Injection Molding Abnormality Detector Using Statistical Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing abnormality detection systems in injection molding machines require manual threshold value setting by operators, which is burdensome, and are prone to erroneous detection due to variations in physical quantity distributions that do not follow a normal distribution.
Innovation Solution
An automated system that calculates threshold values based on distribution index values such as kurtosis, skewness, and high-order moments, adjusting the threshold values dynamically to account for non-normal distributions, thereby reducing operator burden and improving detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual threshold value setting is used for abnormality detection, then the detection system is simple to implement, but the operator burden increases and detection accuracy decreases for non-normal distributions
Solution Approach 1:
The system automatically calculates threshold values using the calculated standard deviation and distribution type, eliminating the need for manual operator input. The controller autonomously adjusts threshold values based on real-time statistical analysis of motor current data, making the system self-configuring and adaptive to different distribution types without operator intervention.
Solution Approach 2:
The system dynamically changes the threshold value parameter based on the calculated standard deviation and identified distribution type. Instead of using a fixed manual threshold, the threshold is automatically adjusted according to statistical parameters derived from actual operating conditions, improving detection accuracy for both normal and non-normal distributions.
2Measurement precision
If fixed threshold values are used for abnormality detection, then the system is simple to operate, but detection accuracy deteriorates when physical quantity distributions are non-normal
Solution Approach 1:
The system dynamically changes the threshold value parameter based on the calculated standard deviation and identified distribution type. Instead of using a fixed manual threshold, the threshold is automatically adjusted according to statistical parameters derived from actual operating conditions, improving detection accuracy for both normal and non-normal distributions.
Solution Approach 2:
The system continuously monitors motor current, calculates statistical parameters including standard deviation and distribution type, and uses this feedback to automatically adjust threshold values. This closed-loop feedback mechanism ensures the threshold adapts to changing operating conditions and distribution characteristics, maintaining high detection accuracy.
3Ease of operation
If automated threshold calculation is implemented, then operator burden is reduced, but the system complexity increases
Solution Approach 1:
The system automatically calculates threshold values using the calculated standard deviation and distribution type, eliminating the need for manual operator input. The controller autonomously adjusts threshold values based on real-time statistical analysis of motor current data, making the system self-configuring and adaptive to different distribution types without operator intervention.
Solution Approach 2:
The patent replaces manual mechanical threshold setting with automated electronic calculation. The controller uses computational algorithms to calculate standard deviation, identify distribution types, and determine threshold values, substituting the mechanical/manual process with an electronic information-processing system that automatically adapts to varying operating conditions.
4Adaptability or versatility
If standard deviation-based threshold adjustment is used, then detection adapts to distribution variations, but calculation complexity increases
Solution Approach 1:
The system dynamically changes the threshold value parameter based on the calculated standard deviation and identified distribution type. Instead of using a fixed manual threshold, the threshold is automatically adjusted according to statistical parameters derived from actual operating conditions, improving detection accuracy for both normal and non-normal distributions.
Solution Approach 2:
The system continuously monitors motor current, calculates statistical parameters including standard deviation and distribution type, and uses this feedback to automatically adjust threshold values. This closed-loop feedback mechanism ensures the threshold adapts to changing operating conditions and distribution characteristics, maintaining high detection accuracy.
Data Source
AI summary
A movable unit in an injection molding machine is started to operate and the current physical quantity and the current time (or current position of the movable unit) are detected. The current physical quantity is stored as a reference physical quantity in association with the elapsed time of operation of the movable unit (or position of the movable unit). A deviation of the current physical quantity from the reference physical quantity is calculated and stored, a distribution index value is then calculated from the physical quantity deviations in the first to n-th cycles, and a threshold value is determined from the distribution index value. When the deviation exceeds the threshold value, alarm processing is performed.