Introduction:

Objective

The CRISP vision is that recent progress in intelligent ICTs offers many novel opportunities to build new operating strategies for high-DG power networks:

• In parallel to the power grid, an extensive data-communication infrastructure emerges through various (and usually hybrid) means including glass-fibre, telephone copper, wireless or the power line itself. This communication infrastructure will cover different system levels of the power grid, even reaching the very fine-meshed level of individual customer equipment.
• Embedded real-time communication and computing processors enable to create smart equipment with “local intelligence” at increasingly affordable prices.
• New software technologies for building distributed online at-a-distance applications have become available, including intelligent communicating agents, electronic market algorithms, and other mechanisms for distributed information systems enabling forms of “global intelligence”. Worldwide standards help to achieve this in a relatively cheap way (e.g. IP).
• System integration using ICT creates vulnerabilities not only in the communication infrastructures but also pose new security requirements on the equipment itself preventing non-authorised back-door access to systems and services. Protection of information from intrusions aiming at disclosure or manipulation is also necessary to create trustworthy operations. However, safeguarding the systems must be motivated from a cost-benefit analysis.

The CRISP project view is that the potential stemming from the mentioned rapid advances in industrial ICT is however not yet fully known, explored and exploited by the utility industry. Effectively, there is a lot of mileage to be gained by combining the capabilities delivered by two different critical infrastructures: the power grid and intelligent ICT networked systems. The roles in operating a power grid in a liberalised energy markets are changing. Stakeholders in trade and distribution take over tasks from each other. In the utility world the concepts of “distributed utilities” and “digital utilities” are attracting strong interest. In the first concept, local generation and storage capacity, at the end of the power distribution hierarchy is utilised in a locally optimised fashion. In the second concept, intensive use is made of information and communication from distributed sources relevant to the utility's business processes. The CRISP project integrates these concepts by developing and demonstrating new operating strategies based on ICT intelligence for high-DG power networks that are “distributed + digital + renewable”.

Within the context of a fine-meshed generation and distribution grid, an extra level of communication and information exchange on the existing electricity distribution infrastructure will emerge. Adding intelligent communication and processing solutions to micro-generation and the lower levels of distribution generation will help to consolidate the power infrastructure without over-dimensioned transport infrastructures to background capacities. From the power consumption side, added intelligence communication possibilities in devices will make them responsive to real-time control and pricing signals and generate a competitive advantage. The predicted or historically measured consumption profile will gain in value as will a future generation profile with a certain time-horizon. Part of power stability will be obtained by balancing flexible consumption and demand patterns. This levelling and scheduling will have to take place on several timescales, from the millisecond to the hourly scale. Through distributed intelligence, load- and demand-side management will be shifted to a lower, more fine-grained local level and market-oriented supply-side management is also made possible. Research of the role and impact this has for various specific European situations is needed. Thus, these developments provide exciting possibilities to invent new distributed configurations of power, communication and computer networks such that they are more robust in the face of disturbances, and are controlled and protected for optimum security/safety, economy, performance and reliability.

The CRISP project will study, develop, and test new operating strategies for high-DG power networks enabled by the mentioned progress in intelligent ICT technologies, focused on a specific set of relevant practical situations, issues and scenarios such as:

• Market-oriented demand-supply matching to increase the cost-effectiveness of DG network operations, integrating demand-side and supply-side management through online agent-mediated e-markets that exploit real-time price/cost information as well as knowledge of load/supply needs and profiles.
• Online distant supervision, fault detection and recovery of distributed generation networks, ensuring the robustness and performance of the system and including the issue of how “self-healing” of the network is possible under major disturbances and blackouts through automatic reconfiguration by means of distributed intelligence.
• Intelligent load shedding in emergency situations such that cost-controlled and prioritised load shedding is done in much more fine-grained ways than currently possible.
• Multi-agent systems based methods for online integration and control strategies for RES supply, demand and storage that achieve bottom-up rather than the usual hierarchical top-down optimisation.
• Architectures and requirements for distributed intelligence in highly distributed power networks that recognize and cope with the fact that such networks contain different decision-making levels of granularity (end-customer devices, low-voltage last mile access and distribution, middle-voltage grid operations) and so are inherently to be treated as networks of networks themselves.
• Intelligent decision support for risk and security analysis of distributed generation networks that not only includes maximum availability considerations but also integrates economic cost-benefit and priority analysis on an equal footing. In effect, we will develop models that recast traditional vulnerability and risk analysis methodologies into terms of investments/insurances to protect value-chains and business processes. This analysis will also influence policies for graceful degradation of services due to a class of incident/threat scenarios.

Contribution to EU-EESD programme/key action objectives

• Medium to long term Target Action I: Integration of RES and distributed generation in energy supply systems, in particular to
  • KA 6.2.1 Assuring electric power flow reliability and stability and increasing power line efficiency, and
  • KA 5.3.2 Integrating renewable energy sources into the grid and standalone systems.
• Strategic Priority ii: Exploiting the potential of new ICTs in energy RTD including e-science issues, in particular to
  • Transfer the latest advances in communicating agent and e-market software technologies, stemming from the EU-IST (Information Society Technologies) programme, to EESD.
  • Implementing cost/benefit models of system safeguarding.

A major innovative way to achieve these key action objectives by the CRISP project is to exploit the latest results of new ICTs and e-science (Strategic priority ii) that are provided by recent EU-IST (Information Society Technologies) projects. CRISP is in the unique position to do this because various consortium partners are involved in relevant IST projects and networks. We mention a few in particular:

• PALAS (IST-1999-11379): telecoms and energy services, business models, and technologies over the power line viewed as an alternative last-mile access infrastructure.
• OnToKnowledge (IST-1999-10132): new infrastructure, tools and standards for the next, intelligent generation of the World-Wide Web (called the Semantic Web, see also the article by Tim Berners-Lee, director of the W3C consortium, in Scientific American, May 2001)
• OBELIX (IST-2001-33144): intelligent Internet/Web tools for advanced e-business scenarios and dynamic value constellations, including energy applications such as large-scale demand-side power bidding in dynamic marketplaces and smart homes.
• The OntoWeb and AgentLink networks of excellence: the two IST networks central to intelligent Web and agent software.

Innovation

1. It will provide novel strategies and solutions for both normal and emergency operations in highly distributed power networks with high DG penetration degrees, exploiting the latest results of intelligent ICTs.
2. It will greatly expand the insight into the performance, safety and security in the architecture of such highly distributed power networks, through scenario analyses, simulations and field tests, in a way that not purely looks at technical availability and functionality, but integrates economic cost-benefit considerations as well.
3. It will, through its development and industrial test work, advance the state of the art of intelligent ICT methods and how they yield novel solutions for utility applications in Europe.
4. It will facilitate the efficient control of energy and heat storage buffers and of partly unpredictable output renewable energy sources. It will also make a contribution to transforming residential areas into energy farms.

A key innovation idea of the CRISP project is to combine advances in several different areas of ICT for the benefit of novel power applications. Progress in communication technology provides the opportunities for almost universal online connectivity needed for real-time distributed systems and networks. Continuing advances in embedded hardware technology have created the possibilities for smart equipment with embedded Internet facilities enabling local forms of intelligence. Ongoing innovations in intelligent software technology such as agents, knowledge systems and electronic markets, take advantage of both connectivity and local intelligence, thereby enabling global intelligence through online optimisation and adaptation to changed circumstances across networked systems. CRISP brings these combined advances in ICT intelligence into the realm of distributed power applications.

¹For these technologies and how they might be used for power applications, we refer the reader to some recent overviews in top computer science journals: “Agents with Power” in Communications of the ACM, March 1999; IEEE Internet Computing, Special issue on Virtual Marketplaces, April 2001; IEEE Intelligent Systems, Special Issue on Intelligent E-business, August 2001.