Explanation of Differences Between Solar Power Generation Simulations and Actual Generation

Table of Contents

• Why do solar power generation simulations and actual generation not match?

• Simulation results are forecasts, not guarantees

• The main causes of differences are irradiance and weather conditions

• Accuracy of equipment input conditions affects the difference in generation

• Don’t overlook site-specific factors such as shading, soiling, and snow

• Effects of degradation, outages, and output curtailment on actual generation

• The correct way to compare simulations and actual performance

• Practical checks to reduce the difference

• How to use generation differences for design, maintenance, and revenue decisions

• Summary

Why do solar power generation simulations and actual generation not match?

Solar power generation simulations are an important task when planning PV systems to estimate annual and monthly generation, feed-in volumes, self-consumption, and utilization rates in advance. They are used in business planning, design review, investment decisions, explanations to financial institutions, and post-construction performance verification. However, the actual generation obtained after starting operation never perfectly matches the values produced by the simulation.

This difference does not necessarily mean the simulation is wrong. PV systems are heavily influenced by natural conditions, and it is impossible to perfectly predict future irradiance, temperature, cloud cover, rainfall, snowfall, wind, equipment outages, and so on. Therefore, simulation values are estimates based on specific assumptions, while actual generation reflects the conditions that actually occurred in that year, month, or day.

What practitioners should beware of is not simply judging “higher than the simulation” or “lower than the simulation.” It is important to separate which assumptions were off, which differences can be tolerated as natural variability, and which differences warrant investigation as design, construction, or maintenance issues. Many people searching for “solar power generation simulation” aim to improve prediction accuracy, explain differences from actuals, find causes of generation decline, or verify the validity of revenue plans.

To correctly understand generation differences, you need to distinguish how simulations work from the nature of actual generation. Simulations estimate generation by inputting irradiance data, tilt, azimuth, module capacity, conversion efficiency, various losses, weather conditions, shading effects, and running calculations based on a certain model. Actual generation, on the other hand, is the result that includes actual weather, equipment condition, operational conditions, maintenance status, grid constraints, and changes in the surrounding environment.

In other words, the difference arises from a combination of “differences in input conditions,” “limitations of the calculation model,” “changes in actual operational conditions,” and “natural variability.” Understanding this structure prevents overreliance on simulation results and overreaction to differences with actuals. The important thing is not only to eliminate differences but to be able to explain why they occurred.

Simulation results are forecasts, not guarantees

The first point to grasp when looking at solar generation simulations is that simulation values are not guaranteed future outputs. Generation simulations are calculated based on historical weather data, average irradiance, equipment specifications, and design conditions. Therefore, the computed annual generation should be understood as “the expected generation under these conditions.”

In practice, this expected value is often used as a basis for revenue planning and equipment design. However, actual generation varies year to year. In years with higher irradiance than average, actual generation may exceed the simulation, while prolonged rainy or cloudy years may result in lower generation. Especially during the rainy season, typhoons, winter snowfall, or prolonged off-season rains, monthly generation can be significantly affected. Even if annual differences are small, large month-to-month deviations are not uncommon.

Simulations often use average weather conditions and thus cannot reflect anomalous weather in a particular year. Comparing a simulation based on long-term historical averages with the actual performance in a specific operating year naturally yields differences. Therefore, when comparing simulations and actuals, it is important not to draw conclusions based on a single year but to check multi-year trends and differences in weather conditions.

Another important point is that simulations include many assumptions made at the design stage. For example, module capacity, tilt, azimuth, wiring losses, conversion losses, shading, soiling losses, degradation, and downtime are affected by input and setting values. If these assumptions do not match site realities, differences in generation will naturally be larger.

To use simulation values effectively, you must retain not only the numbers themselves but also the assumptions that produced them. Recording input conditions, the irradiance data used, equipment capacity, loss rates, shading evaluation methods, and assumed availability makes it easier to trace causes of differences when comparing with actual generation after commissioning. Conversely, if only simulation results with unknown assumptions remain, analyzing differences with actuals becomes difficult.

Simulations are very useful decision-making tools when used correctly. However, they should be treated as conditional forecasts, not guaranteed values. With this mindset, you can calmly separate causes when generation deviates from expectations.

The main causes of differences are irradiance and weather conditions

Irradiance is often the largest factor causing differences between simulated and actual PV generation. Because PV converts sunlight into electricity, generation changes with irradiance even with the same equipment. More sunny days increase generation; more cloudy or rainy days decrease it. This is not an equipment fault but a basic characteristic of solar power.

Irradiance data used in simulations can come from long-term averages, nearby observation stations, satellite-derived estimates, or on-site measurements. While all are useful, they do not exactly match the actual irradiance at a plant in a given year. In mountainous areas, coastal zones, basins, urban areas, and snowy regions, cloud behavior, fog occurrence, snowfall, temperature, and wind conditions can differ even over short distances.

Temperature also affects generation. PV modules generally produce less output as temperature increases. Although irradiance is high in summer, module temperature rises and you cannot assess generation by irradiance alone. In installations with poor wind and trapped heat, temperature losses may be larger than expected. Conversely, in colder seasons with sufficient irradiance, modules may generate efficiently.

Rainfall and wind have indirect effects. Continuous rain reduces irradiance but can also wash dirt from module surfaces. Wind can help cool modules, but strong winds or typhoons can cause temporary shutdowns or require inspections. In snowy regions, snow covering modules causes periods of no generation. If simulations do not sufficiently account for snow losses, winter actual generation can be significantly lower.

Also important is the comparison period. Weather effects are more pronounced on daily or monthly timescales. For example, if a month has a lot of rain, actual generation that month may fall far below the simulation, but annual totals can be compensated by better weather in other months. Conversely, even if annual differences are small, a persistent shortfall in a specific season could indicate issues like shading, soiling, snow, or equipment faults.

Therefore, when analyzing generation differences, start by checking irradiance and weather conditions. Looking at actual generation alone does not allow you to determine whether equipment or weather is the cause. If possible, also check irradiance, temperature, rainfall, snowfall, and generation trends of nearby plants for the same period to distinguish natural causes from equipment issues.

Accuracy of equipment input conditions affects the difference in generation

Simulation accuracy is greatly affected by how accurately equipment conditions are input. No matter how advanced the calculation method, if input conditions differ from site reality, the simulated generation will diverge from actual performance. A common practical issue is that design-drawing conditions and as-built conditions differ slightly, and those differences affect generation.

Typical input items include module capacity, number of modules, tilt, azimuth, racking height, array spacing, series/parallel configuration, inverter capacity, wiring distance, wiring losses, conversion efficiency, temperature coefficients, and shading conditions. Each individual difference may seem small, but accumulated differences appear in annual generation.

Tilt and azimuth are particularly important because they determine how much irradiance is received. If a simulation assumed a near-south orientation but the actual orientation was shifted due to terrain or roof shape, generation will change. For ground-mounted systems, post-site-works slopes and racking installation precision can differ from assumptions. For roof-mounted systems, multiple roof surfaces with different tilts and azimuths make oversimplification likely to cause discrepancies.

Wiring and conversion losses must not be overlooked. DC power produced is sent through wiring to the power conversion device and converted to AC, with losses in the process. If simulation assumed lower loss rates than actual, actual generation will appear lower. Also, inverter sizing may limit output during high irradiance periods. While this can be an intentional design decision, if not properly reflected in simulations it becomes a source of difference.

Pay attention to module specification values. Catalog outputs are measured under standard test conditions and are affected outdoors by temperature, incident angle, soiling, and aging. Even the same module model shows unit-to-unit variation. Simulations commonly use nominal values, so actual site output may not perfectly match.

Post-construction inspection and as-built documentation reconciliation are also important. If you use design-stage simulation values as post-operation evaluation criteria, you must confirm the installed system matches the design. If module count, grid configuration, tilt, azimuth, inverter capacity, wiring route, or shading differ from design, it is desirable to update simulations to reflect as-built conditions.

In short, reducing the difference between simulation and actual generation depends more on how closely input conditions reflect site reality than on the calculation itself. Generation simulations become practical decision-making tools only when they carefully incorporate real site conditions.

Don’t overlook site-specific factors such as shading, soiling, and snow

Site-specific loss factors are critically important for generation differences. Even with the same irradiance and equipment specifications, surrounding conditions can cause large changes in actual generation. Shading, soiling, snow, vegetation growth, changes in nearby buildings, and terrain conditions are particularly likely to cause differences between simulations and actual generation.

Shading is a major loss factor for PV systems. Buildings, utility poles, trees, mountains, adjacent equipment, fences, signs, and inter-row shading of racking can reduce module output. Shading changes by time of day and season. In winter, when solar altitude is low, shading that was insignificant in summer can have large effects. Even shading that occurs only in the morning or evening can create non-negligible annual losses.

If a simulation treats shading simply, differences with actual generation are likely. Examples include insufficient reflection of site obstacles, ignoring terrain-induced shading limits, not accounting for tree growth, or crude inter-row shading evaluation. In roof installations and narrow sites, carefully confirming surrounding structure shading is essential.

Soiling also affects generation. Dust, yellow sand, pollen, bird droppings, fallen leaves, exhaust-related deposits, and dust from nearby farms or factories can adhere to module surfaces and reduce incident light. Rain may naturally wash some dirt away, but shallow tilt installations retain dirt more easily. In regions with little rain or much airborne dust, soiling losses may exceed simulation assumptions.

In snowy regions, snow-induced shutdowns or reduced output are major factors. When modules are covered by snow, generation drops sharply. Whether snow naturally slides off depends on module angle, temperature, irradiance, wind, snow type, and local snow accumulation conditions. Snow can accumulate at lower edges and partially cover modules for extended periods. If winter simulation values differ from actuals, check whether snow loss assumptions were appropriate.

Vegetation growth is often overlooked. Trees or weeds that did not cast shade at construction can grow after several years and create shading. For ground-mounted systems, even weeds reaching the lower edge of modules can affect generation. New buildings nearby, changes in adjacent land use, and accumulation of sediment or fallen leaves also matter in long-term operation.

Because these site-specific factors are not always captured by standard simulation conditions, it is important to conduct site surveys, surveying, photo records, shading checks, post-construction inspections, and regular patrols to bring simulation assumptions closer to site reality. Treat generation differences not merely as numerical issues but with a mindset to verify what is happening on site.

Effects of degradation, outages, and output curtailment on actual generation

PV systems do not maintain the same performance after commissioning. Modules gradually degrade over time, and equipment failures, maintenance shutdowns, communication failures, and grid constraints occur. These operational factors also contribute to differences between simulations and actual generation.

First, module aging and degradation. Modules are exposed outdoors for long periods and are affected by UV, heat, humidity, temperature cycles, wind, and rain. If a simulation provides first-year generation and you compare it with actual generation several years later, the decline due to degradation will appear. Long-term revenue plans typically assume an annual degradation rate, but you should verify whether that assumption matches reality.

Next, the impact of equipment outages. Inverter trips, protective device operation, grid abnormalities, inspection work, communication device troubles, part replacements, and safety checks after disasters can all create periods without generation. Short outages have limited annual impact, but if they occur during high-irradiance periods or recovery is delayed, losses can be large. Simulations often assume a certain availability; without checking the actual outage history you may misidentify causes of differences.

Output curtailment is also important. Depending on grid supply-demand conditions and connection terms, a plant may be required to reduce output even when it could generate. If simulations do not account for curtailment, actual generation will be lower than expected. In regions with high renewable penetration, even if plant performance is fine, operational constraints may reduce generation.

There is also peak clipping due to inverter sizing. If inverter capacity is deliberately set smaller relative to total module capacity, generation exceeding the inverter limit during strong irradiance hours is not delivered. This is sometimes an intentional design trade-off between equipment cost and annual generation. However, if simulations do not correctly reflect this limit, sunny-day actuals may appear lower than expected.

Additionally, watch for missing monitoring data. What you view as actual generation may be the subset for which communication was successfully obtained. Communication outages, meter faults, data loss, or misaligned aggregation periods can make generation appear lower than reality. Before analyzing differences, confirm that measurement data were correctly acquired.

Thus, actual generation is influenced not only by equipment performance but also by operational state, maintenance, grid conditions, and measurement environment. When comparing with simulations, confirm whether equipment operated as planned, whether there were outages or controls, and whether data were recorded correctly.

The correct way to compare simulations and actual performance

When comparing simulations and actual generation, do not simply look at the annual total — align comparison conditions. Without aligned conditions you may treat innocuous differences as problems or miss important anomalies.

First, confirm the comparison period. Check whether the simulation uses a calendar year, fiscal year, or one year from the start of operation. Actual data must be aggregated for the same period for a valid comparison. Differences may arise from month-end closing, meter read dates, monitoring system aggregation periods, or feed-in reporting periods. If a large difference appears, first suspect a misaligned period.

Next, align the definition of generation being compared. Whether the simulation value is DC generation at the modules, AC generation after conversion, feed-in volume, or surplus after self-consumption will change the comparison target. Measured actuals may be monitoring system generation, meter readings, feed-in amounts, or total generated energy including self-consumption. Comparing numbers with different definitions will create apparent differences.

Monthly comparisons are also useful. Annual totals alone do not show when differences occur. Comparing each month’s actual to simulation makes it easier to find seasonal causes. If only winter is underperforming, check for snow or low-solar-altitude shading; if only summer underperforms, consider temperature losses or output limits; if a single month underperforms, consider abnormal weather or outages. If generation is consistently lower across the year, review input conditions, system capacity, wiring losses, and measurement conditions.

Correcting for irradiance is important. If actual generation is below simulation but that year’s irradiance was lower than average, equipment may not be at fault. Conversely, if irradiance is average or above and generation is low, site or equipment causes are likely. If possible, check generation per unit irradiance to separate weather effects from equipment issues.

Comparing with nearby similar plants is also helpful. If multiple plants in the same area show lower generation at the same time, widespread weather or regional irradiance conditions are likely. If only your plant shows low performance, individual equipment faults, shading, soiling, or outages are more probable. However, even with nearby comparisons, differences in tilt, azimuth, capacity, shading, and operation prevent simplistic judgments based only on absolute generation.

When reviewing differences, apply the concept of acceptable tolerance. Given interannual variability in irradiance and weather, exact matches are unrealistic. The issue is whether the difference is persistent or isolated, explainable by weather, or indicative of equipment issues. Clarify the comparison purpose and align period, definitions, weather, and system state to treat generation differences as useful practical information.

Practical checks to reduce the difference

You cannot entirely eliminate differences between simulations and actual generation, but you can reduce them and make them easier to explain. This requires checks at the design, construction, and operation stages.

At the design stage, capture site conditions as accurately as possible. Confirm site location, elevation, terrain, surrounding obstacles, azimuth, tilt, ground conditions, presence of snow, surrounding land use, and potential future shading changes. For roof installations, record each roof surface’s tilt, azimuth, obstacles, load capacity, and wiring routes. Running simulations under standard assumptions with inadequate site surveys often leads to larger post-commissioning differences.

Managing input conditions is also important. Keep records of the irradiance data used in the simulation, equipment capacity, tilt, azimuth, loss rates, shading conditions, degradation rate, and expected downtime. If you cannot later identify the assumptions used, you cannot analyze differences. If design changes occur, update the simulation with post-change conditions.

During construction, verify the system was installed per design. Check module tilt and azimuth, racking height, array spacing, wiring, connection configuration, inverter specifications, and meter installation, and record as-built status. If the site differs from design, evaluate whether those differences affect generation. Post-construction photos and survey records help later analysis.

In operation, regularly review monitoring data. Track daily, monthly, and yearly generation and detect sudden drops or differences among strings. When generation drops, sequentially check weather, irradiance, outage history, output curtailment, equipment failures, communication faults, soiling, shading, vegetation, and snow. Early detection of anomalies reduces losses.

Maintenance planning is crucial for minimizing differences. Appropriate periodic inspections, cleaning, mowing, tree management, fastener checks, electrical inspections, and monitoring system checks help prevent unexpected declines. Shading and soiling often worsen progressively after commissioning. Even if there was no issue at construction, a few years later generation drops may appear, so continual checks are necessary.

Also, do not treat simulation as a one-time deliverable. Accumulate operational data after commissioning and revise assumptions based on actual performance to improve the accuracy of future projects and long-term plans. Analyzing differences is not just a critique; it is essential feedback for improving future design quality.

How to use generation differences for design, maintenance, and revenue decisions

Differences between simulation and actual generation are not only for finding problems. If analyzed appropriately, they can inform design improvements, maintenance planning, revenue decisions, customer explanations, and equipment upgrade choices. In practice, treat those differences not as mere “errors” but as “actionable information.”

In design, reflect past project performance differences in future simulations. If a region consistently underperforms in winter, pay more attention to snow and low-solar-altitude shading. If soiling is significant near coasts or farms, revise soiling loss assumptions and cleaning schedules. If hot installations underperform in summer, consider temperature losses and ventilation conditions. In this way, actual differences help make design assumptions more realistic.

For maintenance, use trend differences to prioritize inspections. Patterns such as low generation in a particular string, missing sunny-day peaks, extreme morning/evening losses, winter-specific shortfalls, or recovery after rain provide clues about causes. Viewing generation by time-of-day, by string, and by season helps narrow down where to inspect. This reduces unnecessary checks and accelerates necessary interventions.

For revenue assessment, the stability of actuals relative to simulations is important. Do not judge long-term revenue based on first-year differences alone; consider irradiance, outages, curtailment, degradation, and maintenance in assessing long-term generation trends. If lower generation is due to temporary weather, long-term impact is limited. But if constant shading, design discrepancies, frequent curtailment, or equipment faults are the cause, reassess revenue plans and implement countermeasures.

For stakeholder communication, difference analysis is useful. If generation is below expectations, simply saying “it’s due to weather” may not satisfy stakeholders. Organizing comparison periods, irradiance, monthly trends, outage history, and inspection results for shading and soiling enhances credibility. Conversely, vague simulation assumptions make it difficult to explain differences convincingly.

Differences can also motivate equipment upgrades or corrective actions. Cleaning, mowing, tree removal, equipment replacement, rewiring, monitoring improvements, or adding meters can be considered based on quantified generation gaps. However, interventions incur costs and operational burdens, so carefully evaluate expected generation recovery and long-term effectiveness. Combining simulation and actual analysis enables evidence-based improvements rather than guesswork.

In PV operations, planning, design, construction, operation, and maintenance are interconnected. Continuously analyzing simulation vs. actual differences enhances the accuracy of the entire workflow. The objective is not merely to find differences but to interpret them, identify causes, and apply findings to future decisions.

Summary

Differences between solar power generation simulations and actual generation are an unavoidable topic for practitioners handling PV systems. Simulations predict generation based on assumptions such as irradiance, weather conditions, equipment specifications, loss rates, shading, degradation, and availability. Actual generation reflects real weather, site environment, equipment condition, outages, output curtailment, maintenance status, and measurement conditions. Therefore, complete agreement between the two is naturally unlikely.

What matters is how you interpret any differences. You must sequentially check whether differences arise from interannual irradiance variability, misaligned input conditions, site-specific factors like shading and soiling, equipment outages or curtailment, or measurement data issues. Simply comparing simulation and actual numbers does not lead to correct conclusions.

To use simulations effectively in practice, clarify input assumptions, reflect design changes and as-built conditions, and keep records in a form that allows comparison with actuals. After commissioning, monitor monthly, seasonal, and string-level generation and generation per unit irradiance to detect anomalies early. Generation differences can be used to improve design, revise maintenance plans, refine revenue assessments, and explain results to stakeholders.

Accurately capturing site conditions is particularly important for simulation accuracy. If you cannot accurately grasp installation location, azimuth, tilt, surrounding obstacles, shading, terrain, snowfall, and vegetation growth, no matter how detailed the calculations, differences from reality will remain. In planning and maintaining PV systems, it is essential not only to perform desk calculations but to measure the site accurately, record it, and enable comparisons.

If you want higher precision for site location and survey accuracy, leveraging LRTK — a GNSS high-precision positioning device that can be attached to an iPhone — is effective. Accurately recording position information for layout verification, site surveys, inspection records, locating objects that cause shading, and post-construction condition management makes it easier to reconcile simulation assumptions with actual site conditions. To analyze differences between simulations and actual generation with stronger evidence, it is important to establish a system that records not only generation data but also accurate site condition information.