PV Array Topology Reconfiguration for Partial Shading Losses
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Photovoltaic (PV) energy systems face significant power losses due to partial shading caused by environmental and man-made obstructions, leading to voltage and current mismatch issues that reduce grid power supply, and existing reconfiguration methods are not scalable or cost-effective.
Innovation Solution
A regularized deep neural network architecture is employed for PV topology reconfiguration, incorporating dropout and batchnorm, which dynamically switches between series-parallel, bridge-link, honeycomb, and total-cross-tied topologies based on observed irradiance data to maximize power output, incorporating wiring losses and eliminating the need for additional unshaded panels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If conventional fixed topology or traditional reconfiguration methods are used, then implementation is simpler, but power loss under partial shading conditions is significant and scalability is limited
Solution Approach 1:
The patent replaces traditional mechanical/optimal control-based reconfiguration methods with a deep neural network system that uses electrical measurements (voltage and current) to predict and optimize topology configurations. The neural network learns optimal reconfiguration strategies from training data without requiring complex real-time calculations or additional mechanical components, thereby reducing power loss while maintaining manageable system complexity.
Solution Approach 2:
The patent creates a virtual model of the PV array through the neural network that copies and learns from historical operating data. This virtual model enables the system to predict optimal topologies without requiring physical prototypes or extensive real-world testing, reducing both power loss and the complexity associated with traditional trial-and-error optimization approaches.
2Productivity
If advanced reconfiguration methods are implemented to reduce power loss, then power output improves, but scalability and cost-effectiveness deteriorate
Solution Approach 1:
The neural network system is designed to be universally applicable across different PV array configurations and shading scenarios. By training the network on diverse datasets representing various topologies and environmental conditions, the same system can be deployed across multiple installations of different sizes and configurations without requiring re-engineering, thereby improving scalability while maintaining high power output through optimized reconfiguration.
3Use of energy by moving object
If topology reconfiguration is performed to maximize power output, then energy efficiency improves, but system complexity and computational requirements increase
Solution Approach 1:
The neural network performs preliminary learning during a training phase where it processes historical voltage, current, and topology data to establish optimal reconfiguration strategies. Once trained, the network makes real-time topology predictions through simple forward propagation calculations rather than complex optimization algorithms, thereby achieving high energy efficiency through pre-computed knowledge rather than real-time computational complexity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system achieves an average test accuracy of 83% and an F1 score of 0.83, resulting in approximately 11% average power improvement by optimizing PV array topology under varying shading conditions, demonstrating scalability and cost-effectiveness.
Implementation Method 1
Photovoltaic (PV) energy systems have played a major part in meeting the renewable energy requirements over the past decade
Data Source
AI summary
A system reconfigures a photovoltaic array used in solar energy based on observed shading conditions to determine an optimal topology of the photovoltaic array to maximize power output. Specifically, the system is designed to reconfigure a photovoltaic array when the photovoltaic array is partially shaded. The system uses a neural network model to determine a topology that maximizes power output of the photovoltaic array based on irradiance data obtained from the photovoltaic array.