RepuTex’s National Electricity Market Renewable Energy Simulator (NEMRES) is our flagship electricity market simulation model, replicating the operation of AEMO’s dispatch engine by simulating market behaviour and supply-demand conditions across the NEM. Various rules, laws and policies govern the operation of the NEM, with the key elements being supply and demand, connected by the electricity network. The supply side is comprised of fossil fuel and renewable generators which offer generation capacity based on their own economic decisions, dispatched by AEMO from the cheapest to more costly generator, subject to system conditions, to meet demand.
Demand is affected by a number of factors such as weather, economic activity, population, etc. Although demand for power has patterns, it remains mostly unplanned and highly inelastic over the short term. System operators rely on demand forecasting for the daily market operation and long term planning. As such, AEMO publishes forecasted demand on a number of timeframes. NEMRES simulates the least cost dispatch process and supply and demand conditions in each forecast period, modelling the resulting generation and emissions from each of the scheduled generation plants. Contracts between generators and retailers/large users impact the percentage of electricity, subject to bidding behaviour and spot price revenue.
NEMRES explicitly models all scheduled power plants, while also allowing for non-market and non-scheduled plants. Figure 1 outlines the main model components and process flows. The central component is the least cost dispatch model, which dispatches the generation of plants based on default bids, adjusted to each generators most recently observed patterns. For each dispatch interval, fossil fuel generator bids are optimised for individual facility profitability. Hydro generation is allocated based on historical inflow and the proportion of run-of-river generation and storable hydro energy.
As shown, input data preparation and model calibration are important blocks, supported by a number of criteria in checking the validity of model outputs, including crosschecks against closing facilities projected to be the least profitable, and the feasibility of new entrants in a given region.
RepuTex NEM Generation Model & Sub-Models
Merit Order Model
A merit order is constructed via the bids offered by all fossil fuel plants. The algorithm orders the price bands offered by plants from the least to highest and accumulates the quantities of corresponding price bands accordingly. When network constraints or inter-temporal constraints are present, an optimisation solver is applied to achieve the role of the algorithm developed for the model.
The bidding model constructs the default four price and quantity pairs. All price and quantity pairs are in percentage of the cost and available capacity of each plant except the price in the first band, which is fixed at $0 per megawatt-hour (MWh). The first band of a bid applies to plant-level minimum generation. The second band applies to short-run marginal cost (SRMC) and the third to long-run marginal cost (LRMC). The last band is related to the value of lost load (VOLL).
The quantity is the percentage that a plant is willing to offer to the market at above given prices. The quantity is incremental, in that the sum of the four quantity components must be 100 per cent. The quantity at the SRMC cost is related to the contract level, while the quantity at the LRMC may be allocated to the normal design level less the amount that has already been allocated in the previous price bands. The last band can be thought as opportunity or gaming bids. There are two default formats. Long-term forecasting calculates dispatch on annual demand duration curve with 200 dispatch intervals per year. High precision forecasting uses the bidding format for the chorological half-hourly dispatch against half-hourly load curves.
The cost of a generator depends on a number of factors: plant characteristics such as plant efficiency/heat rate, plant auxiliary usage, fuel cost, fuel combustion emission factor, variable operating & maintenance (VOM), fixed operating & maintenance cost (FOM), etc. The short-run marginal cost (SRMC) and long-run marginal cost (LRMC) are calculated by summing each cost components as shown in Figure 2, which also shows the contributing factors of each.
To calculate per MWh cost of the fixed cost, a capacity factor is assumed for each plant. This may have impacts on dispatch outcomes. Normally, the impact should be minimal. However, this side-effect can be voided by adjusting bids based on plant profitability because the annual profit of each plant will not use the assumed capacity factors. Annual profit is calculated as total revenue from sent-out energy + fixed subsidies less the variable cost associated with per MWh generation and less the annual fixed cost.
Demand Trace Model
Annual forecast demand comes with three numbers for the NEM. One is for annual energy and the other two are for maximum load in the winter and summer seasons. An annual load shape is chosen to allocate forecast demand into finer time scales. RepuTex aims to mimic the operation of the NEM over 200 periods per year, equivalent to averaging demand over 1.8 days. Once the load shape in a particular historical period is chosen, the Demand Trace Generator can produce a demand trace matching the historical shape and forecasted energy target and the maximum load in the winter and summer season. Weekends and public holidays load profiles are to be checked and matched as required. Forecasted demand for scheduled and semi-scheduled generation is used only at the stage as only scheduled and semi-scheduled plants are modelled.
Wind Trace Model
Wind generation is of high randomness and it is classified as semi-scheduled by AEMO. In addition, new wind farms do not have historical generation data to use. As such a comprised method, which is widely used in simulating power market operation, is used to model the wind generation. For new wind farms, assumptions are made for their capacity factors based on availability of wind resources or similar wind farms located nearby. Once annual energy and potential maximum output for the wind farms is available, it is disaggregated into the wind traces in shape.