Sheet Folding Control with Machine-Learned Quality Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sheet folding apparatuses struggle to appropriately select and execute sheet folding operations due to diverse user conditions and numerous combinations of conditions, leading to inconsistent folding quality and inefficiencies.
Innovation Solution
A sheet folding apparatus equipped with a sheet folder and additional folding part, controlled by circuitry that utilizes machine learning to estimate folding evaluation values based on acquired sheet and folding information, allowing for precise determination of control content in the folding operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the number of combinations of user conditions is enormous, then the sheet folding apparatus can handle diverse user needs, but it becomes difficult to appropriately select and execute the folding operation content
Solution Approach 1:
The sheet folding apparatus performs self-diagnosis and self-adjustment by automatically acquiring sheet information, evaluating folding conditions, and determining optimal folding operation content without user intervention. The control unit automatically selects folding parameters based on acquired sheet data, enabling the system to serve itself in resolving the complexity of diverse user conditions.
Solution Approach 2:
The system dynamically changes folding operation parameters (such as fold-enhancing roller rotation times, folding speed, and pressure) based on acquired sheet information and evaluated folding conditions. This allows the apparatus to adapt to diverse user conditions by adjusting parameters in real-time rather than requiring manual selection from numerous combinations.
2Manufacturing precision
If fold-enhancing operations are performed multiple times on overlapped sheets, then folding quality is improved, but the processing time and resource consumption increase
Solution Approach 1:
The system performs fold-enhancing operations a predetermined number of times (partial action) rather than continuously or excessively, balancing folding quality with processing efficiency. The control unit determines the appropriate number of fold-enhancing operations based on sheet information and folding conditions, avoiding both insufficient and excessive processing.
Solution Approach 2:
The apparatus incorporates feedback mechanisms where the control unit monitors folding operation results and adjusts the number of fold-enhancing operations accordingly. By evaluating folding conditions and acquiring sheet information, the system determines the optimal number of repetitions needed to achieve quality folds without unnecessary resource consumption.
Data Source
AI summary
A sheet folding apparatus includes a sheet folder and circuitry. The sheet folder includes a sheet folding part, and an additional folding part to reinforce a folding portion in the sheet. The circuitry is to control the sheet folder to form and reinforce the folding portion in the sheet; acquire, before completion of a folding operation, first sheet information of the sheet and first folding information indicating a type of the folding operation on the sheet; run a trained model obtained through machine learning executed using second sheet information of a training sheet, second folding information indicating a type of the folding operation on the training sheet, and multiple training datasets including a folding evaluation value; estimate the folding evaluation value after the completion of the folding operation on the sheet; and determine the type of a control content in the folding operation, based on the folding evaluation value.