Search

In order to explore machining characteristics of the soft abrasive flow field in a structured mold surface,the Single Dynamics Model(SPD) was applied to the solution to the motion characteristics of particles in different types of turbulence fields numerically.By taking a U-shaped flow passage for an example, the velocity and pressure of fluid in a two-phase soft abrasive flow field were solved by using the N-S equations,a Realizable k-e model of turbulence and the SIMPLEC method.Then the velocity,locus and the density of particles in the two-phase field were also obtained by the SPD model.The results are that when the initial fluid velocities are 5,10 and 20 m/s,the particle deposition is the largest for the first case.When the particle diameters are 0.01,0.05 and 0.1 mm,the first value causes the particle deposition to be most obvious.In turbulence fields with different viscosities,the water,gas-oil and engine-oil show similar two-phase soft abrasive flow characteristics.It concludes that the initial velocity of the fluid field and the diameters of particles have more severe effect on the moving characteristics of particles,and the viscosity of the fluid influences them the least.

With continuous improvement of computer science and network technology, traditional industry monitoring network system (IMNS) based on server/client pattern can not supply enough self-adapting ability and satisfy the increasing demand of modern industry monitoring, so a kind of intelligent embedded IMNS realization method based on universal plug and play (UPnP) was put forward in this paper. According to the special demand and environment characteristic of industrial monitoring fields, the topological model of IMNS was established around UPnP. System modules' partition and data interaction process were constituted by modularization theory which can make this system reconfigure and optimize easily. IMNS nodes and monitoring center were designed and realized using embedded system technology, and its industrial application instance was finished. Industry fields experiments proved that this system can obtain the anticipated functions of seamless link, zero-configuration and self-recognition using UPnP service program, and it can break through the limitation of enterprise internal gateway and real-IP to realize online remote monitoring and real-time diagnosis by gateway support of UPnP.

Aiming at some problems of embedded industry monitoring system graphic user interface (IMS-GUI) such as large occupation space, bad configuration ability and poor real-time performance, an embedded IMS-GUI components library realization method based on Qt/E was put forward. The Qt/E library file was tailored according to basic requirement of IMS. Using multi-thread method, system parallel processing abilities of data collection, analysis and display were enhanced, and real-time performance and rapidity were guaranteed. By internal communication interface of Qt, the message mapping mechanism which would not require call-back was realized, context switch speed was improved, and GUI development process would become simply. Based on modularization principle, combined with some related industry monitoring and fault diagnostic algorithms, GUI shared components library oriented to IMS was developed by primary C language, that could resolve the problem of cross-platform applicability and be configured and tailored for specific demand of users. Industry experiments proved that GUI developed by the components library has small occupation space and high real-time performance, can stably run in embedded platform environment, and supply powerful technical support for the interactivity of IMS.

Temperature control system (TCS) of texturing machine (TM) has many control channels, requires high control precision, and it is prone to be influenced by peripheral environmental factors. Since traditional temperature control method hardly realize ideal control effect, a kind of TCS realization method based on self-turning PID technology for constant temperature box (CTB) of TM was put forward based on the wide analysis of characteristics and methods of TCS. According to structure characteristic and working environment of TM-CTB, its discrete mathematics model was established by thermodynamics principle. The self-turning PID minimum phase control system for TM temperature controller was designed using pole points placement method, and its application instance was established by embedded system technology. System parameters were identified by recursion least square method, and forgetting factor was introduced to improve the model tracing ability of TCS. Experiments results showed that the method can satisfy the multi-channel requirement of texturing machine TCS, and its temperature control error can be limited under one percent.

Embedded fault diagnosis technology requires high pertinency and small occupation space. Traditional fault diagnosis system can not satisfy the demand above. Aiming at this problem, a kind of embedded fault diagnosis expert system (E-FDES) based on CLIPS and ANN was brought forward. FDES and its relative technology were discussed, developing environment and design tool chain for E-FDES was established in Linux environment. Using modularization theory principle, system re-configuration and expansion ability were guaranteed, and data process components and graphic user interface components could be reduced to meet the embedded system running requirement. By application program interface of CLIPS and Protege plug-in, knowledge base and rule base were realized, that decreased the developing time of system inference engine and reduced source occupation space. Industry experiments prove that this system runs stably in ARM-Linux embedded environment, switch smoothly and can recognize some common faults for the devices monitored.

There are a lot of problems in pouring and pick equipments (PPE) which are widely applied in oil field, such as many different kinds of faults, dispersal of equipments, strict requirement for continuous working etc. But traditional fault diagnosis technology (FDT) based on PC doesn't have adequate flexibility, so it cannot satisfy the requirements as mentioned before. To solve these problems, a new kind of hand-held state monitoring and fault diagnosis system (SM-FDS) based on embedded technology was put forward. After detailed analysis of working situation and operating features of PPE, the realization way which is mainly based on the index of vibration is confirmed. Using multi-channel data collecting technology, hand-held circuit check is realized in real-time data collecting process. Signal analysis component library and graphical user interfaces (GUI) component library which can reconstruct and occupy small space are established to make characteristic extraction of signal, online analysis and diagnosis by multi-kind signal analysis method. According to specific requirements of hand-held instruments, monitoring system which has upper-lower position computer with dual-CPU computer architecture and low power consumption designation is realized. Industry experiments indicated that this system monitors accurately, works stably, and has a bright application future.

This paper is based on the investigation and analysis of today's multi-component force sensor, discussing the mathematic model of six-component force sensor in theory level and establishing the equations set. Then we adopt two algorithms, random-Newton and regression step by step, to solve the equations set. In the two algorithms above, random-Newton is our creation which lead through the genetic algorithm in the traditional Newton iteration and has very high efficiency of calculation. The other one, regression step by step, has better outcome than random-Newton in some condition, but its procedure is more difficult than random-Newton and requires more experiments with high accuracy. So, according to the achievement above, we develop the six-component force sensor and calibration system. After large numbers of experiment, the result is very good.