Adaptive Data Collection for Predictive Maintenance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current predictive maintenance techniques face inefficiencies in determining when, how often, and what data to collect from machinery, leading to excessive time and cost expenditures due to a lack of adaptive and state-driven data collection methods.
Innovation Solution
A method for automatically adjusting data collection parameters in a portable device based on the state of machines, allowing for adaptive and state-driven data collection, where the device configures data collection parameters, analyzes initial data, and prompts technicians to collect additional data based on alerts, optimizing data collection schedules and actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If data is collected as much as possible as frequently as possible, then the understanding of machine condition is improved, but time and money are wasted excessively on monitoring
Solution Approach 1:
The system dynamically adjusts data collection frequency and parameters based on real-time machine state assessment. When machines are operating normally, data collection frequency is reduced. When anomalies or critical conditions are detected, the system automatically increases monitoring intensity, optimizing both information quality and resource utilization
Solution Approach 2:
The patent changes data collection parameters (frequency, duration, type of measurements) based on machine state. Different operational states trigger different data collection regimes, allowing the system to collect sufficient information while avoiding excessive monitoring during stable periods
2Loss of information
If data is collected as much as possible as frequently as possible, then the understanding of machine condition is improved, but money is wasted excessively on monitoring
Solution Approach 1:
The system dynamically adjusts data collection based on machine state, reducing monitoring intensity during normal operation and increasing it when anomalies are detected. This dynamic approach reduces overall resource consumption while maintaining adequate machine condition understanding
Solution Approach 2:
The patent implements state-driven parameter changes where data collection frequency, duration, and type are adjusted according to machine operational state. This ensures adequate monitoring information is obtained while minimizing unnecessary data collection costs
3Stability of the object's composition
If fixed data collection schedules are used, then monitoring consistency is maintained, but adaptability to machine states is lost
Solution Approach 1:
The system transitions from static fixed schedules to dynamic state-driven schedules. The monitoring consistency is maintained through systematic assessment protocols, while adaptability is achieved by adjusting data collection parameters based on real-time machine state evaluation
Solution Approach 2:
The patent changes data collection parameters based on machine state assessment. Different operational states trigger different data collection regimens, allowing the system to adapt to varying machine conditions while maintaining consistent monitoring through structured decision protocols
Data Source
AI summary
Automatically adjusting collection parameters for machines on a route in a collection device, based on states of the machines. For each machine on the route, the machine state is read into the collection device. The machine is included or excluded based on the machine state. The collection device is configured with first collection parameters that are configured based on the machine state. Data is collected from the machine based on the first collection parameters. The data is analyzed using parameters that are configured based on the machine, to determine alerts. Based on the alerts, data is selectively immediately collected from the machine based on second collection parameters that are configured based on the machine state and the alerts. Also based on the alerts, the technician is selectively prompted with the collection device to take a predetermined action and collect data from the machine. The action and data are based on third collection parameters that are configured on the machine state and the alerts.