9 Major Factors That Change Annual Energy Production in PVSyst

Table of Contents

• How differences in annual energy production arise in PVSyst

• Factor 1 Consistency between meteorological data and site

• Factor 2 Azimuth and tilt angle settings

• Factor 3 Shading conditions and how to build the 3D scene

• Factor 4 Module specifications and how to think about initial performance

• Factor 5 PCS settings and how to set the DC/AC ratio

• Factor 6 String configuration and the occurrence of mismatch

• Factor 7 How to estimate temperature losses

• Factor 8 Settings for various loss coefficients

• Factor 9 Availability losses and operational assumptions

• Perspectives to connect annual energy production to practical decisions

How differences in annual energy production arise in PVSyst

When you calculate annual energy production with PVSyst, results can differ more than expected even for what looks like the same project. A common practical mistake is to assume that the annual production is determined roughly by module capacity or PCS capacity alone. In reality, annual energy production is not determined by a single setting; it is formed by the overlap of multiple conditions such as meteorological conditions, installation conditions, shading, temperature, electrical grouping, loss coefficients, and operational assumptions. Because of this, even small differences in settings can accumulate into non-negligible differences over a year.

Also, PVSyst is not merely a generation-calculation tool but a tool for organizing design assumptions. Because the inputs you enter are reflected directly in the results, it is important not only whether the annual production is high or low but whether you can explain why that number was produced. For practitioners, the key is not to create the highest possible number but to verify that the number is reasonable given the site and design conditions. The ability to interpret differences in annual energy production translates directly into the quality of design decisions.

Moreover, differences in annual energy production are not viewed only to simply decide which option is superior. One option may be slightly disadvantaged in annual totals but realistic when considering constructability and maintainability, while another option may look good numerically but rely on idealized assumptions that are likely to fail later. This is why it is meaningful to organize the major factors that change annual energy production. Below, from a practitioner’s perspective, we examine nine factors that tend to create differences in annual production in PVSyst.

Factor 1 Consistency between meteorological data and site

One of the factors with the greatest impact on annual production is consistency between the meteorological data and the site. In PVSyst, the choice of which meteorological dataset to use determines the foundation of the production calculation. Therefore, if the actual site conditions and the location represented by the meteorological data do not match, the overall results will tend to vary even if equipment conditions are set carefully. When differences in annual production are large, you should first question this starting point.

In practice, you may use data from a representative location near the candidate site or reuse nearby data from a previous project. This approach is not uncommon, but proceeding without checking how well that location represents the current site reduces the accuracy of comparisons. Differences in elevation, coastal versus inland location, how open the surrounding terrain is, and seasonal irradiation tendencies will change not only annual production but also monthly output profiles. As a result, what you thought were equipment-condition differences may actually be differences in meteorological assumptions.

Meteorological conditions are not just irradiation; temperature conditions also affect temperature losses, so the choice of meteorological data can significantly change impressions of annual production. When comparing options in practice, you should be clear about which variables you want to examine and align other conditions—especially site and meteorological assumptions. When reading annual energy production in PVSyst, confirming that these basic conditions are realistic is indispensable.

Factor 2 Azimuth and tilt angle settings

Azimuth and tilt angles are also representative factors that easily create differences in annual production. In PVSyst, small changes in orientation and angle alter not only annual totals but also monthly generation trends. Therefore, while you may be tempted to adopt the azimuth and tilt that look best on paper, in practice that orientation may not be adoptable directly due to constraints such as site shape, slopes, earthworks, and maintenance access.

For example, placing a near-ideal azimuth may make the numbers look good, but that orientation could create unnatural array layouts or increase edge losses. The same applies to tilt: an angle that looks advantageous for generation may require larger row spacing, reducing land-use efficiency. When comparing annual production in PVSyst, it is important to evaluate azimuth and tilt not as standalone optimums but within the realistic conditions that can actually be implemented on the site.

Also, slight differences in orientation and tilt assumptions change the meaning of comparison options. One option may be favored by azimuth while facing strict shading or layout constraints. Another option may be slightly disadvantaged by azimuth but may be more natural for the whole array and maintenance. Looking not only at the annual production numbers but also at whether that orientation and tilt can be implemented reasonably in the design makes PVSyst results more useful for practice.

Factor 3 Shading conditions and how to build the 3D scene

Shading conditions strongly affect annual production, yet they are an input that is prone to causing variable results depending on how they are entered. PVSyst allows shading to be represented via 3D scenes or Near Shading, but if the model construction does not match field conditions, annual energy production can appear either too high or too low. If shading is treated too optimistically, results will be optimistic; if treated too conservatively, values may be much lower than reality.

In practice, on-site self-shading, slopes and retaining wall shadows, and shadows from buildings and trees overlap. If these are all lumped together as a single shading factor, improvable shading and unavoidable shading are mixed, making it difficult to prioritize design measures. Self-shading between rows might be mitigated by changing row spacing or tilt, while shadows from off-site buildings may be difficult to improve by small layout changes. When PVSyst shows low annual production, it is very important to separate and examine these components.

Shading also varies by season and time of day, not just annually. Shading that is strong only in winter mornings and evenings has a different design implication than shading that slightly affects generation year-round; even if the annual loss is the same, the impact on system performance differs. The purpose of PVSyst’s 3D scene is not to make a pretty visualization but to organize primary shading causes and read how they affect the annual total. When differences in annual production are large, always revisit shading conditions and the assumptions behind the 3D scene.

Factor 4 Module specifications and how to think about initial performance

Differences in module specifications and how you treat initial performance can also greatly change annual production. In practice, attention often first goes to rated power and dimensions, but when comparing in PVSyst it is important to consider how a module looks when applied to the actual installation conditions. A visible output difference does not necessarily translate directly to an annual production difference.

For example, even on the same site, changing module dimensions alters how many modules can be placed and how arrays are grouped. High-power modules are not always advantageous; slightly smaller modules may reduce edge leftovers and be favorable in total capacity. Including assumptions that affect first-year performance, such as how to treat initial losses, can change the apparent ranking. PVSyst is less a tool for comparing individual equipment performance and more a tool for comparing system performance when equipment is placed in field conditions.

Moreover, module differences tie into temperature losses and mismatch visibility. It is important not to look only at first-year numbers but to observe how a module behaves within shading, temperature, and string configuration. If you can separate whether annual production differences come from equipment differences or from compatibility with layout, the accuracy of comparisons improves significantly. Module selection should not be decided by apparent output alone but by how it appears within PVSyst under realistic conditions.

Factor 5 PCS settings and how to set the DC/AC ratio

PCS settings and the DC/AC ratio are also major factors that change annual production. In practice, module-side capacity is often decided first and PCS is allocated afterward, but whether that balance is natural greatly affects how output clipping and conversion losses appear. When annual production in PVSyst is low, you need to check whether the balance is forcing the design beyond what is reasonable for that project.

For example, loading the DC side heavily can increase annual production totals, but it may also make PCS clipping more noticeable. Conversely, leaving margin on the PCS side might make the configuration straightforward, but it could fail to fully utilize the site or module count potential. When viewing PVSyst results, it is important to understand not just whether production is high or low but under what degree of limitation or margin those numbers are produced.

PCS conditions also interact with string configuration and wiring conditions. A naturally cohesive configuration produces different-quality annual production numbers than a configuration forced to work artificially. In practice, even the same one percent difference can favor the option that is easier to explain and handle. When comparing annual production in PVSyst, view PCS settings not in isolation but as part of the overall system balance including DC/AC ratio, string configuration, and wiring.

Factor 6 String configuration and the occurrence of mismatch

String configuration strongly influences annual production. In practice, if module counts and PCS conditions are satisfied it may look fine, but in PVSyst the visibility of mismatch losses depends on how modules with differing conditions are grouped within the same string. When shading, orientation differences, tilt differences, and temperature differences coexist within a grouping, annual totals may not grow as much as expected.

For example, if rows divided by walkways or terrain conditions are forcibly treated as a single group, the layout may seem orderly but internal variation may be large. If columns with different edge conditions, columns shaded only in the morning, or columns with slightly different azimuths are mixed, the natural coherence of the string can break down. When PVSyst shows low production, reviewing how circuits are grouped can sometimes reveal the cause.

Mismatch is also related to shading, temperature, and soiling. Instead of treating it as an isolated loss, read mismatch as a phenomenon created by layout and circuit partitioning. When comparing options, checking how annual production changes with slight modifications to string configuration makes it easier to see which differing condition is dominant. PVSyst should be used not only to read device performance but also to assess the naturalness of circuit grouping.

Factor 7 How to estimate temperature losses

When annual production is low, how temperature losses are estimated is another factor that must not be overlooked. In practice, ambient temperature is often treated simply as part of meteorological data and temperature losses are handled later in summary, but what you should look at in PVSyst is not only the ambient temperature but the on-site conditions under which the equipment absorbs heat. Array density, ventilation, and the presence of surrounding structures change how temperature is perceived.

For example, even in the same region, an open layout with good airflow and a compact layout with many structures will experience equipment heating differently. Tilt angle and row spacing alone can change this impression. If temperature losses are estimated uniformly, one option may be treated too conservatively while another is too optimistic. When comparing annual production in PVSyst, confirm whether the temperature loss assumptions are natural for the site conditions.

Temperature losses also tie into module differences. An option that looks advantageous by rated power alone may show a smaller advantage once temperature conditions are considered. When annual differences between options are smaller than expected, suspect not only shading and wiring but also how temperature losses are set. Annual production in PVSyst is decided by the overlap of many losses, so pay careful attention to temperature conditions that are easy to overlook.

Factor 8 Settings for various loss coefficients

Settings for various loss coefficients are another factor that greatly change how annual production appears. If shading, temperature, soiling, wiring, mismatch, and availability are each set slightly conservatively, each may be natural on its own but together the overall assumptions can become quite strict. When PVSyst shows low production, a surprisingly common cause is this “over-accumulation of losses.”

In practice, because people prefer to look on the safe side, there is a tendency to set each item conservatively rather than modestly. That approach is not bad in itself, but when loss roles overlap, production can appear unnecessarily low. For example, if shading is modeled in detail while similar adverse conditions are also heavily counted in another item, or if maintenance conditions are layered into both availability losses and soiling losses, the annual total will look more conservative than reality.

Also, when reading comparison options, if loss coefficient settings are not aligned you may end up seeing differences driven by assumptions rather than design. If the same equipment configuration has harsher soiling losses in one option, the meaning of the annual production difference weakens. When investigating why PVSyst shows low output, examine the loss tree and individual loss items to identify which are dominant and which are overlapping.

Factor 9 Availability losses and operational assumptions

The final major factor is availability losses and operational assumptions. In practice, there is a tendency to view equipment as basically running continuously, but on-site reality includes inspection, cleaning, failures, communication interruptions, and delays in recovery, all of which create times when generation does not follow theory. How PVSyst accounts for these will greatly change the realism of the annual production figure.

For example, on a site with easy access and organized sections, inspections and abnormal-response actions may proceed relatively quickly. Conversely, on a site with difficult walkways, multiple separated sections, and time-consuming site understanding, the time from shutdown to recovery may be longer even with the same equipment. If you process such differences with a uniform availability loss, the realistic strengths of each option become hard to see.

Availability losses are also connected to shading, soiling, and maintainability. An option that is easy to clean may be advantageous not only for soiling but also in the assumption of downtime; an option with simple sectioning may be strong in fault isolation and recovery. When PVSyst shows low annual production, it is important not to look only at equipment performance differences but to verify whether that number is natural when operational assumptions are included.

Perspectives to connect annual energy production to practical decisions

What is common to the nine factors examined above is reading annual production not as a mere final result but as the accumulation of assumptions. Meteorological data, site conditions, shading, modules, PCS, strings, temperature, loss coefficients, and availability losses each affect the result a little. Therefore, when you feel production is low, the important action is not to hastily tweak a single setting but to systematically organize which conditions are dominant.

For practitioners, the real importance is not creating the highest annual production. What has value is being able to explain why that number resulted. Even if the production appears low, if it is a natural low reflecting site conditions, the design is still meaningful. Conversely, if a high production number rests on overly ideal assumptions, it is likely to collapse later. It is important to view PVSyst not as a tool for producing ideal values but as a tool for organizing real-world conditions and converging on a convincing number.

Also, improving the accuracy of annual production requires not completing everything with desk-based simulation alone. If site boundaries, slopes, buildings, trees, walkway conditions, existing equipment, and maintenance access are ambiguous, settings tend to be biased toward ideal. To connect PVSyst results to practice, you need to iterate between site understanding and simulation, continually checking which assumptions are realistic and which are optimistic. Annual production is both a number on a screen and a summary of site conditions.

In that sense, when you want to proceed more reliably with on-site position confirmation and coordinate acquisition, using iPhone-mounted GNSS high-precision positioning devices such as LRTK can be effective. If on-site position information and site conditions are easier to organize, placement assumptions, obstacle conditions, and walkway conditions that influence annual production in PVSyst become clearer. If you can improve desktop comparison accuracy with PVSyst and support on-site understanding accuracy with LRTK, improving annual production becomes not just numerical tuning but movement toward site-based practical decisions. Carefully reading the factors that change annual production not only increases simulation accuracy but also enhances design capability that links desk work to the field.