Abstract:Centrifugal compressors are widely used turbomachines employed across various fields as components of systems, such as being part of a turbocharger in an internal combustion engine or a refrigeration cycle. As a result of system analysis, the compressor parameters are determined with 0D or 1D solvers. Simultaneously, the compressor performance, as well as other components, influence the results of system analysis. The system analysis starts with compressor performance assumptions that could be taken from a prototype or based on the engineer’s experience and prior domain knowledge. If the new design deviates significantly from the prototype, its performance assumptions diverge from the actual compressor performance, necessitating numerous iterations to align system and component parameters with overall system requirements and constraints. Therefore, the design and development projects of modern centrifugal compressors are lengthy and very expensive. The research described in this paper aims to develop highly accurate single-stage centrifugal compressor performance models coupled with flow path geometry generation models leveraging artificial intelligence (AI) technology. These models offer the potential to circumvent expensive design-build-test iterations, thereby substantially reducing the time and cost of developing modern centrifugal compressors. In this paper, the various configurations of centrifugal compressors are considered, along with a description of the configurations of interest. The selection of input and output variables that provide sufficient information about compressor design with the respective ranges is justified. There is a detailed description of automated workflows for centrifugal compressor design and performance data generation for neural network training. The approaches for data preprocessing that enable high-accuracy predictions are provided, and the peculiarities of applying the AI technology developed by the authors for centrifugal compressors are discussed. An analysis of the trained model accuracy as well as the technique for quantitative assessment of prediction reliability is provided. The utilization of the centrifugal compressor AI model integrated into a gas turbine engine simulation environment is demonstrated.

Performance Prediction of Turbocharger Using Neural Network

Prediction of performance parameters of a hermetic reciprocating compressor applying an artificial neural network

CFD and Neural Network-Based Predicting for Oil Delivery Performance of Oil Pump in Engines

Centrifugal Compressor Performance and Flow Path Generation With Artificial Intelligence

Towards Predictive Maintenance of a Heavy-Duty Gas Turbine A New Hybrid Intelligent Methodology for Performance Simulation

Improved Performance Prediction Model for Turbocharger Compressor

Dual-convolutional neural network based aerodynamic prediction and multi-objective optimization of a compact turbine rotor

Evaluation of Surge Prediction Capabilities of Body-Force Model on a High-Speed Multi-Stage Axial-Radial Compressor

An Air Filtration Performance Prediction Method of Gas Turbine Based on Dynamic Neural Network

A Surrogate Model for a CAES Radial Inflow Turbine with Test Data-Based MLP Neural Network Algorithm

Thrust Performance Prediction of Scramjet Nozzle Based on Depth Neural Network

A One-Dimensional Unsteady Performance Model for Turbocharger Turbines

Aerodynamic Prediction on the Off-Design Performance of a S-CO2 Turbine Based on Deep Learning

Remaining Useful Life Prediction of an Aircraft Turbofan Engine Using Deep Layer Recurrent Neural Networks

Energy performance prediction of the centrifugal pumps by using a hybrid neural network

Prediction of Remaining Useful Life Using Fused Deep Learning Models: A Case Study of Turbofan Engines

An Explainable Prediction Model for Aerodynamic Noise of an Engine Turbocharger Compressor Using an Ensemble Learning and Shapley Additive Explanations Approach

Energy-Environment evaluation and Forecast of a Novel Regenerative turboshaft engine combine cycle with DNN application

Methods for Gas Turbine’s Compressor General Characteristic Curve Prediction

Research on surrogate models and optimization algorithms of compressor characteristic based on digital twins

A Novel Data-Driven Approach for Predicting the Performance Degradation of a Gas Turbine