Container Handling Machine Control Using Self-Identification Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machine operating methods in processing plants for containers are inefficient due to changes in machine behavior over time, such as wear or altered mass flow, which lead to deviations from the desired operating point, and require high computing power for self-learning algorithms that do not account for current operating conditions.
Innovation Solution
A method that acquires input and output signals to determine a self-identification model representing the current operating point, allowing for automatic configuration and optimization of machine parameters and diagnosis, enabling real-time adaptation to changes and maintaining efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If machine parameters are set during commissioning or retooling to a specific container type and not changed during operation, then the initial setup is simple and quick, but the machine behavior deviates from the desired target behaviour over time due to wear or changed mass flow, leading to reduced efficiency
Solution Approach 1:
The patent implements dynamic adaptation of machine parameters through a self-identification model that continuously learns from input-output signals during operation. The system transitions from static parameters set during commissioning to dynamic parameters that automatically adjust to account for wear and changed mass flow, maintaining optimal operational efficiency without requiring manual reconfiguration
Solution Approach 2:
The system performs self-identification and self-optimization automatically during operation. The self-identification model continuously processes input and output signals to determine current machine behavior and operating points, enabling the machine to self-adjust parameters and maintain efficiency without external intervention or frequent maintenance stops
2Adaptability or versatility
If self-learning algorithms are used to optimize machine parameters, then the machine can adapt to changes in behavior, but high computing power is required which increases system complexity and cost
Solution Approach 1:
The patent replaces complex computational self-learning algorithms with a streamlined self-identification model based on input-output signal processing. This substitution reduces computational complexity and hardware requirements while maintaining the ability to adapt to machine behavior changes through continuous monitoring and parameter optimization
Solution Approach 2:
The system focuses on identifying and adjusting key operational parameters through the self-identification model rather than implementing comprehensive self-learning algorithms. By concentrating on critical parameter optimization based on input-output relationships, the system achieves adaptability with reduced computational burden
3Reliability
If machine parameters are not updated to reflect current operating conditions, then the system operates reliably according to initial specifications, but the current operating point deviates from the desired target behaviour, reducing overall performance
Solution Approach 1:
The patent implements continuous feedback through the self-identification model that monitors input and output signals during operation. This feedback mechanism detects deviations from desired operating points caused by wear or mass flow changes and triggers automatic parameter adjustments, maintaining both reliability and optimal productivity throughout the machine's operational life
Data Source
AI summary
A method for operating a machine in a processing plant for containers, in particular beverage containers, wherein the containers are processed and/or transported by the machine, wherein at least one input signal and at least one output signal of the machine are acquired during the processing and/or the transport, wherein a self-identification model of the machine, which model reproduces at least one current operating point of the machine, is determined based on the at least one input signal and the at least one output signal, wherein at least one machine parameter of the machine and/or of a downstream machine is automatically configured or optimised using the self-identification model, and/or wherein a diagnosis of the machine is automatically carried out using the self-identification model.