7 Ways to Improve the Accuracy of Solar Power Generation Simulations

Table of contents

• Why the accuracy of solar power generation simulations matters in practice

• Method 1 Reconsider how you choose irradiance data

• Method 2 Accurately reflect site conditions

• Method 3 Match azimuth and tilt to on-site conditions

• Method 4 Do not underestimate the impact of shading

• Method 5 Enter equipment specifications and system configuration correctly

• Method 6 Set loss conditions realistically

• Method 7 Update simulations by comparing with measured values

• Common mistakes to avoid to improve accuracy

• Conclusion

Why the accuracy of solar power generation simulations matters in practice

Solar power generation simulation is an important task for grasping the expected generation output in advance when planning a photovoltaic system. In the design stage, you examine annual generation, monthly generation, electricity sold, self-consumption, breakdown of losses, and the appropriateness of system scale. Therefore, if the simulation accuracy is low, it will affect the business plan, investment decisions, system design, and operation planning.

For practitioners, a more frequent problem is not the calculation itself but the validity of the input conditions. Solar generation depends on multiple factors such as irradiance, temperature, installation angle, azimuth, shading, equipment performance, wiring, degradation, soiling, output control, and the surrounding environment. If any one input condition deviates significantly from reality, the annual generation forecast can differ substantially.

Solar power generation simulation is not merely producing numbers. What matters is how accurately you read the on-site conditions, how faithfully you reflect equipment specifications, and how realistically you assume future operating conditions. In other words, improving accuracy requires not only mastering calculation tools but also treating pre-input surveys, organizing design conditions, on-site verification, and post-operation validation as a continuous workflow.

This article explains seven methods to improve the accuracy of generation simulations for practitioners who search for "solar power generation simulation." Rather than memorizing specialized formulas, it focuses on often-overlooked practical checkpoints and organizes approaches that can be used from planning through operation.

Method 1 Reconsider how you choose irradiance data

The most fundamental factor that determines the accuracy of solar power generation simulation is the irradiance data. Because photovoltaics convert energy received from the sun into electricity, if the assumed irradiance is off, no matter how detailed the subsequent inputs are, the generation forecast is likely to diverge from reality.

In practice, it is common to use data from a meteorological observation point near the planned installation site or regional irradiance data. However, even within the same municipality, coastal areas, mountainous areas, basins, urban areas, and snow-prone regions have different cloud formation, fog, temperatures, wind, snowfall, and surrounding topography. In particular, sites near mountains or in valley topography may have sunrise and sunset irradiance conditions that do not match standard regional data.

When looking at irradiance data, it is important not to judge solely by annual averages. Even if annual totals look similar, monthly generation can vary greatly. For example, in regions with low winter irradiance, annual values alone may seem reasonable, but actual revenue or self-consumption plans may suffer from winter generation shortages. Conversely, in areas prone to high temperatures or output curtailment in summer, simply having abundant irradiance does not necessarily translate into correct generation estimates.

To improve accuracy, pay attention to the period covered by the irradiance data. Using data from a single year can bias simulation results depending on whether that year happened to be unusually sunny or rainy. Where possible, it is desirable to use data that reflect multi-year averaged trends. In actual business planning, checking conservative scenarios as well as optimistic generation estimates helps prepare for unexpected shortfalls.

Irradiance data come in various types such as horizontal plane irradiance, tilted plane irradiance, direct irradiance, and diffuse irradiance. Since panels are usually installed at a certain tilt, you should consider irradiance on the tilted plane corresponding to the installation angle and azimuth rather than using horizontal plane irradiance directly. Even when the simulation software automatically converts values, it is important to understand what the input data represent.

Selecting irradiance data is the foundation of generation simulation. If you want to improve accuracy, start by confirming "which location, which period, and which type of irradiance data are being used." If this is left ambiguous and you proceed to detailed settings, it becomes difficult later to trace the cause of errors.

Method 2 Accurately reflect site conditions

In solar power generation simulation, how accurately you can reflect site conditions is crucial. The site is not just the address or latitude/longitude. You need to consider on-site factors that affect generation such as elevation, topography, surrounding buildings, land slope, roof shape, snowfall, ventilation, temperature environment, and surrounding reflectance.

If you run a simulation using only the address, you may capture broad meteorological conditions but fail to reflect site-specific conditions adequately. For example, even within the same area, installing on open flat land differs from installing on a plot surrounded by buildings or trees in terms of how shading occurs at sunrise and sunset. Even for rooftop installations, roof orientation, pitch, ridge position, adjacent roof surfaces, antennas, and rooftop equipment affect performance.

Topography cannot be overlooked. In mountainous areas, the surrounding mountains can block sunlight just after sunrise or before sunset. Even during times that broad-area irradiance data indicate sunny conditions, the site may have periods where generation is not possible due to terrain shadows. Failing to reflect such conditions tends to overestimate generation, especially at dawn, dusk, and in winter.

Elevation and ventilation also affect generation. Panels tend to lose output when they become hot. Therefore, even with the same irradiance, actual generation differs between well-ventilated locations where panel temperature rises less and locations where heat accumulates on the roof surface. Installations close to roof surfaces or with limited ventilation should account for temperature losses carefully.

In snowy regions, snow has a major impact on winter generation. When panels are covered by snow, generation drops sharply during that period. Partial snow coverage can also cause shading and uneven generation. Calculating annual generation without considering snow often results in overestimating winter generation.

To accurately reflect site conditions, avoid relying solely on drawings and maps and conduct on-site verification whenever possible. Organize and reflect in the simulation photos taken on site, the heights of surrounding buildings, tree positions, rooftop equipment layout, ground undulation, azimuth, and slope. Doing so increases the reliability of results. Especially for large-scale systems or simulations used for business decisions, insufficient on-site condition verification can lead to significant errors later.

Method 3 Match azimuth and tilt to on-site conditions

Setting azimuth and tilt is very important for improving simulation accuracy. The amount of irradiance a panel receives changes depending on which direction it faces and at what angle it is installed. Although orientations and angles that efficiently capture sunlight are generally desirable, practical design must consider roof shape, site conditions, racking layout, wind load, snow, constructability, and maintainability.

A common simulation mistake is leaving provisional azimuth and tilt inputs unchanged as the final design evolves. Even if you used standard angles in initial studies, panel layouts may change as design progresses. Panels may be installed along actual roof pitch, arrays may be aligned with site boundaries, maintenance pathways may be provided, or the number of tiers changed to avoid shading. If those changes are not reflected in the simulation, results will differ from actual generation.

Azimuth is particularly prone to errors in rooftop installations. Although a roof may appear south-facing on a drawing, it may be offset from true south in reality. Building layouts that follow roads or site boundaries often mean roof surfaces do not align with exact north-south axes. Even small azimuth deviations affect annual generation, morning and evening generation patterns, and seasonal output.

For tilt, entering a standard value is insufficient. When installing along a roof slope, confirm the actual roof pitch. For flat roofs or ground-mounted systems, enter the designed racking angle, which may also change during design. Increasing tilt can improve winter irradiance reception but can increase wind effects and row-to-row shading issues. Decreasing tilt can allow higher packing density but may affect soiling runoff and drainage.

The treatment of azimuth and tilt also depends on the simulation purpose. Whether you aim to maximize annual generation, prioritize morning and evening self-consumption, or emphasize winter generation, the optimal layout differs. For self-consumption systems, a configuration that generates during demand hours can be more advantageous than one that simply maximizes annual generation. Therefore, in simulation you should check not only annual totals but also hourly and monthly trends.

Azimuth and tilt may look like simple input fields in calculations, but they reflect the design philosophy. To improve accuracy, measure or confirm angles on-site or via drawings, and always update simulation conditions when design changes occur.

Method 4 Do not underestimate the impact of shading

Shading is a frequent source of large errors in solar power generation simulations. Panels generate most stably when light is uniformly incident across their surface. When buildings, trees, utility poles, fences, rooftop equipment, adjacent arrays, or mountain topography cast shadows on part of the array, generation decreases. The impact of shading is not necessarily proportional to the shaded area, so it must be treated carefully.

Time and season are important in shading assessment. A location may be unshaded at one time and shaded at another. Even if summer poses few problems, low solar elevation in winter can create long shadows. Morning and evening shading are often overlooked, but for self-consumption systems morning and evening generation can be critical. Looking only at annual generation can make shading effects appear small, while generation shortages at specific times may create operational problems.

Rooftop equipment-induced shading is a common practical issue. Ventilation, air conditioning, lightning protection, railings, penthouses, antennas, etc., may seem small on drawings but cast long shadows when solar elevation is low. For ground installations, insufficient spacing between panel rows can allow front-row shadows on rear rows. Packing rows too tightly to increase nominal capacity can reduce actual generation efficiency.

Tree shading requires attention due to future changes. A site may look fine at the time of inspection, but trees can grow over several years and increase shading. Deciduous and evergreen trees have different seasonal shading characteristics. Even deciduous trees can still cast shadows from trunks and branches in winter. As neighboring trees and buildings are not under your control, long-term operation requires a margin in evaluation.

To correctly reflect shading, organize on-site photos, the heights, distances, and azimuths of surrounding obstacles, and the sun paths for each season. In simple simulations, shading may be entered as a rough loss rate, but for complex sites a three-dimensional consideration of obstacles is preferable. Especially for rooftop and urban projects, shading inputs strongly affect simulation accuracy.

Shading also affects operational stability in addition to generation. Locations with repeated partial shading are prone to output variability, requiring attention to equipment configuration and circuit design. Checking for shading at the simulation stage contributes to design risk reduction beyond simple generation forecasting.

Underestimating shading can lead to a situation where calculated generation appears sufficient but expected generation is not achieved after operation starts. To improve accuracy, treat shading not as a minor margin but as an important item to carefully verify based on site conditions.

Method 5 Enter equipment specifications and system configuration correctly

In solar power generation simulations, not only irradiance and site conditions but also the accuracy of equipment specifications and system configuration inputs is important. Panel modules, power conversion equipment, wiring, junction boxes, circuit configurations, oversizing ratios, and DC/AC capacity relationships affect final generation.

First check panel specifications. Nominal power, temperature coefficient, power tolerance, degradation rate, and low-irradiance characteristics relate to generation calculations. Entering only nominal power may fail to fully reflect temperature conditions and aging. High-temperature output losses notably affect summer generation. Even in seasons with abundant irradiance, high panel temperatures can prevent expected output, so handling of the temperature coefficient is important.

Power conversion equipment specifications also directly affect generation. Conversion efficiency, input voltage range, maximum input current, rated output, number of circuits, and control characteristics must be correctly reflected. In designs where DC panel capacity exceeds AC conversion capacity, output can hit limits under certain conditions and excess generation is clipped. Failing to consider this may lead to overestimating generation or misjudging the appropriateness of system capacity.

In system configuration, string counts, parallel strings, circuit partitioning, and how surfaces with different azimuths or tilts are treated are important. For the same nominal capacity, installations split across multiple roof surfaces will have different irradiance and generation time profiles. Treating everything as a single averaged condition can fail to represent actual output variability. When east-west, south, north-leaning, or differently tilted surfaces coexist, evaluate them separately.

Wiring losses are also easy to overlook. Losses arise from cable length between panels and inverters, cable cross-section, circuit configuration, and connection condition. Simulations often use standard values, but for large systems or sites with long cables, you need to revise values to match actual design. Underestimating wiring losses can inflate annual generation estimates.

Also be mindful that equipment may be changed during design. Even if provisional specifications were used in initial studies, procurement, construction, delivery, and design changes can result in different actual equipment. Changes affect output, efficiency, temperature characteristics, capacity ratios, and connection conditions. When using simulation results for business plans or reports, always confirm they match the final equipment specifications.

Because equipment specs and system configuration involve many input items, the work can become cumbersome. However, correctly organizing these inputs brings simulations closer to reality. To improve accuracy, do not merely transcribe catalog specs or drawings, but understand how those specifications affect generation and enter them accordingly.

Method 6 Set loss conditions realistically

In solar power generation simulation, irradiated panels do not generate ideally and deliver power without loss. In reality, various losses occur such as temperature rise, wiring, conversion, shading, soiling, snowfall, aging, equipment downtime, output control, and circuit variability. Improving accuracy requires setting these loss conditions realistically.

A common issue is continuing to use default standard values. Standards are convenient for preliminary studies but do not fit every site. For example, in areas with a lot of dust, frequent bird droppings, roofs prone to leaf accumulation, coastal sites exposed to salt, or heavy-snow regions, soiling and deposition losses may be large. Conversely, if regular cleaning and inspection are planned, loss expectations change.

Temperature losses are important at many sites. Although panels generate more under strong irradiance, output drops as temperature rises. Therefore, in hot regions or poorly ventilated installations, you must appropriately account for temperature losses. Panel temperature varies with roof material, installation height, ventilation, and surrounding thermal environment. Higher irradiance does not automatically mean higher generation; evaluate in combination with temperature conditions.

Aging is also important for long-term generation forecasts. When considering not only first-year generation but generation several or many years later, you need to reflect panel output decline. Business planning often considers generation over the operation period rather than only first year, so degradation rate settings affect profitability assessments. Using overly optimistic degradation rates can overestimate long-term generation.

Output control and downtime are also important for realistic generation. Depending on grid conditions and operational constraints, output may need to be curtailed even when irradiance allows generation. Inspections, faults, communication outages, and equipment replacement can cause downtime. Simulations tend to assume ideal continuous operation, but in practice it is realistic to include some stoppage risk.

Loss settings should not simply be large across the board. Excessively conservative assumptions can undervalue the system. Conversely, overly optimistic settings increase the likelihood of shortfalls after operation begins. The important thing is to set explainable loss conditions based on site conditions, equipment specifications, maintenance plans, and operating environment.

High-accuracy simulations do not treat losses as a single lump-sum but break down which losses are likely and how much each contributes. Organizing losses so their breakdown is visible makes it easier to find opportunities for design improvements. For example, if shading loss is large, reconsider layout; if wiring loss is large, revise wiring planning; if temperature loss is large, review ventilation or racking conditions—use simulation results to drive concrete improvement actions.

Method 7 Update simulations by comparing with measured values

To truly improve the accuracy of solar power generation simulations, it is important to compare predictions with measured values after operation begins and update input conditions as necessary. Simulations are predictive at the planning stage and, no matter how carefully you enter conditions, cannot perfectly reproduce the actual operating environment. Comparing measured operational data with predictions provides insights useful for future projects and operational improvements.

When comparing with measured values, check monthly, daily, and hourly data rather than only annual totals. Even if annual generation roughly matches, if there are patterns such as higher summer and lower winter generation, discrepancies in morning/evening timing, or large deviations in specific months, there may be room to improve input conditions. Monthly discrepancies can indicate irradiance data, snow, soiling, or shading issues; hourly discrepancies can point to azimuth, shading, equipment control, or demand relationships.

Consider actual weather conditions during comparisons. If the simulation assumes long-term average conditions, actual years that are sunnier or rainier will change generation. Thus, do not hastily conclude the simulation was wrong just because prediction and result differ. In performance evaluation, interpret differences while checking that year’s irradiance, temperature, snow, typhoons, prolonged rain, and other specific conditions.

Using measured data helps reveal problems that were hard to see at design time. For example, if generation drops at a particular time, on-site checks may reveal unexpected shading. A sudden monthly drop may be caused by soiling, equipment failure, communication issues, or circuit outages. Differences between simulation and measurement are not just forecast errors but diagnostic material to understand system condition.

Comparing with measured values also helps improve internal design standards and input rules. Accumulating per-project differences between predicted and actual values, their causes, and corrective measures clarifies reasonable loss rates and on-site check items for your company. Storing experience such as region-specific irradiance trends, roof-type temperature losses, shading evaluation methods, and soiling expectations improves simulation accuracy for future projects.

When using measured values, pay attention to data quality. Missing generation measurements, interrupted communications, inconsistent measurement units, or mismatched aggregation periods prevent correct comparisons. Decide during planning which data to collect after operation, which units to manage them in, and how to compare so that simulation verification is easier.

Solar power generation simulation is not a one-time task. Updating it through planning, design, construction, operation, and verification produces more reliable generation forecasts. Continuously comparing with measured values is the most reliable long-term method to improve accuracy.

Common mistakes to avoid to improve accuracy

Improving simulation accuracy requires not only knowing the correct methods but also avoiding common mistakes. In practice, users sometimes become complacent simply because they use calculation tools and fail to sufficiently verify input conditions. However, simulation results are heavily influenced by inputs; if assumptions are inaccurate, the generated output is difficult to trust.

First, avoid relying on default settings and looking only at results. Default settings are useful for preliminary estimates but do not necessarily reflect actual site conditions. Adopting annual generation without confirming installation site, azimuth, tilt, shading, equipment configuration, and loss conditions increases the likelihood of deviation. Especially when using results for business plans or proposals, document which standard settings were changed and the rationale.

Second, avoid calculating with overly optimistic assumptions. To make generation look large, underestimating shading, soiling, degradation, and downtime produces attractive figures in planning but undermines reliability if actual performance falls short. Simulations should be used for realistic decision-making, not to create favorable numbers.

Conversely, being overly conservative without basis is also problematic. Setting all losses high without justification can understate generation and misjudge system viability. The key is to set realistic, evidence-based conditions. This requires on-site surveys, drawing checks, equipment specs, maintenance plans, and historical performance to make inputs defensible.

Forgetting to reflect design changes in simulations is another common issue. After initial studies, panel count, layout, equipment, wiring, azimuth, tilt, maintenance access, and shading conditions may change, yet outdated simulation results continue to be used. As design progresses and conditions become specific, simulations must be updated to remain consistent with the final design.

Also avoid judging by total generation alone. Even if annual totals are the same, monthly and hourly generation profiles can differ. For sell-to-grid systems, self-consumption systems, or systems combined with storage, the relevant generation periods differ. Checking alignment with demand, seasonal variation, peak times, and possible output curtailment makes simulations useful in practice.

Finally, underestimating on-site information is a major risk. Maps, aerial photos, and drawings may not reveal actual shading, obstacles, rooftop equipment, topography, trees, soiling propensity, or construction constraints. Verify the site whenever possible and keep photos and measurement records to clarify input rationale. Higher-quality on-site information increases simulation accuracy.

Conclusion

Improving the accuracy of solar power generation simulations requires checking the rationale for input conditions one by one rather than focusing only on calculation outputs. Carefully addressing the seven perspectives—irradiance data selection, site conditions, azimuth and tilt, shading impact, equipment specifications, loss conditions, and comparison with measured values—brings generation forecasts closer to reality.

In practice, there are cases when quick rough estimates are needed. However, for simulations used in business plans, system design, internal approvals, customer proposals, and operational improvement, rough estimates are insufficient. Organizing which conditions were assumed, which were confirmed on-site, and how much loss was expected improves the explanatory power of simulation results.

Particularly important is accurately understanding site conditions. Solar generation can vary with subtle differences in installation location even within the same region and system capacity. Information that cannot be determined without visiting the site—roof orientation and pitch, surrounding buildings, trees, topography, panel-row spacing, maintenance access, and wiring length—must be accurately recorded and reflected in design conditions as the starting point for improving simulation accuracy.

Also indispensable is the practice of comparing with post-operation measured values. Verifying differences between prediction and reality and analyzing their causes informs future design and maintenance. Simulation is not a one-off document but a continuous decision-making tool that supports the system from planning through operation.

To raise simulation accuracy, connect desk-based input procedures with field measurement work. Accurately recording installation location, azimuth, elevation, and surrounding environment and reflecting them in drawings and simulation conditions significantly increases the reliability of generation forecasts. If you want to obtain high-precision location information on site and use survey records in design and simulation, using LRTK (iPhone-mounted GNSS high-precision positioning device) can help more smoothly link planning, on-site verification, construction management, and post-operation record-keeping for photovoltaic systems.