JP2977227B2 - Recovery boiler combustion control method and combustion monitoring method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial application field) The present invention uses a special fuel called black liquor generated in the pulp manufacturing process to perform a special combustion called char bed combustion to generate steam, and a chemical for chip digestion. The present invention relates to a combustion control method and a combustion monitoring method for a recovery boiler to be recovered.

(Prior Art) In a pulp and paper manufacturing plant, black liquor is discharged as waste liquid from a chip cooking process. The black liquor is concentrated in a concentration device, stored in a dark black liquor tank, guided to a black liquor heater, heated by steam, and then injected into the furnace of a recovery boiler by a black liquor injection gun. The black liquor sprayed in the furnace is suspended and dried, and is deposited on the bottom of the furnace to form a char bed. Thus, combustion air is supplied to burn the organic components in the char bed. The generated heat is heat-exchanged in the furnace and in a heat transfer tube in the exhaust gas passage, and is taken out as steam. On the other hand, the inorganic components in the char bed are reduced in a high-temperature reducing atmosphere on the char bed, and are recovered from Spoutro as a smelt of a chemical raw material for cooking.

By the way, the recovery boiler has two main purposes, namely, the reduction recovery of the cooking chemical by the combustion of black liquor and the generation of steam. Fluctuations in the charbed due to fluctuations in the liquid concentration and composition, changes in the spraying state of the black liquor spray gun, irregular dropping of black liquor spray adhering to the peripheral wall, and burst of ash from superheaters and ash hopper And it is very difficult to maintain stable combustion against these disturbances and to accurately predict them. Although the operation was largely dependent on the experience and intuition of a skilled operator, it was still difficult to operate in good combustion conditions.

(Problems to be Solved by the Invention) In the recovery boiler, it is difficult for the various disturbances as described above to change every moment, and it is difficult to accurately grasp the disturbances online. Conventionally, however, the operation has been presumed by estimating the combustion state by comprehensively monitoring process data representing the state of combustion and the recovery boiler by the operator. Therefore, always operating in a good combustion state requires a great deal of skill, and if the operator misses a change in the combustion state, the recovery operation is delayed, making it difficult to operate the recovery boiler stably and efficiently. There was a possibility that it might be.

The present invention constantly monitors fluctuations in the combustion state of the recovery boiler,
An object of the present invention is to provide a recovery boiler combustion control method capable of automatically performing a recovery operation, always maintaining a good combustion state, and easily realizing stable continuous operation and high efficiency operation of the recovery boiler. is there.

Another object of the present invention is to provide a compact and inexpensive combustion monitoring method for a recovery boiler that reduces the mental fatigue of the operator and accurately predicts the combustion state without the need for subjective judgment. .

(Means for Solving the Problems) For this reason, the present invention provides a recovery boiler for charcoal burning black liquor generated in a pulp manufacturing process to generate steam and recovering a chemical for chip digestion. Based on the instantaneous values and tendency values of the process data representing the state and the shape data of the charbed, an evaluation index of the combustion state is calculated using a statistical method, and a disturbance that changes every moment from the change of the evaluation index. Combustion state prediction means for predicting and diagnosing a change in the combustion state caused by the combustion air, and when the combustion state prediction means determines that the combustion state in the furnace is on a downward trend, the combustion air introduced into the furnace In this case, the combustion state is adjusted by adjusting the supply amount and the target value of the black liquor temperature, and this is used as means for solving the problem.

Further, the present invention provides a recovery boiler that burns black liquor generated in a pulp manufacturing process by charbed combustion to generate steam and recovers a chemical for chip digestion, based on instantaneous values of process data representing a combustion state in a furnace. Combustion state estimation means for calculating the combustion state evaluation index using an aggregate method, predicting and diagnosing a change in the combustion state due to a momentarily changing disturbance based on the change in the evaluation index, the combustion state prediction means Displayed by display means, when the combustion state is predicted by the combustion state prediction means to be in a downward trend,
A buzzer or sound is output by the combustion state alarm means, and this is used as means for solving the problem.

(Operation) Process data that indicates the combustion state in the furnace is constantly monitored,
The combustion state is predicted and diagnosed by the combustion state prediction means based on the instantaneous value, the tendency value and the charbed shape data, and when it is predicted that the combustion state in the furnace is lowered, the combustion state is introduced into the furnace by the combustion state adjustment means. The operating conditions such as the supplied amount of combustion air and the temperature of the black liquor are changed to maintain a good combustion state at all times. By taking such measures, when the combustion state is predicted based on the instantaneous value, tendency value and the charbed shape data of the process data representing the combustion state in the furnace, the combustion state introduced into the furnace is Operating conditions such as air supply and black liquor temperature are manipulated to improve the combustion state and a good combustion state is maintained.

In addition, by taking the above-described means, the fluctuation of the combustion state is quantitatively predicted based on the instantaneous value of the process data representing the combustion state in the furnace, and when the combustion state is decreasing, an alarm is generated. Since the operator is notified, it is possible to prevent the combustion recovery operation from being delayed due to the operator overlooking the tendency of the combustion state to be reduced, and the favorable combustion state is easily maintained. Note that the tendency value is a value indicating whether the current instantaneous value has a rising tendency or a falling tendency, determined from the arrangement of past process data.

(Example) Hereinafter, the present invention will be described with reference to examples in the drawings.

Prior to the description of the embodiments of the present invention, the principle of the present invention will be described. Time-series data from the detectors and converters attached to the recovery boiler when the correlated combustion conditions are good.
By applying a statistical method such as a factor analysis method or a principal component analysis method among multivariate analysis methods to Xi (i = 1, 2,..., P), a variable Xi (i = 1,2,…, p) number p
When fewer m new variables, i.e., common factors, are obtained, a small number of principal components or m common factor scores Fk (m) obtained from eigenvectors of the correlation matrix R of Xi (i = 1, 2,..., P) are obtained. k = 1, 2,..., m) indicates the macroscopic combustion behavior of the recovery boiler. For example, in the principal component analysis method, the common factor score Fk (k = 1, 2,..., M) is Fk = ΣAki · Xi (k = 1, 2,..., M) (1) where Aki (i = 1, p) is represented by an eigenvector for the k-th largest eigenvalue of the correlation matrix R of Xi (i = 1, 2,..., P).

When the behavior of the time series data Xi (i = 1, 2,..., P) of the recovery boiler is represented as a trajectory (trajectory) on the m common factor axes obtained by the equation (1), when the combustion state is good The trajectories of the poor and poor times each show a very different aspect with significantly enlarged features. Therefore, in the feature space on the common factor axis, the combustion state can be divided into a plurality of stages, and the boundary can be represented by an elementary function such as a spherical body equation. Note that, here, the expression of the spherical body for area division is referred to as an evaluation expression, and the magnitude of the radius of the spherical body is referred to as an evaluation index J. The evaluation index J is represented by the following formula.

Here, k is an average value of Fk (k = 1, 2,..., M).
The above processing need only be performed once by a large computer or the like.

Now when new data Xi (i = 1,2, ..., p) is obtained,
The common factor score Fk (k = 1, 2,..., M) is obtained by equation (1),
The evaluation index J is calculated from the equation (2). Using the evaluation index J and the time series data when the combustion state is good, the combustion state is predicted and diagnosed in a plurality of stages according to the magnitude of a predetermined evaluation index Jo and the direction of change of the evaluation index J. .

Hereinafter, a first embodiment of the present invention will be described with reference to FIGS. 1 and 8. FIG. 1 is a system diagram showing a schematic configuration of a recovery boiler in which a combustion control device 29 is applied to a combustion control method of the present invention. is there. In FIG. 1, black liquor A discharged from a chip digestion step is passed through a dark black liquor tank 2 and heated by steam flowing through a steam line 4 by a black liquor heater 3 into a recovery boiler 1. Injected into the furnace 1 by the injection gun 6,
The black liquor A 'sprayed into the furnace 1 floats and dries and falls, forming a char bed 8 at the bottom of the furnace.

This char bed 8 is provided with an upper combustion air supply line.
Combustion is promoted by the combustion air B1, B2, B3 supplied from 7a, 7b, 7c. At this time, a reduction reaction is performed in a high-temperature reducing atmosphere on the char bed 8, and smelt C, which is a raw material of the cooking chemical, is recovered from the spout tro 9 at the bottom of the furnace. Further, the combustion exhaust gas D generated by the combustion of the char bed 8
Is discharged from the chimney 15. On the other hand, water E supplied from the water supply line 10 undergoes heat exchange with combustion exhaust gas in each of the heat transfer sections of the economizer 11, the main body bank 12, and the superheater 13, and is taken out as high-temperature and high-pressure steam F from the steam line 14. It is.

Further, the combustion control device 29 includes black liquor concentration data PV1 detected by the black liquor concentration detector 17 provided in the black liquor passage as the process data representing the combustion state in the furnace, the recovery boiler 1
The furnace temperature data PV2 detected by the furnace temperature detector 20 provided on the furnace wall and the like, and the SO _X , O ₂ , NO of the exhaust gas detected by the exhaust gas detector 21 provided in the exhaust gas passage. _{The X} and CO concentration data PV3 and the charbed shape data PV4 detected by the image processing device 19 are fetched. Then, as shown in FIG. 4, the combustion control device 29 takes in a plurality of process data PV1, PV2, PV3 representing the combustion state in the furnace of the recovery boiler 1, and a combustion prediction signal determined from these instantaneous values and tendency values. R1, R2, R3
And a function of predicting and diagnosing the combustion state in a plurality of stages based on the charbed shape determination signal PV4, and a combustion air supply line according to the degree of the combustion state decrease when the combustion state is predicted to decrease. It has a function of changing the target values SV1, SV2, SV3 for the amount of air supplied to the furnace from 7b, 7c and the black liquor temperature in a direction in which the combustion state is recovered. On the other hand, the air amount target values SV1 and SV2 are
The air amount detector 2
The air damper is adjusted so that the air amount detected at 5, 27 matches the target value. The black liquor temperature target value SV3 is given to the steam amount controller 22, and the opening of the steam amount control valve 5 is adjusted so that the black liquor temperature detected by the black liquor temperature detector 16 matches the target value. You.

FIG. 3 is a block diagram showing a specific configuration of the combustion control device 29. The process data processing unit 32 takes in the process data PV1, PV2, PV3, and PV4 at a predetermined cycle, and in particular, performs a filtering process on the process data PV3, converts the process data into engineering values, and stores the result in the process data storage unit.
33, process data PV1 is PV1 ', PV2 is PV2', PV3 is PV
3 'and PV4 are stored as PV4'. The allowable value data setting unit 34 sets the allowable upper limit level and the allowable lower limit level of the process data PV1, PV2, PV3 to the allowable value data LVi (i =
1, 6) are set, and each set allowable value data LVi (i = 1, 6) is stored in the allowable value data storage unit 35. The adjustment rule setting unit 30 includes a plurality of adjustment rules JRi (i = 1, 1) based on the experience and knowledge of a skilled operator and design data.
6) is set, and the set adjustment rules are stored in the adjustment rule storage unit 31. The combustion state predicting section 36 has a specific circuit configuration shown in FIG. That is, the process data PV1 ′, PV2 ′,
PV3 'and PV4' are read, and process data PV1 'and PV
An abnormality determination of 2 'and PV3' is performed, a logical sum of the combustion prediction signals R1, R2, and R3 is obtained, and an ON / OFF output signal S is sent to the combustion state signal output circuit 41. The output signal S is PV1 ', PV2', PV3 '
Are turned on when at least one of them is determined to be abnormal combustion, and turned off when all of them are determined to be normal combustion. Next, the combustion state is divided into four states based on the charbed shape data PV4 'and the output signal S, and the combustion state signal T
Is output. The charbed shape data PV4 'is a contact input signal that is turned on when the shape of the charbed is out of the proper range, and turned off when it is out of the proper range.

The combustion predicting circuit 39 specifically has a functional configuration shown in FIG. That is, the process data PV1 'is read from the process data storage unit 33, and the change amount of the process data is first detected by the change amount detecting means 42. The amount of change represents a tendency value in which the behavior of the time series data of the process is expressed as a trajectory drawn on the m common factor axes obtained by the factor analysis, and the value of the next time is predicted. Based on the amount of change, the output signal Q11 is output by the sign determination unit 43,
When the amount of change is positive, it is a positive value, when it is negative, it is a negative value, and when it is zero, it is 0.
Is set. Next, the output signal Q12 of the output signal Q12 is read by the comparing means 44 from the allowable value data storage unit 35 and compared with the allowable value data corresponding to the process data. A negative value is set when the value is below the lower limit level, and a value of 0 is set when the value is within the allowable value. Finally, the combustion prediction signal R1 is converted by the combustion prediction signal means 45 into the output signal Q1.
When 1 is a positive value and Q12 is a positive value, or when the output signal Q1
It is turned on when 1 is a negative value and the output signal Q12 is a negative value. More specifically, the combustion state signal output circuit 41 determines the combustion state into four states according to the processing flow shown in FIG.

The combustion state adjusting unit 37 calculates target values V1, V2, and V3 for the supply amount of the combustion air for recovering the combustion state and the black liquor temperature using the combustion state signal T as a trigger. Has the circuit configuration shown in FIG. Further, the combustion state adjusting section 37 fetches the combustion state signal T at a predetermined cycle, and the signal T stores the previous status value U1 and the current status value U2 in the combustion state storage circuit 46. The adjustment operation circuit 47 first receives the process data PV1 ', PV2', PV3 ', PV4' from the process data storage unit 33 and the previous combustion status value U1 and the current combustion status value U2 from the combustion state storage circuit 46 at a predetermined cycle. Is read. Next, the combustion status value
The adjustment rule JRi corresponding to the set of the values of U1 and U2 is read from the adjustment rule storage unit 31 based on FIG. Combustion status value
U1 and U2 take the values of "1" in the combustion state 1, "2" in the combustion state 2, "3" in the combustion state 3, and "4" in the combustion state 4 shown in FIG. Has become. The optimal adjustment calculation is performed using the read process data PV1 ', PV2', PV3 ', PV4' and the adjustment rule JRi, and the target values V1, V2, V3 for the supply amount of the combustion air and the black liquor temperature are calculated. Is calculated. The output setting unit 38 receives the output signals V1, V2, V3 of the combustion state adjusting unit 37, and sets the target values SV1, SV2, SV3 to the target value V.
It is designed to be changed stepwise at regular intervals and at a constant level up to 1, V2, V3.

Next, a second embodiment of FIGS. 2 and 11 will be described. FIG. 2 is a system diagram showing a schematic configuration of a recovery boiler in which the combustion monitoring device 31 is applied to the combustion monitoring method of the present invention. In FIG. 2, a black liquor A discharged from a chip digestion step is passed through a dark black liquor tank 2 and heated by a black liquor heater 3 with steam flowing through a steam line 4 in a recovery boiler 1. The black liquor A 'sprayed into the furnace 1 and sprayed into the furnace 1 floats, dries, and falls to form a char bed 8 at the bottom of the furnace. This char bed 8 is provided with combustion air B1, B supplied from upper combustion air supply lines 7a, 7b, 7c.
2, B3 promotes combustion. At this time charbed 8
The reduction reaction is performed in the above high-temperature reducing atmosphere, and smelt C, which is a raw material of the cooking chemical, is recovered from the spout tro 9 at the bottom of the furnace.

Further, the flue gas D generated by the combustion of the char bed 8 is discharged from the chimney 15. On the other hand, water E supplied from the water supply line 10 undergoes heat exchange with combustion exhaust gas in each of the heat transfer sections of the economizer 11, the main body bank 12, and the superheater 13, and is taken out as high-temperature and high-pressure steam F from the steam line 14. .
The combustion monitoring device 58 has black liquor flow rate data X ₁ detected by the black liquor flow rate detector 16 provided in the black liquor passage, and black liquor concentration detector 17 as process data representing the combustion state in the furnace. The detected black liquor density data X ₂ , the charbed shape data X ₃ detected by the image processing device 19, and the furnace temperature detected by the furnace temperature detector 20 provided on the furnace wall of the recovery boiler 1. Temperature data X ₄ , steam flow detector 48 provided in steam line 14
The steam flow rate data X ₅ detected by the exhaust gas detector 49, 50, 51 provided in the exhaust gas flow path, O ₂ , N
O _X and CO concentration data X ₆ , X ₇ , X ₈ , combustion air supply line 7
The primary air flow data X ₉ detected by the air flow detector 52 provided in a, the primary air draft data X ₁₀ detected by the air differential pressure detector 53, and provided in the combustion air supply line 7b The secondary air flow data X ₁₁ detected by the detected air flow detector 54, the secondary air draft data X ₁₂ detected by the air differential pressure detector 55, the air provided in the combustion air supply line 7 c capture the flow detector tertiary air flow rate data X ₁₃ detected by 56 and differential air detected by the pressure detector 57 tertiary air draft data X _14.

FIG. 9 is a block diagram showing a specific configuration of the combustion monitoring device 58. The process data processing unit 59 reads the process data Xi (i = 1, 2,..., 14) at regular intervals,
Especially after process data X _6, X _7, X ₈ is subjected to the filtering process, is converted into engineering values, resulting process data X ₁ in the process data storage unit 60 is X ₁ ', X ₂ is X _2', …, X ₁₄ is X
Stored as ₁₄ '. The evaluation index setting section 61 is a predetermined evaluation index using time series data in a good combustion state.
Jo is set, and the set evaluation index Jo is stored in the evaluation index storage unit 62. The combustion state predicting unit 63 has a specific circuit configuration shown in FIG. That is, the process data Xi '(i = 1, 2,
.., 14) are read out, first, the factor score Fk (k = 1, 2,..., M) is calculated by the operation formula (1), and then the evaluation index J is calculated by the operation formula (2).

Then reads the evaluation index Jo from the evaluation index storage unit 62, as compared with the evaluation index J by comparing means 66, an output signal Y ₁ For JJo is a positive value, J <For Jo output signal Y ₁ Is set to a negative value. Next, change amount detection means
The amount of change in the evaluation index J is detected by 67, and the output signal Y ₂ by the sign determination means 68 is a positive value when the amount of change is positive,
When the value is negative, a negative value is set, and when the value is zero, a value of 0 is set. Finally combustion prediction signal Z by the combustion predicted output unit 69 is made is set to the value shown in Figure 11 by the value of the output signal Y ₁ and Y _2. The combustion state display unit 64 performs color display on the CRT display device in accordance with the value of the output signal Z of the combustion state prediction unit 63. When the value of the output signal Z is 1, the combustion state alarm section 65 outputs a buzzer or a sound. The present invention is not limited to the above embodiment. For example, in the embodiment, black liquor flow rate data, black liquor concentration data, furnace temperature data, as process data representing the combustion state in the furnace,
Evaporation amount data, exhaust gas O _2, NO _X, the density data of CO, air flow rate data, air draft data showed the char bed shape data, is not limited to,
Of course, other data and other synthetic variables may be used. In the figure, 70 is a factor score calculation, and 71 is an evaluation index calculation.

(Effects of the Invention) As described in detail above, the present invention is based on the instantaneous value, tendency value, and charbed shape data of the process data representing the combustion state in the furnace at all times. It is predicted whether or not further improvement can be made, and when it is predicted and diagnosed that the improvement can be made, the combustion state adjusting means according to the experience knowledge of the skilled operator and the adjustment rule based on the setting data, the combustion air supply amount and the black Adjustments are made by changing operating conditions such as liquid temperature, so that fluctuations in the combustion state can be quickly and accurately evaluated with respect to various disturbances.
Since the target value for improving the combustion state has been adjusted,
A combustion control method for a recovery boiler that can maintain a good combustion state at all times and can easily realize a stable operation of the recovery boiler and a high-efficiency operation can be provided.

Further, according to the combustion monitoring method of the present invention, the combustion state of the recovery boiler can be accurately predicted and diagnosed, and when there is a correlation from the existing detector without installing a special detector for the prediction diagnosis. Predictive diagnosis can be made using the series data, so that it is easy to implement and the cost is low. Further, the prediction diagnosis result can be easily displayed to the operator, and when the combustion state is predicted to be decreasing, the combustion state can be accurately transmitted to the operator by a buzzer or an audio output. it can. As described above, according to the combustion monitoring method of the present invention, the combustion state can be accurately predicted and diagnosed at an early stage with respect to various disturbances, which is extremely useful in industry.

[Brief description of the drawings]

FIG. 1 is a system diagram showing a combustion control method for a recovery boiler according to a first embodiment of the present invention, FIG. 2 is a system diagram showing a combustion monitoring method for a recovery boiler according to a second embodiment of the present invention, and FIG. Is a block diagram showing a specific configuration of the combustion control device in the first embodiment, FIG. 4 is a block diagram showing a specific configuration of the combustion state prediction unit, and FIG. 5 is a block diagram showing a specific configuration of the combustion prediction circuit. FIG. 6 is a flowchart showing a schematic process of the combustion state signal output circuit, FIG. 7 is a block diagram showing a specific configuration of the combustion state control unit, and FIG. FIG. 9 is a block diagram of a combustion monitoring device according to a second embodiment, FIG. 10 is a block diagram showing a specific configuration of the combustion state prediction unit, and FIG. FIG. 4 is an explanatory diagram for describing an output signal of a prediction unit. Explanation of main parts in the figure 1 ... Recovery boiler 2 ... Dark black liquid tank 3 ... Black liquid heater 4 ... Steam line 5 ... Steam amount control valve 6 ... Black liquid injection gun 7a, 7b, 7c ... Combustion air supply line 8 Char bed 9 Spoutro 10 Water supply line 11 Energy saving device 12 Main unit bank 13 Super heater 14 Steam line 15 Chimney 16 Black liquid temperature detection Unit 17 Black-concentration detector 18 Furnace monitoring camera 19 Image processing unit 20 Furnace temperature detector 21 Exhaust gas detector 22 Steam controller 23 Primary air flow Detector 24 Primary air flow controller 25 Secondary air flow detector 26 Secondary air flow controller 27 Tertiary air flow detector 28 Tertiary air flow control Total 29: Combustion control device 30: Adjustment rule setting unit 31: Adjustment rule storage unit 32: Process data processing unit 33: Process data storage unit 34 Permissible value data setting section 35 Permissible value data storage section 36 Combustion state predicting section 37 Combustion state adjusting section 38 Output setting section 48 Steam flow rate detector 49, 50, 51 Exhaust gas detector 52 …… Air flow detector 53,55,57 …… Air differential pressure detector 54 …… Flow detector 56 …… Air flow detector 58 …… Combustion monitoring device

Claims (2)

(57) [Claims] 1. A recovery boiler that burns black liquor generated in a pulp manufacturing process by char bed combustion to generate steam and recovers a chemical for chip digestion. Based on the values and the charbed shape data, a combustion state evaluation index is calculated using a statistical method, and the combustion state is predicted and diagnosed based on the change of the evaluation index to predict the fluctuation of the combustion state due to a momentarily changing disturbance. When the combustion state predicting means determines that the combustion state in the furnace has a tendency to decrease, a target value of the supply amount of the combustion air introduced into the furnace and the black liquor temperature is provided. A combustion control method for a recovery boiler, wherein the combustion state is adjusted by adjusting pressure. 2. A recovery boiler which burns black liquor generated in a pulp manufacturing process by charbed combustion to generate steam and recovers a chemical for chip digestion based on instantaneous values of process data representing a combustion state in a furnace. Combustion state estimation means for calculating the combustion state evaluation index using an aggregate method, predicting and diagnosing a change in the combustion state due to a momentarily changing disturbance based on the change in the evaluation index, the combustion state prediction means A combustion monitoring method for a recovery boiler, characterized by displaying on a display means and outputting a buzzer or a sound by a combustion state warning means when the combustion state is predicted to be decreasing by the combustion state prediction means.