Seven Essential Types of Solar Irradiance Data Used in PV Generation Simulations

In photovoltaic (PV) generation simulations, not only the system capacity and panel performance but also the type of solar irradiance data used can greatly affect the apparent annual generation. Even with the same installed capacity, azimuth, and tilt angle, simulation results differ if the input irradiance data vary in type, period, regional resolution, or correction methods.

In practice, generation forecasts are used for business planning, design review, financial checks, post-construction comparisons, and operational improvements. If you accept numbers without understanding what the irradiance data represent, you may make erroneous decisions. This article organizes and explains seven basics that practitioners searching for "solar power generation simulation" should grasp when examining irradiance data.

Table of Contents

• Irradiance data are the starting point of generation simulations

• Basic 1 Understand the difference between irradiance and sunshine duration

• Basic 2 Distinguish between horizontal-plane irradiance and tilted-plane irradiance

• Basic 3 Grasp the meanings of direct, diffuse, and global irradiance

• Basic 4 Understand when to use annual, monthly, and hourly values

• Basic 5 Don’t underestimate regional differences and site corrections

• Basic 6 Confirm the difference between long-term averages and single-year data

• Basic 7 Organize how to treat measured data and modeled data

• Points to note when converting irradiance data to generation

• Practical decision points when reviewing simulation results

• Summary

Irradiance data are the starting point of generation simulations

Put simply, a PV generation simulation calculates how much solar energy a panel at a given location receives and how much of that can be extracted as electricity. The primary input for this calculation is irradiance data.

Irradiance indicates the amount of solar energy reaching the ground or the panel surface. In PV generation, the amount of light arriving at the panels is a major premise for projected output. Therefore, no matter how high-performance the equipment assumed, if the irradiance assumption diverges from reality, the simulation results will also diverge.

A common practical problem is looking only at the displayed annual generation in simulation results without checking the contents of the irradiance data. Even if annual generation appears large, if the irradiance data used are based on optimistic conditions, the actual generation may differ substantially. Conversely, if conservative irradiance data are used, simulated values may appear lower.

In PV planning, you handle many factors—system capacity, panel azimuth, tilt angle, surrounding shading, temperature, and loss rates—but irradiance data are the most fundamental condition among them. Before memorizing calculation formulas or tool operations, understanding the types and characteristics of irradiance data is the first step to improving generation forecast accuracy.

Basic 1 Understand the difference between irradiance and sunshine duration

When handling irradiance data, the first common confusion is between irradiance and sunshine duration. Both relate to sunlight, but their significance for generation simulations differs.

Sunshine duration is an indicator of how long sunlight above a certain threshold was observed. Regions with long periods of clear weather have longer sunshine duration, while cloudy or rainy regions have shorter durations. It is an intuitive metric and useful for rough comparisons of candidate PV sites.

Irradiance, on the other hand, indicates the actual amount of solar energy that arrived, not just how long the sun was present. For PV generation computations, irradiance is a more direct input than sunshine duration. For example, the same one hour of sunlight at noon in summer and at dusk in winter delivers different amounts of energy to the panels due to differences in solar elevation. Also, thin clouds may still count toward sunshine duration while significantly reducing irradiance.

In practice, sunshine duration may appear in regional comparison materials, but it does not necessarily translate directly into usable input for generation simulation. You cannot simply say that longer sunshine duration always means higher generation or that shorter sunshine duration is always disadvantageous. What matters is confirming irradiance as the amount of energy panels receive.

Irradiance is treated as energy per unit area. In practical materials, it is often presented on a daily, monthly, or yearly basis, and generation simulations combine these values with system capacity, conversion efficiency, and various losses. Understanding the difference between irradiance and sunshine duration helps you know which data to check when verifying simulation results.

Basic 2 Distinguish between horizontal-plane irradiance and tilted-plane irradiance

A particularly important distinction in PV generation simulations is between horizontal-plane irradiance and tilted-plane irradiance. Horizontal-plane irradiance refers to the irradiance reaching a plane parallel to the ground. Meteorological observations and regional baseline data are often organized as horizontal-plane irradiance.

However, PV panels are not always installed horizontally. Roof-mounted panels are installed along the roof slope, and ground-mounted systems are typically given a tilt angle for efficiency or site conditions. Therefore, to calculate generation, you need the irradiance that reaches the panel surface—i.e., tilted-plane irradiance.

Converting horizontal-plane irradiance to tilted-plane irradiance involves latitude, azimuth, tilt angle, seasonal solar elevation, and how diffuse irradiance is handled. Panels with a south-facing orientation and an appropriate tilt may outperform the horizontal plane over a year. Conversely, east- or west-facing panels or low-tilt roofs show different patterns for morning/evening or summer generation, and annual generation depends on installation conditions.

When reviewing simulation results, check whether the input irradiance is a horizontal-plane value or has been converted to the tilted-plane value reflecting installation conditions. Judging a region as having high annual irradiance from horizontal-plane data alone may be misleading if the actual roof orientation or tilt is unfavorable.

Roof installations in particular are affected by orientations that are not due south, multiple roof planes, varying pitches, and shading from surrounding buildings. In PV simulations, it is important not only to read regional irradiance data but to look at the conditions after converting them to the installation surface.

Basic 3 Grasp the meanings of direct, diffuse, and global irradiance

Irradiance data are categorized into direct irradiance, diffuse irradiance, and global irradiance. Practitioners need not memorize all formulas, but understanding these meanings deepens comprehension of simulation results.

Direct irradiance is light that reaches the ground or panel directly from the sun. Strong irradiance that produces distinct shadows on clear days is largely due to direct irradiance. Because the sun’s direction is distinct, direct irradiance is strongly affected by panel azimuth and tilt.

Diffuse irradiance is light scattered by water vapor, clouds, dust, and other particles in the atmosphere, arriving from various directions. Even on cloudy days, the surroundings can be bright and panels can produce some power because of diffuse irradiance. Diffuse irradiance is less directional than direct irradiance, so its sensitivity to azimuth and tilt differs.

Global irradiance is the sum of direct and diffuse irradiance. PV generation simulations typically use global irradiance as a basis, then estimate generation by converting to the installation surface and considering shading effects.

These differences matter in atypical installation conditions. For example, east- or west-facing roofs, low-tilt angles, complex shading in mountainous or urban areas, or regions with frequent cloud cover make it difficult to capture generation characteristics using simple annual irradiance alone. In areas with abundant direct irradiance, optimization of azimuth and tilt yields larger differences, whereas in areas where diffuse irradiance constitutes a large share, irradiance distribution is more complex.

Simulation input screens and result reports may not explicitly display these terms. However, since conversion calculations often assume specific irradiance types, understanding direct, diffuse, and global irradiance helps interpret the numbers.

Basic 4 Understand when to use annual, monthly, and hourly values

Irradiance data come in various granularities: annual, monthly, and hourly values. What you can learn from a simulation changes depending on the chosen granularity.

Annual values are used to get a rough estimate of the irradiance over one year. For early-stage PV assessments, they are effective for comparing candidate sites and estimating annual generation. However, annual values alone do not show seasonal fluctuations or time-of-day generation trends.

Monthly values are used to confirm seasonal variations from January through December. In PV generation, output tends to be higher in summer but panel temperature rise can reduce output. In winter, shorter daylight hours and lower solar elevation reduce irradiance, but lower temperatures can reduce thermal losses in panels. Viewing monthly irradiance helps break down annual generation and makes it easier to plan seasonal feed-in, self-consumption, and operations.

Hourly values are finer-grained data that show irradiance variations by hour. Using hourly data makes it easy to confirm generation patterns in the morning, midday, and evening. For self-consumption PV systems, matching generation times with demand is as important as annual generation. Therefore, hourly irradiance-based simulations aid in comparing consumption, evaluating storage systems, and considering peak shaving.

In practice, annual or monthly values may suffice for early assessments, but as you move into business decisions and detailed design, it becomes more common to check hourly data. Especially when there are multiple roof orientations, combinations of east/west faces, or shading effects that vary by time of day, annual values alone cannot capture reality.

The granularity of irradiance data affects not only simulation accuracy but also the purpose of the analysis. Choose the level of detail appropriate for whether the task is a rough estimate, design comparison, financial analysis, or post-operation performance comparison.

Basic 5 Don’t underestimate regional differences and site corrections

Irradiance varies greatly by region. Even within the same prefecture, coastal, inland, mountainous, basin, and urban areas differ in cloud patterns, fog, snowfall, surrounding topography, and temperature. In PV generation simulations, relying on a regional representative value can cause discrepancies with the actual installation site.

Regional irradiance data are often treated as representative values for a fairly wide area. They are convenient for preliminary assessments, but caution is required when the installation site is distant from the representative point or has significantly different terrain. For example, within the same municipality, the north and south sides of a mountain, coastal areas versus high elevation locations, or open plains versus valleys can have different irradiance environments.

Site correction means adjusting representative irradiance data to better reflect the actual installation site. Consider latitude/longitude, elevation, surrounding topography, snowfall, and local tendencies for cloud formation. It is not easy to correct everything precisely, but at least confirming how closely the installation site matches the representative data improves decision quality.

For residential or small facilities, you may approximate using nearby regional data. For commercial PV projects or cases where generation significantly influences revenue, using site-appropriate irradiance data becomes more important. Especially when explaining to financial institutions, making investment decisions, projecting long-term generation, or planning maintenance, carefully checking the regional specificity of irradiance data is essential.

Regional differences affect not only annual generation but seasonal generation patterns. In snowy areas, winter irradiance and panel surface snow effects are important; on coasts, salt damage and fog effects should be considered. In mountainous areas, terrain-induced morning and evening shading is significant, while in urban areas, shading from surrounding buildings is a concern. When reading irradiance data, don’t just compare magnitudes; consider the installation site environment in context.

Basic 6 Confirm the difference between long-term averages and single-year data

In PV simulations, you may use long-term average irradiance data or measured/estimated data for a specific year. Failing to understand this difference can lead to misinterpretation when comparing simulations to actual generation.

Long-term average data are averages of irradiance over multiple years. They are suitable for estimating long-term generation prospects. Because PV systems are operated over long periods, long-term averages help assume standard conditions that are not influenced by a single year’s weather.

Single-year data reflect the weather of a specific year. If a year has many sunny days, generation is likely to be high; if it has many rainy or cloudy days, generation is likely to be low. Therefore, if actual generation in a given year falls below the simulated value, it is premature to immediately assume equipment failure or design faults. That year’s irradiance may have been below average.

In practice, it is important to confirm whether the simulation assumption is based on long-term averages or the weather of a target year. Business planning often uses long-term average generation, while post-operation verification requires comparison adjusted to that year’s irradiance. Rather than simply seeing “actual generation is lower than simulated,” evaluate whether the generation was reasonable given that year’s irradiance conditions.

Long-term averages also have caveats. An average indicates a typical trend but does not mean each year will match it. Generation fluctuates year to year; some years will exceed simulations and others will fall short. Therefore, financial planning and operational evaluations require considering possible variability as well as averages.

When using irradiance data, ideally confirm “how many years of data,” “what period is averaged,” and “how abnormal years are handled.” Even if you cannot grasp detailed statistical processing, just being aware of the difference between long-term averages and single-year data greatly improves how you read simulation results.

Basic 7 Organize how to treat measured data and modeled data

Irradiance data include measured data from observation sites and modeled data estimated from observations or meteorological models. Neither is universally correct; understanding their characteristics and using them appropriately is important.

Measured data are values recorded by actual observation instruments. If the data are from a point near the installation site, they are highly reliable for decision-making. However, observation points are limited, and there may not be a measurement station close to the planned site. Data quality can also be affected by instrument maintenance, missing data, or changes in the instrument environment.

Modeled data are advantageous for covering wide areas. Even in regions with few observation points, irradiance can be estimated via geographic interpolation or meteorological analysis. You can sometimes obtain mesh-based data close to the installation site, which helps capture regional differences. However, modeling methods can introduce errors, so it is necessary to understand accuracy limits when using them.

In practice, combining measured and modeled data for cross-checking is effective. For example, use nearby observation data to verify general plausibility and use modeled data close to the site to supplement regional differences. If multiple datasets show large discrepancies, do not unconditionally adopt one; instead check terrain, climate, distance to observation points, and data periods.

Also note that improving irradiance data accuracy alone does not always greatly improve overall simulation accuracy. Generation is affected by many factors: panel temperature, inverter conversion loss, wiring losses, soiling, snow, aging degradation, shading, and downtime. Irradiance data are an important input, but final generation is determined by the accumulation of multiple conditions.

Therefore, when handling irradiance data, balance precision with practical usability. For initial assessments, approximate with standard modeled data; for detailed studies, confirm regional and measured values; and during operation, compare against actual performance and adjust.

Points to note when converting irradiance data to generation

Merely checking irradiance data does not determine generation. In PV simulations, converting irradiance to generation involves various corrections and losses. Without understanding this conversion, it’s hard to see why high irradiance does not necessarily lead to correspondingly high generation.

First, PV panels cannot convert all received solar energy into electricity. Panels have conversion efficiencies, and actual operating environments reduce output due to temperature rise. In summer, irradiance is high but panel temperature also rises, so actual output can be lower than expected from irradiance alone.

Next, DC power from panels experiences losses when converted to AC. There are wiring losses, equipment conversion losses, and connection-related losses. Additionally, panel soiling, bird damage, fallen leaves, snow, surrounding shading, and equipment downtime reduce generation.

Even with the same irradiance, simulation results can differ depending on assumed loss rates. In practice, you must check both the irradiance and what losses are assumed. If optimistic loss rates are used, annual generation tends to look high. If conservative loss rates are set, actual generation may exceed simulated values.

Shading treatment is also important in generation simulations. Shading from nearby buildings, trees, utility poles, mountains, or rooftop obstacles cannot be fully represented by regional irradiance data. Even in regions with good irradiance, shading on the installation surface will reduce generation. Because solar elevation is lower in mornings, evenings, and winter, shading effects often become larger at those times, so checking hours of shading as well as annual values is desirable.

When converting irradiance to generation, not only the accuracy of input data but the realism of design conditions is tested. Do not take simulation values at face value; confirm that irradiance, installation surface, losses, shading, and operational conditions are consistent to avoid practical failures.

Practical decision points when reviewing simulation results

When reviewing PV simulation results, it is important to confirm the basis step by step, not just the annual generation number. Practitioners should check the type of irradiance data, the period, site appropriateness, installation conditions, loss assumptions, and result variability.

First, identify which region the irradiance data represent. Check whether the data come from a location far from the planned site, and whether regional characteristics like mountainous terrain or coastal conditions have been disregarded. Next, confirm whether horizontal-plane irradiance is used as-is or whether it has been converted to tilted-plane irradiance reflecting panel azimuth and tilt.

Then check whether the data are long-term averages or specific-year data. Business plans typically use long-term averages, while post-construction verification requires comparison adjusted to the year in question. Simple comparisons between long-term average generation and single-year actuals may lead to misattributing weather variability to equipment issues.

Also review monthly generation. Even if annual generation is the same, differences in monthly distribution affect finances and self-consumption assessments. Checking whether generation is concentrated in summer, whether winter drop-off is large, and how much generation occurs in mornings and evenings helps judge equipment usage and alignment with demand.

Once actual performance data are available, compare generation with irradiance for the same period. It is natural for generation to decline in months with less irradiance than average. However, if irradiance was sufficient and generation dropped significantly, inspect shading, soiling, equipment faults, downtime, wiring, and system settings.

When using simulation results in explanatory materials, clearly state assumptions. Reports that omit irradiance data type, installation conditions, loss rates, shading considerations, and calculation periods make later verification difficult. For PV simulations, the assumptions behind the numbers are more important than the numbers themselves.

Summary

Irradiance data used in PV generation simulations are the foundation of generation forecasts. Understanding the difference between irradiance and sunshine duration, distinguishing horizontal-plane and tilted-plane irradiance, and grasping direct, diffuse, and global irradiance can greatly change how simulation results appear.

Choosing annual, monthly, or hourly values determines what you can learn. Annual and monthly values work for early assessments, but for design comparisons, self-consumption, storage, and post-operation verification, hourly data are more useful. Always check regional differences, site corrections, the difference between long-term averages and single-year data, and the characteristics of measured versus modeled data.

Generation does not simply increase in direct proportion to irradiance. Panel azimuth, tilt, shading, temperature, losses, soiling, snow, and equipment downtime combine to determine final generation. Therefore, irradiance data should not be evaluated in isolation but read together with installation and operational conditions.

For site surveys and pre-design checks, accurately recording the planned location, azimuth, tilt, surrounding obstacles, site boundaries, and roof shape is indispensable for translating irradiance data into actual generation. Improving the precision of desktop simulations requires equally precise input of on-site conditions.

As a practical tool to efficiently record on-site conditions and clarify design and simulation assumptions, LRTK, an iPhone-mounted GNSS high-precision positioning device, is effective. If you can retain high-precision location information, surrounding environment details, and records of on-site checks, it becomes easier to link irradiance data with site conditions in your evaluations. To avoid leaving PV generation simulations as mere rough estimates and to support sound practical decisions, it is crucial both to understand irradiance data and to establish a process for accurately capturing site information.