7 Tips to Avoid Mistakes When Setting Meteorological Conditions in PVSyst

Table of Contents

• The impact of PVSyst meteorological condition settings on power generation simulation

• Note 1: Check for location mismatches between site information and meteorological data

• Note 2: Understand the types of irradiance data and choose appropriately

• Note 3: Do not confuse horizontal-plane irradiance with tilted-plane irradiance

• Note 4: Do not overlook temperature and wind speed conditions

• Note 5: Check time resolution and time zone alignment

• Note 6: Understand the difference between long-term average data and measured data

• Note 7: Do not use outliers and missing values as-is

• Practical checklist for avoiding mistakes in meteorological condition settings

• Summary: The accuracy of power generation simulations is greatly affected by understanding local site conditions

The Impact of PVSyst Weather Condition Settings on Power Generation Simulations

One of the first things practitioners who are just beginning to learn how to use PVSyst tend to get stuck on is configuring the meteorological conditions. In solar power simulations there are many input items—panel capacity, PCS capacity, azimuth, tilt angle, shading, loss rates, and so on—but if the underlying meteorological data are inaccurate, no matter how carefully you enter the equipment parameters, the results for annual and monthly energy production will not match reality.

Weather conditions are the starting point for power generation calculations. For photovoltaic power generation, naturally, the more solar irradiance there is, the more power is generated, and the less solar irradiance there is, the less power is generated. Also, even with the same solar irradiance, in regions with higher ambient temperatures panel temperature tends to rise and generation efficiency tends to decrease. In windy locations panels are more easily cooled, and losses due to temperature rise may change. In other words, setting the weather conditions is not simply the task of "choosing a location" or "loading data", but an important process that determines the overall reliability of the power generation simulation.

In practice, people often become reassured as soon as they can select meteorological data on PVSyst's interface. However, unless you check whether the selected meteorological data truly corresponds to the target site, whether the type of solar irradiance is appropriate for the analysis purpose, whether there are discrepancies in the time step or time zone, and whether the data contain missing or anomalous values, the basis for the results will be weakened.

In particular, when the data are used for business feasibility assessments, design studies, customer explanations, preparing materials for financial institutions, or post-construction generation comparisons, it is important to be able to explain the rationale for the meteorological condition settings.

The basic workflow in PVSyst is to set the site, create or import meteorological data, and reflect that in the system design. However, what is difficult for beginners is judging which data to choose, how thoroughly to verify it, and how much it will affect the results. This article organizes seven key points that practitioners should grasp to avoid mistakes when setting meteorological conditions in PVSyst. Rather than being mere operational steps, it explains why those checks are necessary, what kinds of errors are likely to occur, and how to approach them in actual practice.

Note 1: Verify positional discrepancies between location information and weather data

When setting meteorological conditions in PVSyst, the first thing to check is the positional relationship between the site and the meteorological data. When entering a candidate site or installation location for a solar power plant, you set the latitude, longitude, and elevation, but the meteorological data may not exactly match that location. In many cases you will use data from a nearby observation station, a representative point, or regional meteorological datasets. Therefore, you must always check the distance between the site and the meteorological-data location, the elevation difference, and differences in terrain conditions.

For example, even within the same municipality, coastal areas, mountainous areas, basins, and plateaus can differ in solar radiation, temperature, frequency of fog occurrence, snowfall, and wind conditions. If meteorological data from lowland areas are applied directly to a project in a mountainous area, the solar radiation and temperature may differ from actual conditions. Conversely, using data from a high-elevation site for a lowland project can cause temperature conditions to be underestimated or overestimated.

When considering location discrepancies, be particularly careful not to judge based on distance alone. Even if the straight-line distance is short, meteorological characteristics can differ greatly if, for example, a mountain lies between the sites, they are affected differently by sea breezes, snow conditions differ, or the temperature environment differs between urban and suburban areas. Entering location information in PVSyst’s location settings screen and meteorological data management screen is basic practice, but in the field it is important to also check topographic maps, on-site conditions, and the surrounding observation environment.

Elevation is another item that's easy to overlook. When elevation changes, temperature conditions tend to change, and the effects of snowfall and fog may also differ. In solar power generation simulations, solar irradiance tends to attract the most attention, but temperature, wind speed, and the presence or absence of snowfall also affect power output. If the elevation of the target site differs significantly from the elevation of the meteorological data, it is necessary to consider how much that difference is likely to affect the results.

To avoid mistakes in practice, once you have selected meteorological data, first confirm “which location this data represents,” “how far it is from the target site,” “whether the terrain and elevation are similar,” and “whether the environment—coastal, mountainous, urban, etc.—matches.” As you get used to operating PVSyst, it’s easy to focus only on the data-loading step, but the reliability of the power generation simulation depends largely on the validity of the initial site conditions.

Note 2: Understand the types of solar radiation data and choose appropriately

The next important thing in PVSyst's meteorological settings is to understand the types of solar irradiance data. In photovoltaic simulations, even when we refer to irradiance in a single word, there are several concepts: global horizontal irradiance on a horizontal plane, direct irradiance, diffuse irradiance, and irradiance on tilted surfaces. If you load data without understanding these distinctions, you may find that, although input values are present, the calculation results become unrealistic.

In typical photovoltaic analysis, the irradiance incident on the horizontal plane is used as a basis to calculate the irradiance on the tilted plane according to the tilt and azimuth of the installed panels. In PVSyst, the irradiance information contained in the meteorological data is used to calculate energy production by taking into account the sun position and the installation conditions. Therefore, it is important to confirm whether the imported data are values on the horizontal plane or values that have already been converted to the tilted plane.

What beginners often get wrong is looking only at the name of the solar irradiance and assuming any value can be used in the same way. For example, even if a dataset is labeled "global irradiance," you must confirm whether it refers to a horizontal plane value or a value for a specific tilt angle. Also, data that separate direct and diffuse components may be handled differently inside a simulation than data that provide only the total.

When selecting solar irradiance data, you also need to consider the purpose of the analysis. For rough assessments, you may use representative long-term average data to estimate approximate power generation. On the other hand, for detailed design or project feasibility evaluations, it is desirable to use data that are close to the target site and whose period and quality are clearly defined. Furthermore, when comparing with the actual performance of an existing power plant, you must use data that closely match the measured meteorological conditions for the target period; otherwise, comparisons with the generation results cannot be made appropriately.

When using PVSyst, you should not just select meteorological data; you need to examine the data itself. Check whether the monthly irradiation is excessively high, whether seasonal variations match regional characteristics, and whether the annual irradiation deviates significantly from the general trends in the surrounding area. In particular, for overseas projects, mountainous sites, snowy regions, and coastal areas, solar irradiation patterns may differ from those of typical flatland.

Solar irradiation data is the single most significant input for power generation simulations. Even small configuration mistakes can have a large impact on the assessment of annual energy production. When setting meteorological conditions in PVSyst, it is important to understand the type of solar irradiation and to verify which surface the input data refers to, which components it includes, and which period it represents before using it.

Note 3: Do not confuse horizontal-plane solar irradiance with tilted-plane solar irradiance

One thing to be particularly careful about when setting the meteorological conditions in PVSyst is confusing horizontal-plane solar irradiance with tilted-plane solar irradiance. Solar panels are normally installed with a tilt relative to the ground. Whether south-facing, east–west-facing, low-tilt, steep-tilt, roof-mounted, or ground-mounted, the amount of solar irradiance incident on the panel surface varies depending on installation conditions. On the other hand, the solar irradiance commonly provided as meteorological data is often a value observed or estimated on the horizontal plane.

Horizontal plane irradiance is the irradiance that reaches a surface horizontal to the ground. In PVSyst, based on this horizontal plane irradiance, the irradiance on the panel surface is calculated by taking into account the panel tilt and azimuth, the sun’s motion, the direct and diffuse components, and so on. By contrast, tilted plane irradiance is irradiance that has already been converted to a specific tilt and azimuth. If tilted plane irradiance is entered as horizontal plane irradiance, PVSyst will perform the tilt conversion again, which can lead to an unnatural calculation that appears to have been corrected twice.

This mistake tends to occur when importing external data. When the data on hand is organized in table form, column names may be abbreviated, making it hard to tell whether they refer to a horizontal or inclined plane. Also, when you receive data created by another colleague, the conditions under which the values were calculated may not have been shared. Before importing into PVSyst, you should check the data definitions and clarify the units and calculation conditions.

Confusing horizontal-plane irradiance with tilted-plane irradiance can affect the balance of power generation across seasons. In systems with a large tilt angle, irradiance on the tilted surface tends to increase in winter, while in summer the difference from horizontal-plane irradiance becomes smaller, or under certain conditions may even reverse. If incorrect irradiance values are input, not only the annual energy production but also the monthly generation trends will appear unnatural. When reviewing monthly results, if you observe trends that don't match the region's seasonal variations—such as being overestimated only in winter or underestimated only in summer—you should review the surface conditions used for the irradiance.

In practice, before inputting meteorological data, it’s a good idea to document the types of solar radiation as reference material. Record which column corresponds to global horizontal irradiance, whether direct and diffuse components are included, and, for tilted-surface data, the tilt and azimuth angles. When explaining results later, clarifying the assumptions about solar radiation improves the reproducibility of the simulation.

What matters in using PVSyst is not whether the on-screen inputs are complete but whether you understand what the entered values mean. The difference between horizontal-plane irradiance and tilted-plane irradiance is something beginners easily overlook, yet it is a point that can greatly affect the results. In the meteorological condition settings, always verify the irradiance surface conditions and be careful not to duplicate the conversion processing within PVSyst.

Point 4: Do not underestimate temperature and wind speed conditions

In solar power generation simulations, solar irradiance is treated as the most important factor, but temperature and wind speed also affect power output. If, when setting the meteorological conditions in PVSyst, you check only solar irradiance and proceed without adequately considering temperature and wind speed, the loss assessment related to panel temperature may not match the actual site conditions.

Solar panels generate electricity when exposed to sunlight, but their temperature also rises. In general, the higher the panel temperature, the more likely the generation efficiency will decrease. In regions with high ambient temperatures, panel temperature tends to rise more easily even under the same solar irradiance, and temperature-related losses tend to be larger. Conversely, in cold regions, ambient temperatures are lower, so temperature-related losses may be suppressed even under the same solar irradiance. For this reason, when comparing annual energy yields, differences can arise that cannot be explained by solar irradiance alone.

Wind speed should not be overlooked. In windy locations, the front and back surfaces of panels are more easily cooled. In particular, the way panel temperature rises can differ between ground-mounted installations with adequate ventilation and installations positioned close to a roof. PVSyst takes temperature losses into account based on meteorological and installation conditions, but if the wind speed data you enter deviates significantly from actual local conditions, temperature-related calculations can also be off.

A common pitfall for beginners is assuming that once meteorological data are loaded, the temperature and wind speed are automatically valid. However, in some meteorological datasets the wind speed is the value for a representative location, and conditions at the actual installation site may differ. In areas with many surrounding buildings, on mountain ridgelines, in valley terrain, along coasts, on farmland, or on development sites, the way wind flows changes. Differences in wind speed may not be as noticeable as differences in solar radiation, but since they affect evaluations of temperature loss, they should not be ignored.

Also, temperature data can vary depending on the site elevation and surrounding environment. In urban areas, temperatures can tend to be higher due to nearby paving and buildings. In mountainous areas, temperatures can be lower because of elevation differences. In snowy regions, winter temperature conditions can affect power generation and equipment operation. When setting meteorological conditions in PVSyst, it is important to confirm that the monthly average temperatures and seasonal variations are not substantially different from local expectations.

In practice, just as you check solar irradiance, you verify monthly temperatures, wind speeds, and, where necessary, conditions such as snowfall and humidity. Especially when comparing multiple candidate sites during the design phase, judging superiority based solely on irradiance can overlook differences in temperature-related losses and local environmental conditions. When reviewing PVSyst simulation results, be sure to check not only the energy yield but also whether temperature losses or other loss components are unusually large.

The purpose of setting meteorological conditions is not merely to input solar irradiance. It is to model, as realistically as possible, the environment in which a photovoltaic installation will be placed at the site in question. Ambient temperature and wind speed are important elements for enhancing that realism.

Note 5: Check time units and time zone offsets

When importing external meteorological data into PVSyst, verifying the time step and time zone is extremely important. Solar irradiance and temperature data are handled at various granularities, such as monthly, daily, and hourly. In particular, when using hourly data, if the definition of the timestamps is shifted, the correspondence between solar position and irradiance can become inconsistent, which may affect simulation results.

In solar power generation, time, solar altitude, and azimuth are closely related. Whether the solar irradiance occurs in the morning, around noon, or in the afternoon affects the amount of irradiance incident on the panel surface and the influence of shadows. If the timestamps of meteorological data do not match the local standard time of the target region, the peak irradiance will shift away from true noon, and as a result the temporal variation of power generation will appear unnatural.

Time zone discrepancies are particularly likely to occur when handling overseas projects or external data. You need to confirm whether the timestamps in the data are in local standard time or Coordinated Universal Time (UTC), whether they include daylight saving time, and whether the timestamps indicate the start or the end of the period. Even if the data appears to be arranged in hourly intervals, differences in the time reference can cause discrepancies in PVSyst calculations.

Also, solar irradiance data are sometimes recorded as accumulated values over one-hour periods. In such cases, the interpretation depends on whether the timestamp indicates the start of that hour or the end of that hour. For example, if you import data without knowing whether a value at a given time represents the accumulation over the previous hour or over the upcoming hour, the correspondence with the sun position may be offset. When using PVSyst, it is important not only to standardize the file format but also to confirm the time definition.

Attention to time resolution is also necessary. Monthly average data are convenient for rough estimates of long-term annual power generation, but they are limited for evaluating hourly effects, peak output, and detailed loss assessments. Hourly data are suited for detailed analysis, but they are more susceptible to data quality issues, missing values, and time shifts. Depending on the analysis objectives, you must decide which temporal granularity of meteorological data to use.

In practical checks, review monthly graphs and representative days by hour to confirm that the peak in solar radiation occurs at a natural time of day. If solar radiation clearly appears at night, the amount around noon is unnaturally low, or the balance between morning and afternoon does not match local conditions, you should suspect the time or time zone settings.

In PVSyst's meteorological condition settings, time offsets are mistakes that can be hard to notice on the screen. However, their impact on results is far from negligible. Time accuracy is especially important when performing shadow analysis, designing east–west oriented layouts, evaluating tracking systems, or checking hourly generation. Before importing data, confirm the time unit, time reference, time zone, and the handling of daylight saving time, and, if necessary, preprocess the data before use.

Note 6: Understand the difference between long-term average data and measured data

When setting meteorological conditions in PVSyst, it is necessary to understand the difference between long-term average data and measured data. Neither is always correct; it is important to use them appropriately depending on the purpose of the analysis. If you compare results without understanding this difference, you may misinterpret the reasons why simulated values and actual results do not match.

Long-term average data are compiled from meteorological trends over multiple years to represent the conditions of a typical year. They are used in feasibility assessments and in estimating annual power generation during the design phase of solar power plants to grasp the long-term average generation. This is because if a particular year is unusually sunny or, conversely, unusually rainy, the measured values for that year alone make it difficult to judge long-term project viability.

On the other hand, measured data represent the meteorological conditions actually observed during a specific period. When comparing an existing power plant’s generation records with simulations, using data that closely match the measured weather for the target period makes it easier to evaluate equipment performance and losses. For example, if a given year’s generation was lower than expected, you need to check the meteorological conditions for the target period to determine whether the solar irradiance that year was itself lower than the long-term average or whether there was a problem on the equipment side.

A common mistake beginners make is directly comparing the power generation calculated from long-term average data with the actual generation in a specific year. Actual weather varies from year to year. If, in a given year, the rainy season is long, there are many typhoons, winter sunshine is low, or snowfall is heavy, the actual results will naturally differ from the long-term average. If you attribute all of that difference to equipment failure or design mistakes, you will misidentify the cause.

Conversely, caution is also needed when judging long-term business viability using only measured data. If data from a year with unusually high solar irradiance happen to be used, electricity generation may be overestimated. Conversely, if only data from a year with unfavorable weather are used, it may be underestimated. In business viability assessments, it is desirable to use long-term average trends as the benchmark and, where necessary, consider ranges of variability and conservative conditions.

When using PVSyst, it is important at the stage of selecting meteorological data to confirm what the data actually represents. Determine whether it is a long-term average, actual measurements for a specific year, or based on multi-year reanalysis; what the target period is; and whether there has been gap-filling for missing data. When explaining results internally or externally, you can prevent misunderstandings by clarifying assumptions such as "this simulation is based on long-term average meteorological data" or "this comparison reflects the meteorological conditions of the target year."

When setting meteorological conditions, not only the accuracy of the data but also its suitability for the intended purpose is important. Consider which data to use depending on the situation: long-term average data are suitable for planning, while measured data are suitable for performance analysis.

Note 7: Do not use outliers and missing values as-is

When importing meteorological data into PVSyst, you should always check for outliers and missing values. Even if the data format is correct and the import succeeds, the dataset may contain missing or anomalous values. Using it in a simulation as-is can affect monthly energy production and loss calculations, reducing the reliability of the results.

Missing values refer to a situation where data for times or periods that should exist are absent. For example, some hours may be missing in hourly data, there may be blank cells in columns for solar radiation or temperature, or there may be an unusually small amount of data for certain months. If missing values are few, they can sometimes be filled (imputed), but using a period with many missing values as-is can lead to underestimating solar radiation compared with reality or to distorted averages.

Anomalous values are values that are realistically impossible or highly suspicious. If, for example, solar radiation is large at night, solar radiation during the day is extremely high, temperatures are unnatural given the region’s climate, wind speed is always zero, or the same value repeats only over a specific period, you should suspect a data anomaly. Simply mechanically formatting a file may fail to reveal such anomalies.

Anomalous solar irradiance values in particular directly affect power output. If irradiance is recorded at night, it may be caused by a time shift or misidentification of data columns. If daytime values are extremely high, possible causes include differences in units, confusion between cumulative and average values, digit errors, or conversion mistakes. Before importing into PVSyst, checking the annual total, monthly totals, maximum, minimum, and a representative day's time series makes it easier to spot anomalies.

When correcting missing data or outliers, it is also important to record the correction method. Whether you imputed values from neighboring data, interpolated from preceding and following values, excluded the affected period, or replaced them with a long-term mean will change the interpretation of the results. When reviewing simulation results later, reproducibility is lost if it is not clear what processing was applied.

Also, finding an anomalous value does not mean it should automatically be deleted. In some regions and seasons, extremely strong solar radiation, sudden temperature changes, or strong winds can actually occur. The important thing is to assess values that appear suspicious by cross-checking them against local conditions and the data specifications. Deleting them lightly can, conversely, cause you to lose genuine local characteristics.

When configuring meteorological conditions in PVSyst, make it a habit to check data quality both before and after loading the data. The absence of loading errors does not mean the data are valid. Rather than only looking at PVSyst’s result screens, check the distributions and seasonal variations of the meteorological data itself to improve the reliability of the simulation.

How to Conduct Practical Checks to Avoid Failures When Setting Meteorological Conditions

We have explained seven points so far, but in practice it is more efficient not to check them separately every time; instead, check them in a consistent sequence. When you are still getting used to using PVSyst, you tend to be fully occupied with learning the screen operations, but standardizing the procedure for setting meteorological conditions can reduce mistakes.

First, compile the basic information for the target location. Check latitude, longitude, elevation, the terrain of the installation site, the surrounding environment, the presence or absence of snow cover, the distance from the coast, and influences from mountains and buildings. Next, verify the source and generation conditions of the meteorological data to be used, and examine the distance to the target location and any elevation differences. At this stage, determine whether the meteorological data can be considered representative of the target site.

Next, check the type of solar radiation data. Determine whether it is horizontal-plane radiation or tilted-plane radiation, whether it includes direct and diffuse components, what the units are, and whether the time resolution is monthly or hourly. In PVSyst, it is important that the data be in a format that can be read by the settings screen, but even more important that the imported values match the purpose of the analysis.

After that, we check temperature, wind speed, and other meteorological factors. Rather than proceeding based only on solar irradiance, we confirm whether the monthly temperatures match the local climate, whether wind speeds are extreme, and, when necessary, consider conditions such as snowfall and humidity. In power generation simulations, solar irradiance is unquestionably the main factor, but surrounding meteorological conditions also affect temperature-related losses and the evaluation of the installation environment.

When importing external data, check the time resolution and time zone. Verify whether the time reference is local time, whether daylight saving time is handled, and whether the timestamps for hourly values indicate the start or the end of the hour. If possible, check a time-series graph for a representative day and see whether the rise, peak, and evening decline of solar irradiance look natural. If solar irradiance is recorded at night or the daytime peak is unnaturally shifted, re-examine the data's time settings.

Finally, check for anomalous values and missing values and correct them as necessary. If you make corrections, record what processing was performed. When submitting simulation results, it is important to be able to explain the rationale for the input conditions. PVSyst’s calculation results are convincing only when the input conditions are clearly specified.

For practitioners, the important thing is not to treat PVSyst as "software that gives an answer if you input numbers." PVSyst is a powerful tool for power generation simulation, but it is the user's role to assess the validity of the input conditions. When configuring meteorological conditions, proceeding while confirming the meaning of the data, its suitability to the site, and its consistency with the analysis objectives is the most reliable way to prevent failure.

Summary: The accuracy of power generation simulations depends heavily on an understanding of local conditions

To avoid failures when configuring meteorological conditions in PVSyst, it is important not only to load meteorological data but also to verify what the data represent and how well they match the target site. The displacement between site information and meteorological data, the type of solar radiation data, the difference between global horizontal irradiance and tilted-plane irradiance, air temperature and wind speed, time resolution and time zone, the difference between long-term average data and measured data, and the handling of outliers and missing values all affect the reliability of energy yield simulations.

When learning how to use PVSyst, it's easy to focus on which parts of the interface to click. However, what is required in practice is not just the ability to operate the software. It is important to be able to explain why a particular meteorological dataset was used, whether it can be said to represent the location, and whether the results are consistent with on-site conditions. In particular, when explaining expected energy production to clients or stakeholders, the rationale behind the input assumptions determines how convincing the results are.

Also, meteorological conditions cannot be concluded from desk data alone. There is much information that must be verified on site, such as local topography, the surrounding environment, elevation, shading factors, snowfall, wind flow patterns, and ground conditions after development. If the meteorological and terrain conditions set during the design phase differ from the actual on-site conditions, discrepancies between the simulation results and the actual power generation are likely to occur. To improve the accuracy of analyses using PVSyst, it is essential to link data-based conditions with on-site conditions and verify them in the field.

In the design and energy-yield assessment of solar power generation, it is important to consider meteorological conditions, topography, and equipment conditions together. If coordinates and elevations set in the office, on-site terrain variations, equipment layout, and shading factors can be accurately determined, PVSyst input parameters can be made more realistic. Obtaining high-precision positional information during field surveys and surveying is useful for later-stage design, simulation, construction management, and post-completion verification.

Therefore, if you want to tie the meteorological condition settings and power generation simulations in PVSyst more closely to practical work, it is effective to also put in place a system for accurately acquiring on-site location information. LRTK is an iPhone-mounted GNSS high-precision positioning device that acquires coordinates and elevation on-site, offering an option to make the location information needed for surveying, design, and construction management easier to handle. By conducting desktop power generation simulations in PVSyst and using LRTK to capture on-site coordinates and terrain conditions, you can improve the accuracy of solar power generation project assessments and more easily reduce discrepancies between plans and the field.