Model Development Environment for Industrial Asset Health Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Developing models to assess the health and maintenance needs of industrial assets in power systems is complex due to the large number of assets and diverse data sources, often requiring expertise from multiple disciplines and manual data handling, leading to a slow and inefficient process.
Innovation Solution
A model development environment that allows users to define output parameters and associate variables with data without specifying data locations, enabling the creation of models that can be packaged and deployed to automatically map variables and output parameters to relevant data stores, facilitating automated data processing and storage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If models are developed to assess industrial asset conditions, then asset health assessment capability is improved, but model development complexity increases due to diverse data sources and multi-disciplinary requirements
Solution Approach 1:
The patent introduces a model development environment as an intermediary system that mediates between users and the complex model development process. This environment provides pre-configured templates, automated data source discovery, and standardized workflows that simplify asset health assessment model creation while maintaining assessment capability
Solution Approach 2:
The model development environment is designed as a universal platform that can handle diverse data sources (sensor data, maintenance records, asset information) and multiple asset types through standardized interfaces and configurable templates, reducing the need for separate development processes for each asset type
2Measurement precision
If manual data handling and multi-disciplinary collaboration are used, then model development accuracy is improved, but development time increases
Solution Approach 1:
The system performs preliminary actions by pre-configuring model templates with standard data requirements, pre-establishing data source connections, and pre-defining computation methodologies. This allows users to deploy models quickly while maintaining accuracy through proven templates that have been pre-validating against domain expertise
Solution Approach 2:
The model development environment enables self-service model development by providing automated features including data source automatic discovery, variable-to-datastore mapping, and validation checks that ensure model accuracy without requiring manual intervention from multiple disciplines
3Measurement precision
If users specify data locations manually, then data mapping precision is improved, but ease of model deployment deteriorates
Solution Approach 1:
The system automatically performs data mapping by having the model development environment self-configure variable-to-datastore mappings based on data source metadata and model requirements. This eliminates manual data location specification while maintaining mapping precision through automated validation and discovery mechanisms
Data Source
AI summary
Among other things, one or more techniques and/or systems are provided for facilitating development, by a user, of a model describing a condition of an industrial asset. An output parameter of the model can be defined within a model development environment. The output parameter describes a condition of the industrial asset and can be defined using one or more variables. During deployment of the model within a deployment environment, respective variables of the model can be mapped to a data store where data associated with the variable is located. In this way, during development, a user can specify variables to be used within the model without specifying where to fetch and/or push data associated with the variable, for example. Further, the model can be executed within the deployment environment to assess the condition for a plurality of industrial assets.