Modeling the Impacts of Electricity Tariffs on Plug-in Hybrid Electric vehicle Charging, Costs, and Emmissions

In the May-June, 2012 issue of Operations Research Professor Ramteen Sioshansi writes about Plug-In Hybrid Electric Vehicles or PHEVs (full paper available here and electronic companion here). PHEV is a technology that has a potential to revolutionize how we power transportation and the impact of personal transportation on the environment. Gasoline powered cars have a well established distribution network for the fuel they need. PHEVs will draw energy from the same electric power distribution system (or grid) that we use for all other electric power needs. In his paper Sioshansi investigates different strategies for managing the impact of PHEV charging on the power grid. In his abstract he states:

"

Plug-in hybrid electric vehicles (PHEVs) have been touted as a transportation technology with lower fuel costs and emissions impacts than other vehicle types. Most analyses of PHEVs assume that the power system operator can either directly or indirectly control PHEV charging to coordinate it with power system operations. This paper examines the incentives of individual drivers making charging decisions with different electricity tariffs, and compares the cost and emissions impacts of these charging patterns to the ideal case of charging controlled by the system operator. Our results show that real-time pricing performs worst among all of the tariffs we consider, since linear prices are inherently limited in signaling efficient use of resources in a system with non-convexities. We also show that controlling overnight PHEV charging is significantly more important than limiting midday vehicle charging.

"
Ramteen Sioshansi

Invited Comments

The editors have invited comments on this work from several experts.

Dr. Eugene Litinov, ISO New England (full comments pdf LitinovComments ): Dr. Eugene Litvinov is a senior director of Business Architecture and Technology at the ISO New England. He is responsible for advanced System and Markets solutions, Smart Grid and Technology strategy and is a lead of the Research and Development activities in the organization. He has 40 years of professional experience in the area of Power System modeling, analysis and operation; Electricity Markets design, implementation and operation; Information Technology. Dr. Litvinov has extensive expertise in management and technical leadership of large engineering and information technology projects; development of computational methods in power system analysis and operation; market clearing engines, electricity pricing; settlements.

Dr. Litvinov holds BS, MS and PhD in Electrical Engineering. He is an editor of the IEEE Transactions on Power Systems.

Dr. Antonio J. Conejo, Universidad de Castilla, La Mancha (full comments pdf ConejoComments ): Dr. Conejo is a Full Professor of Electrical Engineering at the Universidad de Castilla-La Mancha, Spain. He received the M.S. degree from the Massachusetts Institute of Technology and the Ph.D. degree from the Royal Institute of Technology, Stockholm, Sweden. He has published over 125 papers in SCI journals and is the author or coauthor of books published by Springer, John Wiley, McGraw-Hill and CRC. He has been the principal investigator of many research projects financed by public agencies and the power industry and has supervised 16 PhD theses. He is the Editor-in-Chief of the IEEE Transactions on Power Systems and an IEEE Fellow.

Dr. Derek Lemoine, University of Arizona (full comments pdf LemoineComments ): Dr. Lemoine is an assistant professor in the Department of Economics at the University of Arizona. He holds a Ph.D. in Energy and Resources (2011) and an M.A. in Economics (2010) from UC Berkeley. His interdisciplinary dissertation research was motivated by the uncertainty about the magnitude of temperature change produced by a given greenhouse gas emission path and about the technological change induced by a given policy path. This uncertainty complicates estimation of the benefits and costs of reducing emissions. He explored how the presence of this uncertainty could affect both the target and the form of climate policy. Other research has addressed the transition to electrified vehicles and low-carbon transportation systems, and he has published in journals specializing in environmental and energy economics, climate and atmospheric sciences, environmental science, energy technologies, and forestry.

Dr. Vineet Goyal, Columbia University (full comments pdf GoyalComments ): Dr. Goyal is an Assistant Professor in the Industrial Engineering and Operations Research department at Columbia University. He received his PhD in Algorithms, Combinatorics and Optimization (ACO) in 2008 from Tepper School of Business, CMU. Before joining Columbia,he spent a two years as a postdoctoral associate at the Operations Research Center, MIT. His research interests include dynamic optimization and decision making under uncertainty and their applications in electricity markets, revenue management and inventory management problems.

Overview

Dr. Conejo states that:

"

This is a visionary paper that analyzes the impact of different electricity tariff schemes on the desirable integration of plug-in hybrid electric vehicles (PHEVs) in an electric energy system. The author is to be commended for such a relevant and timely analysis.

"
Conejo

Dr. Lemoine summarizes the main contributions of this paper as follows:

"

First, while many would suppose that real-time electricity pricing provides PHEV owners with optimal charging incentives, Sioshansi recognizes that generators' start-up costs complicate the story by introducing nonlinearities. Because linear real-time prices reflect the variable cost of the marginal plant, real-time pricing provides strong incentives to minimize variable costs. However, by ignoring the nonlinearities due to start-up costs, these high-powered incentives can produce significantly greater total costs than would blunter rate schedules. Second, he shows that the emission implications of PHEVs depend on two factors: the type of plant at the margin, and how the vehicles change inefficient generator start-ups. Real-time pricing produces greater carbon dioxide emissions than the other tariff schedules considered both because it uses more coal (due to that fuel’s low variable cost) and because it increases the natural gas used in starting up generators.

"
Lemoine

Dr.Litinov sees an additional benefit to this work in that it provides a

"

framework that can be used for the analysis of the PHEV integration in power system operation (both on distribution and transmission levels) and electricity markets. It can also be used to select optimal tariff strategy to create appropriate incentives to satisfy technical, economic and environmental objectives.

"
Litinov

He further notes that:

"

It is very important that the industry converges on certain set of models for this kind of studies. We believe that the author makes a noticeable contribution toward this goal. The methodology is capturing sufficient details while preserving certain level of simplicity and ability to analyze the results.

"
Litinov

Discussion

The commentators have also raised several important questions about where the research on this subject needs further development. While the paper shows the limits of one form of real time pricing (RTP) of charging for PHEV it does not rule out the existence of other approaches to RTP.

Conejo says:

"

Marginal prices derived from balance equations need generally to be modified in order to account for the nonconvexities present in most electricity markets (e.g., start-up costs and minimum power outputs) through uplifts, O’Neill et al. (2005), Araoz et al. (2011), or other techniques, Ruiz and Conejo (2012). The author is kindly invited to comment on how a non-convex pricing scheme may alter the main conclusions derived from the analysis carried out, in which convex prices are used.

"
Conejo

While Lemoine suggests that a direction to advance the analysis would be to:

"

consider the optimal design of a decentralized tariff and the incentives to adopt it. How would the ideal tariff provide charging incentives? Would it operate through hourly auctions? Would the gains from that tariff provide the system operator with an incentive to offer it and vehicle owners an incentive to adopt it? This work on optimal tariff design would take the next step beyond describing the implications of tariffs for cost and emission outcomes to arrive at clearer policy implications.

"

The author responds:

"

The need for improved real-time pricing is made apparent by this and other works studying the limitations of using a linear pricing scheme in a system with non-convexities. These limitations can include prices being non-Walrasian and confiscatory. This work takes a relatively simple case of setting prices based on the dual variables of the hourly load-balance constraints (specifically, these are found by fixing the integer variables to their MIP-optimal value and solving the resulting LP relaxation of the unit commitment model). The non-Walrasian nature of these prices leads to the finding that RTP does a poor job of coordinating PHEV charging with power system scheduling. This raises two possible routes of improving the incentives given to PHEV owners. One is to generate discriminatory pricing schemes which are Walrasian. There is some literature demonstrating how this can be done, for instance the works of O'Neill et al. (2005) (cited in the paper), Araoz et al. (2011) and Ruiz and Conejo (2012). The other is to devise decentralized control strategies that improve charging behavior. In my opinion, the latter is the more practical route to go. The reason is that the pricing schemes will require a utility or system operator (SO) to determine optimal charging schedules, and prices based on those. Likewise, vehicle owners would have to monitor price patterns and adjust charging schedules. Since there are likely to be steep transactions costs on both sides, this seems a foolhardy errand. Instead, PHEV charging will most likely be controlled by aggregators, which can provide a deferrable load of vehicle-to-grid type services to the utility or SO. In order to maximize the value of these services, aggregators would need robust control strategies embedded in PHEVs or charging stations, that would adjust charging profiles based on some relevant market- and system-related signals.

"
Sioshansi

Litinov is in agreement that aggregators will play an important role here:

"

It is unlikely for the system operator to track every PHEV on the grid, so more realistic would be to expect aggregators performing the role of the arbitraging between wholesale and retail prices and either implement V2G or other mechanisms of centralized control and smart charging. In the future, we suggest that the model of the aggregator be analyzed as the more likely scenario. It is still a good idea to use central system operator dispatch as a benchmark for the incentive compatibility of the tariffs.

"
Litinov

Litinov and Conejo also have concerns that the distribution network for electricity is also effected by PHEV.

"

One deficiency in the analysis is the lack of the physical constraints that may significantly tailor the economics of the PHEV market integration. More than likely the limitations of the distribution network will dictate the desirable charging patterns up to the point where the distribution network is declared insufficient and will have to be upgraded or “smart charging” patterns are developed and proven to be adequate for the existing infrastructure.

"
Litinov
"

As electric energy systems become increasingly penetrated by renewable non-dispatchable sources (whose productions are variable and uncertain) and stochastic loads such as PHEVs, network bottlenecks become more and more increasingly relevant, and thus a precise representation of the network on the market clearing procedure is generally important. Moreover, network congestion results in different prices in different locations, which may either encourage or discourage the integration of PHEVs.

"
Conejo

The author responds:

"

A third issue raised by the commenters is the need for more integrated analysis between distribution- and bulk system-level effects of PHEVs.

This is perhaps best demonstrated by the vehicle charging patterns shown in Figure 1 (ed. Of the paper). While a spike in overnight PHEV charging between 3 and 6 am may be beneficial from a generation cost perspective, such a charging load may not be feasible (or may impose non-trivial costs) on the distribution system. Indeed, distribution-level effects are understood to be the most immediate impact of PHEVs, since vehicle owners will tend to be geographically clustered--Mohseni and Stevie (2009) (cited in the paper) examine the possible effects of such clustering. Thus a more integrated view of PHEV impacts is needed to understand what is a truly optimal charging pattern. A major shortcoming of current integration analyses is that they focus on charging impacts at either the bulk power system or distribution level.

"
Sioshansi

Goyal and Lemoine both point to emission pricing as a way to incorporate the emission effects of PHEV charging directly into the objective functions.

"

while Sioshansi models a benchmark system operator that controls PHEV charging to minimize total private cost, we would also like to see how the system operator would use PHEV charging to minimize total social cost. How would valuing carbon emissions affect the optimal charging strategy?”

"
Lemoine
"

In all the optimization models (both unit-commitment model of SO and the charging behavior of users), the decisions are based on minimizing the costs. However, the electricity costs do not include the impact of emissions. As a future direction of research, it would be interesting to see if adding a cost to emissions in the price makes RTP a good tariff regime for reducing emissions

"
Goyal

The author responds regarding including carbon costs:

"

Sioshansi and Miller (2011) analyze the effects of imposing such costs in the Texas system under the controlled and fixed immediate scenarios. They find that it may not significantly increase the overall cost of serving PHEV charging loads. This is because the objective function of the unit commitment tends to be rather flat around the optimum, with numerous near-optimal solutions. When emissions costs are imposed, the system shifts loads away from coal-fired generation to higher-cost but lower-emissions natural gas-fired generators. It is, however, able to do so in a way that the efficiency gains revealed in this work outweigh the cost of the higher fuel used. Clearly there is a limit in how much fuel switching can be done before system costs increase noticeably. Moreover, it is not evident if the negative emissions of some of the tariffs, notably RTP, can be completely overcome in this manner.

"
Sioshansi

The commentators also point to the potential for interesting future interactions between the PHEV charging load and the type of power generation technologies that are brought online.

"

how the plant mix might change under various tariff schedules. If real-time pricing consistently ignores additional start-up costs not reflected in consumer prices, do generators go out of business until the marginal plant's variable cost rises sufficiently? And do some tariffs make coal-fired plants more attractive by providing more consistent demand? ……. The main benefit of real-time pricing may in fact be to make that covariance negative: if high outcomes for conventional demand or low outcomes for renewable generation drive up electricity prices, PHEV demand may fall and partially offset these fluctuations. However, determining how much (and how quickly) the vehicles respond to real-time pricing requires their drivers to account for uncertainty.

"
Lemoine

The author acknowledges that:

"

The case examined in this work does not fully consider the benefits of using PHEV charging loads to accommodate higher penetrations of renewables. Indeed, the case study based on 2005 system data has a paltry renewable penetration level compared to today's reality. Renewables place greater strains on power systems, since conventional generators must be used to balance variable renewable supply with demand. Demand-side management, whether direct load control or indirect price-based mechanisms such as RTP, can mitigate these impacts (cf. Sioshansi (2010) and Mohammadi et al. (2011)). PHEV charging loads can be a valuable source of such demand-side flexibility, however there is very little research demonstrating what types of price- or control-based mechanisms could yield the desired coordination between charging and renewable supply.

"
Sioshansi

Summary

This paper and the above discussion expose a set of very complex and rich interactions in the management of the electric power grid in a period of technological innovation. The power grid itself has two components generation and distribution facilities. PHEVs have an impact on both and so the effect of user charging patterns on both needs to be evaluated. Electricity tariff policies can cause pressure on both components in different ways. At the same time power generation uses different technologies these technologies have different startup and operation costs, different expansion costs, and different emissions. As PHEVs become a greater part of the demand for power they will have an impact on the relative economics and feasibility of these various generation technologies. This research shows that the analysis of tariffs and their impact on PHEV user charging behavior cannot be done within the context of a sterile economic model but rather must be partnered with careful modeling of the operations of these systems. At the same time these models cannot be made so complex that they are intractable. Finding that balance is part of the art of Operations Research and Sioshansi has made a fine effort at striking that balance in this article.

References Cited in Commentary

V. Araoz, K. Jörnsten, 2011, “Semi-Lagrangean approach for price discovery in markets with non-convexities,” European Journal of Operational Research, vol. 214, no. 2, pp. 411-417.

Gowrisankaran, G., S.S. Reynolds, and M. Samano. 2011. Intermittency and the value of renewable energy. NBER Working Paper 17086.

Lemoine, D.M. 2010. Valuing plug-in hybrid electric vehicles’ battery capacity using a real options framework. Energy Journal 31(2): 113-143.

J. Mohammadi, A. Rahimi-Kian, and M.S. Ghazizadeh, "Aggregated wind power and flexible load offering strategy," IET Renewable Power Generation, Vol 5, No 6, pp 439-447, 2011.

J. M. Morales, A. J. Conejo, J. Perez Ruiz, 2009, “Economic valuation of reserves in power systems with high penetration of wind power,” IEEE Trans. Power Syst., vol. 24, no. 2, pp. 900–910.

Moura, S.J., H.K. Fathy, D.S. Callaway, and J.L. Stein. 2011. A stochastic optimal control approach for power management in plug-in hybrid electric vehicles. IEEE Transactions on Control Systems Technology 19(3): 545-555. doi:10.1109/TCST.2010.2043736.

R. P. O’Neill, P. M. Sotkiewicz, B. F. Hobbs, M. H. Rothkopf, W. R. Stewart, 2005, “Efficient market-clearing prices in markets with nonconvexities,” European Journal of Operational Research, vol. 164, pp. 269–285.

A. Papavasiliou, S. S. Oren, R. P. O’Neill, 2011, “Reserve requirements for wind power integration: A scenario-based stochastic programming framework,” IEEE Trans. Power Syst., vol. 26, no. 4, pp. 2197-2206.

G. Pritchard, G. Zakeri, A. Philpott, 2010, “A single-settlement, energy-only electric power market for unpredictable and intermittent participants,” Operations Research, vol. 58, no. 4, pp. 1210–1219.

C. Ruiz, A. J. Conejo, "Pricing non-convexities in an electricity pool," IEEE Transactions on Power Systems, in press, 2012.

R. Sioshansi, "Evaluating the Impacts of Real-Time Pricing on the Cost and Value of Wind Generation," IEEE Transactions on Power Systems, Vol 25, No 2, pp 741-748, 2010.

R. Sioshansi and J. Miller, "Plug-in Hybrid Electric Vehicles Can Be Clean and Economical in Dirty Power Systems," Energy Policy, Vol 39, No 10, pp 6151-6161, 2011.

Comments