Information processing apparatus, information processing method, and program
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies for generating learned models to diagnose anomalies in devices are costly and inefficient, requiring unnecessary processing and potentially inaccurate anomaly determination.
Innovation Solution
An information processing apparatus that determines whether to generate a learned model based on sensor measurements, anomaly diagnosis results, control results, and device conditions, using a predetermined estimation method to estimate component values and detect anomalies, with thresholds for error evaluation to decide model generation and anomaly detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a learned model is generated for each device to perform anomaly diagnosis, then anomaly diagnosis accuracy is improved, but processing costs increase
Solution Approach 1:
The patent changes the parameter of model generation strategy from 'generate model for every device' to 'selectively generate model based on error thresholds'. The determining unit compares the absolute error between estimated and actual values against a threshold to decide whether model generation is necessary, thereby optimizing resource allocation while maintaining diagnosis accuracy where needed
Solution Approach 2:
Instead of universally generating learned models for all devices, the patent applies partial action by generating models only for devices where the error threshold condition is met. This selective approach reduces overall processing costs while maintaining adequate anomaly diagnosis capability for devices that require it
2Reliability
If a learned model is generated for each device, then device-specific anomaly detection capability is improved, but processing time increases
Solution Approach 1:
The patent introduces a time-efficient parameter change by using a deterministic error threshold comparison instead of time-consuming model training for all devices. The determining unit quickly evaluates whether |estimated value - actual value| >= threshold to decide on model generation, significantly reducing processing time while maintaining reliability for devices that need it
3Reliability
If learned models are generated for all devices, then anomaly detection coverage is improved, but system complexity increases
Solution Approach 1:
The patent simplifies system complexity by changing the deployment parameter from 'all devices have learned models' to 'only devices meeting error threshold criteria have learned models'. This parameter change maintains adequate anomaly detection coverage across the system while reducing the complexity burden of managing numerous learned models
Solution Approach 2:
The patent applies universality by using a common error threshold mechanism that can be applied across all devices regardless of type. The same determining logic and threshold comparison method works universally for deciding whether to generate learned models, simplifying system architecture while maintaining comprehensive anomaly detection capability where needed
Data Source
AI summary
Provide an information processing apparatus including a determining unit configured to determine whether to generate a learned model that estimates a value relating to a component included in a device, based on at least one of a measurement value measured by each sensor of the device, an anomaly diagnosis result of the device, a result of controlling the device, or a condition relating to the device.