Damage Estimation Device Using Machine Learning for Work Machines

Solution Overview

Problem

Conventional techniques for predicting the lifespan of work machines, such as hydraulic excavators, are cumbersome and prone to inaccuracies due to the need for direct attachment of strain gauges, which can be damaged during operation and lead to unreliable data.

Innovation Solution

A damage estimation device and machine learning device that utilize machine learning models to estimate damage parameters related to the operation of work machines, allowing for accurate and easy estimation of lifespan without the need for direct strain gauge attachment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If strain gauges are directly attached to the surfaces of boom and arm portions to be measured, then strain amount can be detected, but the work becomes troublesome and the strain gauge may be damaged during work

Engineering Contradiction:

Improvestrain amount detection accuracyVSAvoidease of strain gauge attachment

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The patent introduces an intermediary approach by using acceleration sensors and gyro sensors instead of directly attaching strain gauges to the boom and arm surfaces. These sensors are mounted on the work machine's body and measure acceleration and angular velocity, which are then processed to calculate strain amounts indirectly. This mediator approach avoids the troublesome direct attachment while maintaining measurement capability.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical strain gauge system with an sensor-based measurement system. Instead of using mechanical strain gauges that require direct surface attachment, the system uses acceleration sensors and gyro sensors that measure dynamic parameters, which are then converted to strain information through calculation. This substitution eliminates the need for direct mechanical attachment to the boom and arm surfaces.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If strain gauges are directly attached to the surfaces of boom and arm portions to be measured, then strain amount can be detected, but the strain gauge may be damaged during work at the work site

Engineering Contradiction:

Improvestrain amount detection accuracyVSAvoidstrain gauge durability

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent introduces an intermediary approach by using acceleration sensors and gyro sensors instead of directly attaching strain gauges to the boom and arm surfaces. These sensors are mounted on the work machine's body and measure acceleration and angular velocity, which are then processed to calculate strain amounts indirectly. This mediator approach avoids the troublesome direct attachment while maintaining measurement capability.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical strain gauge system with an sensor-based measurement system. Instead of using mechanical strain gauges that require direct surface attachment, the system uses acceleration sensors and gyro sensors that measure dynamic parameters, which are then converted to strain information through calculation. This substitution eliminates the need for direct mechanical attachment to the boom and arm surfaces.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Measurement precision

If multiple strain gauges are attached to measure damage in predetermined portions, then damage amount can be calculated, but the attachment work becomes very troublesome

Engineering Contradiction:

Improvedamage amount calculation accuracyVSAvoidnumber of strain gauges and wiring

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent merges multiple measurement functions into a unified sensor system. Instead of using multiple separate strain gauges attached to different portions of the boom and arm, the system uses acceleration sensors and gyro sensors mounted on the work machine body to capture overall dynamic characteristics. The damage information for multiple portions is derived from this unified sensor data through integrated processing.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces an intermediary approach by using acceleration sensors and gyro sensors instead of directly attaching strain gauges to the boom and arm surfaces. These sensors are mounted on the work machine's body and measure acceleration and angular velocity, which are then processed to calculate strain amounts indirectly. This mediator approach avoids the troublesome direct attachment while maintaining measurement capability.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentEP3889362B1Damage estimation device and machine learning device

Publication Date: 2025.06.18 KOBELCO CONSTR MASCH CO LTD

Figure 1
Figure 2
Figure 3

AI summary

A damage estimation device includes: an operation parameter reception unit (211) that acquires an operation parameter related to an operation of a work machine (1); a damage estimation model storage unit (232) that stores a damage estimation model constructed by machine learning using training data with the operation parameter as an input value and a damage parameter related to damage in a predetermined portion of the work machine as an output value; and a damage parameter estimation unit (223) that estimates the damage parameter by inputting the operation parameter acquired by the operation parameter reception unit (211) to the damage estimation model stored in the damage estimation model storage unit (232).