5 items to review in simulation conditions when power output is low

When a solar power system’s generation is lower than expected, it can be tempting to immediately suspect equipment failure or poor installation. However, before comparing actual generation with the simulation values, the first thing to check is the simulation conditions themselves. Generation simulations are calculated by combining multiple assumptions such as solar irradiance, tilt angle, azimuth, shading, loss rate, system configuration, and operating conditions. Therefore, if those assumptions differ from the actual site, the generation may appear to be “low” even when there are no major problems with the equipment.

In practice, it is common for the design-stage simulation, the sales presentation materials, the post-construction generation performance, the generation figures shown on monitoring screens, and the numbers in monthly reports to be treated under different assumptions. If you make judgments without aligning the conditions of the values being compared, it becomes difficult to isolate causes and can lead to unnecessary on-site checks or incorrect improvement decisions.

This article explains, in five items, the simulation conditions you should review before proceeding to equipment inspection when you feel "power generation is low". To help practitioners more easily compare field data with simulation values, we lay out the order of checks, commonly overlooked assumptions, and considerations when making judgments.

Table of Contents

• Use consistent units and time periods for the power generation being compared

• Reassess the assumptions for solar radiation and weather conditions

• Reconfirm the conditions for orientation, tilt angle, and mounting surface.

• Reassess the loss conditions for shadows, dirt, and the surrounding environment.

• Confirm the conditions of the power conditioner and wiring losses.

• Summary

Standardize the units and time periods of the compared power generation

When you feel the power generation is low, the first thing to check is the units and the period of the figures you are comparing. When placing simulated values and actual results side by side, even though they appear to be the same “power generation,” the underlying assumptions can differ — daily, monthly, yearly, per installed capacity, per PV capacity, based on AC output, etc. If these are not aligned, you cannot accurately assess the condition of the power generation equipment.

For example, even if you think you are comparing monthly power generation, a discrepancy is natural if the simulation uses monthly estimates based on long-term average (typical-year) values while the actual results are influenced by the weather of a specific year. Also, even within the same month, the meter-reading period and the monitoring data aggregation period may not match. Comparing a value aggregated for the calendar month from the 1st to the end of the month with a value from the meter-reading date to the day before the next meter reading mixes differences in the number of days and weather. Before investigating causes of low generation, it is important to first confirm that the periods being compared are truly the same.

Differences in units are also a common oversight. The amount of generated energy is often expressed in kWh, but when comparing sites with different system sizes or multiple systems, it’s easier to judge by generation per kW. However, the result will look different depending on whether the kW used here refers to the PV module capacity or the rated output of the power conditioner (inverter). If you mix up generation based on PV module capacity with generation based on the AC output-side capacity, you can end up with different evaluations for the same system.

In materials that describe simulation conditions, the expected annual power generation is sometimes displayed prominently. However, when confirming a drop in power generation in practice, it is insufficient to judge based solely on the annual value. Some regions generate more in spring and summer and less in winter, while others are susceptible to the effects of the rainy season, snowfall, typhoons, or seasonal shading. Even if there is little difference on an annual basis, if a specific month is low, one should suspect influences such as weather, shading, soiling, or equipment outages. Conversely, a figure that looks low on a monthly basis may not indicate a significant problem on an annual basis.

Be careful when making comparisons by day or time of day. Power generation simulations are often estimates based on average weather data and do not fully reproduce the actual day‑to‑day weather. Naturally, generation varies between sunny, cloudy, rainy, and partly cloudy days. A single day's actual output falling below the simulation does not immediately indicate an anomaly. In practice, it is important to compare multiple days with similar weather or to look at trends on a weekly or monthly basis rather than relying on a single day.

Also, it is important not to confuse the amount of electricity sold with the amount generated. In systems with self-consumption, part of the generated power is used on-site, so looking only at the amount sold can make the generated output appear low. The same applies when battery storage, output control, remote shutdown, or the operating status of on-site loads are involved. Comparing values without confirming whether the simulation assumes power at the generation terminal, expected sold electricity, or values net of consumption may lead to misdiagnosing the cause.

Furthermore, the values displayed on the monitoring screen do not necessarily match the figures used in reports. The monitoring screen may show the AC energy for each power conditioner, or it may aggregate values from meters or from the incoming power point. On days when there were communication losses, or in months when meter corrections were applied, the on‑screen generation and the actual report values can differ. Before concluding that generation is low, you should confirm which measurement point’s values are being used.

Thus, in the initial check, clarify "which period, which unit, and which measurement point's values are being compared to what." A judgment that power generation is low only becomes meaningful once the comparison conditions are aligned. If you suspect an equipment-side fault before units and periods are aligned, the prioritization of inspections can easily become disordered. Reviewing simulation conditions is easier if you think of it as first preparing the foundation for comparison.

Revisit assumptions about solar radiation and weather conditions

The key condition at the center of power generation simulations is solar irradiance. Because photovoltaic systems generate electricity from incoming sunlight, if actual solar irradiance is lower than assumed, the generated output will also be lower. When you notice lower generation output, you need to check not only for equipment faults but also to what extent the solar irradiance assumptions used in the simulation differed from the actual weather conditions.

The solar radiation used in simulations is generally based on historical meteorological data and long-term averages. This is useful as a long-term guideline, but it does not guarantee the weather for any particular year or month. In years when the rainy season is prolonged, in months when cloudy weather persists due to typhoons or frontal systems, or during periods with heavy snowfall or frequent yellow sand, actual power generation may fall below projections. In such cases, although the low power output is a factual observation, it does not necessarily indicate an equipment malfunction.

When evaluating solar radiation, pay attention not only to the monthly total but also to day-to-day variability. Even if a month appears average overall, if clouds are frequent during hours favorable for power generation, the amount of electricity generated will be limited. Conversely, even if mornings and evenings are clear, if many days have cloud cover around noon, generation during peak hours will be suppressed. Because simulations sometimes treat the temporal distribution of solar radiation under fixed assumptions, differences in actual cloud cover can appear as differences in power output.

Ambient temperature conditions are another point to review. Solar panels tend to generate more power with stronger solar irradiance, but at the same time their output tends to decrease when panel temperature rises. Therefore, in summer, despite high irradiance, energy yield may not increase as much as expected. Checking how temperature losses are accounted for in the simulation and whether heat dissipation from the back of the panels is adequately ensured in the actual installation environment will make it easier to identify the reasons for the discrepancy.

In snowy regions, it is important not only how much solar irradiance there is but also how the effects of snow are handled. In simulations, even on days with solar irradiance, if snow remains on the panel surface the power output will drop significantly. Actual performance varies depending on whether the slope allows snow to slide off naturally, whether the roof or racking geometry tends to retain snow, and whether there is drifting from surrounding areas. If snow losses are not adequately reflected in the simulation conditions, winter generation may appear lower than expected.

In addition, differences in weather and environmental conditions at installation sites—such as coastal areas, mountainous regions, industrial zones, and areas around farmland—also have an impact. On the coast, winds carrying salt and humidity; in mountainous areas, fog and localized clouds; around farmland, dust and pollen; in industrial regions, deposition of particulates—these can all affect power generation. These effects are not always fully represented by simple solar radiation data. If simulation conditions use wide-area meteorological data, you should assume there may be differences compared with the actual, site-specific conditions.

When checking whether low power generation is due to solar irradiance, comparing with nearby facilities is also effective. If facilities in the same area with similar orientation, tilt, and capacity show similar declines, the likelihood of weather-related causes increases. On the other hand, if only your facility’s output is significantly lower, you should proceed to check not only solar irradiance but also shading, soiling, equipment outages, and measurement anomalies. However, even when comparing with nearby facilities, simple comparisons are not valid if capacity, installation angle, output control, or power conditioner configuration differ. It is safest to use such comparisons only as an aid to isolate meteorological factors.

In reviewing simulation conditions, it is important to understand that the assumed solar irradiance is an "average expected value." Even if there are months with low power generation, that may be a natural outcome if solar irradiance was low. On the other hand, if solar irradiance is around normal yet only the power generation is significantly low, you need to look for other loss factors. By first organizing solar irradiance and weather conditions, it becomes easier to decide whether to proceed to equipment inspection or to observe the trend for a while.

Reconfirm the azimuth, tilt angle, and mounting surface conditions

In power generation simulations, the orientation and tilt angle of solar panels are important parameters. If the orientation or angle shown in design documents differs from the actual installation, the gap between the expected and actual energy output can become large. When generation is low, before checking whether there is a problem with on-site equipment, it is necessary to verify that the orientation, tilt angle, and mounting-surface conditions entered into the simulation are correct.

Orientation deviations tend to appear clearly in time-of-day generation curves. If a system simulated as south-facing is actually installed slightly southeast or southwest, the peak generation time and the morning/afternoon output profile will change. If it is oriented toward the southeast, generation tends to be stronger in the morning and fall off in the afternoon. If it is toward the southwest, generation tends to be biased toward the afternoon. Even when monthly generation totals make it hard to tell, checking the time-of-day graph makes it easier to detect deviations in orientation.

The tilt angle is equally important. Installations with a shallow tilt tend to generate more easily in summer and are less likely to increase output in winter. Installations with a steep tilt tend to receive more winter solar radiation, while their generation behavior can change with the high solar altitude in summer. Even if the simulation conditions use the design angle, differences in the actual roof pitch or mounting angle will lead to variations in seasonal generation. For rooftop installations in particular, it is advisable to check for any discrepancy between the pitch shown on the drawings and the measured value.

For installations with multiple mounting surfaces, verifying the conditions becomes even more important. When divided into multiple orientations such as east-, west-, and south-facing surfaces, if the capacity allocation and tilt angle for each are not entered correctly, the expected power generation will not match reality. Simulations that only input the total capacity and treat it as a single orientation can produce time-of-day generation curves that do not align with on-site conditions. In such cases, not only the total generated energy but also the balance of generation between morning and afternoon will differ.

Also, the surface on which the panels are installed may not be completely uniform. Even when it appears to be the same roof plane, there can be cases where only certain rows have a different tilt, only the extended section has different layout conditions, or the racking heights vary by location. Even for ground-mounted installations, site slope, the grade of the prepared surface, and height differences between rows can affect generation conditions. If these factors are simplified in simulations, you need to assume there will be some degree of discrepancy when comparing with actual results.

Panel layout density should also be reviewed as part of the installation surface conditions. In installations with narrow row spacing, shadows from the front rows can fall on the rear rows depending on the season and time of day. If inter-row shading is not accounted for in simulations, generated power may appear lower in the mornings, evenings, or during winter. Even for rooftop installations, changes in elevation, roof ridges, adjacent roof surfaces, or equipment can cast shadows on some panels. It is necessary to check not only orientation and tilt but also the relative positions of surrounding objects on the installation surface.

When checking orientation and tilt angle, do not rely solely on drawings and as-built documents; confirming them with site photos and measured data will improve accuracy. The design-stage drawings and the post-construction condition do not necessarily match exactly. Adjustments made during construction, component fit, avoidance of obstacles on the roof, and changes in placement due to site conditions can alter the actual installation conditions. If a simulation was created based on the initial design conditions, it is important to recheck it against the post-construction conditions.

The purpose of reviewing the azimuth and tilt angle when generation is low is not simply to look for input errors. It is also to understand the actual generation characteristics and to correctly interpret how output varies by time of day and by season. If the azimuth and tilt match reality, it becomes easier to narrow down the causes of reduced generation to other factors. Conversely, if these are off, no matter how detailed the inspections, differences from the simulation may persist. As a prerequisite for comparison, rechecking the installation conditions is a task that should always be carried out.

Review loss conditions for shadows, soiling, and the surrounding environment

When power generation is low, loss conditions such as shading, soiling, and the surrounding environment are easily overlooked. In simulations, a uniform loss rate is sometimes entered as a lump sum, but on actual sites the way shadows fall and the accumulation of dirt can vary greatly from place to place. Therefore, if loss conditions are set more leniently than the real site situation, power output can fall below expectations even without equipment failures.

The impact of shading is one of the factors in reduced energy production that is particularly prone to misjudgment. When buildings, trees, utility poles, racking, fences, nearby equipment, or mountain shadows cast shade only during certain times of day, it can be difficult to identify the cause from monthly energy production alone. If there is a tendency for low output only in the morning, only in the evening, or only in winter, it is worth re-evaluating the shading conditions. If simulations did not account for shading, or if obstacles that did not exist at the design stage have increased, discrepancies with actual performance will occur.

The growth of trees and changes to surrounding buildings are also important. Branches that were not a problem at the time of installation can grow over several years and cast shadows during specific seasons or times of day. If a new building or facility is built on adjacent land, generation conditions will also change. Simulations reflect the conditions at the time they are created and do not include all future changes in the surrounding environment. If power generation appears to be decreasing year after year, you should check not only age-related degradation but also changes in the surrounding environment.

Soiling conditions can also lead to discrepancies between expected and actual results. If dust, yellow sand, pollen, bird droppings, fallen leaves, or splashed mud adhere to the panel surface, the amount of solar radiation received can be reduced and power generation may decrease. Some dirt is naturally washed away by rain, but when the panel tilt is shallow or the structure tends to accumulate dirt along the edges, soiling is more likely to remain. If a simulation assumes small losses from soiling, actual power generation may appear lower depending on the real environmental conditions.

One particularly important point is that even if soiling or shading affects only some panels, it can impact the output of an entire circuit. In solar power installations, multiple panels are electrically connected, so poor conditions on some panels can influence the power generation of that circuit. If simulations assume uniform solar irradiance conditions, they may not fully capture reductions caused by partial shading or soiling. When there are strings or circuits with low generation, comparing them with site photos and inspection records makes it easier to narrow down the cause.

Losses from the surrounding environment include reflection conditions and ventilation. The rate at which panel temperature rises can differ between open areas and locations surrounded by buildings or walls. Poor ventilation makes it harder for panels to cool, increasing the likelihood of output reduction at high temperatures. Even if simulations assume temperature losses under certain conditions, if the actual installation environment is severe, the generated power—especially in summer—may be lower than expected.

Also, for ground-mounted installations, the growth of grass and the condition of weed-control measures should be checked. If grass grows in front of the panels, it can cast shadows on the lower edge. In particular, for low-mounted racks or installations with a shallow tilt, even a small grass height can cause shading effects. Because the frequency and timing of mowing are rarely reflected in simulation conditions, actual power generation can vary depending on the state of site management. When generation is low, it is necessary to review not only the monitoring data but also the on-site operation and maintenance conditions.

When reviewing shading and soiling loss conditions, simply judging "shaded" or "soiled" is insufficient. It is important to grasp at which times of day, over what area, and to what extent they are affecting the system. The impact on generation differs depending on whether the shading occurs only in the morning and evening or also around midday when generation is at its peak. Likewise, soiling affects output differently depending on whether it is thinly distributed across the entire surface or heavily concentrated in parts. Checking time-of-day graphs, site photos, and inspection records together makes it easier to explain differences from the simulation.

Because loss rates are often entered collectively in simulation conditions, it is important for practitioners to check their breakdown. When shading, soiling, temperature, wiring, equipment conversion, and aging are treated as a single lump sum, it becomes difficult to understand which losses are being assumed and to what extent. To isolate the causes of low power generation, losses should be considered separately by item as much as possible. By bringing the shading and soiling assumptions closer to actual on-site conditions, the gap between simulated and actual values can be evaluated more realistically.

Check the conditions of the power conditioner and wiring losses

When reviewing simulation conditions, you need to check not only the solar panel side but also the conditions for the power conditioner and wiring losses. The DC power generated is converted to AC by the power conditioner and then delivered through wiring and measurement points for use or sale. Because losses occur during this process, if you do not clarify which point's generation you are simulating and which losses you are accounting for, comparisons with actual performance will be misaligned.

First, check the relationship between the photovoltaic (PV) capacity and the power conditioner (inverter) capacity. If the PV capacity exceeds the rated output of the power conditioner, the output may become capped during periods of strong irradiance. This is not necessarily an abnormality and can be an expected design outcome. However, if simulations do not adequately reflect this capping, the generation on clear days may appear lower than expected. If time-of-day graphs show a flattened output around noon, confirm not only equipment faults but also restrictions caused by the capacity configuration.

Power conditioner conversion efficiency also affects the gap between simulation and actual results. Conversion efficiency is not constant; it varies with input voltage, input current, output magnitude, temperature conditions, and so on. Simulations may use representative efficiencies, but if actual operating conditions differ from those assumptions, the generated power will differ. In particular, facilities with long periods of low output or facilities with poor input balance across multiple circuits may show different apparent efficiencies.

Wiring losses should also be reviewed. On the DC-side wiring from the solar cells to the power conditioner, and on the AC-side wiring from the power conditioner to the point of interconnection, losses occur depending on distance, cable thickness, and the magnitude of the current. Even if a simulation assumes standard wiring losses, actual losses can be larger than expected when the actual wiring distance is long or the installation routing is complex. For systems with low power generation, it is important to check the wiring routes and the single-line diagram to ensure the simulation’s loss assumptions match the site.

Be aware of differences in circuit configuration. Even with the same system capacity, differences in the number of strings, the number of modules connected in series, and the allocation of input circuits change the operating conditions of the power conditioner. Even if a simulation assumes uniform inputs, in reality panels with different azimuths or tilts mixed on the same input can reduce power generation efficiency. Also, if only some strings are affected by shading or soiling, the output of the entire input circuit can become unstable. When generation is low, it is important to check conditions not only at the equipment level but also at the input-circuit level.

You should also verify whether output curtailment and grid-side constraints are included in the simulation conditions. Even when generation equipment is capable of normal operation, output may be curtailed due to grid-side constraints or control directives. If a simulation represents pure generation capability while actual results include output curtailment, the actual performance will appear low. In that case, it can be concluded that the equipment’s generation performance is not low, but that generation has been suppressed by operating conditions. Checking monitoring screens and operation logs for any history of control actions or shutdowns makes it easier to assess.

Also, communication and measurement conditions cannot be ignored. Because actual generation values are obtained from measuring instruments and monitoring devices, communication loss or differences in measurement points can make the apparent generation lower. In cases where data is missing for only a certain period, only specific power conditioners are not recorded, or corrections are not reflected during aggregation, the facility’s actual generation and the displayed values will not match. You also need to confirm whether the actual values used for comparison with simulations can be trusted.

Handling aging effects is also important. Photovoltaic systems can experience gradual changes in the performance of equipment and panels over time. Whether a simulation indicates assumed generation for the first year or values that account for aging will affect your judgment. If you simply compare a system that has been operating for several years with the assumed first-year output, its generation may appear lower. When there are assumed values for each year, it is important to compare against the value corresponding to the target year.

The purpose of reviewing the conditions for power conditioners and wiring losses is to understand where differences in generation are occurring. Even if the panels are producing sufficiently, discrepancies in conversion, wiring, control, or measurement will reduce the final actual output. By checking whether these losses are included in the simulation conditions, and if so whether they match on-site conditions, you can set equipment inspection priorities more accurately.

Summary

When you feel the power generation is low, rather than immediately suspecting a malfunction or poor installation, reviewing the simulation conditions can make it easier to sort out the causes. Power generation simulation is a useful basis for comparison, but it is only an estimate based on certain assumptions. Actual generation varies due to many factors such as weather, solar irradiance, installation conditions, shading, dirt, temperature, wiring, equipment configuration, operational control, and measurement conditions. Therefore, instead of immediately treating the discrepancy between simulation and actual results as an anomaly, it is important to first confirm that the comparison conditions are aligned.

The first thing to check is the unit and period of the generation being compared. You need to clarify whether the comparison is on a daily, monthly, or yearly basis, whether it refers to generated energy or electricity sold, and whether it is a value per unit of installed capacity—if you don’t sort these out, the basis for any judgment will collapse. Next, confirm the assumptions about solar irradiance and weather conditions. Simulations based on long‑term average values can differ from actual results in a particular year. For months with poor weather or periods affected by snow cover, yellow dust, pollen, typhoons, and the like, you need to consider causes other than equipment faults.

Orientation, tilt angle, and the conditions of the mounting surface are also important. If the design conditions differ from the actual installation state, there will be differences in power generation by time of day and by season. For installations split across multiple surfaces, or installations prone to inter-row shading, simple simulation conditions may not adequately represent the real site conditions. It is also necessary to confirm whether shadows, soiling, and the surrounding environment have changed from the conditions at the time of creation. Tree growth, changes to nearby buildings, grass growth, and the accumulation of dirt can all be causes of reduced power generation.

Also, don’t forget conditions such as power conditioners and wiring losses. Output capping due to capacity configuration, conversion efficiency, wiring distance, circuit configuration, output control, missing measurements, and so on all affect actual energy production. By checking which losses the simulation includes and which measurement points’ generation it assumes, you can more accurately assess the discrepancy with actual results.

In practice, what matters is not just looking at the result that power output is low, but breaking down, one by one, the reasons why it appears low. By reviewing the simulation conditions, you can determine whether the difference is a natural effect of the weather, a mistaken assumption about installation conditions, the impact of shading or soiling, or whether checks of equipment or measurements are needed. If you align the comparison conditions and then verify the field data, you can reduce unnecessary inspections and more easily decide which areas to prioritize.

To continuously manage declines in power generation, it is important to have simulation values, actual performance data, site photos, inspection records, weather information, and per-equipment operating data consolidated and ready for review. Keeping not only daily generation figures but also monthly and annual trends, differences between installations, and comparisons with past data helps enable early detection of anomalies. Rather than scrambling to find the cause when generation appears low, the foundation of stable generation management is to routinely standardize comparison conditions and accumulate information to support decision-making.

If you want to carry out verification of generation performance, recording of site conditions, and organization of information during inspections more efficiently, it is also effective to utilize management methods and inspection-support tools that make it easy to connect site management with the inspection tasks for photovoltaic power generation equipment. By organizing the differences between simulation conditions and actual results and putting in place a system that clarifies the items to be checked on site, initial decision-making when generation appears low will tend to become more stable.