Keeping The Lights On

This workshop has been convened to consider, as its title says, the 'Implications of Decentralized Power Generation'. I can summarize the answer to the suggested question in a sentence. The implications of decentralized power generation are that electricity systems are going to change almost beyond recognition. However, whether they are going to do so smoothly and gradually, while keeping the lights on, is a lot less certain. We could be in for major trouble, even in places that now take electricity and electricity services for granted. Electricity research, both on technology and on institutions, has never before faced such a severe challenge. In my presentation this morning I'll try to explain why I think so, and why it matters. The starting point is what I call 'traditional electricity', the common technical model that was replicated all over the world during the past century. In this model, large central stations, remotely sited, generate electricity as alternating current and deliver it to users over networks including long high-voltage transmission lines. The traditional electricity system holds a monopoly, granted by government, in its franchise area. Within this area no one else is permitted to generate electricity for sale. The monopoly franchise guarantees a revenue stream, and allows system planners to make very large long-term investments, notably in power stations and networks, because the captive customers of the monopoly bear all the risks. The cost of capital for investment is thus low; and since the capital investment makes up a substantial part of the cost of the electricity, the cost of electricity is likewise low. Throughout the past century this traditional model was remarkably successful, making electric light, electric motive power and other electricity services part of the everyday fabric of industrial society. But the traditional model of electricity arose for one key reason: the economies of scale of the generating technologies then available, particularly steam turbines and water turbines. Making these generators larger steadily lowered the cost of the electricity they produced, up to generators of enormous size, even when the cost of the accompanying delivery network was deducted. The technical configuration of the traditional electricity system, with centralized production and a network for delivery, was modeled on that for town gas, for similar financial reasons; and the analogy was effective and useful for most of a century. Now, however, it is seriously misleading. The emergence of small-scale decentralized generation as an economic option invalidates the key premise underlying the traditional model of electricity. Moreover the traditional model has failed to deliver electricity services to some two billion people - one-third of humanity; and its key technologies - large dams, large coal-fired and nuclear steam-cycle stations, and overhead transmission lines - all face problems, particularly financial and environmental, that may become insuperable. For all these reasons, including decentralized power generation, the