Printing Machine Data Evaluation for Anomaly-Based Fault Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern printing machines generate vast amounts of data, making it difficult to identify the causes of machine malfunctions or production disruptions, especially with increasing complexity and automation.
Innovation Solution
A computer-implemented method that stores actual machine data during production, forms standard machine values using artificial intelligence for disruption-free production periods, and identifies anomalies by comparing actual data to standard values, facilitating the detection of potential errors and disruptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If comprehensive monitoring of printing machine parameters is implemented, then productivity and quality control are improved, but data volume and analysis complexity increase significantly
Solution Approach 1:
The patent segments the large volume of machine data into different categories (machine data, production data, interference signals) and processes them through hierarchical filtering. The system divides data analysis into multiple levels: raw data collection, preliminary filtering of normal values, anomaly detection, and detailed analysis only of suspicious data points. This segmentation reduces the complexity of comprehensive data analysis while maintaining reliable production monitoring.
Solution Approach 2:
The patent introduces an intermediary computing device that acts as a mediator between the printing machine's sensors and the operators. This intermediary automatically processes, filters, and evaluates the data, presenting only relevant anomalies and insights to operators. The intermediary system handles the complex data analysis tasks, freeing operators from direct engagement with raw data while maintaining production reliability.
2Measurement precision
If the number of sensors and monitoring parameters is increased, then detection precision is improved, but the difficulty of identifying error causes increases
Solution Approach 1:
The patent implements feedback mechanisms where the system continuously monitors machine data and provides automatic feedback when anomalies are detected. The computing device compares real-time data against learned patterns and provides feedback signals when deviations occur, highlighting specific parameters and time points that require attention. This feedback approach helps operators quickly identify error causes among numerous sensors by directing attention to the most relevant anomalies.
Solution Approach 2:
The system performs preliminary actions by pre-processing and filtering data before presentation to operators. It automatically identifies and flags anomalies, pre-evaluates data patterns, and prepares summarized reports that highlight potential error causes. This preliminary analysis reduces the effort required by operators to identify error causes among the large number of monitoring parameters.
3Manufacturing precision
If more components and control processes are integrated, then manufacturing precision is improved, but the number of potential errors increases
Solution Approach 1:
The patent enables the printing machine to monitor and evaluate its own operational data through an integrated computing device. The system performs self-diagnosis by automatically analyzing machine data, detecting anomalies, and identifying potential issues before they affect printing quality. This self-service capability allows the complex multi-component system to maintain reliability by autonomously detecting and reporting problems, reducing the impact of increased component count on system stability.
Data Source
AI summary
The invention relates to a computer-implemented method for analyzing data of a printing machine during or after a production, wherein a number of printed products with a specified target printing technology value are produced by the printing machine during at least one print job, wherein actual machine data and/or actual printing technology values and/or production data and/or machine interference signals and/or control signals and/or an operating signal are detected and/or stored over a period of time by a computing device across a portion of the at least one print job. It is the object of the invention to evaluate the large data volumes and the very large number of measuring values and signals by means of a computer-implemented method, in order to find causes for production disruptions and to be able to increase the productivity of a printing machine.