Application of Fuzzy-PID Control System in Full-Mechanized Coal Face

Mar 2005J. China Univ. of Mining Tech (English Edition)VoL 15 No. 1Application of Fuzzy-PID Control System inFull-Mechanized Coal faceLU Kui", TANG Pei-rong, YANG Wei-miDepartment of Computer Technology Engineering, Anhui University of Science and TechnologyHuainan, Anhui 232001. China2 China University of mining Technology, Beijing 100083, China;The Eighth institute of Electron Technology Group Company of China, Huainan, Anhui 232001, ChinaAbstract: The control system, which includes structure, the composition of software and hardware, the form of PIDcontrol syd its systematic closed-loop, was used in No 4236 full-mechanized coal face of Xinlongzhuang mineThe typical fuzzy PID control system structure was investigated, and a simplified fuzzy PID control system was takenthe place of the complex three-dimension fuzzy controller. Based on the parameter relation between fuzzy controllerand normal PID controller, a common method of parameter adjustment of Pid controller was summed up and thecomputer simulation was realized. This system can overcome the problems of large delay, nonlinear, poor running en-vironment and great load change in the full-mechanized coal face. The simulating investigation indicates that the de-signing method of fuzzy controller is simple and feasible.Key words: full-mechanized coal face; nonlinear; computer; fuzzy control; PID controlCLCr:TP2734;TD821 Introductioncomputer(PM3 I/Ze/a)computer(PM31/Ze/a)In Xinglongzhuang coal mine the computer controlling system of No 4326 full-mechanized coal faceModbuswas used to realize the modern production and man-agement. The computer controlling system in full-mechanized coal face is a long-delay, non-linearembedded(PC104)and controllingTK200)systemin bad environment and with a high fluctu-Fig. 1 The control systemating load. Therefore, in this system using generalPID controlling system the good effect can't be ob-The composition of hardware of this system istained. This paper introduces the fuzzy-PID control1)The PM31/Ze/a is a type of industry contrsystem in No 4326 full-mechanized coal face of Xi- lling computer to serve electro-hydraulic controllininglongzhuang coal minesystem, which is a general type with intrinsic safe.Because in coal mine some explosive danger exists,2 Construction of Controlling System中国煤化工 ing for supervisingThe construction of control systemis shown inCNMHGu PlIer SCM was used asFig 1fuzzy PID controller 3. The reason is: high reliability,Received 6 July 2004: accepted 30 October 2004Corresponding author: Tel: +86-554-6631951; E-mail address: klu@ aust. eduLU Kui et alapplication of Fuzzy- PID Control System in Full-Mechanized Coal Faceuninterrupted working; adaptability for bad working function buttons can be set on the main interfaceenvironment; lower power consumption; friendly1)Selection of strategies for fuzzy controllingfigure interface; smaller volume; standard vo port to select the fuzzy controlling method, such as, basicand easy developmentconsulting-tables, triangle subj3)The AT89C52-SCM was used as supervising able field fuzzy algorithm and Basseville fuzzyand controlling unit. By means of series A/D 8 chan- trolling etc.nel 4-20 mA current was converted into digital sig-2)Selection of shape for subjection functionsnalto involve triangle, Gauss and trapezium3)Setting the fuzzy parameters: to selecfor electronic controller. By this protocol, communvariable of input, and to define the valuecation can be realized between controller and other each variable and its fuzzy subsetdevices4- It has become a standard industrial proto-4)Setting the precision parameters: to select thecan industrial web in which supervision can be car- each variable and its fuzzy subser e value range ofcol. All controlling devices can be connected with variables of output, and to define the value range ofried outIn the coal mine for computer fuzzy controlling5)The TK200 is a system of communication, the basic algorithm includes: Firstly, the fuzzy concontrol, and protection. On the basis of the TK100 trolling algorithms are compiled using C languagesystem it has some new functions such as protecting dLL by iFix's OLE container function; Secondly, thecoal carrier in coal face, controlling, communication, specified Active X controller is called; Finally, theseexploring fault point, display and alarmActive X controllers are embedded in iFix systemand every algorithm is conveniently implemented3 Software Designthrough iFix platform3.2 Function of the software running in3.1 Basic scheme of supervising computer in coal1)Collecting and managing data by MODBUSThe pm31/zela computer was supplied by Ger2)Through Win Socket setting up the TCP/P li-many, which is compatible with X86. The operating nk, sending data, receiving information, configuringsystem is windows 2000. The configuration softwarestate and controlling actionis iFix which is used in both ground and underground3) Realizing visible production process by graphTherefore, in the ground the configuration schemecan be carried out and transferred to the underground4) Managing the alarm, such as early warningThe TCP/IP(PPP/SLIP) protocol was used as com- and alarm of the system fault and the dangerous reIter communication in ground, and the MODBUs sultsprotocol was used as communication between super-5)Controlling or locking supervision sets of thevising computer and embedded PC104-SCM in the system6)Coordinating each sub-systemThe iFix-Dynamics system platform provides a 3.3 The software scheme of PC104 embeddedgood developing environment for industrial control,SCMwhich is developed by American Interlution Group1)Operation system: Because PC104 embeddedand can connect with Modbus. Using its graph-base, SCM中国煤化工emoy. indowsthe manufacturing flow can be drawn, and each in- CE wdustrial parameter can be showed in time with dy- tionsCNMHGalize their funcnamic fashion otherwise the current trend curve and2)Interface software: iFix Dynamihistorical trend curve can be drawn as3)Protocol and IO: MODBUSJ. China Univ of Mining Tech (English Edition)Vol 15 No. 13.4 Software function of PC104 embedded SCMe controllery1)Collecting and managing data by R485controlled2)Cby MODBUSFig 3 Two-dimension fuzzy PID control system3)Realizing visible supervision in real time bygraph configuration5 Analysis of Fuzzy PID Control System4)Managing the alarm such as warning andClosed Looplarm of the system fault and the dangerous resultsAccording to the investigation, the fuzzy PIDcontrolling system is simplified as following4 Construction of Fuzzy PID Control Sys-S: if y(k)is A and athentemy(k+1)=a0+a0 y(k)+.+an y(-n+1)+Fig. 2 schematically shows the typical fuzzyh1u(k)+…+bnk-m+1,PID control system". Its control regulation isd xk-n+1) iswhere:y()=L(k),y(k-1),y(k-n+DTA thenA=A'×A2x…×A,A=A×Ax…×A.Superscript 1 is the Ist rule of fuzzy modenk)=k,e+k∑e+k∴which can be regarded as a sub-system; Superscript jController's output formis the jth rule, which can be regarded as a sub-controllerw u(k)From (4), equivalent closed loop system is∑signed for each sub-system as l=j, andsubsystem and its controller form into the closed loopwhere:W is the subjection function of the first subsystem" one controller controls other subsys-tem, and then they form into the closed loop subsysfuzzy regulation w,=IA[x-i+1); u; is the uni- tem with other system. Because each controller istary form of wi 2u=1Ka, Ku, Ku are respec- ity of system is greatly simplified. Every controllertively coefficient of PId in responding to propormust be verified by its capability. And it must be deintegral and differentialsigned again if it badly performs in controlling procK[12k.-∑kK6 Simulation StudyKM=>u KThe fuzzy model of nonlinear system as follow13].FuzzycontrollercontrolledIf y()is ather中国煤化工)+0603(k)Fig 2 Typical fuzzy PID controlling systenIfCNMHGComparing with general fuzzy controller, thisy(k+1)=14y(k)-048y(k-1)+0.5(k),control system was simplified Fig. 3)where: A"and a"are showed in Fig 4LU Kui et alApplication of Fuzzy-PID Control System in Full-Mechanized Coal FaceFig 4 Affiliated functions of input variablesThe PID parameters of two subsystems are re-Fig 6 Output curve of fuzzy PID controlling systemspectively0.240,4712,3.613and0.706,2608,1.3Fig5 and Fig. 6 indicate that the design of fuzzyThe performance of system is showed in Fig. 5controller is simple, practical and feasible.7 ConclusionThe fuzzy PID control system of No 4326 full-mechanized coal face was investigated, and the software and hardware of system were established. Bymeans of iFIX Dynamics software pictures the flowof controlling the production of coal face was realFig 5 Performance of fuzzy PID controlling systemzed. All technical parameters can be dynamicallyshowed in screen, as well as real-time and historicalThe output curve of closed loop system is trend graphs. The field effect has indicated that thisshowed in Fig. 6system is successful and satisfiedReferences[1] Ma L Intelligent Control and Lonnetwork Development Technology Beijing: Qinghua University Publisher, 2003. (In Chinese)[2] Byungkook AGM. Design of a fuzzy controller using input and ouput mapping. IEEE Trans Fuzzy Syst, 1991, 6(2): 315-321[3] Byuangkook A G M. Adaptive fuzzy sliding mode control of nonlinear system IEEE Trans Fuzzy Syst, 1998, 6(2): 315-321[4] Cao S G, Rees N W, Feng G. Analysis and design for a class of complex control systems(Part ID). Fuzzy Controller Designautomatica,1997,33(6):1029-1039[5] HuY M. Nonlinear Control System Theoty and Its Application. Changsha: National Defence Industry Publisher, 2002. (InChinese)[6] Barada S, Singh H Generating optimal adaptive fuzzy-neural models of dynamic systems with applications to control. IEEETrans Sysf Man Cybern, 1998, 28(3): 371-391[7] Ding Y S Analytical Analysis and Design of Fuzzy Systems with Application to Laser Thermal therapy[Ph D. Dissertation]Shanghai: China Textile University. (In Chinese)3] Byungkook AG M. Adaptive fuzzy sliding mode control of nonlinear system. IEEE Trans. fuzzy Syst, 1998, 6(2): 315-321[9] Cao J F, Han CZ, Fang Y W Nonlinear Control System Theoty and Its Application. Xian: Xian Universuty Publisher, 2001. (InChinese)[10] Pang FS. Fuzzy-Linear Compound Control, Fuzzy Control. Beijing:H中国煤化工iy[ll] LiHX, Gatland G B Conventional fuzzy control and its enhancemerCNMHG[12] Buckley JJ. Universal fuzzy controller. Automatica, 1992 28(60): 124[13] Buckley JJ. Surgeon type controllers are universal approximations. Fuzzy Sets Syst, 1993 53(3): 299-303