Evaluation Model Generation Using Raw Material Similarity

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing evaluation models for equipment like oil refineries need to be frequently updated due to variations in crude oil properties, leading to increased processing loads and time losses, as each change in oil region or production time requires a new model generation.

Innovation Solution

An evaluation model generating apparatus that determines whether to generate a new evaluation model based on crude oil properties, using machine learning to create a model only when necessary, thereby reducing processing loads and time losses by reusing models for similar properties.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If evaluation models are frequently updated to adapt to variations in crude oil properties, then adaptability is improved, but processing load and time loss increase

Engineering Contradiction:

Improveadaptability to crude oil property variationsVSAvoidtime loss from frequent model generation

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent changes the parameter of model generation frequency from frequent updates to conditional updates based on property similarity. By introducing a similarity threshold parameter (e.g., 80% similarity), the system determines whether to generate a new model or reuse an existing one, thereby reducing unnecessary model generation while maintaining adaptability to significant property variations.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent replaces the mechanical approach of frequent model generation with a computational similarity assessment mechanism. Instead of automatically generating new models for every crude oil batch, the system computes property similarity metrics and uses this information to intelligently decide whether model regeneration is necessary, substituting brute-force model generation with smart decision-making.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If evaluation models are frequently updated to adapt to variations in crude oil properties, then reliability is improved, but processing load increases

Engineering Contradiction:

Improvereliability of evaluation accuracyVSAvoidprocessing load for model generation

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent introduces a similarity threshold parameter to control when model generation occurs. By setting this parameter (e.g., 80% similarity threshold), the system balances reliability and productivity: models are regenerated only when property variations exceed the threshold, ensuring adequate reliability while avoiding unnecessary processing load from frequent regenerations.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent replaces the computationally intensive mechanical process of frequent model generation with a lighter computational approach: calculating property similarity metrics. This substitution maintains evaluation reliability by ensuring models are updated when necessary, while dramatically reducing average processing load by avoiding redundant model generation for similar crude oils.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Adaptability or versatility

If new evaluation models are generated for each change in oil region or production time, then adaptability is improved, but device complexity increases

Engineering Contradiction:

Improveadaptability to different oil regions and timesVSAvoidcomplexity of model management system

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent simplifies the model management complexity by introducing a similarity threshold parameter that automates the decision-making process. Instead of complex manual judgments about when to generate new models, the system uses this parameter to automatically determine whether property variations warrant model regeneration, reducing management complexity while maintaining adaptability.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent replaces complex manual model management procedures with an automated similarity assessment system. The computational mechanism automatically compares crude oil properties, calculates similarity metrics, and determines model generation needs, eliminating the need for complex human judgment and simplifying the overall system architecture.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS20240012967A1Evaluation model generating apparatus, evaluation model generating method, and non-transitory computer readable medium

Publication Date: 2024.01.11 YOKOGAWA ELECTRIC CORP

AI summary

An evaluation model generating apparatus, including: a property data obtaining unit which obtains property data representing a property of a raw material to be used in equipment that manufactures a product from the raw material; a determining unit which determines, based on the property data, whether to generate an evaluation model that outputs an indicator obtained by evaluating a state of the equipment; and an evaluation model generating unit which generates, according to a result of the determining, the evaluation model by machine learning, is provided. In the evaluation model generating apparatus, the determining unit may determine, if it is judged that a target raw material is not similar in property to a raw material that has been used in the past based on the property data, to generate the evaluation model.