CN1912588A - Coal on-line analyse equipment based on laser induced spectral and nerve network technology - Google Patents
Coal on-line analyse equipment based on laser induced spectral and nerve network technology Download PDFInfo
- Publication number
- CN1912588A CN1912588A CN 200610121134 CN200610121134A CN1912588A CN 1912588 A CN1912588 A CN 1912588A CN 200610121134 CN200610121134 CN 200610121134 CN 200610121134 A CN200610121134 A CN 200610121134A CN 1912588 A CN1912588 A CN 1912588A
- Authority
- CN
- China
- Prior art keywords
- coal
- analysis
- neural network
- laser
- powder
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
An on-line analyzing and detecting device of coal quality is prepared as shining sample to be analyzed by laser to generate induced light spectrum; collecting said spectrum by CCD image transducer and image collection control circuit and calculating to obtain vector E of character element content in raw coal by signal processor; sending vector E into BP neural network for carrying out conversion to obtain heating value, water content, ash content, volatile composition and fixed carbon of raw coal; utilizing neural network to predict slagging index and temperature of ash melted point of raw coal.
Description
One, technical field
The invention belongs to automatic measurement technology field and artificial intelligence technology application, is the instrument and the system of the on-line measurement of a kind of ultimate analysis that is used for the industrial processes ature of coal and technical analysis index specifically.
Two, background technology
Along with development and national economy, the sustainable growth of raw coal consumption, over the past two years, situation in short supply had appearred in national coal resources, and especially electric power is in an emergency with coal comprehensively, and part power plant has been absorbed in and has shut down the awkward condition for the treatment of coal; Simultaneously, because electric coal supply is the marketization, diversification day by day, cause the ature of coal fluctuating range to increase, coal is mixed, is gone into stove ature of coal control difficulty and strengthens, make boiler fired coal off-design coal, boiler smooth combustion is damaged, and the equipment deficiency showed increased of initiation comprises that the ratio that faults such as boiler scaling, fire extinguishing, heating surface overtemperature tube burst account for the unplanned stoppage in transit of unit obviously rises.Industries such as other industry such as coking, chemical industry, metallurgy also have similar problem, cause decrease in efficiency, and fault increases, and has influenced the ordinary production of enterprise.In this case, the production run personnel will in time grasp the coal analysis situation into the stove coal, particularly go into fugitive constituent, thermal value, the content of ash and the melting characteristic of coal ash of stove coal, so that accomplish to shoot the arrow at the target when the control boiler operatiopn.
Traditional coal analysis all adopts the artificial sample sample preparation, utilize Laboratory Instruments that it is analyzed, analysis speed is slow like this, the analytical cycle of a collection of coal sample is 6～8h, the coal sample on most of power plant same day just can go out examining report in second day, can not satisfy the needs of boiler combustion adjustment and crash analysis far away.Because the restriction of detection means, it is more and more outstanding to make actual needs and detection data quote the contradiction of time lag.
In ature of coal on-line monitoring instrument, chemical technology and nuclear technology have generally been adopted both at home and abroad, United States Patent (USP) U.S.Pat.No.4,562,044 has described a kind of method that adopts chemico-analytic method that raw coal is analyzed, this method adopts four mechanical arms to carry four sampling cups, utilizes motor-driven to take a sample and sampling cup is placed the position of chemical examination, and the method for analysis adopts chemical analysis to carry out.United States Patent (USP) U.S.Pat.No.4,841,153 adopt the radioactive source gamma rays that coal sample is shone analysis, the security protection means and the measure of such system requirements strictness make the bulky of equipment, cost an arm and a leg, the user scarcely is ready to accept for the nuclear feared state of mind.Use the method for ash content in the nuclear technology detection coal and conclude following 4 kinds: the gamma ray scattering method of dual energy gamma ray through transmission technique, 60keV, electron pair method and neutron activation analysis method.Using more is dual energy gamma ray through transmission technique and neutron activation analysis method.The neutron activation analysis ratio juris is to utilize thermal neutron to excite in the tested coal sample each atoms of elements nuclear, and the C ray energy spectrum that sends when measuring the atomic nucleus transition of these excited state can obtain the content of each element.The major product of neutron activation analysis ash measurer is the Model 3612C type ash measurer of U.S. GammaMetrics company and the COALSCAN9000 type ash measurer of Australian Scantech company.The neutron source of utilizing of U.S. GammaMetrics company production is measured the online survey coal of the 1218 types instrument of multiple coal index in addition, can directly measure sulphur, ash content, carbon, hydrogen, nitrogen, fluorine, silicon and moisture etc., and can measure thermal value and sulphuric dioxide etc. indirectly.
The laser-induced spectrum technology shines on the measuring object after utilizing a branch of intense pulse laser to focus on, and the measuring object ionization of focus point produces high temperature, highdensity plasma.In such high temperature system, all materials can be evaporated into molecule or atom, fierceness collision in the high temperature system between the particle makes molecule or atomic ionization become ion again, and molecule, atom or ion can population on each energy level, high level makes laser plasma produce very strong spectrum to the low-lying level transition.The light that plasma sends removes the light of environment light source and excitation source through filter filtering.Through monochromator splitting, make mixed light become the monochromatic light of arranging by wavelength then, at the light-emitting window of monochromator highly sensitive photodetector is installed and is detected.Pulse producer and time delay generator constitute a time schedule controller, control effectively that laser pulse sends and optical signal detecting between time delay, thereby reach the continuous background light that effective removal plasma sends, the purpose of differentiating the characteristic spectral line of atom.At last, draw the element kind of being analyzed according to the spectral signature line wavelength, by calibration, the intensity of corresponding spectral line is just represented the concentration of analytical element, so the laser-induced spectrum technology can be used for highly sensitive element detection.
The laser-induced spectrum technology has obtained certain progress at aspects such as testing laboratory's ultimate analyses since proposing the seventies in last century, but in the application of industry spot, still there are some problems, research work personnel have launched the practical research of laser-induced spectrum technology, for example, the patent CA2299046 that Canada PHARMALASER INC company declares has introduced the kind of utilizing LIBS technology for detection timber, compares with the data in being stored in computing machine by the spectral line that obtains to be correlated with and determines the kind of timber.And the other a patent of the said firm also is to adopt this technology to be used for detecting the composition of tablet, relates generally to the automatic transmission and the measurement of multiple medicines sample.United States Patent (USP) U.S.Pat.No.6,532, the pulse laser of two varying strengths of 068 employing is used for having realized applying the substance classes measurement of skin-material, the laser beam of the first bundle broad is used to puncture the top layer, the hole is ablated on the top layer, measured matter is come out, utilize the narrower laser beam of second bundle to produce the laser spectrum of inducting then, compose to determine the character of measured matter by the laser of inducting, this method can be used for the depth of material analysis.United States Patent (USP) U.S.Pat.No.6,657,721 provide a kind of elements are contained technology that need not the laser-induced spectrum of reference curve calibration, this technology is mainly based on the saha-Boltzmann equation, utilize the Physical Electronics principle to calculate, and need not reference sample and do not need to calibrate the quantitative analysis that just can realize constituent content.United States Patent (USP) U.S.Pat.No.6,771,368 have proposed a kind of method, the fundamental purpose of this method is the phase mutual interference that reduces between the spectral line, adopts three detecting devices, detects the part of fluorescence spectrum respectively, after data enter computing machine, analyze the constituent content that obtains measured matter.Chinese invention patent CN1480722A discloses a kind of laser-induced spectrum coal quality analyzer, this instrument comprises sampling part and analysis part, utilize cyclone separator to collect coal dust, put into quartzy analysis cell, then quartzy analysis cell is fixed on the fixed station of motor driven, utilizes the laser-induced spectrum technology to carry out the ultimate analysis of coal dust.
Carry out in the process of online adjustment the fuel-burning power plant, that the result of coal analysis is needed is the result of technical analysis, such as: thermal value, volatile matter and the ash index of grading, and the result that the laser-induced spectrum coal quality elemental analyser obtains is the content of each element in the coal, can't be directly used in and instruct burning adjustment, need be converted into the result of technical analysis.Because coal is a kind of very complicated material, is a kind of potpourri, is to be mixed by various organism and inorganics, can not be explained by single simple substance compound.Concern complexity between the coal analysis data, classic method is the contact that utilizes between the linear regression method research coal analysis characteristic index of mathematical statistics, obtain the binary contact between the coal index in the different mining areas of different coals, these associations occur with the experimental formula form, can reflect the contact between the coal characteristic index, but this contact is essence, accurate inadequately inadequately, and need set up different formula for the coal of different regions, consequently experimental formula is many, be difficult to unified, and the result calculated error is bigger, is not easy to promote the use of.Under the situation of not knowing former coal district in advance, the resultant error that adopts a certain formula to calculate can be very big, and this method is difficult to be applicable to different coals, can not satisfy and produce actual needs.Therefore need set up the relation that nonlinear model characterizes their complexity, can realize the accurate transformation of ultimate analysis, and be applicable to the measurement of the raw coal ature of coal of various places to technical analysis.
In addition the melt temperature of ash content also is a main parameter in the raw coal, but does not also have effectively this parameter of On-line Estimation of method at present, and the safe and stable operation of it and boiler is closely bound up.Ash content in the coal has three kinds of forms after burning, the one, keep solid-state, discharge from flue with the flying dust form, the 2nd, at high temperature evaporate into gaseous state, on colder water-cooling wall, superheater or reheater tube face, condense then, and combine with flying dust and to be deposited on the pipe, form the contamination or the dust stratification of heating surface.Also having a kind of is to be fused into liquid state, be bonded on the furnace heating surface then, form coke button through progressively depositing, boiler heating surface dust stratification and slagging scorification have very big negative effect to the efficient and the safe and stable operation of boiler, because the coefficient of heat conductivity of ash is very low, thermal resistance is very big, and thermal insulation is very strong, can influence the heat-transfer capability of heating surface, the heat-transfer capability of general dust stratification water-cooling wall after a few hours can reduce by 30%～60%, move after making the furnace flame center, the furnace outlet temperature raises, and boiler is moved away from design load, do not reach accordingly and exert oneself, cause the boiler av eff to reduce by 1～2%, increase coal consumption, the waste resource.Under the effect of high-temperature flue gas, complicated chemical reaction takes place with tube wall in the dust stratification of heating surface or slagging scorification meeting, form high temperature corrosion, under the slagging scorification serious situation, the coke button whereabouts can injure water-cooling wall by a crashing object, causes the accident, or is heated irregular because of slagging scorification district and non-slagging scorification district, cause the boiler water wall booster, cause being forced to blowing out.
The slagging scorification of boiler heating surface, contamination, dust stratification, not only relevant with the composition and the fusing point of coal and coal ash, and with the boiler design relating to parameters, under the situation that design parameter is determined, if can know the relevant composition of coal and the melt temperature point and the slagging index of ash in advance, just can be by adjusting the operational factor of boiler, change the boiler operatiopn operating mode, or the feed coal of the different ash deposition propensities of choose reasonable can make the ash deposition propensity of coal blending in being controlled in the light or slight partially scope, thereby the slagging scorification situation of boiler is reduced to minimum, boiler is used safely and life-saving.
Problems of the prior art, sampling system is difficult to adapt to different industrial environments, for example in adopting Transport Machinery course of conveying such as belt, the conveying of pipeline wind-force, automobile, need change sampler, could satisfy the requirement of analyzing automatically according to different conveying machineries.Adopted optical lens and catoptron in the laser spectral analysis relevant device, the slight variation meeting of optical lens, reflector position produces considerable influence to measurement result under the situation that has vibration, thereby influences the normal use of equipment.In addition, because in the industrial processes, that the operations staff generally is concerned about is the technical analysis result of raw coal, because changing feasible the conversion to the technical analysis result by ultimate analysis, coal is difficult to adopt unified formula to be expressed, existing technical scheme does not provide the needed technical analysis result of operations staff, has limited the use of this kind equipment.
Three, summary of the invention
The technical problem to be solved in the present invention provides the online check and analysis equipment of a kind of ature of coal, overcome the defective that prior art exists, especially analyse uncertainty in the transfer process as a result by ultimate analysis to industry under the coal situation of change, belong to automatic measurement technology and artificial intelligence technology application, be used for solving the difficult problem of the on-line measurement of the ultimate analysis of industrial processes ature of coal and technical analysis.
The online check and analysis equipment of a kind of ature of coal based on laser-induced spectrum technology and artificial neural network technology of the present invention is characterized in that: comprise negative pressure sampling assemble 1, analyze case 2, spectra collection analytic unit 5, signals collecting control module 3, based on the signal processor 6 of artificial neural network; This equipment is characterised in that:
Negative pressure sampling assemble 1, (26A～26F), be used to fetch the coal dust sample is positioned over it in analysis case 2 to comprise negative pressure generator 8 and sampling line;
Spectra collection analytic unit 5, comprise pulse laser source component 44, image acquisition control circuit 52 and ccd image sensor 50, image acquisition control circuit 52 sends signal to pulse laser source component 44 under the control of signal processor 6, make pulse laser source component 44 send laser pulse, on the coal dust sample of vertical irradiation in Measurement and analysis pond 31, inspire the spectral signal of inducting, the spectrum picture signal of 52 pairs of ccd image sensors 50 of image acquisition control circuit is gathered simultaneously, and the signal of collection is transferred to signal processor 6;
Signal processor 6 based on artificial neural network, be used for the information via data processing of laser-induced spectrum is obtained coal dust sample carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, the results of elemental analyses of copper, and results of elemental analyses is input to neural network A in the signal processor 6, convert former calorific value of coal to by neural network A, moisture, ash content, fugitive constituent, fixed carbon technical analysis result, with resulting carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, the results of elemental analyses of copper and thermal value, moisture, ash content, fugitive constituent, fixed carbon technical analysis result is input to the neural network B in the signal processor 6, convert the melt temperature point and the slagging index of coal ash to, with above-mentioned results of elemental analyses, the technical analysis result, coal ash melting temperature point and slagging index pass through remote signal transmission interface 7 to outside device transmission.
In order better to realize also needing the present invention the data of laser-induced spectrum are carried out suitable processing, the present invention uses two artificial neural networks, and the step that data are handled comprises following three steps:
The first step: the element that obtains analyzed former coal sample by laser-induced spectrum.
Under the situation that measuring condition is determined, the content of each element has definite quantitative relationship with the intensity of corresponding spectral line in the raw coal, after eliminating the influence of laser power to the spectrum of inducting, by spectral line is demarcated, we can obtain, and element in the raw coal is formed and the content of each element.The present invention obtains the content of following element by the analysis to the spectrum of inducting: carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper.
Second step: utilize artificial neural network that results of elemental analyses is converted to the technical analysis result.
Described in this instructions technical background,, need set up nonlinear model because the complicacy of raw coal composition and chemical constitution utilizes the experimental formula of linear model can't realize the accurate transformation of ultimate analysis to technical analysis.Artificial neural network has memory capability, pattern classification ability, BP neural network more than three layers and RBF artificial neural network have the approximation capability to arbitrary function, neural network after the process training can realize memory, the classification to the raw coal feature, and the result of ultimate analysis can be converted to the result of technical analysis.The present invention utilizes the BP neural network to realize the conversion of results of elemental analyses to technical analysis.The constituent content that the first step is calculated is formed the input vector E={E of neural network
C, E
H, E
O2, E
N, E
S, E
Si, E
AL, E
Ca, E
Mg, E
K, E
Na, E
Ti, E
Fe, E
Mn, E
P, E
Cu, corresponding respectively: carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper.Realize that results of elemental analyses is designated as neural network A to the neural network of technical analysis result conversion, the structure of neural network A is: input layer has 16 unit, respectively corresponding carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper.The number of unit of hidden layer is greater than 50, and the unit number of output layer is 5, and is corresponding respectively: thermal value, moisture, ash content, fugitive constituent, fixed carbon.Detailed process is:
1 obtains the raw coal analyzing samples collection of some, each sample comprises that raw coal is through the result after the ultimate analysis, contain: the content of carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper, sample also comprises the technical analysis result, contains the numerical value of thermal value, moisture, ash content, fugitive constituent, fixed carbon.
2 data normalizations with sample set are transformed into the scope of the input signal that neural network can accept.
3 utilize the data after the normalization that neural network is trained, and make the output of neural network enough little with the quadratic sum of the technical analysis result's of reality error, set up neural network model, preserve the weights of neural network.
4 models are used, and the data of the raw coal coal analysis that measures are formed the input vector of neural network, and carry out normalization, are input to neural network, and the output of neural network is converted to needed result, can obtain the technical analysis result of raw coal.
The 3rd step: the fusion point temperature and the slagging index that utilize ash content in the neural network prediction coal.
The melt temperature point of coal ash comprises deformation temperature DT, softening temperature ST, hemispherical temperature HT and flow temperature FT, depend primarily on the chemical composition of raw coal, the size of slagging index and coal ash melting temperature, components of coal ash, coal ash content size and volatile matter has certain relation, the data processing first step and resulting result of second step according to the present invention can utilize the melt temperature point of ash content in the slagging index of neural network prediction raw coal and the coal.The result of the constituent content that the first step is calculated and second resulting ash, thermal value, volatile matter forms the input vector E={E of neural network
C, E
H, E
O2, E
N, E
S, E
Si, E
AL, E
Ca, E
Mg, E
K, E
Na, E
Ti, E
Fe, E
Mn, E
P, E
Cu, E
Ash, E
Heat, E
Daf, corresponding respectively: carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper, ash content, thermal value, fugitive constituent.Realize that the neural network of returning powder fusion point temperature and slagging index prediction in the coal is designated as neural network B, the structure of neural network B is: input layer number of unit 19, the hidden layer number of unit is greater than 68, the output layer unit number is 5, respectively corresponding deformation temperature DT, softening temperature ST, hemispherical temperature HT, flow temperature FT and slagging index.Detailed process is:
1 obtains the raw coal analyzing samples collection of some, each sample comprises that raw coal is through the result after the ultimate analysis, contain: the content of carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper, sample also comprises the technical analysis result, contains the numerical value of thermal value, ash content, fugitive constituent, coal ash deformation temperature DT, softening temperature ST, hemispherical temperature HT, flow temperature FT and slagging index.
2 data normalizations with sample set are transformed into the scope of the input signal that neural network can accept.
3 utilize the data after the normalization that neural network is trained, and the output that makes neural network is enough little with the quadratic sum of the error of actual coal ash melting temperature point and slagging index, sets up neural network model, the weights of preservation neural network.
4 models are used, the input vector of the data of the raw coal coal analysis that measures being formed neural network, and carry out normalization, be input to neural network, the output of neural network is converted to needed result, can obtains predicting the outcome of coal ash deformation temperature DT, softening temperature ST, hemispherical temperature HT, flow temperature FT and slagging index.
The present invention compared with prior art has following advantage and beneficial effect:
1. adopt the negative pressure sampling system, go for belt transportation sampling, pipeline strength conveying sampling, haulage vehicle sampling, equipment is realized automation mechanized operation fully, need not human intervention.
2. adopt neural network that measurement data is handled, utilize the non-linear mapping capability of neural network, solved the difficult problem of ultimate analysis to technical analysis result conversion, the conversion accuracy height has two kinds of outputs of results of elemental analyses and technical analysis result.
3. adopt neural network that coal ash melting temperature point and raw coal slagging index are predicted, help the operations staff in time to adjust combustion system, the thermal efficiency of avoiding boiler scaling to cause reduces and accident.
4. the equipment optical system is simple, is not afraid of vibration, more can adapt to the rugged surroundings of industry spot.
5. adopt nitrogen to set up the single-measurement gaseous environment, reduce the ground unrest influence, protect laser instrument and photosensitive optical fiber simultaneously, utilize the low temperature module to eliminate the influence of thermonoise, make measurement result stable, the adaptability of environment is strengthened ccd image sensor.
Four, description of drawings
Fig. 1 is the technical solution of the present invention block diagram;
Fig. 2 A is a principle of the invention structural drawing, and Fig. 2 B is the cross sectional arrangement figure that analyzes case;
Fig. 3 is the graph of a relation of required collection of signals collecting control module of the present invention and control signal;
Fig. 4 is the workflow diagram of touring 6 passages of check and analysis of the present invention;
Fig. 5 is the present invention carries out check and analysis to single passage a process flow diagram;
Fig. 6 is that the present invention rotates back up pad and Measurement and analysis pond working position figure;
Fig. 7 is flow chart of data processing figure of the present invention;
Fig. 8 is embodiment of the present invention 1 synoptic diagram;
Fig. 9 is embodiment of the present invention 2 synoptic diagram;
Figure 10 is embodiment of the present invention 3 synoptic diagram;
Figure 11 is a sampling head structural drawing of the present invention;
Figure 12 is a sampling head structure upward view of the present invention.
Number in the figure
1, negative pressure sampling assemble 2, analysis case
3, signals collecting control module 4, protection assembly
5, spectra collection analytic unit 6, signal processor
7, remote signal transmission interface 8, negative pressure generator
9, dust-precipitator 10, time powder processing components
11, row's powder valve 12, row's tube cell
13, put tube cell 14, blow down valve
15, put powder valve 16, following coal powder silo with level meter sensor
17, Vib. 18, last coal powder silo with level meter sensor
19, whirlwind powder collector 20, negative pressure control valve
21, exhaust pipe 22, the female pipe of sampling
23, abrasion-proof bent tube A～abrasion-proof bent tube F 24, sampling operation valve A～sampling operation valve F
25, blowback operation valve A～blowback operation valve F 26, sampling line A～sampling line F
27, analyze casing 28, put the powder position transducer
29, turning axle 30, rotation back up pad
31, Measurement and analysis pond 32, coal dust scraper plate
33, protection barrier driving motor 34, protection chamber
35, focus lamp 36, protection baffle plate
37, cross-brace beam 38, measuring position sensor
39, control motor 40, filtering pressure reducer
41, cleaning operation valve 42, nitrogen gas generator
43, protection gas control valve 44, pulse laser source component
45, protection tube 46, polychromator
47, photosensitive optical fiber 48, fiber optic protection sleeve pipe
49, Transmission Fibers 50, ccd image sensor
51, low temperature module 52, image acquisition control circuit
53, target spot 54, duff pipe
55, loudspeaker sampling head 56, band conveyor
Five, embodiment
Below in conjunction with drawings and Examples, the present invention is described further, but embodiments of the present invention are not limited only to this.
As shown in Figure 1, the online check and analysis equipment of a kind of ature of coal of the present invention comprises negative pressure sampling assemble 1; analyze case 2, signals collecting control module 3, protection assembly 4; spectra collection analytic unit 5, remote signal transmission interface 7 and based on the signal processor 6 of artificial neural network.
Negative pressure sampling assemble 1 comprises negative pressure generator 8, dust-precipitator 9, return powder processing components 10, row's powder valve 11, blow down valve 14, put powder valve 15, Vib. 17, following coal powder silo with level meter sensor 16, last coal powder silo with level meter sensor 18, whirlwind powder collector 19, negative pressure control valve 20, exhaust pipe 21, the female pipe 22 of sampling, abrasion-proof bent tube 23A, 23B, 23C, 23D, 23E, 23F, sampling operation valve 24A, 24B, 24C, 24D, 24E, 24F, blowback operation valve 25A, 25B, 25C, 25D, 25E, 25F and corresponding sampling line 26A, 26B, 26C, 26D, 26E, 26F.Negative pressure sampling assemble 1 links to each other with analysis case 2 with row's tube cell 12 by putting tube cell 13, and Vib. 17 is used for avoiding coal dust to be flocked in the pipeline to putting 15 vibrations of powder valve.Under the situation of putting powder valve 15 and negative pressure control valve 20 shutoffs, the row's of opening powder valve 11, the coal dust of being controlled after pressurized air are finished measurement by blow down valve 14 blows out in Measurement and analysis pond 31, through dust-precipitator 9 coal powder collection that measures is entered back powder processing components 10 then.Each sampling line 26A, 26B, 26C, 26D, 26E, 26F finally can be connected to the female pipe 22 of sampling, because the flow velocity of breeze airflow is higher in the negative pressure sampling system, tube wall is produced bigger wearing and tearing, in order to reduce the resistance of pipeline in the negative pressure sampling system, adopt abrasion-proof bent tube 23A, 23B, 23C, 23D, 23E, 23F to connect at the meeting place of connecing.
Analyze case 2 and comprise analysis casing 27; put powder position transducer 28; measuring position sensor 38; cross-brace beam 37; turning axle 29; rotation back up pad 30; Measurement and analysis pond 31; coal dust scraper plate 32; protection barrier driving motor 33; protection chamber 34; protection baffle plate 36; focus lamp 35; control motor 39; wherein control the below that motor 39 is fixed on cross-brace beam 37; link to each other with rotation back up pad 30 by turning axle 29; driven rotary back up pad 30 is rotated; Measurement and analysis pond 31 is fixed on the rotation back up pad 30; under the drive of motor, can carry out circular motion; put powder position transducer 28 and measuring position sensor 38 is fixed on the cross-brace beam 37, be used for the position of rotation back up pad 30 is detected.The effect of protection chamber 34 is when coal dust being placed on Measurement and analysis pond 31 and coal dust blown out in Measurement and analysis pond 31, provides protective effect to focus lamp 35, photosensitive optical fiber 47, prevents that dust from entering.Protection barrier driving motor 33 is loaded on protection 34 outsides, chamber; drive protection baffle plate 36 and rotate, put powder and with coal dust from the process that Measurement and analysis pond 31 blows out, protection baffle plate 36 will protect chamber 34 to close; protection baffle plate 36 is in 36 ' position during measurement, makes laser can shine sample.
Spectra collection analytic unit 5 comprises pulse laser source component 44, protection tube 45, polychromator 46, photosensitive optical fiber 47, fiber optic protection sleeve pipe 48, Transmission Fibers 49, ccd image sensor 50, low temperature module 51, image acquisition control circuit 52, wherein, pulse laser source component 44 links to each other with image acquisition control circuit 52, protection tube 45 is installed on the below of pulse laser source component 44, this protection tube 45 is paths of pulsed laser signal, prevents that operating personnel from being accidentally injured by superlaser; The analyzed sample of pulse laser vertical incidence of pulse laser source component 44 emissions does not adopt catoptron in the laser signal transmission process, reduced the complicacy of optical system; The head of photosensitive optical fiber 47 is aimed at the laser focusing point, photosensitive optical fiber 47 is loaded in the protection fiber optic protection sleeve pipe 48, and link to each other with polychromator 46, the output signal of polychromator 46 is connected to ccd image sensor 50 through Transmission Fibers 49, and ccd image sensor 50 links to each other with image acquisition control circuit 52; Low temperature module 51 is installed on around the ccd image sensor 50; Image acquisition control circuit 52 links to each other with signal processor 6, the control of acknowledge(ment) signal processor 6, when signal processor 6 after image acquisition control circuit 52 sends instruction, send signal by image acquisition control circuit 52 to pulse laser source component 44, make pulse laser source component 44 send pulse laser, the irradiation sample, the generation spectrum of inducting, image acquisition control circuit 52 is being gathered the spectrum picture signal of ccd image sensor 50 through the small back of delaying simultaneously, convert digital image sequence to, be sent to signal processor 6; Because 50 pairs of thermonoises of ccd image sensor are relatively more responsive, for eliminating the influence of thermonoise, adopt low temperature module 51 to provide the cryogenic thermostat environment for ccd image sensor 50 to measuring results, the temperature range of low temperature is-10 ℃～-25 ℃; Pulse laser source component 44 has the laser energy output signal, and the electric signal of laser energy is represented in output when sending laser pulse, and this signal is gathered by signal acquisition module 33.
Signals collecting control module 3 and signal processor 6, pulse laser source component 44, last coal powder silo with level meter sensor 18, following coal powder silo with level meter sensor 16, put powder position transducer 28, measuring position sensor 38, blow down valve 14, negative pressure control valve 20, row's powder valve 11, put powder valve 15, Vib. 17, protection gas control valve 43, cleaning operation valve 41, sampling operation valve 24A, 24B, 24C, 24D, 24E, 24F, blowback operation valve 25A, 25B, 25C, 25D, 25E, 25F, control motor 39, protection barrier driving motor 33 links to each other; Wherein, the laser power signal of pulse laser source component 44 output and last coal powder silo with level meter sensor 18, coal powder silo with level meter sensor 16, the signal of putting powder position transducer 28,38 outputs of measuring position sensor all insert signals collecting control module 3 down, are gathered by signals collecting control module 3; Blow down valve 14, negative pressure control valve 20, row's powder valve 11, put powder valve 15, Vib. 17, protection gas control valve 43, cleaning operation valve 41, sampling operation valve 24A, 24B, 24C, 24D, 24E, 24F, blowback operation valve 25A, 25B, 25C, 25D, 25E, 25F, control motor 39, protection barrier driving motor 33 by 3 controls of signals collecting control module.
Remote signal transmission interface 7 links to each other with signal processor 6, and the control of acknowledge(ment) signal processor 6 transmits results of elemental analyses, technical analysis result and coal ash melting temperature point and slagging index to miscellaneous equipment.
A kind of ature of coal on-line analysis equipment of the present invention has six negative pressure sampling passages, is respectively passage A, channel B, channel C, passage D, passage E, passage F, can go the rounds sampling and detection to six sampling passages, and as shown in Figure 4, its course of work is:
Before starting working, to take a sample operation valve 24A, 24B, 24C, 24D, 24E, 24F and blowback operation valve 25A, 25B, 25C, 25D, 25E, 25F is in closed condition, then passage A is taken a sample and detect, open sampling operation valve 24A, operate according to single passage sample analysis subprocess then, specifically as shown in Figure 5.In single passage sample analysis subprocess, at first close and put powder valve 15 and row's powder valve 11, open negative pressure control valve 20 and set up negative pressure sampling path, then start negative pressure generator 8, coal dust is drawn onto whirlwind powder collector 19 under the effect of the suction that negative pressure produces, fall the pipeline of whirlwind powder collector 19 belows then, after last coal powder silo with level meter sensor 18 sends signal, the coal dust that expression is fetched has satisfied analyzes required quantity, close negative pressure control valve 20 and stop negative pressure generator 8 this moment, then opens and put powder valve 15, Vibration on Start-up device 17, make coal dust fall into the Measurement and analysis pond 31 of below, instantly coal powder silo with level meter sensor 16 is output as at 0 o'clock, and the coal dust that expression is fetched has been put, and all drops in the Measurement and analysis pond 31, close then and put powder valve 15, stop Vib. 17.Position such as Fig. 6 A in rotation back up pad 30 and Measurement and analysis pond 31.Start-up control motor 39 driven rotary back up pads 30 are to right rotation; when forwarding the position of Fig. 6 B to; through coal dust scraper plate 32; the coal dust of 31 tops, Measurement and analysis pond is calibrated; when forwarding the position of Fig. 6 C to; measuring position sensor 38 output signals 1; stop to control motor 39; open protection baffle plate 36; carry out the laser-induced spectroscopic analysis first time; give an order to image acquisition control circuit 52 by signal processor 6 based on neural network; send laser pulse by image acquisition control circuit 52 gating pulse lasing light emitter assemblies 44, laser pulses irradiate produces the spectrum of inducting at target spot 53 places; laser-induced spectrum is gathered, and the laser-induced spectrum that collects is stored.Control motor 39 driven rotary back up pads 30 then to right rotation 15 degree, arrive the position of Fig. 6 D, carry out the laser-induced spectroscopic analysis second time, the spectrum of inducting that collects is stored.Control motor 39 driven rotary back up pads 30 then again to right rotation 15 degree, arrive the position of Fig. 6 E, carry out laser-induced spectroscopic analysis for the third time, the spectrum of inducting that collects is stored.The spectroscopic data of inducting that collects for three times is handled the ultimate analysis and the technical analysis result that can obtain the ature of coal that the present invention will obtain.At this moment; close protection baffle plate 36; start-up control motor 39 when rotating to the position of Fig. 6 A, is put powder position transducer 28 output signals 1 to right rotation; stop to control motor 39; the row's of opening powder valve 11 and blow down valve 14 utilize the coal dust after pressurized air finishes Measurement and analysis to blow out from Measurement and analysis pond 31, delay time after 2 seconds; shut-down purge valve 14; start negative pressure generator 8, the coal dust that blows out is drawn onto dust-precipitator 9, be discharged to back in the powder processor 37; delay time and close negative pressure generator 8 after 10 seconds; so far, to the sample analysis end of a passage, switch the sampling passage then; repeat said process, can realize touring processing six passages.
In order better to realize also needing the present invention the data of laser-induced spectrum are carried out suitable processing, the present invention uses two artificial neural networks, and the step that data are handled comprises following three steps:
The first step: the element that obtains analyzed former coal sample by laser-induced spectrum.
Under the situation that measuring condition is determined, the content of each element has definite quantitative relationship with the intensity of corresponding spectral line in the raw coal, after eliminating the influence of laser power to the spectrum of inducting, by spectral line is demarcated, we can obtain, and element in the raw coal is formed and the content of each element.The present invention obtains the content of following element by the analysis to the spectrum of inducting: carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper.
Second step: utilize artificial neural network that results of elemental analyses is converted to the technical analysis result.
The present invention utilizes the BP neural network to realize the conversion of results of elemental analyses to technical analysis.The constituent content that the first step is calculated is formed the input vector E={E of neural network
C, E
H, E
O2, E
N, E
S, E
Si, E
AL, E
Ca, E
Mg, E
K, E
Na, E
Ti, E
Fe, E
Mn, E
P, E
Cu, corresponding respectively: carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper.Realize that results of elemental analyses is designated as neural network A to the neural network of technical analysis result conversion, the structure of neural network A is: input layer has 16 unit, respectively corresponding carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper.The number of unit of hidden layer is greater than 50, and the unit number of output layer is 5, and is corresponding respectively: thermal value, moisture, ash content, fugitive constituent, fixed carbon.Detailed process is:
1 obtains the raw coal analyzing samples collection of some, each sample comprises that raw coal is through the result after the ultimate analysis, contain: the content of carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper, sample also comprises the technical analysis result, contains the numerical value of thermal value, moisture, ash content, fugitive constituent, fixed carbon.
2 data normalizations with sample set are transformed into the scope of the input signal that neural network can accept.
3 utilize the data after the normalization that neural network is trained, and make the output of neural network enough little with the quadratic sum of the technical analysis result's of reality error, set up neural network model, preserve the weights of neural network.
4 models are used, and the data of the raw coal coal analysis that measures are formed the input vector of neural network, and carry out normalization, are input to neural network, and the output of neural network is converted to needed result, can obtain the technical analysis result of raw coal.
The 3rd step: the fusion point temperature and the slagging index that utilize ash content in the neural network prediction coal.
The result of the constituent content that the first step is calculated and second resulting ash, thermal value, volatile matter forms the input vector E={E of neural network
C, E
H, E
O2, E
N, E
S, E
Si, E
AL, E
Ca, E
Mg, E
K, E
Na, E
Ti, E
Fe, E
Mn, E
P, E
Cu, E
Ash, E
Heat, E
Daf, corresponding respectively: carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper, ash content, thermal value, fugitive constituent.Realize that the neural network of returning powder fusion point temperature and slagging index prediction in the coal is designated as neural network B, the structure of neural network B is: input layer number of unit 19, the hidden layer number of unit is greater than 68, the output layer unit number is 5, respectively corresponding deformation temperature DT, softening temperature ST, hemispherical temperature HT, flow temperature FT and slagging index.Detailed process is:
1 obtains the raw coal analyzing samples collection of some, each sample comprises that raw coal is through the result after the ultimate analysis, contain: the content of carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper, sample also comprises the technical analysis result, contains the numerical value of thermal value, ash content, fugitive constituent, coal ash deformation temperature DT, softening temperature ST, hemispherical temperature HT, flow temperature FT and slagging index.
2 data normalizations with sample set are transformed into the scope of the input signal that neural network can accept.
3 utilize the data after the normalization that neural network is trained, and the output that makes neural network is enough little with the quadratic sum of the error of actual coal ash melting temperature point and slagging index, sets up neural network model, the weights of preservation neural network.
4 models are used, the input vector of the data of the raw coal coal analysis that measures being formed neural network, and carry out normalization, be input to neural network, the output of neural network is converted to needed result, can obtains predicting the outcome of coal ash deformation temperature DT, softening temperature ST, hemispherical temperature HT, flow temperature FT and slagging index.
Online detection of a kind of ature of coal of the present invention and analytical equipment can be used for boiler and go into online detection of stove coal ature of coal and analysis, as shown in Figure 8, general power plant has four to six coal pulverizers, the coal dust that coal pulverizer grinds is sent in the boiler under the effect of wind-force by duff pipe 54 and burns, because every pairing run coal bin difference of coal pulverizer, and the raw coal in the run coal bin also might be different, therefore need carry out online detection and analysis to the coal dust that every coal pulverizer grinds, because the present invention has 6 negative pressure sampling passages, therefore can realize the online detection to coal dust that every coal pulverizer grinds.
This example is described at the situation of four coal pulverizers, utilize four sampling passages of the present invention, sampling line is welded on the duff pipe 54 of every coal pulverizer, can realizes the coal dust that every coal pulverizer produces is carried out online detection and analysis according to method provided by the present invention.
Because coal pulverizer has horizontal pipeline section to the duff pipe 54 that boiler transports coal dust, and vertical pipeline section is also arranged, accompanying drawing 8 adopts horizontal pipeline section to describe, but also suitable fully at vertical pipeline section.
Online detection of a kind of ature of coal of the present invention and analytical equipment can be used for online detection and the analysis that belt is carried ature of coal under the raw coal situation.As shown in Figure 9, reliability in order to ensure transportation system, band conveyor 56 generally is the parallel settings of two belt feeders, can adopt among the present invention six sampling passages wherein two respectively two belts are taken a sample, with be installed in duff pipe on different, because the raw coal on the coal conveyer belt is the mixture of block and powdery coal dust, in order to guarantee to fetch coal dust, at the head of sampling pipe a loudspeaker sampling head 55 as shown in Figure 11 need be installed, loudspeaker sampling head 55 has the filtration unit of lattice-shaped, coal dust can pass through, and the bigger coal dust piece of particle can't pass through.
Online detection of a kind of ature of coal of the present invention and analytical equipment can be used for online detection and the analysis that Transport Machinery such as automobile are transported ature of coal under the raw coal situation.As shown in Figure 10, arrange two sampling pipes at haulage vehicle above through the highway section, the head of pipeline is installed loudspeaker sampling head 55, can realize the sampling and analysing to raw coal in the process that vehicle slowly travels.
Claims (11)
1. online check and analysis equipment of the ature of coal based on laser-induced spectrum technology and artificial neural network technology, comprise negative pressure sampling assemble (1), analyze case (2), spectra collection analytic unit (5), signals collecting control module (3), based on the signal processor (6) of artificial neural network; This equipment is characterised in that:
Negative pressure sampling assemble (1), (26A～26F), be used to fetch the coal dust sample is positioned over it in analysis case (2) to comprise negative pressure generator (8) and sampling line;
Analyze case (2), comprise Measurement and analysis pond (31) and control motor (39), the coal dust sample of being fetched by negative pressure sampling assemble (1) is placed in this Measurement and analysis pond (31), signals collecting control module (3) drive controlling motor (39) makes its drive Measurement and analysis pond (31) arrive the measuring position and drive Measurement and analysis pond (31) arrival by control motor (39) after finishing laser-induced spectrum collection and storage puts the powder position, and the coal dust sample is discharged;
Spectra collection analytic unit (5), comprise pulse laser source component (44), image acquisition control circuit (52) and ccd image sensor (50), image acquisition control circuit (52) sends signal to pulse laser source component (44) under the control of signal processor (6), make pulse laser source component (44) send laser pulse, on the coal dust sample of vertical irradiation in Measurement and analysis pond (31), inspire the spectral signal of inducting, image acquisition control circuit (52) is gathered the spectrum picture signal of ccd image sensor (50) simultaneously, and the signal of collection is transferred to signal processor (6);
Signal processor (6) based on artificial neural network, be used for the information via data processing of laser-induced spectrum is obtained coal dust sample carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, the results of elemental analyses of copper, and results of elemental analyses is input to neural network A in the signal processor (6), convert former calorific value of coal to by neural network A, moisture, ash content, fugitive constituent, fixed carbon technical analysis result, with resulting carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, the results of elemental analyses of copper and thermal value, moisture, ash content, fugitive constituent, fixed carbon technical analysis result is input to the neural network B in the signal processor (6), convert the melt temperature point and the slagging index of coal ash to, with above-mentioned results of elemental analyses, the technical analysis result, coal ash melting temperature point and slagging index pass through remote signal transmission interface (7) to outside device transmission.
2. the online check and analysis equipment of a kind of ature of coal according to claim 1 based on laser-induced spectrum technology and artificial neural network technology, it is characterized in that: negative pressure sampling assemble (1) also comprises dust-precipitator (9), return powder processing components (10), row's powder valve (11), blow down valve (14), put powder valve (15), Vib. (17), following coal powder silo with level meter sensor (16), last coal powder silo with level meter sensor (18), whirlwind powder collector (19), negative pressure control valve (20), exhaust pipe (21), the female pipe of sampling (22), abrasion-proof bent tube (23A～23F), the sampling operation valve (24A～24F), the blowback operation valve (25A～25F); Negative pressure sampling assemble (1) links to each other with analysis case (2) with row's tube cell (12) by putting tube cell (13); (24A～24F) is installed in corresponding sampling line respectively, and (on the 26A～26F), (25A～25F) is installed on the sampling operation valve (on the parallel pipeline of the top of 24A～24F) to the blowback operation valve to the sampling operation valve; (26A～26F) adopts abrasion-proof bent tube (23A～23F) can be connected to the female pipe of sampling (22) to sampling line; The female pipe of sampling (22) links to each other with whirlwind powder collector (19), and the port of whirlwind powder collector (19) top links to each other with negative pressure control valve (20), is connected to dust-precipitator (9) by exhaust pipe (21); Dust-precipitator (9) links to each other with time powder processing components (10) with negative pressure generator (8) simultaneously; The port of whirlwind powder collector (19) below with put tube cell (13) and be connected, put tube cell (13) and enter analysis case (2) through putting powder valve (15), Vib. (17) is installed on to be put above the powder valve (15); Following coal powder silo with level meter sensor (16) is installed on puts powder valve (15) top, and last coal powder silo with level meter sensor (18) is installed on down the top of coal powder silo with level meter sensor (16); On row's powder valve (11) row's of being installed on tube cell (12), blow down valve (14) is installed on the parallel pipeline of putting powder valve (15) below.
3. the online check and analysis equipment of ature of coal according to claim 1, it is characterized in that: analyze case (2) and also comprise analysis casing (27), put powder position transducer (28), measuring position sensor (38), cross-brace beam (37), turning axle (29), rotation back up pad (30), coal dust scraper plate (32), protection barrier driving motor (33), protection chamber (34), protection baffle plate (36), focus lamp (35); Wherein cross-brace beam (37) is installed on the middle part of analyzing casing (27), control motor (39) is fixed on the below of cross-brace beam (37), link to each other with rotation back up pad (30) by turning axle (29), driven rotary back up pad (30) is rotated, Measurement and analysis pond (31) is fixed on the rotation back up pad (30), under the drive of motor, can carry out circular motion, put powder position transducer (28) and measuring position sensor (38) and be fixed on the cross-brace beam (37); The inner upper right side of case (2) is being analyzed in protection chamber (34), and focus lamp (35) level is installed on protection inside, chamber (34); Protection tube (45) vertically enters protection chamber (34), the central lines of the axis of protection tube (45) and focus lamp (35) from the top of analyzing case (2); Photosensitive optical fiber (47) in the spectra collection analytic unit (5) and fiber optic protection sleeve pipe (48) enter protection chamber (34) from the upper right corner of analyzing case (2); Cleaning operation valve (41) in the protection assembly (4) links to each other with protection chamber (34) by pipeline; Protection barrier driving motor (33) is loaded on the outside, protection chamber (34), drives protection baffle plate (36) and rotates.
4. the online check and analysis equipment of a kind of ature of coal according to claim 1 based on laser-induced spectrum technology and artificial neural network technology, it is characterized in that: spectra collection analytic unit (5) also comprises protection tube (45), polychromator (46), photosensitive optical fiber (47), fiber optic protection sleeve pipe (48), Transmission Fibers (49) and low temperature module (51), wherein, pulse laser source component (44) links to each other with image acquisition control circuit (52), protection tube (45) is installed on the below of pulse laser source component (44), photosensitive optical fiber (47) is loaded in the protection fiber optic protection sleeve pipe (48), and link to each other with polychromator (46), the head of photosensitive optical fiber (47) is aimed at the laser focusing point, the output signal of polychromator (46) is connected to ccd image sensor (50) through Transmission Fibers (49), and ccd image sensor (50) links to each other with image acquisition control circuit (52); Low temperature module (51) is installed on ccd image sensor (50) on every side; Image acquisition control circuit (52) links to each other with signal processor (6), the control of acknowledge(ment) signal processor (6).
5. the online check and analysis equipment of a kind of ature of coal according to claim 1 based on laser-induced spectrum technology and artificial neural network technology, it is characterized in that: pulse laser source component (44) has the laser energy output signal, the electric signal of laser energy is represented in output when sending laser pulse, and this signal is connected to signals collecting control module (3).
6. the online check and analysis equipment of ature of coal according to claim 1, it is characterized in that: this analytical equipment also comprises protection assembly (4), this protection assembly (4) is made up of filtering pressure reducer (40), cleaning operation valve (41), nitrogen gas generator (42), protection gas control valve (43), and wherein nitrogen gas generator (42) links to each other with fiber optic protection sleeve pipe (48) with protection tube (45) through overprotection gas control valve (43); Protection gas control valve (43), cleaning operation valve (41) link to each other with signals collecting control module (3), by signals collecting control module (3) drive controlling; Nitrogen gas generator (42) produces the protection gas of the nitrogen of pressure-fired as pulsed laser source and photosensitive optical fiber (47) probe, this protection gas has three effects, the first provides the nitrogen stream of pressure-fired, prevent coal dust and contamination by dust laser via and photosensitive optical fiber (47), it two is that single gaseous environment is provided when the laser radiation sample, reduce the ground unrest of laser-induced spectrum, can prevent the coal dust detonation in addition; Pressurized air inserts protection chamber (34) through filtering pressure reducer (40) back under the control of cleaning operation valve (41); the head of focus lamp (35) and photosensitive optical fiber (47) is aimed in the pipeline outlet, and the head to focus lamp (35) and photosensitive optical fiber (47) behind the operation certain hour cleans.
7. the online check and analysis equipment of ature of coal according to claim 2, it is characterized in that: the signal of going up coal powder silo with level meter sensor (18), following coal powder silo with level meter sensor (16) in this negative pressure sampling assemble (1) inserts signals collecting control module (3), gather by signals collecting control module (3), and be transferred to signal processor (6); Blow down valve (14), negative pressure control valve (20), row's powder valve (11), (24A～24F), (25A～25F) link to each other with signals collecting control module (3) is by signals collecting control module (3) drive controlling for the blowback operation valve to put powder valve (15), Vib. (17), sampling operation valve.
8. the online check and analysis equipment of ature of coal according to claim 3, it is characterized in that: the signal of putting powder position transducer (28), measuring position sensor (38) in this analysis case (2) inserts signals collecting control module (3), gather by signals collecting control module (3), and be transferred to signal processor (6); Protection barrier driving motor (33) links to each other with signals collecting control module (3), by signals collecting control module (3) drive controlling.
9. the online check and analysis equipment of ature of coal according to claim 1, it is characterized in that: this equipment also comprises remote signal transmission interface (7), it links to each other with signal processor (6), the control of acknowledge(ment) signal processor (6) transmits results of elemental analyses, technical analysis result and coal ash melting temperature point and slagging index to other external unit.
10. the online check and analysis equipment of ature of coal according to claim 1 is characterized in that: signal processor (6) uses two artificial neural networks to carry out data processing, and treatment step comprises following three steps:
The first step: the element that obtains analyzed former coal sample by laser-induced spectrum;
After eliminating the influence of laser power to the spectrum of inducting, utilize the content of each element in the raw coal definite quantitative relationship to be arranged with the intensity of corresponding spectral line, calculate the content of element composition in the raw coal and each element; The present invention obtains the content of following element by the analysis to the spectrum of inducting: carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper;
Second step: utilize the BP artificial neural network to realize the conversion of results of elemental analyses to technical analysis; The constituent content that the first step is calculated is formed the input vector E={E of neural network
C, E
H, E
O2, E
N, E
S, E
Si, E
AL, E
Ca, E
Mg, E
K, E
Na, E
Ti, E
Fe, E
Mn, E
P, E
Cu, corresponding respectively: carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper; Realize that results of elemental analyses is designated as neural network A to the neural network of technical analysis result conversion, the structure of neural network A is: input layer has 16 unit, respectively corresponding carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper; The number of unit of hidden layer is greater than 50, and the unit number of output layer is 5, and is corresponding respectively: thermal value, moisture, ash content, fugitive constituent, fixed carbon;
The 3rd step: utilize the fusion point temperature and the slagging index of ash content in the BP neural network prediction coal, the result of the constituent content that the first step is calculated and second resulting ash, thermal value, volatile matter forms the input vector E={E of neural network
C, E
H, E
O2, E
N, E
S, E
Si, E
AL, E
Ca, E
Mg, E
K, E
Na, E
Ti, E
Fe, E
Mn, E
P, E
Cu, E
Ash, E
Heat, E
Daf, corresponding respectively: carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, copper, ash content, thermal value, fugitive constituent; Realize that the neural network of returning powder fusion point temperature and slagging index prediction in the coal is designated as neural network B, the structure of neural network B is: input layer number of unit 19, the hidden layer number of unit is greater than 68, the output layer unit number is 5, respectively corresponding deformation temperature DT, softening temperature ST, hemispherical temperature HT, flow temperature FT and slagging index.
11. an analytical approach that is used for the online check and analysis equipment of ature of coal, the online check and analysis equipment of this ature of coal comprise negative pressure sampling assemble (1), analyze case (2), signals collecting control module (3), based on the signal processor (6) and the remote signal transmission interface (7) of artificial neural network; It is characterized in that this method comprises:
Use negative pressure sampling assemble (1) to be used to fetch the coal dust sample, be positioned over the Measurement and analysis pond of analyzing in the case (2) (31);
Drive the control motor of analyzing in the case (2) (39) by signals collecting control module (3) and drive arrival measuring position, Measurement and analysis pond (31);
Send instruction by signal processor (6) to image acquisition control circuit (52), image acquisition control circuit (52) gating pulse lasing light emitter assembly (44) sends laser pulse, be radiated on the coal dust sample in the Measurement and analysis pond (31), inspire the spectral signal of inducting, this laser-induced spectrum signal is transferred to ccd image sensor (50), carry out the laser-induced spectrum information acquisition by image acquisition control circuit (52), the signal of gathering is transferred to signal processor (6), finishes the collection and the storage of laser-induced spectrum signal for the first time;
Drive the control motor of analyzing in the case (2) (39) by signals collecting control module (3) and drive Measurement and analysis pond (31) rotation predetermined angular, carry out the laser-induced spectrum collection second time and storage;
Drive control motor (39) the drive Measurement and analysis pond of analyzing in the case (2) (31) by signals collecting control module (3) and rotate predetermined angular once more, carry out laser-induced spectrum collection for the third time and storage;
Put the powder position by control motor (39) drive Measurement and analysis pond (31) arrival of analyzing in the case (2), so that the coal dust sample is discharged;
Signal processor (6) obtains carbon in the coal dust sample with the information via data processing of three laser-induced spectrums, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, the results of elemental analyses of copper, results of elemental analyses is input to neural network A in the signal processor (6), convert former calorific value of coal to by neural network A, moisture, ash content, fugitive constituent, fixed carbon technical analysis result, with resulting carbon, hydrogen, oxygen, nitrogen, sulphur, silicon, aluminium, calcium, magnesium, potassium, sodium, titanium, iron, manganese, phosphorus, the results of elemental analyses of copper and thermal value, moisture, ash content, fugitive constituent, fixed carbon technical analysis result is input to the neural network B in the signal processor (6), converts the melt temperature point and the slagging index of coal ash to; Above-mentioned results of elemental analyses, technical analysis result, coal ash melting temperature point and slagging index are transferred to external unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006101211344A CN100526859C (en) | 2006-08-23 | 2006-08-23 | Coal on-line analyse equipment based on laser induced spectral and nerve network technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006101211344A CN100526859C (en) | 2006-08-23 | 2006-08-23 | Coal on-line analyse equipment based on laser induced spectral and nerve network technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1912588A true CN1912588A (en) | 2007-02-14 |
| CN100526859C CN100526859C (en) | 2009-08-12 |
Family
ID=37721584
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2006101211344A Expired - Fee Related CN100526859C (en) | 2006-08-23 | 2006-08-23 | Coal on-line analyse equipment based on laser induced spectral and nerve network technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100526859C (en) |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101833283A (en) * | 2010-04-21 | 2010-09-15 | 武汉钢铁集团江南燃气热力有限责任公司 | Coked material balancing method |
| CN101995891A (en) * | 2010-09-17 | 2011-03-30 | 南京工业大学 | Online analysis method for moisture of solid mother material recovery system in aromatic acid production |
| CN102033043A (en) * | 2010-10-19 | 2011-04-27 | 浙江大学 | Grain moisture content detecting method based on hyperspectral image technology |
| CN101694430B (en) * | 2009-10-13 | 2011-08-24 | 北京金隅红树林环保技术有限责任公司 | Method for detecting raw coal |
| US20110269079A1 (en) * | 2010-04-28 | 2011-11-03 | Enviromental Energy Services, Inc. | Process for operating a utility boiler and methods therefor |
| CN101639412B (en) * | 2008-08-01 | 2011-11-09 | 中国神华能源股份有限公司 | Method for determining slagging prevention capacity of boiler by zoning |
| CN102313722A (en) * | 2011-09-05 | 2012-01-11 | 华南理工大学 | Proximate analyzing method for coal quality base on multivariate linear regression |
| CN101661001B (en) * | 2008-08-27 | 2012-03-28 | 华东电力试验研究院有限公司 | Method for measuring components of coal ash |
| CN102567785A (en) * | 2011-11-28 | 2012-07-11 | 浙江省电力试验研究院 | Numeric model-based coal element analysis method |
| CN102591370A (en) * | 2012-02-17 | 2012-07-18 | 中冶焦耐自动化有限公司 | Full-automatic coke furnace coal added water controlling system |
| CN102661935A (en) * | 2012-05-15 | 2012-09-12 | 广东电网公司电力科学研究院 | LIBS (Laser-induced Breakdown Spectroscopy) belt type powdery material detector |
| CN103728268A (en) * | 2013-12-17 | 2014-04-16 | 沈阳工程学院 | Near-infrared on-line coal quality measurement device |
| CN105445240A (en) * | 2016-01-11 | 2016-03-30 | 中国神华能源股份有限公司 | Monitoring system of process of producing aluminum oxide through coal ash |
| CN105823863A (en) * | 2016-03-28 | 2016-08-03 | 华北电力大学（保定） | Coal quality on-line industrial analysis and measurement method based on constant temperature thermogravimetric analysis |
| CN106442470A (en) * | 2016-08-31 | 2017-02-22 | 广州博谱能源科技有限公司 | Coal quality characteristic quantitative analysis method based on LIBS (laser induced breakdown spectrum) and genetic neural network |
| CN107589093A (en) * | 2017-09-08 | 2018-01-16 | 邓雷 | A kind of ature of coal on-line checking analysis method based on regression analysis |
| CN107870166A (en) * | 2017-01-04 | 2018-04-03 | 广东省特种设备检测研究院顺德检测院 | A kind of heating value of coal rapid measurement device without burning |
| CN109060349A (en) * | 2018-07-19 | 2018-12-21 | 洛阳轴承研究所有限公司 | The testing equipment and test method of screw type refrigerating compressor rotor supports bearing |
| CN110057735A (en) * | 2019-05-24 | 2019-07-26 | 太原市海通自动化技术有限公司 | A kind of thermoelectricity station-service concentration of pulverized coal measurement method and measuring device |
| CN111751370A (en) * | 2020-06-17 | 2020-10-09 | 神华神东电力有限责任公司 | System and method for testing thermal efficiency of boiler |
| CN111754028A (en) * | 2020-06-08 | 2020-10-09 | 吉林大学 | Hyperspectrum-based coal ash content and moisture detection system and method |
| CN111896544A (en) * | 2020-08-05 | 2020-11-06 | 合肥约翰芬雷矿山装备有限公司 | Online combustion value detection method and detection device for coal dressing |
| CN112881455A (en) * | 2021-01-15 | 2021-06-01 | 山西格盟中美清洁能源研发中心有限公司 | Method for predicting coal ash melting temperature based on mineral phase and neural network composite model |
| WO2021203675A1 (en) * | 2020-04-08 | 2021-10-14 | 西安热工研究院有限公司 | Design and control method for flue gas temperature at boiler hearth outlet |
| CN114324469A (en) * | 2021-12-29 | 2022-04-12 | 华中科技大学 | Analysis and test method for organic waste liquid with high water content |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103278653B (en) * | 2013-05-02 | 2015-11-25 | 太原市海通自动化技术有限公司 | A kind of ature of coal on-line analysis and analytic system |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4562044A (en) * | 1982-07-06 | 1985-12-31 | The Babcock & Wilcox Company | On-line coal analyzer |
| IT1306112B1 (en) * | 1998-03-20 | 2001-05-29 | Consiglio Nazionale Ricerche | METHOD FOR THE QUANTITATIVE ANALYSIS OF THE ATOMIC COMPONENTS OF MATERIALS BY MEANS OF LIBS SPECTROSCOPY MEASURES WITHOUT CALIBRATION |
| AUPP573098A0 (en) * | 1998-09-04 | 1998-10-01 | Generation Technology Research Pty Ltd | Apparatus and method for analyzing material |
| US6532068B2 (en) * | 2001-07-17 | 2003-03-11 | National Research Council Of Canada | Method and apparatus for depth profile analysis by laser induced plasma spectros copy |
| CN1243233C (en) * | 2003-07-18 | 2006-02-22 | 华中科技大学 | Analyser for spark through spectrum medium by laser induced |
-
2006
- 2006-08-23 CN CNB2006101211344A patent/CN100526859C/en not_active Expired - Fee Related
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101639412B (en) * | 2008-08-01 | 2011-11-09 | 中国神华能源股份有限公司 | Method for determining slagging prevention capacity of boiler by zoning |
| CN101661001B (en) * | 2008-08-27 | 2012-03-28 | 华东电力试验研究院有限公司 | Method for measuring components of coal ash |
| CN101694430B (en) * | 2009-10-13 | 2011-08-24 | 北京金隅红树林环保技术有限责任公司 | Method for detecting raw coal |
| CN101833283A (en) * | 2010-04-21 | 2010-09-15 | 武汉钢铁集团江南燃气热力有限责任公司 | Coked material balancing method |
| US20110269079A1 (en) * | 2010-04-28 | 2011-11-03 | Enviromental Energy Services, Inc. | Process for operating a utility boiler and methods therefor |
| CN101995891B (en) * | 2010-09-17 | 2012-09-19 | 南京工业大学 | Online analysis method for moisture of solid mother material recovery system in aromatic acid production |
| CN101995891A (en) * | 2010-09-17 | 2011-03-30 | 南京工业大学 | Online analysis method for moisture of solid mother material recovery system in aromatic acid production |
| CN102033043A (en) * | 2010-10-19 | 2011-04-27 | 浙江大学 | Grain moisture content detecting method based on hyperspectral image technology |
| CN102313722A (en) * | 2011-09-05 | 2012-01-11 | 华南理工大学 | Proximate analyzing method for coal quality base on multivariate linear regression |
| CN102313722B (en) * | 2011-09-05 | 2013-03-20 | 华南理工大学 | Proximate analyzing method for coal quality based on multivariate linear regression |
| CN102567785A (en) * | 2011-11-28 | 2012-07-11 | 浙江省电力试验研究院 | Numeric model-based coal element analysis method |
| CN102591370A (en) * | 2012-02-17 | 2012-07-18 | 中冶焦耐自动化有限公司 | Full-automatic coke furnace coal added water controlling system |
| CN102661935A (en) * | 2012-05-15 | 2012-09-12 | 广东电网公司电力科学研究院 | LIBS (Laser-induced Breakdown Spectroscopy) belt type powdery material detector |
| CN103728268B (en) * | 2013-12-17 | 2015-12-30 | 沈阳工程学院 | Near infrared ature of coal on-line measurement device |
| CN103728268A (en) * | 2013-12-17 | 2014-04-16 | 沈阳工程学院 | Near-infrared on-line coal quality measurement device |
| CN105445240A (en) * | 2016-01-11 | 2016-03-30 | 中国神华能源股份有限公司 | Monitoring system of process of producing aluminum oxide through coal ash |
| CN105823863A (en) * | 2016-03-28 | 2016-08-03 | 华北电力大学（保定） | Coal quality on-line industrial analysis and measurement method based on constant temperature thermogravimetric analysis |
| CN106442470A (en) * | 2016-08-31 | 2017-02-22 | 广州博谱能源科技有限公司 | Coal quality characteristic quantitative analysis method based on LIBS (laser induced breakdown spectrum) and genetic neural network |
| CN107870166A (en) * | 2017-01-04 | 2018-04-03 | 广东省特种设备检测研究院顺德检测院 | A kind of heating value of coal rapid measurement device without burning |
| CN107589093A (en) * | 2017-09-08 | 2018-01-16 | 邓雷 | A kind of ature of coal on-line checking analysis method based on regression analysis |
| CN109060349B (en) * | 2018-07-19 | 2020-09-25 | 洛阳轴承研究所有限公司 | Testing equipment and testing method for rotor support bearing of screw type refrigeration compressor |
| CN109060349A (en) * | 2018-07-19 | 2018-12-21 | 洛阳轴承研究所有限公司 | The testing equipment and test method of screw type refrigerating compressor rotor supports bearing |
| CN110057735A (en) * | 2019-05-24 | 2019-07-26 | 太原市海通自动化技术有限公司 | A kind of thermoelectricity station-service concentration of pulverized coal measurement method and measuring device |
| WO2021203675A1 (en) * | 2020-04-08 | 2021-10-14 | 西安热工研究院有限公司 | Design and control method for flue gas temperature at boiler hearth outlet |
| CN111754028A (en) * | 2020-06-08 | 2020-10-09 | 吉林大学 | Hyperspectrum-based coal ash content and moisture detection system and method |
| CN111751370A (en) * | 2020-06-17 | 2020-10-09 | 神华神东电力有限责任公司 | System and method for testing thermal efficiency of boiler |
| CN111896544A (en) * | 2020-08-05 | 2020-11-06 | 合肥约翰芬雷矿山装备有限公司 | Online combustion value detection method and detection device for coal dressing |
| CN112881455A (en) * | 2021-01-15 | 2021-06-01 | 山西格盟中美清洁能源研发中心有限公司 | Method for predicting coal ash melting temperature based on mineral phase and neural network composite model |
| CN112881455B (en) * | 2021-01-15 | 2024-03-12 | 山西格盟中美清洁能源研发中心有限公司 | Method for predicting coal ash melting temperature based on mineral phase and neural network composite model |
| CN114324469A (en) * | 2021-12-29 | 2022-04-12 | 华中科技大学 | Analysis and test method for organic waste liquid with high water content |
| CN114324469B (en) * | 2021-12-29 | 2024-05-07 | 华中科技大学 | Analysis and test method for high-water-content organic waste liquid |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100526859C (en) | 2009-08-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100526859C (en) | Coal on-line analyse equipment based on laser induced spectral and nerve network technology | |
| Legnaioli et al. | Industrial applications of laser-induced breakdown spectroscopy: a review | |
| Sheta et al. | Coal analysis by laser-induced breakdown spectroscopy: a tutorial review | |
| CN100526854C (en) | Beta-ray soot concentration direct-reading monitor and method for determining effective sample | |
| CN101509873B (en) | Coal quality detecting method based on active marker method | |
| CN101949852B (en) | Spectral standardization-based coal quality on-line detection method | |
| CN1020000C (en) | Calorimetry system | |
| Chansa et al. | Determination of alkali release during oxyfuel co-combustion of biomass and coal using laser-induced breakdown spectroscopy | |
| CN110133007B (en) | Coal consumption on-line metering system | |
| Yao et al. | Characterization of fly ash laser-induced plasma for improving the on-line measurement of unburned carbon in gas–solid flow | |
| CN115008006B (en) | Laser welding detection evaluation system and method | |
| Li et al. | Online dynamic prediction of potassium concentration in biomass fuels through flame spectroscopic analysis and recurrent neural network modelling | |
| CN204287043U (en) | Based on soil heavy metal cadmium and the copper content detection device of induced with laser spectral technique | |
| Li et al. | Simultaneous determination of Na concentration and temperature during Zhundong coal combustion using the radiation spectrum | |
| E. Romero et al. | LIBS analysis for coal | |
| CN202661379U (en) | Laser-induced breakdown spectroscopy (LIBS) powdery material belt type detection device | |
| CN110095562A (en) | Dual-Phrase Distribution of Gas olid visualization combustion furnace and corrosion chamber with purging function | |
| CN105424557A (en) | Online measurement device and method for boiler pulverized coal particle parameters | |
| CN209513693U (en) | A kind of calorimeter assembly and system for coal dust calorific value on-line measurement | |
| CN109270120B (en) | Heat measuring device and system for on-line measurement of calorific value of pulverized coal | |
| Sthel et al. | Pollutant gas and particulate material emissions in ethanol production in Brazil: social and environmental impacts | |
| Dong et al. | A review of laser-induced breakdown spectroscopy and spontaneous emission techniques in monitoring thermal conversion of fuels | |
| CN85101184A (en) | Combustion system | |
| Skvaril et al. | Utilization of near infrared (NIR) spectrometry for detection of glass in the waste-based fuel | |
| Hedlund | Detection of glass in RDF using NIR spectroscopy |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090812 Termination date: 20150823 |
|
| EXPY | Termination of patent right or utility model |