PV Module Control Using Machine Learning for Real-Time Fault Isolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current photovoltaic power generation systems face challenges in identifying and addressing issues with individual solar cell modules, leading to reduced power generation due to module trouble or abnormalities, and lack real-time data for efficient control and management.
Innovation Solution
A machine-learning-based control system that collects real-time voltage and electric power information from photovoltaic modules, uses machine learning to analyze and model control data, and adjusts module operations to maintain uniform power production, enabling real-time control and troubleshooting of connected devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional monitoring systems are used to measure only inverter output, then system complexity is reduced, but measurement precision and ability to detect individual module issues deteriorates
Solution Approach 1:
The system divides the photovoltaic power generation system into individual module-level monitoring units. Each photovoltaic module is equipped with its own monitoring device that collects voltage, current, and power data independently, enabling precise detection of individual module issues without requiring a completely complex centralized system.
Solution Approach 2:
A cloud-based machine learning platform serves as an intermediary that receives data from multiple simple module-level sensors, performs complex analysis and modeling, and returns control instructions. This intermediary handles the computational complexity while keeping individual module devices simple.
2Productivity
If real-time data collection from all photovoltaic modules is implemented, then productivity and power generation optimization improve, but device complexity and cost increase
Solution Approach 1:
The monitoring device at each module is designed to perform multiple functions: collecting electrical parameters (voltage, current, power), communicating with the cloud platform, and receiving control instructions. This multi-functionality reduces the need for separate dedicated devices for each function, optimizing power generation while managing system complexity.
Solution Approach 2:
The system implements a closed-loop feedback mechanism where the cloud-based machine learning platform analyzes real-time module data, generates control instructions based on learned patterns and models, and sends them back to individual modules. This feedback enables continuous optimization of power generation while maintaining manageable system complexity through automated decision-making.
3Manufacturing precision
If machine learning modeling is performed for various service functions, then control precision and power generation management improve, but device complexity and computational requirements increase
Solution Approach 1:
The cloud-based machine learning platform acts as an intermediary that performs complex modeling and analysis tasks. Instead of embedding complex machine learning models in each module device, the system centralizes computational complexity in the cloud, allowing precise control while keeping individual module devices simple and cost-effective.
4Reliability
If individual module monitoring and control is implemented, then reliability and power generation stability improve, but device complexity increases
Solution Approach 1:
The system monitors and controls each photovoltaic module independently through dedicated monitoring devices. This segmentation enables precise identification of faulty modules and allows selective control actions on individual modules or groups, improving power generation stability while managing complexity through modular architecture.
Solution Approach 2:
Real-time feedback from individual module monitoring enables the system to detect performance degradation, faults, or anomalies at the module level and respond with targeted control actions. This feedback mechanism improves reliability by enabling rapid response to issues while maintaining manageable complexity through automated decision-making algorithms.
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
This system ensures optimal electric power production by identifying and addressing module issues in real-time, balancing power output, and providing real-time analysis and management of photovoltaic power generation systems.
Implementation Method 1
Photovoltaic power generation is a power generation method of converting sunlight into DC current to produce electricity
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention are a system and a method for controlling solar photovoltaic power generation on the basis of machine learning, the system comprising: solar photovoltaic modules; node control units for switching off a connected solar photovoltaic module when measured current, voltage and power data do not satisfy control data; a gateway unit for storing measured data; a real-time control module for classifying, comparing and analyzing data and storing same, and transmitting a control command to the gateway unit; and machine learning for monitoring a device and data, learning on the basis of machine learning, and extracting functional data required for controlling solar photovoltaic power generation so as to provide control service data according to the result of performed modeling.