Self-Awareness in Self-Organising Institutions for Autonomic Power Systems Management
•
0 recomendaciones•271 vistas
Presentation by Sapere at the Awareness Inter-Project Day, Bologna, 2012
Recomendados
Recomendados
Más contenido relacionado
Más de FET AWARE project - Self Awareness in Autonomic Systems
Más de FET AWARE project - Self Awareness in Autonomic Systems (20)
Último
Último (20)
Self-Awareness in Self-Organising Institutions for Autonomic Power Systems Management
- 1. Self-Awareness in Self-Organising Institutions for Autonomic Power Systems Management Julia Schaumeier and Jeremy Pitt AWARE Inter-Project Day 23–24 January 2012
- 2. The Problem: conflicting goals Government Implement carbon emission regulations (i.e. by renewable energy sources) Consumers Power when needed and at an affordable cost Saving of energy only done when convenient Generators / Distributors Satisfy instantaneous demand (i.e. by excess assets) Obey regulatory targets by the government Make profit Julia Schaumeier and Jeremy Pitt Self-Awareness in Self-Organising Institutions 1/7
- 3. The Problem: conflicting goals macro-level vs. micro-level meeting global carbon emissions vs. comfort/utility of an individual Example: energy consumption during peak-hours peak demand to be met from conventional and renewable resources resources provided by consumers (i.e. ‘prosumers’) capacity of renewable resources unpredictable need: incentives for consumers to lower or ‘flatten’ their consumption Julia Schaumeier and Jeremy Pitt Self-Awareness in Self-Organising Institutions 2/7
- 4. Proposed Solution view APS as a Common Pool Resource with exogenous and endogenous resource provision regularisation of the system through institutions as described by Elinor Ostrom self-governing agents ensure responsible handling of the resource institutional rules: conventionally agreed by affected agents, mutually understood, monitored and enforced, mutable and nested within higher-order rules contribution / demand / allocation / appropriation → account for individual needs Julia Schaumeier and Jeremy Pitt Self-Awareness in Self-Organising Institutions 3/7
- 5. Proposed approach Sociologically-inspired computing Norm-governed system specification for Multiagent Systems Physical power, institutionalised power, and permission Obligations, and other complex normative relations Sanctions and penalties Roles and actions (communication language) Protocol stack: object-/meta-/meta-meta-/etc. level protocols Specification space (DoFs) Axiomatisation in the Event Calculus general purpose action language for representing events, and for reasoning about effects of events Julia Schaumeier and Jeremy Pitt Self-Awareness in Self-Organising Institutions 4/7
- 6. Awareness loop environment environment self-aware loop {specification agent {specification agent space} population space} population specification specification instance instance identify and react on nature of population manage times when there are not plenty of resources, avoid times of excess and identify times of crisis Julia Schaumeier and Jeremy Pitt Self-Awareness in Self-Organising Institutions 5/7
- 7. How can we ensure the agents act the way we would like them to? Appeal on or influence . . . . . . private Self-Awareness £ representation of self-aspects £ ˆˆˆ £ system’s macro-goal → agents’ private goal £ ˆ£ £ £ . . . persistent Self-Awareness £ ˆˆ £ £ ˆˆ£ reflection over time £ £ learn from previous experience £ˆˆˆ £ £ £ ˆ£ £ . . . predictive Self-Awareness £ˆˆˆ £ simulate future environmental states £ ˆ£ trade-off alternatives £ £ . . . meta Self-Awareness £ˆˆ £ £ ˆˆ£ £ incentivise change of (private) goal revise incentivising mechanisms Julia Schaumeier and Jeremy Pitt Self-Awareness in Self-Organising Institutions 6/7
- 8. Thank you! Julia Schaumeier and Jeremy Pitt Self-Awareness in Self-Organising Institutions 7/7