Monocrystal Pulling Decision Control for Real-Time Abnormality Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the production of photovoltaic monocrystals by pulling-up, manual decision-making for addressing bract breakage and other abnormalities is time-consuming, inefficient, and prone to errors, leading to potential safety hazards and wastage.
Innovation Solution
An automatic decision-making method using deep learning to process and analyze data from pulling nodes, establishing models for optimal monocrystal temperature and pulling length, and making real-time decisions on process abnormalities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual inspection and decision-making are used for pulling process abnormalities, then human judgment can handle complex situations, but the timeliness and efficiency are low and the risk of missing inspection is high
Solution Approach 1:
The patent replaces the manual mechanical inspection system with an automated computer-based monitoring and decision-making system. The system automatically collects data from the pulling process, analyzes it using pre-established decision rules, and generates decisions without human intervention, thereby improving both reliability and efficiency simultaneously
Solution Approach 2:
The system enables the pulling process to monitor and evaluate itself automatically. By continuously collecting process data and comparing it against established criteria, the system performs self-diagnosis and self-decision-making regarding process abnormalities, eliminating the need for external manual inspection
2Measurement precision
If manual inspection is repeated until decision is made, then thorough assessment can be achieved, but time is wasted and production efficiency decreases
Solution Approach 1:
The patent establishes decision rules and evaluation criteria in advance before the pulling process begins. These pre-established rules enable the system to immediately assess abnormalities as they occur without requiring repeated manual inspections, thus achieving both accurate assessment and rapid decision-making
Solution Approach 2:
The system implements continuous real-time feedback by monitoring pulling process parameters and immediately comparing them against established criteria. This closed-loop feedback mechanism provides instant assessment of abnormalities, eliminating the time delay associated with repeated manual inspections while maintaining assessment accuracy
3Productivity
If pulling continues after bract breakage, then production can proceed, but diameter fluctuation and crystal instability increase
Solution Approach 1:
The system continuously monitors pulling process parameters including diameter measurements and provides real-time feedback on crystal stability. When abnormalities such as bract breakage are detected, the system immediately signals the need for process adjustment, enabling continuous production while maintaining diameter stability through prompt corrective action
4Productivity
If pulling length exceeds certain interval, then more production is achieved, but cost increases and monocrystal explosion risk increases due to thermal stress
Solution Approach 1:
The system establishes safe pulling length intervals and thermal stress thresholds in advance. By continuously monitoring pulling length and temperature parameters against these pre-set limits, the system proactively identifies when to pause or adjust the pulling process, preventing excessive thermal stress accumulation and potential explosions while optimizing production output
Solution Approach 2:
The system takes preliminary protective action by detecting early signs of thermal stress buildup and automatically triggering process adjustments before dangerous conditions develop. This preventive approach counteracts the accumulation of thermal stress that could lead to monocrystal explosion, allowing sustained high-volume production without compromising safety
Data Source
AI summary
The present application relates to automatic decision-making for pulling. Multi-dimensional data cleaning is performed and dimensional data warehouse is established by processing, filtering and converting basic source data of pulling nodes in a pulling process for monocrystal pulling-up into data sets easily identified and marked and establishing respective models based thereon. Basic source data of a current pulling nodes are obtained and converted into process parameters. The process parameters are compared with respective models in the dimensional data warehouse to obtain a first determination result. Data analysis is performed on the first determination result to determine whether an abnormality occurs in the current pulling process to obtain a second determination result. Decision is made automatically based on the second determination result.