5 Methods for Calculating Solar Power Generation Assuming Self-Consumption

Table of Contents

• Why calculations assuming self-consumption differ from standard generation calculations

• Method 1: Estimating self-consumption from annual generation

• Method 2: Calculate by overlaying monthly generation and consumption

• Method 3: Calculate by matching time-of-day loads and generation

• Method 4: Adjust the self-consumption ratio using measured data

• Method 5: Calculate by comparing optimal capacities across multiple scenarios

• Common mistakes when calculating assuming self-consumption

• How practitioners should proceed to improve the accuracy of self-consumption calculations

• Summary

Why calculations that assume self-consumption differ from typical power generation calculations

When calculating solar power generation, people often start by determining how much will be generated in a year in kWh. If you know the system capacity, you can estimate the annual total generation by multiplying it by the regional reference generation. This approach is correct in itself and is useful for comparing system sizes. However, when assuming self-consumption, that alone is not sufficient, because for self-consumption it is more important to know how much can be used at which times than the total amount of electricity generated.

For example, even if a system generates 10,000 kWh per year, that does not mean all of it can be self-consumed. Facilities or homes that use a lot of electricity during the daytime tend to have higher self-consumption rates and smaller surpluses. Conversely, facilities with low daytime usage have lower self-consumption rates, and the same generation can lead to more electricity sold or surplus. In other words, in calculations assuming self-consumption, you need to look at the temporal distribution of consumption as much as the generation itself.

What I want to clarify first is that generation, self-consumption, and electricity sold are separate figures. Generation is the total amount of electricity produced by the equipment. Self-consumption is the portion of that which the building or facility uses on-site. The portion that cannot be used and remains becomes the electricity sold or surplus electricity. In terms of an equation, the surplus is what remains when you subtract self-consumption from generation. However, in practice, if you do this subtraction roughly on an annual, lump-sum basis, it is easy to overlook seasonal and time-of-day variations.

Also, in calculations of self-consumption, increasing the system capacity does not necessarily make it more advantageous. A larger system will increase generation, but if the extra output cannot be used during the daytime, the self-consumption rate will fall. As a result, the expected economic benefit may not increase as much as anticipated. Conversely, even a slightly smaller system capacity can achieve higher self-consumption efficiency if it aligns well with the timing of demand. In other words, when assuming self-consumption, it is essential to consider not only the amount of generation but also how well it overlaps with demand.

Therefore, when calculating solar power generation on the assumption of self-consumption, it is more practical not to stop at the annual total but to also adopt perspectives such as monthly breakdowns, hourly breakdowns, adjustments based on actual performance, and comparisons across multiple scenarios. In this article, we organize that approach into five methods and explain which method should be used at each stage.

Method 1: Estimate self-consumption from annual power generation

The simplest method is to estimate self-consumption from the annual electricity generation. This is very convenient as a first step when you begin considering self-consumption. The idea is to first determine the annual generation, then assume a self-consumption rate and calculate the self-consumption amount. In terms of the formula, you obtain an estimate of self-consumption by multiplying the annual generation by the self-consumption rate.

For example, suppose the system capacity is 10 kW and the annual generation, taking local conditions into account, is expected to be 10,500 kWh. If you assume a self-consumption rate of 40%, the self-consumed amount is 4,200 kWh. The remaining 6,300 kWh is expected to go to surplus or be sold. The advantage of this method is that it requires few input figures and makes it easy to grasp the overall picture in a short time. It is very useful for comparing system sizes, initial consultations, and confirming direction within the company.

However, this method is only a rough initial estimate. Because it represents the self-consumption rate as a single number, it significantly averages out seasonal and time-of-day differences. The difference between weekdays with high daytime loads and holidays with almost no usage is also not visible. In other words, even if you assume a self-consumption rate of 40% or 50%, if it remains unclear under what conditions that number was derived, discrepancies with actual on-site conditions are likely to arise later.

What makes this method important is that it provides a very clear entry point for calculations based on self-consumption. Because it fosters the habit of separating self-consumption from surplus, as well as considering generation output, evaluations of system size become considerably more practice-oriented. It also makes it easier to roughly grasp the relationship between self-consumption and surplus when comparing multiple capacity options.

For practitioners, this method is suitable for an initial comparison. When detailed load data or monthly data are not yet available, it is realistic to start here and later improve accuracy with the next method. Rather than stopping at annual power generation alone, applying a self-consumption rate even once has significant meaning for estimates that assume self-consumption.

Method 2: Calculate by overlaying monthly power generation and consumption

The second method is to calculate by overlaying monthly generation and usage. When assuming self-consumption, annual aggregate figures often make it difficult to see the real situation, so simply lining up monthly generation and monthly electricity usage can significantly improve accuracy. Even if the annual generation is sufficient, differences such as shortages in the months when power is needed or unevenly concentrated surpluses become much clearer when broken down by month.

For example, in May power generation is high and cooling load is still not very strong, so surpluses may be more likely. In August generation is also large, but if daytime cooling demand is high the self-consumption rate may increase. In winter generation falls while heating and hot-water usage may increase. If you estimate using only annual averages without looking at these month-to-month differences, you may overestimate or, conversely, underestimate the self-consumption rate.

With this method, the monthly power generation is calculated first. Multiply the system capacity by the month’s average generation-equivalent hours and the number of days in the month, and after applying any necessary corrections you can obtain an estimate of the monthly power generation. Then compare that with the month’s daytime consumption and total monthly consumption to see how much will be used for self-consumption and how much will be surplus. The important point here is that the character of electricity use differs by month. For facilities like factories and offices that have high weekday daytime loads and for residences that are often unoccupied during the day, the same monthly power generation has different implications.

An advantage of viewing data by month is that it makes it easier to explain the value of the installation by season. You can organize it as: plenty of surplus in spring, increased self-consumption in summer, and reduced generation that is insufficient in winter, so the financial and operational outlook becomes much more concrete. This method is especially valuable for projects premised on self-consumption, where seasonal differences in electricity usage are often large.

Also, if you overlay them by month, you can see in which months power generation shortages or surpluses are concentrated. As a result, it becomes easier to determine whether you should increase system capacity, whether the current capacity is sufficient, or whether improvements can be made through operational adjustments. It requires a little more effort than using annual values, but for calculations assuming self-consumption it is a highly practical method.

Method 3: Calculate by Matching Load and Generation by Time of Day

The third method is to calculate by matching load and generation by time of day. If you truly want to improve the accuracy of self-consumption, this way of thinking is very important. That is because self-consumption is not determined solely by the monthly total, but by how much the times when generation occurs overlap with the times when consumption occurs. Looking month by month is closer to the reality of self-consumption than looking at annual totals, and looking by time of day is closer still.

For example, even if the monthly generation is the same 500 kWh, the ease of self-consumption differs between an installation where most of it is concentrated around noon and one where generation is widely distributed from morning through afternoon. Furthermore, on the demand side it varies greatly depending on whether a facility operates from the morning, a facility has high loads in the afternoon, or a residence is often unoccupied during the daytime. In other words, when seriously calculating self-consumption, temporal overlap is more important than the total amount of generation.

With this method, you look at how much the installation is likely to generate during each period of the day and overlay that with the periods of demand. For example, east-facing systems tend to be stronger in the morning, while west-facing systems tend to be stronger in the afternoon. South-facing systems tend to concentrate around midday. By overlaying the demand-side load profile on this generation profile, you can see the periods when self-consumption is possible and the periods when a surplus will occur.

In practice, when you can take the analysis this far, the value of an installation becomes fairly clear. If you look only at annual kWh, south-facing installations may have the advantage in total. However, considering the timing of demand ramp-up and the overlap with end-of-business hours, east-west distributed installations can be more favorable for self-consumption. In other words, time-of-day calculations can not only improve accuracy but also change the evaluation of the system configuration itself.

Of course, time-of-day calculations are time-consuming. However, for projects with large-scale installations or projects that prioritize the self-consumption rate, that effort is well worth it. In particular, for facilities that have demand time-of-day data, simply using this method can significantly change the way the financials look. For calculations based on self-consumption, it is one of the most practical methods.

Method 4: Calculate by adjusting the self-consumption rate based on measured data

The fourth method is to calculate by correcting the self-consumption rate using measured data. If there are existing facilities or if similar facilities have records of power generation and electricity usage, using those to correct the self-consumption rate greatly improves the accuracy of the estimates. This is because site-specific conditions that are difficult to capture through desk calculations alone are already included in the actual records.

For example, even in projects where the theoretical annual power generation was expected to be 20,000 kWh and the self-consumption rate was thought to be around 50%, actual measurements sometimes showed a self-consumption rate of only 35%. This gap arises from a buildup of factors that are difficult to capture on paper, such as misalignment with demand time periods, low holiday operation, the time-of-day characteristics of the equipment, shading, and high-temperature conditions. If these actual performance values are reflected in the next estimate, the forecast will be much closer to the on-site reality.

Also, using measured data makes it easier to identify where you were overestimating. If the self-consumption rate is high only in summer, it points to cooling loads; if it is low only in winter, it points to reduced generation; if it is high only on weekdays, it indicates a coincidence with operating hours. In this way, the context behind the numbers becomes clear. You come to understand that the self-consumption rate is not just a simple percentage but a figure that reflects the site’s actual operation.

The strength of this method is that theoretical values can be adjusted to match the actual site. While annual or monthly coefficients may be sufficient for preliminary studies, when it comes to deciding whether to implement a system or evaluating return on investment, it's more reassuring to base estimates as closely as possible on actual performance. In particular, when considering deployment in other buildings on the same site or expansion of existing equipment, it would be a waste not to use measured data.

For practitioners, measured data is one of the most powerful inputs for calibration. That's because it provides numbers showing not only the theory but how it was actually used. If you want to improve calculation accuracy under a self-consumption assumption, the quickest way is to incorporate actual performance records as much as possible.

Method 5 Compare and calculate the optimal capacity across multiple scenarios

The fifth method is to calculate by comparing optimal capacities across multiple scenarios. In estimates that assume self-consumption, it is very important not to produce a single value for generation and stop there, but to see how the self-consumption rate and surplus change when the installed capacity is varied. This is because increasing the installed capacity will raise generation, but the self-consumption rate does not necessarily increase.

For example, with a 5 kW system, annual power generation may be relatively low, but the self-consumption rate might be high. Increasing to 8 kW will raise annual generation, but surplus may also begin to increase. If you increase it further to 10 kW or 20 kW, generation will increase even more, but there may be many periods when output exceeds daytime demand, and the self-consumption rate could fall. In other words, the absolute amount of generation and the self-consumption rate do not necessarily move in the same direction.

When assuming self-consumption, it's better to break the analysis into scenarios by system capacity and compare them. For example, place three systems of 5 kW, 8 kW, and 10 kW side by side and compare annual generation, self-consumed energy, surplus energy, and the self-consumption rate. This makes it easier to see whether simply installing a larger system is advantageous or whether a medium-sized system offers the best balance. If you install only one system capacity without this comparison, it becomes difficult to determine whether it is truly optimal.

Also, this method is very well suited for equipment proposals and internal explanations. This is because it allows you to explain why a given capacity is reasonable not merely by equipment cost or intuition, but from the balance between the self-consumption rate and surplus. For example, setting it to 10 kW increases generation but also increases surplus, and you can conclude that 8 kW is easier to manage when assuming self-consumption. Such explanations tend to increase stakeholders’ confidence in the equipment.

When performing calculations that assume self-consumption, it is more practical to compare multiple system capacities to find the optimal outcome than to look for the “right” answer from a single capacity. Consider the three together: total generation, self-consumption rate, and surplus. Simply adopting this way of thinking will make decisions about system size much clearer.

Common mistakes when calculating based on self-consumption

One common mistake in estimates that assume self-consumption is to look only at total generation and conclude that the economic benefit will automatically be large. While it is true that increasing installed capacity tends to raise generation, if the additional electricity cannot be used up during the daytime the self-consumption rate will drop. Bigger systems are not necessarily more advantageous. This misunderstanding is particularly likely to occur in households and small-scale facilities.

Another common mistake is to assume a single fixed self-consumption rate. For example, if you casually set it at around 50%, you omit the usage patterns specific to that facility or residence. In reality, it can vary considerably by month, by day of the week, and by time of day. Even when looking at it on an annual basis, if you don’t know under what conditions that figure was derived, it’s hard to explain.

Also, thinking of the impact of orientation and shading as only a matter of energy output is a mistake. When self-consumption is assumed, the time of day when generation occurs is important, so differences between east- and west-facing orientations and the times when shading occurs directly affect the self-consumption rate. Even if the annual kWh changes only slightly, if the periods that are convenient for self-consumption are reduced, the economic value can change significantly.

Furthermore, it is risky to discuss self-consumption without looking at demand data. Even if the power generation calculation is done carefully, if daytime usage is unknown, estimates of the self-consumption rate will be rough. In simulations that assume self-consumption, it is necessary to always consider the generation side and the demand side together as a set. The fact that it cannot be concluded by discussing only generation is a major difference from ordinary power generation calculations.

To prevent such mistakes, it is effective to first consider generation, self-consumption, and surplus separately, and to break the values down not only by annual totals but also by month and by time of day. Under a self-consumption assumption, the way they overlap is more important than the total amount alone. Simply adopting this way of thinking makes the estimates much more practical.

How practitioners should proceed to improve the accuracy of self-consumption calculations

If practitioners want to improve calculation accuracy for self-consumption, it's important to start with a simple annual estimate, then move on to monthly, time-of-day, and measured-data adjustments. Jumping into detailed analysis from the outset makes the workload too large, but stopping at annual values alone doesn't reveal practical usability. Gradually increasing accuracy in stages is the most realistic and robust way to proceed.

First, derive the generation input from the system capacity and the region's annual coefficient. Then, by assuming a rough self-consumption rate, grasp the overall figures for generation, self-consumption, and surplus. Next, overlay monthly generation and monthly usage to check which seasons have strong generation and which seasons have large surpluses. If necessary, further examine the overlap between load by time of day and generation to make the self-consumption rate more site-specific.

If possible, using performance data from existing facilities or similar projects greatly improves accuracy. Even if something looks good in theory, the self-consumption rate can turn out low in practice due to the ramp-up of daytime demand or holiday operation. Conversely, daytime demand can be larger than expected, resulting in a higher self-consumption rate. These differences only appear in actual measurements. In other words, performance data is a very strong corrective factor for self-consumption calculations.

Moreover, the accuracy of on-site conditions is also important. If the candidate equipment locations, surrounding obstacles, roof surface orientation, and elevation differences are unclear, estimates of the power generation time windows will also be coarse. Especially when assuming self-consumption, the times when shadows occur are highly significant, so the accuracy of positional relationships directly affects estimation accuracy. In other words, accurately capturing site conditions is as important as refining the formulas.

Summary

When calculating solar PV generation on a self-consumption basis, five methods are practical in real-world applications: estimating self-consumption from annual generation, overlaying monthly generation and usage, matching load and generation by time of day, adjusting the self-consumption rate with measured data, and comparing optimal capacities across multiple scenarios. Each method serves a different role, and it is easier to use if you rely on simple methods in the initial stages and shift toward detailed methods during the decision-making stage.

What’s important in calculations assuming self-consumption is not just the total amount of generation, but also when and how much of that electricity is used. By paying attention not only to annual kWh but also to how monthly and time-of-day patterns overlap, the value of the system becomes much more concrete. It must not be overlooked in these calculations that increasing system capacity is not necessarily always advantageous, and that the balance between the self-consumption rate and the amount of surplus energy is crucial.

Also, to truly improve the accuracy of the estimate, it is essential not only to rely on desk formulas but to accurately understand on-site conditions. If the candidate equipment locations, the positions of surrounding obstructions, elevation differences, or the orientation of roof surfaces are unclear, forecasts of generation time periods and projections of the self-consumption rate will be rough. In particular, when and how shadows fall is critically important for estimates based on self-consumption.

In that regard, the iPhone-mounted GNSS high-precision positioning device LRTK is extremely effective as a means of accurately determining on-site positional relationships. Because it makes it easier to accurately record candidate equipment locations and the positions of surrounding obstructions on site, it facilitates moving to self-consumption-based estimates that take shading and placement conditions into account. If you want solar power generation figures to be truly usable on a self-consumption basis, accurately capturing site conditions with a method like LRTK is a major advantage.