Automatic Machine Predictive Maintenance Using Anomaly Matrices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing predictive maintenance systems for automatic machines are costly, inefficient, and fail to accurately predict malfunctions due to reliance on single-dimensional sensor data and high-frequency sampling limitations, often missing combined factor influences and requiring extensive operator intervention.
Innovation Solution
A method involving high-frequency sampling and recording of motorization metrics by local control units, with data processing units updating anomaly matrices using statistical features and synchronization signals to detect anomalies within a multidimensional tolerance horizon, allowing for efficient predictive maintenance without real-time data transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If traditional one-dimensional sensor systems are used to detect local features, then the system complexity is reduced, but the ability to detect combined factor influences and predict malfunctions is insufficient
Solution Approach 1:
The patent transitions from one-dimensional sensor detection to two-dimensional analysis by comparing current signal oscillations with reference signal oscillations across multiple dimensions (time, frequency, amplitude). This allows detection of combined factor influences that single sensors cannot capture, improving malfunction prediction while maintaining reasonable system complexity through signal processing rather than adding numerous sensors.
Solution Approach 2:
The reference signal serves multiple functions: it provides a baseline for comparison, enables detection of various malfunction types through oscillation analysis, and adapts to different operating conditions. This multi-functionality allows comprehensive monitoring without requiring separate specialized sensors for each parameter, resolving the contradiction between system simplicity and prediction accuracy.
2Measurement precision
If high-frequency sampling is performed to capture detailed signal variations, then the detection precision is improved, but the data management burden and transmission requirements increase significantly
Solution Approach 1:
The patent extracts only the essential features from high-frequency sampling data by comparing oscillations against a reference signal. Instead of managing all raw high-frequency data, the system extracts relevant deviation information that indicates malfunctions, thereby maintaining high detection precision while significantly reducing data management burden and transmission requirements.
Solution Approach 2:
The system performs high-frequency sampling (excessive action) but only processes and transmits the necessary comparison results rather than all sampled data. This partial processing approach captures sufficient detail for accurate detection while avoiding the full data management overhead, resolving the contradiction between precision and complexity.
3Device complexity
If local averaging of high-frequency values is performed to reduce data transmission, then the data management load is reduced, but the accuracy is lost as peaks indicating malfunctions are smoothed out
Solution Approach 1:
Instead of averaging data before transmission (which loses peak information), the patent inverts the approach by transmitting high-frequency sampling data and performing the comparison operation at the central unit. This allows preservation of peak information for accurate malfunction detection while still reducing transmission load by only sending relevant sampling intervals rather than continuous streams.
4Ease of operation
If threshold-based comparison with reference signals is used, then the system operation is simplified, but maintenance predictions become inefficient when oscillations indicate malfunction without exceeding thresholds
Solution Approach 1:
The patent moves beyond simple threshold comparison by analyzing multiple signal parameters simultaneously (oscillation frequency, amplitude patterns, phase relationships) when comparing current and reference signals. This multi-parameter analysis maintains operational simplicity through systematic comparison while improving maintenance prediction reliability by detecting malfunctions that don't exceed single-parameter thresholds but show characteristic oscillation patterns.
Data Source
AI summary
A method for the predictive maintenance of an automatic machine for manufacturing or packing consumer articles comprising the steps of: detecting and recording at least a sampling series relating to at least one motorization metric of at least one electric actuator, by means of at least one respective local control unit; transmitting the recorded sampling series to a data processing unit; defining at least one multidimensional tolerance horizon within an anomaly matrix having as dimensions at least two statistical features based on at least one sampling series detected and relative at least to the detected motorization metric; calculating the two statistical features in order to define the position of an actual condition within the anomaly matrix; determining, based on the position of the actual condition in the anomaly matrix and the multidimensional tolerance horizon, the imminence of necessary maintenance.