Design optimization of a heat‐to‐cool Stirling cycle using artificial neural network

Abstract

This study focuses on applying artificial neural network (ANN) to investigate different parameters and operating conditions that affect the cooling flux and efficiency of a heat-to-cool gamma-type Stirling machine. ANN model is developed to predict output parameters of the Stirling machine based on sample data generated by a MATLAB code-named Nlog code. The code was previously developed and validated by the authors. To reach the addressed goal, three input parameters, including Stirling refrigerator frequency (f), displacer stroke (d), and phase angle (φ) were selected. The output parameters are coefficient of performance (COP) and cooling power of the Stirling refrigerator, and efficiency (η) and output power (P e ) of the Stirling engine. The ANNs proved to have the most effective performance with 10 neurons and the ANN structure as 4-10-1 since it could perfectly map the input parameters to the output parameters. The presented ANNs were compared with four different design cases of a Stirling refrigerator against the experimental data. The ANN results had good agreement with experimental data with an average error of less than 1% for the cooling power and an average error of less than 8% for the COP.Besides, the ANN model and Nlog code error for the efficiency and output power of the Stirling engine were 1.33% and 0.21%, respectively.

Parametric Investigation and Univariant Optimization of Domestic Stirling Refrigerator

“…Artificial neural network (ANN) has been applied to study the effect of different parameters and operating conditions for the heat-to-cool gamma-type Stirling machine [28]. It has been reported that ANN model could predict the performance of a Stirling machine accurately and within a reasonable time.…”

The Role of Artificial Intelligence in Civil Engineering Applications and Programs

“…AI algorithms can automate the analysis and optimization of complex structures, enabling engineers to quickly evaluate different design options and identify optimal solutions. Additionally, AIbased predictive modeling techniques allow engineers to assess the performance of structures under different loads and environmental conditions [3].…”

“…Integration Complexity: Integrating AI techniques into existing civil engineering programs and drawing techniques may pose technical challenges. Compatibility issues, software integration, and interoperability need to be addressed to ensure smooth integration and seamless workflow[3]. 7.…”