Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Tin-bismuth alloy performance prediction method based on transfer learning

A tin-bismuth alloy and performance prediction technology, which is applied in the field of tin-bismuth alloy performance prediction and modeling, can solve problems such as difficulty in training machine learning models, and achieve the effect of improving the prediction rate

Pending Publication Date: 2020-03-24
云南锡业集团(控股)有限责任公司研发中心
View PDF5 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, because of this method that requires a large amount of data to model with high precision, the training of machine learning models under small samples is extremely difficult

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Tin-bismuth alloy performance prediction method based on transfer learning
  • Tin-bismuth alloy performance prediction method based on transfer learning
  • Tin-bismuth alloy performance prediction method based on transfer learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] Embodiment 1: The performance prediction method of this tin-bismuth alloy based on transfer learning is as follows:

[0045] 1. Build a deep belief network

[0046] Use python language and build DBN based on tensorflow framework, see figure 1 ;

[0047] (1) Load the deep belief network library

[0048] Import the following packages in the programming code: urllib, math, tensorflow, numpy, utils, PIL;

[0049] (2) Initialize a DeepBeliefNetwork object, and pass in parameters including pretrain, rbm_layers, rbm_learning_rate;

[0050] (3) Initialize n RBM objects in the DeepBeliefNetwork object, where dbn.sizes=[100 100] is set;

[0051] Among them, in DBN, the probability distribution of system random variables, that is, the energy relationship corresponding to the objective function is

[0052]

[0053] W ij is the connection weight from visible layer neuron i to hidden layer neuron j, b j is the bias of the jth neuron in the visible layer, c i is the bias of t...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a tin-bismuth alloy performance prediction method based on transfer learning. According to the invention, the method comprises the steps: constructing a deep belief network, transferring the learned priori knowledge to the feature learning of the new tin-bismuth alloy through the model, and predicting the performances of the new tin-bismuth alloy at different proportions. According to the method, the existing tin-bismuth alloy performance data and prior knowledge of a prediction model are utilized, and under the experiment condition based on a small amount of data, theproblem that data characteristics cannot be effectively learned due to the fact that the sample amount of a deep learning algorithm is too small can be effectively solved; according to the method, thecorresponding performance of the new tin-bismuth alloy can be predicted through transfer learning, and the vacancy of the tin-bismuth alloy performance prediction field to a certain degree is made up; the prediction rate of the performance prediction model achieved through the method is greatly increased compared with the prediction rate of a performance prediction model without transfer learning.

Description

technical field [0001] The invention relates to a performance testing method of tin-bismuth alloy based on transfer learning, which belongs to the field of performance prediction modeling of tin-bismuth alloy. Background technique [0002] In recent years, the use of lead-free solders such as tin-bismuth alloys in the electronics industry has become an inevitable trend. They are widely used in all levels of soldering processes in electronic packaging and have good wettability. At present, the technical formulation of lead-free tin solder used in the domestic electronic packaging industry mainly relies on the United States and Japan. Companies represented by Japan Senju and American Alpha hold most of the patent rights of lead-free solder. The domestic development of tin-bismuth lead-free solder is subject to foreign influence. Patents are severely constrained, and there is an urgent need to accelerate the development of tin-bismuth alloy series products with excellent perfor...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G16C20/30G16C20/70G06N3/04G06N3/08
CPCG16C20/30G16C20/70G06N3/084G06N3/045
Inventor 马朝君王旖旎张文兵陈光云彭巨擘沈韬刘英莉朱艳
Owner 云南锡业集团(控股)有限责任公司研发中心
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products