Shusheng Gu

A method of the knowledge acquisition using rough set knowledge reduction algorithm based on PSO

An variable precision rough set (RS) knowledge acquisition based on discrete particle swarm optimization (DPSO-VPRS) are proposed to solve rough set is lack of the ability of anti-jamming, which is used the information entropy is considered as a suitable function in discrete particle swarm algorithm and the attribute dependent degree of variable precision rough set is optimized, and make the classification rules more reliable in the case of noisy data. The study of knowledge acquisition method based on DPSO-VPRS algorithm which is applied into the grate-kiln system in order to acquire knowledge. The mass production process data is deeply analyzed, and find the key factor which determined the finished pellets quality, then attain manufacturing rule of production process control. The results showed that the grate-kiln expert method is effective and has great value as a reference to the palletizing production process control.

Development of DeviceNet fieldbus intelligent node

The DeviceNet is a low-cost and high-performance fieldbus. It has been widely applied in the industrial automatic fields in recently years, so the demands of DeviceNet product are increased gradually. To satisfy market needs, the hardware and software designs have been proposed for developing intelligent node of DeviceNet in this paper. The hardware design is mainly composed of controller (SJAl000), microprocessor (80CS1), bus transceiver (82C251), photoelectrical isolation (6N137), switches and display portion, etc. The software design is mainly composed of controller initialization, data receiving, data transmitting, etc. Hardware circuit and software flow charts have been given and some problems that should be paid attention to in realizing are indicated at the same time.

Predictive Method for Traffic Flow of Elevator Systems Based on Neural Networks

Traffic flow prediction is an important part of elevator systems. Generally, the traffic flow of elevator systems has high complexity and randomicity and the passenger flow possesses nonlinear feature, which is difficult to be expressed by a certain functional style. In this paper, we intend to construct a predictive model of traffic flow for elevator systems using time series prediction theory based on wavelet neural network. The Morlet wavelet has been chosen in this study as the activation function. The simulation results show that the novel model has much advantages over conventional model based on linear exponential smoothing method and the novel model has such properties as simple structure of network, fast convergence and higher forecast precision.

Deep Network Based on Stacked Orthogonal Convex Incremental ELM Autoencoders

Extreme learning machine (ELM) as an emerging technology has recently attracted many researchers’ interest due to its fast learning speed and state-of-the-art generalization ability in the implementation. Meanwhile, the incremental extreme learning machine (I-ELM) based on incremental learning algorithm was proposed which outperforms many popular learning algorithms. However, the incremental algorithms with ELM do not recalculate the output weights of all the existing nodes when a new node is added and cannot obtain the least-squares solution of output weight vectors. In this paper, we propose orthogonal convex incremental learning machine (OCI-ELM) with Gram-Schmidt orthogonalization method and Barron’s convex optimization learning method to solve the nonconvex optimization problem and least-squares solution problem, and then we give the rigorous proofs in theory. Moreover, in this paper, we propose a deep architecture based on stacked OCI-ELM autoencoders according to stacked generalization philosophy for solving large and complex data problems. The experimental results verified with both UCI datasets and large datasets demonstrate that the deep network based on stacked OCI-ELM autoencoders (DOC-IELM-AEs) outperforms the other methods mentioned in the paper with better performance on regression and classification problems.

An improved extreme learning algorithm based on truncated singular value decomposition

With respect to the ill-posed problem when calculating output weights of the ELM (Extreme Learning Machine), an improved ELM algorithm based on TSVD (Truncated Singular Value Decomposition) is proposed in this paper. The degree of ill-condition is severe if the hidden layer output matrix has a large condition number. In such case, the output weights computed by general SVD (Singular Value Decomposition) method will be large and unevenly distributed, which would result in a worsened stability and anti-interference ability. Also, the over-fitting phenomenon presented easily. TSVD is an effective regularization method. It can eliminate the influence caused by small singular values and enhance the generalization ability of the model. As for selecting truncation parameter, it is determined by minimizing the GCV (Generalized Cross-Validation) function with the relationship between TSVD and Tikhnovo Regularization. Simulation results illustrate that TSVD-ELM performs higher prediction accuracy than original ELM on data with noise and increases the models robustness. Finally, the proposed method is used to build a soft-sensor model to predict the quality of iron ore pellet and gets an acceptable error rate.

Improved Delphi method with weighted factor and its application

In view of long duration, high costs and expert evaluation value with many extreme points of the Delphi method, using the fuzzy method, this paper proposes an improved Delphi method with weighted factor in order to reduce the influence of subjective factors and shorten the consulting time. Then, it applies the improved Delphi method to simulate in the context of stroke rehabilitation evaluation system. Simulation results show the effectiveness of the proposed method.

An Evaluation Model of Stroke Rehabilitation

At present, the evaluation of stroke rehabilitation is mainly based on physician’s experience and evaluates the degree of one of function injury for motor function, nerve function, and activities of daily living. There is no a quantitative method. Therefore, the evaluation model of stroke rehabilitation is proposed and evaluates the condition of stroke patients quantitatively. Experimental results show that the evaluation model of stroke rehabilitation correlates well with the Fugl-Meyer assessment, which indicates that the model is effective.

Research on multiple features of sEMG combination and dimension reduction

Feature Extraction for sEMG signals is the key technique of the sEMG-based recognition system. More and more different types of features are extracted at present. Aiming at the problem of blind selection and low information utilization rate of features, an improved extraction method based on the multiple features combination and dimension reduction is proposed by this paper to build new features with high information proportion to promote the accuracy of recognition system. The basic sEMG features of time domain and AR model coefficients are used in this paper to be projected on the direction of largest error via PCA method, then the proportion and cumulative of the components on each direction are calculated for selecting the main components that contain most information to rebuild the feature vectors while reducing the dimension of the vectors. In the process of extraction, aiming at the contradiction between the stable principles of the signal in short time window and fluctuation of feature values, the length of the time window is optimized. Experiment shows that the feature extraction method proposed in this paper can promote the accuracy of recognition, which proves the efficiency of the improved method.

Model of rehabilitation training program for stroke by fuzzy reasoning

At present, using the rehabilitant robot can achieve the stroke rehabilitation training for patients, which is high intensity, targeted and repetitive rehabilitation training. It needs sound rehabilitation training program to guide the training. Due to different patients have different condition, and the same patients condition changes as time change, however, there are some uncertainties of patients in the process of drawing up a rehabilitation training program, At the same time, the process of drawing up the program needs long time, and requires lots of manpower and material resources. In view of these issues, the model of rehabilitation training program for stroke was proposed, combing with the fuzzy reasoning (FR). Experimental results show that the model can make a rehabilitation training program and improve rehabilitation efficiency effectively.

Improvement and application of the Delphi method

In view of extreme points of the Delphi method, this paper introduces the penalty factor to improve it and overcome the influence of subjective factors. Then, it applies the improved Delphi method to determine weight factor of stroke rehabilitation evaluation indicator in the context of stroke rehabilitation evaluation system. Simulation results show the effectiveness of the improved method.