Packaging Machine Fault Prediction Using Motion Error Statistics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current condition monitoring and fault prediction methods for packaging machines with independently moving objects are complex, inaccurate, and time-consuming, particularly due to assumptions about constant rotation speeds, leading to sub-optimal fault prediction and increased maintenance costs.
Innovation Solution
A method and system that register position error data from sensors, calculate central tendency and distribution shape, and determine dispersion trends over time to predict faults in packaging machines with independently movable objects, eliminating the need for complex frequency analysis and assumptions about constant speeds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If frequency analysis is used to characterize bearing faults, then fault detection capability is improved, but implementation complexity increases and accuracy is limited by assumptions of constant rotation speed
Solution Approach 1:
The patent extracts the essential fault characteristics from vibration signals by focusing on statistical parameters (mean, standard deviation, skewness, kurtosis) rather than performing complete frequency analysis. This extraction approach captures the most relevant information while avoiding the complexity of full spectral analysis and its underlying assumptions about constant rotation speed.
Solution Approach 2:
The patent transforms the vibration signal characterization from frequency domain parameters to time domain statistical parameters. By changing the parameter space from frequency spectra to statistical moments (mean, std, skewness, kurtosis), the method achieves fault detection without requiring constant rotation speed assumptions, thereby reducing implementation complexity while maintaining accuracy.
2Reliability
If empirical evaluation of vibration levels is performed, then fault prediction is attempted, but the activity is error-prone and may lead to significant underestimation or overestimation of remaining lifetime
Solution Approach 1:
The patent implements a feedback mechanism by continuously monitoring statistical parameters of vibration signals and comparing them against threshold values or historical data. This feedback loop enables dynamic adjustment of maintenance decisions based on actual machine condition, reducing both underestimation and overestimation of remaining lifetime while improving prediction reliability.
Solution Approach 2:
The patent performs preliminary analysis by calculating multiple statistical parameters (mean, standard deviation, skewness, kurtosis) that collectively characterize the vibration signal before making fault predictions. This preliminary multi-parameter assessment provides a more robust basis for prediction than single-parameter empirical evaluation, reducing errors in remaining lifetime estimation.
3Adaptability or versatility
If solutions are employed to accommodate variable speeds in fault prediction, then applicability to independently movable objects is improved, but implementation complexity and limitations increase
Solution Approach 1:
The patent replaces mechanical assumptions about constant rotation speed with a statistical analysis approach that is inherently adaptable to variable speeds. By substituting the mechanical model (which requires constant speed) with a statistical model (analyzing vibration signal characteristics), the system naturally accommodates variable speed profiles without additional complexity.
Solution Approach 2:
The patent creates a universal fault prediction method that works across different operating conditions including variable speeds, different machine components, and various types of packaging machines. The statistical parameter approach serves multiple functions: it detects faults, characterizes their severity, and adapts to different speed profiles, eliminating the need for separate solutions for each scenario.
Data Source
Figure 1
Figure 2a~2d
Figure 3
AI summary
A method of fault prediction in a packaging machine is disclosed. The method comprises registering data values associated with the motion of independently movable objects along a track in the packaging machine; determining a distribution of the data values; calculating a measure of central tendency of the data values in the distribution; calculating a quantified measure of a shape of the distribution; associating the measure of central tendency with said quantified measure of the shape as a coupled set of condition parameters; determining a degree of dispersion of a plurality of coupled sets of condition parameters associated with a plurality of motion cycles of the independently movable objects; and comparing the degree of dispersion with a dispersion threshold value, or determining a trend of the degree of dispersion over time, for said fault prediction.