When designing an off-grid solar energy system, one of the key considerations is determining the appropriate battery capacity. Battery storage ensures that you have a consistent energy supply even when the sun isn’t shining. Most of the time we are starting from scratch where we have limited data. We can look at historical energy usage for a client and assume some solar production figures from weather models. But ultimately we have to make some judgement calls to determine how much battery capacity is appropriate.
Oversizing a battery is great – if you have the budget, the space, and you can do it within building code limitations. For practical purposes, we need to make adjustments to stay within reasonable limits. This is often done using rules of thumb and anecdotal information from experience and previous clients. Ultimately, it’s about offering a great user experience while respecting budgetary constraints. The idea is to ensure there will be enough battery capacity and solar production to provide backup power on the vast majority of days when needed.
By examining daily solar production and consumption figures over an entire year and applying statistical analysis, we can make informed decisions about the ideal battery capacity for your needs.
Analyzing a Year’s Solar Production and Consumption to Optimize Off-Grid Battery Capacity
When you have great data, you can make great decisions. We took one client who has had solar panels for years and analyzed solar production and household consumption for a full calendar year (2023). We did this to simulate how we can approach sizing a battery if this client wanted to add battery backup to their existing solar energy system.
The data below comes from daily solar energy production and daily household electricity usage that we gathered from their Enphase Energy Enlighten web portal. The Enphase data monitoring allows you to download reports. We then analyzed the data statistically.
We took data from a single year (2023) to simplify the analysis. The site we used is known to be approximately net zero, meaning the homeowner produces roughly the same amount of solar energy per year as the electricity they consume in the household.
Yearly Solar Production and Consumption Data
For this analysis, we examined daily solar production and consumption data for an entire calendar year. Here’s a statistical summary of the net energy (production minus consumption) per day:
- Count: 365 days
- Mean (Average): 3.253 kWh
- Standard Deviation: 21.457 kWh
- Minimum: -119.130 kWh
- 25th Percentile: -8.430 kWh
- Median (50th Percentile): 4.890 kWh
- 75th Percentile: 17.610 kWh
- Maximum: 46.450 kWh
This data helps us understand the typical range of daily net energy and the variability we can expect over the year. The mean tells us that the homeowner produces with solar power a little more than 3 kWh per day than they consume. But that is just the average.
You could look at the 25th percentile, which tells us that there is a shortfall of solar energy production of 8 kWh per day 25% of the time. You could then provide extra battery capacity to reduce the impact of this shortfall so that enough battery capacity is virtually guaranteed 75% of the time. It would be unreasonable to size a battery for the minimum because that might only happen once in a blue moon. So conservation measures might be a better solution.
Visualizing the Data
To visualize this data, we created a histogram of daily energy consumption, daily solar energy production, and net energy, overlaid with a normal distribution curves and standard deviation lines. This helps illustrate the spread of daily net energy values and how they cluster around the mean.
– The green dashed line represents the mean.
– The purple, orange, and yellow dashed lines indicate the 1st, 2nd, and 3rd standard deviations from the mean, respectively.
Electricity Consumption
In this chart you can see the average consumption is around 55 kWh. We are mostly concerned with the number of days where the average consumption is greater than the average, because this is what we have to account for in a grid outage situation. There are not too many days above one standard deviation, or 70 kWh, and just a handful of days above 80 kWh that we have to consider.
Solar Energy Production
This chart shows the number of days for which each whole number of kWh of solar production was achieved during the year. In this data, the average is 58 kWh per day, and we are most concerned with the number of days when solar production is below this figure. There are not too many days where solar production is less than one standard deviation, or 41 kWh per day. This tells us that it is highly likely that we will be able to produce at lease this amount of energy on any given day.
Net Energy
Where things get really interesting is looking at the net energy production on a daily basis. As we already know, on average, we have 3 kWh excess of solar energy per day. By looking at one standard deviation, we can see that it is highly likely that the daily shortfall to surplus will fall in the range of -18 kWh to +25 kWh. It is probable that there will be days with an 18 kWh shortfall of solar energy production, but that could easily be followed by a day where solar production exceeds household consumption by 25 kWh. It is fairly unlikely that the result will fall outside this range.
Understanding the Implications
The mean daily net energy of 3.253 kWh gives us a central point around which most daily net energy values cluster. However, the standard deviation of 21.457 kWh indicates significant variability. On some days, the net energy can be as low as -119.130 kWh (indicating a deficit), while on other days, it can be as high as 46.450 kWh (indicating a surplus).
What this doesn’t tell us is how many days in a row we had bad weather, or how many days in a row we had excess solar production. It does tell us that there is a likely range of outcomes, however.
Determining Ideal Battery Capacity
To ensure reliable off-grid power, battery capacity should cover the expected daily energy needs, accounting for days with net energy deficits. Here’s how we can use this statistical analysis to determine the ideal battery capacity:
- Base Capacity on Mean Net Energy:
– A battery capacity around the mean daily net energy (3.253 kWh) ensures you have some storage for an average day. However, this doesn’t account for days with significant deficits. - Incorporate Variability:
– To handle days with net energy deficits, we can consider adding capacity based on the standard deviation. For instance, a capacity of 1 standard deviation above the mean (3.253 + 21.457 = 24.7 kWh) would cover most days, except for those with unusually large deficits. So we might choose a battery sized for the expected nighttime usage plus 24.7 kWh as an additional buffer. - Plan for Worst-Case Scenarios:
– For maximum reliability, design the system to handle the lowest observed net energy (-119.130 kWh) or even 2-3 standard deviations below the mean. This ensures the system can handle extended periods of net energy deficits. However, that would be cost-prohibitive and unreasonable. It would be better to determine why these outliers exist and eliminate them. - Factor in Energy Consumption:
– Match the battery capacity with your household’s daily energy consumption patterns. If your average consumption is 40 kWh per day, and considering the net energy statistics, a larger capacity (e.g., 60-80 kWh) may be required to ensure reliability during periods of low production.
Complicating Factors
The analysis is complicated by time of year. We have done the above analysis without regard to when an outage is most likely to occur, especially an extended outage. History tells us that in Southwest Florida, hurricanes in August and September are most likely to result in extended power outages. So, it is wise to hone in on these months to consider the worst-case scenario before settling on an appropriate battery capacity.
Another factor complicating the analysis is whether the homeowner will change their habits when operating off-grid. We have to ask what kinds of lifestyle changes are likely or acceptable to the homeowner. What things will be done to reduce energy consumption? For example, if we can turn off the pool equipment during extended outages or forego water heating, that could drastically change the analysis. Does the consumption data include electric vehicle charging, which is not advised when off-grid? That could further skew the analysis.
Realistically, will the homeowner even be present after a hurricane? Are they likely to bug out and get away from the home, meaning the electricity usage could be drastically lower when they are not present, especially if they turn things off before they leave? Or is the homeowner likely to hunker down and stay in their home during and after a storm?
The analysis also ignores costs and suitable pairings of inverters, solar panels, and batteries, which might shift the conversation.
Landing On The Right Number
Unfortunately, there is no “right” answer. We can use this excellent statistical data to determine likely outcomes, but we also have to take into account other factors. Once we determine the likely use case for the property during an extended outage and marry that with acceptable lifestyle changes and loads that we can ignore in a grid-outage scenario, the battery capacity needs may be more modest or may even increase. The statistical analysis will serve as a starting point, but the human factors will also play a large role.
I know people will comment on why we didn’t come up with a single number here if I don’t stick my neck out and give an answer. So I will attempt to placate those people. In a case like this, where there is ample solar power for the average needs of the homeowner, I would say that a battery of 40kWh would be a good starting point. With some minor lifestyle changes, omitting the swimming pool from the backed up loads, and being cautious with energy use depending on the weather, this client would have a good experience.
I know this owner to be fairly tech savvy, and willing to accept and implement a variety of energy conservation measures if a prolonged outage is expected. If it was a different client who was less willing to accept lifestyle changes, I would recommend a battery in the range of 60-80 kWh to meet their needs. It all comes down to expectations and understanding how the system works.
The great thing about modern battery backup systems is we can build them to be modular, and adding battery capacity at a later date is an option that does not cause significant re-engineering and cost aside from the additional batteries themselves. That’s why I mentioned 40kwh as a starting point.
Conclusion
By analyzing daily solar production and consumption data over an entire year and understanding its statistical properties, we can make informed decisions about the ideal battery capacity for your off-grid solar energy system. This approach ensures you have sufficient energy storage to maintain a reliable power supply, even during periods of low solar production and high consumption.
Designing an off-grid system requires careful consideration of both energy production and consumption patterns. Using statistical analysis to guide decisions helps optimize system performance and reliability, ensuring you can enjoy the benefits of solar power without interruptions.
However, there are complicating factors and human factors that are not considered in the raw data. This is where an experienced solar contractor can step in and explain how lifestyle changes, automatic load shedding, monitoring, and other factors can reduce the investment required to achieve energy independence in a grid outage scenario. I tell people frequently that there is quite a bit of art involved in designing systems. It’s not all math and science.
If you’re considering a new solar energy system with battery backup or need help analyzing your existing data to determine what it would take to add battery backup, we are here to help. Contact us today to learn more about how we can assist you in designing a system that meets your energy needs.
