What is PVSyst? A 3-minute explanation of why it's used for power generation forecasting

Table of Contents

• What PVSyst Is

• Reason 1: Can Be Based on Meteorological Conditions

• Reason 2: Can Perform Detailed Hourly Simulations

• Reason 3: Can Reflect Installation Conditions and Equipment Configuration

• Reason 4: Can Explain the Breakdown of Losses

• Reason 5: Can Analyze Shading Effects in Depth

• Reason 6: Facilitates Comparison and Performance Verification

• Limitations to Know Before Implementation

• Summary

What is PVSyst?

PVSyst is dedicated software for examining photovoltaic power generation systems, sizing capacity, evaluating performance, and analyzing data. The official documentation describes it as an environment that handles not only grid-connected systems but also stand-alone systems, pumping applications, and DC-system applications, and it provides a weather database, an equipment database, various solar energy–related tools, and functions for comparing measured data. In other words, PVSyst is not simply a tool to calculate annual energy production once; it is more accurate to view it as practical software for building the evidential basis for design decisions while organizing a project’s conditions.

When practitioners search for "What is PVSyst", what they want to know is less the meaning of the name than why it is widely used for energy yield forecasting. The short answer is that PVSyst can handle meteorological conditions, the orientation of the installation surface, system configuration, losses, shading, and post-operation performance comparisons all within a single workflow. In the official documentation, a simplified design for preliminary studies is separated from a full design that performs detailed time-step simulations, so you can change how you use it according to the progress of the project. This flexibility makes it easy to link a single approach from early-stage estimates through detailed design and even post-commissioning verification.

The important point here is that PVSyst is not a "black box" that returns a single answer. In formal project design, each project contains geographic conditions and time-series meteorological data, and you can save multiple simulation conditions as alternative scenarios and compare them. In other words, the value of this software lies not in the final numbers themselves but in visualizing how results change depending on the assumptions. It is used for power generation forecasting not because it is famous, but because it makes it easy to explain the link between design conditions and outcomes.

Reason 1: Can be based on meteorological conditions

The starting point for power generation forecasting is how much solar energy a location can receive. In PVSyst's official documentation, a project is essentially described as a framework that retains geographic conditions and hourly meteorological data; regarding meteorological data, it organizes everything from creating the geographical site and generating hourly datasets to visualizing and comparing meteorological data and importing external files. This is an approach that builds the foundation of power generation forecasting from the site's meteorological conditions rather than from the equipment-side conditions. Power generation forecasts are more likely to be trusted because they do not leave the initial inputs ambiguous and treat meteorological assumptions as elements under management.

In practice, even with the same equipment capacity, results can vary greatly depending on the site. Furthermore, even when the total annual solar irradiance is similar, differences in seasonal distribution and time-of-day characteristics change how generation behaves. Because PVSyst relies on hourly meteorological data in detailed design, it becomes easier to reveal differences that aren’t visible from annual totals alone. It’s straightforward to see how much output drops during high summer temperatures, how much can be captured in the morning and evening, and how much generation can be secured in winter, since meteorological conditions can be analyzed by month and by hour. In generation forecasting, being able to view conditions over the course of time is far more meaningful than using simple averages.

Furthermore, PVSyst does not use horizontal-plane irradiance data directly as the input for the generating surface. The official irradiance conversion model explicitly defines the method for calculating the irradiance incident on a tilted installation surface from horizontal-plane irradiance. This conversion treats the direct component, the diffuse component, and the ground-reflected component separately, reflecting how differences in tilt and azimuth affect them. In other words, PVSyst does not simplify things to “more irradiance means more generation,” but instead evaluates “how that irradiance actually reaches a generating surface of a given orientation.” This is why it is straightforward to explain during the calculation why south-facing and east–west-facing orientations yield different results, and why differences in angle cause variations even within the same region.

Being able to base analyses on meteorological conditions also means, conversely, that the choice of meteorological assumptions has a strong impact on the results. For this reason, PVSyst provides mechanisms for comparing and importing meteorological data, making it easy to use initial assumption data for the first simulation and to replace it with more appropriate data as needed. What matters for practitioners is not just looking at the power generation forecast numbers, but being able to explain "under which meteorological conditions those numbers were produced." One reason PVSyst is used is that it makes securing this explainability easy from the very first stage.

Reason 2: Enables detailed hour-by-hour simulations

One of the main reasons PVSyst is widely used for energy yield prediction is its detailed hourly simulations. In the official project design, this feature is intended to perform the design and performance analysis of photovoltaic (PV) systems through detailed hourly simulations. Moreover, cases are organized within a "project" framework, where multiple alternative proposals can be created and compared. This implies that the intended use is not merely to obtain a single annual energy figure, but to examine which option is more appropriate by varying design conditions.

In practical work, power generation fluctuates greatly both within a day and over the course of a year. The solar altitude and incidence angle differ between morning, midday, and evening, and between summer and winter not only the irradiance but also the temperature conditions change. PVSyst computes these changes over time and provides a mechanism to review the results by month, by day, and by hour. The official overview also explains that the results include numerous simulation variables and can be displayed or exported monthly, daily, and hourly. In other words, what makes PVSyst excellent is that it lets you track the intermediate steps leading to the final annual value, not just the final annual figure. This enhances the explanatory power of the power generation forecast.

Also, PVSyst is not merely an assemblage of empirical rules but is founded on physical models. In the official list of physical models, solar position, incident irradiance models, photovoltaic modules, conversion equipment, batteries, pumps, and so on are organized as the main models used in the calculations. In other words, PVSyst links, via each of these models, how radiation is received, how the received light becomes electricity, and how that electricity is converted on the AC side. This software is chosen for generation forecasting not because the results look flashy, but because it makes it easier to create coherent explanations for changes in conditions.

For photovoltaic modules, PVSyst implements a single-diode model and handles not only the basic values listed in the datasheet but also additional parameters. The official documentation explains that this model involves elements that are not fully reflected in standard datasheets, such as series and shunt (parallel) resistances and temperature-related corrections. In particular, behavior under low irradiance and changes due to temperature significantly affect actual power generation, so it is important that such internal models exist. What matters in energy-yield prediction is not being able to calculate with high accuracy only at sunny noon, but how the model can respond to real operating conditions such as low irradiance and high temperatures. PVSyst is structured with that in mind.

PVSyst also treats the conversion-equipment side quite concretely. In the official converter model, the concepts of maximum power point tracking, the input-side voltage window, power and current limits, and the fact that efficiency mainly varies with power and can also depend on the input voltage are explained. In other words, it is not assumed that the power extracted on the PV side is delivered to the AC side unchanged. Losses when operating outside the tracking range, limitations when exceeding ratings, and efficiency reductions at low load are also included in the predictions. PVSyst is used for generation forecasting because it can collectively handle these elements that are close to actual equipment behavior within its hourly calculations.

Reason 3: Can reflect installation conditions and equipment configuration

The quality of power generation forecasts is not determined solely by weather conditions. The same location can yield very different results depending on which direction it faces, at what angle it is installed, whether tracking is used, whether it is arranged in rows, and what equipment configuration is chosen. In PVSyst’s project design you can define the orientation of the installation surface and handle tracking and row arrangements. Furthermore, after selecting a specific equipment configuration, you can also design the array layout, such as the number in series and the number in parallel. This means it is not software that simply inputs capacity to produce an energy estimate, but that the forecast can reflect “how the system is to be realized.”

In practice, even with the same nominal output, predictions can vary considerably depending on the configuration. If the orientation of the installation surface differs, the daytime output curve changes; if the arrangement of series and parallel connections or the way inputs are split changes, the constraints on the conversion side and how losses occur also change. PVSyst allows these configuration conditions to be entered as part of the "design", so you can produce forecasts that include not only "how many kilowatts to install" but also "how the system will actually be realized." This leads to a generation forecast deeper than rough per-unit calculations. PVSyst is commonly used in design practice because it lets you carry the design conditions behind the numbers.

On the other hand, in the early stages of a project it is not uncommon for the detailed configuration to be undecided. What makes PVSyst convenient is that it can easily handle that stage. In the official preliminary design, you can quickly make monthly estimates using only a few general conditions. Moreover, the documentation explicitly states that this is a rough early-stage estimate not to be used for detailed design, that accuracy should be expected to have a spread of roughly 10% or more, and that more precise results should be obtained through time-step simulations. In other words, PVSyst can be used in a two-tiered way: to grasp the broad picture in the initial stage and to refine the configuration in the detailed stage.

This two-stage approach has considerable practical significance. In the initial planning phase, you often just want a quick look at candidate sites and a sense of capacity, while in the subsequent basic design phase there are more opportunities where you want to work out layout, shading, equipment configuration, and losses in detail. Using separate tools makes it easy for assumptions to be lost during handover, but PVSyst allows you to move from rough assessments to detailed studies while keeping the same way of thinking. The reason this software is favored for energy yield forecasting is not simply that it is highly functional, but that it is easy to switch how you use it in line with a project’s development.

Reason 4: Able to explain the breakdown of losses

What is truly useful for generation forecasting is not software that merely returns a single annual energy output number. If it is to be used for design decisions or in explanatory materials, it must be able to explain why that number was produced. PVSyst’s official overview explains that the results include numerous simulation variables, can be displayed by month, by day, and by hour, and that the "Loss Diagram" is particularly useful for identifying weaknesses in the design. In other words, PVSyst adopts an approach that visualizes at which stages and by how much losses accumulated prior to the final energy output.

The reason visualizing these losses is important is that the power output of a solar installation is not determined by a single factor. There is the irradiance reaching the installation surface, then optical losses due to the angle of incidence, and on top of that temperature rise, variations between modules, mismatch, wiring losses, reductions in conversion efficiency, and impacts from shutdowns or downtime accumulate. In PVSyst project design, you can set detailed losses such as soiling, angle of incidence loss, module temperature, wiring resistance, module quality differences, mismatch, and downtime. The reason it is used in energy yield forecasting is that these real-world losses can be incorporated into the calculation flow from the start, rather than being "bundled together later with a safety factor."

By carefully examining the loss diagram, it becomes easier to see where there is room for improvement. If you can tell whether the solar irradiance conditions themselves are harsh, optical losses are large, temperature losses exceed expectations, or losses are being incurred in wiring and power conversion equipment, you can determine which parts of the design should be reviewed. A simple comparison of annual energy production tends to end with just “high” or “low,” but PVSyst makes it easier to move to the next actions of “why is it low?” and “what should be fixed?” For practitioners, this difference is quite significant. It is therefore natural that PVSyst is valued when you want to demonstrate the validity of a design not only with numbers but also with the underlying reasons.

Furthermore, PVSyst provides a metric called the performance ratio. The official documentation explains that the performance ratio includes, in addition to optical losses such as shading, incidence-angle losses, and soiling, array losses such as solar-cell conversion, aging, module quality differences, mismatch, and wiring, and further system losses such as power converters and storage elements. Because this indicator, unlike energy generation per unit capacity, is less directly dependent on weather and installation orientation, it is considered easy to use for quality comparisons between systems with different locations or orientations. The fact that it makes the completeness of the installation as an asset easier to see, not just the amount of generation, is also a reason PVSyst is valued as a prediction tool.

The background for why the performance ratio is emphasized on site should not be overlooked. Officially, the performance ratio is calculated from predictive simulations and is described as being widely used in the context of performance guarantees by comparing it with the performance ratio measured on site. In other words, PVSyst is not software confined to desk-based energy yield predictions; it makes it easy to link evaluation metrics from the prediction stage with those from the measurement stage. The reason it is used in energy yield prediction is not just that "a predicted value is produced" but also that "the prediction results are easy to link later with on-site operation."

Reason 5 You can see the effects of shadows more deeply

PVSyst's particular strength in energy production forecasting is its handling of shading. In official project design it is set up to handle, in stages, distant horizon shading, partial shading from nearby obstacles, and even module layouts used for detailed calculations of electrical shading losses. For near shading, the environment around the power-generating surface can be constructed in 3D, and, when necessary, the position and electrical interconnection of each module can be defined. In other words, PVSyst does not treat shading as a simple "percentage reduction" applied across the board, but has mechanisms to evaluate it in depth according to site conditions.

Why this point matters is that shading losses in a photovoltaic system are not determined simply by the proportion of shaded area. The official terminology explains that electrical shading loss is the additional loss obtained by subtracting the linear irradiance shortfall from the actual shading loss: when part of a cell or module is shaded, the I-V characteristics are disturbed and the string current is limited by the weakest cell, causing losses beyond a simple reduction in irradiance. In other words, the intuition that "a little shading will only cause a small loss" can be wrong. PVSyst is trusted for energy-yield prediction because it can address this practically troublesome phenomenon head-on.

Furthermore, the official module layout feature can calculate electrical shading mismatch losses in detail by defining the exact position of each module and which string and which input it connects to. It considers not only the direct component but also the scattered component, treating cases where even if the direct light is completely blocked, scattered light remains and a certain current still flows. In other words, PVSyst does not reduce shading to a binary present/absent condition; it separates the optical components from their effects on the circuit. The more a project wants to examine the impact of row spacing, surrounding structures, and how wiring is grouped on the results, the more valuable this feature becomes.

Of course, achieving this level of detail requires precise input data. Even the official documentation explains that the module layout should be used only after both the 3D shading definition and the electrical sub-array definition are in place—essentially the final stage of system design. Conversely, in the early stages you don’t need to be that detailed; you can increase the depth of shading evaluation as the project becomes more defined. The ability to progressively deepen the analysis is another reason PVSyst is easy to use for energy yield prediction. Because it’s not just a simple look-up table but allows deeper shading evaluation when needed, it remains convenient to use consistently from preliminary studies through detailed design.

Reason 6: Makes it easier to compare and verify results

In the field of power output forecasting, what is truly useful is software that does not stop at calculation but connects through to comparison and validation. In PVSyst’s official project design, you can create multiple variants within the same project and compare them while changing the orientation of the installation surface, system conditions, loss conditions, and so on. Furthermore, there is a mechanism to run a series of simulations with varied parameters together and list the key results. This implies that the intended use is not so much to evaluate a single proposal as to search for the optimal one while observing differences in conditions.

In practice, it is uncommon for a project to be decided on a single plan from the outset. It is normal to refine design decisions by repeatedly comparing questions such as what happens if you change the orientation slightly, how much shading losses are reduced if you widen the row spacing, or how much the figures change if you assume more severe soiling or downtime. PVSyst is used because it makes it easy to organize these comparison exercises within the scope of a project. With a simple spreadsheet, differences in assumptions become scattered and hard to manage, but with PVSyst it is easier to track the differences between the changed conditions and the results. When preparing presentation materials, it is also easier to show the process of comparison, not just the conclusion, which helps build consensus both inside and outside the company.

Furthermore, PVSyst includes a feature to compare post‑operation actual performance with simulations in a closely matched way. The official documentation explains that measured data can be imported from almost any text format, viewed in tables and graphs, and compared closely with simulation variables. Moreover, the comparison process is similar to a regular simulation, allowing you to limit the period based on the dates in the measured data and to define which measured values correspond to which simulation variables. This makes it easier to verify how well the assumptions made during the forecasting stage held up in actual operation.

The reason this comparison of actual performance is important is that the value of generation forecasting is not just about "producing numbers in advance." By comparing forecasts with actual results, it provides an opportunity to consider whether the discrepancy is due to differences in weather conditions, overly optimistic loss assumptions, or small equipment-side faults. Official guidance also explains that such comparisons serve as a means to analyze actual system parameters and detect very small anomalies. Even in organizations where design and operation tend to be siloed, using PVSyst makes it easier to connect the people who produced the forecasts with those who oversee operations. Forecasting is used not only for design but also because it makes it easier to feed operational learnings back into subsequent projects.

Limitations You Should Be Aware of Before Implementation

Up to this point, PVSyst may seem extremely versatile. However, before deployment you need to understand its limitations. First, the preliminary design is only a rough estimate at an early stage, and the official documentation explicitly states that it should not be used for detailed system design. Furthermore, its accuracy should generally be expected to have a margin of error of 10% or more, and more precise results should be obtained through hourly simulations. In other words, the fact that PVSyst has simplified features does not mean those simplified features alone can produce precise answers. If you decide to use it, you must proceed on the premise of properly distinguishing between coarse assessments and detailed analysis.

Even detailed simulations strongly depend on the quality of the input assumptions. In formal project design, it is recommended to first create a coarse variant with default values and then sequentially refine it by addressing soiling, angle-of-incidence losses, module temperature, wiring resistance, module quality differences, mismatch, downtime, horizon shading, near-field shading, module layout, and so on. This shows that PVSyst is not software that will return the correct answer with a single click. The quality of the numbers depends on how carefully the assumptions are defined. Conversely, organizations that have operational processes to organize input conditions are better able to realize the benefits of PVSyst.

Moreover, just because an equipment database exists does not mean you can unconditionally trust the recorded values. The official module database description states that database values cannot be guaranteed and, because of the possibility of transcription errors or specification changes, strongly recommends carefully cross-checking them against the latest datasheets when actually using them. It also explains that some of the parameters required by the model are not listed in standard datasheets, so PVSyst may include assumptions or estimates. Internal parameters that affect behavior under low irradiance are particularly important elements in the model. In other words, PVSyst is highly capable, but the responsibility to verify the assumptions ultimately remains with the user.

There are limits to how deeply shading can be evaluated. The official module-layout feature assumes the typical rectangular cell of crystalline modules for detailed electrical shading calculations, and therefore cannot be applied without limitation to all geometries and ultra-large-scale systems. Approximate models suited to regular row arrangements are also provided, but they are noted to underestimate shading in special cases such as thin, long shadows. In other words, while PVSyst can handle shading in depth, the choice of method and the interpretation depend on the input conditions and the characteristics of the target system. It is used for power-yield forecasting because it is powerful, but rather than overrelying on it, those responsible who understand what it can represent and where caution is required can use it more effectively.

Summary

To summarize in one sentence why PVSyst is used for energy yield forecasting: it can handle the factors that determine energy production in a single workflow from meteorological conditions to performance comparison. It can base analyses on meteorological conditions, perform detailed hourly simulations, reflect installation conditions and equipment configuration, explain the breakdown of losses, examine shading effects in depth, and facilitate comparisons and validation against actual performance. Because these six capabilities are present, PVSyst is positioned not as mere estimation software but as practical software that supports design decision-making.

From the perspective of practitioners, the value of PVSyst is not the mere production of numbers. Its value lies in being able to organize everything — which assumptions are used, which losses are anticipated, which options are compared, and how the results are explained. If your objectives are to improve the accuracy of generation forecasts, strengthen the explanatory power of design, or connect post‑operation actuals and forecasts to enable improvements, PVSyst is a highly suitable choice. On the other hand, if you intend to make a final decision based only on a simple estimate, you may not be able to fully leverage the depth of its functionality. Before implementation, it is important to clarify why your company is conducting forecasts.

The more carefully you refine power generation forecasts in desk-based analyses, the more important the accuracy of on-site location information and equipment layout becomes. While simulations treat the orientation of the mounting surface and shading conditions in detail, if the actual on-site layout or positioning deviates, the gap between desk-based assumptions and reality widens. That is why, after firming up the design-side outlook in PVSyst, combining it with an on-site setup that can handle high-precision positional information makes it easier to improve consistency from planning through construction and maintenance. If you are aiming for that level of on-site accuracy, it is natural to also consider LRTK, an iPhone-mounted GNSS high-precision positioning device. The idea of refining predictive accuracy with PVSyst and aligning on-site positional accuracy with LRTK helps stabilize the practical execution of solar projects.