Get Complete Project Material File(s) Now! »



Cleaner electricity production is likely to play even more important roles in the future. It has become a key goal of governments to reduce carbon emission form electricity generation. Solar and wind power can improve power generation by diversifying renewable electricity production yet need to be carefully incorporated into the system to minimise system cost, and avoid from over-generation whilst meeting demand. In the New Zealand market hydro can help with managing an increasing amount of intermittent generation as it can act to some extent as a battery, generating more when wind and solar generation is low and less when it is high. However, as discussed above there is limited hydro storage capacity and inflows are low during winter months when demand is highest. The focus of the previous essay was to introduce solar power and examine how it works in the power system with hydro inflow. The focus of this essay is on intermittent resources, solar power and wind power, in New Zealand. While there is a strategy to expand the size and geographic spread of wind farms, solar PV installation has also been increasing. It is therefore timely to examine how these resources can affect the electricity market. There are several important advantages to adding combined solar and wind into electricity generation systems. Firstly, wind and solar are abundant resources and they do not need ongoing exploration. Secondly, their availability is not limited by their usage for other purposes such as irrigation, conservation and recreation yet it requires land to set up the technology. Thirdly, combining these resources together could alleviate concerns about periods that they are not available individually. Last but not least, the cost of generation from wind or solar is not affected by volatility in international fossil fuel markets. Capturing energy from solar PV and selection of sites across the country was explained in essay 1. The following section provides a brief review of wind power on a global scale, wind generation in New Zealand, and an explanation of solar and wind integration into the existing hydro-based network. Wind energy is freely available and wind power is one of the least-cost options of new power generation. It is one of the world’s fastest-growing electricity resources. Wind power is largely affected by wind speed. The relation between wind power and hub height wind speed produces the wind turbine power curve (Figure 28). The minimum speed at which wind speed results in wind power is called “cut-in speed”, and the maximum speed at which the turbine is allowed to generate power is the “cut-out speed”. The maximum output power is almost fixed at a constant rate that is known as “rated speed” (Lydia et al. 2014). Nowadays, 75 countries worldwide have developed commercial wind power installations. Wind farms often contain identical turbines; each turbine works independently (New Zealand Wind Energy Association (2012). Wind power deployment has more than doubled since 2008 to reach 300 gigawatts (GW) of cumulative installed capacity International Energy Agency (2013). In 2012, wind power’s contribution to meeting global electricity demand was 2.5%. Wind installation was more than 51 GW in 2015, with 44% annual market growth as the industry recovered from its biggest ever fall in 2013. Wind power is expected to grow to satisfy about 16% of global electricity demand by 2050 (International Energy Agency, 2013). The majority of new installations are happening outside OECD countries. While development in wind power has remained consistent in the OECD, Asia has been the largest market for seven consecutive years (Renewables, 2015). The Asian market is led by China, which has surpassed Europe in terms of total wind capacity (Global Wind Energy Council, 2015). New Zealand’s wind farms are very productive, generating electricity at 40% of rated output on average. This is due to geographic conditions and strong wind speeds. Most of the wind stations are located in the Waikato, Manawatu, and Wellington regions of the North Island. The first wind farm was built in Brooklyn in 1993 and by 2004 the wind industry had grown ten-fold. West Wind near Wellington is one of the largest farms, generating power equivalent to the demand of all homes in Wellington city (New Zealand Wind Energy Association, 2013). New wind farms are planned across both the North and South Islands which may entail an upgrade of transmission lines to high-demand locations. Due to intermittent characteristic of wind power, more flexible generation such as hydro and gas peaking plants are needed. Geographic diversification of wind farms is also known to smooth fluctuations of wind power generation. The cost of wind generation has decreased and made it more cost competitive. Moreover, it has been calculated that when the emission-trading scheme is taken into account, thermal generation is more expensive (New Zealand Wind Energy Association, 2011). Similar to solar power, wind power provides electricity at the price level of zero once it is available. Taking into account improvements in wind technology and a fuel cost of zero, wind generation costs are stable, in contrast to the variable fuel costs of thermal generation. In the last decade, the wholesale electricity price has increased largely because of increase in gas prices. For example, in 2008 the price of natural gas increased 31.6%, reaching 6.96 $/GJ. (New Zealand Wind Energy Association, 2011). Demand for electricity is expected to grow 1-1.2% on average per year to 2030, with corresponding impacts on peak demand. The New Zealand government aims to achieve 90% renewable generation by 2025 and increasing wind generation is one of the key strategies in this regard. Wind power is expected to contribute 20% of generation by the year 2030, indicating an increase in wind capacity of about 3500 MW (New Zealand Wind Energy Association, 2013). As discussed in the previous chapter, solar power combined with wind power put further strain on other plants to ramp up once solar is not available. As a result it pushes up the system cost. For example, the results of the simulation under competitive conditions indicated that medium and high solar scenario crowds out other renewables such as wind, hydro and geothermal. In the high solar scenario with diminished lake levels, the national price increases. In designing a portfolio of intermittent energies, it is essential to capture the most effective mix of generation. While the aim is to expand wind power across New Zealand, there could be potential locations, particularly in the sunnier North Island, where aggregating solar into the system provides a more cost-effective solution. There is a possibility that sunny moments and windy times counterbalance each other. If sunny hours or days are less windy, this is an indication that these resources could complement each other well. However, this is an insufficient basis from which to propose an effective energy mix. Different sites are expected to benefit from different shares of wind and solar due to varied climatic and geographic condition, as well as different nodal demand and price of electricity. Some commentators are cautious about high reliance on renewables, mainly due to the sudden variations in power production. In the case of wind power, many studies agree that wind power significantly increases variability in generation, which works against the security of supply (Mills et al., 2009, Mills and Wiser, 2010 and Budischak et al., 2013).
While individually large solar or wind generation will increase power variation, combining them helps to control variability and reduces the early need for infrastructural investment. Nandi and Ghosh (2010) showed that in the Bangladesh electricity network, wind and solar together provide a reliable alternative system as they alleviate sudden fluctuations in generation. Jacobson et al. (2013) investigated whether it would be possible by 2030 to generate all New York City’s power requirements from renewables, including solar, wind, geothermal, hydro, tidal and wave. They found that it was not only possible, but that such a system would mitigate demand from the grid by 37% and electricity price would be reduced as the fuel cost is mainly zero. In a hydro-based grid the intermittency of solar and wind generation can be offset to some extent at least by hydro generation ramping up or down Suomalainen et al. (2015). In the New Zealand electricity market, the relative stability of hydro-power makes it a good back-up for intermittent generation, resulting in higher supply reliability and lower generation cost. However, the New Zealand hydro network has limited lake storage. Another possible solution to lessen variation in power generation is utilising battery banks, as highlighted by Ekren and Ekren (2010). Barnhart et al. (2013) explained that the cost of storage in the case of PV power is greater than curtailment; however, this is not the case for wind power33. Dong Energy et al. (2006) and Budischak et al. (2013) demonstrated similar outcomes. Budischak et al’s (2013) model for the year 2030 showed that solar and wind meet nearly all power requirements, and supply could be managed to cover load over seasonal variations by aggregating storage for only a few days, rather than months, which would reduce total system cost. This essay extends the work described in the literature and in the previous essay by focusing on the relationship between the two intermittent power sources, solar and wind, and studying them in relation to demand and price of electricity. It investigates the market price in a competitive network setting by defining three more scenarios for different share of resources to simulate the power market for the year 2025. There is not a comprehensive study to analyse mixes of generation by statistical, and market analysis at the same time. This provides detailed examination of data and their relationships as well as delivers more robust analysis when combined with market simulation. Section 4.2. explains the data applied. In the next section, 4.3. statistical analysis is discussed. In the section 4.4, the method utilised to examine the correlation between data and the simulation model is presented. In general, the applied method is similar to the previous essay; however, it is amended to address the question of this essay. Section 4.5 results are discussed and finally, this chapter ends with a summary and conclusion in section


Each set of data is presented below in its own sub-section.Wind data are simulated for a number of potential wind farm sites by the New Zealand National Institute for Water and Atmospheric Research (NIWA). The data represent wind speeds for 15 actual or potential wind stations across New Zealand. Wind speeds are measured in m/s in 10-minute intervals between September 2003 and August 2008. Data are synthetic, inferred from actual measurements at nearby sites using the local area MM5 model developed by Pennsylvania State University and the National Centre for Atmospheric Research (Browne et al., 2015). For the correlation analysis, wind and solar data for the years 2003-2008 is used and the simulation analysis uses data from the years 2004 and 2006 with anticipated generation mix and network capacity in 2025 for different scenarios. In order to match wind farms in the models to the dataset above, the location of each wind farm built by the New Zealand Electricity Authority’s Generation Expansion Model (GEM) is linked with the nearest wind farm site reported in the NIWA database, as described in (Brown et al., 2015, see Figure 29). NIWA only provides information on wind farms by region; it does not identify their exact location. The actual wind farms are able to be accurately linked to wind sites by looking at wind output data in the New Zealand Authority’s Central Dataset (CDS) and comparing them to the synthetic data. Note that GEM allows a limited number of wind farms in each region. Working with half-hourly data, three 10-minute measurements on and prior to that half hour are averaged.To enable aggregating energy from wind and solar in this study, their energies are converted to power. A simulated power curve developed by Mclean (2008) for the European Wind Energy Association is applied to convert wind speed to power. The simulation model which has a two-sided normal distribution provides a good fit for the values, as shown in Figure 30. This formula has been applied in previous studies and the robustness of data has been supported (Brown et al., 2014 and 2015). Using the formula below, wind speed in m/s is converted to power in MW. With wind speed denoted as w, the power curves are Sun radiation data are also taken from NIWA34 and are measured on a horizontal plane in MJ/m2 every 10-minutes. As explained earlier in essay 1, this study focuses on solar stations set up to provide energy for residential use. Since PV modules are normally installed at a fixed angle on rooftops, the converted data for the inclined surface is calculated taking into account the optimum angle that results in the highest absorbed energy for each location. The 15 sites selected as described before are the sunniest and/or sites close to the major cities across New Zealand.Demand and price of electricity Data on electricity price are from CDS35, provided in half-hourly NZD/MWh values for the wholesale market from 2001 to 2013. Demand for electricity, measured half-hourly in MW from 1996 to the end of March 2011, is also from CDS. Both demand and price of electricity data are from a sample of 18 main nodes throughout New Zealand.

1.1. Overview of electricity generation globally
1.2. Electricity generation in New Zealand
1.3. Research problems
1.4. Structure of the thesis
1.5. Objectives of the study
2.1. Availability of resources and their relationship
2.2. Methods to analyse electricity market
2.3. Market Analysis
2.4. Summary and conclusion
3.1. Introduction
3.2. Data
3.3. Indicative analysis
3.4. Method
3.5. Results
3.6. Summary and Conclusion
4.1. Introduction
4.2. Data
4.3. Statistical Analysis
4.4. Method
4.5. Results
4.6. Summary and conclusions
5.1. Introduction
5.2. Method
5.3. Results
5.4. Summary and conclusion


Related Posts