Condition Monitoring Model Reuse for Similar Machine Components
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing condition monitoring systems require significant manhours to create new models for machines with differences from those for which models have already been created, limiting the efficiency of model reuse and increasing introduction costs.
Innovation Solution
A condition monitoring system that collects operation data from machines, stores information about similar machines, and uses this data to create model candidates, allowing for the selection and display of suitable models for new machines, thereby reducing the need for manual model creation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a model is newly created for each component of a machine, then the model accurately reflects the specific machine's characteristics, but the manhours and costs for model creation increase significantly
Solution Approach 1:
The system performs preliminary actions by pre-creating models for representative machines and storing them in a database. When a new machine requires a model, the system searches for and reuses pre-created models from similar machines, adjusting them as needed rather than creating from scratch. This preliminary preparation of models for potential reuse directly reduces the manhours required for new model creation while maintaining accuracy through proper model selection and adaptation.
Solution Approach 2:
The system creates copies of existing models by retrieving models from the database that were created for similar machines. Instead of creating entirely new models, the system copies proven models from the database and adapts them to the specific machine context. This copying approach maintains model accuracy while dramatically reducing the time and resources needed for model creation, as the core model structure and parameters are reused from reliable sources.
2Adaptability or versatility
If a correction function is used to combine simulation models with measurement data, then some differences between machines can be absorbed, but it is still necessary to newly create models for each machine
Solution Approach 1:
The system creates a universal model database where models are stored that can be applied to multiple different machines. Instead of creating machine-specific models that cannot be reused, the system develops universal models that can serve multiple machines through the search and retrieval process. This multi-functionality allows a single model to be adapted to various machines, reducing the need for continuous new model creation while maintaining adaptability through the correction function and model selection process.
Solution Approach 2:
The system introduces an intermediary mechanism - the model database - that sits between model creation and model application. The database stores models that can be retrieved and adapted for different machines, serving as an intermediary repository. This intermediary allows the correction function to work more effectively by providing a base model from the database that can be adjusted with measurement data, rather than starting from a simulation model each time, thereby improving model creation efficiency.
3Productivity
If models are reused from similar machines, then introduction costs of the condition monitoring system are reduced, but the system must manage and store model information efficiently
Solution Approach 1:
The system replaces manual model management mechanisms with an automated information processing system. Instead of manually tracking and managing model versions, compatibility, and retrieval, the system uses automated database management, search algorithms, and information retrieval mechanisms. This substitution of automated electronic management for manual processes reduces the complexity of model management while enabling efficient model reuse, as the automated system handles the bookkeeping and matching tasks that would otherwise require significant human effort and complex procedural management.
Data Source
AI summary
A condition monitoring system which collects operation data from a machine and monitors a condition of the machine includes: a storage unit that stores information indicating components of a first machine for which a model for a sensor data analysis has been created and components of a second machine for which the model is newly created, information indicating a correspondence relationship between the components of the first machine and the components of the second machine, and information relating to the model; a model creation unit that creates model candidates of the second machine from the model similar to the second machine by using the information stored in the storage unit, and creating information relating to a model candidate selected via an input unit out of the model candidates as the model of the second machine; and a display unit that displays the model candidates.