Backpropagation Neural Network-Based Prediction Model of Marble Surface Quality Cut by Diamond Wire Saw

At a Glance

Metadata	Details
Publication Date	2025-08-23
Journal	Micromachines
Authors	Dong Hui, Feng Cui, Zhipu Huo, Yufei Gao
Institutions	Shandong University, Shandong Jiaotong University
Analysis	Full AI Review Included

Technical Documentation: IWOA-BP Optimization for Diamond Wire Sawing

This document analyzes the research paper “Backpropagation Neural Network-Based Prediction Model of Marble Surface Quality Cut by Diamond Wire Saw” and connects its findings regarding precision cutting of hard, brittle materials to the advanced MPCVD diamond solutions offered by 6ccvd.com.

Executive Summary

The research successfully developed an intelligent prediction model for controlling the surface quality of hard, brittle materials (marble) during diamond wire sawing (DWS).

Core Achievement: Establishment of an IWOA-BP (Improved Whale Optimization Algorithm-Backpropagation) neural network model to accurately predict surface roughness (Ra) and waviness (Wa).
Superior Performance: The IWOA-BP model demonstrated optimal prediction performance, significantly reducing error metrics compared to traditional BP and other optimized neural networks.
Key Error Reduction: The Root Mean Square Error (RMSE) for Ra prediction was minimized to 0.0342, and for Wa prediction to 0.0570.
Methodology: Systematic data collection via single-factor, orthogonal, and Box-Behnken design experiments varying Feed Speed (Vf), Wire Speed (Vs), and Sawing Length (L).
Application Relevance: Provides a robust, data-driven framework for optimizing DWS parameters, crucial for high-precision slicing of ultra-hard materials, including MPCVD diamond substrates.
6CCVD Connection: The findings directly support the need for high-quality, customizable diamond materials and precision processing techniques essential for advanced diamond wafer manufacturing.

Technical Specifications

The following table summarizes the critical experimental parameters and the optimized model performance metrics extracted from the research.

Parameter	Value	Unit	Context
Roughness Prediction RMSE (Ra)	0.0342	-	IWOA-BP Model (Minimum Error)
Waviness Prediction RMSE (Wa)	0.0570	-	IWOA-BP Model (Minimum Error)
Ra Prediction MAPE	1.5614%	-	Mean Absolute Percentage Error
Wa Prediction MAPE	1.7028%	-	Mean Absolute Percentage Error
Optimal Hidden Layer Nodes	6	Nodes	Selected for minimal RMSE
Correlation Coefficient (R)	> 0.99	-	High fitting degree across all datasets
Saw Wire Core Diameter	220	µm	Nickel-plated diamond wire
Abrasive Particle Size	25-35	µm	Diamond grit range used in DWS
Particle Distribution Density	70-85	grits/mm²	Abrasive concentration
Feed Speed (Vf) Range	0.6 to 3	mm/min	Experimental input variable range
Wire Speed (Vs) Range	600 to 1400	m/min	Experimental input variable range
Achieved Surface Roughness (Ra)	1.113 to 2.685	µm	Measured roughness range on marble slices

Key Methodologies

The prediction model was developed through a systematic experimental and computational process focused on optimizing the cutting of hard, brittle materials.

Experimental Design: Sawing experiments were conducted on marble using a diamond wire saw (SH300). Input parameters (Feed Speed Vf, Wire Speed Vs, Sawing Length L) were systematically varied using single-factor, orthogonal, and Box-Behnken designs to generate a comprehensive dataset.
Surface Quality Measurement: Surface roughness (Ra) and waviness (Wa) were measured post-sawing using a VK-X200K laser confocal microscope. Five random positions were measured per slice, and the average was used as the characterization index.
Neural Network Structure: A three-layer Backpropagation (BP) neural network was adopted, featuring 3 input nodes (Vf, Vs, L) and 1 output node (Ra or Wa, trained separately). The Sigmoid function was used for hidden layer activation.
IWOA Optimization Strategy: The BP network’s initial weights and thresholds were optimized using the Improved Whale Optimization Algorithm (IWOA), incorporating five hybrid strategies to enhance global search and convergence speed:
- Sine chaotic mapping and quasi-reverse learning for population initialization.
- Improved nonlinear convergence factor.
- Adaptive weighting strategy (w_min = 0.4, w_max = 0.9).
- Archimedean spiral position update (replacing the logarithmic spiral).
- Random differential mutation strategy.
Model Validation: Prediction accuracy was verified by comparing the IWOA-BP model against four other neural network models (BP, WOA-BP, IMWOA-BP, TIWOA-BP) using RMSE, MAE, and MAPE metrics.

6CCVD Solutions & Capabilities

The research highlights the critical need for precise process control when slicing hard, brittle materials using diamond wire. As the global leader in MPCVD diamond, 6CCVD provides the foundational materials and customization services necessary to replicate, extend, and surpass this research, particularly in the context of cutting diamond itself.

Applicable Materials for Advanced DWS Research

The paper used diamond abrasives to cut marble. For engineers developing next-generation DWS tools or requiring ultra-high-quality substrates for subsequent processing, 6CCVD offers materials with unmatched purity and mechanical stability.

6CCVD Material	Relevance to DWS Research	Key Specifications
Optical Grade SCD	Ideal for high-power optics, quantum applications, and substrates requiring ultimate surface quality (Ra < 1 nm).	High purity (Type IIa), Thickness: 0.1 µm - 500 µm, Polishing: Ra < 1 nm.
High-Purity PCD	Excellent for large-area thermal management, high-wear tooling, and large-scale DWS experiments.	Plates/wafers up to 125 mm diameter, Thickness: 0.1 µm - 500 µm, Substrates up to 10 mm.
Boron-Doped Diamond (BDD)	Used for electrochemical sensing and high-conductivity applications, often requiring precision slicing and metalization.	Customizable doping levels, available in both SCD and PCD formats.

Customization Potential for Tool Development

The paper optimized parameters based on a specific abrasive size (25-35 µm) and density (70-85 grits/mm²). 6CCVD’s capabilities allow researchers to test new tool designs and material interfaces directly.

Custom Dimensions: We provide custom diamond plates and wafers up to 125 mm in diameter (PCD) and substrates up to 10 mm thick, enabling large-scale DWS research platforms.
Ultra-Precision Polishing: While the marble achieved Ra < 3 µm, 6CCVD guarantees surface roughness of Ra < 1 nm for SCD and Ra < 5 nm for inch-size PCD. This capability is essential for applications where DWS is used as a preliminary step before final high-precision finishing.
Advanced Metalization: For integrating diamond substrates into complex systems (e.g., thermal sinks, sensors), 6CCVD offers in-house metalization services, including: Au, Pt, Pd, Ti, W, and Cu. This is critical for researchers studying the interface between diamond and tool bonding materials.

Engineering Support

The successful implementation of the IWOA-BP model demonstrates the value of intelligent optimization in processing hard, brittle materials. 6CCVD’s in-house PhD team specializes in the growth, processing, and characterization of MPCVD diamond.

Application Focus: Our experts can assist researchers applying similar AI/ML optimization techniques to diamond processing itself (e.g., optimizing laser cutting parameters, chemical-mechanical polishing recipes, or metal adhesion for similar Diamond Wire Sawing projects).
Global Logistics: We ensure reliable, global delivery of custom diamond materials, with DDU (Delivery Duty Unpaid) as the default and DDP (Delivery Duty Paid) available upon request.

For custom specifications or material consultation, visit 6ccvd.com or contact our engineering team directly.

View Original Abstract

Marble is widely used in the field of construction and home decoration because of its high strength, high hardness and good wear resistance. Diamond wire sawing has been applied in marble cutting in industry due to its features such as low material loss, high cutting accuracy and low noise. The sawing surface quality directly affects the subsequent processing efficiency and economic benefit of marble products. The surface quality is affected by multiple parameters such as process parameters and workpiece sizes, making it difficult to accurately predict through traditional empirical equations or linear models. To improve prediction accuracy, this paper proposes a prediction model based on backpropagation (BP) neural network. Firstly, through the experiments of sawing marbles with the diamond wire saw, the datasets of surface roughness and waviness under different process parameters were obtained. Secondly, a BP neural network model was established, and the mixed-strategy-improved whale optimization algorithm (IWOA) was used to optimize the initial weight and threshold of the network, and established the IWOA-BP neural network model. Finally, the performance of the model was verified by comparison with the traditional models. The results showed that the IWOA-BP neural network model demonstrated the optimal prediction performance in both the surface roughness Ra and waviness Wa. The minimum predicted values of the root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) were 0.0342%, 0.0284% and 1.5614%, respectively, which proved that the model had higher prediction accuracy. This study provides experimental basis and technical support for the prediction of the surface quality of marble material cut by diamond wire saw.

Tech Support

Original Source

DOI: https://doi.org/10.3390/mi16090971

References

2022 - The use of waste marble for cleaner production of structural concrete: A comprehensive experimental study [Crossref]
2023 - A review of dimension stone extraction methods [Crossref]
2025 - A review of silicosis and other silica-related diseases in the engineered stone countertop processing industry [Crossref]
2025 - An AHP and BEES model-based sustainability assessment of small-scale marble processing industries: Indian prospects [Crossref]
2025 - Effects of marble waste on shear strength and pullout resistance of sand reinforced with Glass Fiber-Reinforced Polymer (GFRP) bars
2025 - Mixed-mode fracture behavior of a marble exposed to acidic environments [Crossref]
2021 - The mechanics of sawing granite with diamond wire [Crossref]
2022 - Research progress of wire saw machining technology [Crossref]