A Demand Responsive Customized Price Function for Real-Time Pricing – The Need for Pushing the Boundaries of Grid Modernization.
image credit: © Adonis1969 | Dreamstime.com
- Jun 19, 2020 8:45 pm GMTJun 19, 2020 8:45 pm GMT
- 797 views
This item is part of the Special Issue - 2020-06 - Grid Modernization, click here for more
The modernization of the power grid inevitably needs to incorporate the best of the information transfer network in it. This will help in the enhancement of efficiency of the modern grid. One of the oldest problems of the electricity industry is the demand-supply mismatch which creates peaks and troughs in the load profile. One of the effective ways of addressing this problem is through dynamic pricing of electricity. Dynamic pricing is applied in a large number of power utilities worldwide. However, almost all of these implementations are characterized by low level of dynamism in the pricing, namely, block pricing, peak pricing or time-of-day pricing (Spiller, 2020). There is an urgent need of extensively implementing real-time pricing in electricity (Zethmayr & Kolata, 2017), which will enhance the consumers’ responsiveness to the price change and lead to load flattening. This will require huge data transfers in very short intervals of time and hence a very strong data transfer network infrastructure must run parallel to the power grid.
Let me present an idea that needs to be implemented and tested. A demand responsive price function is one where the per unit price of electricity is proportional to the demand. That means when the demand goes up, so does the price and vice versa. How will it impact the consumption? If the price goes up due to a forecasted high demand, the actual consumption at that high price is expected to go down in order to reduce the overall expenditure of the consumer. Similarly, if the price is low, there can be an increase in consumption (psu, 2020). Thus, peaks will reduce and troughs will increase to develop a more flattened load profile. If the demand is forecasted for shorter intervals of time, say fifteen minutes, the response is supposed to be better. This is because, the demand profile can be modified better with a demand response in every fifteen minutes than that in every one hour, since the opportunity of demand modification increases from one time in an hour to four times in an hour. This shortening of the time slots for price change will provide a more effective real-time pricing scheme. Now think of a situation when this demand forecasting is done for every individual consumer. There will be a demand forecast for every fifteen minutes and a set of ninety-six price points per day for each consumer to enable customization of the pricing. Well, that is a huge amount of data that will need to be transferred and processed very quickly. What will this customization look like? At any given timeslot, different consumers will pay different prices. One who is consuming more will pay a higher price while one who is consuming less will pay a lower price at the same time. This is unlike most present-day dynamic pricing schemes where the price changes are based on grid demand alone and not on individual consumers. In those cases, during a high-priced period every consumer has to pay a high price irrespective of the level of his/her individual consumption.
All these above features can be translated into a demand responsive customized price function as shown in the diagram alongside. Such a price function has to be made for each consumer every day. For doing it, the pricing authority will require an estimate of the consumption that each consumer plans to have in the next day. Hence a smart meter needs to be at every consumer’s place to register the number of appliances that a consumer is planning to use in the next day. It also needs to collect the data on the power ratings, the time span of operation and the earliest possible start time and the latest possible stop time of each appliance. This can provide the estimated consumption by a consumer in each timeslot. A suitable optimization model can use this data and develop a price function as shown in the diagram above for each consumer. The convexity of the curve needs to be determined by the supplier based on several factors. The smart meter in each consumers place should receive the customized price function thus created and schedule the operation of the appliances to flatten the individual consumer’s load profile over the day. With such a pricing function, the only way to minimize the consumer’s expenditure is to flatten his/her individual load curve (Mitra & Dutta, 2018). A flat price would also have brought in the load flattening effect but higher peak prices as obtained from such non-linear functions are more likely to impact a demand response.
The mere scale of information that needs to be handled in such a situation demands a highly efficient data transfer network in the grid. The computation power that needs to be attached along with it will enable the processing and calculation of the data and the optimization of the results. Moreover, an enormous storage facility will be necessary to store the information for future reference. While we think of pushing the boundaries of the power grid through its modernization, how feasible will it be to implement such a huge infrastructure and how long will it take to do so? While some utilities have already started real-time pricing implementations but customizing the prices seems difficult (comed, 2020). I have tried to work on this idea but need to know if any automated customized pricing scheme like this has been implemented anywhere. There are challenges in terms of policy as well as consumer behaviour apart from the technological challenges that still keep real-time pricing from extensive implementation (Allcott, 2012). Customized pricing like this can flatten the individual load profile of each consumer which in turn can be expected to flatten the overall grid load profile. Distributed generation and the evolution of prosumers will further enhance the data capacity requirement of the modern grid. Are we still technologically capable enough to push the boundaries of the grid modernization?
Allcott, H. (2012). Real-Time Pricing and Electricity Market Design [Ebook]. Retrieved 16 June 2020, from https://simsee.org/simsee/biblioteca/MercadoSpot/Allcott%202012%20-%20Re....
comed. (2020). Live Prices | ComEd's Hourly Pricing Program. Hourlypricing.comed.com. Retrieved 16 June 2020, from https://hourlypricing.comed.com/live-prices/.
Mitra, K., & Dutta, G. (2018). A two-part dynamic pricing policy for household electricity consumption scheduling with minimized expenditure. International Journal Of Electrical Power & Energy Systems, 100, 29-41. https://doi.org/10.1016/j.ijepes.2018.01.028
psu. (2020). Real-time Pricing for Electricity. Penn State Extension. Retrieved 16 June 2020, from https://extension.psu.edu/real-time-pricing-for-electricity.
Spiller, B. (2020). All Electricity is Not Priced Equally: Time-Variant Pricing 101. Energy Exchange. Retrieved 16 June 2020, from http://blogs.edf.org/energyexchange/2015/01/27/all-electricity-is-not-pr....
Zethmayr, J., & Kolata, D. (2017). The costs and benefits of real-time pricing. The Citizens Utility Board, Environmental Defense Fund. Retrieved 16 June 2020, from https://citizensutilityboard.org/wp-content/uploads/2017/11/20171114_Fin....