【How to Check Solar Irradiance Data in PVSyst｜5 Key Metrics to Monitor】

Table of Contents

• Purpose of checking solar radiation data in PVSyst

• Basics to grasp before checking solar radiation data

• Indicator 1 to check: Global horizontal irradiance

• Indicator 2 to check: Direct and diffuse solar radiation

• Indicator 3 to check: Tilted plane solar radiation

• Indicator 4 to check: Monthly variability of solar radiation

• Indicator 5 to check: Missing and anomalous values

• Specific process for checking solar radiation data

• Common mistakes beginners make when checking

• Practical approach to using solar radiation data in practice

• Summary

Purpose of checking solar irradiance data in PVSyst

When running a solar power plant simulation in PVSyst, the first important information to check is the solar irradiance data. Calculations of energy production proceed on the premise of how much sunlight reaches the installation site. Therefore, even if you finely configure panel capacity, PCS capacity, tilt angle, azimuth, and loss conditions, an insufficient understanding of the underlying irradiance data can lead you to misinterpret the simulation results.

Many practitioners researching how to use PVSyst are unsure which numbers on the screen to look at, which items have a major impact on energy production, and how much verification is sufficient for practical purposes.

Solar radiation data contains many technical terms, and similar items are listed: horizontal plane, tilted plane, direct, diffuse, monthly values, annual values, etc. However, by organizing the key points to check, even beginners can more easily assess the validity of the data.

This article, focusing on five indicators to check when reviewing solar irradiance data in PVSyst, explains practical verification procedures and points to watch. The goal is not simply to read the numbers on the screen, but to be able to determine whether the irradiance data, as a precondition for power generation simulations, is appropriate for the design conditions.

In solar power generation simulations, differences of a few percent can affect annual energy production and project viability assessments. In particular, when comparing candidate sites, during basic design, when calculating approximate generation, for documents intended for financial institutions, and for post-construction generation verification, the handling of solar irradiance data is important. By getting into the habit of checking irradiance data from the stage when you start using PVSyst, you can reduce rework in later stages.

Fundamentals to Grasp Before Checking Solar Radiation Data

Before checking irradiance data in PVSyst, you first need to understand what irradiance represents. Irradiance is the amount of solar energy that reaches the ground or a panel surface over a given period. It is generally expressed as the amount of energy per square meter. In photovoltaic power generation, the higher the irradiance, the more energy is available for generation, making it a key determinant of annual energy production.

However, higher solar irradiance does not necessarily lead to a proportional increase in power output. Actual power generation is affected by ambient temperature, panel temperature, shading, soiling, wiring losses, PCS conversion losses, equipment specifications, installation angle, orientation, terrain, and other factors. PVSyst combines these conditions to perform simulations, but the starting point is meteorological data, and among that solar irradiance data is particularly important.

In PVSyst, the solar irradiance handled includes irradiance incident on a horizontal plane, direct irradiance arriving from the sun, diffuse irradiance scattered in the atmosphere, and irradiance reaching a panel’s tilted surface. These are similar but have different meanings. Horizontal-plane irradiance is the basic value for understanding a site’s climatic characteristics. Direct and diffuse irradiance are used to understand the state of the sky and atmospheric conditions. Tilted-surface irradiance is important for determining how much light actually reaches the panel surface.

Also, rather than looking only at the annual total, it is important to check monthly variations. Even if the annual solar irradiation is similar, locations that are stable in spring and autumn, locations that are heavily skewed toward summer, and locations that experience a large drop in winter will produce different impressions of the generation curve and capacity factor. Business plans tend to emphasize annual values, but in practice, checking monthly values is indispensable.

When reviewing solar radiation data, it is important not to judge solely by the magnitude of the numbers. If there are extremely high or low values, they may reflect actual climatic characteristics, but they may also result from input data errors, incorrect site settings, unit mix-ups, or the effects of missing-data imputation. When using PVSyst, rather than taking the values displayed on the screen at face value, you should interpret them while checking site conditions, the data period, monthly trends, and consistency with other settings.

Key Metric 1: Global Horizontal Irradiance

The first metric to check is the global horizontal irradiance. This is the basic solar radiation measure that represents the total amount of sunlight reaching a horizontal surface on the ground. When evaluating candidate sites for solar power generation, it is used to understand the overall solar potential of the region. In PVSyst, after loading the meteorological data, you can view this global horizontal irradiance on the site information and monthly meteorological data screens.

Global horizontal irradiance is easiest to understand if you think of it as the sum of direct solar radiation and diffuse solar radiation. It tends to be larger in regions with many clear skies and strong sunlight, and smaller in regions with frequent cloudiness and rainfall. Of course, it also varies with latitude, elevation, season, cloud cover, and atmospheric clarity. Even within Japan there are regional differences, so placing the same power generation equipment at different sites will result in differences in annual power output.

When viewing these values in PVSyst, first check whether the annual total is not significantly out of line with the expected climate of the area. For example, if you assume the region has good solar irradiation conditions but the annual value is extremely low, or if the region is mountainous or experiences frequent overcast skies yet the value is abnormally high, you should review the site settings and the selection of weather data. Possible causes include errors in entering latitude and longitude, differences in elevation, choosing data from a nearby station, or differences in the conditions used when the data were generated.

Next, examine the monthly global horizontal irradiance. The annual total alone does not tell you which seasons receive more solar radiation. In photovoltaic power generation there tends to be more irradiance in summer and less in winter, but in some regions this pattern changes due to the rainy season, snowfall, seasonal winds, and cloudiness. Checking monthly values makes it easier to understand the breakdown of annual generation and makes later simulation results easier to interpret.

Global horizontal irradiance is the entry point for checking irradiance in PVSyst. However, actual solar panels are generally not horizontal but are installed at a certain tilt angle. Therefore, making a final judgment based solely on global horizontal irradiance is insufficient. It should be used only as an indicator to grasp the basic irradiance conditions at the site, and after that it is natural to check the tilted-surface irradiance and loss items.

Also, when comparing multiple candidate sites, it is useful to view global horizontal irradiance as a common reference. As a preliminary step before running simulations under the same conditions, this lets you grasp the solar resource potential at each site. However, when comparing sites, a location with higher irradiance is not necessarily the optimal choice. Landform, site development conditions, grid connection, shading effects, constructability, maintainability, and other factors also come into play. Solar irradiance is an important metric, but in practice it should be treated as one part of the overall design.

Key Metric 2: Direct and Diffuse Solar Irradiance

The second set of metrics I want to check are direct solar irradiance and diffuse solar irradiance. Direct solar irradiance is the radiation that reaches the ground directly from the sun after passing through the atmosphere. By contrast, diffuse solar irradiance is radiation scattered by atmospheric water vapor, clouds, dust, and other particles, arriving from the entire sky. While global horizontal irradiance indicates the total amount of solar radiation on a horizontal surface, direct and diffuse solar irradiance are metrics used to break that total down.

In photovoltaic power generation, the proportion of direct solar radiation to diffuse solar radiation changes the generation characteristics. In regions with many clear days, the direct component tends to be larger, while in regions with many cloudy days, the proportion of the diffuse component tends to be higher. In typical fixed photovoltaic systems, not only direct radiation but also diffuse radiation contributes to power generation. Therefore, it is important not to look only at irradiance on clear days, but to include the diffuse light that reaches the panels on cloudy days.

The purpose of checking direct and diffuse irradiance in PVSyst is to understand the characteristics of the meteorological data. For example, even if the annual global horizontal irradiance appears normal, if the balance between direct and diffuse components is extreme, it may be advisable to verify the data’s origin and the gap-filling methods used. In particular, when importing external meteorological data or using locally observed data that has been processed, examining the consistency between the direct and diffuse components makes it easier to assess the quality of the input data.

Also, the ratio of direct to diffuse solar irradiance is related to the effects of shading and to inclined-surface conversion. The component arriving directly from the sun is more susceptible to shadows cast by obstacles and by terrain. In contrast, the diffuse component arrives from the whole sky, so it is affected by shading in a different way. In PVSyst, irradiance is transposed to the panel surface and the energy production is then calculated taking into account shading and other losses. Therefore, understanding the breakdown between direct and diffuse components makes it easier to trace the reasons behind loss results later.

Beginners should be careful not to judge a site's suitability based solely on direct solar irradiance. In photovoltaic power generation, the total solar irradiance received by the panels is what matters. A high direct component can be advantageous, but you need to consider the total energy — including the diffuse component — together with seasonal variations and the relationship with the installation angle. Especially in regions like Japan, where weather varies greatly by season, checking the monthly balance between direct and diffuse irradiance makes it easier to understand seasonal variations in power generation.

In practice, when checking direct and diffuse solar irradiance, we look for any anomalies in the monthly pattern. For example, it is natural for the share of the diffuse component to increase during rainy or cloudy seasons and for the direct component to increase during periods with more clear skies. Of course there are regional differences, but if the data strongly contradicts seasonal expectations, it serves as a prompt to recheck the site or data settings.

Indicator 3 to Watch: Solar Irradiance on Inclined Surfaces

The third metric to check is the tilted-surface irradiance. This represents the solar irradiance that actually reaches the surface where the photovoltaic panels will be installed. While the global horizontal irradiance indicates the basic solar conditions at the site, the tilted-surface irradiance is a metric that more closely reflects the expected energy production because it accounts for the designed panel tilt and orientation. This tilted-surface irradiance is extremely important when evaluating energy production in PVSyst.

Solar panels are usually installed at a fixed tilt rather than horizontally. Changing the tilt angle and azimuth alters the solar irradiance reaching the panel surface at the same location. Installing at an orientation close to due south versus tilting toward the east or west changes not only the annual total but also the generation patterns in the mornings and evenings and across seasons. In PVSyst, when you enter the design conditions, it calculates the irradiance on the tilted surface by converting from the horizontal-plane irradiance.

When checking the irradiation on a tilted surface, first confirm that the designed azimuth and tilt angles are correctly reflected. If the angles you entered are incorrect, both the tilted-surface irradiation and the energy yield will be significantly off. For example, if you mix up the azimuth reference or misunderstand the sign of the angles, you may end up performing calculations for an orientation different from the one intended. When you are not yet familiar with using PVSyst, it is important to always review the settings after entering the values.

Next, we examine the relationship between global horizontal irradiation and tilted-surface irradiation. In general, when installed at an appropriate tilt angle, the annual tilted-surface irradiation can be more favorable than that on a horizontal plane. However, the relationship varies with installation angle, region, and season. In particular, designs with a small tilt angle or a large azimuth deviation may not yield as much increase in tilted-surface irradiation as expected. Because this directly affects differences in power generation, it is an item that should always be checked when comparing designs.

Tilted-plane irradiance has different meanings depending on whether it is before considering the effects of shadows or after accounting for shadows and other losses. First, check it as the pure irradiance incident on the panel surface, and then look at the results that reflect losses such as near shading, far shading, ground reflection, and soiling. In PVSyst’s result screens, it is important to be aware which stage’s irradiance you are looking at. Simply saying “tilted-plane irradiance” can be interpreted differently depending on whether it is before or after losses.

In practice, it is common to compare design proposals using tilted surface irradiance. For example, you check how much the amount of solar irradiation captured changes when you slightly change the tilt angle, change the azimuth, or change the racking layout. However, you should not decide a design based only on tilted surface irradiance; you need to judge it also considering installable capacity, spacing distances, shading effects, constructability, wind loads, maintenance access, and so on. PVSyst is effective for comparing irradiance and energy production, but it will produce results usable in practice only if the site conditions are entered correctly.

Key Metric 4: Monthly Variations in Solar Radiation

The fourth metric to look at is the monthly variation in solar radiation. When checking solar radiation data, you may be tempted to feel reassured by looking only at the annual total, but in practice you should always check the monthly variations. Even if the annual solar radiation is the same, which months are high and which are low will change how generation looks, revenue planning, maintenance scheduling, and approaches to equipment operation.

In photovoltaic power generation, solar irradiance generally increases from spring to summer and decreases in winter. However, midsummer is not necessarily the month with the highest generation. While summer brings higher irradiance, panels are more affected by temperature rise, which increases panel temperature. In addition, monthly irradiance shows regional characteristics due to the rainy season, typhoons, snowfall, and seasonal cloud cover. By viewing monthly data in PVSyst, you can grasp the generation trends unique to that location.

When checking monthly variations in solar radiation, first look for any extreme months. If a particular month is unnaturally high or low, missing-data imputation or data-entry errors may be suspected. This is especially true when external data have been imported; scanning the sequence of monthly values makes it easier to spot anomalies. Issues that are not obvious in the annual total can become clearly apparent when examined month by month.

Next, check whether the monthly solar irradiation and the monthly energy production naturally correspond. In principle, months with higher irradiation should also show increased production, but when there are temperature-related losses, shading effects, assumed output curtailment, soiling, snow, etc., the relationship will not be a simple proportional one. When reading PVSyst results, it is important to view the monthly variation of irradiation and the monthly variation of production side by side to understand which losses are affecting which seasons.

Also, monthly solar irradiation is useful for explaining project plans. If you present only the annual generation, it is hard to convey the seasonality of power production. By understanding monthly trends, it becomes easier to explain why generation is lower in winter, why it drops in specific months, and where changes to design conditions will lead to improvements. For practitioners, using PVSyst is not just about running calculations; it is also about being able to present and explain the results to stakeholders.

When examining monthly variations, caution is also required when comparing candidate sites. Even locations with high annual values may experience declines in power generation during certain periods. Conversely, a site with average annual values may have stable seasonal variation and be easier to operate. Which is better depends on the project’s objectives, the terms for electricity sales, its relationship to demand, and the maintenance arrangements. Therefore, in PVSyst it is standard practice to check both annual and monthly values.

Metric 5 to Monitor: Missing Values and Outliers

The fifth metric to check is missing and anomalous values. This is not a type of solar irradiance itself, but it is a very important check for assessing data quality. No matter how carefully you configure PVSyst, if there are problems in the original data, the reliability of the simulation results will decrease. Especially when using on-site observation data or meteorological data processed externally, it is important not to skip checking for missing or anomalous values.

Missing values refer to instances in which data were not recorded for some reason. They can arise from instrument stoppages, communication failures, recording errors, or errors during data conversion. Short-term missing data can sometimes be imputed, but long-term missing data or missing data biased toward particular seasons can affect annual solar radiation and monthly trends. Before importing data into PVSyst, or after import, it is necessary to check for the presence of missing values and the imputation method.

Anomalous values refer to values that are so high or low as to be unlikely in reality, or values that are unnatural compared with surrounding data. For example, solar radiation being recorded at night, values that are extremely small for clear-sky conditions, briefly abnormally large values, or monthly totals that differ drastically from adjacent months are all examples. Causes can include unit conversion mistakes, time offsets, instrument malfunctions, or errors during data processing.

When checking solar irradiance data in PVSyst, pay attention not only to annual and monthly values but also to the quality of the data. Just because values are displayed on the screen does not mean the data are necessarily correct. In particular, when comparing multiple datasets or using local observations, it is important to verify the data period, time reference, units, missing-data imputation, and the approach to handling outliers.

What beginners often overlook is the difference in units. Solar irradiance data includes values that represent intensity per unit time and values that represent the accumulated amount over a given period. If you use the wrong input format, the simulation results can be significantly off. Handling of time is also important. If the solar position and the time are mismatched, the irradiance distribution and the power generation curve can become unrealistic. When working with external data in PVSyst, make sure to check the data format before importing.

Checking for missing or anomalous values is a verification task that, once you get used to it, doesn't take much time. However, if you neglect it, it becomes difficult to explain the power output later. For example, when simulation results are lower than expected, you need to determine whether the design conditions are poor, the loss settings are too high, or there is a problem with the solar irradiance data. If you check data quality at the outset, later verification steps will go more smoothly.

Specific procedure for checking solar irradiance data

When checking solar irradiance data in PVSyst, it becomes easier to understand if you don't just open the screen and stare at the numbers but instead check in a certain order. The first thing to do is to check the site settings. Confirm that latitude, longitude, elevation, time zone, and the target region are set correctly. Because solar irradiance data is highly dependent on the site, if the location information is off, all subsequent checks become meaningless.

Next, verify the period and type of meteorological data being used. Whether it is long-term average data, observational data from a specific year, or representative values based on multiple years will change how the results are interpreted. Long-term average data are convenient for standard design assessments, but they may not capture extreme meteorological variations in a particular year. Local observational data can reflect conditions close to the site, but caution is needed if the observation period is short or there are missing observations.

After that, check the annual value of the global horizontal irradiance. The purpose here is to see whether it is broadly consistent with regional expectations. Check whether areas assumed to have strong insolation show an unexpectedly low value, or conversely whether regions with frequent cloudy skies show an unnaturally high value. If anything seems off, review the site settings, selection of meteorological data, units, and import format.

Next, examine the monthly solar radiation. Even if the annual total appears reasonable, the monthly pattern may show anomalies. For example, a particular month may be unusually high, the seasonal trend may not match the region’s climate, or the difference between winter and summer may be unnatural. Monthly values are useful for verifying both data quality and regional characteristics.

Next, examine the balance between direct and diffuse solar radiation. Here, determine whether the data are clear-sky type or contain a large diffuse component. The breakdown between direct and diffuse is useful for understanding slope conversion and the effects of shading. When using external data, check for extreme values because results can vary depending on how direct and diffuse are calculated or on gap-filling procedures.

Finally, after entering the design conditions, check the irradiance on the tilted surface. Even if the irradiance on the horizontal plane looks reasonable, mistakes in setting the tilt angle or azimuth can make the tilted-surface irradiance appear unnatural. By confirming that the solar irradiance incident on the panel surface matches the design intent before viewing the power generation simulation results, you can detect configuration errors at an early stage.

If you follow this sequence, even users who are not familiar with operating PVSyst will be less likely to be unsure where to look. By examining, in order, location, data type, horizontal irradiance, monthly variation, direct and diffuse components, irradiance on tilted surfaces, and missing and anomalous values, you can gain a three-dimensional understanding of the solar irradiance data.

Checkpoints Beginners Often Get Wrong

When checking irradiance data in PVSyst, a common mistake beginners make is to consider the check complete by looking only at the annual energy production results. When the simulation displays the energy production, it can appear to be the final answer. However, that energy production is calculated based on the irradiance data, design conditions, and loss settings. If you don't verify the underlying irradiance data, you cannot judge whether the results are reasonable.

A common mistake is confusing global horizontal irradiance with tilted-surface irradiance. Global horizontal irradiance is the basic value that represents the meteorological conditions at a location, while tilted-surface irradiance is the irradiance incident on the panel surface. Tilted-surface irradiance is closer to actual power generation, but global horizontal irradiance is also important for site comparisons and data verification. Distinguishing between the two prevents misinterpretation of results.

Also, failing to check monthly data is a major mistake. Annual values alone do not reveal seasonal generation trends. In practice, there are often situations where an explanation of monthly generation is required. To explain the drop in generation during winter, temperature-related losses in summer, and the reduction in solar irradiance during the rainy season, you need to understand the relationship between monthly solar irradiance and generation.

Mistakes in entering azimuth or tilt angles are also common. In PVSyst, because installation conditions must be entered numerically, misunderstanding the reference azimuth or the direction/sign of the angles can lead to unintended design conditions. After inputting the data, it is important to check the changes in plane-of-array irradiance and energy production to confirm they match the design intent. This is especially critical when comparing multiple scenarios: an error in the angle inputs can invalidate the comparison itself.

When importing external meteorological data, you must also pay attention to units and time handling. If the units of solar irradiance, the time interval, the time reference, or the treatment of missing data are not appropriate, the data may be read into PVSyst yet yield implausible results. After importing the data, always check monthly values and daily trends to ensure there are no extreme inconsistencies.

Also, be careful not to judge power generation based solely on solar irradiance. Power generation is strongly affected by irradiance, but it is influenced by many factors such as temperature, shading, equipment specifications, soiling, wiring, PCS, and terrain. If irradiance is high but generation does not increase, temperature losses or shading losses may be significant. Conversely, even if irradiance is average, good design conditions can result in stable power generation.

Approaches to Utilizing Solar Radiation Data in Practical Work

In practical use of PVSyst, solar irradiance data is not merely an input value but forms the basis for design decisions and explanatory materials. In initial site screening, global horizontal irradiance is used to confirm the area's generation potential. In basic design, irradiance on tilted surfaces is reviewed to assess the appropriateness of the azimuth and tilt angles. In detailed design, after accounting for shading and losses, the expected energy production and the breakdown of losses are verified.

What's important when using solar irradiance data is not to look at the numbers in isolation but to interpret them in conjunction with the design conditions. Even with the same irradiance, power generation can vary depending on the racking layout, panel tilt angle, terrain, and surrounding obstacles. When comparing multiple scenarios in PVSyst, separating the differences into irradiance, irradiance on the tilted surface, shading losses, and temperature losses makes it easier to understand which conditions are influencing the results.

Also, when explaining to stakeholders, it is important not to simply list technical terms but to organize and convey their meanings. Explain that global horizontal irradiance is the basic amount of solar energy in the region; irradiance on an inclined (tilted) surface is the energy actually received by the panels; direct irradiance and diffuse irradiance are the breakdown of how light arrives; monthly variation indicates seasonal trends in power generation; and missing data and outliers are items for checking the reliability of the data. Presenting the information this way makes it easier for non-expert stakeholders to understand.

Solar irradiance data are also useful for verifying post-construction power generation. When comparing the planned simulation with the actual generation, the first thing to check is how the irradiance conditions during the actual period compared with those at the time of planning. If irradiance was lower than planned, a reduction in generation is natural. Conversely, if irradiance conditions were similar but generation is low, you need to check for shading, soiling, equipment malfunctions, wiring, the PCS, and measurement conditions.

Correctly reflecting on-site conditions is also an important point when utilizing solar irradiance data. Desktop simulations may not be able to fully capture terrain and obstacles. The actual site includes trees, buildings, slopes, utility poles, fences, surrounding equipment, and changes in ground elevation after land development. When these affect shading and layout conditions, discrepancies can arise between the assumptions in PVSyst and the actual power generation conditions. Therefore, it is important to establish design assumptions by combining site surveys and surveying data.

Especially at large-scale solar power plants and sites with uneven terrain, it is important to accurately understand panel layout, ground elevation, and shadowing conditions. Even if solar irradiance data are correct, discrepancies with actual power generation are likely to occur if the terrain conditions are off. To make practical use of the irradiance and energy yield results obtained from PVSyst, the accuracy of the site information entered is equally important.

Summary

When checking irradiance data in PVSyst, the indicators to look at are five: global horizontal irradiance, direct and diffuse irradiance, irradiance on the tilted plane, monthly irradiance variability, and missing and anomalous values. By checking these in order, it becomes easier to judge whether the assumptions for the power generation simulation are reasonable.

Global horizontal irradiance is a basic indicator for assessing a location's solar radiation potential. Direct irradiance and diffuse irradiance help to understand how sunlight is delivered and the characteristics of the meteorological data. Irradiance on tilted surfaces is an important indicator that is closer to actual power generation, as it represents the solar radiation that actually reaches panel surfaces. By looking at monthly irradiance variations, you can grasp seasonal generation trends and detect inconsistencies in the data. Furthermore, checking for missing and abnormal values can improve the reliability of simulation results.

What is important when using PVSyst is not to accept the calculation results as they are, but to read the results while checking the underlying conditions. Solar irradiance data are the entry point for power generation calculations and relate to design decisions, site comparisons, stakeholder explanations, and post-construction verification. By checking not only annual values but also monthly values, breakdowns, and consistency with the design conditions, you move closer to simulations that can be used in practice.

Also, to configure PVSyst correctly, it is essential to accurately understand the site location, terrain, and surrounding environment. Even if the solar irradiance data are valid, if the installation position, ground elevation, or information about nearby obstacles is ambiguous, discrepancies will occur in the assessment of shading and layout conditions. In the design and on-site verification of a solar power plant, it is important to integrate desk-based simulations with on-site positioning and surveying information.

In that regard, by utilizing LRTK, an iPhone-mounted GNSS high-precision positioning device, you can streamline candidate site verification, survey point recording, current condition assessment, and pre- and post-construction checks based on the high-precision positional data acquired on-site. By organizing assumptions about solar irradiance and energy production in PVSyst and accurately capturing on-site location data with LRTK, it becomes easier to reliably connect the planning, design, construction, and maintenance of solar photovoltaic systems.