[1]覃树宏,梁 锦.doi: 10.3969/j.issn.1001-3849.2026.01.015面向Ni-SiC纳米镀层耐磨性能预测的GA-BP神经网络模型[J].电镀与精饰,2026,(01):116-122.
　QIN Shuhong,LIANG Jin.Predicted GA-BP neural network model for wear resistance of Ni-SiC nano coatings[J].Plating & Finishing,2026,(01):116-122.
点击复制

doi: 10.3969/j.issn.1001-3849.2026.01.015面向Ni-SiC纳米镀层耐磨性能预测的GA-BP神经网络模型()

《电镀与精饰》[ISSN:1001-3849/CN:12-1096/TG]

卷:
期数:
2026年01
页码:
116-122
栏目:
出版日期:
2026-01-31

文章信息/Info

Title:
Predicted GA-BP neural network model for wear resistance of Ni-SiC nano coatings
作者:
覃树宏; 梁 锦
(广西民族大学相思湖学院 招标采购办公室,广西 南宁,530225)
Author(s):
QIN Shuhong;  LIANG Jin
(Bidding and Procurement Office, Xiangsihu College of Guangxi Minzu University, Nanning 530225, China)
关键词:
Ni-SiC纳米镀层; 耐磨性能预测; GA算法; BP神经网络; 摩擦磨损
Keywords:
Ni-SiC nano coating;  wear resistance prediction;  GA algorithm;  BP neural network;  friction and wear
分类号:
TG174.4；TQ153
文献标志码:
A
摘要:
Ni-SiC纳米镀层的耐磨性能与其制备工艺参数之间存在复杂的非线性关系，需要具有很强的非线性拟合能力，才能捕捉输入参数与耐磨性能之间的复杂关系，在进行模型求解时可避免陷入局部最优而降低预测精度。为此，提出遗传算法-反向传播(Genetic Algorithm-Backpropagation，GA-BP)神经网络模型，对Ni-SiC纳米镀层的耐磨性能预测方法展开研究。选用50?mm×50?mm×5?mm 304 不锈钢板材作为基体材料进行预处理，使用电镀液配方对镀液进行配置；采用恒电流脉冲电镀模式完成复合电镀，并利用多功能摩擦磨损试验机进行耐磨性能试验；构建基于BP神经网络的Ni-SiC纳米镀层耐磨性能预测模型，并引入遗传算法对BP神经网络模型的阈值和权值展开寻优，将磨损量作为模型输出，实现Ni-SiC纳米镀层的耐磨性能预测。试验表明，利用本文方法获取的磨损量预测值与磨损量真实值之间的误差最大仅为0.2?mg，预测后的R2为0.988，预测结果的拟合优度较高，应用效果较好。
Abstract:
There is a complex nonlinear relationship between the wear resistance of Ni-SiC nano coating and its preparation process parameters, capturing this relationship requires strong nonlinear fitting capabilities. When solving the model, it can avoid falling into local optima and reducing prediction accuracy. Therefore, a GA-BP neural network model is proposed to study the prediction method of wear resistance of Ni- SiC nano coatings. 5 0?mm×50?mm×5?mm 30 4 stainless steel plate is selected as the substrate material for pretreatment, and the plating solution is configured using a plating solution formula. The composite electroplating was completed using a constant current pulse electroplating mode, and the wear resistance was tested using a multifunctional friction and wear tester. A prediction model for the wear resistance of Ni-SiC nano coating based on BP neural network is constructed, and a genetic algorithm is introduced to optimize the threshold and weight of the BP neural network model. The wear amount is used as the model output to achieve the prediction of the wear resistance of Ni- SiC nano coating. The experiment shows that the maximum error between the predicted wear amount obtained by the method in this article and the true wear amount is only 0.2?mg, and the predicted R2 is 0.988. The goodness of fit of the predicted results is higher, and the application effect is better

参考文献/References:

[1].郭明忠, 谭世光, 任爽, 等. Ni-P/纳米SiC复合镀层制备及性能研究[J]. 广州化工, 2024, 52(2): 97-98, 105.
[2].宿辉, 栾柏瑞, 张春波. SiCP添加量对AZ91D镁合金表面纳米环保复合沉积层的影响[J]. 材料工程, 2023, 51(2): 116-123.
[3].李心源, 马春阳, 赵旭东. 基于BP神经网络模型的Ni-SiC纳米镀层耐磨性能预测研究[J]. 功能材料, 2020, 51(1): 1126-1130.
[4].李蕴, 刘海. 基于大数据分析的MEMS电镀铜薄膜耐磨性预测[J]. 科技通报, 2017, 33(12): 121-124.
[5].邓羽, 张杰, 彭中波, 等. 基于人工神经网络的Ni-ZrO2纳米镀层耐腐蚀性能预测[J]. 装备环境工程, 2022, 19(2): 98-105.
[6].王亚利, 于继明, 王艺. 遗传算法-模糊径向基神经网络模型预测自润滑镀层耐磨性[J]. 电镀与精饰, 2021, 43(7): 30-34.
[7].朱慧泉, 车银超. 基于果蝇优化算法的电镀槽液温度自

更新日期/Last Update: 2026-01-16