Research And Development Of AR-Based Turbine Maintenance Assistance System

As a large and complex power generation equipment,the turbine overhaul process is very tedious.The traditional turbine overhaul method is mainly based on single outsourced overhaul,which is inefficient and ineffective,and there are often problems such as chaotic personnel management and incomplete information in the overhaul process,all of which will lead to a decline in the quality of turbine overhaul and seriously affect the safety of power plant unit operation.In this paper,we design and develop an AR-based turbine overhaul auxiliary system for the above problem.On the one hand,the system is based on Holo Lens2 glasses and combines augmented reality technology and knowledge representation technology to superimpose turbine overhaul knowledge and operational processes on top of real scenes in the form of 3D virtual information to provide overhaul auxiliary functions for overhaul personnel.On the other hand,the system can both digitally store paper-based information and unify the on-site overhaul personnel,thus providing a solution to on-site management chaos.It is conducive to creating a new model of turbine overhaul,making the turbine overhaul process more standardized and standardized,and playing a role in the construction of digital smart power plants.The research work in this paper focuses on the following points.(1)Study the concept of ontology and common methods of ontology modeling,conduct knowledge analysis of turbine overhaul process package,then design the whole knowledge base structure according to the characteristics and structure of turbine overhaul knowledge,after the structure design is completed,organize the knowledge content of each part of the knowledge base and the attribute relationship and data relationship between each part,and finally,through protégé ontology modeling software to establish the knowledge base of turbine overhaul process.After the knowledge base is constructed,the knowledge base is stored in an OWL format file.In order to make the knowledge information in the OWL document better applied by the system,this paper designs and develops the process flow parser and process flow node division function according to the structure characteristics of the OWL document,and completes the conversion from ontology knowledge in OWL format to object-oriented knowledge.The OWL document is parsed into a configuration file that can be called by the system in this paper as the underlying data support.(2)Design and develop an AR-based turbine maintenance assistance system.In terms of Holo Lens2 terminal,through the analysis of Holo Lens2 functions and structure,the system network access method was determined and the overall architecture of the system was designed;through the study of Holo Lens2 interaction and 3D registration tracking technology,the human-computer interaction method within the system platform was designed in conjunction with the system requirements;in terms of model building,3ds Max was used to build In terms of model building,the 3D model of the device and the skeletal animation file are built using 3ds Max,and the model is appropriately optimized in order to solve the problem of large space occupation of the model and slow loading of the system.Finally,in terms of system implementation,Unity engine and VS2017 are selected as the system’s integrated development environment(IDE),and C# scripting language is used as the system’s development language to complete the development of the whole system.