Simulator of US Wholesale Power Market

Electricity markets are auction markets where power generators (suppliers) and wholesale consumers (buyers) submit their bids in the form of price-quantity pair. We assume two kinds of traders: generators and wholesalers. The traders can participate in two markets: Day-Ahead (DA) market and Real-Time (RT) market. The demonstration contains 4 charts: Day Ahead Curve, Real Time Curve, Price vs. Iterations, and Volume vs. Iterations (Click here for a complete demonstration of the power trader).

The DA market is open to the traders one day before the actual transaction. Both wholesalers and generators submit bids and the corresponding prices. The supply curve (generators) and demand curve (wholesalers) are plotted and the MCP (Market Clearing Price) is determined by the point of intersection of the two curves. This is illustrated in the ‘Day Ahead Curve'. The x-axis indicates quantity and y-axis indicates price. The green curve is the demand curve. The blue curve is the supply curve. The point of intersection is projected on both the axis by a red line. The projection on the y-axis gives the market-clearing price of the DA market. This is the price that all the winning traders pay for the amount of electricity that is allocated to them.

The ‘Real Time Curve' chart describes the RT market. In the case of the RT market, generators can submit bids in the form of price and quantity. The wholesalers can only specify the demand (quantity) and not the price. The wholesalers have to accept whatever price is decided on the RT market as a result of the bids of generators. The supply curve is denoted by the blue color. The red line shows the total demand of all the wholesalers in the real time market. The projection on the y-axis gives the MCP for RT market.

The ‘Price vs. Iterations' chart dynamically shows the trend of market price in DA and RT markets at every iteration. The ‘Volume vs. Iterations' chart dynamically shows the quantity that is transacted among the traders in DA and RT markets for each iteration.

The software illustrates an artificial market in a computer where many traders make their decisions regarding selling and buying electricity, based upon not only external information on the electrical power market but also various economic conditions (e.g., an oil price and a demographic change). The software also includes information about initializing the behavior of the various traders. This simulation study explains that a risk-averse trader with a learning system described in the paper performs better in terms of rewards.

For a more technical description of the power trader, refer to A Wholesale Power Trading Simulator with Learning Capabilities.