis and simulation results presented in this paper show that both the OTRA and THS features of the proposed MAC results in a Power Outage Data Collection Latency (PODCL) that is 1/10th of the 4G LTE PODCL.