CN105930302B - A kind of coal pulverizer method for diagnosing faults based on Fast Genetic Algorithm and grey-box model - Google Patents

A kind of coal pulverizer method for diagnosing faults based on Fast Genetic Algorithm and grey-box model Download PDF

Info

Publication number
CN105930302B
CN105930302B CN201610209933.0A CN201610209933A CN105930302B CN 105930302 B CN105930302 B CN 105930302B CN 201610209933 A CN201610209933 A CN 201610209933A CN 105930302 B CN105930302 B CN 105930302B
Authority
CN
China
Prior art keywords
mrow
msub
coal pulverizer
coal
mover
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.)
Active
Application number
CN201610209933.0A
Other languages
Chinese (zh)
Other versions
CN105930302A (en
Inventor
梁修凡
沈炯
李益国
刘西陲
吴啸
潘蕾
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southeast University
Original Assignee
Southeast University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Southeast University filed Critical Southeast University
Priority to CN201610209933.0A priority Critical patent/CN105930302B/en
Publication of CN105930302A publication Critical patent/CN105930302A/en
Application granted granted Critical
Publication of CN105930302B publication Critical patent/CN105930302B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10Complex mathematical operations
    • G06F17/11Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/12Computing arrangements based on biological models using genetic models
    • G06N3/126Evolutionary algorithms, e.g. genetic algorithms or genetic programming

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Biophysics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Pure & Applied Mathematics (AREA)
  • Evolutionary Biology (AREA)
  • Computational Mathematics (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Biomedical Technology (AREA)
  • Operations Research (AREA)
  • Algebra (AREA)
  • Physiology (AREA)
  • Genetics & Genomics (AREA)
  • Artificial Intelligence (AREA)
  • Databases & Information Systems (AREA)
  • Computational Linguistics (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • Disintegrating Or Milling (AREA)
  • Other Investigation Or Analysis Of Materials By Electrical Means (AREA)

Abstract

The invention discloses a kind of coal pulverizer method for diagnosing faults based on Fast Genetic Algorithm and grey-box model, Fast Genetic Algorithm and grey-box model are combined, a kind of novel coal mill method for diagnosing faults is obtained, including off-line model identification and the two processes of online breakdown judge.The present invention need not obtain the fault data of coal pulverizer with regard to that can carry out the judgement of fault type.Due to considering the mechanism characteristic of coal pulverizer internal physical process, there is higher precision than conventional method for diagnosing faults.Model is recognized using Fast Genetic Algorithm, effectively reduces the time overhead of identification process.

Description

A kind of coal pulverizer method for diagnosing faults based on Fast Genetic Algorithm and grey-box model
Technical field
The present invention relates to thermal power plant control fields, and Fast Genetic Algorithm and grey-box model are based on more particularly to one kind Coal pulverizer method for diagnosing faults.
Background technology
A Rapid development stage is passed through in China's power industry at present, and the number of thermal power plant and installed capacity have reached Very huge quantity.In July, 2015, national whole month power generation have reached 4536.85 hundred million kilowatt hours, average daily power generation 146.35 hundred million Kilowatt hour, generate electricity maximum day 152.81 hundred million kilowatt hours (July 12), adds up power generation year up to 28089.47 hundred million kilowatt hours.Although combustion The technology of coal unit develops constantly;But at the same time, the reliability of some key equipments of thermal power plant, availability, can repair Property the problem of security with also becoming increasingly conspicuous.
Power station coal pulverizer is the core equipment of boiler combustion pulverized coal preparation system, and working condition runs entire power plant system Safety and economy have important influence.Such as the pressure of milling of coal pulverizer, ventilation quantity, the load in coal pulverizer etc., they Size and exception will all influence production efficiency.As far as possible these states to be avoided to be abnormal in process of production, the exception of state On the one hand productivity can be influenced, the failure of equipment on the other hand will be also caused, causes the loss of bigger.Therefore timely discovering device It is abnormal, timely maintained equipment, it appears particularly important.It for example, during the hydraulic performance decline of coal pulverizer, should be safeguarded in time, to reduce Powder-grinding energy consumption avoids pulverizer capacity deficiency that the power of the assembling unit is caused to decline;It, should be timely when the coal pulverizer of operation breaks down It is overhauled, avoids causing the accident of unit outage due to coal pulverizer failure.Coal pulverizer generally has to work in the presence of a harsh environment, Its course of work is a complicated process.Therefore the condition monitoring and fault diagnosis for realizing power station coal-grinding machine equipment is a tool Challenging work.This is also always thermal power plant's coal-grinding production urgent problem to be solved.
The coal pulverizer method for diagnosing faults studied both at home and abroad at present can be mainly divided into two classes:Kernel model based diagnosis method With the diagnostic method based on fault mode classification.The main distinction of the two is whether need the dynamic reference mould for establishing equipment Type.
According to the method based on fault mode classification, it is necessary to while possess the fault mode data and normal mode of coal pulverizer Formula data could the different fault type of enough grader differentiations.The data of coal pulverizer normal mode can be readily available, but Data when being failure must could be obtained when coal pulverizer breaks down, therefore the coal pulverizer for not yet breaking down, this Kind method is difficult to judge out of order type.
First have to establish the reference model of coal pulverizer normal operation mode using the method for diagnosing faults based on model.Root again Fault diagnosis is carried out according to reference model output valve and the residual error of measured value.In this way, data during even if without coal pulverizer failure The type of residual error can be drawn according to the analysis result of residual error.At home, the method based on grey-box model is not yet applied to coal-grinding The fault diagnosis of machine.Grey-box model combines mechanism and data, can reflect real physical process inside coal pulverizer, compare cardinar number There is higher precision according to the method for modeling.Therefore the precision of fault diagnosis can be improved using grey-box model.
For traditional genetic algorithm when recognizing grey-box model, time overhead is very big, and being often required to calculate a whole day could be to Go out identification result.
The content of the invention
Goal of the invention:The object of the present invention is to provide it is a kind of can effectively save identification the time based on Fast Genetic Algorithm With the coal pulverizer method for diagnosing faults of grey-box model.
Technical solution:To reach this purpose, the present invention uses following technical scheme:
Coal pulverizer method for diagnosing faults of the present invention based on Fast Genetic Algorithm and grey-box model, including offline mould Type recognizes and online breakdown judge;
Off-line model identification based on genetic algorithm comprises the following steps:
S1:Based on the mass balance of coal inside coal pulverizer, the differential of raw coal quality and quality of pc inside coal pulverizer is established Equation:
In formula (1), (2), mcFor the raw coal quality inside coal pulverizer, qm,cFor coal pulverizer inlet coal quality flow, mpfFor mill Quality of pc inside coal machine, qm,pfPulverized coal mass flow is exported for coal pulverizer;
Coal pulverizer outlet pulverized coal mass flow qm,pfIt is calculated according to formula (3):
qm,pf=K11Δpampf (3)
In formula (3), Δ paFor the pressure difference that primary air fan generates, calculated according to formula (4);
In formula (4), tinFor air temperature of coal pulverizer inlet, qm,airFor the mass flow of coal pulverizer inlet First air;
S2:Calculate coal pulverizer internal moisture evaporation capacity
In formula (5), θCMFor raw coal moisture, and
S3:Calculate coal pulverizer internal energy balance equation:
In formula (6), toutAn air temperature is exported for coal pulverizer, I is coal pulverizer electric current, and I is calculated according to formula (7):
I=K6mpf+K7mc+K8 (7)
S4:Calculate coal pulverizer outlet pressure equation:
More than K1~K18It is parameter to be identified;
S5:The data of following measuring point are gathered from the DCS system of power plant:Coal pulverizer inlet coal quality flow qm,c, coal pulverizer The mass flow q of entrance First airm,air, coal pulverizer export an air temperature tout, coal pulverizer electric current I and coal pulverizer outlet pressure P;
S6:The initial parameter of genetic algorithm is set, including generation gap, crossing-over rate, population dimension, aberration rate, maximum hereditary generation Number and population at individual number;Then the upper and lower bound of Genetic algorithm searching matrix is set;The parameter of each individual is in population One one-dimensional matrix, the one-dimensional matrix share 22 degree of freedom, and preceding 4 degree of freedom are the raw coal quality m inside coal pulverizerc, mill Quality of pc m inside coal machinepf, raw coal moisture θCMAn air temperature t is exported with coal pulverizerout, rear 18 degree of freedom are to treat Identified parameters K1~K18, the initial value of individual parameter is random in the range of the upper and lower bound of Genetic algorithm searching matrix in population Value;
S7:The primary population of initial time genetic algorithm calculates the fitness function value of each individual of primary population:It will be from The coal pulverizer inlet coal quality flow q of DCS acquisitionsm,c, coal pulverizer inlet First air mass flow qm,airIt is and individual in population The initial value of parameter is input in formula (1)-formula (8), coal pulverizer electric current I and coal pulverizer outlet pressure P is obtained, then according to formula (9) fitness function is calculated:
In formula (9), W1、W2And W3For the weights of artificial settings, N is the data sample number gathered from DCS,For tout's Measured value,For the measured value of I,For the measured value of P,ForMaximum,ForMaximum,For's Maximum;
S8:Selection, restructuring, mutation operator are acted on into group, obtain progeny population;
S9:Calculate the fitness function value of all individuals in progeny population;
S10:Weight insertion operation is performed to progeny population, judges whether genetic algebra reaches the maximum of setting:If not Reach, then go to step S8;If reached, off-line identification process terminates.
On-line fault diagnosis comprise the following steps:
S11:By individual parameter in the population of the fitness function minimum obtained during off-line identification by genetic algorithm It brings into formula (1)-formula (8);
S12:Sampling time is set, a coal pulverizer fan-in evidence is gathered from Power Plant DCS System every the sampling time With fan-out evidence;Coal pulverizer fan-in is according to including coal pulverizer inlet coal quality flow qm,c, air temperature of coal pulverizer inlet tinWith the mass flow q of coal pulverizer inlet First airm,air, coal pulverizer fan-out is according to including the air temperature in coal pulverizer outlet tout, coal pulverizer outlet pulverized coal mass flow qm,pf, coal pulverizer electric current I and coal pulverizer outlet pressure P;
S13:The coal pulverizer fan-in evidence collected is input to formula (1)-formula to be solved in (8);
S14:The output valve of wushu (1)-formula (8) and the coal pulverizer fan-out evidence gathered from DCS are compared, and are obtained The residual error e of the twok, then ekIt is compared with the fault threshold set, if ekMore than the threshold value of setting, then it is assumed that system Failure has occurred.
Further, the step S9 is:The fitness function value of each individual in progeny population is calculated successively, if calculated It obtains some individual fitness function value and reaches upper limit value Ω as shown in formula (10), then stop calculating the individual fitness Functional value, and using the individual fitness function value as upper limit value Ω, then proceed to calculate remaining individual fitness function Value;
Further, in the step S5, the sampling time of data acquisition is arranged to 1s, and each measuring point gathers 10000 groups of numbers According to.
Advantageous effect:The present invention need not obtain the fault data of coal pulverizer with regard to that can carry out the judgement of fault type.Due to The mechanism characteristic of coal pulverizer internal physical process is considered, there is higher precision than conventional method for diagnosing faults.Using fast Fast genetic algorithm recognizes model, effectively reduces the time overhead of identification process.
Description of the drawings
Fig. 1 is the flow chart of the off-line model identification of the present invention;
Fig. 2 is the flow chart of the online breakdown judge of the present invention;
Fig. 3 is to carry out the model output value of electric current and pair of measured value that off-line model recognizes using the method for the present invention Than figure;
Fig. 4 is to carry out the model output value of temperature and pair of measured value that off-line model recognizes using the method for the present invention Than figure;
Fig. 5 is to carry out the model output value of pressure and pair of measured value that off-line model recognizes using the method for the present invention Than figure;
Fig. 6 is to carry out the model output value of electric current and pair of measured value that online breakdown judge obtains using the method for the present invention Than figure;
Fig. 7 is between the model output value of the electric current obtained using the online breakdown judge of the method for the present invention progress and measured value Residual sum threshold value comparison diagram;
Fig. 8 is to carry out the model output value of temperature and pair of measured value that online breakdown judge obtains using the method for the present invention Than figure;
Fig. 9 is between the model output value of the temperature obtained using the online breakdown judge of the method for the present invention progress and measured value Residual sum threshold value comparison diagram;
Figure 10 is the model output value and measured value that the pressure that online breakdown judge obtains is carried out using the method for the present invention Comparison diagram;
Figure 11 be using the method for the present invention carry out the pressure that online breakdown judge obtains model output value and measured value it Between residual sum threshold value comparison diagram.
Specific embodiment
Technical scheme is further introduced With reference to embodiment.
The invention discloses a kind of coal pulverizer method for diagnosing faults based on Fast Genetic Algorithm and grey-box model, including from Line model recognizes and online breakdown judge;
Off-line model identification comprises the following steps:
S1:Based on the mass balance of coal inside coal pulverizer, the differential of raw coal quality and quality of pc inside coal pulverizer is established Equation:
In formula (1), (2), mcFor the raw coal quality inside coal pulverizer, qm,cFor coal pulverizer inlet coal quality flow, mpfFor mill Quality of pc inside coal machine, qm,pfPulverized coal mass flow is exported for coal pulverizer;
Coal pulverizer outlet pulverized coal mass flow qm,pfIt is calculated according to formula (3):
qm,pf=K11Δpampf (3)
In formula (3), Δ paFor the pressure difference that primary air fan generates, calculated according to formula (4);
In formula (4), tinFor air temperature of coal pulverizer inlet, qm,airFor the mass flow of coal pulverizer inlet First air;
S2:Calculate coal pulverizer internal moisture evaporation capacity
In formula (5), θCMFor raw coal moisture, and
S3:Calculate coal pulverizer internal energy balance equation:
In formula (6), toutA temperature is exported for coal pulverizer, I is coal pulverizer electric current, and I is calculated according to formula (7):
I=K6mpf+K7mc+K8 (7)
S4:Calculate coal pulverizer outlet pressure equation:
More than K1~K18It is parameter to be identified;
S5:The data of following measuring point are gathered from the DCS system of power plant:Coal pulverizer inlet coal quality flow qm,c, coal pulverizer The mass flow q of entrance First airm,air, coal pulverizer export an air temperature tout, coal pulverizer electric current I and coal pulverizer outlet pressure P;
S6:The initial parameter of genetic algorithm is set:
Generation gap is 0.8;
Crossing-over rate is 1;
Population dimension is 22, wherein initial state value of preceding 4 dimension for grey-box model, rear 15 dimension is genetic algorithm parameter;
Aberration rate is 1/22;
Maximum genetic algebra is 40;
Individual amount is 20;
Then the upper and lower bound of Genetic algorithm searching matrix is set:
The algorithm search matrix upper limit is:
[50,20,100,15,0.0035,0.25,0.015,0.0055,0.12,0.3,1,30,0.05,0.55, 0.008,-0.045,0.01,0.3,0.2,6,0.075,0.025];
Algorithm search matrix lower limit is:
[10,5,30,5,0.002,0.05,0.006,0.0008,0.07,0.1,0.7,5,0.01,0.3,0.0004,- 0.09,0.006,0.05,0.05,4,0.055,0.01];
The parameter of each individual is an one-dimensional matrix in population, which shares 22 degree of freedom, preceding 4 freedom Degree is the raw coal quality m inside coal pulverizerc, quality of pc m inside coal pulverizerpf, raw coal moisture θCMWith coal pulverizer outlet one Secondary air temperature tout, rear 18 degree of freedom are parameter K to be identified1~K18, the initial value of individual parameter is searched in genetic algorithm in population Random value in the range of the upper and lower bound of rope matrix;
S7:The primary population of initial time genetic algorithm calculates the fitness function value of each individual of primary population:It will be from The coal pulverizer inlet coal quality flow q of DCS acquisitionsm,c, coal pulverizer inlet First air mass flow qm,airIt is and individual in population The initial value of parameter is input in formula (1)-formula (8), coal pulverizer electric current I and coal pulverizer outlet pressure P is obtained, then according to formula (9) fitness function is calculated:
In formula (9), W1、W2And W3For the weights of artificial settings, N is the data sample number gathered from DCS,For toutReality Measured value,For the measured value of I,For the measured value of P,ForMaximum,ForMaximum,ForMost Big value;
S8:Selection, restructuring, mutation operator are acted on into group, obtain progeny population;
S9:Calculate the fitness function value of all individuals in progeny population;
S10:Weight insertion operation is performed to progeny population, judges whether genetic algebra reaches the maximum of setting:If not Reach, then go to step S8;If reached, terminate.
The results are shown in Table 1 for Model Distinguish.
1 Model Distinguish result of table
K1=0.0034 K2=0.2345 K3=0.0137 K4=0.0049 K5=0.0821 K6=0.2232
K7=0.7053 K8=20.0000 K9=0.0465 K10=0.3570 K11=0.0032 K12=-0.0464
K13=0.0073 K14=0.1759 K15=0.0758 K16=4.6200 K17=0.0701 K18=0.0214
mc=41.24 mpf=13.68 θCM=8.61 tout=78.21
The verification result of off-line model identification is as shown in Fig. 3,4 and 5, it is seen then that carries out off-line model identification using the present invention Precision it is very high, it is basic to coincide with Power Plant DCS measured data.
On-line fault diagnosis comprise the following steps:
S11:By individual parameter in the population of the fitness function minimum obtained during off-line identification by genetic algorithm It brings into formula (1)-formula (8);
S12:Sampling time is set, a coal pulverizer fan-in evidence is gathered from Power Plant DCS System every the sampling time With fan-out evidence;Coal pulverizer fan-in is according to including coal pulverizer inlet coal quality flow qm,c, air temperature of coal pulverizer inlet tinWith the mass flow q of coal pulverizer inlet First airm,air, coal pulverizer fan-out is according to including the air temperature in coal pulverizer outlet tout, coal pulverizer outlet pulverized coal mass flow qm,pf, coal pulverizer electric current I and coal pulverizer outlet pressure P;
S13:The coal pulverizer fan-in evidence collected is input to formula (1)-formula to be solved in (8);
S14:The output valve of wushu (1)-formula (8) and the coal pulverizer fan-out evidence gathered from DCS are compared, and are obtained The residual error e of the twok, then ekIt is compared with the fault threshold set, if ekMore than the threshold value of setting, then it is assumed that system Failure has occurred.
In coal pulverizer normal operation, the residual error of coal pulverizer outputting measurement value and model output value becomes in smaller scope It is dynamic.When a failure occurs, which can exceed the threshold value of setting.Fault type may determine that according to the feature of residual error.It gives below The most common three kinds of catastrophic discontinuityfailure diagnostic methods of coal pulverizer are gone out, have been respectively:Coal pulverizer inlet breaks coal failure, such as Fig. 6 and Fig. 7 It is shown;Coal pulverizer coal powder ignition failure, as shown in Figure 8 and Figure 9;Coal pulverizer blocks up powder failure, as shown in Figure 10 and Figure 11.
It can be seen from the figure that when an error occurs, residual error generates tendency variation, as long as setting certain residual error threshold Value, it is possible to the generation of troubleshooting, and according to the type of type of residual error failure judgement.If occur there are many failure or unknown Failure occurs, and can be judged according to residual error.Therefore the present invention can effectively diagnose the fault type of coal pulverizer.

Claims (2)

1. a kind of coal pulverizer method for diagnosing faults based on Fast Genetic Algorithm and grey-box model, it is characterised in that:Including offline Model Distinguish and online breakdown judge;
Off-line model identification based on genetic algorithm comprises the following steps:
S1:Based on the mass balance of coal inside coal pulverizer, the differential equation of raw coal quality and quality of pc inside coal pulverizer is established:
<mrow> <msub> <mover> <mi>m</mi> <mo>&amp;CenterDot;</mo> </mover> <mi>c</mi> </msub> <mo>=</mo> <msub> <mi>q</mi> <mrow> <mi>m</mi> <mo>,</mo> <mi>c</mi> </mrow> </msub> <mo>-</mo> <msub> <mi>K</mi> <mn>10</mn> </msub> <msub> <mi>m</mi> <mi>c</mi> </msub> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow>
<mrow> <msub> <mover> <mi>m</mi> <mo>&amp;CenterDot;</mo> </mover> <mrow> <mi>p</mi> <mi>f</mi> </mrow> </msub> <mo>=</mo> <msub> <mi>K</mi> <mn>10</mn> </msub> <msub> <mi>m</mi> <mi>c</mi> </msub> <mo>-</mo> <msub> <mi>q</mi> <mrow> <mi>m</mi> <mo>,</mo> <mi>p</mi> <mi>f</mi> </mrow> </msub> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </mrow>
In formula (1), (2), mcFor the raw coal quality inside coal pulverizer, qm,cFor coal pulverizer inlet coal quality flow, mpfFor coal pulverizer Internal quality of pc, qm,pfPulverized coal mass flow is exported for coal pulverizer;
Coal pulverizer outlet pulverized coal mass flow qm,pfIt is calculated according to formula (3):
qm,pf=K11Δpampf (3)
In formula (3), Δ paFor the pressure difference that primary air fan generates, calculated according to formula (4);
<mrow> <msub> <mi>&amp;Delta;p</mi> <mi>a</mi> </msub> <mo>=</mo> <mn>100</mn> <mo>&amp;times;</mo> <mfrac> <mn>22.4</mn> <mn>28.8</mn> </mfrac> <mo>&amp;times;</mo> <mfrac> <mrow> <mn>273</mn> <mo>+</mo> <msub> <mi>t</mi> <mrow> <mi>i</mi> <mi>n</mi> </mrow> </msub> </mrow> <mn>273</mn> </mfrac> <mo>&amp;times;</mo> <msup> <mrow> <mo>(</mo> <mfrac> <msub> <mi>q</mi> <mrow> <mi>m</mi> <mo>,</mo> <mi>a</mi> <mi>i</mi> <mi>r</mi> </mrow> </msub> <mn>10</mn> </mfrac> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>4</mn> <mo>)</mo> </mrow> </mrow>
In formula (4), tinFor air temperature of coal pulverizer inlet, qm,airFor the mass flow of coal pulverizer inlet First air;
S2:Calculate coal pulverizer internal moisture evaporation capacity
<mrow> <msubsup> <mi>W</mi> <mrow> <mi>f</mi> <mi>r</mi> <mi>e</mi> <mi>e</mi> </mrow> <mrow> <mi>w</mi> <mi>a</mi> <mi>t</mi> <mi>e</mi> <mi>r</mi> </mrow> </msubsup> <mo>=</mo> <msub> <mi>K</mi> <mn>13</mn> </msub> <mrow> <mo>(</mo> <msub> <mi>q</mi> <mrow> <mi>m</mi> <mo>,</mo> <mi>c</mi> </mrow> </msub> <msub> <mi>&amp;theta;</mi> <mrow> <mi>C</mi> <mi>M</mi> </mrow> </msub> <mo>)</mo> </mrow> <msub> <mi>t</mi> <mrow> <mi>o</mi> <mi>u</mi> <mi>t</mi> </mrow> </msub> <mrow> <mo>(</mo> <mn>1</mn> <mo>-</mo> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mfrac> <msub> <mi>q</mi> <mrow> <mi>m</mi> <mo>,</mo> <mi>a</mi> <mi>i</mi> <mi>r</mi> </mrow> </msub> <msub> <mi>K</mi> <mn>15</mn> </msub> </mfrac> </mrow> </msup> <mo>)</mo> </mrow> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>5</mn> <mo>)</mo> </mrow> </mrow>
In formula (5), toutAn air temperature, θ are exported for coal pulverizerCMFor raw coal moisture, and
S3:Calculate coal pulverizer internal energy balance equation:
<mrow> <msub> <mover> <mi>t</mi> <mo>&amp;CenterDot;</mo> </mover> <mrow> <mi>o</mi> <mi>u</mi> <mi>t</mi> </mrow> </msub> <mo>=</mo> <mrow> <mo>(</mo> <msub> <mi>K</mi> <mn>1</mn> </msub> <msub> <mi>t</mi> <mrow> <mi>i</mi> <mi>n</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>K</mi> <mn>2</mn> </msub> <mo>)</mo> </mrow> <msub> <mi>q</mi> <mrow> <mi>m</mi> <mo>,</mo> <mi>a</mi> <mi>i</mi> <mi>r</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>K</mi> <mn>3</mn> </msub> <msub> <mi>q</mi> <mrow> <mi>m</mi> <mo>,</mo> <mi>c</mi> </mrow> </msub> <mo>-</mo> <mrow> <mo>(</mo> <msub> <mi>K</mi> <mn>4</mn> </msub> <msub> <mi>t</mi> <mrow> <mi>o</mi> <mi>u</mi> <mi>t</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>K</mi> <mn>5</mn> </msub> <mo>)</mo> </mrow> <mrow> <mo>(</mo> <msub> <mi>q</mi> <mrow> <mi>m</mi> <mo>,</mo> <mi>a</mi> <mi>i</mi> <mi>r</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>q</mi> <mrow> <mi>m</mi> <mo>,</mo> <mi>c</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>+</mo> <msub> <mi>K</mi> <mn>9</mn> </msub> <mi>I</mi> <mo>-</mo> <msub> <mi>K</mi> <mn>14</mn> </msub> <msubsup> <mi>W</mi> <mrow> <mi>f</mi> <mi>r</mi> <mi>e</mi> <mi>e</mi> </mrow> <mrow> <mi>w</mi> <mi>a</mi> <mi>t</mi> <mi>e</mi> <mi>r</mi> </mrow> </msubsup> <mo>+</mo> <msub> <mi>K</mi> <mn>12</mn> </msub> <msub> <mi>t</mi> <mrow> <mi>o</mi> <mi>u</mi> <mi>t</mi> </mrow> </msub> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>6</mn> <mo>)</mo> </mrow> </mrow>
In formula (6), toutAn air temperature is exported for coal pulverizer, I is coal pulverizer electric current, and I is calculated according to formula (7):
I=K6mpf+K7mc+K8 (7)
S4:Calculate coal pulverizer outlet pressure equation:
<mrow> <mi>P</mi> <mo>=</mo> <mrow> <mo>(</mo> <msub> <mi>K</mi> <mn>16</mn> </msub> <mo>+</mo> <msub> <mi>K</mi> <mn>17</mn> </msub> <msub> <mi>m</mi> <mi>c</mi> </msub> <mo>)</mo> </mrow> <msubsup> <mi>q</mi> <mrow> <mi>m</mi> <mo>,</mo> <mi>a</mi> <mi>i</mi> <mi>r</mi> </mrow> <mn>2</mn> </msubsup> <mo>+</mo> <msub> <mi>K</mi> <mn>18</mn> </msub> <msub> <mi>m</mi> <mrow> <mi>p</mi> <mi>f</mi> </mrow> </msub> <msub> <mi>q</mi> <mrow> <mi>m</mi> <mo>,</mo> <mi>a</mi> <mi>i</mi> <mi>r</mi> </mrow> </msub> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>8</mn> <mo>)</mo> </mrow> </mrow>
More than K1~K18It is parameter to be identified;
S5:The data of following measuring point are gathered from the DCS system of power plant:Coal pulverizer inlet coal quality flow qm,c, coal pulverizer inlet The mass flow q of First airm,air, coal pulverizer export an air temperature tout, coal pulverizer electric current I and coal pulverizer outlet pressure P;
S6:The initial parameter of genetic algorithm is set, including generation gap, crossing-over rate, population dimension, aberration rate, maximum genetic algebra and Population at individual number;Then the upper and lower bound of Genetic algorithm searching matrix is set;The parameter of each individual is one in population One-dimensional matrix, the one-dimensional matrix share 22 degree of freedom, and preceding 4 degree of freedom are the raw coal quality m inside coal pulverizerc, coal pulverizer Internal quality of pc mpf, raw coal moisture θCMAn air temperature t is exported with coal pulverizerout, rear 18 degree of freedom are to be identified Parameter K1~K18, the initial value of individual parameter takes at random in the range of the upper and lower bound of Genetic algorithm searching matrix in population Value;
S7:The primary population of initial time genetic algorithm calculates the fitness function value of each individual of primary population:It will be from DCS The coal pulverizer inlet coal quality flow q of acquisitionm,c, coal pulverizer inlet First air mass flow qm,airAnd individual ginseng in population Several initial values is input in formula (1)-formula (8), obtains coal pulverizer electric current I and coal pulverizer outlet pressure P, then according to formula (9) Calculate fitness function:
<mrow> <mi>F</mi> <mi>i</mi> <mi>t</mi> <mo>=</mo> <mfrac> <mn>1</mn> <mi>N</mi> </mfrac> <munderover> <mi>&amp;Sigma;</mi> <mn>1</mn> <mi>N</mi> </munderover> <mrow> <mo>(</mo> <msub> <mi>W</mi> <mn>1</mn> </msub> <mo>|</mo> <mrow> <mfrac> <msub> <mi>t</mi> <mrow> <mi>o</mi> <mi>u</mi> <mi>t</mi> </mrow> </msub> <mrow> <msup> <msub> <mover> <mi>t</mi> <mo>^</mo> </mover> <mrow> <mi>o</mi> <mi>u</mi> <mi>t</mi> </mrow> </msub> <mi>max</mi> </msup> </mrow> </mfrac> <mo>-</mo> <mfrac> <msub> <mover> <mi>t</mi> <mo>^</mo> </mover> <mrow> <mi>o</mi> <mi>u</mi> <mi>t</mi> </mrow> </msub> <mrow> <msup> <msub> <mover> <mi>t</mi> <mo>^</mo> </mover> <mrow> <mi>o</mi> <mi>u</mi> <mi>t</mi> </mrow> </msub> <mi>max</mi> </msup> </mrow> </mfrac> </mrow> <mo>|</mo> <mo>+</mo> <msub> <mi>W</mi> <mn>2</mn> </msub> <mo>|</mo> <mrow> <mfrac> <mi>I</mi> <msup> <mover> <mi>I</mi> <mo>^</mo> </mover> <mi>max</mi> </msup> </mfrac> <mo>-</mo> <mfrac> <mover> <mi>I</mi> <mo>^</mo> </mover> <msup> <mover> <mi>I</mi> <mo>^</mo> </mover> <mi>max</mi> </msup> </mfrac> </mrow> <mo>|</mo> <mo>+</mo> <msub> <mi>W</mi> <mn>3</mn> </msub> <mo>|</mo> <mrow> <mfrac> <mi>P</mi> <msup> <mover> <mi>P</mi> <mo>^</mo> </mover> <mi>max</mi> </msup> </mfrac> <mo>-</mo> <mfrac> <mover> <mi>P</mi> <mo>^</mo> </mover> <msup> <mover> <mi>P</mi> <mo>^</mo> </mover> <mi>max</mi> </msup> </mfrac> </mrow> <mo>|</mo> <mo>)</mo> </mrow> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>9</mn> <mo>)</mo> </mrow> </mrow>
In formula (9), W1、W2And W3For the weights of artificial settings, N is the data sample number gathered from DCS,For toutActual measurement Value,For the measured value of I,For the measured value of P,ForMaximum,ForMaximum,ForMaximum Value;
S8:Selection, restructuring, mutation operator are acted on into group, obtain progeny population;
S9:Calculate the fitness function value of all individuals in progeny population;
S10:Weight insertion operation is performed to progeny population, judges whether genetic algebra reaches the maximum of setting:If not up to, Then go to step S8;If reached, off-line identification process terminates;
Online breakdown judge comprises the following steps:
S11:Individual parameter in the population of the fitness function minimum obtained during off-line identification by genetic algorithm is brought into In formula (1)-formula (8);
S12:Sampling time is set, a coal pulverizer fan-in evidence and defeated is gathered from Power Plant DCS System every the sampling time Go out end data;Coal pulverizer fan-in is according to including coal pulverizer inlet coal quality flow qm,c, air temperature t of coal pulverizer inletinWith The mass flow q of coal pulverizer inlet First airm,air, coal pulverizer fan-out is according to including the air temperature t in coal pulverizer outletout, mill Coal machine outlet pulverized coal mass flow qm,pf, coal pulverizer electric current I and coal pulverizer outlet pressure P;
S13:The coal pulverizer fan-in evidence collected is input to formula (1)-formula to be solved in (8);
S14:Both the output valve of wushu (1)-formula (8) and the coal pulverizer fan-out evidence gathered from DCS are compared, be obtained Residual error ek, then ekIt is compared with the fault threshold set, if ekMore than the threshold value of setting, then it is assumed that system occurs Failure.
2. the coal pulverizer method for diagnosing faults according to claim 1 based on Fast Genetic Algorithm and grey-box model, special Sign is:In the step S5, the sampling time of data acquisition is arranged to 1s, and each measuring point gathers 10000 groups of data.
CN201610209933.0A 2016-04-06 2016-04-06 A kind of coal pulverizer method for diagnosing faults based on Fast Genetic Algorithm and grey-box model Active CN105930302B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610209933.0A CN105930302B (en) 2016-04-06 2016-04-06 A kind of coal pulverizer method for diagnosing faults based on Fast Genetic Algorithm and grey-box model

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610209933.0A CN105930302B (en) 2016-04-06 2016-04-06 A kind of coal pulverizer method for diagnosing faults based on Fast Genetic Algorithm and grey-box model

Publications (2)

Publication Number Publication Date
CN105930302A CN105930302A (en) 2016-09-07
CN105930302B true CN105930302B (en) 2018-05-18

Family

ID=56840162

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610209933.0A Active CN105930302B (en) 2016-04-06 2016-04-06 A kind of coal pulverizer method for diagnosing faults based on Fast Genetic Algorithm and grey-box model

Country Status (1)

Country Link
CN (1) CN105930302B (en)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106439199B (en) * 2016-11-09 2019-04-09 北京化工大学 A kind of control valve failure monitoring method based on DCS data
CN106567964B (en) * 2016-11-09 2018-10-23 北京化工大学 A kind of flow control valve monitoring method based on DCS data
CN109299507A (en) * 2018-08-20 2019-02-01 浙江大学 A kind of different degrees of Analysis on Fault Diagnosis method under the coal pulverizer multi-state based on dynamic modeling
CN109634211B (en) * 2018-12-18 2020-06-30 华中科技大学 Alternating current servo system model identification method based on frequency data and control system
CN110208696A (en) * 2019-06-25 2019-09-06 华北电力大学 Based on model from the coal pulverizer method for diagnosing faults of typical sample of drawing
CN111780149B (en) * 2020-06-01 2022-12-27 大唐东北电力试验研究院有限公司 Remote diagnosis method for equipment state of pulverizing system of thermal power plant
CN113012412B (en) * 2021-03-03 2022-10-18 福建鸿鹄环境发展有限公司 Intelligent data acquisition method and system based on dynamic acquisition statistical analysis of instrument and video data

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101334666A (en) * 2008-07-15 2008-12-31 西安艾贝尔科技发展有限公司 Double-inlet double-outlet steel ball coal mill straight blowing type milling system optimized control method

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101334666A (en) * 2008-07-15 2008-12-31 西安艾贝尔科技发展有限公司 Double-inlet double-outlet steel ball coal mill straight blowing type milling system optimized control method

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
MPS型中速磨煤机建模与仿真;曾德良等;《动力工程学报》;20150115;第35卷(第1期);55-61 *
Nonlinear coal mill modeling and its application to model predictive control;Andrea Cortinovis et al.;《Control Engineering Practice》;20130331;第21卷(第3期);308-320 *
Observer-based fault detection and moisture estimating in coal mills;Peter Fogh Odgaard;《Control Engineering Practice》;20080831;第16卷(第8期);909-911 *
电站磨煤机状态监测与故障诊断的研究;杨雁梅;《中国优秀硕士学位论文全文数据库 工程科技II辑》;20080915;C039-12 *

Also Published As

Publication number Publication date
CN105930302A (en) 2016-09-07

Similar Documents

Publication Publication Date Title
CN105930302B (en) A kind of coal pulverizer method for diagnosing faults based on Fast Genetic Algorithm and grey-box model
CN111308991B (en) Coal mill operation fault identification method and application
CN102566505B (en) Intelligent fault diagnosis method for numerical control machine
CN102182671B (en) State analysis monitoring method of gas compressor
CN110346160B (en) SVR-based coal mill efficiency abnormity detection and diagnosis method
CN108664010A (en) Generating set fault data prediction technique, device and computer equipment
CN101446831B (en) Decentralized process monitoring method
CN109238760B (en) Online monitoring method of coal mill of intelligent power plant coal-fired generator set based on typical correlation analysis and slow characteristic analysis
CN106156485A (en) Method for diagnosing fault of power transformer and device
CN109670242B (en) Transformer winding deformation unsupervised online monitoring method based on elliptical envelope curve
CN103810328A (en) Transformer maintenance decision method based on hybrid model
CN110531303B (en) Batch fault early warning method and system for in-transit intelligent electric energy meters
CN104777830A (en) Multi-work-condition process monitoring method based on KPCA (kernel principal component analysis) mixture model
CN104536439B (en) A kind of method for diagnosing faults based on nested iterations Fei Sheer discriminant analysiss
CN105242155A (en) Transformer fault diagnosis method based on entropy weight method and grey correlation analysis
CN111159844A (en) Abnormity detection method for exhaust temperature of gas turbine of power station
CN104793606A (en) Industrial fault diagnosis method based on improved KPCA (kernel principal component analysis) and hidden Markov model
CN105404979A (en) Multi-source information-based power grid equipment quality rating method
CN112001511A (en) Equipment reliability and dynamic risk evaluation method, system and equipment based on data mining
CN106441843B (en) A kind of rotating machinery fault method for waveform identification
CN105334823A (en) Supervision-based industrial process fault detection method of linear dynamic system model
CN110443481B (en) Power distribution automation terminal state evaluation system and method based on hybrid K-nearest neighbor algorithm
CN109184821B (en) Online monitoring method for closed-loop information analysis of steam turbine of large-scale generator set
CN107895972B (en) Thermal power unit load capacity assessment method and system
CN105989435A (en) Estimation method for maintenance period of equipment based on RCM theory

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant