Article PostSign in to Vote 0
Future Power Systems - what will they look like and the 'Smart Enterprise'
- Posted on May 25, 2010
- 72 views
The Power System of the Future will be radically different from today, with high penetration of unpredictable variable output renewables. We have a lot of potential new technology and control mechanisms to utilise.
We need to determine what combination of Generation (Big+Little), Storage, Distributed Resources management, Interconnection trading, Ancillary services provision and Transmission/Distribution management will deliver the goods as regards safe, secure, efficient, and economic operation with reduced fossil fuel burn. All parts of the electricity business are involved; generation, supply, transmission, distribution, system operation and the market, together with the customer, in a commercial/technical framework in which the target can be hit while maintaining system security, quality of supply and accuracy of settlement.
Better commercial and technical interfaces are required between member states and Electricity Pools and across Interconnectors. This all forms the 'Smart Enterprise', incorporating Smart Grid and Smart Meter initiatives.
Future Power - Active Retail markets and Customer Engagement.
The main thing the customer, the industry and the community wants to see out of the Smart Enterprise is 'value'; to put together the infrastructure from two way premises communications device (meter+) through aggregator/disseminator mechanisms to supplier trading/back office and distribution/transmission/system operator systems is quite a large and expensive undertaking.
In summary, we are making the biggest change to the retail electricity market 'philosophy' in nearly 130years of electricity supply. Goodbye to 'pay me 10p/10cts/10Eurcents/1Rupee.... per unit and you can use what you like when you like'. In the new world, time and day both matter.
There are a number of important points to note in the customer engagement debate.
1) The change from fixed preset single or two rate daily tariff to ToU has a major impact. Try going in stages via more preset rates (time of day, day of week, season etc).Future Power and other Energy sector Technology - What's it worth? The main issue is the need to evaluate the various scenarios, for each nation, union and system correctly.
2) Put up a safety net at each progression with the old and next tariff in parallel. Allow the customer to pay the cheaper at each billing point but try and help them see where behavioural change would make the new tariff more advantageous.
3) Predictive ToU, being sequences of firm then non-firm prices updated at regular intervals, would seem to be more effective in getting useful Customer reaction. This however has a big impact on Meter design and upstream processing. Make the meter (Customer Interface) unit flexible as regards data content.
4) Get the data together for customer reaction vs price (Site Import/export change) from the interface; quite a complex function which varies depending on preceding profiles of price level and activity. Use the Smart interface to get intelligence on appliance action.
5) Don't try using raw marginal prices (10x at peak etc) as a tariff signal to a large part of the customer demand base or you will throw the load curve all over the place. That will result in inefficient operation of generation. Set the price to get the customer reaction to remove just the high price generation.
6) What we are aiming to do is to 'flatten the fossils'; i.e. reduce running periods for this plant but also make sure that the remaining fossil plant doesn't need to start too often, runs up and down at max rate and is then loaded flat out when on the bars. That requirement shape will not necessarily follow the demand curve, especially where large amounts of variable output renewables (e.g. Big Wind) are installed.
7) All the Customer controls need to be automatic; we need good HAN, BMIS and Industrial Control mechanisms with suitable AI to help the customer set up the strategy logic.
8) There is also of course the concept of supplier or operator trading kW 'blocks' across time with the customer, where the resultant customer action can be proved and therefore charged/credited.
9) The intelligent Customer-Utility Interface (CUI - 'meter' is too restrictive a term), needs to be a flexible unit within a flexible Standards 'framework' for data content. This will ensure that new data structures can be incorporated as we develop the methodologies for customer participation.
10) One size will most definitely not fit all. Different solutions will be appropriate on different Power systems and for different customer groups.
11) Don't try making too many changes at once. If, for instance, you hike electricity prices while installing smart meters, the customer will target the new meter as the reason for any bill increases whereas it is in fact the price hike which is responsible,
For the Power System we need full nested time series simulations (Commitment-Schedule-Dispatch-Outturn), with Fuel supply allocation and emissions calculations and changing data forecasts (especially for variable, unpredictable renewables) applied in sequence.
We also need to run the full simulation of Heating and Cooling provision, again back to the fuel inputs and with emissions calculations, with the various proposed changes to energy sources (boilers vs heat pumps vs CHP etc) covered.
To cover the proposed large scale introduction of Electric vehicles, we need to also have the full Vehicle Internal Combustion model (Well to Exhaust), again to produce fuel burn and emissions data.
Obviously, each scenario will represent different data exchanges between the three main energy sector simulations.
This whole exercise will determine the value of each scenario and thus allow the correct judgements to be made as regards practicability, impact and cost. This exercise will determine the value of each scenario and thus allow the correct judgements to be made as regards practicability, impact and cost.
Each major Energy system will of course have its own characteristics and modelling requirements.
Brave New World; here we come. Let's make sure we put together the right solution.