Optimization of Copper Electroforming Process Parameters Based on Double Hidden Layer BP Neural Network.

Area of Science:

• Materials Science
• Electrochemistry
• Computational Modeling

Background:

• Optimizing electroforming processes is crucial for achieving desired material properties.
• Pulse electroforming offers advantages over direct current methods but requires precise parameter control.
• Predictive modeling can accelerate the optimization of complex electrodeposition parameters.

Purpose of the Study:

• To develop a predictive model for optimizing the pulse electroforming copper process.
• To establish the relationship between process parameters and the microhardness and tensile strength of electrodeposited copper.
• To identify optimal operating conditions for enhanced copper layer properties.

Main Methods:

• Construction of a double hidden layer back propagation (BP) neural network.
• Training the neural network with electroforming process data to map conditions to properties.
• Experimental verification of predicted microhardness and tensile strength using copper pyrophosphate solutions and pulse power supply.

Main Results:

• The "3-4-3-2" structure neural network accurately predicted microhardness and tensile strength, with a relative error below 2.82%.
• Predicted microhardness ranged from 100.3–205.6 MPa, and tensile strength ranged from 165–485 MPa.
• Optimal parameters for enhanced properties were identified: current density (2–3 A·dm⁻²), pulse frequency (1.5–2 kHz), and pulse duty cycle (10–20%).

Conclusions:

• A double hidden layer BP neural network effectively models the pulse electroforming copper process.
• The predictive model enables accurate estimation of copper layer properties, facilitating process optimization.
• The study provides a data-driven approach to determine optimal parameters for achieving desired microhardness and tensile strength in electroformed copper.