JP2590261B2 - Char bed shape control device for recovery boiler - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial application field) The present invention relates to a charcoal shape control device of a recovery boiler used for operation management of a recovery boiler in a pulp manufacturing plant, and in particular, to recovery. The present invention relates to a charbed shape control device for a recovery boiler that controls the position, shape, and the like of a charbed in a boiler.

(Prior Art) In general, a recovery boiler is used in a pulp manufacturing plant from the viewpoint of energy saving and labor saving.
In this recovery boiler, black liquor discharged as waste liquid in the chip digestion process is injected into the furnace using a black liquor injection gun, and at this time, the black liquor drops to the furnace bottom while floating and drying due to the temperature in the furnace. Since a char bed is formed, air is sent into the furnace to burn the char bed, thereby generating steam from the furnace upper side and, at the bottom of the furnace, a chemical material for chip cooking obtained by a reduction reaction generated at the time. It is configured to be collected from the side.

By the way, in recent years, in a pulp manufacturing plant, it is important to maintain a proper position and shape of a char bed and to stably operate a recovery boiler and improve thermal efficiency in order to promote energy saving and labor saving. It has been recognized that there is.

Therefore, conventionally, a viewing window was provided on the furnace wall of the recovery boiler,
The operator monitors the position and shape of the charbed from the viewing window while patroling around the recovery boiler from time to time, and if necessary, manually adjusts the black liquor injection amount, black liquor temperature, combustion air flow rate, etc. based on past experience etc. The position and shape of the charbed are controlled while making appropriate adjustments.

(Problems to be Solved by the Invention) However, the above-mentioned shape bed control means manually adjusts the black liquor temperature and the like injected into the furnace while monitoring the position and shape of the char bed artificially. Therefore, the position and shape of the charbed cannot be constantly monitored. As a result, for example, the shape of the charbed often becomes too high or steep, and conversely, the shape of the charbed often becomes too low. Here, if the charbed is too high, combustion failure occurs due to the collapse of the charbed, and if it is too low, there is a problem that the recovery efficiency of the chemical material for chip digestion is reduced and carryover of dust is caused. .

In addition, the operator can see the position of the char bed from the viewing window,
In order to monitor the shape, if there is a lack of proper attention or the viewing window is slightly cloudy, the position and shape of the charbed may be misaligned or the charbed may be locally raised or depressed. I can't figure out. As a result, for example, when the char bed is not aligned, blackout occurs due to a blockage of the air inlet port, a large amount of unburned carbon flows out of the spout trolley, or dust fusion occurs due to an increase in the gas temperature in the recovery boiler. In addition, if the charbed is in a position or shape such as a local bulge or dent, there is a risk of combustion fluctuations, and the charcoal bed may move in a short time. It is difficult to accurately grasp the phenomenon in advance.

Furthermore, if the charbed collapses or deposits fall from the furnace upper part and the furnace side wall, the charbed partially causes poor combustion, so-called blackout. In addition, if the detection of the abnormal state is delayed, a quick recovery process becomes impossible, and there is a risk that the operation of the plant may be adversely affected.

Accordingly, the manual adjustment work not only requires a large burden of labor but also makes it difficult to achieve stable operation and high efficiency of the plant because the position and shape of the charbed cannot be grasped with high accuracy.

The present invention has been made in view of the above circumstances, and can constantly and automatically monitor the position and shape of the bed, can quickly determine an abnormal state of the bed and take appropriate measures, and thus, the recovery boiler An object of the present invention is to provide a charbed shape control device of a recovery boiler that can achieve stable operation and high efficiency and can always operate a plant in an optimal state.

[Constitution of the Invention] (Means for Solving the Problems) In order to solve the above problems, the present invention is to inject black liquor discharged from a chip cooking step of a pulp manufacturing plant into a furnace by a black liquor injection gun. A charcoal bed is formed at the bottom of the furnace, and combustion air is injected into a plurality of locations in the furnace to burn the charbed, thereby generating the above and collecting the chemical raw material for chip digestion. Charbed shape detecting means for detecting the position and shape of the bed, and multi-point information for each pixel representing the presence or absence of the charbed from two-dimensional image data related to the position and shape of the charbed obtained by the charbed shape detecting means Image signal converting means for converting, and a switch having a predetermined height, left and right position, and width in advance in terms of stable operation and recovery efficiency of the recovery boiler. -The synapse variable coupling coefficient of each neuron element is set so as to be the position and shape of the bed, and the height, left and right position and width deviation are determined and output based on the multipoint information input from the image signal conversion means. A neural network having a multi-layer structure with a learning function to perform, and at least an operation element that affects at least the height, the left / right position, and the width of the charbed based on the discrimination result of the neural network, and a variable position of the proper charbed. And a control device for controlling the shape of the char bed of the recovery boiler provided with operation control means for correcting the shape of the recovery boiler.

(Operation) Therefore, according to the present invention, when the position and shape of the charbed are detected by the charbed shape detecting means and input to the image signal converting means, the image signal converting means The two-dimensional image data relating to the position and shape of the bed is converted into multi-point information for each pixel representing the presence or absence of a char bed, and is introduced into the input layer of the neural network. This neural network stores in advance the synapse variable coupling coefficient of each neuron element so that the charbed has a predetermined height, left and right position, and width, in terms of stable operation and recovery efficiency of the recovery boiler. Therefore, by taking in and calculating the multipoint information for each pixel, it is possible to immediately discriminate and output the height, the left / right position, and the deviation of the width at the same time, for example, operating elements such as injection black liquor, combustion air and auxiliary fuel. By selectively changing the position and shape of the bed, it is possible to quickly correct the position and shape of the bed from a multifaceted surface, so that the position and shape of the bed can always be formed properly, and the position and shape of the bed can be properly adjusted. The learning and holding system achieves stable operation and high efficiency of the recovery boiler.

(Embodiment) Hereinafter, in describing an embodiment of the apparatus of the present invention, first, a principle for understanding the present invention will be described.

When the black liquor discharged from the chip digestion process is heated by a black liquor heater and then injected into the furnace with a black liquor injection gun, when the black liquor heater is operated to raise the black liquor temperature to near the boiling point In addition, the black liquor injected from the black liquor injection gun is dried while floating in the furnace and burns at a high rate, thereby reducing the amount of bed of the char bed. Therefore, in this case, the top of the charbed can be lowered. Conversely, when the black liquor temperature is lowered,
It becomes difficult to burn while floating, and the top of the charbed can be raised.

Usually, the black liquor injection gun is fixed to the furnace wall. For example, if the black liquor injection gun is installed to be able to swing and the black liquor injection angle is directed upward or the injection pressure is increased,
The black liquor injected from the black liquor injection gun descends while floating far away. As a result, the charbed moves to the opposite furnace wall side, and the floating time becomes longer, so that the amount of bed of the charbed is reduced and the charbed can be lowered. Conversely, if the angle of the black liquor injection gun is turned downward or the pressure of the black liquor is included, the char bed can be moved toward the furnace wall on the black liquor injection gun side, and the char bed can be raised. Can be.

Attach black liquor injection guns to the front and rear furnace walls of the recovery boiler, respectively, to reduce the black liquor injection amount of the front black liquor gun and increase the black liquor injection amount of the opposite black liquor injection gun Then, the char bed can be moved to the opposing side wall surface.

Increasing the combustion air flow rate improves the flammability and lowers the char bed, while decreasing the air flow rate lowers the flammability and increases the char bed.

When a plurality of combustion air inlets are provided in the height direction of the recovery boiler, if the flow rate of air supplied from the combustion air inlet of the stage corresponding to the top of the charbed is relatively increased, the top of the charbed is The flammability of the part is improved, so that the height of the charbed can be suppressed and the top of the charbed can be made flat.

When the flow rate of air supplied from the combustion air inlet of the stage corresponding to the bottom of the charbed is relatively increased, the combustion at the bottom of the furnace is improved, and the bottom of the charbed can be separated from the furnace wall, and It can be shaped into a steep char bed. Conversely, if the flow rate of air supplied from the combustion air inlet of the stage corresponding to the bottom of the charbed is relatively reduced, combustion at the furnace bottom becomes worse, and the bottom of the charbed can be drawn to the furnace wall, And it can be shaped into a gentle char bed.

Also, increase the amount of air to be injected from a certain furnace wall side,
When the amount of air introduced from the opposite furnace wall side is reduced, the combustion on the side where the amount of air in the charbed is increased becomes active,
The combustion on the side where the air amount is reduced becomes inactive. As a result, the char bed can be brought closer to the opposing wall surface.

Among the first furnace walls facing each other, the amount of air introduced from one furnace wall is increased, the amount of air introduced from the other furnace wall is decreased, and the first air on the side orthogonal to the first furnace wall is further reduced. The char bed can be moved to the other furnace wall side as the amount of air supplied from the furnace wall side decreases as the distance from the one furnace wall increases, that is, as the air amount decreases toward the other furnace wall.

If the charbed has irregularities, the amount of air injected from the combustion air inlet near the convex part is increased to improve the combustion of the convex part, and the flow rate of air injected from the combustion air inlet near the concave part is reduced to reduce the amount of air. The charbed surface can be shaped into a smooth state by reducing the combustion of the charbed.

When heavy oil or the like is burned in the upper part of the bed, the radiant heat activates the burning of the projections on the bed surface, so that the bed surface can be shaped into a smooth shape. In addition, since the entire combustion of the charbed is improved, the charbed can be lowered as a whole.

By increasing the combustion of the combustion burner on a specific wall and reducing the combustion of the combustion burner on the opposing wall, the charbed combustion becomes active on the side with more combustion, and the charbed is moved to the opposing wall. Can be.

The present invention has been realized based on the above principle,
Hereinafter, the embodiment will be described with reference to the drawings. FIG. 1 is a system configuration diagram showing one embodiment of the present invention. In the figure, reference numeral 10 denotes a recovery boiler for producing and recovering steam and a chemical material for chip cooking, which is not shown in the drawing but is a black liquor discharged as waste liquid in the chip cooking process.
11 is injected into the furnace by a black liquor injection gun 14 via a black liquor heater 12 and a black liquor injection pressure control valve 13. The black liquor jetted by the black liquor jet gun 14 is float-dried and landed to form a char bed 15. Therefore, in this black liquor supply system, the black liquor 11 from the chip digestion step is heated to a desired temperature by the black liquor heater 12, and is adjusted to a desired pressure or flow rate by the black liquor injection pressure control valve 13, The gun is operated at a desired injection angle from the black liquor injection gun 14 by the gun angle operation valve 16. 17 is a heating medium and 18 is a black liquor temperature control valve. Although only one black liquor spray gun 14 is shown in the figure, a plurality of black liquor spray guns 14 are usually provided at different positions on the furnace wall.

The recovery boiler 10 is provided with a combustion air supply system. This combustion air supply system is provided with combustion air inlets 21a, 21b, 21c over a plurality of stages in the vertical direction of the furnace wall,
Also, as shown in the lower part of the figure, a plurality of combustion air inlets 21a,..., 21b,.
b, 21c are not shown). That is, the plurality of combustion air inlets are grouped into a plurality of units, and the combustion air operating means 22 is grouped for each group.
Is provided. The combustion air operation means 22 has a function of charging the combustion air 23 into the furnace and burning the charbed 15 while appropriately controlling the air flow rate of the combustion air 23 to each combustion air inlet.

Further, the recovery boiler 10 is provided with an auxiliary fuel supply system. In this auxiliary fuel supply system, an auxiliary burner 31 is attached to the furnace wall of the recovery boiler 10, and an auxiliary fuel 32 such as heavy oil is supplied to the auxiliary burner 31 via an auxiliary operation valve 33.

Reference numeral 40 denotes a charbed shape detecting means for monitoring the position and the shape of the charbed 15 in the furnace, for example, using an ITV camera, a scanning pilot, or the like. Is output. Note that a plurality of such bed detecting units 40 are mounted on the same wall surface or a plurality of wall surfaces as necessary.

Reference numeral 41 denotes image signal conversion means for converting the image signal a sent from the charbed shape detection means 40 into a pixel (pixel) signal h. More specifically, the image signal a is divided into meshes and Determine the presence or absence of char bed 15,
It has a function of outputting, for each pixel, a signal corresponding to “1” when there is a charbed, and a signal corresponding to “0” when there is no charbed.

42 is a pixel signal h sent from the image signal conversion means 41
This is a neural network that determines whether the position and shape of the char bed 15 are appropriate in response to the request. That is, as shown in FIG. 2, the neural network 42 has a multilayer structure of an input layer, at least one intermediate layer, and an output layer, and each layer uses a plurality of neuron elements. The input layer is composed of I neuron elements 43 ₁ , 43 ₂ ,..., 43 _I corresponding to the total number of pixels.
A signal h indicating the presence or absence of the char bed of each pixel is supplied to the corresponding neuron elements 43 ₁ , 43 ₂ ,..., 43 _I. On the other hand, the middle layer is M
Neurons elements 44 _1, 44 _2, ..., 44 is composed of _M, the output of all or required number of neuron elements in the input layer is inputted respectively with the synaptic variable coupling coefficient as desired,
Here, the following equation (1) is calculated and output. Next, the output layer is composed of P neuron elements 45 ₁ , 45 ₂ ,..., 45 _P , and the outputs of all or the required number of neurons in the intermediate layer are input with desired synapse variable coupling coefficients. Here, the calculation is performed by the following equation (1), and the output is obtained. In particular, the neuron elements 45 ₁ , 45 ₂ ,
..., a combination of 45 to 45 _1, 45 ₂ of the _P "1", "0" combination char bed height for determination signal consists of, also from 45 _3, 45 ₄ "1", "0" right Katayose for discriminating signals of the char bed 15, also from 45 _5, 45 ₆ "1", "0" combined width for discriminating signals of the char bed 15 consisting of, ..., from 45 _P "1", " A determination signal indicating the presence / absence of abnormalities in the charbed shape detection means of 0 "is output.

As shown in FIG. 3, each neuron element constituting the intermediate layer and the output layer includes a plurality of outputs xj of the preceding layer.
(J = 1,2, ..., N) are the synaptic variable coupling coefficients wj
(J = 1,2,..., N), and when xj · wj is input, the plurality of xj · wj are added and the sum is subtracted by a threshold θ to obtain an output function. f (eg sigmoid function) And outputs it to each neuron element in the next layer. Further, the values of the synapse variable engagement coefficient w and the threshold value θ of each neuron element constituting the intermediate layer and the output layer are determined based on the position and shape of the char bed 15 based on the principle described above. Is determined in advance through a learning function so that is obtained, and it is appropriately changed while learning is performed thereafter. That is, a test signal or each pixel signal h of an actual image signal accompanied by the position and shape of the char bed 15 is input to the input layer, and the output of the output layer at that time is compared with the teacher signal to output an erroneous output. If you judge that
A properly functioning neural network 42 is constructed by correcting the values of w and θ little by little using a back propagation learning rule and repeating the correction until a correct discrimination result is obtained.

Reference numeral 50 denotes a charbed 15 having an appropriate position and shape based on the discrimination signal i sent from the neural network 42.
This is an operation control means for outputting an operation signal for obtaining the determination signal i. Specifically, as shown in FIG. 4, the discrimination signal i from the neural network 42 is varied for each individual operation element by a logical operation. A logic circuit 51 that converts the signals into necessary signals, and selects one of the operation valves 13, 18, 33 and the operation means 16, 22 based on the output of the logic circuit 51, and issues command signals b, c,
Command signal generating means 52b to 52f for outputting d, e, and f are provided.

Next, the operation of the device configured as described above will be described. The black liquor 11 discharged from the chip cooking process is a black liquor heater
After being heated at 12, the required pressure or flow rate is set by the black liquor injection pressure control valve 13 and further the gun angle control means 16
At the required injection angle. At this time,
The sprayed black liquor in the furnace is suspended and dried, and is then placed on the bottom of the furnace.
A char bed 15 as shown in the figure is formed. At this time, the combustion air inlets 21a, 21a are
When the required amount of combustion air 23 is supplied from b and 21c, steam is generated by the combustion of the char bed 15, and a chemical reaction material for chip digestion can be recovered by a reduction reaction occurring at that time.

When the charbed 15 is burned, one or more charbed shape detecting means 40 captures an image signal a indicating the position and shape of the charbed 15 in the measurement visual field, and then this image signal a is used as a subsequent image. The signal is sent to the signal conversion means 41. The image signal conversion means 41 decomposes the image signal a into a pixel signal h, and outputs a signal of "1" for each pixel when there is a charbed and a signal of "0" for no pixel of the neural network 42. The input neuron elements 43 ₁ , 43 ₂ ,..., 43 _I are individually supplied to predetermined input layer side neurons 43 ₁ , 43 ₂ ,.

Here, each of the neuron elements 43 ₁ , 43 ₂ ,..., 43 in the input layer subtracts a signal obtained by multiplying the pixel signal h with or without the charbed by the synapse variable coupling coefficient w to a threshold θ, and obtains a signal obtained by subtracting the signal. To the entire middle layer or the required number of neuron elements 44
_1, 44 _2, ..., and sends it to 43 _M. The neuron elements in the intermediate layer and the output layer perform the above-described signal processing, and output a discrimination signal as a processing result from each of the neuron elements 45 ₁ , 45 ₂ ,..., 45 _{P in} the output layer. Incidentally, it is assumed that the dotted line (b) in FIG. 5 (a) is an appropriate shape of the char bed 15, and for example, when the shape of the char bed 15 is a solid line (a) in the figure, the neuron elements 45 ₁ , 45 in the output layer ₂ outputs a “1” signal, and outputs a discrimination signal indicating that the so-called char bed 15 is too high. Conversely, in the case of FIG. 5B, the neuron elements 45 ₁ and 45 ₂ respectively output A "0" signal is output, and a discrimination signal indicating that the so-called char bed 15 is too low is output. If none of these belongs, "1",
Outputs a "0" signal. Further, when the char bed 15 if is offset to the left as in the fifth diagram (c) of outputting each example, "1" signal from the neuron element 45 _3, 45 _4, are offset to the right in the opposite Is the neuron element 45 ₃ , 45 ₄
Output a “0” signal from each of them, and if they do not belong to any of them, “1”,
Outputs “0” signal. Further, each output layer neuron elements 45 _5, 45 ₆ in the case the width of the char bed is too wide "1" signal outputs, respectively when determined too thin reversed to be a steep shape " A signal "0" is output, and if it does not belong to any of these, "1", "0", that is, a determination signal indicating that the shape is appropriate is output. Furthermore, for example, the char bed shape detecting unit "1" from the neuron element 45 _P when the detection window is determined to dirty 40 outputs a signal, and outputs a "0" signal otherwise.

Therefore, when the above-described determination signal is output from the neural network 42, the logic circuit of the operation control means 50
At 51, the discrimination signal “1” sent from the output layer,
After confirming "0", predetermined command signal generating means 52b to 52f are selected and a command signal for changing the operation element is transmitted.

That is, the logic circuit 51 of the operation control means 50, for example, when a determination signal indicating that the shape of the char bed 15 is too high is input from the output layer, sets the black liquor temperature set values to the command signal generation means 52b and 52e, respectively. A command signal for increasing the pressure and a command signal for increasing the air flow rate at the top of the charbed are transmitted. As a result, an operation output b for increasing the valve opening is sent from the command signal generating means 52b to the black liquor temperature control valve 18, and the black liquor heater 12 is controlled to increase the temperature of the black liquor 11 (the principle described above). Section). Further, a signal e for increasing the air flow rate at the top of the char bed is sent from the command signal generating means 52e to the combustion air operating means 22. Here, the combustion air operation means 22 controls the total air flow rate of the combustion air 23 so as to send more air flow rate to the combustion air input port 21a near the top of the charbed than the lower combustion air input port, for example, 21c. I do. The arrow (c) in FIG. 5 (a) shows that state. As a result, the combustion at the top of the charbed is improved as described in the above-mentioned principle, and the charbed 15 is shaped into an appropriate contour as shown by the dotted line (b) in FIG. 5 (a).

On the other hand, for example, the outline of FIG. 5 (b) to the measured image signal as shown is shaped from neuron elements 45 _1, 45 ₂ of the char bed in the output layer of the neural network 42 is too low in the case of (b) Since the “0” signal is output, the operation control unit 50 performs an operation opposite to that described above.

Next, in the operation control means 50, the neuron elements 45 ₃ and 45 ₄
When "0" is input to both of them, it is determined that the char bed 15 is deviated to the right, and as a result, the control signal e for increasing the air flow rate at the right combustion air inlet is transmitted from the command signal generation means 52e to the combustion air operation. To the means 22. Here, the combustion air operating means 22 increases the air flow rate at the right combustion air inlet (refer to the above-mentioned principle). When the char bed 15 is as shown in FIG.
Since both "1" signal from the neuron element 45 _3, 45 ₄ of the output layer are outputted, it is determined that the char bed 15, the operation control unit 50 is offset to the left side performs the inverse control with the.
Further, in the case of the situation as shown in FIG. 5 (d), if the control is performed based on the above principle, the char bed 15 can be moved to an appropriate position indicated by the dotted line in the figure.

Therefore, according to the configuration of the embodiment described above, the charbed shape detecting means 40 is provided on the furnace wall of the recovery boiler 10, and the neural network 42 is used based on the measured image signal obtained from the charbed shape detecting means 40. In char bed
Judge the position and shape deviation of 15 and operate various operating elements that affect the position and shape of the char bed 15 based on the judgment result, and correct the char bed 15 to the appropriate shape and position by the feedback control method As a result, it is possible to avoid adverse effects such as poor combustion due to the collapse of the charbed, a decrease in the recovery rate of the chemical raw material for chip digestion, and an increase in carryover, as well as to automatically control the shape of the charbed. Therefore, even if an abnormality occurs in the charbed, the desired shape can be quickly corrected. Therefore, there is no fear that the operation of the plant is hindered, the stable operation and high efficiency of the recovery boiler can be achieved, and further energy saving and labor saving can be achieved.

It is needless to say that the present invention is not limited to the operation elements shown in FIG. In addition, the present invention can be implemented with various modifications without departing from the scope of the invention.

[Effects of the Invention] As described above, according to the present invention, the recovery boiler is provided with the charbed shape detecting means, and based on the charbed presence / absence signal for each pixel of the image signal obtained from the charbed shape detecting means. The neural network determines the presence / absence of various abnormalities and controls the operating elements that affect the position and shape of the bed according to this determination signal. Control of recovery boiler char bed that can take appropriate measures promptly for abnormalities such as the shape of the char bed and contribute to the stable operation and high efficiency of the recovery boiler, and can realize energy saving and labor saving reliably. Equipment can be provided.

[Brief description of the drawings]

1 to 5 are views for explaining an embodiment of a char bed control device for a recovery boiler according to the present invention. FIG. 1 is a block diagram of the device of the present invention, and FIG. FIG. 3 is a schematic diagram of the neuron element, FIG. 4 is a schematic diagram of the operation control means shown in FIG. 1, and FIG. 5 is an operation control based on a deviation between a measured image of the charbed and an appropriate image. It is a figure explaining an example. 10… Recovery boiler, 11… Black liquor, 12… Black liquor heater, 13
…… Black liquor injection pressure control valve, 14 …… Black liquor injection gun, 15 ……
Charbed, 16 ... Gun angle operating means, 21a, 21b, 21c
... combustion air inlet, 22 ... combustion air operating means, 23 ...
Combustion air, 31 ... burning burner, 32 ... burning fuel, 33 ...
... Auxiliary operation valve, 40 ... Charbed shape detection means, 41 ...
... image signal conversion means, 42 ... neural network, 50 ... operation control means, 51 ... logic circuit, 52b-52f ...
... Command signal generating means.

Claims (3)

(57) [Claims] 1. A black liquor discharged from a chip digestion process of a pulp manufacturing plant is injected into a furnace by a black liquor injection gun to form a char bed at the bottom of the furnace and to supply combustion air to a plurality of locations in the furnace. A char bed shape detecting means for detecting the position and shape of the char bed, wherein the char bed shape detecting means for detecting the position and shape of the char bed; Image signal converting means for converting two-dimensional image data relating to the position and shape of the charbed obtained by the means into multi-point information for each pixel representing the presence or absence of the charbed; The synapse of each neuron element can be adjusted so that the position and shape of the charbed with a predetermined height, left and right position, and width A neural network having a multi-layer structure having a learning function in which coupling coefficients are set, and a difference between height, left / right position and width is determined and output based on multipoint information input from the image signal converting means; Operation control means for selectively changing operation elements that affect at least the height, the left / right position, and the width of the charbed based on the determination result of the network and correcting the position and shape of the charbed to an appropriate position and shape. Char bed shape control device for recovery boiler. 2. The operating element is at least one of the injected black liquor, the combustion air, and the auxiliary fuel.
A charbed shape control device for the recovery boiler described in the above. 3. An operating element selection variable means includes: a black liquor injection angle of the black liquor injection gun, a black liquor temperature to be injected, a pressure of the black liquor to be injected, a flow distribution ratio of the black liquor to be injected from a plurality of locations, 2. The char bed shape control device for a recovery boiler according to claim 1, wherein one or two or more of the input flow rate of the working air, the flow rate ratio of the combustion air at a plurality of locations, and the flow rate of the auxiliary fuel are selectively changed.