Solar Energy Element Simulation Using Group-Based Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for simulating the quantity of interest in solar energy installations, such as power output, are inefficient due to high computational costs and lack of standardization, especially for large and complex installations with emerging technologies like bifacial photovoltaic modules and sun tracking systems.
Innovation Solution
A method that identifies groups of solar energy elements with similar performance variables, selects representative elements, and calculates the quantity of interest for these representatives to determine the overall installation performance, using physical parameters and clustering algorithms like k-means or k-nearest neighbors, reducing computational resources and eliminating empirical tuning coefficients.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the most accurate approach is used to determine the quantity of interest for each solar energy element, then the accuracy of the simulation is improved, but the computational cost increases significantly
Solution Approach 1:
The patent divides the solar energy installation into multiple groups based on performance variables (such as irradiation conditions, shading patterns, and orientation). By segmenting the installation into homogeneous groups, the method enables selective simulation of representative elements from each group rather than every element, thereby reducing computational cost while maintaining accuracy within each segment.
Solution Approach 2:
The patent introduces performance variables as grouping criteria to classify solar energy elements with similar characteristics. By changing the parameter space from individual element analysis to group-based analysis using performance variables (irradiation, shading, orientation), the method achieves computational efficiency without sacrificing the accuracy needed for each element's quantity of interest.
2Loss of energy
If the lumped approach is used to reduce computation costs, then the computational cost is reduced, but the accuracy and standardization are worsened due to dependence on engineer experience
Solution Approach 1:
The patent segments solar energy elements into groups based on objective performance variables rather than subjective engineer selection. This segmentation provides a systematic, reproducible approach that eliminates dependence on individual engineer experience while maintaining computational efficiency through group-based representation.
Solution Approach 2:
The patent transforms the lumped approach from a subjective, experience-based method to an objective, parameter-driven method by using performance variables (irradiation, shading, orientation) as grouping criteria. This parameter change enables standardization and reproducibility while preserving computational efficiency.
3Ease of operation
If representative solar energy elements are selected based on engineer experience, then the selection process is simple, but the method cannot be standardized and lacks reproducibility
Solution Approach 1:
The patent replaces subjective engineer judgment with objective performance variables as the basis for selecting representative elements. By using measurable parameters (irradiation conditions, shading patterns, orientation angles), the method achieves both simplicity in selection and full standardization, enabling reproducible results across different engineers and projects.
4Productivity
If the lumped approach is used for complex installations with irregular terrain and shading, then the computation is faster, but the accuracy deteriorates due to heterogeneity of solar energy elements
Solution Approach 1:
The patent segments heterogeneous solar energy elements into homogeneous groups based on performance variables that capture local conditions (irradiation, shading, orientation). This segmentation allows the method to handle complex, heterogeneous installations by treating each homogeneous group separately, maintaining accuracy for diverse conditions while preserving computational efficiency through group-based simulation.
Solution Approach 2:
The patent applies local quality by allowing different groups to have different representative elements and simulation parameters based on their specific performance variables. This enables the method to adapt to local conditions (irregular terrain, shading patterns) within each group while maintaining overall computational efficiency, rather than applying a uniform approach to the entire installation.
Data Source
Figure 1
Figure 2~3
Figure 4~5B
AI summary
A method for simulating a quantity of interest of a solar energy installation comprising a plurality of solar energy elements (2) and/or of parts thereof, comprising: identifying (103) a plurality of groups of solar energy elements within the solar energy installation, wherein a performance variable is similar for each of the solar energy elements (2) within the group; obtaining (104), for each group, a representative solar energy element (2) of which the performance variable is representative for the group; simulating (105) a quantity of interest of the representative solar energy element (2) of each group; and determining (106), from the quantity of interest of the representative solar energy element (2) of each group, the quantity of interest of the solar energy installation and/or of at least one solar energy element (2) of the solar energy installation.