Load Characteristic Estimation for Driving Machines

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing load characteristic estimation methods for driving machines, such as those used in machine tools and robots, face challenges in accurately estimating friction and other load characteristics under various operation conditions, particularly during high acceleration and low-speed operations, leading to increased estimation errors and instability.

Innovation Solution

A load characteristic estimating apparatus that includes an action-command generating unit, a driving-force-command generating unit, a sign determining unit, a load-driving-force estimating unit, a normal-rotation-load calculating unit, and a reverse-rotation-load calculating unit, which generate and process signals to estimate load driving forces and friction characteristics by averaging torque signals during normal and reverse rotation states, thereby stabilizing the estimation process across different operation conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If disturbance observer method is used to sequentially estimate friction, then real-time estimation capability is improved, but estimation error increases during high acceleration and low-speed operations

Engineering Contradiction:

Improvereal-time estimation capabilityVSAvoidestimation accuracy

Core Design Contradiction:

SpeedVSMeasurement precision

Solution Approach 1:

The patent segments the operation range into multiple regions (high acceleration region, low-speed region, normal operation region) and applies different estimation methods or weightings for each region. This allows the system to handle each operational condition optimally, reducing estimation errors that occur when a single method is used across all conditions.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent dynamically changes estimation parameters such as filtering coefficients, weighting factors, or model parameters based on the current operational state (acceleration level, speed range). By adapting parameters to match the operational conditions, the system maintains high estimation accuracy across varying operating conditions while preserving real-time capability.

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If speed trapezoidal wave command method is used to estimate load characteristics, then estimation accuracy is improved, but productivity decreases due to required machine stops

Engineering Contradiction:

Improveestimation accuracyVSAvoidmachine operational continuity

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent enables continuous estimation of load characteristics during normal machine operation without requiring stops or special test sequences. By processing torque and speed data from regular operation, the system maintains both high estimation accuracy and continuous productivity, eliminating the need to interrupt manufacturing processes for characterization measurements.

Inventive Principle:

Principle #20Continuity of useful action

Solution Approach 2:

The system uses the machine's own operational data (torque commands, speed measurements from normal operation) to perform self-characterization and friction estimation. This self-service approach eliminates the need for external test equipment or dedicated calibration procedures, maintaining both accuracy and productivity.

Inventive Principle:

Principle #25Self-service

3Speed

If disturbance observer is used with inertia error, then real-time estimation is maintained, but harmful factors increase due to inertia error disturbance

Engineering Contradiction:

Improvereal-time estimation capabilityVSAvoidinertia error disturbance

Core Design Contradiction:

SpeedVSObject-generated harmful factors

Solution Approach 1:

The patent acknowledges the presence of inertia errors and uses them to identify and compensate for other disturbances. By modeling the inertia error disturbance and incorporating it into the estimation framework, the system converts this harmful factor into useful information that improves overall estimation robustness and accuracy.

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

Solution Approach 2:

The patent prepares for inertia errors by incorporating compensation mechanisms in advance. The estimation algorithm includes terms that account for potential inertia mismatches, cushioning against their harmful effects before they degrade estimation accuracy. This proactive approach maintains real-time capability while mitigating the impact of inertia errors.

Inventive Principle:

Principle #11Beforehand cushioning (Prior cushioning)

Data Source

PatentUS9124212B2Load characteristic estimating apparatus for driving machine

Publication Date: 2015.09.01 MITSUBISHI ELECTRIC CORP

AI summary

A load characteristic estimating apparatus for a driving machine includes an action-command generating unit for generating an action command for a position and speed of a driving machine, a driving-force-command generating unit for generating a driving force command to cause an action of the driving machine to follow the action command, a driving unit for generating driving force corresponding to the driving force command and drive the driving machine, a sign determining unit for determining, based on driving speed of the driving machine, the state of the driving machine, a load-driving-force estimating unit for calculating, based on the driving force command, a load driving force signal, and a normal reverse-rotation-average calculating unit and a reverse-rotation-average calculating unit, configured to calculate a sequential average of the load driving force signal, respectively when the determination result is a normal and reverse rotation action.