scispace - formally typeset
Search or ask a question

Showing papers on "Distributed algorithm published in 2015"


Posted Content
TL;DR: This work presents the first massively distributed architecture for deep reinforcement learning, using a distributed neural network to represent the value function or behaviour policy, and a distributed store of experience to implement the Deep Q-Network algorithm.
Abstract: We present the first massively distributed architecture for deep reinforcement learning. This architecture uses four main components: parallel actors that generate new behaviour; parallel learners that are trained from stored experience; a distributed neural network to represent the value function or behaviour policy; and a distributed store of experience. We used our architecture to implement the Deep Q-Network algorithm (DQN). Our distributed algorithm was applied to 49 games from Atari 2600 games from the Arcade Learning Environment, using identical hyperparameters. Our performance surpassed non-distributed DQN in 41 of the 49 games and also reduced the wall-time required to achieve these results by an order of magnitude on most games.

464 citations


Journal ArticleDOI
TL;DR: A novel distributed coordinated controller combined with a multiagent-based consensus algorithm is applied to distributed generators in the Energy Internet, which keeps voltage angles and amplitudes consensus, while providing accurate power-sharing and minimizing circulating currents.
Abstract: With the bidirectional power flow provided by the Energy Internet, various methods are promoted to improve and increase the energy utilization between Energy Internet and main grid (MG). This paper proposes a novel distributed coordinated controller combined with a multiagent-based consensus algorithm, which is applied to distributed generators in the Energy Internet. Then, the decomposed tasks, models, and information flow of the proposed method are analyzed. The proposed coordinated controller installed between the Energy Internet and MG keeps voltage angles and amplitudes consensus, while providing accurate power-sharing and minimizing circulating currents. Finally, the Energy Internet can be integrated into the MG seamlessly if necessary. Hence, the Energy Internet can be operated as a spinning reserve system. Simulation results are provided to show the effectiveness of the proposed controller in an Energy Internet.

329 citations


Journal ArticleDOI
TL;DR: This paper provides sufficient conditions under which the optimization problem can be solved via its convex relaxation, and demonstrates the operation of the algorithm, including its robustness against communication link failures, through several case studies involving 5-, 34-, and 123-bus power distribution systems.
Abstract: This paper addresses the problem of voltage regulation in power distribution networks with deep-penetration of distributed energy resources, e.g., renewable-based generation, and storage-capable loads such as plug-in hybrid electric vehicles. We cast the problem as an optimization program, where the objective is to minimize the losses in the network subject to constraints on bus voltage magnitudes, limits on active and reactive power injections, transmission line thermal limits and losses. We provide sufficient conditions under which the optimization problem can be solved via its convex relaxation. Using data from existing networks, we show that these sufficient conditions are expected to be satisfied by most networks. We also provide an efficient distributed algorithm to solve the problem. The algorithm adheres to a communication topology described by a graph that is the same as the graph that describes the electrical network topology. We illustrate the operation of the algorithm, including its robustness against communication link failures, through several case studies involving 5-, 34-, and 123-bus power distribution systems.

314 citations


Proceedings ArticleDOI
01 Dec 2015
TL;DR: A new augmented distributed gradient method (termed Aug-DGM) based on consensus theory is developed that will be able to seek the exact optimum even with constant stepsizes assuming that the global objective function has Lipschitz gradient.
Abstract: We consider distributed optimization problems in which a number of agents are to seek the optimum of a global objective function through merely local information sharing. The problem arises in various application domains, such as resource allocation, sensor fusion and distributed learning. In particular, we are interested in scenarios where agents use uncoordinated (different) constant stepsizes for local optimization. According to most existing works, using this kind of stepsize rule for update, which is necessary in asynchronous scenarios, will lead to some gap (error) between the estimated result and the exact optimum. To deal with this issue, we develop a new augmented distributed gradient method (termed Aug-DGM) based on consensus theory. The proposed algorithm not only allows for using uncoordinated stepsizes but also, most importantly, be able to seek the exact optimum even with constant stepsizes assuming that the global objective function has Lipschitz gradient. A simple numerical example is provided to illustrate the effectiveness of the algorithm.

300 citations


Journal ArticleDOI
TL;DR: A distributed energy management approach based on the consensus + innovations method is presented and used to coordinate local generation, flexible load, and storage devices within the microgrid.
Abstract: Distributed energy resources and demand-side management are expected to become more prevalent in the future electric power system. Coordinating the increased number of grid participants in an efficient and reliable way is going to be a major challenge. A potential solution is the employment of a distributed energy management approach, which uses intelligence distributed over the grid to balance supply and demand. In this paper, we specifically consider the situation in which distributed resources and loads form microgrids within the bulk power system in which the load is supplied by local generation. A distributed energy management approach based on the consensus + innovations method is presented and used to coordinate local generation, flexible load, and storage devices within the microgrid. The approach takes advantage of the fact that, at the optimal allocation settings, the marginal costs given as a function of the power output/consumption need to be equal for all nonbinding network resources. Solutions for single time step, as well as multitime step optimization including intertemporal constraints, are presented.

286 citations


Journal ArticleDOI
TL;DR: Both theoretical and numerical results show that the optimal load sharing can be achieved within both generation and delivering constraints in a distributed way.

276 citations


Journal ArticleDOI
TL;DR: In this survey, the vertex-centric approach to graph processing is overviewed, TLAV frameworks are deconstructed into four main components and respectively analyzed, and TLAV implementations are reviewed and categorized.
Abstract: The vertex-centric programming model is an established computational paradigm recently incorporated into distributed processing frameworks to address challenges in large-scale graph processing. Billion-node graphs that exceed the memory capacity of commodity machines are not well supported by popular Big Data tools like MapReduce, which are notoriously poor performing for iterative graph algorithms such as PageRank. In response, a new type of framework challenges one to “think like a vertex” (TLAV) and implements user-defined programs from the perspective of a vertex rather than a graph. Such an approach improves locality, demonstrates linear scalability, and provides a natural way to express and compute many iterative graph algorithms. These frameworks are simple to program and widely applicable but, like an operating system, are composed of several intricate, interdependent components, of which a thorough understanding is necessary in order to elicit top performance at scale. To this end, the first comprehensive survey of TLAV frameworks is presented. In this survey, the vertex-centric approach to graph processing is overviewed, TLAV frameworks are deconstructed into four main components and respectively analyzed, and TLAV implementations are reviewed and categorized.

267 citations


Journal ArticleDOI
TL;DR: A sequential detector based on the generalized likelihood ratio is proposed to be robust to a variety of attacking strategies, and load situations in the power system, and its computational complexity linearly scales with the number of meters.
Abstract: We consider the sequential (i.e., online) detection of false data injection attacks in smart grid, which aims to manipulate the state estimation procedure by injecting malicious data to the monitoring meters. The unknown parameters in the system, namely the state vector, injected malicious data and the set of attacked meters pose a significant challenge for designing a robust, computationally efficient, and high-performance detector. We propose a sequential detector based on the generalized likelihood ratio to address this challenge. Specifically, the proposed detector is designed to be robust to a variety of attacking strategies, and load situations in the power system, and its computational complexity linearly scales with the number of meters. Moreover, it considerably outperforms the existing first-order cumulative sum detector in terms of the average detection delay and robustness to various attacking strategies. For wide-area monitoring in smart grid, we further develop a distributed sequential detector using an adaptive sampling technique called level-triggered sampling. The resulting distributed detector features single bit per sample in terms of the communication overhead, while preserving the high performance of the proposed centralized detector.

258 citations


Journal ArticleDOI
TL;DR: A fully distributed control strategy based on the consensus algorithm for the optimal resource management in an islanded microgrid is proposed through a multiagent system framework, which only requires information exchange among neighboring agents through a local network.
Abstract: A microgrid is a promising approach to provide clean, renewable, and reliable electricity by integrating various distributed generations and energy storage systems into power systems. However, highly intermittent renewable generations and various load demands pose new challenges to the optimal resource management in a microgrid. This paper proposes a fully distributed control strategy based on the consensus algorithm for the optimal resource management in an islanded microgrid. The proposed strategy is implemented through a multiagent system framework, which only requires information exchange among neighboring agents through a local network. The objective is achieved through a two-level control strategy. The upper control level is a consensus-based optimization algorithm that discovers the reference of optimal power generation or demand while maintaining the supply–demand balance. The lower control level is responsible for reference tracking of the associated component. Simulation results in the IEEE 14- and 162-bus systems are presented to demonstrate the effectiveness of the proposed control strategy.

251 citations


Journal ArticleDOI
TL;DR: A class of distributed Laplacian-gradient dynamics that are guaranteed to asymptotically find the solution to the economic dispatch problem with and without generator constraints are proposed.
Abstract: This paper considers the economic dispatch problem for a group of generator units communicating over an arbitrary weight-balanced digraph. The objective of the individual units is to collectively generate power to satisfy a certain load while minimizing the total generation cost, which corresponds to the sum of individual arbitrary convex functions. We propose a class of distributed Laplacian-gradient dynamics that are guaranteed to asymptotically find the solution to the economic dispatch problem with and without generator constraints. The proposed coordination algorithms are anytime, meaning that its trajectories are feasible solutions at any time before convergence, and they become better solutions as time elapses. In addition, we design the provably correct determine feasible allocation strategy that handles generator initialization and the addition and deletion of units via a message passing routine over a spanning tree of the network. Our technical approach combines notions and tools from algebraic graph theory, distributed algorithms, nonsmooth analysis, set-valued dynamical systems, and penalty functions. Simulations illustrate our results.

247 citations


Journal ArticleDOI
TL;DR: This work proposes a cross-layer distributed algorithm called interference-based topology control algorithm for delay-constrained (ITCD) MANETs with considering both the interference constraint and the delay constraint, which is different from the previous work.
Abstract: As the foundation of routing, topology control should minimize the interference among nodes, and increase the network capacity. With the development of mobile ad hoc networks (MANETs), there is a growing requirement of quality of service (QoS) in terms of delay. In order to meet the delay requirement, it is important to consider topology control in delay constrained environment, which is contradictory to the objective of minimizing interference. In this paper, we focus on the delay-constrained topology control problem, and take into account delay and interference jointly. We propose a cross-layer distributed algorithm called interference-based topology control algorithm for delay-constrained (ITCD) MANETs with considering both the interference constraint and the delay constraint, which is different from the previous work. The transmission delay, contention delay and the queuing delay are taken into account in the proposed algorithm. Moreover, the impact of node mobility on the interference-based topology control algorithm is investigated and the unstable links are removed from the topology. The simulation results show that ITCD can reduce the delay and improve the performance effectively in delay-constrained mobile ad hoc networks.

Journal ArticleDOI
TL;DR: A cellular computing model in the slime mold physarum polycephalum is exploited to solve the Steiner tree problem which is an important NP-hard problem in various applications, especially in network design.
Abstract: Using insights from biological processes could help to design new optimization techniques for long-standing computational problems. This paper exploits a cellular computing model in the slime mold physarum polycephalum to solve the Steiner tree problem which is an important NP-hard problem in various applications, especially in network design. Inspired by the path-finding and network formation capability of physarum, we develop a new optimization algorithm, named as the physarum optimization, with low complexity and high parallelism. To validate and evaluate our proposed models and algorithm, we further apply the physarum optimization to the minimal exposure problem which is a fundamental problem corresponding to the worst-case coverage in wireless sensor networks. Complexity analysis and simulation results show that our proposed algorithm could achieve good performance with low complexity. Moreover, the core mechanism of our physarum optimization also may provide a useful starting point to develop some practical distributed algorithms for network design.

Proceedings Article
06 Jul 2015
TL;DR: The algorithm is based on an inexact damped Newton method, where the inexact Newton steps are computed by a distributed preconditioned conjugate gradient method, and its iteration complexity and communication efficiency for minimizing self-concordant empirical loss functions are analyzed.
Abstract: We propose a new distributed algorithm for empirical risk minimization in machine learning. The algorithm is based on an inexact damped Newton method, where the inexact Newton steps are computed by a distributed preconditioned conjugate gradient method. We analyze its iteration complexity and communication efficiency for minimizing self-concordant empirical loss functions, and discuss the results for distributed ridge regression, logistic regression and binary classification with a smoothed hinge loss. In a standard setting for supervised learning, where the n data points are i.i.d. sampled and when the regularization parameter scales as 1/√n show that the proposed algorithm is communication efficient: the required round of communication does not increase with the sample size n, and only grows slowly with the number of machines.

Journal ArticleDOI
He Chen1, Yonghui Li1, Yunxiang Jiang, Yuanye Ma1, Branka Vucetic1 
TL;DR: Simulation results show that the proposed game-theoretical approach can achieve a near-optimal network-wide performance on average, especially for the scenarios with relatively low and moderate interference.
Abstract: In this paper, we consider simultaneous wireless information and power transfer (SWIPT) in relay interference channels, where multiple source-destination pairs communicate through their dedicated energy harvesting relays. Each relay needs to split its received signal from sources into two streams: one for information forwarding and the other for energy harvesting. We develop a distributed power splitting framework using game theory to derive a profile of power splitting ratios for all relays that can achieve a good network-wide performance. Specifically, non-cooperative games are respectively formulated for pure amplify-and-forward (AF) and decode-and-forward (DF) networks, in which each link is modeled as a strategic player who aims to maximize its own achievable rate. The existence and uniqueness for the Nash equilibriums (NEs) of the formulated games are analyzed and a distributed algorithm with provable convergence to achieve the NEs is also developed. Subsequently, the developed framework is extended to the more general network setting with mixed AF and DF relays. All the theoretical analyses are validated by extensive numerical results. Simulation results show that the proposed game-theoretical approach can achieve a near-optimal network-wide performance on average, especially for the scenarios with relatively low and moderate interference.

Journal ArticleDOI
TL;DR: In this article, an alternating direction method of multipliers combined with sequential convex approximations is proposed to solve the non-convex OPF problem, which is based on semidefinite programming relaxation.
Abstract: The optimal power-flow (OPF) problem, which plays a central role in operating electrical networks, is considered. The problem is nonconvex and is, in fact, NP hard. Therefore, designing efficient algorithms of practical relevance is crucial, though their global optimality is not guaranteed. Existing semidefinite programming relaxation-based approaches are restricted to OPF problems where zero duality holds. In this paper, an efficient novel method to address the general nonconvex OPF problem is investigated. The proposed method is based on an alternating direction method of multipliers combined with sequential convex approximations. The global OPF problem is decomposed into smaller problems associated with each bus of the network, the solutions of which are coordinated via a light communication protocol. Therefore, the proposed method is highly scalable. The convergence properties of the proposed algorithm are mathematically substantiated. Finally, the proposed algorithm is evaluated on a number of test examples, where the convergence properties of the proposed algorithm are numerically substantiated, and the performance is compared with a global optimal method.

Journal ArticleDOI
TL;DR: A resource-aware hybrid scheduling algorithm suitable for Heterogeneous Distributed Computing, especially for modern High-Performance Computing (HPC) systems in which applications are modeled with various requirements (both IO and computational intensive), with accent on data from multimedia applications.

Journal ArticleDOI
TL;DR: A novel coordinated power controller design framework is proposed to optimize the active power output of multiple generators in a distributed network and the distributed control and management strategies enhance the redundancy and the plug-and-play capability in microgrids.
Abstract: A novel coordinated power controller design framework is proposed to optimize the active power output of multiple generators in a distributed network. Each bus in the distributed generation systems includes two function modules: a distributed economic dispatch (DED) module and a cooperative control (CC) module. By virtue of the distributed consensus theory, a DED algorithm is proposed and utilized to calculate the optimal active power generation references for each generator. The CC module receives and tracks the active power generation references such that the generation–demand balance is guaranteed at minimum operating cost while satisfying all generation constraints. The distributed control and management strategies enhance the redundancy and the plug-and-play capability in microgrids. Optimal properties and convergence rates of the proposed distributed algorithms are strictly proved. Simulation studies further demonstrate the effectiveness of the proposed approach.

Journal ArticleDOI
TL;DR: Simulation results for simulations using real-data demonstrate the ability of the optimization framework to respond dynamically in real-time to external conditions while maintaining the functional requirements of all DERs.
Abstract: As we transition toward a power grid that is increasingly based on renewable resources like solar and wind, the intelligent control of distributed energy resources (DERs) including photovoltaic (PV) arrays, controllable loads, energy storage, and plug-in electric vehicles (EVs) will be critical to realizing a power grid that can handle both the variability and unpredictability of renewable energy sources as well as increasing system complexity. Realizing such a decentralized and dynamic infrastructure will require the ability to solve large scale problems in real-time with hundreds of thousands of DERs simultaneously online. Because of the scale of the optimization problem, we use an iterative distributed algorithm previously developed in our group to operate each DER independently and autonomously within this environment. The algorithm is deployed within a framework that allows the microgrid to dynamically adapt to changes in the operating environment. Specifically, we consider a commercial site equipped with on-site PV generation, partially curtailable load, EV charge stations and a battery electric storage unit. The site operates as a small microgrid that can participate in the wholesale market on the power grid. We report results for simulations using real-data that demonstrate the ability of the optimization framework to respond dynamically in real-time to external conditions while maintaining the functional requirements of all DERs.

Journal ArticleDOI
TL;DR: A detailed transient analysis of the learning behavior of multiagent networks reveals how combination policies influence the learning process of networked agents, and how these policies can steer the convergence point toward any of many possible Pareto optimal solutions.
Abstract: This paper carries out a detailed transient analysis of the learning behavior of multiagent networks, and reveals interesting results about the learning abilities of distributed strategies. Among other results, the analysis reveals how combination policies influence the learning process of networked agents, and how these policies can steer the convergence point toward any of many possible Pareto optimal solutions. The results also establish that the learning process of an adaptive network undergoes three (rather than two) well-defined stages of evolution with distinctive convergence rates during the first two stages, while attaining a finite mean-square-error level in the last stage. The analysis reveals what aspects of the network topology influence performance directly and suggests design procedures that can optimize performance by adjusting the relevant topology parameters. Interestingly, it is further shown that, in the adaptation regime, each agent in a sparsely connected network is able to achieve the same performance level as that of a centralized stochastic-gradient strategy even for left-stochastic combination strategies. These results lead to a deeper understanding and useful insights on the convergence behavior of coupled distributed learners. The results also lead to effective design mechanisms to help diffuse information more thoroughly over networks.

Journal ArticleDOI
TL;DR: In this article, the authors present a survey of distributed data aggregation algorithms, providing three main contributions: the concept of aggregation, characterizing the different types of aggregation functions, organizing the main aggregation techniques, and summarizing their principal characteristics.
Abstract: Distributed data aggregation is an important task, allowing the decentralized determination of meaningful global properties, which can then be used to direct the execution of other applications. The resulting values are derived by the distributed computation of functions like Count , Sum , and Average . Some application examples deal with the determination of the network size, total storage capacity, average load, majorities and many others. In the last decade, many different approaches have been proposed, with different trade-offs in terms of accuracy, reliability, message and time complexity. Due to the considerable amount and variety of aggregation algorithms, it can be difficult and time consuming to determine which techniques will be more appropriate to use in specific settings, justifying the existence of a survey to aid in this task. This work reviews the state of the art on distributed data aggregation algorithms, providing three main contributions. First, it formally defines the concept of aggregation, characterizing the different types of aggregation functions. Second, it succinctly describes the main aggregation techniques, organizing them in a taxonomy. Finally, it provides some guidelines toward the selection and use of the most relevant techniques, summarizing their principal characteristics.

Journal ArticleDOI
TL;DR: This paper presents the theory and implementation of a distributed algorithm for controlling differential power processing converters in photovoltaic (PV) applications that achieves true maximum power point tracking of series-connected PV submodules by relying only on local voltage measurements and neighbor-to-neighbor communication between the differential power converters.
Abstract: This paper presents the theory and implementation of a distributed algorithm for controlling differential power processing converters in photovoltaic (PV) applications. This distributed algorithm achieves true maximum power point tracking of series-connected PV submodules by relying only on local voltage measurements and neighbor-to-neighbor communication between the differential power converters. Compared to previous solutions, the proposed algorithm achieves reduced number of perturbations at each step and potentially faster tracking without adding extra hardware; all these features make this algorithm well-suited for long submodule strings. The formulation of the algorithm, discussion of its properties, as well as three case studies are presented. The performance of the distributed tracking algorithm has been verified via experiments, which yielded quantifiable improvements over other techniques that have been implemented in practice. Both simulations and hardware experiments have confirmed the effectiveness of the proposed distributed algorithm.

Journal ArticleDOI
TL;DR: Kia et al. as mentioned in this paper introduced a continuous-time dynamic average consensus algorithm for networks whose interaction is described by a strongly connected and weight-balanced directed graph, which allows agents to track the average of their dynamic inputs with some steady-state error whose size can be controlled using a design parameter.
Abstract: arXiv:1401.6463v1 [math.OC] 24 Jan 2014 Dynamic Average Consensus under Limited Control Authority and Privacy Requirements ∗ Solmaz S. Kia, Jorge Cort´es, Sonia Mart´inez Department of Mechanical and Aerospace Engineering, University of California San Diego, La Jolla, CA 92093, USA Abstract This paper introduces a novel continuous-time dynamic average consensus algorithm for networks whose interaction is described by a strongly connected and weight-balanced directed graph. The proposed distributed algorithm allows agents to track the average of their dynamic inputs with some steady-state error whose size can be controlled using a design parameter. This steady-state error vanishes for special classes of input signals. We analyze the asymptotic correctness of the algorithm under time-varying interaction topologies and characterize the requirements on the stepsize for discrete-time implementations. We show that our algorithm naturally preserves the privacy of the local input of each agent. Building on this analysis, we synthesize an extension of the algorithm that allows individual agents to control their own rate of convergence towards agreement and handle saturation bounds on the driving command. Finally, we show that the proposed extension additionally preserves the privacy of the transient response of the agreement states and the final agreement value from internal and external adversaries. Numerical examples illustrate the results. Keywords: dynamic average consensus; time-varying input signals; directed graphs; rate of con- ∗ under review in International Journal of Robust and Nonlinear Control

Journal ArticleDOI
TL;DR: Simulation results show that the proposed schemes can provide effective management for household electricity usage and bidirectional transactions.
Abstract: In this paper, the electricity cost minimization problem is considered for a residential microgrid which consists of multiple households (users) equipped with renewable-based distributed energy resource (DER). Each user has a set of nonshiftable and shiftable loads. Bidirectional electricity transactions are allowed, and a dynamic pricing model for the purchasing/selling of electricity from/to the grid is proposed. In order to reduce the electricity cost, the following decisions needed to be made: 1) scheduling decisions for the shiftable loads; 2) purchasing/selling decisions for each user at each time slot; and 3) amount decisions of the electricity purchased/sold by the users. An optimization problem to minimize the total electricity cost is formulated to obtain the optimal amount of electricity consumed, sold, and purchased for each user, respectively. A centralized algorithm based on dynamic programming, $Q$ -learning, and Lyapunov methods are proposed to solve the optimization problem with perfect information, with partial information, and without information of any time-varying parameters, respectively. For the latter two cases, distributed algorithms are designed for practical implementation. Simulation results show that the proposed schemes can provide effective management for household electricity usage and bidirectional transactions.

Journal ArticleDOI
TL;DR: A fully distributed bisection algorithm for the economic dispatch problem (EDP) in a smart grid scenario, with the goal to minimize the aggregated cost of a network of generators, which cooperatively furnish a given amount of power within their individual capacity constraints is presented.
Abstract: In this paper, we present a fully distributed bisection algorithm for the economic dispatch problem (EDP) in a smart grid scenario, with the goal to minimize the aggregated cost of a network of generators, which cooperatively furnish a given amount of power within their individual capacity constraints. Our distributed algorithm adopts the method of bisection, and is based on a consensus-like iterative method, with no need for a central decision maker or a leader node. Under strong connectivity conditions and allowance for local communications, we show that the iterative solution converges to the globally optimal solution. Furthermore, two stopping criteria are presented for the practical implementation of the proposed algorithm, for which sign consensus is defined. Finally, numerical simulations based on the IEEE 14-bus and 118-bus systems are given to illustrate the performance of the algorithm.

Journal ArticleDOI
TL;DR: Experimental results with comparisons show that the DAC- based learning algorithm performs favorably in terms of effectiveness, efficiency and computational complexity, followed by the ADMM-based learning algorithm with promising accuracy but higher computational burden.

Posted Content
TL;DR: This work proposes a distributed algorithm and establishes consistency, as well as a nonasymptotic, explicit, and geometric convergence rate for the concentration of the beliefs around the set of optimal hypotheses.
Abstract: We consider the problem of distributed learning, where a network of agents collectively aim to agree on a hypothesis that best explains a set of distributed observations of conditionally independent random processes. We propose a distributed algorithm and establish consistency, as well as a non-asymptotic, explicit and geometric convergence rate for the concentration of the beliefs around the set of optimal hypotheses. Additionally, if the agents interact over static networks, we provide an improved learning protocol with better scalability with respect to the number of nodes in the network.

Journal ArticleDOI
TL;DR: An effective online algorithm to solve the first- tier problem and prove its asymptotic optimality, as well as a distributed optimal algorithm for solving the second-tier problem are developed.
Abstract: Although considerable advances have been made in single microgrid (MG) systems, the problem of cooperation among MGs and the macrogrid has attracted considerable interest only recently. As in wireless communications systems, exploiting the temporal, spatial, and technological diversities in multiple cooperative MGs could bring about more efficient power generation and distribution. This paper investigates a hierarchical power scheduling approach to optimally manage power trading, storage, and distribution in a smart power grid with a macrogrid and cooperative MGs. We first formulate the problem as a convex optimization problem and then decompose it into a two-tier formulation. The first-tier problem jointly considers user utility, transmission cost, and grid load variance, while the second-tier problem minimizes the power generation and transmission cost, and exploits distributed storage in the MGs. We develop an effective online algorithm to solve the first-tier problem and prove its asymptotic optimality, as well as a distributed optimal algorithm for solving the second-tier problem. The proposed algorithms are evaluated with trace-driven simulations and are shown to outperform several existing schemes with considerable gains.

Journal ArticleDOI
TL;DR: A new constrained tensor factorization framework is proposed in this paper, building upon the Alternating Direction Method of Multipliers (ADMoM).
Abstract: Tensor factorization has proven useful in a wide range of applications, from sensor array processing to communications, speech and audio signal processing, and machine learning. With few recent exceptions, all tensor factorization algorithms were originally developed for centralized, in-memory computation on a single machine; and the few that break away from this mold do not easily incorporate practically important constraints, such as non-negativity. A new constrained tensor factorization framework is proposed in this paper, building upon the Alternating Direction Method of Multipliers (ADMoM). It is shown that this simplifies computations, bypassing the need to solve constrained optimization problems in each iteration; and it naturally leads to distributed algorithms suitable for parallel implementation. This opens the door for many emerging big data-enabled applications. The methodology is exemplified using non-negativity as a baseline constraint, but the proposed framework can incorporate many other types of constraints. Numerical experiments are encouraging, indicating that ADMoM-based non-negative tensor factorization (NTF) has high potential as an alternative to state-of-the-art approaches.

Proceedings ArticleDOI
24 Aug 2015
TL;DR: The concept of e-dominant dataset is defined, which is only a small data set and can represent the vast information carried by big sensory data with the information loss rate being less than e, where e can be arbitrarily small.
Abstract: The amount of sensory data manifests an explosive growth due to the increasing popularity of Wireless Sensor Networks. The scale of the sensory data in many applications has already exceeds several petabytes annually, which is beyond the computation and transmission capabilities of the conventional WSNs. On the other hand, the information carried by big sensory data has high redundancy because of strong correlation among sensory data. In this paper, we define the concept of e-dominant dataset, which is only a small data set and can represent the vast information carried by big sensory data with the information loss rate being less than e, where e can be arbitrarily small. We prove that drawing the minimum e-dominant dataset is polynomial time solvable and provide a centralized algorithm with 0(n3) time complexity. Furthermore, a distributed algorithm with constant complexity (O(l)) is also designed. It is shown that the result returned by the distributed algorithm can satisfy the e requirement with a near optimal size. Finally, the extensive real experiment results and simulation results are carried out. The results indicate that all the proposed algorithms have high performance in terms of accuracy and energy efficiency.

Journal ArticleDOI
TL;DR: Analytical and experimental results are provided that verify the effectiveness of the proposed architecture for generation control in islanded microgrids, and illustrate the performance of the aforementioned distributed algorithms under a variety of scenarios.
Abstract: In this paper, we propose a distributed architecture for generation control in islanded ac microgrids with both synchronous generators and inverter-interfaced power supplies. Although they are smaller and have lower ratings, the generation control objectives for an islanded microgrid are similar to those in large power systems, e.g., bulk power transmission networks; specifically, without violating limits on generator power output, frequency must be regulated and generation costs should be minimized. However, in large power systems, the implementation of the generation control functions is centralized, i.e., there is a computer that resides in a centralized location, e.g., a control center, with measurements and control signals telemetered between the generating units and the centrally located computer. The architecture for generation control that we propose in this paper does not rely on such a centrally located computer. Instead, the implementation of the control functions is distributed and relies on iterative algorithms that combine local measurements and certain information acquired from neighboring generating units with local, low-complexity computations. We provide analytical and experimental results that verify the effectiveness of the proposed architecture for generation control in islanded microgrids, and illustrate the performance of the aforementioned distributed algorithms under a variety of scenarios.