Estimation of solar radiation distribution is crucial for the performance, design, and economic evaluations of solar panels and/or collector systems operating under various climatic conditions, tilt angles, and geographic locations. A comprehensive study involving the combined effect of tilt angle as well as ambient temperature for maximizing the PV array power output was performed. At first, we present a comparison between different isotropic and anisotropic models showing that the anisotropic model gains 5% more energy than the isotropic one. We report monthly and yearly optimum tilts derived from the anisotropic model. Utilizing the optimum tilt derived from the selected anisotropic model, a case study of a mono-crystalline silicon PV array with 2.76 kWp of the rated power is carried out to evaluate the PV performance in five cities of the Kingdom of Saudi Arabia. The results show that the estimated yearly optimum tilt angle is close to the latitude of the studied cities. For the city of Dhahran, a gain of 4.2% power generation is achieved at ambient temperature through monthly adjustment of the PV module instead of yearly adjustment. The estimated yearly tilt angles are as follow: 27.3° for Dhahran, 26.0° for Riyadh, 22.7° for Jeddah, 32.7° for Arar, and 20.1° for Abha. Although Riyadh and Arar receive the same annual average GHI of 6.0 kW/m ² , the yearly average PV power output is ~7.1% higher for Arar (1.50kW) compared with Riyadh (1.40 kW). This is mainly attributed to the fact that Riyadh has a higher annual average ambient temperature of 29°C compared with 23°C for Arar. Thus, in addition to panel orientation, the ambient temperature was found to have a significant impact on the performance of the PV system and should be taken into consideration when designing the system.

In the above article [1] , there has been an error in the last reference section of the final version of publication. Ref. [31] was duplicated with Ref. [34], and Ref. [36] was duplicated with Ref. [30]. Hence in the reference section of [1] , Refs. [31] and [36] should be replaced with the new Refs. [2] and [3] , respectively, without any modifications in the text and the order of other references.