ANALYSISMANUFACTURING COST
then try different spacing layouts with the arrays to optimise the amount of unshaded collector area; in other words, the fraction that provides income, per unit cost, where the cost of all the collector area and land area is included. Optimum spacing is often surprisingly sparse and in one analysis for a particular site in the sunbelt region, the balance led to an optimal packing density (ratio of collector to land area) of about 13%. For this case, the cost saving on $/kWh between 40% and 13% packing density was found to be up to 7%.
By using a measure of output yield per unit unshaded collector area we can then estimate the area of land required, the total area of collector and the costs associated with these. Using the model then shows how a cost reduction in the collector modules reduces not only collector costs, but also the cost of land, as the optimal spacing becomes denser. This may reveal some optimisations that might otherwise have been ruled out if assessed without considering which interactions are actually cost effective. As well as being able to adjust inputs based on changes to other parts of the system, the modular nature of the model allows more detail to be added once known. For instance, once the exact site has been identified weather data can be added to include known weather patterns. Weighting the shading average with data on sunshine levels by time of day and year gives an estimate of the average proportion of collector shaded during the sunny periods where revenue is generated.
Minimising the impact of potential failures Many solar systems are installed in very aggressive environments with sand storms eroding exposed parts, strong UV from the bright sun and heavy thermal cycling between hot and cold. Enclosures are designed to give good protection from the elements but it is only cost effective to protect modules to a degree. So another challenge to system designers is what to do if the worst does happen and a module fails. It needs to be considered: when to replace failed modules, when not to, and whether it is worth designing the system to allow for easy replacement at all. There will always be a failure rate of collector modules, especially in large, multi-megawatt installations when the number of modules can be counted in hundreds of thousands. The way to decide whether it is cost effective to replace a failed module is similar to analysing how cost effective a complete system is in the first place because there are similar upfront costs and payback times. First, the upfront cost needs to be adjusted for the cost of raising money a length of time ahead of the payback – effectively the cost of delaying that return. Secondly, the payback must occur within
the remaining life of the installation. This immediately highlights that units should not be replaced towards the end of the solar farm’s life. Although the cost of replacing modules is likely to be lower than installing the same number of modules as additional capacity at the start, additional upfront modules will be present for full life, and so are likely to achieve greater return on investment. Another difficulty with replacing modules is finding which module has failed. This means that it might well be worth installing additional capacity at the start and accepting a decline in output over time.
On top of this, if the systems are designed not to have replaceable modules, the complexity and cost of the module and the array holding it will be reduced, and with fewer seals and electrical connections the overall reliability may increase too. On the other hand, if the module costs are less significant than the rest of the system such as the tracker and inverter costs, then it may be worth designing modules to be changed quickly and cheaply to maximise use of the more expensive components.
41
A closely related design target is reducing the implications of potential failures, as some components are critical for an entire module to function, and a failure would prevent the whole module from generating power. In this case, it may be worth arranging the system in such a way that the implications of a component failure are reduced; for example by reducing the number of cells wired in series so that one local failure ‘wastes’ fewer connected cells in a string.
Figure 3: Deep drawn module with 30 cells is easily handled by one person to facilitate assembly and maintenance. 96 modules are mounted together on a tracker to make up each array
www.solar-pv-management.com Issue VI 2011
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48