How to Use the Self-Consumption Model in PVSyst | 5 Steps for Beginners

Table of Contents

• What can be checked with PVSyst's self-consumption model

• Step 1: Organize the PV plant's basic conditions

• Step 2: Prepare demand data and load profiles

• Step 3: Set self-consumption conditions in PVSyst

• Step 4: Check the key indicators in the simulation results

• Step 5: Compare multiple cases to evaluate the optimal option

• Common pitfalls in the self-consumption model

• Understanding site conditions affects the accuracy of self-consumption simulations

• Summary

What can be verified with PVSyst's self-consumption model

In solar power system design, there is an increasing need not only to assume that all generated electricity will be sold externally, but also to check how much can be used within the building or facility. In factories, warehouses, offices, commercial facilities, public facilities, agricultural facilities, and the like, the greater the overlap between daytime electricity demand and solar generation, the easier it is to reduce the amount of electricity purchased from outside. For this kind of design, simply calculating annual generation is insufficient. It is necessary to verify how much of the generated electricity is consumed on-site and how much flows out as surplus.

PVSyst's self-consumption model is a feature used for this kind of analysis. In a standard photovoltaic generation simulation, annual and monthly energy production are calculated based on solar irradiance, panel capacity, tilt angle, azimuth, loss conditions, and so on. In contrast, the self-consumption model overlays the facility's power demand onto that. In other words, it compares the generation-side output curve and the demand-side consumption curve on the same time axis to estimate the amount of energy available for use at the same time.

What beginners should first understand is that a self-consumption model is meant not only to look at the performance of the solar power system itself, but to assess how effectively it can be used at a given facility. Even with the same solar power installation, the self-consumption rate can vary greatly between a facility that uses most of its electricity during the daytime and one that uses most of its power at night. Also, in facilities with low loads on weekends or holidays, there may be large surpluses on those days even if self-consumption works well on weekdays.

Typical items to check in a self-consumption model include annual total power generation, the amount of electricity consumed on-site, the amount of surplus electricity exported off-site, the share of demand met by solar power, time-of-day surplus trends, and monthly self-consumption trends. Reviewing these makes it easier to judge whether the design capacity is too large or too small, whether the generation peak aligns with the demand curve, and whether there is scope to consider energy storage.

When using PVSyst, rather than chasing detailed numbers from the outset, it is important first to look at the relationship between "generation" and "demand." Even if annual generation is high, surplus will increase if it does not overlap with the facility's demand. Conversely, even if generation is somewhat low, if it matches daytime demand well, the design can be efficient for self-consumption. The self-consumption model is a simulation to support such practical decision-making.

Step 1: Organize the basic conditions of the power plant

Before using the self-consumption model in PVSyst, first organize the basic conditions of the photovoltaic power generation system itself. In discussions of self-consumption, demand data tends to attract attention, but if the generation-side conditions remain unclear, no matter how accurate the load data you enter is, the reliability of the results will not be high. It is important to first clarify the installation site, system capacity, module layout, racking conditions, azimuth, tilt angle, surrounding obstructions, wiring losses, temperature losses, soiling losses, and so on.

The installation site directly affects solar irradiance and temperature conditions. Even with the same installed capacity, annual energy production varies by region. Also, environmental conditions—such as mountainous areas, coastal areas, urban areas, and snow-prone regions—change the way losses are considered. In self-consumption models, because the temporal overlap with demand is examined, not only the annual total but also seasonal generation trends are important. Facilities with high demand in summer, facilities with high demand in winter, and facilities with steady demand throughout the year require different design decisions.

When deciding equipment capacity, it is important not to judge solely by the maximum capacity that can be installed on the roof or on the premises. If the goal is self-consumption, more generation is not necessarily better. If the system capacity is too large relative to the facility’s daytime demand, surplus power will increase and the self-consumption rate may not rise as much as expected. Conversely, if capacity is kept too low, demand may not be sufficiently met even during generation periods, and the benefits of installation may appear small.

Azimuth and tilt angle are also important factors in self-consumption design. An orientation close to due south tends to increase annual energy production, but if a facility’s demand peaks are skewed to the morning or afternoon, a mismatch with the generation peak can occur. For example, for a facility whose operations are concentrated in the morning, an eastward orientation may better match demand. Facilities with high afternoon power usage may benefit from a westward generation tendency. PVSyst allows you to compare these azimuth and tilt differences on a case-by-case basis, so when combined with a self-consumption model it is useful for practical decision-making.

Checking for obstructions is also essential. Shading from rooftop equipment, adjacent buildings, trees, fences, chimneys, handrails, and surrounding terrain not only reduces power generation but also affects the generation curve by time of day. In particular, morning and evening shading can change how generation overlaps with demand. For example, at a facility with high morning demand, if strong morning shading occurs, the expected self-consumption effect may be smaller than assumed.

At this stage it is not necessary to finalize the complete detailed design, but you should at least decide which conditions will be the standard case. When using the self-consumption model in PVSyst, if the generation-side conditions are not fixed it becomes difficult to tell whether changes in the results are due to the demand data or to the generation conditions. For beginners, it is easiest to first create one standard case and then create comparison cases by varying capacity, tilt, orientation, and loss conditions.

Step 2: Prepare demand data and load profiles

In a self-consumption model, the most important factor is the facility's demand data. In PVSyst's self-consumption calculations, because solar generation and demand are matched along the time axis, the accuracy of the load profile has a major impact on the results. A load profile is data that shows how much electricity a facility uses at different times of day. Total annual electricity consumption alone is insufficient for evaluating self-consumption.

For example, even for facilities with the same annual electricity consumption, the amount that can be self-consumed differs greatly between facilities that use more power during the daytime and those that use more at night. Because solar power generation basically produces electricity during the daytime, the greater the daytime demand, the easier it is to use the generated power on-site. Conversely, in facilities that operate mainly at night, the generation times of solar power tend to be misaligned with demand times, making surplus power more likely. This difference cannot be determined from annual consumption alone.

Ideally, demand data should be measured data that shows power consumption at regular time intervals. If you have data at hourly intervals or at finer intervals, it becomes easier to reflect the facility’s operating patterns. In practice, load profiles are created based on power metering data, incoming power data, facility management data, equipment operation records, and so on. The longer the data period the better, and at a minimum you need a range that can capture representative weekdays, holidays, and seasonal variations.

When measured data are not available, we may also consider using an assumed load profile. In this case, based on typical operating hours for each facility use, we set demand from morning through evening, nighttime base load, low load on holidays, seasonal HVAC loads, and so on. However, because assumed data introduce large uncertainty into the results, it is necessary to make clear in proposal documents and internal reviews that they are "assumed values rather than measured values."

When preparing load profiles, pay attention to units and time intervals. If you do not confirm whether the values represent energy or average power, or what the time interval is in minutes or hours, calculation results after input can be significantly off. You should also check date and time formats, how holidays are treated, missing data, and outliers. In particular, if measurement instrument outages or data gaps exist, using those periods as-is can lead to underestimating demand.

Also, in solar power self-consumption, not only the peak demand but also the minimum demand is important. In facilities where the daytime minimum demand is small, even a slight increase in solar generation can create a surplus. Conversely, facilities that have a constant base load during the day tend to be able to consume more of the generated power even if the system capacity is increased to some extent. When looking at load profiles, it is important to check not only peak demand but also the daytime minimum demand, variations on operating days, and demand drops on holidays.

Before using the self-consumption model in PVSyst, we recommend not loading the demand data as-is but graphing it and visually inspecting it. Check whether the peaks and troughs by time of day match the actual situation of the facility, whether demand on holidays or at night is not unnaturally high, and whether seasonal variations are reasonable. Performing this preliminary check lets you detect input errors or unit mistakes early. In self-consumption simulations, input errors on the demand side are more likely to be overlooked than those on the generation side, so it is important to proceed carefully here.

Step 3: Set self-consumption conditions in PVSyst

Once the basic generation-side conditions and the demand data are prepared, set the conditions for the self-consumption model in PVSyst. What beginners should be aware of is that deciding in advance "by what approach to evaluate self-consumption" is more important than the act of entering numbers on the settings screen. Simply inputting the demand data and looking at the results may not readily lead to design decisions.

First, clarify whether the target system assumes that the generated power is used preferentially within the facility, that surplus is exported externally, or that control is performed so that it is not exported externally. In self-consumption type solar power, the handling when generation exceeds demand is important. Whether surplus can be exported externally or cannot changes the optimal equipment capacity and evaluation metrics. Under the assumption that surplus is not exported, generation curtailment may occur, and if equipment capacity is made too large the amount of generated power that cannot be utilized may increase.

Next, prepare the demand data so it can be handled by PVSyst. Check the data format to be used, the ordering of timestamps, the time interval, and the units, and verify that the period covered is consistent with the period of the power generation simulation. For an annual simulation, it is desirable that the demand data also be representative of the entire year. If data are available for only part of the period, decide in advance whether to expand that data to a full year, treat it as representative day(s), or adjust it seasonally.

When configuring a self-consumption model, matching demand and generation is important. During periods when generation is lower than demand, the electricity produced is essentially self-consumed. During periods when generation is higher than demand, only the portion equal to demand is self-consumed, and the remainder becomes surplus or is subject to curtailment. Understanding this relationship makes it easier to interpret the self-consumption and surplus values displayed on the results screen.

Beginners often get confused about the balance between power generation system capacity and load conditions. For example, if you simply enter the maximum capacity that can be installed on the roof, annual energy production will increase, but the self-consumption rate may decrease. Conversely, if you reduce capacity to match demand, the self-consumption rate may appear higher, but total generation and the amount of externally purchased electricity that can be reduced may become smaller. In other words, maximizing only the self-consumption rate is not necessarily the optimal design.

When using PVSyst's self-consumption model, it's easier to understand if you first create a baseline case and then make multiple cases with different system capacities. For example, comparing three capacities—smaller, standard, and larger—lets you see how much self-consumption increases as capacity grows and at what point surplus generation becomes significant. In addition, adding cases with different orientations and tilts lets you assess whether the generation peak times align with demand.

When configuring the system, whether to include energy storage should also be considered. If storage is present, surplus power generated during the daytime can be used at other times, affecting the self-consumption rate and the amount of electricity purchased externally. However, including storage increases the number of conditions to consider, such as charge/discharge efficiency, capacity, power output, control strategy, and discharge time periods. For beginners, it is easier to first understand the basic relationship between generation and demand using a self-consumption model without storage, and then consider cases that include storage.

Also, when assuming generation curtailment or output limits, you should clarify that approach as well. Whether you treat all generation that exceeds demand as surplus, or truncate anything above a certain threshold as unusable, will change the results. For facilities that cannot export power externally, if you do not set the handling of surplus electricity to reflect reality, you may overestimate the benefits of implementation.

Step 4: Confirm the metrics to check in the simulation results

Once the self-consumption model is set up, review the simulation results. The important thing here is not to make a judgment based solely on annual generation. For self-consumption applications, in addition to having a large amount of generation, you need to see how much that generation overlapped with demand. On PVSyst’s results screen, comprehensively check generation, demand, self-consumption, surplus, the equivalent amount purchased from external sources, and monthly and time-of-day trends.

First, what we want to confirm is the annual electricity generation. This is the basic indicator showing how much power a solar PV system produces over the course of a year. However, under a self-consumption model, annual generation does not directly equate to the amount of electricity that can be used on-site. If there is no demand at the time electricity is generated, that portion can become surplus or be curtailed. Therefore, annual generation should be viewed only as the overall upper limit.

Next, check the self-consumption amount. This is the amount of the electricity generated that is considered to have been actually used within the facility. In self-consumption–oriented designs, this figure is extremely important. If you increase installed capacity but the self-consumption amount does not increase much, a large portion of the added capacity may be surplus. Conversely, if the self-consumption amount increases in line with the capacity increase, it indicates that the facility has sufficient demand and is effectively using the generated power.

The self-consumption rate is also a commonly checked metric. It shows the proportion of generated electricity that was consumed on-site. A higher self-consumption rate makes it appear that the generated power is being used without waste. However, judging solely by the self-consumption rate is risky. Because reducing installed system capacity tends to raise the self-consumption rate, you also need to consider whether the generation volume and the amount of externally purchased electricity that can be reduced are sufficient.

The concept equivalent to the demand coverage rate is also important. This is an indicator that shows how much of a facility’s electricity demand was met by solar power generation. Even if the self-consumption rate is high, if the installed capacity is small, the contribution to the facility’s overall demand may be small. Conversely, even if the self-consumption rate is somewhat low, solar power may greatly reduce the amount of electricity purchased from external sources. In practice, it is important to consider both the self-consumption rate and the demand coverage rate to assess the balance.

Always check the surplus electricity as well. If the surplus is large, the installed capacity may be too big relative to demand. However, having a surplus is not necessarily a bad thing in itself. If the surplus can be exported externally, it can have a certain value. Also, when anticipating future demand increases, changes in operating hours, the addition of storage equipment, or the introduction of electric equipment, it is common to design with some allowable surplus. The important thing is to understand why the surplus is occurring.

Don't overlook monthly results. Solar power generation varies with the seasons. Facility demand also fluctuates seasonally. In facilities with large air conditioning loads, demand can increase in summer or winter, and in facilities with closure periods, demand can fall in specific months. Viewing data by month lets you identify mismatches that are not visible in annual totals.

Time-of-day trends are also important. In a self-consumption model, the essence is the overlap between generation and demand. Checking whether surpluses are concentrated around midday, whether demand shortfalls occur in the morning and evening, or whether surpluses are high only on holidays will lead to the next design improvements. For example, if there is a large midday surplus, possible measures include reviewing equipment capacity, considering battery storage systems, shifting load operation times, and adjusting control strategies.

When interpreting results, the shape of the graph also matters. Numbers alone can make it difficult to see the temporal relationship between demand and generation. By overlaying the generation curve with the demand curve, you can intuitively understand which time periods are covered by self-consumption, where surpluses occur, and where external purchases are required. When operational staff make internal or customer presentations, using representative-day graphs and monthly trends, not just annual total figures, makes explanations easier.

Step 5: Compare multiple cases and determine the best option

PVSyst's self-consumption model does not end with calculating a single case. What matters in practice is exploring design options that meet your objectives while comparing multiple conditions. In solar power system design there are several factors that affect the results, such as system capacity, azimuth, tilt, loss assumptions, demand profile, handling of surplus, and the presence or absence of storage. Changing all of these at once makes it difficult to identify the causes, so it is important to decide the order of comparisons and proceed accordingly.

The easiest thing to compare first is the installed capacity. Using the standard case as a baseline, create cases with smaller and larger capacities. Then compare annual generation, self-consumption, self-consumption rate, surplus, and demand satisfaction rate. As capacity increases, generation increases, but self-consumption does not necessarily increase in the same proportion. Beyond a certain capacity, surplus can suddenly increase and self-consumption efficiency may decline. Identifying this tipping point is important when considering installed capacity.

Next, compare differences in azimuth and tilt angles. A layout that maximizes annual energy production is not necessarily optimal for self-consumption. For facilities with high demand in the morning, high demand in the afternoon, or reduced demand during lunch breaks, shifting the generation peak slightly may better match self-consumption. By comparing multiple cases in PVSyst, you can check not only differences in generation but also differences in overlap with demand.

Differences in demand profiles should also be compared. If a facility’s operating hours may change in the future, it is practical to develop scenarios that assume future demand as well as current demand. For example, installing new equipment, extending operating hours, increasing holiday operation, upgrading HVAC systems, or adding electric equipment can increase daytime demand. If future demand growth is anticipated, sizing capacity based only on the current situation can later prove insufficient.

The treatment of surplus electricity should also be included in case comparisons. The assessment differs greatly between conditions where surplus can be exported externally and where it cannot. Under conditions where export is not possible, the more surplus there is, the greater the amount of generation that cannot be utilized, so the optimal equipment capacity may be somewhat smaller. Conversely, when export is possible, even with a certain amount of surplus the overall benefits of deployment may still hold. In self-consumption models, it is important not to leave this assumption ambiguous.

When considering battery storage, compare the case without storage to the case with storage. With battery storage, daytime surplus can be used in the evening and afterward, so on-site consumption may increase. However, because battery storage involves charge/discharge losses, not all surplus power will be effectively utilized. Also, if capacity is too small it may not absorb all surplus, while if capacity is too large there may be more periods when it cannot be fully used. In comparisons that include storage, you should check not only the self-consumption rate but also charge/discharge amounts, unutilized surplus, and reductions in purchases from external sources.

When organizing comparison results, decide which indicator to prioritize. The optimal option changes depending on whether you want to raise the self-consumption rate, reduce external purchases, curb surplus, or accommodate future demand. For example, prioritizing the self-consumption rate tends to result in smaller equipment capacity, while placing more importance on reducing external purchases may call for larger capacity. If you compare options while the objective is unclear, it becomes difficult to judge which option is best.

In practice, rather than presenting only a single final design proposal, it is important to be able to explain the differences among multiple proposals. Laying out cases with different approaches—such as a standard option, a surplus-suppression option, a generation-focused option, and a future-demand-response option—makes it easier for stakeholders to make decisions. PVSyst’s self-consumption model is useful as baseline documentation for carrying out such comparative evaluations.

Common Pitfalls of the Self-Consumption Model

When using the self-consumption model in PVSyst, the mistake beginners most often make is judging based only on annual totals. Comparing annual generation and annual demand and assuming there's no problem because demand is larger is risky. What matters for self-consumption is whether generation and demand overlap in the same time periods. Even if annual demand is sufficiently large, if demand is mainly at night, daytime generation may not be usable as-is.

Another common mistake is failing to verify the representativeness of demand data. Using data from a period with unusually low operation, or treating data from special operating days as typical days, will cause results to diverge from reality. In factories and facilities, electricity consumption varies with season, day of the week, production volume, operating schedule, holidays, equipment inspections, and so on. You must confirm that the demand data entered into the self-consumption model is representative of the facility’s normal operation.

Unit errors are also a very common point of caution. Mistakes such as confusing energy with power, using incorrect time intervals, misplacing digits in demand data, or failing to notice time shifts can have a major impact on the results. In particular, if the timestamps of the demand data are shifted relative to the timestamps of the generation simulation, the overlap between generation and demand may be calculated incorrectly. After inputting the data, be sure to check the representative day's graph and confirm that the daytime and nighttime demand patterns match reality.

It is also problematic to leave the handling of surplus power ambiguous. Whether surplus can be exported externally, cannot be exported, or is subject to restrictions affects the optimal system capacity. If calculations assume surplus can be used but in reality it cannot be exported, the expected benefits of the installation may be overstated. Conversely, if calculations are performed assuming surplus is not taken into account at all, the results can be more conservative than reality.

Also, you should avoid trying to explain simulation results with a single number. If you simply conclude that a high self-consumption rate is good, that a small surplus is good, or that a large annual generation is good, you may stray from the design objectives. In practice, it is necessary to look at generation, self-consumption amount, reduction in external purchases, surplus amount, monthly trends, and time-of-day trends together. When explaining to stakeholders, showing the reasons for your decisions using multiple indicators increases their level of acceptance.

Another point to be careful of is ignoring future changes. Solar power generation systems are used for long periods. Demand may change not only because of current needs but also due to future equipment expansions, changes in operating hours, energy-saving measures, electrification, air-conditioning updates, and so on. A design that fits current demand exactly may not be optimal a few years later. Because PVSyst allows you to create multiple cases, separating current and future scenarios for consideration will lead to more practical decision-making.

Understanding local conditions determines the accuracy of self-consumption simulations

PVSyst's self-consumption model can be very useful for analysis if the input conditions are well organized. However, the accuracy of the results is not determined solely by the settings in the software. What matters is how accurately you can grasp the actual site conditions. In solar power generation simulations, topography, building shape, roof height, orientation, tilt, obstructions, available installation area, and constraints around equipment all affect power generation. If these remain ambiguous, the results of the self-consumption model are also likely to deviate from reality.

When installing on rooftops or around facilities in particular, there are many conditions that drawings alone cannot reveal. Rooftop equipment, piping, railings, adjacent buildings, signs, trees, changes in elevation, fences, delivery access routes, and maintenance spaces can be overlooked unless checked on site. These affect panel layout and shading conditions, and consequently the power generation curve. In self-consumption models, because the overlap between the generation curve and the demand curve is examined, time-of-day declines in generation caused by shading are important.

Also, it is important to grasp the installable area. Even if the drawings appear to show sufficient space, there may in reality be equipment to be avoided or required safety clearances that prevent installing the assumed capacity. Conversely, by accurately measuring the site you may find a more efficient layout than originally anticipated. If the assumptions about equipment capacity change, self-consumption, surplus, and the demand fulfillment rate will also change.

In on-site surveys, the accuracy of positioning and recording is also important. If the range of potential installation areas and the positions of obstructions can be determined precisely, modeling and case comparisons in PVSyst become easier. Photos and notes alone can make it difficult to reconstruct spatial relationships later. Especially when design review involves multiple stakeholders, being able to share on-site positional information and point cloud data helps reduce discrepancies in judgment.

In on-site assessment situations like this, using an iPhone-mounted high-precision GNSS positioning device such as LRTK makes it easier to record candidate installation sites and surrounding conditions together with high-precision location information. If coordinates, photos, point clouds, and other data collected on site can be used in design reviews, it becomes easier to verify the assumptions to be entered into PVSyst and leads to improved reliability of self-consumption simulations. In studies of self-consumption for solar power generation, it is important not only to operate software but also to correctly measure, record, and share on-site conditions with stakeholders.

Summary

When using the self-consumption model in PVSyst, you can check not only how much a photovoltaic system will generate but also how effectively that generated power can be used within the facility. For self-consumption-oriented design, it is important to comprehensively consider not only annual generation but also the temporal overlap with demand, the amount of self-consumption, surplus energy, the demand satisfaction rate, monthly trends, and time-of-day trends.

For beginners, it is easier to understand if you proceed in this order: organize the plant’s basic conditions, prepare the demand data, set the self-consumption conditions in PVSyst, check the result indicators, and compare multiple cases. In particular, the accuracy and representativeness of the demand data have a large impact on the results. Rather than judging by annual consumption alone, checking the load profile by time of day is the first step to using a self-consumption model correctly.

Also, when interpreting simulation results, it is important not to focus solely on the self-consumption rate. Even if the self-consumption rate is high, if the system capacity is small, the effect on reducing purchases from external sources may be limited. Conversely, even with a somewhat lower self-consumption rate, a design can be advantageous in terms of total power generation or demand coverage rate. Practically, you should clarify which metrics to prioritize according to the design objectives and compare multiple cases.

PVSyst's self-consumption model is a powerful tool for considering the connection between the generation side and the demand side. However, its results are influenced by the quality of the input conditions. Correctly understanding on-site installation conditions, shading/obstructions, roof shape, azimuth, tilt, and the available installation area, and carefully organizing demand data, greatly enhances the credibility of the simulation. Accurately capturing on-site information, not just desk calculations, is indispensable in the design of self-consumption solar PV systems.

If you want to improve the accuracy of power generation simulations, it is important to accurately acquire location information from the site-survey stage and establish a system that can reflect it in the design. LRTK, as a GNSS high-precision positioning device that can be attached to an iPhone, can be used on site for coordinate acquisition, photo recording, point cloud acquisition, and checking candidate installation areas. As a preliminary step before evaluating a self-consumption model in PVSyst, recording site conditions with high accuracy makes it easier to organize assumptions about system capacity and shading conditions, and helps lead to a solar power plan that is closer to reality.