Solar PV & Thermal: Seasonal Variability - So what's it like in Winter
Nowadays, even among some of those who have historically been most resistant to change, it's almost indisputable that renewable energy technologies at a national scale act to both seriously reduce reliance on fossil fuel generation and benefit the environment, however, there still remains a considerable level of dispute related to such benefits due to seasonal generation variability, particularly that related to using the sun as the energy source.
We've all heard or seen debate or argument criticising renewable energy using phrases along the lines of "... when the wind doesn't blow and the sun doesn't shine" and, of course, "... the short days of deepest winter". Yes, there's no argument here, it's a fact that the sun really does only shine during the day and wind really does tend to blow intermittently, not only that, but due to both being seasonally variable, the power generated is not always available when it's needed. Providing a degree of balance to that, isn't it also the case that the windiest months in the UK are during the darkest months of winter and early spring when daylight hours are short and the calmest months are during the summer when daylight hours are long?. Maybe there's a need to use a more inclusive approach when considering renewable energy so as not to concentrate on the weaknesses of individual technologies in isolation, a need to stand back and appreciate the entire picture, but that's the subject for a future article: For now, let's look at the perceived weaknesses related the phrase "... the short days of deepest winter" and see the extent to which they really affect technologies based on solar energy, whether it's solar pv for generation of electricity or the direct collection of heat via a solar thermal system.
We all know that solar pv and solar thermal systems use energy derived from sunlight and that during the winter daylight hours are shorter, the sun's arc through the sky is lower and therefore light intensity is diluted, so just how much effect does have on your expectations for solar PV or solar thermal contributions during those drab and cold months ?
Technical
Both solar photovoltaic (PV) and solar thermal systems rely on collecting radiation emitted by the sun. It's really important at this stage to note that both forms of solar power rely on the intensity of light falling on their collectors, not heat.
Solar PV
Although a considerable number of people think that PV systems will generate more electricity on a hot mid-summer day than a colder sunny day in spring, it's just not the case. The operating efficiency of crystalline solar PV panels reduces as the surface temperature of the cells increases, typically this is somewhere around 0.5% per degree C of panel temperature rise, which during the summer even in the UK can easily reach 50°C. Considering that panels are output performance rated at a standard 25°C, a temperature of 50°C would result in a typical 12.5% reduction in power, therefore cool, bright and breezy conditions which combine to provide lower panel operating temperatures prove to be more suitable for PV generation. Considering that lower operating temperatures in bright conditions are good, then generation in the colder months should be better?, well no because of two major factors, the incidence angle of the sun to the panel surface and the number of daylight hours available. The lower the sun is in the sky, the further light travels through the planet's atmosphere, the higher the refraction and reflection of energy back into space, therefore reducing the energy reaching the panels. Although solar PV panels operate more efficiently at lower temperatures, the sun will always be lower in the sky in winter months than in summer, therefore the higher efficiency is being applied to lower insolation. Maximum power levels attained by solar PV in high latitude countries such as the UK are heavily impacted by reduced irradiation due to the sun's angle of incidence, but this is partially offset by operating temperature efficiencies, however, it's not so much the power which is relevant, it's the effect of that power over time, which is the definition of energy. The nearer to the winter solstice, the shorter the daylight hours, therefore the less time available to accumulate power (kW) into energy (kWh), therefore the lower the generation total.
Variability
Various sources suggest that a south facing solar PV system would be expected to generate around 100kWh per kWp of installed capacity during June and July, falling to 25kWh/kWp in December and January, so a ratio of around 4:1. Our own system which faces approximately South-West has a ratio of over 5:1, this being due to the comparative reduction in direct sunlight falling on the panels as the combination of degrees separation in orientation from South increases and the available daylight hours decreases. So there's two figures, but in general terms, what's the average generation ratio likely to be.
To address that we look to a well populated data source representing systems across the UK, PVOutput. The analysis represented by the figure below shows the actual monthly contribution of generation towards the annual total for a group of 549 solar PV system of various orientations and sizes installed throughout the UK. Of the 549 systems, approximately 260 supply data daily. To compensate for changes in installed capacity, the analysis is based on average monthly efficiency in kWh/kWp for the total data provided for each month. The inner ring of the doughnut represents 2011. the first full year of the UK Feed-in-Tariff (FiT) scheme, progressively incrementing through to 2016, the last full calendar year for which data is currently available. The outermost ring represents the average for all six years, showing the relevant percentages for each month and should be read clockwise..
On a like-for-like basis with previous comparisons, the analysis clearly shows that total generation for June and July accounts for 26% of the annual total, whilst December and January combined make up 5%, a ratio of just over 5:1.
The above data can be further analysed -
By Quarter - Q1 : 17% / Q2 : 38% / Q3 : 34% / Q4 : 11%
By Calendar Half Year - H1 : 55% / H2 : 45%
By Daylight Half Year - Long Daylight Half : 72% / Short Daylight Half : 28%
The conclusion to be made from the analysis above is that although there is a significant weighting in generation towards the warmer months, solar PV is far from being totally ineffective during "... the short days of deepest winter" as long as domestic energy consumption is efficient. As an example, the figure below represents our own household imports of electricity for the last 12 years, note that peak Winter electricity imports have recently been in the region of 150kWh, with the Summer minimum being around 60kWh.
Taking this in context, the average generation for the PVOutput team 'United Kingdom' used in the analysis above is 2.42kWh/kWp/day, therefore a typically average 4kWp system would reasonably be expected to generate an average of 3533kWh/year, therefore applying 2% for December and 13% for June resolves to monthly generation figures of 70kWh & 459kWh respectively. Obviously, generating 70kWh in December won't cover a significant proportion of total demand due to the reduced daylight hours, however, 70kWh of generation will certainly cover baseload requirements for the majority of those daylight hours resulting in a high level of generated power self-consumption.
Solar Thermal From the above we have deduced the seasonal effect on solar PV in terms of generation balance and can see that the technology has the ability to cover a reasonable level of daytime energy demand over the winter months. Let’s now look at solar thermal which directly heats water using sunlight.
Unlike solar PV, solar thermal technology doesn't have the ability to make energy available immediately. Heat needs to be collected by the panels over a period of time before being pumped into storage, therefore overcast and intermittently cloudy conditions will both reduce energy collection and increase the time available for panels to lose what has already been collected, thereby have a considerably larger impact than apparent with solar PV. Considering that the important term here is heat and that winter months are by definition cold, it would be reasonable to expect higher winter month heat-loss, resulting in a solar thermal system showing a higher seasonal imbalance than solar PV. To counter this it is relevant to note that, unlike grid connected solar PV, solar thermal system heat provision on long sunny summer days is limited by demand, not availability: when the domestic hot water (DHW) storage reaches a pre-set temperature the system shuts down, effectively acting to cap the seasonal imbalance. Some solar thermal systems are designed to divert excess heat to some form of heat dump, but this seems to be relatively rare in more modern installations. Whether the solar thermal system shuts down and 'stagnates' or dumps excess heat when the the storage temperature set point is reached, it is important to note that these systems need to be sized correctly to match individual household requirements, if too small then they will not meet demand, whereas over-sizing a system simply results in the extra capacity being unused for much of the year. The Energy Saving Trust (EST) field trial report of 2011 regarding solar thermal systems suggests that solar thermal should provide around 60% of annual domestic hot water requirements when installed and used correctly, the average (median) across the trial being 39% - the poorest performing systems contributing as little as 9% and the best 98%. Personally, I have a number of very strong reservations regarding the report, particularly the system specifications, selection criteria, lack of supporting data and highly relevant to this article, the absence of meaningful seasonality analysis combined with overlooking the synergy advantages of operating both solar thermal and PV systems.
Various manufacturer sources suggest that systems which are correctly sized and matched to household demand provide around 90% of domestic hot water (DHW) in high Summer, and around 25% during the winter, but how does this stack up against how our own system performs. The following figure represents the monthly contribution towards the total annual hot water provision from a reasonably sized evacuated tube solar thermal system. The inner ring of the doughnut represents aggregated monthly gas usage as a percentage of the aggregated total consumption of 1831kWh.t (DHW and space heating) for the three years 2014-2016, with the outer ring showing the relevant monthly proportion of annual solar thermal DHW provision over the same three years. Again the percentages for each month should be read clockwise..
To ensure consistency with previous comparisons, the analysis clearly shows that total generation for June and July accounts for 26.7% of the annual total, whilst December and January combined make up 6.2%, a ratio of just over 4:1.
The above data can be further analysed -
By Quarter - Q1 : 16.6% / Q2 : 36.4% / Q3 : 34.4%* / Q4 : 12.6%
By Calendar Half Year - H1 : 53% / H2 : 47%*
By Daylight Half Year - Long Daylight Half : 70.8%* / Short Daylight Half : 29.2%
( * : Rounding Adjusted +0.1% )
The conclusion to be made from the analysis above is that although there is a significant difference in the technologies employed, the monthly weighting for this particular evacuated tube system is almost identical to the average profile across a considerable number of PV systems. Note that over the three years analysed there was no backup energy provision through gas water heating in the months April to September (Daylight Half Year) and very little in March, October & November, which tends to convey that excess energy collected and represented as higher percentages in Summer months mainly leads to higher DHW storage temperatures and therefore proportionally higher heat-loss. Looking at the relative provision of heated DHW between January (2.9%) and the marginal months of March and October (8.3% & 6.3%. average 7.3%) suggests that around a third of DHW requirement is supported by solar thermal even in the poorest month. These figures tend to reinforce my reservations regarding the EST field trial and suggest that the majority of the systems were either incorrectly specified to meet DHW demand, were inefficient in operation, were used incorrectly, or the household heat demand pattern could be better optimised towards solar thermal heat provision, for example, not automatically heating water in the morning prior to renewable heat collection.
Flat Plate Collectors
Flat plate panels normally comprise of a dark colour flat plate to absorb energy which is contained in an insulated box with a clear glass cover. Heat is transferred from the panel to a water storage cylinder within a domestic property by pumping a liquid, which normally contains a form of antifreeze, through tubes which are attached to the plate absorber. These types of panels tend to balance heat-loss inefficiencies through increasing physical dimensions and therefore collector surface area, however, in high latitude locations, cooler month heat-loss efficiency is far more relevant than in sunnier and warmer climes.
Evacuated Tube Collectors Evacuated tube panels use a cylindrical glass double wall construction incorporating a partial vacuum, to insulate an internal collector surface against ambient temperatures, thereby reducing heat-loss as a Thermos flask would. This collector technology generally performs well in lower light or intermittent irradiation conditions caused by clouds and is therefore better suited to use in high latitude installations such as in the UK, extending the exclusive solar thermal heating season towards the winter solstice from both sides of the calendar. In the deepest, darkest months of winter evacuated tubes will contribute towards energy savings more effectively than flat panels, the small amount of heat collected being used to pre-heat the cold water fed into the DHW system thus reducing the load on other water heating energy sources.
Seasonal Optimisation During the winter months, the optimal angle for your panels to receive energy will be steeper in order to compensate for the sun being lower in the sky. It's very rare for roof mounted solar PV or thermal systems to have the ability to optimise the collector angle, this normally being defined by roof angles and orientations, therefore forms of seasonal optimisation are normally limited to ensuring that the sunlight lower angle of incidence in winter months doesn't create unnecessary shade issues.
Where panels are to be installed on the ground or a flat roof, there are mounting systems which allow for manual adjustment of panel tilt into preset lockable position, normally every two or three months, in order to optimise generation. The overall efficiency of ground mounted solar PV panels can also be improved through the use of automated single or multi-axis tracking to maintain an optimal generation angle to the sun throughout each day of the year, however, within the UK domestic environment tracking arrays are subjected to considerable planning restrictions.
As panel prices have fallen considerably over the past few years, the viability of investing in optimisation has become more and more questionable. It is now generally considered to be cheaper to increase overall capacity to meet an annual generation target than to invest in manual or automated optimisation.
Conclusion
Taking the above into consideration, should we really have concerns about the seasonal variability inherent in renewable technologies based on collecting sunlight?. In effect, as long as the technology is doing it's primary job, reducing the need for fossil fuel or other forms of non-renewable consumption why should there be a concern at all as every little helps.
As we have seen, even in "... the short days of deepest winter" both solar PV and solar thermal systems have the ability to contribute a reasonable proportion of required energy during daylight hours as long as they're specified, designed and sized to meet demand, but in common with other energy saving technologies, their effectiveness depends heavily on the efficiency of the environment in which the energy will be consumed. The combination of some or all of these factors provide the only logical explanation for the EST field trial's wide range of reported performance. The majority of energy aware users of solar panels seem to be content as long as generation covers their household base-load for the majority of daylight hours and tend to shift high-load appliance or thermal demand to make best use of self generation or collection on sunny days, even in the winter, which invariably reduces winter peak evening electricity demand on the grid.
Read more about Solar PV & Thermal systems in the relevant menu sections
