Furnace temperature control device, furnace temperature control method, and coke manufacturing method

Abstract

This furnace temperature control device comprises: a storage unit that stores at least one among a first model and a second model; and a control unit. The control unit: acquires operation information including temperature measurement values of a plurality of combustion chambers, temperature measurement values of a plurality of carbonization chambers, actual values of the quantity of heat supplied to the plurality of combustion chambers, and actual values and scheduled values of the state of coal filling the plurality of carbonization chambers; calculates a temperature prediction value of the combustion chamber for each combustion chamber among the plurality of combustion chambers by inputting information included in the operation information to a first individual model, or calculates a temperature prediction value of the carbonization chamber for each carbonization chamber among the plurality of carbonization chambers by inputting information included in the operation information to a second individual model; and calculates the quantity of heat supplied to each combustion chamber among the plurality of combustion chambers on the basis of a target temperature of the combustion chamber and the calculated temperature prediction value of the combustion chamber, or on the basis of a target temperature of the carbonization chamber and the calculated temperature prediction value of the carbonization chamber.

Description

Furnace Temperature Control Device, Furnace Temperature Control Method, and Coke Manufacturing Method

[0001] The present disclosure relates to a furnace temperature control device, a furnace temperature control method, and a coke manufacturing method.

[0002] There is a coke oven in which a plurality of combustion chambers and a plurality of carbonization chambers are alternately connected to form a furnace group. In such a coke oven, coke is produced by carbonizing the coal charged into the carbonization chamber with heat from an adjacent combustion chamber.

[0003] In order to achieve stabilization of the quality of coke produced in a coke oven, improvement in production efficiency, and reduction of the heat quantity for carbonization, it is important to control the temperature of the combustion chamber or the carbonization chamber to a desired temperature.

[0004] A coke oven has a large heat capacity of the furnace body and a long time constant for the response to an action for operating the furnace temperature. Therefore, in order to control the temperature of the combustion chamber or the carbonization chamber with high precision, it is necessary to accurately predict the future temperature and perform an action for operating the furnace temperature.

[0005] However, the temperature of a coke oven fluctuates complexly due to a number of factors such as the supply status of fuel gas to the combustion chamber, the properties and carbonization status of the coal charged into the carbonization chamber, and the states of adjacent carbonization chambers and combustion chambers. Therefore, it has been difficult to accurately predict the future temperature of a coke oven.

[0006] For example, Patent Document 1 proposes a method for improving the future temperature prediction accuracy by reflecting the furnace temperature prediction error in the previous carbonization cycle in the furnace temperature prediction calculation at the current time in order to reduce the uncertainty of the furnace temperature prediction mathematical model and the furnace temperature prediction error due to disturbance for accurately predicting the future temperature of a coke oven.

[0007] Japanese Patent Application Laid-Open No. 5-255668

[0008] The technology disclosed in Patent Document 1 improves the accuracy of future temperature predictions by reflecting the furnace temperature prediction error in the pre-carbonization cycle in the furnace temperature prediction calculation at the current time. However, there are multiple uncertainties and disturbance factors in the furnace temperature prediction model. Therefore, as in Patent Document 1, a method of correcting one temperature prediction of the combustion chamber or carbonization chamber in a coke oven with one pattern of error correction logic cannot fully correct for the multiple expected disturbance factors, and thus the prediction accuracy is not sufficient.

[0009] External disturbances that may affect the temperature prediction of a coke oven include, for example, fluctuations in the properties of coal (particle size, moisture content, ash content, volatile matter, etc.), variations in the amount or distribution of coal packed, fluctuations in the combustion gas components or flow rate, deterioration or wear of bricks (refractory materials) over time, carbon buildup on the brick walls, changes in operating conditions (operating patterns of the combustion chamber and carbonization chamber, air ratio, gas flow rate adjustment, etc.), fluctuations in ambient temperature or the environment surrounding the furnace, decreased accuracy or malfunction of measuring instruments, and variations in the timing of fire extinguishing.

[0010] In particular, a coke oven is composed of multiple combustion chambers and carbonization chambers connected together. Each combustion chamber and carbonization chamber has different heat transfer characteristics due to factors such as the aging or wear of the bricks (refractory materials), carbon buildup on the brick walls, and variations in the properties or packing state of the coal. These characteristics also change over time. Therefore, unless the individual changes in heat transfer characteristics of each combustion chamber and carbonization chamber can be appropriately compensated for, highly accurate temperature prediction is difficult.

[0011] The purpose of this disclosure is to provide a furnace temperature control device, a furnace temperature control method, and a coke manufacturing method that can predict the future temperature of a coke oven with high accuracy.

[0012] [1] A furnace temperature control device for controlling the temperature of a combustion chamber or a carbonization chamber in a coke oven in which a plurality of combustion chambers and a plurality of carbonization chambers are alternately connected to form a furnace group, comprising: a storage unit that stores at least one of a first model for calculating a predicted temperature value of the combustion chamber and a second model for calculating a predicted temperature value of the carbonization chamber; and a control unit, wherein the storage unit, when storing the first model, stores an individual first model for each of the plurality of combustion chambers; and when storing the second model, stores an individual second model for each of the plurality of carbonization chambers; and the control unit acquires operational information including the temperature measured values ​​of the plurality of combustion chambers, the temperature measured values ​​of the plurality of carbonization chambers, the actual value of the amount of heat supplied to the plurality of combustion chambers, and the actual and planned values ​​of the coal filling status of the plurality of carbonization chambers. A furnace temperature control device that inputs the information contained in the operational information into an individual first model to calculate a predicted temperature for each of the plurality of combustion chambers, or inputs the information contained in the operational information into an individual second model to calculate a predicted temperature for each of the plurality of carbonization chambers, and calculates the amount of heat to be supplied to each of the plurality of combustion chambers based on the target temperature of the combustion chamber and the calculated predicted temperature of the combustion chamber, or based on the target temperature of the carbonization chamber and the calculated predicted temperature of the carbonization chamber.

[0013] [2] The furnace temperature control device according to [1] above, wherein the control unit, when calculating the predicted temperature of the combustion chamber, inputs information relating to the combustion chamber and at least information relating to the adjacent carbonization chamber from the information contained in the operation information to the first model corresponding to the combustion chamber, and when calculating the predicted temperature of the carbonization chamber, inputs information relating to the carbonization chamber and at least information relating to the adjacent combustion chamber from the information contained in the operation information to the second model corresponding to the carbonization chamber.

[0014] [3] The furnace temperature control device according to [1] or [2] above, wherein the first model includes a first parameter, the second model includes a second parameter, the control unit calculates the first parameter so as to minimize a first evaluation function that represents the error between a predicted temperature of the combustion chamber and a measured temperature of the combustion chamber over a predetermined past period, and the control unit calculates the second parameter so as to minimize a second evaluation function that represents the error between a predicted temperature of the carbonization chamber and a measured temperature of the carbonization chamber over a predetermined past period.

[0015] [4] The furnace temperature control device according to any one of the above [1] to [3], wherein the first evaluation function is the difference or the average value of the absolute values ​​of the difference between the measured temperature of the combustion chamber and the predicted temperature of the first model at each point in time during the most recent predetermined period, and the second evaluation function is the difference or the absolute value of the difference between the predicted coal core temperature of the second model and the temperature that should theoretically be reached at the time of fire extinguishing, for each timing of fire extinguishing in the carbonization chamber.

[0016] [5] The furnace temperature control device according to any one of the above [1] to [4], wherein the identification of the first parameter and the second parameter is performed by, as a first loop, optimizing the first parameter for each of the combustion chambers so that the first evaluation function is less than a predetermined threshold, then, as a second loop, using the optimized first parameter, optimizing the second parameter for each of the carbonization chambers so that the second evaluation function is less than a predetermined threshold, and further, when either the first parameter or the second parameter is changed, the first loop and the second loop are repeated a predetermined number of times.

[0017] [6] The furnace temperature control device according to any one of [1] to [5] above, wherein the most recent predetermined period is adjusted according to the magnitude of the fluctuation of the first parameter or the second parameter, the most recent predetermined period is set to be longer when the fluctuation of the first parameter or the second parameter is large, and the most recent predetermined period is set to be shorter when the fluctuation of the first parameter or the second parameter is small.

[0018] [7] The furnace temperature control device according to any one of [1] to [6] above, wherein the first model further includes the second parameter, and the second model further includes the first parameter.

[0019] [8] The furnace temperature control device according to any one of the above [1] to [7], wherein the first model is a model for calculating a predicted temperature of the combustion chamber based on the heat conduction equation, and the second model is a model for calculating a predicted temperature of the carbonization chamber based on the heat conduction equation.

[0020] [9] The furnace temperature control device according to any one of the above [1] to [8], wherein the first model is a model that calculates a predicted temperature of the combustion chamber based on a machine learning model, and the second model is a model that calculates a predicted temperature of the carbonization chamber based on a machine learning model.

[0021]

[10] The furnace temperature control device according to any one of the above [1] to [9], wherein the first model is a model that calculates a predicted temperature of the combustion chamber based on a heat conduction equation, and the second model is a model that calculates a predicted temperature of the carbonization chamber based on a machine learning model.

[0022]

[11] The furnace temperature control device according to any one of the above [1] to

[10] , wherein the first model is a model that calculates a predicted temperature of the combustion chamber based on a machine learning model, and the second model is a model that calculates a predicted temperature of the carbonization chamber based on a heat conduction equation.

[0023]

[12] A furnace temperature control method in a furnace temperature control device for controlling the temperature of a combustion chamber or a carbonization chamber in a coke oven in which a plurality of combustion chambers and a plurality of carbonization chambers are alternately connected to form a furnace group, wherein the furnace temperature control device includes a storage unit that stores at least one of a first model for calculating a predicted temperature value of the combustion chamber and a second model for calculating a predicted temperature value of the carbonization chamber, the storage unit, when storing the first model, stores an individual first model for each of the plurality of combustion chambers, and when storing the second model, stores an individual second model for each of the plurality of carbonization chambers, the furnace temperature control method includes the step of acquiring operational information including temperature measurement values ​​of the plurality of combustion chambers, temperature measurement values ​​of the plurality of carbonization chambers, actual values ​​of the amount of heat supplied to the plurality of combustion chambers, and actual and planned values ​​of the coal filling status of the plurality of carbonization chambers, A furnace temperature control method comprising: inputting the information contained in the operational information into an individual first model to calculate a predicted temperature for each of the plurality of combustion chambers, or inputting the information contained in the operational information into an individual second model to calculate a predicted temperature for each of the plurality of carbonization chambers; and calculating the amount of heat to be supplied to each of the plurality of combustion chambers based on the target temperature of the combustion chamber and the calculated predicted temperature of the combustion chamber, or based on the target temperature of the carbonization chamber and the calculated predicted temperature of the carbonization chamber.

[0024]

[13] The furnace temperature control method according to

[12] above, wherein the step of calculating the predicted temperature of the combustion chamber is to input information relating to the combustion chamber and at least information relating to an adjacent carbonization chamber from the information contained in the operation information into the first model corresponding to the combustion chamber, and the step of calculating the predicted temperature of the carbonization chamber is to input information relating to the carbonization chamber and at least information relating to an adjacent combustion chamber from the information contained in the operation information into the second model corresponding to the carbonization chamber.

[0025]

[14] The furnace temperature control method according to

[12] or

[13] , wherein the first model includes a first parameter, the second model includes a second parameter, the first parameter is calculated to minimize a first evaluation function representing the error between a predicted temperature of the combustion chamber and a measured temperature of the combustion chamber over a predetermined past period, and the second parameter is calculated to minimize a second evaluation function representing the error between a predicted temperature of the carbonization chamber and a measured temperature of the carbonization chamber over a predetermined past period.

[0026]

[15] The furnace temperature control method according to any one of

[12] to

[14] above, wherein the first model further includes the second parameter, and the second model further includes the first parameter.

[0027]

[16] The furnace temperature control method according to any one of the above

[12] to

[15] , wherein the first model is a model for calculating a predicted temperature of the combustion chamber based on the heat conduction equation, and the second model is a model for calculating a predicted temperature of the carbonization chamber based on the heat conduction equation.

[0028]

[17] The furnace temperature control method according to any one of the above

[12] to

[16] , wherein the first model is a model that calculates a predicted temperature of the combustion chamber based on a machine learning model, and the second model is a model that calculates a predicted temperature of the carbonization chamber based on a machine learning model.

[0029]

[18] The furnace temperature control method according to any one of the above

[12] to

[17] , wherein the first model is a model for calculating a predicted temperature of the combustion chamber based on a heat conduction equation, and the second model is a model for calculating a predicted temperature of the carbonization chamber based on a machine learning model.

[0030]

[19] The furnace temperature control method according to any one of the above

[12] to

[18] , wherein the first model is a model that calculates a predicted temperature of the combustion chamber based on a machine learning model, and the second model is a model that calculates a predicted temperature of the carbonization chamber based on a heat conduction equation.

[0031]

[20] A method for producing coke, comprising controlling the amount of heat supplied to each of the plurality of combustion chambers based on the amount of heat supplied to each of the plurality of combustion chambers calculated by the furnace temperature control method described in any one of the above items

[12] to

[19] , to produce coke.

[0032] According to the furnace temperature control device, furnace temperature control method, and coke manufacturing method described herein, it is possible to appropriately correct for the differences and changes over time in the heat transfer characteristics of each of the multiple combustion chambers and carbonization chambers, and to predict the future temperature of the coke oven with high accuracy.

[0033] This is a schematic diagram showing an example of a furnace temperature control system according to one embodiment of the present disclosure. This is a block diagram showing an example of the configuration of a furnace temperature control device according to one embodiment of the present disclosure. This is a flowchart showing an example of a two-stage optimization procedure. This is a flowchart showing an example of a two-stage optimization procedure. This is a flowchart showing an example of the operation of a furnace temperature control device according to one embodiment of the present disclosure.

[0034] In this embodiment, in a coke oven configured with multiple combustion chambers and carbonization chambers connected together, individual heat transfer model parameters are set for each combustion chamber and each carbonization chamber, and these parameters are identified and updated sequentially. By minimizing the error between the measured values ​​and model values ​​for each part using assimilation technology, it is possible to correct with high accuracy the different heat transfer characteristics or changes over time for each combustion chamber and carbonization chamber, such as the aging deterioration or carbon deposition of bricks, and variations in coal properties or packing conditions, thereby improving the accuracy of future temperature predictions.

[0035] The embodiments of this disclosure will be described below with reference to the drawings.

[0036] Figure 1 is a schematic diagram showing an example of a furnace temperature control system 1 according to one embodiment of the present disclosure. The furnace temperature control system 1 comprises a coke oven 10, a furnace temperature control device 20, and a control terminal 30.

[0037] The coke oven 10 comprises combustion chambers 11-1 to 11-N, carbonization chambers 12-1 to 12-(N-1), a first thermometer 13-1 to 13-N, a second thermometer 14-1 to 14-(N-1), a control valve 15, individual control valves 16-1 to 16-(N-1), a main gas pipe 17, and gas branch pipes 18-1 to 18-(N-1). N is an integer of 2 or more, but is not limited to a specific number.

[0038] In the following, unless otherwise necessary, combustion chambers 11-1 to 11-N may simply be referred to as combustion chamber 11. Similarly, carbonization chambers 12-1 to 12-(N-1) may simply be referred to as carbonization chamber 12. Furthermore, the first thermometers 13-1 to 13-N may simply be referred to as the first thermometer 13. Also, the second thermometers 14-1 to 14-(N-1) may simply be referred to as the second thermometer 14. Individual control valves 16-1 to 16-(N-1) may simply be referred to as individual control valve 16. Finally, gas branch pipes 18-1 to 18-(N-1) may simply be referred to as gas branch pipe 18.

[0039] In the coke oven 10, the combustion chambers 11-1 to 11-N and the carbonization chambers 12-1 to 12-(N-1) are connected alternately to form a furnace group.

[0040] Fuel gas G is supplied to the combustion chamber 11 from a gas supply source. As shown in Figure 1, the fuel gas G supplied from the gas supply source is supplied to the main gas pipe 17 via a control valve 15. The main gas pipe 17 branches into N-1 gas branch pipes 18-1 to 18-(N-1).

[0041] Each of the gas branch pipes 18-1 to 18-(N-1) is equipped with an individual control valve 16-1 to 16-(N-1).

[0042] Further, the gas branch pipes 18-1 to 18-(N-1) are respectively piped to the combustion chambers 11 adjacent to both sides of the carbonization chambers 12-1 to 12-(N-1). For example, the gas branch pipe 18-1 is piped to the combustion chambers 11-1 and 11-2 adjacent to both sides of the carbonization chamber 12-1. Also, for example, the gas branch pipe 18-2 is piped to the combustion chambers 11-2 and 11-3 adjacent to both sides of the carbonization chamber 12-2.

[0043] The regulating valve 15 is a valve that adjusts the flow rate of the fuel gas G supplied to the entire furnace group of the coke oven 10. The opening degree of the regulating valve 15 is controlled by the control terminal 30. The control terminal 30 can adjust the flow rate of the fuel gas G supplied to the entire furnace group of the coke oven 10 by controlling the opening degree of the regulating valve 15. The control terminal 30 adjusts the flow rate of the fuel gas G supplied to the entire furnace group so that the average temperature value of the entire furnace group of the coke oven 10 becomes the target temperature. The control terminal 30 controls the opening degree of the regulating valve 15 according to a command from the furnace temperature control device 20.

[0044] The individual regulating valve 16 is a valve that individually adjusts the flow rate of the fuel gas G supplied to the combustion chambers 11 adjacent to both sides of the carbonization chamber 12. For example, the individual regulating valve 16-1 can individually adjust the flow rate of the fuel gas G supplied from the gas branch pipe 18-1 to the combustion chambers 11-1 and 11-2 adjacent to both sides of the carbonization chamber 12-1. Also, for example, the individual regulating valve 16-2 can individually adjust the flow rate of the fuel gas G supplied from the gas branch pipe 18-2 to the combustion chambers 11-2 and 11-3 adjacent to both sides of the carbonization chamber 12-2.

[0045] The opening degree of the individual regulating valve 16 is controlled by the control terminal 30. The control terminal 30 can individually adjust the flow rate of the fuel gas G supplied to the combustion chambers 11-1 to 11-N by controlling the opening degree of the individual regulating valve 16. The control terminal 30 individually adjusts the flow rate of the fuel gas G supplied to the combustion chambers 11-1 to 11-N so that the temperatures of the combustion chambers 11-1 to 11-N and the carbonization chambers 12-1 to 12-(N-1) respectively become the individually set target temperatures. The control terminal 30 controls the opening degree of the individual regulating valve 16 according to a command from the furnace temperature control device 20.

[0046] Coal, which is the raw material for coke, is charged into the carbonization chamber 12. The carbonization chamber 12 is heated by the heat generated by the combustion chambers 11 adjacent to both sides. The coke oven 10 manufactures coke by heating the carbonization chamber 12 with the heat generated by the combustion chamber 11 and carbonizing the coal charged into the carbonization chamber 12.

[0047] The first thermometers 13-1 to 13-N are respectively installed in the combustion chambers 11-1 to 11-N. The first thermometers 13-1 to 13-N respectively measure the temperatures of the combustion chambers 11-1 to 11-N. The first thermometer 13 may be any thermometer capable of measuring the temperature of the combustion chamber 11. The first thermometer 13 transmits the measured temperature value of the combustion chamber 11 to the control terminal 30.

[0048] The second thermometers 14-1 to 14-(N-1) are respectively installed in the carbonization chambers 12-1 to 12-(N-1). The second thermometers 14-1 to 14-(N-1) respectively measure the temperatures of the carbonization chambers 12-1 to 12-(N-1). The second thermometer 14 may be any thermometer capable of measuring the temperature of the carbonization chamber 12. The second thermometer 14 transmits the measured temperature value of the carbonization chamber 12 to the control terminal 30.

[0049] Note that the coke oven 10 does not necessarily have to have both the first thermometer 13 and the second thermometer 14.

[0050] For example, the coke oven 10 may have the first thermometer 13 and not have the second thermometer 14. In this case, instead of measuring the temperature of the carbonization chamber 12 with the second thermometer 14, the control terminal 30 may calculate the temperature of the carbonization chamber 12 based on other physical quantities or the like. At this time, the control terminal 30 may calculate the temperature of the carbonization chamber 12 based on, for example, a model that associates other physical quantities with the temperature of the carbonization chamber 12. In this embodiment, the temperature of the carbonization chamber 12 calculated in this way is also referred to as the temperature measurement value of the carbonization chamber 12.

[0051] Other physical quantities may include, for example, the temperature measured at a location spatially distant from the carbonization chamber 12. Alternatively, other physical quantities may include, for example, the measured temperature of the gas discharged from the carbonization chamber 12. Furthermore, other physical quantities may include, for example, the measured components of the gas discharged from the carbonization chamber 12.

[0052] Furthermore, for example, the coke oven 10 may be equipped with a second thermometer 14 and not a first thermometer 13. In this case, instead of measuring the temperature of the combustion chamber 11 with the first thermometer 13, the control terminal 30 may calculate the temperature of the combustion chamber 11 based on other physical quantities measured for the combustion chamber 11. In this case, the control terminal 30 may calculate the temperature of the combustion chamber 11 based on a model that associates other physical quantities with the temperature of the combustion chamber 11. In this embodiment, the temperature of the combustion chamber 11 calculated in this way will also be referred to as the measured temperature of the combustion chamber 11.

[0053] The furnace temperature control device 20 controls the temperature of the combustion chambers 11-1 to 11-N or the carbonization chambers 12-1 to 12-(N-1). When controlling the temperature of the combustion chambers 11-1 to 11-N, the furnace temperature control device 20 controls the temperature of each combustion chamber 11 of the combustion chambers 11-1 to 11-N individually. When controlling the temperature of the carbonization chambers 12-1 to 12-(N-1), the furnace temperature control device 20 controls the temperature of each carbonization chamber 12 of the carbonization chambers 12-1 to 12-(N-1) individually.

[0054] Alternatively, the furnace temperature control device 20 may control both the temperature of the combustion chambers 11-1 to 11-N and the temperature of the carbonization chambers 12-1 to 12-(N-1).

[0055] The furnace temperature control device 20 may be a general-purpose computer such as a workstation or personal computer, or it may be a dedicated computer configured to function as a furnace temperature control device 20.

[0056] Figure 2 is a block diagram showing an example of the configuration of a furnace temperature control device 20 according to one embodiment of the present disclosure.

[0057] The furnace temperature control device 20 comprises a control unit 21, an input unit 22, an output unit 23, a storage unit 24, and a communication unit 25.

[0058] The control unit 21 includes at least one processor, at least one dedicated circuit, or a combination thereof. The processor is a general-purpose processor such as a CPU (Central Processing Unit) or GPU (Graphics Processing Unit), or a dedicated processor specialized for a specific process. The dedicated circuit is, for example, an FPGA (Field-Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit).

[0059] The control unit 21 reads programs, data, etc., stored in the storage unit 24 and executes various functions.

[0060] The input unit 22 includes one or more input interfaces that detect user input and acquire input information based on user operations. The input unit 22 includes, for example, physical keys, capacitive keys, a touchscreen integrated with the display of the output unit 23, or a microphone that accepts voice input.

[0061] The output unit 23 includes one or more output interfaces for outputting information and notifying the user. The output unit 23 includes, for example, a display for outputting information as an image, a speaker for outputting information as sound, etc. The display included in the output unit 23 may be, for example, an LCD (Liquid Crystal Display), a CRT (Cathode Ray Tube) display, etc.

[0062] The storage unit 24 is, for example, a flash memory, a hard disk, or an optical memory. Part of the storage unit 24 may be located outside the furnace temperature control device 20. In this case, part of the storage unit 24 may be a hard disk, memory card, or the like, connected to the furnace temperature control device 20 via any interface.

[0063] The memory unit 24 stores programs for the control unit 21 to execute various functions, data used by those programs, and so on.

[0064] The communication unit 25 includes at least one of a communication module that supports wired communication and a communication module that supports wireless communication. The furnace temperature control device 20 can communicate with the control terminal 30 via the communication unit 25.

[0065] Details of the operation of the furnace temperature control device 20 will be described later.

[0066] When the control terminal 30 obtains the temperature measurement value of the combustion chamber 11 from the first thermometer 13, it transmits the obtained temperature measurement value of the combustion chamber 11 to the furnace temperature control device 20. When the control terminal 30 obtains the temperature measurement value of the carbonization chamber 12 from the second thermometer 14, it transmits the obtained temperature measurement value of the carbonization chamber 12 to the furnace temperature control device 20.

[0067] Furthermore, if the control terminal 30 calculates the temperature of the combustion chamber 11 from other physical quantities, it also transmits the calculated temperature of the combustion chamber 11 to the furnace temperature control device 20. Similarly, if the control terminal 30 calculates the temperature of the carbonization chamber 12 from other physical quantities, it also transmits the calculated temperature of the carbonization chamber 12 to the furnace temperature control device 20.

[0068] The control terminal 30 controls the opening degree of the control valve 15 in response to a command from the furnace temperature control device 20. The control terminal 30 also controls the opening degree of the individual control valves 16 in response to a command from the furnace temperature control device 20.

[0069] The control terminal 30 may be a general-purpose computer such as a workstation or personal computer, or it may be a dedicated computer configured to function as a control terminal 30.

[0070] Next, the operation of the furnace temperature control device 20 will be explained.

[0071] The memory unit 24 stores a first model for calculating a predicted temperature for the combustion chamber 11. The memory unit 24 also stores a second model for calculating a predicted temperature for the carbonization chamber 12. Here, the predicted temperature for the combustion chamber 11 is the predicted temperature of the combustion chamber 11 after a predetermined time. The predicted temperature for the carbonization chamber 12 is the predicted temperature of the carbonization chamber 12 after a predetermined time.

[0072] Here, "temperature of combustion chamber 11" refers to the gas temperature inside combustion chamber 11 (for example, the gas temperature in the center of combustion chamber 11), and "temperature of carbonization chamber 12" refers to the core temperature of the coal filled inside carbonization chamber 12. Furthermore, when the first and second models described later are based on the heat conduction equation, the temperature distribution of each part of the combustion chamber, carbonization chamber, and the brick separating them can be calculated, and the gas temperature inside combustion chamber 11 and the core temperature of the coal can be used as predicted temperature values, respectively. In addition, in the model, physical parameters of each part, such as the heat transfer coefficient between the combustion chamber gas and bricks, and between bricks and coal, as well as the heat capacity and thermal conductivity of the coal, can be clearly defined.

[0073] Note that the memory unit 24 does not need to store both the first model and the second model; it only needs to store at least one of the first model and the second model. In this embodiment, the case in which both the first model and the second model are stored will be explained as an example.

[0074] The memory unit 24 stores a separate first model for each of the combustion chambers 11-1 to 11-N. That is, the memory unit 24 stores N first models, each with different parameters, such as a first model for calculating the predicted temperature of combustion chamber 11-1, a first model for calculating the predicted temperature of combustion chamber 11-2, ..., a first model for calculating the predicted temperature of combustion chamber 11-N.

[0075] Furthermore, if the first and second models are based on the heat conduction equation, the parameters in the first and second models shall correspond to the physical characteristics of each part, such as the heat transfer coefficient between the combustion chamber gas and the bricks, the thermal conductivity of the bricks, the heat transfer coefficient between the bricks and the coal, and the heat capacity and thermal conductivity of the coal.

[0076] The memory unit 24 stores a separate second model for each of the carbonization chambers 12-1 to 12-(N-1). That is, the memory unit 24 stores N-1 second models, each with different parameters, such as a second model for calculating the predicted temperature of carbonization chamber 12-1, a second model for calculating the predicted temperature of carbonization chamber 12-2, ..., a second model for calculating the predicted temperature of carbonization chamber 12-(N-1).

[0077] The control unit 21 acquires operational information for the coke oven 10. The control unit 21 may acquire operational information through user input operations to the input unit 22, or it may acquire operational information transmitted by other devices such as the control terminal 30 via the communication unit 25.

[0078] The operational information includes, at a minimum, temperature measurements of combustion chambers 11-1 to 11-N, temperature measurements of carbonization chambers 12-1 to 12-(N-1), actual values ​​of the amount of heat supplied to combustion chambers 11-1 to 11-N, and actual and planned values ​​of the coal filling status of carbonization chambers 12-1 to 12-(N-1).

[0079] The temperature measurement value of the combustion chamber 11 may be the value measured by the first thermometer 13, or it may be a value calculated by the control terminal 30 based on other measured physical quantities. If the temperature measurement value of the combustion chamber 11 is calculated based on other physical quantities, the control unit 21 may perform the calculation.

[0080] The temperature measurement value of the carbonization chamber 12 may be the value measured by the second thermometer 14, or it may be a value calculated by the control terminal 30 based on other measured physical quantities. If the temperature measurement value of the carbonization chamber 12 is calculated based on other physical quantities, the control unit 21 may perform the calculation.

[0081] Furthermore, if direct temperature measurements of the carbonization chamber 12 cannot be obtained, for example, if the core temperature of the coal cannot be measured while the carbonization chamber 12 is filled with coal, the timing of burnout (completion of carbonization) can be determined by utilizing the temperature changes and gas component changes of the gas generated from the carbonization chamber 12. The burnout timing is determined when the temperature of the generated gas drops sharply or when changes in gas component appear. At this burnout time, the core temperature of the coal is theoretically thought to have reached around 900°C, so the temperature can be indirectly estimated by assuming the core temperature of the carbonization chamber 12 at the burnout timing to be 900°C. This allows for highly accurate estimation of the temperature of the carbonization chamber 12 using theoretical or model values, even when actual measurements cannot be obtained.

[0082] The actual values ​​of the amount of heat supplied to the combustion chamber 11 may be expressed as information such as the actual opening degree of the control valve 15 and the opening degree of the individual control valves 16.

[0083] The actual and planned values ​​for the coal filling status of the carbonization chamber 12 may include information on the actual and planned weight of the coal charged into the carbonization chamber 12, and information on the actual and planned moisture content of the coal charged into the carbonization chamber 12.

[0084] In addition to the information above, the operational information may further include information such as the flow rate and composition of the fuel gas G supplied to the entire reactor group.

[0085] The control unit 21 inputs the information contained in the operational information into individual first models and calculates a predicted temperature value for each combustion chamber 11 of combustion chambers 11-1 to 11-N. Alternatively, the control unit 21 inputs the information contained in the operational information into individual second models and calculates a predicted temperature value for each carbonization chamber 12 of carbonization chambers 12-1 to 12-(N-1).

[0086] When the control unit 21 calculates the predicted temperature of the combustion chamber 11, it inputs not only the information about the target combustion chamber 11 from the operational information, but also information about the adjacent carbonization chamber 12 and combustion chamber 11 into the first model to calculate the predicted temperature of the combustion chamber 11.

[0087] Furthermore, information regarding the combustion chamber 11 may include the measured temperature of the combustion chamber 11 and the actual amount of heat supplied to the combustion chamber 11. In addition, information regarding the carbonization chamber 12 may include the measured temperature of the carbonization chamber 12 and the actual and planned values ​​of the coal filling status in the carbonization chamber 12.

[0088] For example, when calculating the predicted temperature of combustion chamber 11-i, the control unit 21 may input not only the information about combustion chamber 11-i from the operational information, but also information about the adjacent carbonization chambers 12-(i-1) and 12-i into the first model corresponding to combustion chamber 11-i. In addition, the control unit 21 may input information about combustion chambers 11-(i-1) and 11-(i+1), and information about carbonization chambers 12-(i-2) and 12-(i+1), and so on, into the first model, which are further adjacent combustion chambers 11 and 12. By inputting information about the adjacent combustion chambers 11 and 12 into the first model in this way, the control unit 21 can calculate the predicted temperature of combustion chamber 11 while also considering the influence of the adjacent combustion chambers 11 and 12.

[0089] In the case of a heat conduction equation model, "input" refers to measured values ​​and operational parameters used to describe physical phenomena, such as temperature measurements for each part, heat supply, coal filling status, and temperature or operational information for adjacent parts. On the other hand, in the case of a machine learning model, "input" refers to all the features (input data) that the model uses for learning and prediction, such as temperature measurements, heat supply, coal filling status, and temperature or operational information for adjacent parts.

[0090] When the control unit 21 calculates the predicted temperature of the carbonization chamber 12, it inputs not only the information about the target carbonization chamber 12 from the operational information, but also information about the adjacent combustion chamber 11 and carbonization chamber 12 into the second model to calculate the predicted temperature of the carbonization chamber 12.

[0091] For example, when calculating the predicted temperature of carbonization chamber 12-i, the control unit 21 may input not only the information about carbonization chamber 12-i from the operational information, but also the information about the adjacent combustion chambers 11-i and 11-(i+1) into the second model corresponding to carbonization chamber 12-i. In addition, the control unit 21 may input information about carbonization chambers 12-(i-1) and 12-(i+1), and information about combustion chambers 11-(i-1) and 11-(i+2), and so on, into the second model, which are further adjacent carbonization chambers 12 and 11. By inputting information about the adjacent combustion chambers 11 and 12 into the second model in this way, the control unit 21 can calculate the predicted temperature of carbonization chamber 12 while also considering the influence of the adjacent combustion chambers 11 and 12.

[0092] The first model includes a first parameter. The control unit 21 calculates the first parameter to minimize a first evaluation function that represents the error between the predicted temperature of the combustion chamber 11 and the measured temperature of the combustion chamber 11 over a predetermined period in the past.

[0093] Here, the first evaluation function can be, for example, the difference at each point in time between the temperature measurement value of the combustion chamber 11 over the most recent 10 or 20 hours and the temperature prediction value by the first model over the same period, or the average of the absolute values ​​of the differences, or the sum of the squares. This allows for the sequential identification of parameters using time-series data over a certain period in the past. The temperature measurement value can be acquired, for example, every hour, and parameter identification can be performed at a similar frequency.

[0094] The second model includes a second parameter. The control unit 21 calculates the second parameter to minimize a second evaluation function that represents the error between the predicted temperature of the carbonization chamber 12 and the measured temperature of the carbonization chamber 12 over a predetermined past period.

[0095] Furthermore, the second evaluation function can, for example, use the difference or the absolute value of the difference between the predicted coal core temperature by the model and the temperature that should theoretically be reached at the time of fire extinction (e.g., 900°C or 1000°C or higher) for each fire extinction timing of the carbonization chamber 12. This makes it possible to identify parameters for each fire extinction timing and correct the changes in heat transfer characteristics for each carbonization chamber 12 with high accuracy.

[0096] In this embodiment, separate heat transfer model parameters (first parameter, second parameter) are set for each combustion chamber 11 and carbonization chamber 12. When identifying and updating these parameters sequentially, data from multiple points in time over a predetermined period is utilized, rather than simply using the most recent measured values.

[0097] Specifically, for the combustion chamber 11, for example, temperature measurement data of the combustion chamber 11 over the past 10 or 20 hours is used to identify a correction coefficient (first parameter) that minimizes the difference between the model calculation and the actual temperature every hour as an evaluation function.

[0098] For the carbonization chamber 12, a correction coefficient (second parameter) is identified at each fire extinction timing so that the model coal core temperature approaches the theoretical value (for example, 900°C or 1000°C or higher). In this way, by designing an evaluation function by combining data from multiple time points and multiple locations, and successively optimizing the parameters, it is possible to correct the different heat transfer characteristics and changes over time for each combustion chamber 11 and carbonization chamber 12 with high accuracy.

[0099] The control unit 21 may calculate the first parameter and the second parameter at an arbitrarily set time cycle and update the first parameter and the second parameter.

[0100] For example, the control unit 21 may update the first parameter when it acquires a temperature measurement value from the combustion chamber 11. Also, for example, the control unit 21 may update the second parameter when it acquires a temperature measurement value from the carbonization chamber 12.

[0101] The predetermined period in the past can be any period, but if the fluctuations of the first and second parameters are large, it is preferable to set a longer predetermined period in the past. Conversely, if the fluctuations of the first and second parameters are small, it is preferable to set a shorter predetermined period in the past. By setting a longer predetermined period in the past, the effects of short-term noise and disturbances can be averaged out, improving the stability and reliability of parameter estimation. On the other hand, by setting a shorter period, the parameters can be quickly adapted to changes in the furnace state and disturbances. Therefore, by appropriately setting the period according to the fluctuations of the parameters, the balance between noise immunity and responsiveness can be optimized, and the accuracy of temperature prediction can be improved.

[0102] The first evaluation function may be the average of the difference or the absolute value of the difference between the predicted temperature of the combustion chamber 11 and the measured temperature of the combustion chamber 11 over a predetermined period in the past, or it may be a function that is weighted over a predetermined period in the past in a time series.

[0103] The second evaluation function may be the average of the difference or the absolute value of the difference between the predicted temperature of the carbonization chamber 12 and the measured temperature of the carbonization chamber 12 over a predetermined period in the past, or it may be a function that is weighted over a predetermined period in the past in a time series.

[0104] Furthermore, the first model may include a second parameter in addition to the first parameter. Similarly, the second model may include the first parameter in addition to the second parameter. By including each other's parameters in the first and second models, the thermal interaction between the combustion chamber 11 and the carbonization chamber 12 can be modeled, and changes in the heat transfer characteristics of both can be corrected with higher accuracy.

[0105] Here, "mutual parameters" refer to physical parameters commonly used in both models, such as the thermal conductivity, heat capacity, and thickness of the brick wall separating the combustion chamber 11 and the carbonization chamber 12 in the case of a heat conduction equation model. By sharing these parameters between both models, the thermal interaction between the combustion chamber 11 and the carbonization chamber 12 can be modeled faithfully, and changes in heat transfer characteristics can be corrected with higher accuracy. On the other hand, in the case of a machine learning model, the combustion chamber model and the carbonization chamber model can use some input features and learning parameters in common (for example, features indicating the degree of deterioration of the brick wall and weights of the shared layer), thereby reflecting the state changes and mutual influences between the two within the model, and improving the overall prediction accuracy and generalization performance.

[0106] The first model may be, for example, a model that calculates a predicted temperature of the combustion chamber 11 based on the heat conduction equation. The second model may be, for example, a model that calculates a predicted temperature of the carbonization chamber 12 based on the heat conduction equation.

[0107] In this case, the first model may be, for example, a model that calculates a predicted temperature of the combustion chamber 11 by solving the heat conduction equation for the alternately connected combustion chambers 11 and carbonization chambers 12 and the bricks separating the combustion chambers 11 and carbonization chambers 12. The first parameter included in the first model may be, for example, a parameter that corrects the heat conductivity between the combustion chamber 11 and the bricks.

[0108] Furthermore, since the thermal conductivity between the combustion chamber 11 and the bricks changes over time due to the aging and wear of the bricks (refractory material), carbon deposits on the brick walls, and changes in operating conditions, it is possible to reflect the actual changes in heat transfer characteristics in the model by setting up parameters to correct for these changes and sequentially identifying and updating them. This makes it possible to maintain the accuracy of the predicted temperature of the combustion chamber 11 and to improve the accuracy of furnace temperature control.

[0109] In this case, the second model may be, for example, a model that calculates a predicted temperature of the carbonization chamber 12 by solving the heat conduction equation for the alternately connected combustion chambers 11 and carbonization chamber 12 and the bricks separating the combustion chambers 11 and carbonization chamber 12. The second parameter included in the second model may be, for example, a parameter that corrects the heat conductivity between the carbonization chamber 12 and the bricks.

[0110] Furthermore, the thermal conductivity between the carbonization chamber 12 and the bricks changes over time or from one carbonization chamber 12 to another due to factors such as the aging and wear of the bricks (refractory material), carbon deposits on the brick walls, and variations in the packing state and properties of the coal. Therefore, by setting up a parameter to correct the thermal conductivity between the carbonization chamber 12 and the bricks and sequentially identifying and updating it, the actual changes in heat transfer characteristics can be reflected in the model. This improves the accuracy of estimating the coal core temperature at the fire extinguishing timing, maintains the accuracy of the predicted temperature value of the carbonization chamber 12, and enables more precise furnace temperature control.

[0111] Furthermore, in the first and second models, the first and second parameters can be sequentially identified for each combustion chamber 11 and carbonization chamber 12. Specifically, the first parameters included in the first model are, for example, the heat transfer coefficient between the combustion chamber 11 and the bricks, heat loss, and gas temperature, while the second parameters included in the second model are, for example, the heat transfer coefficient between the carbonization chamber 12 and the bricks, the heat capacity and thermal conductivity of the carbonization chamber 12, etc. These parameters can then be sequentially optimized based on temperature measurements and operational information for each chamber.

[0112] The optimization of the first and second parameters can be achieved by using an algorithm such as a local search method to search for the evaluation function (the error between the measured value and the model value) to be minimized. The parameter updates can be performed as follows: for the combustion chamber 11, corrections can be made each time a temperature measurement is obtained (e.g., every hour); for the carbonization chamber 12, corrections can be made at times such as when the fire goes out.

[0113] Furthermore, to improve the accuracy of the model, data assimilation techniques can be used to identify the first and second parameters in a way that sequentially minimizes the error between measured values ​​and model values. This allows for flexible adaptation to temporal disturbances and aging of equipment, thereby achieving higher accuracy in future temperature predictions.

[0114] Furthermore, in identifying the first and second parameters, if temperature measurements or other data are unavailable, it is also possible to identify the parameters using theoretical values ​​or other model values.

[0115] Furthermore, regarding the second parameter, even if, for example, temperature measurements of the carbonization chamber cannot be obtained, the parameter can be identified using the indirectly estimated temperature of the carbonization chamber 12 (for example, assuming the core temperature at the time of fire extinguishing to be 900°C). In this way, even when actual measured values ​​cannot be obtained, parameter identification can be performed using theoretical or model values, thereby improving the accuracy of temperature prediction.

[0116] Furthermore, the optimization of the parameters in the first and second models can be performed using a two-step optimization procedure, such as that shown in Figures 3A and 3B. Figures 3A and 3B are flowcharts of the algorithm for finding the optimal values ​​of the correction coefficients (first parameter, second parameter) according to this embodiment, with each step indicated by S1 to S12.

[0117] <First Loop (Figure 3A)> The gas temperature in the combustion chamber 11 is measured at a fixed interval of 30 minutes to 1 hour using a thermometer (thermocouple, etc.) installed at the top of the furnace, and the optimal value of the correction coefficient is searched for each measurement.

[0118] Based on the calculated and measured temperatures of each combustion chamber 11-i (i = 11-1 to 11-N) over the most recent predetermined period, the first parameter (combustion chamber correction coefficient) of the first model for all combustion chambers 11 is optimized.

[0119] First, using a model of all combustion chambers 11 and all carbonization chambers 12, temperature calculations are performed for a predetermined period, such as the most recent 10 hours (S1). A first evaluation function is calculated using the difference between the measured temperature of each combustion chamber 11-i and the temperature predicted by the first model (S2). If the value of the first evaluation function for each combustion chamber 11-i is greater than or equal to a predetermined threshold (No. in S3), the first parameter of each combustion chamber 11-i is updated using a local search method (S4). The values ​​of the first parameters are repeatedly updated and optimized until the first evaluation function for all combustion chambers 11 falls below a predetermined threshold (No. in S5). The number of iterations is limited to a predetermined upper limit (for example, 5 times) (S6).

[0120] <Second Loop (Figure 3B)> Using the first parameter optimized in the first loop, the second parameter (carbonization chamber correction coefficient) is optimized for all carbonization chambers 12-j (j = 12-1 to 12-(N-1)) that have failed within the most recent predetermined period.

[0121] Targeting the carbonization chamber 12-j that has failed (Yes in S7), a second evaluation function is calculated using the difference between the coal core temperature calculated using the first parameter optimized in the first loop and the temperature that should theoretically be reached at the time of failure (e.g., 900°C or 1000°C or higher) (S8). If the value of the second evaluation function for the carbonization chamber 12-j that has failed is greater than or equal to a predetermined threshold (No in S9), the second parameter for that carbonization chamber 12-j is updated using a local search method (S10). The value of the second parameter is repeatedly updated and optimized until the value of the second evaluation function for all carbonization chambers 12-j that have failed in the most recent predetermined period is less than the predetermined threshold (No in S11). The number of repetitions is limited to a predetermined upper limit (e.g., 5 times) (S12).

[0122] Thus, in this embodiment, in a coke oven configured by connecting multiple combustion chambers 11 and multiple carbonization chambers 12, individual parameters are provided to adjust the temperature accuracy in each combustion chamber 11 and each carbonization chamber 12. Because the combustion chambers 11 and carbonization chambers 12 influence each other thermally in a complex manner via bricks, it is difficult to maintain the temperature accuracy of the entire furnace with a simple overall correction.

[0123] In this embodiment, the gas temperature of the combustion chamber 11 can be directly measured at a fixed interval of 30 minutes to 1 hour using a thermocouple at the top of the furnace, etc. Therefore, the first parameter (combustion chamber correction coefficient) can be accurately identified for each combustion chamber 11 using past temperature data for a predetermined period (e.g., 10 hours).

[0124] On the other hand, since the carbonization chamber 12 is filled with coal and its core temperature cannot be directly measured, the coal core temperature for each fire extinction timing is estimated by model calculation, and the difference between this estimated temperature and the theoretically expected temperature at fire extinction (e.g., 900°C or 1000°C or higher) is used to identify a second parameter (carbonization chamber correction coefficient).

[0125] Furthermore, because there are many combustion chambers 11 and carbonization chambers 12 (for example, 53 combustion chambers 11 and 52 carbonization chambers 12), and it is necessary to accurately identify parameters for each part, a two-stage optimization procedure is adopted: first, the parameters of the combustion chambers 11 are optimized overall, and then the parameters of the carbonization chambers 12 are optimized using the optimized parameters of the combustion chambers 11. Since changing one parameter affects the temperature accuracy of other parts, by repeatedly performing this two-stage optimization, the changes and interactions of the heat transfer characteristics of each combustion chamber 11 and carbonization chamber 12 can be corrected with high accuracy as a whole, and the temperature prediction accuracy and parameter identification accuracy of the coke oven as a whole can be greatly improved.

[0126] In addition to using local search methods for updating parameters, other optimization algorithms such as simulated annealing, hill climbing, or genetic algorithms, or methods that gradually adjust the parameter variation range and search range, can also be applied as needed.

[0127] Furthermore, the first model may be, for example, a model that calculates a predicted temperature of the combustion chamber 11 based on a machine learning model. The second model may be, for example, a model that calculates a predicted temperature of the carbonization chamber 12 based on a machine learning model.

[0128] In this case, the first model and the second model may be machine learning models based on past operational information. The first parameter included in the first model may be, for example, a parameter that corrects the error between the predicted temperature of the combustion chamber 11 and the measured temperature of the combustion chamber 11. The second parameter included in the second model may be, for example, a parameter that corrects the error between the predicted temperature of the carbonization chamber 12 and the measured temperature of the carbonization chamber 12. In the machine learning model, the first parameter included in the first model and the second parameter included in the second model correspond to learning parameters (weights, biases, regression coefficients, etc.) of a neural network or regression model, for example, and these are learned and updated to minimize the error (loss function) between the measured temperature of the combustion chamber 11 or carbonization chamber 12 (training data) and the temperature predicted by the model. This improves the prediction accuracy of the model and allows it to respond flexibly to actual changes in the state of the furnace and disturbances.

[0129] Furthermore, the first model may be a model that calculates the predicted temperature of the combustion chamber 11 by solving the heat conduction equation, and the second model may be a model that calculates the predicted temperature of the carbonization chamber 12 based on a machine learning model. Alternatively, the first model may be a model that calculates the predicted temperature of the combustion chamber 11 based on a machine learning model, and the second model may be a model that calculates the predicted temperature of the carbonization chamber 12 based on the heat conduction equation. By using the heat conduction equation model, the physical phenomena inside the furnace can be faithfully reproduced and it can also handle changes in heat transfer characteristics over time. In addition, by using a machine learning model, it becomes possible to make highly accurate temperature predictions, including complex nonlinear relationships and unknown disturbance factors, by utilizing past big data.

[0130] Furthermore, for the combustion chamber 11, the gas temperature can be directly measured using thermometers (such as thermocouples) installed at the top of the furnace, so either a machine learning model utilizing the measured data or a heat conduction equation model can be applied. On the other hand, for the carbonization chamber 12, the coal core temperature cannot be directly measured, so it is necessary to indirectly estimate the core temperature from the timing of fire extinguishing and the gas temperature, etc., and using a heat conduction equation model that can faithfully reproduce physical phenomena is particularly effective.

[0131] As described above, the control unit 21 inputs the information contained in the operational information into individual first models and calculates a predicted temperature value for each of the combustion chambers 11-1 to 11-N. Alternatively, the control unit 21 inputs the information contained in the operational information into individual second models and calculates a predicted temperature value for each of the carbonization chambers 12-1 to 12-(N-1).

[0132] When calculating the predicted temperature of the combustion chamber 11, the control unit 21 calculates the amount of heat to be supplied to the combustion chamber 11 based on the target temperature of the combustion chamber 11 and the calculated predicted temperature of the combustion chamber 11. The target temperature of the combustion chamber 11 may be a predetermined temperature set in advance. The target temperature of the combustion chamber 11 may be a target temperature set individually for each of the combustion chambers 11-1 to 11-N.

[0133] The control unit 21 calculates a temperature prediction value for each of the combustion chambers 11-1 to 11-N.

[0134] The control unit 21 calculates the amount of heat to supply to each combustion chamber 11, based on the difference between the target temperature of the combustion chamber 11 and the calculated predicted temperature of the combustion chamber 11, so that the difference between the temperature of the combustion chamber 11 after a predetermined time and the target temperature of the combustion chamber 11 falls within a predetermined range.

[0135] When the control unit 21 calculates the amount of heat to be supplied to the combustion chamber 11, it may calculate the opening degree of the control valve 15 and the opening degree of the individual control valves 16.

[0136] When calculating the predicted temperature of the carbonization chamber 12, the control unit 21 calculates the predicted temperature of the carbonization chamber 12 and then calculates the amount of heat to be supplied to the combustion chamber 11 based on the target temperature of the carbonization chamber 12 and the calculated predicted temperature of the carbonization chamber 12. The target temperature of the carbonization chamber 12 may be a predetermined temperature set in advance. The target temperature of the carbonization chamber 12 may be a target temperature set individually for each of the carbonization chambers 12-1 to 12-(N-1).

[0137] The control unit 21 calculates a temperature prediction value for each of the carbonization chambers 12-1 to 12-(N-1).

[0138] The control unit 21 calculates the amount of heat to supply to the combustion chamber 11 for each carbonization chamber 12, based on the difference between the target temperature of the carbonization chamber 12 and the calculated predicted temperature of the carbonization chamber 12, so that the difference between the temperature of the carbonization chamber 12 after a predetermined time and the target temperature of the carbonization chamber 12 falls within a predetermined range.

[0139] When the control unit 21 calculates the amount of heat to be supplied to the combustion chamber 11, it may calculate the opening degree of the control valve 15 and the opening degree of the individual control valves 16.

[0140] The control unit 21 calculates the opening degree of the control valve 15 and the opening degree of the individual control valves 16, and then transmits the calculated opening degrees of the control valve 15 and the individual control valves 16 to the control terminal 30 via the communication unit 25.

[0141] When the control terminal 30 receives the opening degree of the control valve 15 and the opening degree of the individual control valve 16 from the furnace temperature control device 20, it adjusts the opening degree of the control valve 15 and the opening degree of the individual control valve 16 based on the received opening degree of the control valve 15 and the opening degree of the individual control valve 16.

[0142] In this case, the control unit 21 may display information such as the calculated opening degree of the control valve 15 and the opening degree of the individual control valves 16 on the output unit 23. This allows the operator to understand how to adjust the opening degree of the control valve 15 and the opening degree of the individual control valves 16. In this case, instead of the control terminal 30 adjusting the opening degree of the control valve 15 and the opening degree of the individual control valves 16, the operator may manually adjust the opening degree of the control valve 15 and the opening degree of the individual control valves 16.

[0143] The furnace temperature control device 20 according to this embodiment calculates a predicted temperature for the combustion chamber 11 or the carbonization chamber 12 based on operational information including measured temperatures for the combustion chambers 11-1 to 11-N, measured temperatures for the carbonization chambers 12-1 to 12-(N-1), actual values ​​of the amount of heat supplied to the combustion chambers 11-1 to 11-N, and actual and planned values ​​of the coal filling status in the carbonization chambers 12-1 to 12-(N-1). Therefore, it can calculate a predicted temperature for the combustion chamber 11 or the carbonization chamber 12 with high accuracy.

[0144] Furthermore, in this embodiment, when calculating the predicted temperature of the combustion chamber 11, the furnace temperature control device 20 inputs not only information about the target combustion chamber 11 but also information about the adjacent carbonization chamber 12 and combustion chamber 11 into the first model to calculate the predicted temperature of the combustion chamber 11, thereby enabling the calculation of the predicted temperature of the combustion chamber 11 with even greater accuracy. Furthermore, in this embodiment, when calculating the predicted temperature of the carbonization chamber 12, the furnace temperature control device 20 inputs not only information about the target carbonization chamber 12 but also information about the adjacent combustion chamber 11 and carbonization chamber 12 into the second model to calculate the predicted temperature of the carbonization chamber 12, thereby enabling the calculation of the predicted temperature of the carbonization chamber 12 with even greater accuracy.

[0145] Furthermore, the furnace temperature control device 20 according to this embodiment controls the control valve 15 and the individual control valve 16 based on the predicted temperature of the combustion chamber 11 or the predicted temperature of the carbonization chamber 12, which have been calculated with high accuracy, thereby enabling high-precision control of the temperature of the combustion chamber 11 and the temperature of the carbonization chamber 12.

[0146] Furthermore, in a coke oven 10 configured by connecting multiple combustion chambers 11 and carbonization chambers 12, the furnace temperature control device 20 sets individual heat transfer model parameters for each combustion chamber 11 and each carbonization chamber 12, and sequentially identifies and updates these parameters. By minimizing the error between the measured values ​​and model values ​​for each part using assimilation technology, it is possible to correct with high accuracy the different heat transfer characteristics and changes over time for each combustion chamber 11 and carbonization chamber 12, such as the aging deterioration and carbon deposition of bricks, and variations in coal properties and packing conditions, thereby improving the accuracy of future temperature predictions.

[0147] The operation of the furnace temperature control device 20 according to this embodiment will be explained with reference to the flowchart shown in Figure 4.

[0148] Step S101: The control unit 21 of the furnace temperature control device 20 acquires operational information about the coke oven 10.

[0149] Step S102: The control unit 21 inputs the information contained in the operation information into individual first models and calculates the predicted temperature of each combustion chamber 11 of combustion chambers 11-1 to 11-N. Alternatively, the control unit 21 inputs the information contained in the operation information into individual second models and calculates the predicted temperature of each carbonization chamber 12 of carbonization chambers 12-1 to 12-(N-1).

[0150] The control unit 21 may update the first parameters included in the first model and the second parameters included in the second model based on the operational information acquired in step S101 before processing in step S102. Alternatively, the control unit 21 may execute the processing in step S102 based on the first and second models created previously, without updating the first and second parameters after step S101.

[0151] Step S103: The control unit 21 calculates the amount of heat to supply to the combustion chamber 11 based on the target temperature of the combustion chamber 11 and the calculated predicted temperature of the combustion chamber 11. Alternatively, the control unit 21 calculates the amount of heat to supply to the combustion chamber 11 based on the target temperature of the carbonization chamber 12 and the calculated predicted temperature of the carbonization chamber 12.

[0152] When the control unit 21 calculates a predicted temperature for the combustion chamber 11, it calculates the amount of heat to supply to the combustion chamber 11 based on the difference between the target temperature of the combustion chamber 11 and the calculated predicted temperature for the combustion chamber 11, so that the difference between the temperature of the combustion chamber 11 after a predetermined time and the target temperature of the combustion chamber 11 is within a predetermined range. Alternatively, when the control unit 21 calculates a predicted temperature for the carbonization chamber 12, it calculates the amount of heat to supply to the combustion chamber 11 based on the difference between the target temperature of the carbonization chamber 12 and the calculated predicted temperature for the carbonization chamber 12, so that the difference between the temperature of the carbonization chamber 12 after a predetermined time and the target temperature of the carbonization chamber 12 is within a predetermined range.

[0153] When calculating the amount of heat to be supplied to the combustion chamber 11, the control unit 21 may calculate the amount of heat to be supplied to the combustion chamber 11 based on a model that represents the relationship between the amount of heat to be supplied and the change in temperature. This model may, for example, model the relationship between the amount of heat to be supplied and the change in temperature based on the mass and specific heat of the fuel gas in the combustion chamber 11, the coal in the carbonization chamber 12, and the materials related to the furnace wall. Alternatively, this model may be, for example, a model constructed based on the relationship between the change in the amount of heat supplied and the change in temperature of the combustion chamber 11 or the carbonization chamber 12 in past performance data.

[0154] In step S103, when the control unit 21 calculates the amount of heat to be supplied to the combustion chamber 11, it may calculate the opening degree of the control valve 15 and the opening degree of the individual control valves 16.

[0155] When calculating the opening degree of the control valve 15 and the opening degree of the individual control valves 16, the control unit 21 may calculate the opening degree of the control valve 15 and the opening degree of the individual control valves 16 based on a model that represents the relationship between the opening degree of the control valve 15 and the opening degree of the individual control valves 16 and the amount of heat supplied. This model may, for example, be a model that models the relationship between the opening degree of the control valve 15 and the opening degree of the individual control valves 16 and the amount of heat supplied based on physical laws. Alternatively, this model may be, for example, a model constructed based on the relationship between the opening degree of the control valve 15 and the opening degree of the individual control valves 16 and the amount of heat supplied in past performance data.

[0156] Step S104: The control unit 21 transmits the calculated opening degrees of the control valve 15 and the individual control valves 16 to the control terminal 30 via the communication unit 25. The control terminal 30 adjusts the opening degrees of the control valve 15 and the individual control valves 16 based on the acquired opening degrees of the control valve 15 and the individual control valves 16.

[0157] As described above, in the furnace temperature control device 20 according to this embodiment, the control unit 21 acquires operational information including temperature measurements of multiple combustion chambers 11, temperature measurements of multiple carbonization chambers 12, actual values ​​of the amount of heat supplied to the multiple combustion chambers 11, and actual and planned values ​​of the coal filling status of the multiple carbonization chambers 12. The control unit 21 inputs the information contained in the operational information into individual first models to calculate predicted temperatures for each of the multiple combustion chambers 11, or inputs the information contained in the operational information into individual second models to calculate predicted temperatures for each of the multiple carbonization chambers 12. The control unit 21 then calculates the amount of heat to be supplied to each of the multiple combustion chambers 11 based on the target temperature of the combustion chamber 11 and the calculated predicted temperature of the combustion chamber 11, or based on the target temperature of the carbonization chamber 12 and the calculated predicted temperature of the carbonization chamber 12. As described above, the furnace temperature control device 20 according to this embodiment calculates a predicted temperature value for the combustion chamber 11 or the carbonization chamber 12 based on the acquired operational information. Since the operational information includes multiple furnace temperature fluctuation factors, it is possible to predict the temperature of the combustion chamber 11 or the carbonization chamber 12 with high accuracy. Therefore, the furnace temperature control device 20 according to this embodiment can predict the future temperature of the coke oven 10 with high accuracy and control the future temperature of the coke oven 10 with high accuracy.

[0158] Furthermore, the furnace temperature control device 20 according to this embodiment can accurately correct for changes in characteristics over time and individual differences by sequentially identifying and updating individual heat transfer model parameters for each combustion chamber 11 and each carbonization chamber 12, thereby improving the accuracy of future temperature predictions.

[0159] This disclosure is not limited to the embodiments described above. For example, multiple blocks described in the block diagram may be combined, or a single block may be divided. Instead of executing multiple steps described in the flowchart in chronological order as described, they may be executed in parallel or in a different order, depending on the processing capacity of the device performing each step, or as necessary. Other modifications are possible without departing from the spirit of this disclosure.

[0160] For example, in the above-described embodiment, the case in which the furnace temperature control device 20 and the control terminal 30 are separate devices was explained, but the furnace temperature control device 20 and the control terminal 30 may be an integrated device. That is, the furnace temperature control device 20 may have the functions of the control terminal 30.

[0161] 1 Furnace temperature control system 10 Coke oven 11 Combustion chamber 12 Carbonization chamber 13 First thermometer 14 Second thermometer 15 Control valve 16 Individual control valve 17 Main gas pipe 18 Branch gas pipe 20 Furnace temperature control device 21 Control unit 22 Input unit 23 Output unit 24 Memory unit 25 Communication unit 30 Control terminal

Claims

1. A furnace temperature control device for controlling the temperature of a combustion chamber or a carbonization chamber in a coke oven in which a plurality of combustion chambers and a plurality of carbonization chambers are alternately connected to form a furnace group, comprising: a storage unit that stores at least one of a first model for calculating a predicted temperature value of the combustion chamber and a second model for calculating a predicted temperature value of the carbonization chamber; and a control unit, wherein the storage unit, when storing the first model, stores an individual first model for each of the plurality of combustion chambers; and when storing the second model, stores an individual second model for each of the plurality of carbonization chambers; and the control unit acquires operational information including the temperature measured values ​​of the plurality of combustion chambers, the temperature measured values ​​of the plurality of carbonization chambers, the actual value of the amount of heat supplied to the plurality of combustion chambers, and the actual and planned values ​​of the coal filling status of the plurality of carbonization chambers. A furnace temperature control device that inputs the information contained in the operational information into an individual first model to calculate a predicted temperature for each of the plurality of combustion chambers, or inputs the information contained in the operational information into an individual second model to calculate a predicted temperature for each of the plurality of carbonization chambers, and calculates the amount of heat to be supplied to each of the plurality of combustion chambers based on the target temperature of the combustion chamber and the calculated predicted temperature of the combustion chamber, or based on the target temperature of the carbonization chamber and the calculated predicted temperature of the carbonization chamber.

2. The furnace temperature control device according to claim 1, wherein when the control unit calculates the predicted temperature of the combustion chamber, it inputs information relating to the combustion chamber and at least information relating to the adjacent carbonization chamber from the information included in the operation information to the first model corresponding to the combustion chamber, and when the control unit calculates the predicted temperature of the carbonization chamber, it inputs information relating to the carbonization chamber and at least information relating to the adjacent combustion chamber from the information included in the operation information to the second model corresponding to the carbonization chamber.

3. The furnace temperature control device according to claim 1 or 2, wherein the first model includes a first parameter, the second model includes a second parameter, the control unit calculates the first parameter to minimize a first evaluation function representing the error between a predicted temperature of the combustion chamber and a measured temperature of the combustion chamber over a predetermined past period, and the control unit calculates the second parameter to minimize a second evaluation function representing the error between a predicted temperature of the carbonization chamber and a measured temperature of the carbonization chamber over a predetermined past period.

4. The furnace temperature control device according to claim 3, wherein the first evaluation function is the difference or the average value of the absolute values ​​of the difference between the measured temperature of the combustion chamber and the predicted temperature of the first model at each point in time during the most recent predetermined period, and the second evaluation function is the difference or the absolute value of the difference between the predicted coal core temperature of the second model and the temperature that should theoretically be reached at the time of fire extinguishing, for each timing of fire extinguishing in the carbonization chamber.

5. The furnace temperature control device according to claim 4, wherein the identification of the first parameter and the second parameter is performed by: firstly optimizing the first parameter for each of the combustion chambers so that the first evaluation function is less than a predetermined threshold; then, as a second loop, using the optimized first parameter, optimizing the second parameter for each of the carbonization chambers so that the second evaluation function is less than a predetermined threshold; and further, if either the first parameter or the second parameter is changed, the first loop and the second loop are repeated a predetermined number of times.

6. The furnace temperature control device according to claim 4 or 5, wherein the most recent predetermined period is adjusted according to the magnitude of the fluctuation of the first parameter or the second parameter, and the most recent predetermined period is set to be longer when the fluctuation of the first parameter or the second parameter is large, and the most recent predetermined period is set to be shorter when the fluctuation of the first parameter or the second parameter is small.

7. The furnace temperature control device according to claim 3, wherein the first model further includes the second parameter, and the second model further includes the first parameter.

8. The furnace temperature control device according to claim 1 or 2, wherein the first model is a model for calculating a predicted temperature of the combustion chamber based on the heat conduction equation, and the second model is a model for calculating a predicted temperature of the carbonization chamber based on the heat conduction equation.

9. The furnace temperature control device according to claim 1 or 2, wherein the first model is a model that calculates a predicted temperature value for the combustion chamber based on a machine learning model, and the second model is a model that calculates a predicted temperature value for the carbonization chamber based on a machine learning model.

10. The furnace temperature control device according to claim 1 or 2, wherein the first model is a model for calculating a predicted temperature of the combustion chamber based on a heat conduction equation, and the second model is a model for calculating a predicted temperature of the carbonization chamber based on a machine learning model.

11. The furnace temperature control device according to claim 1 or 2, wherein the first model is a model that calculates a predicted temperature value of the combustion chamber based on a machine learning model, and the second model is a model that calculates a predicted temperature value of the carbonization chamber based on a heat conduction equation.

12. A furnace temperature control method in a furnace temperature control device for controlling the temperature of a combustion chamber or a carbonization chamber in a coke oven in which a plurality of combustion chambers and a plurality of carbonization chambers are alternately connected to form a furnace group, wherein the furnace temperature control device includes a storage unit that stores at least one of a first model for calculating a predicted temperature value of the combustion chamber and a second model for calculating a predicted temperature value of the carbonization chamber, the storage unit, when storing the first model, stores an individual first model for each of the plurality of combustion chambers, and when storing the second model, stores an individual second model for each of the plurality of carbonization chambers, and the furnace temperature control method includes the step of acquiring operational information including the temperature measured values ​​of the plurality of combustion chambers, the temperature measured values ​​of the plurality of carbonization chambers, the actual value of the amount of heat supplied to the plurality of combustion chambers, and the actual and planned values ​​of the coal filling status of the plurality of carbonization chambers, A furnace temperature control method comprising: inputting the information contained in the operational information into an individual first model to calculate a predicted temperature for each of the plurality of combustion chambers, or inputting the information contained in the operational information into an individual second model to calculate a predicted temperature for each of the plurality of carbonization chambers; and calculating the amount of heat to be supplied to each of the plurality of combustion chambers based on the target temperature of the combustion chamber and the calculated predicted temperature of the combustion chamber, or based on the target temperature of the carbonization chamber and the calculated predicted temperature of the carbonization chamber.

13. The furnace temperature control method according to claim 12, wherein the step of calculating the predicted temperature of the combustion chamber is to input information relating to the combustion chamber and at least information relating to an adjacent carbonization chamber from the information contained in the operation information into the first model corresponding to the combustion chamber, and the step of calculating the predicted temperature of the carbonization chamber is to input information relating to the carbonization chamber and at least information relating to an adjacent combustion chamber from the information contained in the operation information into the second model corresponding to the carbonization chamber.

14. A method for producing coke, comprising controlling the amount of heat supplied to each of the plurality of combustion chambers based on the amount of heat supplied to each of the plurality of combustion chambers calculated by the furnace temperature control method described in claim 12 or 13, thereby producing coke.

Images