Boiler operation support device and boiler operation support system

Abstract

To operate a boiler while further increasing a mixed fuel burning ratio of biomass, low-grade coal, etc.SOLUTION: The present invention relates to a boiler operation support device which constructs a predictive model defining a mixed fuel burning ratio of a first class fuel desired to increase the mixed fuel burning ratio and a setting of an operation end of a boiler or an accessory of the boiler as input parameters and defining a constraint parameter which is changed considerably in a case where the mixed fuel burning ratio is increased relatively, as an output parameter. A prediction value in a case where the mixed fuel burning ratio is increased is calculated from the predictive model and compared with a reference value determined from a facility specification aspect corresponding to the kind of the constraint parameter. Setting of the operation end is adjusted on the basis of a comparison result; the constraint parameter is changed in a direction with a tolerance for the reference value; and an upper limit value of the mixed fuel burning ratio is estimated.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a boiler operation support device and a boiler operation support system, and more particularly to a boiler operation support device and an operation control device for a pulverized fuel-fired boiler that pulverizes coal, biomass, etc. and burns them in a burner.

Background Art

[0002] In a boiler installed in a thermal power plant, a multi-coal type control logic is used to correct the set value set at the operation terminal while coping with changes in boiler characteristics such as coal properties and dirt, and to improve the controllability of the boiler for multiple coal types. In the multi-coal type control logic, constraint parameters (rotation speed of the rotary classifier, hydraulic pressure, etc.) are feedback-controlled by a predetermined logic based on various process values (for example, mill motor current, mill table differential pressure).

[0003] Further, as an example of operation control when co-firing different types of fuels, Patent Document 1 discloses a steam temperature control device for a boiler that prevents the fluid temperature at the outlet of the furnace evaporation tube from exceeding the allowable range even when the co-firing ratio changes in a boiler that co-fires different types of fuels.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] When co-firing coal and biomass in a boiler, normal operation control is performed by diverting the multi-coal type control logic.

[0006] However, when the co-firing rate of biomass or low-grade coal is increased, certain process values that do not vary much during the combustion of ordinary coal change significantly, reaching the limitations in equipment design, which results in the situation that the co-firing rate cannot be increased due to this factor.

[0007] The present invention has been made in view of the above situation, and an object thereof is to provide an operation support device and an operation support system for operating a boiler by further increasing the co-firing rate of biomass or low-grade coal.

Means for Solving the Problems

[0008] To achieve the above object, the present invention has the configuration described in the claims. For example, it is an operation support device for a boiler that co-fires multiple types of fuels, taking as input parameters the co-firing ratio of a first type of fuel for which an increase in the co-firing ratio is desired and a second type of fuel different from the first type of fuel, and the set value of at least one operation terminal of the boiler or an auxiliary machine of the boiler, and having as an output parameter a constraint parameter that changes significantly when the co-firing ratio of the first type of fuel is relatively increased. It is a prediction model that machine-learns using, as teacher data, operation data when the first type of fuel and the second type of fuel are co-fired in the boiler. A model storage unit that stores the prediction model; a reference value storage unit that stores a reference value corresponding to the type of the constraint parameter, the reference value being determined from the equipment specifications; an optimization unit used to increase the co-firing ratio, or a co-firing ratio upper limit estimation unit used to estimate the upper limit of the co-firing ratio; and an output unit that outputs the calculation result of the optimization unit or the co-firing ratio upper limit estimation unit. In the operation support device for a boiler, the optimization unit increases the virtual co-firing ratio input to the prediction model to calculate the predicted value of the constraint parameter, and based on the comparison result between the predicted value and the reference value, performs an operation to adjust the setting of the operation terminal to change the constraint parameter in a direction equal to or having a margin over the reference value. The co-firing ratio upper limit estimation unit increases the virtual co-firing ratio input to the prediction model to calculate the predicted value of the constraint parameter, and performs an operation to estimate the upper limit value of the co-firing ratio within a range where the predicted value of the constraint parameter is equal to or has a margin over the reference value of the constraint parameter. The output unit outputs the set value of the operation terminal obtained from the calculation in the optimization unit or the upper limit value of the co-firing ratio obtained from the calculation in the co-firing ratio upper limit estimation unit.

Advantages of the Invention

[0009] According to the present invention, it is possible to provide an operation support device and an operation control device for operating a boiler by further increasing the co-firing ratio of biomass or low-quality coal. Other problems, configurations, and effects than those described above will be clarified by the description of the following embodiments.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Mode for Carrying Out the Invention

[0011] Hereinafter, with reference to the accompanying drawings, preferred embodiments of the present invention will be described in detail. Note that the present invention is not limited by this embodiment, and when there are a plurality of embodiments, those configured by combining each embodiment are also included. The same reference numerals are given to the same configurations and steps throughout the drawings, and redundant explanations are omitted.

[0012] FIG. 1 is a schematic configuration diagram of the boiler 1. The boiler 1 of this embodiment is a boiler 1 capable of pulverizing solid fuels such as coal and biomass, and performing exclusive firing operation of coal and co-firing operation of a plurality of types of fuels such as coal and biomass.

[0013] The boiler 1 has a furnace 11, a combustion device 12, and a flue 13. The furnace 11 has, for example, a hollow shape of a square cylinder and is installed along the vertical direction. The furnace 11 has a wall surface composed of evaporation tubes (heat transfer tubes) and fins connecting the evaporation tubes. The feed water and steam flowing in the evaporation tubes and the combustion gas in the furnace 11 exchange heat to suppress the temperature rise of the furnace wall. Specifically, a plurality of evaporation tubes are arranged, for example, along the vertical direction on the side wall surface of the furnace 11 and are arranged side by side in the horizontal direction. The fins block the space between the evaporation tubes. The furnace 11 is provided with an inclined surface 62 at the furnace bottom, and a furnace bottom evaporation tube 70 is provided on the inclined surface 62 to form the bottom surface.

[0014] The combustion device 12 is provided on the lower vertical side of the furnace wall constituting the furnace 11. In this embodiment, the combustion device 12 has a plurality of combustion burners (for example, 21, 22, 23, 24, 25) attached to the furnace wall. For example, a plurality of these combustion burners (burners) 21, 22, 23, 24, 25 are arranged at equal intervals along the circumferential direction of the furnace 11. However, the shape of the furnace, the arrangement of the burners, the number of combustion burners in one stage, and the number of stages are not limited to this embodiment.

[0015] Each of these combustion burners 21, 22, 23, 24, 25 is connected to pulverizers (pulverized coal machines / mills: corresponding to auxiliary machines) 31, 32, 33, 34, 35 via pulverized coal supply pipes 26, 27, 28, 29, 30. When coal is conveyed by a conveying system (not shown) and fed into these pulverizers 31, 32, 33, 34, 35, it is pulverized here to a predetermined fine size and the pulverized coal (pulverized coal) of the coal pulverized in the pulverizers 31, 32, 33, 34, 35 can be supplied to the combustion burners 21, 22, 23, 24, 25 together with the conveying air (primary air) from the pulverized coal supply pipes 26, 27, 28, 29, 30.

[0016] Further, the furnace 11 is provided with a wind box 36 at the mounting positions of the respective combustion burners 21, 22, 23, 24, 25. One end of an air duct 37b is connected to this wind box 36, and the other end is connected to an air duct 37a for supplying air at a connection point 37d.

[0017] Above the furnace 11 in the vertical direction, a flue 13 is connected, and a plurality of heat exchangers (41, 42, 43, 44, 45, 46, 47) for generating steam are arranged in this flue 13. Therefore, the combustion burners 21, 22, 23, 24, 25 inject a mixture of pulverized fuel and combustion air into the furnace 11 to form a flame, generate combustion gas, and flow into the flue 13. Then, the combustion gas heats the feed water and steam flowing through the furnace wall and the heat exchangers (41 - 47) to generate superheated steam, supply the generated superheated steam to rotate a steam turbine (not shown), and rotate a generator (not shown) connected to the rotating shaft of the steam turbine to generate electricity. Further, this flue 13 is connected to an exhaust gas passage 48, and is provided with a denitration device 50 for purifying the combustion gas, an air heater 49 for performing heat exchange between the air sent from the blower 38 to the air duct 37a and the exhaust gas sent through the exhaust gas passage 48, a dust treatment device 51, an induced draft fan 52, etc., and a chimney 53 is provided at the downstream end. Note that the denitration device 50 may not be provided as long as the exhaust gas standard can be satisfied.

[0018] The furnace 11 of this embodiment is a so-called two-stage combustion furnace that performs fuel lean combustion by newly introducing combustion air (after air) after fuel-rich combustion by the conveying air (primary air) of pulverized coal and the combustion air (secondary air) introduced into the furnace 11 from the air box 36. Therefore, the furnace 11 is provided with an after-air port 39, one end of the air duct 37c is connected to the after-air port 39, and the other end is connected to the air duct 37a that supplies air at the connection point 37d. Note that when the two-stage combustion method is not adopted, the after-air port 39 may not be provided.

[0019] The air sent from the blower 38 to the air duct 37a is heated by heat exchange with the combustion gas in the air heater 49, and branches into secondary air guided to the air box 36 via the air duct 37b and after-air guided to the after-air port 39 via the air duct 37c at the connection point 37d.

[0020] FIG. 2 is a schematic explanatory diagram of an operation support system 10 for a boiler 1. The operation support system 10 includes a boiler 1, an operation support device 100 for the boiler 1, and an operation control device 120 for the boiler 1.

[0021] The operation support device 100 constructs, by machine learning, a prediction model for predicting various process values that occur when the boiler 1 performs exclusive coal combustion operation or co-combustion operation of different fuels, and uses the prediction model to optimize operation conditions suitable for the fuel properties or calculate the upper limit value of the co-combustion ratio. High-moisture coal (low-grade coal) and biomass fuel, which have a relatively high moisture content as different fuels, are fuels for which it is desired to increase the co-combustion ratio, and thus correspond to the first type of fuel. Also, high-grade coal co-combusted with high-moisture coal or biomass fuel corresponds to the second type of fuel. The operation support device 100 aims to support operation with an increased co-combustion ratio of the first type of fuel.

[0022] The operation support device 100 includes a data acquisition unit 110, an operation data storage unit 112, a data extraction unit 114, a soft sensor value calculation unit 116, an RTC 118, a model construction unit 220, a model storage unit 222, an optimization unit 230, a reference value storage unit 232, a co-combustion ratio upper limit estimation unit 240, an operation condition evaluation unit 242, and an output unit 250. The functions of each unit will be described later.

[0023] Figure 3 is a diagram showing the hardware configuration of the driving support device 100. The driving support device 100 includes a processor 301, a RAM (Random Access Memory) 302, a ROM (Read Only Memory) 303, an HDD (Hard Disk Drive) 304, an input I / F 305, an output I / F 306, and a communication I / F 307, and is configured using a computer in which these are connected to each other via a bus 308. The processor 301 may be a GPU (Graphics Processing Unit) or a CPU (Central Processing Unit), and may be of any type as long as it is a device that executes an arithmetic function. Further, the hardware configuration of the driving support device 100 is not limited to the above, and may be configured by a combination of a control circuit and a storage device. The driving support device 100 is configured by the processor 301 executing a driving support program that realizes each function of the driving support device 100, or by the control circuit performing arithmetic operations.

[0024] An input device 311 such as a mouse, a keyboard, or a touch panel is connected to the input I / F 305.

[0025] A display 312 composed of an LCD or the like is connected to the output I / F 306.

[0026] The boiler 1 and the operation control device 120 are each connected to the communication I / F 307.

[0027] Figure 4 is a flowchart showing the main flow from model construction to operation by the driving support system 10 of the boiler 1.

[0028] <S1: Construction of Prediction Model> The model construction unit 220 of the driving support device 100 adds the mixed combustion rate to the input parameter (explanatory variable in the regression model) and models the constraint process value, that is, constructs a prediction model for the constraint process value (S1). The constructed prediction model is stored in the model storage unit 222. Figure 5 is a flowchart showing the details of the processing from acquisition of driving data to construction of the prediction model.

[0029] The data acquisition unit 110 acquires operation data from the boiler 1 and stores it in the operation data storage unit 112 (S101). The data acquisition unit 110 acquires the actual process values measured by each of the sensors 1, 2, ···, M during actual operation, the operation end parameters (operation end set values) set by the operation control device 120 for each of the operation ends 1, 2, ···, N, and the actual mixing ratio of the first type of fuel and the second type of fuel during co-firing operation, generates operation data associating the process values, the operation end parameters, and the mixing ratio, and stores it in the operation data storage unit 112.

[0030] In addition to the constraint process values, the above process values include, for example, trace components such as the nitrogen oxide concentration contained in the gas discharged from the thermal power plant, and the metal temperature of the heat transfer tubes.

[0031] The data acquisition unit 110 may generate operation data consisting of time-series data by adding time information from the RTC 118 to each of the operation end parameters and the actual process values.

[0032] The operation data acquired in this embodiment serves as teacher data for constructing a prediction model that calculates predicted values of constraint process values and other process values. The teacher data may use not only the actual process values and the operation conditions (operation end parameters and mixing ratio) when they are obtained, but also calculated values obtained by analyzing the boiler 1.

[0033] The preprocessing by the data extraction unit 114 is executed (S102). The data extraction unit 114 reads the operation data described above in the operation data storage unit 112, and compensates for missing data by causing the soft sensor value calculation unit 116 to execute variable calculation to obtain soft sensor values. Further, the operation conditions including the mixing ratio are read from the operation data, and the tuning data is extracted.

[0034] The soft sensor value calculation unit 116 calculates the values of sensors (soft sensors) not installed in the boiler 1 using the actual process values actually measured by the sensors 1, 2, ···, M, and outputs the soft sensor values consisting of the measured values to the data acquisition unit 110.

[0035] The model construction unit 220 acquires the tuning data extracted by the data extraction unit 114 to generate learning data (S103), and sets learning conditions (S104).

[0036] The model construction unit 220 constructs a machine learning model based on the learning data and the set learning conditions.

[0037] FIG. 6 is a diagram showing an example of a constraint process value. The constraint process value is a process value that changes significantly when co-firing high-moisture coal (so-called low-grade coal) or biomass fuel. By suppressing the behavior of the constraint process value, the co-firing rate can be increased. Here, the constraint process value is a process value that may make it difficult to continue operation by reaching the management limit value beyond the appropriate range when the co-firing rate is increased. Examples of the constraint process value include the temperature of the air for drying and transporting pulverized fuel and the motor current of the pulverized coal mill (mill).

[0038] FIG. 7 is a diagram showing an example of a prediction model created by the model construction unit 220. The model construction unit 220 constructs a prediction model corresponding to each of the constraint process values shown in FIG. 6. Specifically, the model construction unit 220 uses, as input parameters, the control values (operation end parameters) set at the operation ends 1 to N, the soft sensor values, and the co-firing rate of the first type of fuel, and uses, as output parameters, the respective constraint process values (measured values) obtained by actually operating the boiler 1 with the input parameters set therein. The model construction unit 220 performs machine learning on the prediction model corresponding to each constraint process value and constructs the prediction model.

[0039] When the model construction unit 220 constructs a prediction model using a regression model, in the example of the prediction model 1 in FIG. 7, the operation end parameters, fuel parameters (including the co-firing rate), and other parameters (such as environmental conditions such as air temperature) are used as explanatory variables, and the pulverized fuel drying / transport air temperature is used as the target variable to perform machine learning on the regression model. The constructed prediction model is stored in the model storage unit 222. In the example of FIG. 7, in this example, a plurality of prediction models including the prediction model 2 for calculating the predicted value of the motor current of the pulverized coal machine (mill) are constructed, but one prediction model that models the most notable one constraint process value may be constructed.

[0040] <S2: Optimization of Operating Conditions> The optimization unit 230 improves the constraint process value (S2). FIG. 8 is a flowchart showing the flow from the details of the optimization of S2 to S5.

[0041] The optimization unit 230 sets the optimization conditions (S201). Specifically, it is set which operating conditions are optimized when performing co-firing operation in which operating mode. Further, the optimization unit 230 also performs score setting as one of the settings of the optimization conditions. In this embodiment, it is set to the soundness mode, and the score setting is adjusted so that the constraint process value is improved (the score is made larger than other soundness process values).

[0042] The optimization unit 230 executes optimization (S202). The optimization unit 230 executes a process of changing the weighting of the constraint process value to give a margin to the constraint process value in order to improve the constraint process value that varies greatly when the co-firing rate is increased.

[0043] The timing at which the optimization unit 230 performs the optimization process is a) When the margin of a specific constraint process value disappears, b) When the co-firing rate reaches an area that has not been learned (when exceeding the co-firing rate range of the learning data at the time of model construction), Either one, or both, of the above are satisfied.

[0044] The countermeasures taken by the optimization unit 230 at the above timing are as follows: c) Review the weighting for specific process values with little margin and increase the margin. Here, for each individual process value, review the score setting. d) Perform optimization again in the soundness mode (increase the margin for all constraint process values). The optimization unit 230 performs one or both of the above. To this end, based on the comparison result between the predicted value of the constraint process value obtained from the prediction model and the reference value, the optimization unit 230 adjusts the setting of the operation end parameter to change the constraint process value in the direction of having a margin with respect to the reference value.

[0045] The "reference value" is the limit value of each constraint process value determined from the equipment specification aspect. As a modification example, it may be an alarm value that has a margin from the limit value but for which an alarm is issued.

[0046] The optimization unit 230 checks whether the constraint process value has been improved by optimization (S203). It also checks that the optimal setting in S202 is not inconsistent with the concept of equipment design and past operation results. The reference value storage unit 232 stores in advance the reference values to be compared with the individual constraint process values. The optimization unit 230 checks the presence or absence of improvement based on the comparison result with the reference value. If the improvement is not sufficient (S203: No), it returns to step S201 to review the optimization conditions. If the improvement is sufficient (S203: Yes), it proceeds to the estimation process of the co-firing rate.

[0047] <S3: Estimation process of co-firing rate> The co-firing rate upper limit estimation unit 240 uses the prediction model constructed in step S105 and the optimal setting in step S107 to obtain the constraint process values when the co-firing rate is increased, and estimates the maximum co-firing rate until the reference value is reached (S3).

[0048] The co-firing rate upper limit estimation unit 240 calculates the predicted value of the constraint process value by increasing the virtual co-firing rate among the operation conditions input to the prediction model. Then, it estimates the upper limit value of the co-firing rate within the range where the predicted value of the constraint process value is less than or equal to the reference value of the constraint process value.

[0049] Based on the predicted value of the constraint process value obtained in S3, the operation condition evaluation unit 242 performs an operation balance evaluation and an economic efficiency evaluation when the co-firing rate increases, and sets a target co-firing rate (S4). The operation condition evaluation unit 242 evaluates the operation conditions according to the score setting set in step S201. A radar chart may be created from the predicted values of each constraint process value, and the operation balance may be evaluated from its shape.

[0050] Based on the operation balance evaluation and the economic efficiency evaluation, the operation condition evaluation unit 242 performs an evaluation calculation of the operation conditions, and outputs the calculation result (high-evaluation operation conditions or increased co-firing rate) to the output unit 250.

[0051] <S5: Start co-firing on the actual machine> The output unit 250 outputs the set value of the operation terminal obtained from the calculation in the optimization unit 230 or the upper limit value of the co-firing rate obtained from the calculation in the co-firing rate upper limit estimation unit 240 to the operation control device 120, and starts co-firing on the actual machine (S5). FIG. 9 is a flowchart showing the detailed flow of S5.

[0052] When both the first condition and the second condition are satisfied (S501: Yes), the output unit 250 outputs an instruction signal for starting an increase in the co-firing rate to the operation control device 120 in order to start an operation for increasing the co-firing rate on the actual machine (S505). e) The actual co-firing rate is within the learning range (first condition). f) There is a margin for all constraint process values (second condition).

[0053] On the other hand, when either or both of the above first condition and second condition are No (S501: No), the output unit 250 requests optimization from the optimization unit 230.

[0054] In steps S502 to S504, the optimization unit 230 performs the same optimization process as in steps S201 to S203. If it is determined in step S504 that the constraint process value has not improved (S504: No), the process returns to step S502 to reset the optimization conditions.

[0055] On the other hand, when the optimization unit 230 determines in step S504 that the constraint process value has improved (S504: Yes), it returns the result to the output unit 250. In response to this, the output unit 250 outputs an instruction signal for starting the co-firing rate increase operation in the actual machine to the operation control device 120 (S505).

[0056] FIG. 10 is a diagram showing the operation process flow of the boiler 1 using the operation support system 10.

[0057] When the operation support device 100 acquires operation data, it performs preprocessing to generate learning data. Using the learning data, it constructs a prediction model. As a result of optimizing the soundness mode using the prediction model, the constraint process value decreases, and a margin with respect to the upper limit value of the constraint process value is secured (at time t1).

[0058] Next, the operation support device 100 performs an estimation process of the upper limit of the co-firing rate, and based on the operation balance evaluation at the upper limit of the co-firing rate estimated from the prediction model, increases the co-firing rate in the actual machine (at time t2). As the co-firing rate increases, the constraint process value also increases. Therefore, the operation support device 100 checks the behavior of the constraint process value, and as a result of reflecting the optimized setting with the optimized score setting reviewed in the actual machine, the constraint process value decreases again, and a margin with respect to the upper limit value of the constraint process value is secured (at time t3).

[0059] When the operation data after the increase in the co-firing rate of the actual machine is re-learned and optimized and the optimized setting is reflected in the actual machine, while increasing the co-firing rate more than before t2, a margin with respect to the upper limit value of the constraint process value is secured (at time t4).

[0060] According to the present embodiment, when biomass fuel or low-grade coal is co-fired with high-grade coal, the constraint process value, which conventionally fluctuates greatly, has been a constraint factor for the increase in the co-firing rate. According to the present embodiment, the operation support device 100 determines the set value of the operation terminal so that the margin of the constraint process value when the co-firing rate is increased increases, so that the co-firing rate can be increased within a range where the constraint process value does not reach the alarm value.

[0061] The above-described embodiments do not limit the present invention, and there are various modification modes without departing from the spirit of the present invention. For example, in the above-described operation support device 100, an example in which both the optimization unit 230 and the co-firing rate upper limit estimation unit 240 are provided has been shown, but only one of them may be provided. For example, only the optimization unit 230 may be provided, the operation conditions may be optimized, and the operation end parameters when the co-firing rate at that time is increased may be passed to the output unit 250. Further, only the co-firing rate upper limit estimation unit 240 may be provided, and the co-firing rate upper limit value obtained as a result of performing a simulation for estimating the co-firing rate upper limit value may be passed to the output unit 250.

Explanation of Reference Numerals

[0062] 1: Boiler 10: Operation support system 11: Firebox 12: Combustion device 13: Flue 21 - 25: Combustion burner 26 - 30: Pulverized coal supply pipe 31 - 35: Pulverizer 36: Wind box 37a - 37c: Air duct 37d: Connection point 38: Blower 39: After air port 48: Exhaust gas passage 49: Air heater 50: Denitration device 51: Coal dust treatment device 52: Induced draft fan 53: Chimney 62: Inclined plane 70: Furnace bottom evaporation pipe 100: Operation support device 110: Data acquisition unit 112: Operation data storage unit 114: Data extraction unit 116: Soft sensor value calculation unit 118: RTC 120: Operation control device 220: Model construction unit 222: Model Memory Unit 230: Optimization Unit 232: Reference Value Memory Unit 240: Co-firing Rate Upper Limit Estimation Unit 242: Operating Condition Evaluation Unit 250: Output Unit 301: Processor 305: Input I / F 306: Output I / F 307: Communication I / F 308: Bus 311: Input Device 312: Display

Claims

A boiler operation support device for co-firing a plurality of types of fuels including a first type of fuel for which an increased co-firing rate is desired and a second type of fuel different from the first type of fuel, wherein the co-firing rate of the first type of fuel and the set value of at least one operation terminal of the boiler or an auxiliary machine of the boiler are used as input parameters, a prediction model having a constraint parameter that may exceed an appropriate range when the co-firing rate of the first type of fuel is relatively increased as an output parameter, and storing a prediction model obtained by machine learning using operation data when the first type of fuel and the second type of fuel are co-fired in the boiler as teacher data, a reference value storage unit that stores a reference value corresponding to the type of the constraint parameter and determined from the equipment specification, an optimization unit used to increase the co-firing rate or a co-firing rate upper limit estimation unit used to estimate the upper limit of the co-firing rate, a boiler operation support device including an output unit that outputs an operation result of the optimization unit or the co-firing rate upper limit estimation unit, the optimization unit performs an operation of adjusting the set value of the operation terminal based on a comparison result between a predicted value of the constraint parameter obtained from the prediction model and the reference value, and changing the constraint parameter in a direction equal to or more margin than the reference value, the co-firing rate upper limit estimation unit increases a virtual co-firing rate input to the prediction model to calculate a predicted value of the constraint parameter, and performs an operation of estimating an upper limit value of the co-firing rate within a range where the predicted value is equal to or more margin than the reference value, the output unit outputs the set value of the operation terminal obtained from the operation of the optimization unit or the upper limit value of the co-firing rate obtained from the operation of the co-firing rate upper limit estimation unit, a boiler operation support device.

2. The boiler operation support device according to claim 1, wherein the model storage unit stores a prediction model using a regression model having the co-firing rate and the set value of the operation terminal as explanatory variables and the constraint parameter as an objective variable. a boiler operation support device.

3. The boiler operation support device according to claim 1, wherein the auxiliary machine is a mill that pulverizes the first type of fuel, and the constraint parameter is at least one of the temperature of air for drying and transporting pulverized fuel and the motor current of the mill. a boiler operation support device.

4. The boiler operation support device according to claim 1, The first type of fuel is biomass fuel or low-grade coal with a relatively high water content, and the second type of fuel is high-grade coal with a relatively low water content. A boiler operation support device.

5. An operation support system for a boiler that co-fires a plurality of fuels, The boiler operation support device according to any one of claims 1 to 4, And an operation control device that sets a set value at an operation terminal of the boiler. The operation control device of the boiler, Obtains information indicating the target co-firing rate of the first type of fuel from the operation support device, and sets operation terminal parameters at each operation terminal of the boiler and the boiler auxiliaries based on the target co-firing rate. A boiler operation support system.

Images