Component Life Prediction Using Operation-State Cumulative Time
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods fail to accurately predict the end of life for components in apparatuses due to varying operating states and use tendencies, leading to unpredictable component failures and reduced productivity.
Innovation Solution
A method and system that determine the operation state of components, predict remaining useful life, and cumulative time in that state, and ultimately forecast the component's life end based on these predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If remaining usable time is predicted based on average service life, then component replacement can be planned, but prediction accuracy deteriorates due to varying operating states and use tendencies
Solution Approach 1:
The patent segments the prediction process into multiple independent prediction units: one for remaining useful life and another for cumulative time in operation states that reduce service life. Each unit processes specific input data and produces dedicated output, which are then combined to achieve accurate overall prediction without requiring a single complex system
Solution Approach 2:
The patent adds a new dimension to the prediction model by separately predicting cumulative time in specific operation states (printing, heating, cleaning) that reduce service life. This dimensional expansion allows the system to account for varying operating conditions without increasing overall system complexity, as each dimension operates independently
2Measurement precision
If service life data is collected in advance, then prediction accuracy improves, but data collection difficulty increases for components with unknown service life
Solution Approach 1:
The patent performs preliminary actions by predicting remaining useful life and cumulative time in operation states before actual component failure occurs. This allows the system to build prediction models using operational data collected during normal service, eliminating the need for advance collection of service life data and enabling accurate predictions even for components with unknown service life
Solution Approach 2:
The patent implements feedback mechanisms where actual component service life data is continuously collected and used to refine and update prediction models. This feedback loop enables the system to improve prediction accuracy over time without requiring advance data collection, as the model learns from actual operational patterns during service
3Adaptability or versatility
If operation state varies between components, then prediction flexibility improves, but prediction reliability deteriorates due to unknown service life for some components
Solution Approach 1:
The patent applies local quality by tailoring prediction parameters to specific operation states (printing, heating, cleaning) that have different impacts on component service life. Each operation state has its own prediction model with state-specific parameters, allowing the system to adapt to varying operational conditions while maintaining reliable predictions through state-specific accuracy rather than relying on general unknown service life data
Data Source
AI summary
A component life prediction method that makes it possible to enhance, when predicting a time at which a component provided in an apparatus reaches its life end, the accuracy of this prediction. A method of predicting the life end of a component provided in an apparatus. An operation state of the apparatus is determined, which can reduce the service life of the component. A remaining useful life is predicted in which the component can be used until the component reaches the life end. Cumulative time is predicted which is determined by accumulating time in the operation state. A time at which the component reaches the life end is predicted based on a result of predicting the remaining useful life and a result of predicting the cumulative time.