AI Analysis of Sun Tracking Efficiency

Engineering notes

Engineering notes will be added by the Power for Solar editorial team.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为光伏效率、钙钛矿太阳能电池、储能技术、太阳能热利用、BIPV、并网技术等高价值论文补充中文说明。

Original abstract

For two years in a row, 2025 and 2026, the electrical energy generated by PV constitutes about 10-12% of all electricity produced across the world [1]. Significant amount of solar energy was produced by the PV systems with fixed tilt. Only 40-45% of the PV energy was generated by sun tracking systems [2]. The amount of energy produced (in kWh/m2/day) is defined mostly by efficiency of solar panels. The second important factor of PV energy collection is the productivity of sun tracking, i.e. the control by algorithms of sun tracking. In the cost of energy units produced by solar panels there are some other economic factors to be considered, such as basic cost of tracking equipment, cost of maintenance and cleaning of panel surfaces. The initial low cost and maintenance of single-axis tracking systems (SATS), smaller basic size made this machinery dominant in sun tracking [3]. The methods of control of SATS and algorithms of usage at different weather conditions define real efficiency of tracking systems. Recently we developed instant measurements of sun irradiation using three optical sensors installed at sun tracker which cover entire sun spectrum [4]. As the scanner expected to make step from position “A” to position “B”, the control algorithm compares the sun intensity in both positions. The step is allowed if in the position “B” solar panels would gain energy. Various weather conditions are programmed in the algorithm what allows flexible tracking control [5]. In the current brief study, we subjected our novel control of sun tracking to be compared with other scanning systems by AI. Conclusions of AI analysis are presented.

Links and sources

• PDF from original source

Comments