Chemical State Estimation via Machine Learning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods, such as PTL 1 and PTL 2, face difficulties in estimating the state of a chemical substance during production, particularly in chemical reaction processes, as they struggle to accurately determine the internal state of production devices and product characteristics.

Innovation Solution

A state estimation system comprising an acquisition unit, an extraction unit, an estimation unit, and an output unit that acquires and processes time series data using machine learning to estimate the state of a chemical substance by training an estimation model on the relationship between feature amounts of production environment data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If production is stopped to confirm the state of the chemical substance, then the internal state can be directly confirmed, but production efficiency decreases and sufficient characteristics cannot be obtained when restarted

Engineering Contradiction:

Improvestate confirmation accuracyVSAvoidproduction efficiency

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces physical intervention (stopping production to take samples) with a field-based measurement system using sensors and machine learning models. The system continuously monitors production environment parameters and estimates chemical substance state in real-time, eliminating the need to halt production for state confirmation.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces an intermediary estimation system that indirectly determines the chemical substance state through production environment parameters. Instead of directly measuring the chemical substance, the system uses sensors to monitor environmental parameters and employs machine learning models to infer the substance state, enabling continuous monitoring without production interruption.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If traditional measurement methods are used to determine chemical substance state, then direct confirmation is possible, but continuous monitoring during production cannot be achieved

Engineering Contradiction:

Improvestate determination reliabilityVSAvoidmonitoring time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent implements continuous monitoring by maintaining the production process without interruption. Sensors continuously capture production environment parameters, and the machine learning model continuously estimates the chemical substance state, ensuring both production continuity and ongoing state monitoring.

Inventive Principle:

Principle #20Continuity of useful action

Solution Approach 2:

The patent replaces discrete, interruptive measurement methods with a continuous field-based estimation system. The machine learning model processes continuous sensor data streams to provide ongoing state estimates, eliminating the need to stop production for measurements.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Measurement precision

If production is interrupted for state confirmation, then the chemical substance state can be verified, but product characteristics become insufficient and production quality may deteriorate

Engineering Contradiction:

Improvestate verification accuracyVSAvoidproduct characteristic quality

Core Design Contradiction:

Measurement precisionVSManufacturing precision

Solution Approach 1:

The patent uses production environment parameters as intermediary indicators to infer the chemical substance state and product characteristics. By monitoring environmental parameters continuously and correlating them with substance state through machine learning, the system verifies state and assesses product quality without interrupting production, thereby maintaining manufacturing precision.

Inventive Principle:

Principle #24Intermediary (Mediator)

4Measurement precision

If direct measurement of chemical substance state is performed, then accurate state information is obtained, but production process complexity increases and real-time monitoring is difficult

Engineering Contradiction:

Improvestate measurement accuracyVSAvoidmeasurement system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces complex direct measurement systems with a simpler field-based estimation approach. Instead of inserting sensors into the chemical process or taking physical samples, the system uses existing production environment sensors and machine learning to estimate state, reducing system complexity while maintaining measurement capability.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS20240044764A1State estimation device, state estimation method, and program recording medium

Publication Date: 2024.02.08 NEC CORP

AI summary

A state estimation device includes an acquisition unit, an extraction unit, an estimation unit, and an output unit. The acquisition unit acquires first time series data pertaining to a generation environment of the targeted chemical substance. The extraction unit extracts a feature amount of the first time series data. The extraction unit extracts a feature amount of the first time series data. The estimation unit estimates, based on the feature amount of the first time series data, the state of the targeted chemical substance by using an estimation model trained, through machine learning, on the relationship between the state of the targeted chemical substance in the generation process and a feature amount of second time series data pertaining to the generation environment. The output unit outputs the estimated state.