Wang Hai's share is mainly divided into six parts, introducing his research topics and achievements in the master stage. 

I. Research background and significance

1. The Report on Stroke Prevention in China (2018) shows that the total number of stroke patients over 40 years old in China is as high as 12.42 million, among which the number of patients with brain tissue damage and local dysfunction caused by cerebral vascular blockage and bleeding is the second highest.

2. According to the 2010 China Disabled Persons' Federation, the number of disabled people in China has reached 85.02 million, of which 29.07% are physically disabled.

Ⅱ. Research on generating principle and preprocessing of surface electromyographic signal

1. Mechanism of EMG generation

2.EMG eigenvalues

· Time domain analysis method

· Frequency domain analysis method

· Time-frequency analysis method

3. Research ideas

Ⅲ. Construction of feature extraction model

Ⅳ. Continuous decoding of finger joint Angle based on deep forest algorithm

1. Introduction to deep forest algorithm

2. Neural network structure design

Design the empirical formula of neural network

Generally speaking, a neural network should be designed, giving priority to a three-layer network (that is, one hidden layer).

The following conditions must be met when determining the number of hidden layer nodes:

(1) The number of nodes in the hidden layer must be less than N-1 (where N is the number of training samples)

(2) The number of training samples must be more than the connection weight of the network model, which is generally 2~10 times

3. Design of finger joint Angle decoding model

Ⅴ. Design of high degree of freedom EMG prosthetic hand

Ⅵ.online control effect display