State Estimation Apparatus Learning Common Deterioration Data

The patent combines deterioration data from multiple similar structural objects to estimate the deterioration state of a target object. When the target object has insufficient historical data, the system merges data from other similar objects (e.g., same type, age, material, or usage conditions) to supplement the estimation, thereby improving accuracy without requiring extensive data from the target object alone.

The patent introduces an intermediary learning model that processes and integrates data from multiple sources. This learning model acts as a mediator between the available deterioration data and the estimation result, enabling the system to effectively utilize data from similar objects while accounting for differences between individual structures through learned patterns and relationships.

The patent segments the data processing task into distinct modules: data acquisition from multiple objects, data preprocessing and cleaning, feature extraction, learning model training, and estimation output. This segmentation allows each module to handle specific aspects of the complex task independently, making the overall system more manageable and maintainable while achieving high accuracy through comprehensive data utilization.

The patent develops a universal learning model that can be applied across different types of structural objects (bridges, buildings, etc.) by learning common deterioration patterns. This multi-functional approach allows the same system architecture to handle various object types, reducing overall system complexity compared to having separate specialized systems for each object type.