Comparative study of electrostatic sensors with circular and probe electrodes for velocity measureme Comparative study of electrostatic sensors with circular and probe electrodes for velocity measureme

Comparative study of electrostatic sensors with circular and probe electrodes for velocity measureme

  • 期刊名字:仪器仪表学报
  • 文件大小:383kb
  • 论文作者:Shao Jiaqing,Krabicka Jan,Yan
  • 作者单位:Department of Electronics
  • 更新时间:2020-06-12
  • 下载次数:
论文简介

第28卷第I1期仪器仪表学报VoL 28 No 112007年11月Chinese journal of scientific InstrumentNov.2007Comparative study of electrostatic sensors with circular and probe electrodesfor velocity measurement of pulverised coalka Jan Yan YoDepartment of Electronics, University of Kent, Canterbury, Kent CT2 7NT, UK)Abstract: This paper presents recent progress on the velocity measurement of pulverised coal in pneumatic pipelinesusing electrostatic sensors in combination with correlation signal processing techniques. A comparative study of elecprostatic sensors with circular and probe electrodes was conducted on a 94 mm bore horizontal pipeline in a 4 Mwfumace. The advantages and limitations of both sensors are discussed. Experimental results demonstrate that bothsensors are capable of providing pulverised coal velocity measurement with excellent repeatability and dynamic re-sponse.Keywords: electrostatic sensor; circular electrode; probe electrode; pulverised coal; flow measurementstudies of both sensors have been reporled for the velocity1 Introductionmeasuremenerybout the performances of both sensors under indIn order to achieve efficient pneumatic conveyance of pul- trial installations, This paper addresses some of the designversed coal in coal fired power plants, it is necessary to issues of the electrostatic sensors. A series of experimentalmeasure the velocity of coal particles in an on line continu- tests is recently conducted on a 4 MW coal fired test facilityous manner Sensors based on a variety of principles such as to study the performances of the circular and probe sensorscapacitance(o, modulation/attenuation of ultrasonic waves The experimental results are reported and discussed in thisor radiation[23, optics[4-51 and electrostatics() have been paperinvestigated. However, only some of them are suitable forthe velocity measurement of pulverised coal under dilute flow 2 Measurement principleconditions typically seen in the power generation industryCapacitance sensors lack sufficient sensitivity for such dilute The movement of particulate materials in a pneumaticflows, which often contain moisture, and systems based pipeline generates a net electrostatic charge on the pon the attenuation or modulation of a signal source are gener- through the interaction of the pipeline and the conveyingally regarded as difficult or expensive to install. As a result, each other. It is believed that the amount of charge on thethe electrostatic sensing approach has attracted a major atten- particles depends on many factors, including physical andtion in the flow measurement community. Since the electro chemical properties of the particles and their surrounding en-atatic sensors do not require an extemal signal source they vironment in the pipeline Although the charge on the parti-are therefore normally robust and relatively inexpensivecles is often unpredictable, the dynamic changes in chargeCircular-shaped electrostatic sensors have been widely can be detected by an insulated electrode in conjunction withtudied largely due to their non-intrusive natuProbet. a random signal is producedbased intrusive sensors have also been investigated re- from the fluctuations in the electric field caused by the pacently for their advantages such as easy installation and cost. sage of the charged particles. Two identical sensors, one be-ffectiveness. Although some progress has been made on the ing positioned downstream of the other, are used to measurese of both circular and probe electrostatic sensors for flow themeasurement of solids in a laboratory environment, many is- illustral中国煤化工ocity measurement.sues remain to be examined. In particular, no comparative SinceCNMHGsors is known, theReceived Date:200708收稿日期:20074081922仪器仪表学报第28卷correlation velocity( V )is derived fromat various locations as shown in fig. 2. The installation ofL(1) such electrodes only requires suitable holes to be drilled toaccommodate the probes. The drawback of such electrodes iswhere TL is the transit time taken by the particles moving that they are intrusive to the flow and subject to wear prob-from the upstream sensor to the downstream sensorlem. As the surface of the probe can be easily coated withanti-abrasive material, the wear problem is no longer a biggnal processorconditioningconditioningUpstreamDownstreamelectrode○QFig. 1 Principle of electrostatic cross-corelationFig. 2 Electrodes used in electrostatic sensorsvelocity measurementIt is known that the electrostatic flow noise is a wide-bandd downstream signals, the velocitysignal. Howehigh frequency components in theurement is independent of the signal amplitude. Thus, ansignal are inevitably lost during the sensing processlectrostatic measurement system using the cross-correlation lectrostatic electrode acts like a low-pass filter).Therefotechnique can be used to monitor a wide rang of flow condi. it is acceptable that the signal conditioning unit has a limitedtions and the system is, in principle, free from the variations bandwidth as long as it is wider than the bandwidth of the e-of the particle properties and their surrounding environmentin the pipeline.A suitable sensor spacing(L)is essential for optimizationof the velocity measurement. In general, it should be small3 Sensor designenough to achieve higher signal similarity. Meanwhile, ex-The sensors are used to detect random fluctuations oftion error of the cross correlation signal processor would bewide-band electrostatic flow noise and convert them into cor-ignificant while the measured time delay(Ti)is too smallresponding electrical signals, which can be processed by aThe axial width ofcross-correlation signal processor. A number of different geo-ter. It should generally be as short as possiblemetrical shapes of electrodes have been used as sensing ele- sensor bandwidth and reduce the uncertainty of the sensorments.Fig. 2 shows the arrangements of circular-shaped andsPainprobe electrodes used in this study. Thnormally embedded in the pipe wallsula. 4 Experimental results and discussiontons and are flush with the rest of pipe wall. This arrange-ment is completely non-intrusive to the flow. Circular elec- 4.1 Test conditionsodes are also able to average the flow streams circumferenTwo electrostatic flow sensors with circular and probe elec-tially and have a relatively uniform sensitivity. Therefore, trodes have been designed and manufactured. In order tohis kind of arrangement has been widely researched. How- vestigate the effects of different sensor designs on the meaever, circular electrodes usually take form of a spool piece urement results, the signal conditioning units for both serinstalled inline with the pipeline. The installation of theof their handwidth and gainspool piece can be expensive in power plants, in particular, trol中国煤化工 de of stainleswhile the diameter ofCN MH Ga spacing of 50 mm aprobe electrodes can be placed around the pipeline perimeter part. They are embedded in the pipe wall via PTFE insula第11期Shao Jiaqing et al: Comparative study of electrostatic sensors with circular and probe electrodesfor velocity measurement of pulverised coaltors and are flush with the pipe wall. The probe electrodes as shown in fig. 3(c)(n. Pressure and temperature sensorsare 4 mm stainless steel rods with an intrusive depth of were also installed at the test section, allowing the volumet47 mm into pipeline and a sensor spacing of 50 mm. Both e- ric flow rate of air(Q) through this section to be evaluatedlectrode assemblies are flanged into the steel pipeline of an assuming ideal gas behaviour. The average velocity of airintemal diameter 94 mm as shown in fig. 3. During the tests within the test section could then be estimated from the su-the sensor with probe electrodes was rotated to locations 1, 2 perficial air velocityis definedand 3( Fig. 2), respectively, to detect localised flow infor-Q(2)where A is the eross sectional area of the pipe and the pres-ence of particles within the test section is ignored. y. wasmaintained around 18 m/s to satisfy the requirements of theRapid variations in Va could only bethe start-up and shutdown of the furnace, V. was updatedevery 15 s. The volumetric concentration of pulverised coal(B)is estimated using the following equation(a)Circular electrostatic sensoβ=QPM8 is found to be in the range of0. 04%-0.05%, indicating that the flow in the pipeline isprimary air has to be fed into the fumace steadily forreasons and facility restrictions during the tests. Thereforeonly limited flow conditions could be obtained(b)Probe electostatic sensorFig 4 Outline of the test facilityBoth circular and probe sensors gave consistent reading(e) OverviewFig 3 Sensor installations on a 4MW fumacethroughout the tests. Clearly defined correlogram peaks wereproduced by both systems. Fig. 5 shows typical correlogramsfor the circular and probe sensors, respectively, togetherFig. 4 illustrates a simplified diagram of the test facility. with their waveforms of the electrostatic signals. The correla-The electrostatic sensors were installed on the horizontal sec-tion of the pipeline approximately 7 m away from the outlet of tion coelficient was around 0. 2-0. 3 for the circular sensormill that is used to produce pulverised coal from raw coaland 0.6-0 7 for the probe sensor regardless of the installa-The particle size distribution of the pulverised coal (90%

论文截图
版权:如无特殊注明,文章转载自网络,侵权请联系cnmhg168#163.com删除!文件均为网友上传,仅供研究和学习使用,务必24小时内删除。