Nodal Pricing in Electric Transmission: A Customers View

Various methods of determining local prices for access to power on electricity grids have been proposed since deregulation became widespread. The aim of power regulatory authorities is to ensure fairness by restrictions on collusion. Nodal pricing is used in various transmission networks around the world but has a flaw in the opportunity it offers suppliers to exploit the nexus between historical topology in transmission networks and inevitable line load limits. Introduction The past 15 years of market deregulation in the power industry has generated considerable research into efficient transmission and pricing mechanisms. Robinson (2003, 2005) has commented on the role played by mathematics in analysing the networks and explaining some of the more public failures of the system, such as the collapse of the transmission grid in the North East of North America on August 14 2003 and the pricing debacle faced by operators in California in 2001. Deregulation of electric power industries worldwide has led to various pricing dilemmas facing the network operators in those markets. One aspect of this is seen in the trading market where the spot price of power is set from those producers bidding what they will supply at a raft of prices termed the offer stack (see e.g. Kliman, 1994). Another is the affect that the transmission network itself has on the local price dictated by available generation and demand. Nodal pricing is a process applied to set prices at the nodes of a transmission 16 network. Despite some inevitable 'freak' pricing this may generate from the so-called Springwasher effect, nodal pricing has proved popular and is widely used (see e.g. Oren et al, 1994; Whiten and Marion, 2005). There has been some pressure to introduce this to the Australian market (Biggar, 2005). The pricing of electric power is a contentious issue. Generally markets have been organised with standard auction designs and rules on collusion to yield simple demand driven single pricing. This inevitably leads to price spikes (see e.g Fabra et al, 2004). Conversely however, it has been observed that demand itself is not so elastic with price (see e.g. Sapio, 2004) because actual end users of power are generally in regulated price structures, unlike the power retailers. In a complex network, prices are often set at network nodes by use of the demand/auction bid system. This nodal price paradigm operates in New Zealand and many other deregulated markets (Sotkiewiecz and Vignolo, 2005). One perspective says that customers should pay an average price for the electrical power that is supplied to them via the transmission network. The other side of the argument says that the customer should pay a price that is dependant upon the node at which they join the network. There are several unsatisfactory aspects of nodal pricing for electrical power transmission networks from the viewpoint of the customer. The basic objection of the customer who demands the electrical power is that the marginal price is applied to the total supply, and For the set of demand nodes I d and generating nodes I g, the nodal pricing and dispatch problem (ignoring transmission losses and ancillary services) is given by the following LP. We use a well-known linearised form of the power flow equations, known as the DC approximation (Oren et al, 1994). This approximation ignores transmission losses and ancillary services. The dispatch problem seeks to find the pattern for generating the supply of electricity to clear the greatest proportion of the total demand at the least total cost, subject to the constraints on the generators and the links. The solution to this problem can be found by linear programmmg. Consider the following variables and parameters in the network. gi = Cleared generation at node i d, = Cleared demand at node i D, = Specified load at node i Gi = maximum generation at node i Pi = net injection at node i (,j = powerflow along link (i,j) L, j = maximum powerflow along link (i,j) Bj , j = admittance of link (ij) C, = generation offer cost at node i not just the marginal supply. In particular, we show how low bidding generators could manipulate the marginal price to their advantage, and customers who are not well served by the network (possibly though accidents of history and geography) can face excessive charges. Adverse consequences of nodal prices have been discussed in Wu et al (1995), where a number of 'folk' myths on nodal pricing are exposed. Cho and Meyn (2005; also reported in Robinson, 2005) have argued that decentralised power markets have inherent instability that arises from neglect of some constraints within the pricing models. The Tr-ansrrrisaion Network An electrical transmission network can be represented as a set of nodes where power stations and/or consumers join the network, and a set of links between the nodes along which the power is transmitted. The links between the nodes have two properties that are important to this discussion. The first is the admittance of the line, which is a physical property relating the electromotive force (emf) applied at the ends of the line to the power transmitted along the line. The second is the maximum safe power limit, which relates to the heat losses on the line during transmission that can cause failure of the link