Deep belief network (DBN) is a kind of recurrent neural network superposed by several layers of unsupervised restricted Boltzmann machines (RBMs). To solve the defect of RBM during sample training that the fixed learning rate is unfavorable to find the optimal value, a principle of dynamical learning rate was proposed to improve the RBM network so as to increase the accuracy of eigenvector mapping. Moreover, the authors also constructed a DBN with two layers of RBMs, of which the neural networks were self-trained by using contrastive divergence (CD). The improved network was applied to control BLDCM..The experimental results indicate that the improved DBN can effectively accelerate response and also improve the accuracy.

Heavy duty computer numerical control machine tools have characteristics of large self-weight, load and. The insufficiency of foundation bearing capacity leads to deformation of lathe bed, which effects machining accuracy. A combined-layer foundation model is created to describe the pile group foundation of multi-soil layer in this paper. Considering piles and soil in pile group as transversely isotropic material, equivalent constitutive relationship of composite foundation is constructed. A mathematical model is established by the introduction of boundary conditions, which is based on heavy duty computer numerical control machine tools-composite pile foundation-soil interaction system. And then, the response of different soil and pile depth is studied by a case. The model improves motion accuracy of machine tools.

With the continuous development of artificial intelligence technology,the expert system,fuzzy theory,artificial neural network and many other latest intelligent control achievements start to enter the field of motor control,and they provide a brand-new way for the further improvement of the brushless direct current motor controlling.This paper combines the research status of the controlling system of both domestic and foreign brushless direct current motor,explores optimization,which is the most cutting-edge modern computer applications,the intelligent control and traditional PID controller in combination.It is difficult for traditional PID controller to achieve satisfactory results when aiming at the controlled objects or processes with complicated nonlinear features.The research makes use of the optimization algorithm to optimize intelligent PID control,such as fuzzy-PID,BP-PID,adaptive-PID control,expert-PID and PID controller based on genetic algorithm setting),to perfect the application of the intelligent control method in BLDCM controlling system.

The area of knowledge discovery and data mining are growing rapidly. A large number of methods are employed to discrete data, however, most of the existing discretion methods are applied in the case of attributes with real-value. In the practical application, the attribute value is interval number in many cases. Aiming at this problem, a new discretization algorithm applied to interval numbers is proposed. Similarity degree of interval number is used to describe the similar relation of two interval numbers. Threshold degree is defined to ensure discrete relationship between the data to implement algorithm. A new variable-associated degree is proposed through analysing action of similarity degree in the algorithm, and associated degree is used to improve algorithm. A group of data set is applied to testing the performance of the algorithm and the experiment result is compared with other discretization algorithms. The experiment result shows that the algorithm is effective.

College laboratory is an ideal environment for students to cultivate their scientific research abilities and to highlight their individual personality and speciality. So the sutudents experiment management is very important for the cultivation of new type talents. To deal with the above problems,the paper explorate the students management methods that can change the students' attitude of study from passive learning into active learning, arouse the students' enthusiasm of study, and also train the students' abilities of analyzing and solving problems.

人本教育管理是信息化时代对高校人力资源管理提出的新要求,是教育面向现代化的必然要求。本文以信息化为视角分析了人本教育管理的涵义、主要作用,探讨了高校人本教育管理的应用策略。

the paper introduces the fundamental principles and realization procedure of edge detection by canny operator. With Mat lab, it use Roberts, Sobel, Prewitt, LOG and Canny operator to deal with the car plate's image. According to the result, it is confirmed to be a suitable method for the edge detection of car plate's image.

To deal with the various and complicated index in university evaluation, the paper proposed a new method based on rough set. Depending on the relationship between conditon attributes and decision attributes, selected the important attributes by using the discernible vector, realized the zttribute reduction and finally established the evaluation system. The experiment indicates the method is effective.

This paper introduces a newly-structured capacitance sensor which is used for water-containing measure for grain. The special circuit is designed and measure result is also given. It analyzes the factors that affect the measure and puts forward reasonable method.

Ball mill blending process is one of the key producing processes in the sintering alumina production.In this blending process,raw materials are ground and blended in the mill to produce raw slurry.The automatic control system was designed and developed by using S7-300 PLC of Siemens Company in consideration of this industry features and technique requested.The structure,functions and control strategy of this system were all discussed,and the control software was developed.Control software includes the logic control program,continuous control program and picture monitoring.The proposed control system can not only improve the quality of raw slurry,but also can lower energy consumption and create more profit for its enterprise.

Discretization is important in data preprocess.After researching on the interval numbers discretization,a new algorithm is proposed.Firstly,associated degree of interval number is introduced to describe the associated relationship of two interval number,then threshold degree is defined to value discretization relationship between the datas.Finally,the algorithm defermine is applied in a group of data sets to test the performance.The experiment results compared with other discretization algorithm show that the algorithm is effective and effcient.

Computer organization is one major course in computer science and technology field. How to master the computer basic technology of computer architecture,Some problems are the common difficulties in the teaching process,such as how to analyze the computer system, how to design the computer system, etc. To deal with the above problems,the paper proposes some methods,for example, teaching books integration, hierarchical teaching, developing the students' enthusiasm, adopting the modern teaching methods and so on. These methods can increase the students' academic motivation, develop their ability of analyzing and solving problems and their ability of innovation can be improved,which are very useful to build the course and develop the teching quality.

Vehicle license plate location is the important procedure of license plate identification, there are lots of usual algorithm. The paper introduces method of preparation before location. It analyses the advantage and disadvantage of usual algorithm. According to the features of RGB of vehicle license in color image, it can deal with vehicle-license pixels' distribution of x-velocity and y-velocity statistically. It finds the position of vehicle-license in image, and shows the result's concrete analysis, summarizes the corrective method and factors which has effect on location result.

高职院校的不断发展,使高职院校的教学管理问题备受关注。师资队伍建设及教育教学观念等跟不上发展的速度,使高职教育在教学管理中存在一定的问题。本文分析了目前高职教育的教学管理存在的问题,并针对这些问题提出了相应的解决方案。

According to experiences on teaching "Embedded Systems" course, status of school students and employment needs,this paper summarizes Some new ideas and experiences of "embedded systems" course from the teaching content, teaching methods,laboratory equipment, teaching staff construction, to better to promote the teaching and learning, these measures is proved effective in improving the students practical ability, learning ability, communication skills and teamwork in teaching.

The "SQL Server database" is the one of specialized course that software engineering profession opened at institutions of higher learning colleges and universities. The course emphasizes the actual operation,and has strong practicality. The paper describes the course teaching situation and analyzes the problems existing at present. The paper made innovation in the teaching method: the teaching method of project driving, improving the teaching methods and evaluation mechanism combining lingbo teaching software system. combining 1x teaching software system.

Knowledge acquisition is an important research area of knowledge discovery database and machine leaning, which includs knowledge reduction and knowledge extraction from large number of original data. Researchers in these fields are very interested in this new research topic since it offers opportunities to discover useful knowledge in information systems.Many algorithms demand information system must be complete. To deal with the problem in an incomplete information system, this paper proposed a method based on rough set theory. Based on tolerance relationship, the concept of tolerance relationship similar matrix via using an extension of equivalence relationship of rough set theory are defined in incomplete information systems. It calculates the core attributes of incomplete information systems via the tolerance relationship similar matrix. To overcome its drawback of NP-hard time complexity,It applies attribute significance, which this paper puts forward based on attribute frequency in the tolerance relationship similar matrix, as the heuristic knowledge, makes use of binsearch heuristic algorithm to calculate the candidate attribute expansion so that it can reduce the expansion times to speed up reduction. Experiment results show that the algorithm is simple and effective.

This paper defines a concept of attribute significance based on the granular fineness,and uses the significance as the inspiration for attribute reduction.From the original uncertainty information systems,a space of granular based on equivalence relation of rough sets is defined.This method greatly reduces the space occupied by the original system to avoid duplication of the calculation object.And the computation scale can be decreased.

For the majority of attribute reduction algorithm,their time complexity are relatively high.So this pap 1 er proposes a new solution using rough set.A formula to compute attribute importance is put forward based on the concept of similar matrix and attribute frequency in the matrix,which is used as heuristic knowledge in deciding the sequence of attribute reduction.The idea of half-inquery is used in the algorithm to accerate reduction.Experiments show that the algorithm is simple and effective.

高校行政管理专业在课堂教学有效知识量、实用时间、课堂学习投入度等课堂教学方面的有效性存在诸多问题,本文认为只有大力推进教学改革,不断更新传统教学观念、转变教学方式、创新教学手段,逐步适应当前人才培养的现实需求,才能提高行政管理专业的教学质量。