A range of variables produced many scenarios for analysis. Variables were each tested for sensitivity. This method enabled assessment of each variables relationship to the output. Producing scenarios for every variation would mean an assessment of over a thousand scenarios. It was decided to only vary the inputs, which would have a significant effect on the final output. These were used for sensitivity. This reduced the range of scenarios. The 4 main areas of cost analysis were :-
1. The present and future cost of wind generation.
Our research has found reported evidence of an expected drop in the future costs of wind power. This would be due to :-
- economies of scale within a larger market share,
- improvements in technology, and
- lower startup costs for 'second generation' installation at existing approved sites.
2. The present and future cost of Pumped Hydro Storage.
We see a future for utilisation of pumped storage as an intermediate stage in the delivery of wind generated electricity, but this would be cost sensitive. Both upgrade of existing hydro schemes and new built pumped hydro systems are considered. An estimate of the capital cost of pumped hydro was made by means of deducting certain civil costs, which would not now be required, from documented construction costs. The same criteria for interest rate and accounting period were then applied.
3. The future cost of Grid Upgrade.
This is documented in the RETS report and shows future costs of grid upgrade relating to the increase in capacity to cope with 2GW, 4GW, and 6GW of wind capacity.
4. The present and future cost of Conventional Generation.
For the cost of conventional generation we obtained an average buying price from 'www.bmreports.com'. This would be the price to pay if we needed to replace any wind generation, which may not be there due to low or no wind.
(link to spread sheet model at page end)
After weighing up options that could have been used we opted for Excel. This tool was easily accessible from university and home PCs, allowed for relatively easy communication of problems and/or results, and had a good range of output options for presentation purposes.
The main project component parts of Wind costs, Storage costs, Conventional backup costs, and Grid upgrade costs were the main headings for the data in the model. The combination of these would give a resulting overall cost of electricity generation.
Research of the subjects gave the inputs used under the main headings. Development of the storage model as a sub-analysis gave the input required for that area.
The data was displayed in one sheet with the graphical output in another
From researched data it was assumed that :-
The introduction of a dynamic aspect to the analyses meant that there had to be an appreciation of how costs would change with time. This would be very hard to predict with any great degree of confidence. This affected how we approached conclusions.
The starting point of 52 TWh as the annual electricity consumed in Scotland was used for the purposes of this study.
A 'control' scenario was predicted based on no development of renewable electricity generation further than the current level of about 3%. This 3% is not solely provided by wind generation. Other basic scenarios were based on how quickly targets were met. We would hit our target of 40% by 2020, or be a bit behind and reach our target by 2030, or be well behind our target and reach it by 2040.
After the initial spreadsheet was constructed, an analysis of the variables was undertaken and proved to be time consuming. Each variable was altered in turn to assess the effect that each had on the overall output. This produced some very large spreadsheet results. Variations in capital cost of pumped hydro, for example, or the Regulator contributing to the capital cost of Grid upgrade, were applied to the overall model and it was found that these did not make a great deal of difference to the overall output.
The main finding were presented in our presentation, such as the increasing costs of electricity generation with greater levels of wind penetration, how does each cost factor contribute to the overall cost, and a comparison of our storage technology used against conventional backup being used.
The increasing costs of generation can be viewed in the data sheet of the excel model.
These can be accessed by downloading the model from the downloads section. The following two graphs are examples of the sort of visual outputs from the model.
Analysis of outputs .
We analysed the outputs and found that increasing costs of electricity generation is unavoidable. The greater sustainability of a generation profile with higher levels of Wind penetration combined with a pumped hydro storage system would incur extra costs. The extra cost would vary intertemporally depending on how quickly we reach our targets, between an extra £2 and £4 per MWh. Using storage techniques as indicated would be significantly more costly than conventional generation as backup.