Industrial Process Knowledge Acquisition via Weighted Logic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial processes face challenges in acquiring and representing unstructured domain knowledge, such as expert experience and mechanism knowledge, due to human cognitive bias and complexity, which hinders automation in optimization decision-making.
Innovation Solution
The method establishes a domain rule base using weighted first-order logic rules and probability soft logic to extract and refine knowledge from multi-source data, creating a semantic knowledge base that enhances the comprehensibility of optimization decision-making models by learning first-order logic rules through machine learning algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If unstructured domain knowledge (expert experience, mechanism knowledge) is directly used in industrial processes, then knowledge representation flexibility is improved, but automation in optimization decision-making deteriorates due to human cognitive bias and complexity
Solution Approach 1:
The patent introduces a knowledge base as an intermediary between unstructured domain knowledge and automated decision-making systems. Expert experience and mechanism knowledge are formalized into structured knowledge representations (ontologies, rules, logic programs) that serve as a bridge, enabling automation while preserving the flexibility of original knowledge forms. This intermediary structure eliminates human cognitive bias by providing a standardized, machine-processable knowledge representation.
Solution Approach 2:
The patent transforms knowledge from unstructured to structured forms by changing its representation parameters. Expert experience transitions from natural language descriptions to formal logical expressions. Mechanism knowledge changes from qualitative descriptions to quantitative mathematical models. This parameter transformation enables automated processing while maintaining the essential characteristics of original knowledge.
2Loss of information
If massive multi-source data is mined to establish semantic knowledge base, then comprehensibility of optimization decision-making model is improved, but device complexity worsens due to machine learning technology requirements
Solution Approach 1:
The patent extracts essential knowledge from massive multi-source data by applying machine learning algorithms to identify and separate meaningful patterns from noise. The semantic knowledge base extracts only the most relevant information needed for decision-making comprehensibility, rather than processing all raw data. This extraction approach improves model interpretability while managing computational complexity by focusing on key knowledge elements.
Solution Approach 2:
The patent creates a simplified copy of complex data relationships through semantic knowledge representation. Instead of working directly with massive raw data, the system generates a condensed semantic knowledge base that replicates essential decision-making logic in a more comprehensible form. This copying approach maintains the core insights while reducing complexity for human understanding and model interpretation.
3Loss of information
If first-order logic rules are learned from semantic knowledge base using machine learning, then knowledge interpretability is improved, but learning difficulty worsens due to ILP complexity
Solution Approach 1:
The patent segments the complex Inductive Logic Programming task into manageable components by breaking down first-order logic rule learning into smaller sub-tasks. The semantic knowledge base is divided into modular knowledge units that can be processed independently. This segmentation reduces the overall learning complexity while maintaining interpretability by creating a structured approach to rule acquisition from multi-source data.
Data Source
AI summary
Disclosed is an acquisition method for domain rule knowledge of an industrial process. The method comprises the steps of: establishing a domain rule base, establishing a semantic knowledge base, and combining the domain rule base and the semantic knowledge base so as to realize an augmented update of a domain rule knowledge base; describing the domain knowledge of the industrial process by using weighted first-order logic rules so as to form a training sample set of the first-order logic rules; performing a weight learning by applying probability soft logic and the training sample set of the first-order logic rules so as to realize weight to non-weighted rules; performing rule learning through a machine learning algorithm so as to obtain a first-order logic rule on a change in optimization decision-making semantic when multi-source data semantic information changes.