Search

This paper considers a class of multi-agent distributed convex optimization with a common set of constraints and provides several continuous-time neurodynamic approaches. In problem transformation, l 1 and l 2 penalty methods are used respectively to cast the linear consensus constraint into the objective function, which avoids introducing auxiliary variables and only involves information exchange among primal variables in the process of solving the problem. For nonsmooth cost functions, two differential inclusions with projection operator are proposed. Without convexity of the differential inclusions, the asymptotic behavior and convergence properties are explored. For smooth cost functions, by harnessing the smoothness of l 2 penalty function, finite- and fixed-time convergent algorithms are provided via a specifically designed average consensus estimator. Finally, several numerical examples in the multi-agent simulation environment are conducted to illustrate the effectiveness of the proposed neurodynamic approaches. [ABSTRACT FROM AUTHOR]

In order to solve the problem that it is difficult to get out of the local optimum to achieve further optimization in large-scale dense Distributed Constraint Optimization Problems (DCOPs). This paper proposed a local parallel search framework (LPOS) for solving algorithms of DCOPs. In LPOS, the Agent searches all the value assignments of its local neighbors in parallel according to its current value assignment to further expand the search of solution space, so as to avoid the algorithm falling into local optimum prematurely. In order to ensure the convergence of the algorithm, this paper designed an adaptive equilibrium factor K to balance the exploration and exploitation ability of the DCOPs solving algorithms. At the theoretical analysis, it proved that the LPOS can expand the search of solution space and the K can achieve the balance. The experimental results demonstrate that the performance of the proposed LPOS is better than the-state-of-the-art algorithms in both low density and high density. Specifically, LPOS significantly superior to the competitors when solving high density DCOPs. [ABSTRACT FROM AUTHOR]

The on-demand transport (ODT) systems have developed worldwide as they have significant social, environmental, and economic benefits. Even with those benefits, it's still important to gain popular acceptance. The acceptance key is the reactivity of the system in providing fast and reliable solutions whilst respecting vehicles' and clients' constraints. This paper presents a decentralized multi-agent approach to model and solve the ODT problem in a static road network. The agents interact with each other using the A* algorithm to find an optimal solution for each transport demand. The optimal solution is expressed by the fastest trajectory taken by the cheapest vehicles. We utilize factual data from a Lebanese city to do experiments evaluating the proposed approach. [ABSTRACT FROM AUTHOR]

The consensus control method based on a multi-agent system has been widely applied in the distributed control and optimization of microgrids. However, the following drawbacks are still common in current research: (1) ignoring the influence of consensus control commands on the synchronization stability of the physical grid under primary control; (2) only focusing on improving one property ofcontrol performance, lacking comprehensive considerations of multiple properties. With the aim of solving these problems, in this paper we propose a weight-adaptive robust control strategy for implementing distributed frequency regulation of islanded microgrids. Firstly, the frequency synchronization stability of the physical layer is analyzed by means of a coupled oscillator theory and the design objectives of the controllable parameters for the information layer are formed. Subsequently, the relationship between the weight coefficients and the two important control performances of convergence speed and delay robustness is strictly analyzed. Based on this, an adaptive coefficient that can be autonomously adjusted according to the frequency deviation is designed to achieve a trade-off between convergence speed and delay robustness. Finally, three simulation studies are presented to verify the effectiveness of the proposed control strategy. [ABSTRACT FROM AUTHOR]

• The data-driven inverse optimization corrects the DR execution online. • The proposed strategy is realized in a distributed manner and free of initialization. • This paper provides rigorous stability proof for the proposed optimization. • D-PPDS algorithm solves the DR optimization with global inequality constrain. This paper provides an insight into the demand response (DR) optimization in distribution markets consisting of a retailer and multiple demand response aggregators (DRA), where a retailer determines DR incentives based on power consumption profile. Conventional DR optimization with global states and constraints is intractable to be implemented in a distributed framework, which restricts the application feasibility and the potential profit of DR. To handle these limitations, we design a multi-agent architecture for distributed demand response (DDR). An online data-mining method is developed to identify the characteristics of DR. A leader–follower structure decomposes the original problem into a leader problem with global variables and aggregators of sub-problems, where discrete singular consensus is designed to broadcast the leader's strategy to followers in real-time. The distributed perturbation primal–dual sub-gradient (D-PPDS) algorithm is proposed to solve the DDR problem with global inequality constraints in a completely distributed fashion. The proposed DDR strategy is tested by an actual case. The simulation results demonstrate that the asynchronous D-PPDS algorithm can obtain the near-optimal solution of the problem with global inequality constraints, and is robust against delay or plug-and-play. [ABSTRACT FROM AUTHOR]

To achieve large-scale generalized demand-side resources to participate in the "peak-shaving and valley-filling" of the power system, a distributed cooperative control strategy for discrete-time linear multi-agents considering historical data is proposed, which comprehensive consider the changes in load commands, topology, multi-objective function coefficients, communication delays and input delays. This paper combines the distributed multi-agent system with the hierarchical architecture to solve the problems of extensive data calculations in centralized control, which may easily cause communication channel blockage and delay. Considering the cost of changing the generalized demand-side resources and the user's comfort, the pre-learning historical data storage strategy based on K-means clustering is adopted before the actual task, and the idea of transfer learning is introduced into the consensus algorithm. Finally, through experimental statistics, the number of iteration steps of the consensus algorithm is reduced and the effectiveness of the proposed algorithm is proved. [ABSTRACT FROM AUTHOR]

• Defining a distributed multi-agent approach to solve power system restoration problem. • Bus agents achieve the general agreement on load feeding sequence by using Dynamic Programming. • Considering uncertainty in the wind and photovoltaic sources by Monte Carlo simulation. • Demonstrating the outperformance of the multi-agent approach rather than the top-down method. In this paper, a decentralized multi-agent system (MAS) has been proposed to solve the power system restoration problem. In the proposed MAS, an agent with its own specific logic and interactions with other agents is devoted to any piece of equipment in the grid, including bus, black start, non-black start, photovoltaic and wind generating units. Power system restoration is devised as a single objective problem to minimize the energy not supplied (ENS), which is solved by bus agents using dynamic programming. The uncertainty of the wind and photovoltaic sources is considered in the corresponding agents, which is dealed by the Monte Carlo method. In addition, not only the genetic algorithm but also dynamic programming are employed in a top-down approach to solve the problem. The proposed algorithms are applied successfully to the IEEE 39-bus system. Comparing the results of both MAS and top-down approaches demonstrate that the proposed MAS outperforms the centralized method either optimized by genetic algorithm or dynamic programming in the sense of ENS. [ABSTRACT FROM AUTHOR]