Machine Condition Monitoring With Adaptive Performance Baselines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complex machines pose challenges in monitoring the state of health of critical components, particularly due to difficulty in accessing highly loaded parts and varying effects of loadings based on installation location, configuration, and ambient conditions, which complicates the determination of meaningful loading indicators and maintenance strategies.
Innovation Solution
A method and monitoring device that continuously capture current operating data, simulate the operating behavior of machine components, derive performance values, and adapt performance reference values dynamically to changes in operating conditions, surroundings, and component alterations, enabling accurate state-of-health assessment and prediction of future performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensors are installed on highly loaded rotor components to directly measure loading, then measurement precision is improved, but device complexity and installation expenditure increase significantly
Solution Approach 1:
The patent uses easily accessible components (such as motors, couplings, or gearboxes) as intermediary elements to measure loading indirectly. Instead of placing sensors directly on the difficult-to-access rotor components, the system measures loading at these intermediary locations where sensors can be easily installed, thereby avoiding complex installation while still obtaining meaningful loading data for health monitoring
Solution Approach 2:
The patent replaces direct mechanical sensor installation on rotating components with an indirect measurement approach using computational models. Physical simulation models and machine learning algorithms substitute for direct mechanical measurement, allowing loading determination without the need for complex sensor installations on highly loaded rotor parts
2Adaptability or versatility
If performance reference values are kept static, then system complexity is reduced, but adaptability to changing operating conditions deteriorates
Solution Approach 1:
The patent implements dynamic performance reference values that automatically adapt to changing operating conditions. Instead of using fixed static thresholds, the system continuously updates reference values based on actual operating data, load patterns, and environmental conditions, enabling the monitoring system to remain accurate across varying operational scenarios without manual intervention
Solution Approach 2:
The patent employs feedback mechanisms where measured performance data is continuously compared against reference values, and the reference values are subsequently updated based on this feedback. This closed-loop approach allows the system to learn from actual operating conditions and automatically adjust its baseline expectations, improving adaptability while maintaining systematic control over the reference value updates
Data Source
AI summary
In order to monitor the condition of a machine, current operating data relating to the machine are continuously captured and are taken as a basis for continuously simulating a current operating behavior of a machine component by a concurrent simulation module. Furthermore, performance values quantifying a current performance of the machine component are continuously derived from the simulated operating behavior and are stored over time. In addition, a performance normal value is regularly determined on the basis of a multiplicity of performance values which were derived earlier. The operating data and/or the performance values are monitored in order to determine whether a predefined first change pattern occurs. Detection of the first change pattern then causes a performance reference value to be updated with the current performance normal value. In addition, the respective current performance values are continuously compared with the current performance reference value in each case.