Real-Time Integrated Performance Analysis of Photovoltaic Systems under Environmental Stressors in Baghdad

This section constitutes an analytic review of the responses achieved concerning the practical part of this research by means of detailing a few graphical curves that prove this way in which the system behaved based on certain rigorous field data measured using sensors; for 11/11/2025, power generation by the solar panel was integrated with climate data imported from Open Meteo database to increase accuracy in how these proposed models came into existence.

3.1 Realistic data results

Figure 5 illustrates the temporal variation in solar radiation and electrical output throughout the day. A gradual increase in radiation and power is observed during the morning, as the angle of incidence of sunlight increases. Radiation reaches its maximum around midday, while electrical power peaks in the late morning. It then gradually decreases despite radiation remaining at relatively high levels, a behavior attributed to the effect of high temperatures on solar cells and the accompanying decrease in efficiency. As daylight hours progress toward evening, both radiation and power decrease significantly as the sun's angle declines and direct radiation diminishes, until production reaches nearly zero at sunset.

Figure 5. Time correlation between radiation intensity and output powe

Figure 6 illustrates Daily changes in solar radiation, panel surface humidity, and atmospheric humidity. It is noted that the radiation intensity rises rapidly during the morning hours to reach its maximum values at midday and then begins to gradually decrease towards the evening, while atmospheric humidity records high values in the early morning and then decreases with increasing temperature and increasing radiation as a result of evaporation processes, while the panel surface humidity shows a similar behavior as it decreases during periods of high radiation As a result of the surface temperature rising and then rising again in the evening with a decrease in radiation and an increase in atmospheric humidity, this behavior reflects the thermal and environmental interaction between solar radiation and humidity and their direct impact on the surface properties of solar panels.r

Figure 6. The daily relationship between solar radiation, solar panel humidity, and atmospheric humidity in terms of time

Calculations of solar radiation, solar panel temperature, and ambient temperature throughout the daytime hours are displayed in Figure 7. This data indicates solar radiation is available early in the morning, peaks at noon, and decreases gradually as we approach evening. When solar radiation strikes a solar panel, the panel's temperature is typically higher than the ambient air temperature. In fact, that increase can be significant. In contrast, the rise in ambient temperature is slower and smaller. From noon onwards, the panels cool down more quickly than the home air, either because of lower radiation intensity and/or greater heat loss from their outer surfaces. This shows how solar radiation affects panel temperature and illustrates the external weather conditions around this time of day that can skew that data.

Figure 7. Daily change in solar radiation, solar panel temperature, and air temperature in terms of time

Figure 8. Daily change in wind speed and solar radiation

Figure 8 shows the daily variation of solar radiation and wind speed between 6:00 a.m. and 6:00 p.m. Solar radiation begins at approximately 36 W/m² around 6:00 a.m. and soon increases rapidly, reaching, on average, around 450 W/m² by 8:00 a.m., before gradually increasing to levels around solar noon, during the range of 11 a.m. to 1 p.m., with values from about 545 W/m² to 550 W/m². By now, radiation levels have been reduced further into the normal range, with a mean of about 150 W/m² at 4:00 p.m. and about 20 W/m² at around 6:00 p.m. That’s where the sun takes its path through the sky. The wind speed, on the other hand, has a certain profile with a moderate 5 m/s speed at the origin (6:00 a.m.); it's gone down to 2–3 m/s almost every single time, again and again, at the same hour (7:00 and also 8:00 a.m.), and yes, it reaches its lowest (0 m/s) during time phases from continued up until the very recently mentioned time period (on the side of an exercise during one such era). That's after which there might as well have been some rise. But that means there is a certain quiet in the air when solar radiation is at its most intense. With a jump at 3:00 p.m., wind speeds rise to 7 m/s and settle down at 4–5 m/s by the end of the afternoon; heating and local convection explain the increase in wind speed around 3:00 p.m. It highlights the high temporal variability of solar radiation and wind speed, the intermittent nature of both aspects, and their corresponding effects on the performance of renewable energy systems over an average day.

Figure 9 illustrates the daily behavior of both electrical power and output current based on experimental data, where values start near zero in the early morning hours and then rise significantly with increasing solar radiation intensity, reaching about 36 watts at 10 in the morning. In contrast, the current stabilizes at approximately 2.1 amperes. After that, the power and current gradually decrease throughout the afternoon and beyond as intensity, radiation, and plate temperature decrease. This behavior reflects the constant changes in the PV system's electrical operating conditions during daylight hours. illustrates in Figure 10 the daily behavior of the output voltage and electrical power as both parameters increase rapidly during the morning hours to reach peak values around 10:00 a.m. with the voltage approaching 20 V and the power around 22 W and then gradually decrease after midday as the voltage and power steadily decrease towards the afternoon and evening to reach nearly zero by 6:00 p.m., reflecting the direct dependence of the system outputs On the energy inputs available during the daily operating period.

Figure 9. Daily variation of photovoltaic (PV) output power (W) and output current (A) from 6:00 a.m. to 6:00 p.m.

Figure 10. Daily variation of photovoltaic (PV) output voltage (V) and output power (W) from 6:00 a.m. to 6:00 p.m.

As shown in Figure 9, the highest power output was achieved in the late morning. In contrast, the subsequent decrease in power and current indicates the combined effects of reduced solar radiation, high plate temperature, and gradual dust accumulation during the afternoon.

Figure 11 shows the daily change in the concentration of dust suspended in the air, dust deposition on the panels, and the capacity generated. Air dust concentrations begin at about 10 mg/m³ at 6:00 a.m., rise to an average of about 40 mg/m³ at 11:00 a.m., and then decrease significantly during the late afternoon. Airborne dust concentrations and dust concentrations on solar panels are shown daily, along with the predicted impact on power generation. Dust concentrations in the atmosphere begin at ~10 mg/m³ around 6:00 a.m., then increase to an average of ~40 mg/m³ at 11:00 a.m., and decrease significantly to ~2 mg/m³ by ~5:00 p.m. By comparison, deposition behavior on the panels is more predictable, peaking at ~34 mg/m² at 4:00 p.m., since the panels still accumulate dust and practically no removal occurs in the afternoon, compared with around 2 mg/m² in the morning. The power output, rising to 36 W at 10:00 a.m. (the a.m. start), has shown a steady trend as dust builds in the air and solar radiation increases. By the late afternoon, however, production gradually decreases towards approximately 3:00 p.m. and ~20 W, and the final output drops to almost 0 by the end of the day. Reduced transmission in direct sunlight is attributed to dust deposition on the PV panels reducing transmittance and the amount of solar radiation reaching the panels. Airborne dust particles will underperform panels sooner than settled dust because most of the dirt accumulates on panels in the afternoons. This highlights how often these require cleaning in order for the solar system to run at its maximum.

Figure 11. Daily variation of airborne dust concentration (mg/m³), panel dust deposition (mg/m²), and generated power (W) from 6:00 a.m. to 6:00 p.m.

According to measured data, dust accumulation on the plate surface increased from about 2 mg/m² in the morning to about 34 mg/m² in the late afternoon, accompanied by a significant decrease in production capacity. These results suggest that the cleaning process should be performed when dust accumulation reaches 30 mg/m². From the collected data, considering the recorded conditions, this equals about one cleaning cycle per 1–2 days during dusty periods. In addition to periodic cleaning, dust mitigation measures are preventive measures such as anti-soiling coatings, self-cleaning hydrophobic surfaces, and increased angles on panel inclination to limit dust adhesion. Such technologies also aid in diminishing optical losses and the volume of maintenance work per day, particularly in dry environmental conditions.

The main observation in Figure 11 is that during dust deposition on the panel surface, the amount of generated power diminishes as the dust deposit rises and is in fact very close to the maximum measured value. This means that they should wash or use Anti-Soiling Mitigation Techniques if dust accumulates and/or is needed; this is an acceptable operational aspect. Furthermore, the findings also show that PV performance is influenced by the combined effect of environmental stressors as opposed to each factor alone. High plate temperature leads to low effectiveness as a means of electric conversion, and dust accumulation decreases the optical transmittance. When the two are present simultaneously, the power loss increases. The measured power was reduced from about 36 W to approximately 13 W as dust deposition reached its maximum measurement value of approximately 34 mg/m² due to the combined effects of dust accumulation, heat stress, and suppressed solar radiation emitted during the afternoon. The simultaneous presence of relatively high humidity and dust during the early hours of the morning may enhance particle adhesion to the panel surface, reducing light penetration, increasing monitoring and cleaning work on the panels for dust-infested climates.

A simplified mathematical model may be constructed that explains the quantitative relationships between output power and these primary environmental factors with respect to solar radiation intensity, panel surface temperature, and dust buildup. This relationship can be deduced in a simpler version in Eq. (1), which contains positive radiation-related cross-line energy entering the system, whilst negative effects are introduced into the system, such as high temperature, which leads to low electrical efficiency, and increased dust accumulation, which leads to a decrease in optical permeability. According to this model, the capacity generated does not come from any single component but from the intricate interdependences between ecosystem components that regulate the functioning of the system in the environment [21, 22].

$P=k\cdot G\cdot \left( 1-\beta \left( T-Tref \right) \right)\cdot \left( 1-\gamma D \right)$                 (1)

3.2 Forecast data with realistic data results

Figure 12 shows a rigorous evaluation of the credibility of meteorological data, real vs. expected values for humidity and wind speed throughout the day. The actual humidity curve illustrates s a concave behavior that starts with high values in the morning at 33 g/m² and then gradually descends to reach its lowest levels at noon, recording about 18 g/m² as a result of rising temperatures and evaporation of suspended humidity before rising again to reach 36 g/m² in the evening. The predictive data for humidity simulate the same directional path, but with lower values, reflecting a fixed margin of error in the estimate. As for wind speed, the real data revealed a period of stagnation and complete stillness during which the speed disappeared to reach zero in the period extending from nine in the morning until two in the afternoon, which the predictive data failed to detect, as the speed during that period was estimated at an average of 3–5 m/s.

Figure 12. Daily comparison between measured and expected values of humidity and wind speed in terms of time

Figure 13 reveals that the imported atmospheric data matches the actual reality of radiation and temperature; as you can tell in this figure, the solar radiation values differ fundamentally: the predictive model used (in yellow) overestimated its output, peaking at about 660 W/m² around 12 o'clock in the afternoon, whereas the actual results (in orange) show that the actual irradiance was below 550 W/m². This difference could be due to local environmental conditions—such as dust or passing clouds—that didn’t enter the forecasts. For the temperature data, relative convergence is observed along the general path. However, actual readings fluctuate, dipping slightly from 2:00 p.m. to 24 ℃, while the prediction curve displays more moderate and stable values.

Figure 13. Analysis of the standard gap between realistic and predictive data for solar radiation rates and temperatures

The higher-than-real estimate of projected solar radiation values is mainly due to subtle local climatic factors that are not fully represented in large-scale atmospheric models. These factors include dust concentration, temporary shading, Atmospheric Aerosols, and site-specific environmental conditions.

The accuracy of predictions can be improved in the future by integrating local sensor measurements into the prediction process and applying machine learning-based correction models trained on historical measurements and past predictions.

Integrating forecast data with real-time measurements enhances PV performance monitoring by enabling direct comparison between forecasted and actual operating conditions. This approach enables the identification of local environmental impacts, such as dust accumulation and microclimate changes, that large-scale forecast models cannot fully represent, thereby contributing to more efficient and accurate performance assessment and maintenance planning.

The discrepancy between the forecast and measured data indicates a systematic overstatement by the forecast model, particularly in solar radiation. In this study, the Root Mean Square Error (RMSE) was about 85 W/m². In comparison, the Mean Absolute Error (MAE) was about 72 W/m², which confirms that the model was unable to represent local operating conditions fully. The main sources of this bias may include local dust, Atmospheric Aerosols, temporary shading, wind stagnation near the Earth's surface, and microclimatic influences surrounding the test site. Therefore, forecasting models for PV applications in dry areas should be improved by incorporating high-spatial-resolution weather data, local dust measurements, and real-time feedback from local sensors. Future models could also rely on machine learning-based correction algorithms trained using measured and forecast data to reduce forecast bias and improve the reliability of results.

3.3 Optical monitoring and digital processing system

Table 2 illustrates the analytical outputs of a direct optical monitoring system, in which an installed electronic camera scans the solar panel's surface and processes images algorithmically to determine dust accumulation rates and any optical distortions that may impede optical absorption. The embedded data illustrates the system's response to different test scenarios, with the first image recording the highest observed distortion rate. This is due to a deliberate sensitivity test in which white pieces were placed on the plate surface to verify detection accuracy. The table also highlights a special case in which strong reflections of sunlight appear on the surface of the painting, as the camera deals with it and reads it as a distortion ratio, which explains the mechanism of the system's work in classifying any change in the optical homogeneity of the surface, whether physical, such as dust and foreign objects, or optical, such as sharp reflections.

Table 2. Analytical data for digital camera outputs

The optical monitoring system adds another method for consumers to monitor the health of their panels. Unlike ordinary electrical measurements, its features allow anyone to monitor the panels’ surfaces at all times. By implementing image processing features, it’s much easier to notice things like dust, foreign objects, or strong reflections — things that electrical checks might ignore. This helps users find faults early and plan maintenance, since they can actually see the panel’s condition. In large PV setups, the system also has high ease of use. This is a low-cost option for real-time monitoring and predictive maintenance – especially in dusty areas where surface dirt may seriously hinder performance or the efficiency of equipment.

3.4 Statistical analysis paragraph

The statistical assessment was utilized to determine the correlation between the environmental variables and the electrical performance of the PV system. Pearson's correlation coefficient was employed to measure the strength of the relationship between solar radiation, panel temperature, dust accumulation, and production capacity. Such a tendency is supported by the output data; for the two variables, production and solar intensity, the correlation is significantly similar (r = 0.91), which means that the production of power is mainly from solar input. In contrast, high panel temperature was only moderately negatively related to the system's production capacity (−0.6336), confirming that high temperatures reduce system yield even when there is considerable radiation. Dust deposition on the surfaces of the panels also negatively correlated with production capacity, demonstrating that deposition of dust is not only detrimental to the transmittance of light but also impacts productivity at various points of the day — notably in the afternoon, when dust accumulation is most inevitable. Wind speed had a weak positive correlation with production capacity (~0.32), suggesting that surface cooling from wind slightly improved performance but also measured performance. Errors (the difference between observed and predicted) were used to measure prediction accuracy. The intercept values for solar radiation were 85 W/m² with RMSE, with an MAE of about 72 W/m² indicating that the predictive data could be overestimated. The differences have been attributed to more local environmental conditions (dust, microclimates) that general predictive models do not consider. Generally speaking, these evaluations support the observation of solar radiation as the major force of energy generation while temperature and dust build-up are its principal challenges. All of these were significant points indicating that the performance in desert conditions was only sustained by good thermoregulation and frequent cleaning of the surface.

This dataset was chosen as the most representative observation day, since it demonstrated maximum completeness and encapsulated most of the response behaviour of the PV system to environmental changes across multiple measurement days. The findings indicated that the PV unit's power output was dependent on solar energy. Meanwhile, panel temperatures above around 40 ℃ displayed a significant reduction in power output even with ample solar irradiance. Likewise, as dust accumulation on the panel surfaces increased, energy generation decreased, especially when deposition was near the greatest readings during that observation time. Temperature control and minimizing dust collection were mentioned as crucial parameters to achieve the best performance of PV systems in these scenarios. Future work could be performed using large, multi-day datasets, as well as developing predictive models using Regression and machine learning models to pinpoint reliable operational limits and scenarios that have improved cooling with less dust settling on solar panels.