以機器學習建構冷媒壓縮機的性能預測模型

Performance Prediction Model Construction of Refrigerant Compressors Using Machine Learning

10.29911/JEHVACE_NEW.202409_(147).0004

陳俊宇(Jiun-Yu Chen)

冷媒壓縮機 ； 機器學習 ； 性能預測 ； refrigerant compressors ； machine learning ； performance prediction

冷凍空調&能源科技

147期（2024 / 09 / 10）

51 - 60

繁體中文;英文

本文提出使用機器學習來建立冷媒壓縮機性能預測模型，以兩台壓縮機性能測試的實驗結果為例，透過開源分析軟體來嘗試多種演算法模型以進行性能預測，並根據壓縮機原廠規格數據判別適用的演算法，比較各個模型之間的表現後找出適合的模型來預測冷媒壓縮機性能，確認所建立的迴歸模型可做廣泛的性能分析預測。

In this article, machine learning is used to establish performance prediction models of refrigerant compressors. Taking the performance test experimental results of two compressors as examples, an open-source analysis software is utilized to try a variety of algorithm models for compressor performance prediction. Based on the specification data provided by the compressor manufacturers, appropriate algorithms are determined, and the results of each model are compared to find a suitable model to predict the performance of a refrigerant compressor. Besides, the regression models are identified to be used for extensive performance analysis prediction.