Hanno Hildmann

Influence of variable supply and load flexibility on Demand-Side Management

Demand-Side Management (DSM) refers to the shaping of the electrical load to improve the fit with the supply side [1]. One way to achieve this is to confront the user with a dynamic financial incentive to influence the demand. In previous work (see [2]) a crude reactive pricing algorithm has been shown to shave the peak of the aggregated load by over 25%. In this paper this approach is extended and evaluated in more detail. The algorithm is extended to cover scenarios where the supply is no longer fixed; its performance is investigated under varying conditions such as the percentage of the flexible load as well as the extent of the loads flexibility. The expected benefit of applying the algorithm to real world scenarios is predicted, allowing for an informed decision on whether or not to employ it in specific situations and under which pricing conditions.

A Mobile Phone Based Virtual Pet to Teach Social Norms and Behaviour to Children

The paper presents a Tamagotchi-like mobile phone game that uses an artificial neural network driven character model and is designed to teach positive moral values to children. The behavioral model is explained and the approach is supported by a proof of concept implementation. Our results suggest the feasibility of the approach.

A Bio-Inspired Resource-Saving Approach to Dynamic Client-Server Association

This novel method for client-server association draws inspiration from the nest-building behavior of social insects. It focuses on reducing active, and thus resource-consuming, servers.

Modeling quality of service vs. Peak Reduction trade-offs in A/C-based Demand-Side Management

A large number of A/C units potentially represents a substantial flexible load susceptible to be leveraged for Demand-Side Management (DSM) purposes. However, in such a scenario, a drop in quality of service (QoS), i.e. an uncomfortable increase in temperature, is to be expected as some cooling-down is delayed to achieve peak reduction. We present the results of a simulation study designed to test a decentralised control framework based on probabilistic activation of individual A/C units and reactive pricing [1]. Performance is evaluated and discussed, primarily in terms of the trade-off between QoS and DSM efficiency.

Ant-Colony Based Heuristics to Minimize Power and Delay in the Internet

In networks like the Internet, it is desirable to reduce both power consumption and delays. In this paper, we propose an ant-colony heuristic algorithm which aims to minimize the product of the following two measures: consumed-power to available-bandwidth (PWR) and delay. Our current study is conducted using simulations on a topology which represents a 2-dimensional torus. We also compare the ant-colony based method with the well known Dijkstras shortest path algorithm. As compared to Dijkstras algorithm, we observe that the ant-colony heuristic can be computationally less intensive in obtaining a near optimal solution. This could be especially useful in large scale systems. As the PWR metric can be used across all levels of the Internets routing hierarchy, thus methods like ant-colony algorithms could lead to a reduction in both power consumption and delays.

A Mobile Device Based Multi-agent System for Structural Optimum Design Applications

This paper will discuss a design, implementation and evaluation of a multiagent system (MAS) for a structural optimum design application (SODA) on mobile devices. The paper commences by defining the relevant concepts of SODAs and MAS. It will then present an architecture and framework combining both. It will explore in detail an implementation running on a mobile device and compare that to a version on a standard PC. The computational evaluation results obtained from a simulation on an actual mobile device indicate that running a complex system on such a device is feasible and usable. Finally, the paper suggests that current mobile devices can be used as more than data displays, and their processing power can be efficiently utilised.

Resource Allocation and Scheduling based on Emergent behaviours in Multi-Agent Scenarios

We present our observations regarding the emergent behaviour in a population of agents following a recently presented nature inspired resource allocation / scheduling method. By having agents distribute tasks among themselves based on their local view of the problem, we successfully balance the work across agents, while remaining flexible to adapt to dynamic scenarios where tasks are added, removed or modified. We explain the approach and within it the mechanisms that give rise to the emergent behaviour; we discuss the model used for the simulations, outline the algorithm and provide results illustrating the performance of the method.

A mobile device based, and multi-agent driven structural design solution

This paper will discuss a design and implementation of a multi - agent system (MAS) for a structural optimum design application (SODA) on mobile devices. The paper commences by defining the relevant concepts of SODAs and MAS as well as the problem. It explores an implementation running on a mobile device and compare it to a version on a standard PC. The results from both computational and experimental evaluation, obtained from a simulation and from an actual mobile device, indicate that running a complex system on such a device is feasible and usable. Finally, the paper suggests that mobile devices can be used as more than data displays, and their processing power can be efficiently utilised.

A self-organizing client / server allocation algorithm for applications with non-linear cost functions

We subject a previously presented client-server allocation algorithm to a revised performance criterion to account for non-linear cost functions. We identify the conditions under which the original approach is not guaranteed to outperform the benchmark approach and propose a revised cost prediction method to improve performance for those scenarios. We present empirical results to show the measured benefit of the revised approach over the benchmark as well as the original work.

Adaptive scheduling in dynamic environments

We present a method for fair agent scheduling in transportation scenarios. The approach is designed to first ensure the scheduling of all required task locations and, once this is achieved, focus on a balancing the workload across the population of transportation units. This, while almost certainly sub-optimal in the context of efficiency, facilitates the speedy allocation of new geographically located tasks due to the distribution of the remaining capacity across the agent population. We discuss our method, present results from simulations and discuss the advantages and disadvantages of the approach.