Coke manufacturing process control device, method and program

Abstract

To enable a physical amount representing a carbonized state of coke to be adjusted according to a target value and reduce variations in the physical amount representing the carbonized state of coke.SOLUTION: A controller for a coke production process 100 controls a heat input so as to bring a coke temperature to close to a target temperature. A target furnace temperature calculator 102 uses a coke temperature prediction model to predict a coke temperature on the basis of an influencer including an oven battery temperature. On the basis of an evaluation function including a term representing a difference between the coke temperature predicted by the coke temperature prediction model and the target temperature, the target furnace temperature calculator calculates a target oven battery temperature for a plurality of inter-block times to come, so that the coke temperature is adjusted according to the target temperature. A heat input calculator 103 calculates a heat input according to the target oven battery temperature calculated by the target furnace temperature calculator 102. Thus, considering how the coke temperature for each block will change depending on the oven battery temperature, it is possible to dynamically calculate the target oven battery temperature for a plurality of inter-block times to come.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a control device, method, and program for a coke manufacturing process. [Background technology]

[0002] Patent Document 1 describes a technology related to the control of the coke production process. In Patent Document 1, a block discharge method is used, in which all carbonization chambers in a coke oven are grouped into multiple groups (blocks). Then, a target oven temperature for each group is calculated using a model based on the predicted amount of coal to be charged in the future, the predicted moisture content, the planned carbonization time, and the actual oven temperature. The target oven battery temperature (optimum oven temperature) is determined by a weighted average, and the input heat amount is calculated to achieve the target oven battery temperature. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 9-302350 [Patent Document 2] Japanese Patent Application Laid-Open No. 2009-75737 [Non-patent literature]

[0004] [Non-Patent Document 1] Flower Pollination Algorithm for Global Optimization, arXiv.org, Dec 19, 2013 [Non-patent document 2] Model Predictive Control-III: Generalized Predictive Control (GPC) and Related Topics, Shiro Masuda, Toru Yamamoto, Masahiro Oshima, Systems / Control / Information, 2002, Vol. 46(9), pp. 578-584 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the above-mentioned Patent Document 1, the target furnace battery temperature is calculated by taking a weighted average of each target furnace temperature, so the target furnace battery temperature is different from each target furnace temperature. In other words, the value of each coke temperature is not taken into consideration depending on the finally determined target furnace battery temperature, and the calculation is static. For this reason, it is difficult to bring the coke temperature after carbonization close to the target temperature and to suppress the variation in the coke temperature after carbonization.

[0006] The present invention has been made in consideration of the above points, and aims to make it possible to set physical quantities representing the carbonization state of coke to values ​​corresponding to target values, and to suppress variation in the physical quantities representing the carbonization state of coke. [Means for solving the problem]

[0007] The control device for a coke production process of the present invention is a control device for a coke production process in a coke oven having a plurality of carbonization chambers and a plurality of combustion chambers, and controls an amount of heat input to the combustion chambers so that a physical quantity representing a carbonization state of coke becomes a value corresponding to a target value. The control device includes: a target oven temperature calculation unit that predicts the physical quantity using a physical quantity prediction model that predicts the physical quantity based on a first influencing factor including an oven temperature, which is the temperature of the combustion chamber, and at least one of an actual value and a schedule value of the first influencing factor, calculates a value of a first evaluation function that includes a term that represents a difference between the predicted physical quantity and a target value of the physical quantity, and calculates a target oven temperature based on the calculated value of the first evaluation function; and an input heat amount calculation unit that calculates an input heat amount corresponding to the target oven temperature calculated by the target oven temperature calculation unit. The physical quantity is the coke temperature or the furnace wall temperature. do. The control method for a coke production process of the present invention is a control method for a coke production process in a coke oven having a plurality of carbonization chambers and a plurality of combustion chambers, in which an input heat amount to the combustion chambers is controlled so that a physical quantity representing a carbonization state of coke becomes a value corresponding to a target value, the control method including: a target oven temperature calculation step of predicting the physical quantity using a physical quantity prediction model that predicts the physical quantity based on a first influencing factor including an oven temperature, which is the temperature of the combustion chambers, and at least one of an actual value and a schedule value of the first influencing factor; calculating a value of a first evaluation function that includes a term representing a difference between the predicted physical quantity and a target value of the physical quantity; and calculating a target oven temperature based on the calculated value of the first evaluation function; and an input heat amount calculation step of calculating an input heat amount corresponding to the target oven temperature calculated in the target oven temperature calculation step. The physical quantity is the coke temperature or the furnace wall temperature. do. The program of the present invention causes a computer to function as each part of the control device for the coke manufacturing process. [Effects of the Invention]

[0008] According to the present invention, it is possible to set the physical quantity representing the carbonization state of the coke to a value corresponding to the target value and to suppress the variation in the physical quantity representing the carbonization state of the coke. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a diagram illustrating a functional configuration of a control device for a coke production process according to a first embodiment. FIG. [Figure 2] FIG. 1 is a diagram showing a schematic configuration of a coke oven. [Figure 3A] FIG. 1 is a diagram showing an outline of a coke oven during carbonization. [Figure 3B] FIG. 1 is a diagram showing an outline of a coke oven during the unloading (extrusion) operation. [Figure 4] FIG. 10 is a diagram showing the changes in coke temperature and furnace bed temperature in one carbonization chamber from loading to extrusion. [Figure 5] 5 is a flowchart showing the processing of a target furnace temperature calculation unit in the first embodiment. [Figure 6]FIG. 6 is a diagram for explaining an outline of the processing in the flowchart of FIG. 5. [Figure 7] 4 is a flowchart showing the processing of an input heat amount calculation unit in the first embodiment. [Figure 8] FIG. 1 is a diagram showing the relationship between furnace battery temperature, coke temperature, and input heat amount. [Figure 9] FIG. 10 is a diagram for explaining a coke temperature prediction model in the second embodiment. [Figure 10] FIG. 1 is a characteristic diagram showing the results of Example 1. [Figure 11] 11 is a flowchart showing the processing of a target furnace temperature calculation unit and an input heat amount calculation unit in the third embodiment. [Figure 12] FIG. 12 is a diagram for explaining an outline of the processing in the flowchart of FIG. [Figure 13] FIG. 10 is a characteristic diagram showing the results of Example 2. [Figure 14] FIG. 1 is a diagram for explaining the influence of a delay in unloading from the oven on a coke temperature. [Figure 15] FIG. 10 is a diagram showing the functional configuration of a control device for a coke production process according to a fourth embodiment. [Figure 16] 10 is a flowchart showing the processing of a target furnace temperature calculation unit, an input heat amount calculation unit, and a carbonization time correction unit in the fourth embodiment. [Figure 17] FIG. 10 is a diagram illustrating an example of a delay in unloading from the kiln. [Figure 18] FIG. 10 is a diagram for explaining another example of a delay in unloading from the kiln. [Figure 19] FIG. 10 is a characteristic diagram showing the results of Example 3. [Figure 20] FIG. 10 is a diagram for explaining the influence on the coke temperature of an operation based on the operator's judgment when a delay in unloading from the oven occurs. [Figure 21] FIG. 10 is a diagram for explaining a method for correcting the estimated dry distillation time. [Figure 22] FIG. 10 is a diagram showing the functional configuration of a control device for a coke production process according to a sixth embodiment. [Figure 23]13 is a flowchart showing the processing of a target furnace temperature calculation unit, an input heat amount calculation unit, and a target furnace temperature correction unit in the sixth embodiment. [Figure 24] FIG. 10 is a diagram for explaining an outline of the process for correcting the target furnace battery temperature pattern. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings. [First embodiment] The schematic configuration of the coke oven 1 and an outline of the coke manufacturing process will be described with reference to FIGS. 2, 3A, and 3B. In a coke oven 1, carbonization chambers (kilns) 2 and combustion chambers 3 are arranged alternately with furnace walls 4 interposed therebetween. The carbonization chambers 2 carbonize charged coal to obtain coke. The combustion chambers 3 burn fuel gas to keep the carbonization chambers 2 at high temperatures.

[0011] In the coke production process using the coke oven 1, the so-called block unloading method is adopted for the unloading coal charging operation. The unloading coal charging operation consists of the operation of pushing coke out of the coke chamber 2 using an extruder (unloading operation), and the subsequent operation of supplying coal to the coke chamber 2 (charring operation). Note that hereinafter, unloading is also referred to as "extrusion." In the block unloading method, the entire coke chamber 2 is divided into multiple blocks (hereinafter referred to as "blocks"). In this embodiment, it is divided into five blocks. Specifically, it is divided into Blocks 1 (coke chambers No. 1, 6, 11, 16, etc.), Block 2 (coke chambers No. 2, 7, 12, 17, etc.), Block 3 (coke chambers No. 3, 8, 13, 18, etc.), Block 4 (coke chambers No. 4, 9, 14, 19, etc.), and Block 5 (coke chambers No. 5, 10, 15, 20, etc.), with intervals of five blocks. The order of loading the coals before unloading is, for example, 1, 3, 5, 2, and 4, to prevent a sudden drop in temperature. Furthermore, within each sequence, the loading work is carried out in order from the earliest sequence. The time from the timing at which loading work is completed for one sequence to the timing at which loading work is completed for the next sequence (for example, 3, which follows 1 in the above example) is called the loading time. The loading time is generally about 3 to 6 hours. This embodiment is not limited to the block unloading method. For example, if the sequence (block) is treated as an individual coking chamber 2 in the following explanation, it can also be applied to a case where loading work is carried out for one coking chamber 2 at a time.

[0012] In addition, in the coke oven manufacturing process, furnace battery control is performed to simultaneously adjust the heat input to all combustion chambers 3 and control the average carbonization state for each type. That is, the heat input to the coke oven 1 is controlled by operating a single regulating valve 5 installed for all combustion chambers 3. The regulating valve 5 is a valve for adjusting the flow rate of a mixture of fuel gas and combustion air. The regulating valve 5 is operated via an actuator (not shown) under the control of a coke manufacturing process control device 100 (described later). The representative value of the temperature of all combustion chambers 3 is called the furnace battery temperature. For example, thermometers 6 measuring the ambient temperature of the combustion chambers 3 are installed in multiple combustion chambers 3, and the average temperature of the combustion chambers where the thermometers 6 are installed is defined as the furnace battery temperature. In this embodiment, the furnace temperature, which is the temperature in the combustion chambers 3 of the coke oven 1, is defined as the furnace battery temperature. Note that the method of this embodiment is not limited to simultaneously adjusting the heat input to all combustion chambers 3. For example, when the unloading and loading work is performed in units of one carbonization chamber 2, an adjustment valve and an actuator may be installed in each combustion chamber 3, and the carbonization state (input heat amount) may be controlled for each carbonization chamber 2. Also, the thermometer 6 may be installed in each of all combustion chambers 3, or only in some of the combustion chambers 3. For example, the thermometer 6 may be installed in all combustion chambers 3, and the temperature of each combustion chamber 3 may be used as the furnace temperature instead of the furnace battery temperature.

[0013] As described above, coke is pushed out of the coke chamber 2 by the pusher. In the example shown in FIG. 3B, coke 10 pushed out of the coke chamber 2 by the pusher ram 7 of the pusher is discharged via a guide car 9 into a fire extinguishing car (not shown) located below the guide car 9, and then transported to a downstream process by the fire extinguishing car. The guide car 9 moves to the position of the coke chamber 2 where the unloading and loading work is performed. In FIG. 3B, the coke 10 produced in the coke chamber 2 located at the bottom of FIG. 3B is discharged via the guide car 9 into a fire extinguishing car (not shown) to complete the unloading and loading work, and then the guide car 9 moves to the coke chamber 2 located at the top of FIG. 3B. This is indicated by the two-dot chain line after the guide car 9 has moved. In addition, in FIG. 3B, a thermometer 8 for non-contactly measuring the temperature of the coke 10 is installed inside the guide car 9. The thermometer 8 is installed so as to view the path of the coke 10 inside the guide car 9 through a window provided in the guide car 9. In this manner, in this embodiment, the temperature of the coke 10 is measured immediately after it leaves the coke chamber 2 during the unloading operation (at the time of extrusion). However, the temperature of the coke 10 does not necessarily have to be measured in this manner as long as the temperature of the coke leaving the coke chamber 2 is measured. The temperature of the coke 10 when it is discharged from the coke chamber 2 (at the time of extrusion) is called the coke temperature.

[0014] The coke temperature is calculated, for example, from the value measured by the thermometer 8 shown in FIG. 3B. When the extrusion ram 7 is pushing the coke 10 out of the coke chamber 2, the temperature of the coke 10 sequentially discharged from the coke chamber 2 is measured by the thermometer 8, and the average value of the temperatures measured at each time and each position (the sum of the temperatures measured at each time and each position divided by the number of temperature measurements) is taken as the temperature of the coke 10 produced in the coke chamber 2. The average value of the temperatures of the coke 10 produced in the coke chambers 2 belonging to one run is taken as the coke temperature (run average value). The run average value is the arithmetic mean value (the sum of the temperatures of the coke produced in the coke chambers 2 belonging to one run divided by the number of coke chambers 2 belonging to that run). It should be noted that the coke temperature is preferably the temperature of the coke 10 immediately after it is discharged from the carbonization chamber 2, and therefore the coke temperature is determined as shown in FIG. 3B. However, the thermometer for measuring the coke temperature and the method for determining the coke temperature may be, for example, any of those employed in a coke factory, and are not limited to those described above.

[0015] 1 shows the functional configuration of a control device 100 (hereinafter simply referred to as the control device) for a coke production process according to a first embodiment. The hardware of the control device 100 is realized, for example, by using an information processing device including a processor such as a central processing unit, a main memory device, an auxiliary memory device, an input device, and an output device. The hardware of the processing device 300 may be realized by a PLC (Programmable Logic Controller) or dedicated hardware such as an ASIC (Application Specific Integrated Circuit). The control device 100 includes an input unit 101, a target furnace temperature calculation unit 102, an input heat amount calculation unit 103, and an input heat amount setting unit 104, and controls the input heat amount so that the coke temperature (i.e., the coke temperature at the time of extrusion) becomes a value corresponding to a predetermined target temperature.

[0016] The input unit 101 inputs operation data of the cokemaking process. The operation data includes past and present operation performance values ​​and future operation schedule values. More specifically, the operation performance values ​​include data indicating the control content and operation status of the cokemaking process, such as first influence factors that affect the coke temperature, such as the measurements of the thermometers 8 and the oven battery temperature, second influence factors that affect the oven battery temperature, such as the input heat amount, and information on the coke before being pushed into the coke oven 1. The operation schedule values ​​include various schedule values, such as the target value (target temperature) of the coke temperature, the availability rate of the coke oven, and the amount of coal charged. The memory unit 200 stores operation data of the cokemaking process together with time-series information such as time, and the input unit 101 inputs the operation data from the memory unit 200. For example, the sensor values ​​of various sensors installed to monitor the coke production process, such as measurements by the thermometer 8, are periodically stored in the storage unit 200, thereby enabling real-time monitoring of the state of the coke production process. The coke temperature (average value of the run) obtained from the measurements by the thermometer 6 as described above is also stored in the storage unit 200 as one piece of operation data (actual value of the coke temperature). While the example in which the input unit 101 inputs operation data from the storage unit 200 has been described, the input unit 101 may input operation data from an external device via a network, or the user may input operation data directly. The control device 100 may be provided with the storage unit 200.

[0017] The target oven temperature calculation unit 102 uses a coke temperature prediction model for predicting the coke temperature based on influencing factors (first influencing factors) on the coke temperature, including the oven battery temperature, to calculate target oven battery temperatures for multiple future pass times (= N × pass times (N is an integer of 2 or more; N = 5 in Figure 2)) so that the coke temperature measured as described above will be a value corresponding to the target temperature (target value). The target oven temperature calculation unit 102 has a coke temperature prediction function and an oven battery temperature optimization function, details of which will be described later. The coke temperature prediction function predicts the coke temperature using the coke temperature prediction model. The oven battery temperature optimization function calculates the target oven battery temperature based on an evaluation function that includes a term representing the difference between the coke temperature predicted by the coke temperature prediction model and the target coke temperature.

[0018] The input heat quantity calculation unit 103 calculates the input heat quantity according to the target furnace battery temperature calculated by the target furnace temperature calculation unit 102. The input heat quantity calculation unit 103 has a furnace battery temperature prediction function and an input heat quantity optimization function, details of which will be described later. The furnace battery temperature prediction function predicts the furnace battery temperature using a furnace battery temperature prediction model that predicts the furnace battery temperature based on influence factors (second influence factors) that include the input heat quantity. The input heat quantity optimization function calculates the input heat quantity using the target furnace battery temperature calculated by the target furnace temperature calculation unit 102, based on an evaluation function that includes a term that represents the difference between the furnace battery temperature predicted by the furnace battery temperature prediction model and the target furnace battery temperature.

[0019] The input heat amount setting unit 104 outputs the input heat amount calculated by the input heat amount calculation unit 103 to a control device of an actuator (not shown) so that the input heat amount is reflected in the coke production process. The control device of the actuator operates the regulating valve 5 (see FIG. 2) via the actuator (not shown) to set the opening degree of the regulating valve 5 to an opening degree corresponding to the input heat amount calculated by the input heat amount calculation unit 103.

[0020] The target furnace temperature calculation unit 102 and the input heat amount calculation unit 103 will be described in detail below. First, the furnace battery temperature prediction model used in the input heat amount calculation unit 103 will be explained. The furnace battery temperature prediction model is a model that predicts the furnace battery temperature based on a second influencing factor including the input heat amount. In this embodiment, as shown in equation (1), a regression model is used that includes the furnace battery temperature in the past as an explanatory variable, which is the furnace battery temperature to be predicted. T ro represents the furnace temperature [℃], Q represents the input heat [GJ / h], S represents the amount of coal charged [tons], and W represents the coal moisture content [%]. Also, t represents the time ahead of the specified time from the current time (the following explanation assumes that the specified time is 1 hour). The objective variable ΔT ro (t), ΔQ(ti) and ΔT as explanatory variables ro (ti), ΔS(ti), and ΔW(ti) are the furnace battery temperatures T ro , input heat Q, furnace temperature T ro , the amount of coal charged S, and the amount of coal moisture W are expressed as the change from one hour before.

[0021]

number

[0022] Table 1 shows the relationship between the explanatory variables (○) and the objective variables (★) of the furnace battery temperature prediction model used in this embodiment. In order to take into account the time delay, the actual values ​​for multiple consecutive periods in one-hour cycles (furnace battery temperature T ro For all other cases, the schedule value for one hour ahead is used as the explanatory variable, and the furnace battery temperature prediction model predicts the furnace battery temperature for one hour ahead, which is the target variable. In detail, each item marked with a circle from one hour ahead (t) to five hours ahead (t-6) is used as an explanatory variable. Since the furnace battery temperature is subject to periodic fluctuations caused by operation, it is preferable to apply a moving average filter as preprocessing to suppress periodic fluctuations.

[0023] [Table 1]

[0024] The specific contents of the objective variables and explanatory variables in the furnace battery temperature prediction model are explained below. <Objective variable> ΔT ro(t): Furnace temperature T from time t-1 to time t ro The change in (ΔT ro (t)=T ro (t)-T ro (t-1) <Explanatory variables> ΔQ(ti): The change in the input heat quantity Q from time ti-1 to time ti (ΔQ(ti) = Q(ti) - Q(ti-1) ΔT ro (ti): Furnace temperature T from time ti-1 to time ti ro The change in (ΔT ro (ti)=T ro (ti)-T ro (ti-1) ΔS(ti): Change in coal amount S from time ti-1 to time ti (ΔS(ti) = S(ti) - S(ti-1) ΔW(ti) is the change in coal moisture content W from time ti-1 to time ti (ΔW(ti) = W(ti) - W(ti-1)

[0025] Coefficient a in Equation (1) i , b i , c i , d i are the explanatory variables ΔQ(ti) and ΔT ro (ti), ΔS(ti), and ΔW(ti). Coefficient a i , b i , c i , d i The coefficients that best fit the form of equation (1) to the past operation results of the coke oven 1 are separately obtained. For example, a set of ΔQ(ti), ΔT ro The data for (ti), ΔS(ti), and ΔW(ti) are used as training data to create a large number of training data, and multiple regression analysis is performed using the training data to obtain the coefficient a i , b i , c i , d i All we need to do is find the answer.

[0026] The coal amount S is the total value of the values ​​in all the coking chambers 2 included in the coke oven 1 to be predicted, and the coal moisture content W is the average value of the values ​​in all the coking chambers 2 included in the coke oven 1 to be predicted. The coal moisture content W is expressed, for example, as the mass ratio (mass%) of coal. In addition, the explanatory variables ΔQ(ti) and ΔT ro When calculating ΔS(ti), ΔW(ti), and ΔT(ti), actual values ​​are used for past values, and for future values, schedule values ​​(values ​​determined in the operation schedule) or values ​​already calculated using formula (1) are used. ro (t) is the furnace battery temperature T ro (t) is used. By updating the value of t every hour, the furnace battery temperature T ro In this embodiment, the formula (1) corresponds to the furnace temperature prediction model of the present invention.

[0027] In applying the present invention, the furnace battery temperature prediction model does not necessarily have to be in this form, and for example, the way in which the amount of change is given may be changed, or the absolute value of the dependent variable may be predicted directly. Also, the model construction method does not necessarily have to be the linear form (linear regression model) of equation (1), and may be, for example, a physical model (a model including a differential equation (or an equation obtained by discretizing the differential equation) that represents a physical phenomenon), or a prediction model based on machine learning other than equation (1).

[0028] Next, a coke temperature prediction model used in the target furnace temperature calculation unit 102 will be described. Here, the time from the start of the loading operation in one coking chamber 2 to the end of the unloading operation is called a carbonization cycle. In this embodiment, one cycle in the carbonization cycle is the sum of the time for each of the five consecutive unloading and loading operations. The coke temperature prediction model is a model that predicts the coke temperature based on the first influencing factor including the oven battery temperature. In this embodiment, a regression model based on the physical phenomenon of a single coking chamber 2 is adopted, as shown in equation (2). T co is the coke temperature [℃], T ro is the furnace temperature [℃], t t is the time [hr], t krepresents the carbonization time [hr], S represents the amount of coal charged [tons], and W represents the coal moisture content [%]. Furthermore, nj (n-9 to n shown in FIG. 4) is a variable specifying a period obtained by dividing the carbonization cycle by the number of passes, and each period is the same length as the pass time. In this embodiment, since coal is discharged from the kiln and charged at intervals of five kilns, the period from charging coal in one coke chamber 2 to extrusion is divided into five pass times (five periods from n to n-4 and five periods from n-5 to n-9). When the values ​​of the two variables nj are consecutive, this indicates that the periods specified by the two variables are consecutive periods (for example, period n and period n-1 are consecutive periods). For consecutive periods n and n-1, the kiln loading sequence is 1, 3, 5, 2, and 4, and the kiln loading sequence for sequence 1 is carried out in period n. To explain this more specifically, period n is the period in which the kiln loading sequence for sequence 1 is carried out, and is the same length of time as sequence 1. Furthermore, period n-1 is the period in which the kiln loading sequence for sequence 4 is carried out, and is the same length of time as sequence 4. Hereinafter, the sequence in which the kiln loading sequence for sequence nj is carried out will be referred to as the sequence corresponding to period nj.

[0029]

number

[0030] ΔT, which is the objective variable in equation (2), co (n), ΔT as explanatory variable ro (n), Δt t (n), Δt k (n), ΔS(n), and ΔW(n) are the coke temperature T co , furnace temperature T ro , street time t t , dry distillation time t kIt is expressed as the change in the coal loading amount S and coal moisture content W from the previous cycle (five cycles ago) in the carbonization cycle of the same carbonization chamber. Figure 4 shows the change in coke temperature and furnace battery temperature from loading to extrusion in one carbonization chamber. Figure 4 shows two cycles (two carbonization cycles) in the carbonization cycle from loading to extrusion in one carbonization chamber.

[0031] Table 2 shows the relationship between explanatory variables (○) and objective variables (★) of the coke temperature prediction model used in this embodiment. Using operational data during carbonization as explanatory variables, the coke temperature of each coke chamber, which is the objective variable, is predicted using the coke temperature prediction model. In detail, the values ​​for each pass from coal loading to extrusion are used for the furnace battery temperature and pass time. The values ​​for each carbonization cycle, given at the time of the first coal loading in the carbonization cycle, are used for the carbonization time, coal loading amount, and coal moisture content.

[0032] [Table 2]

[0033] Specific details of the dependent variable and explanatory variables in the coke temperature prediction model will be described below with reference to FIG. 4. In FIG. 4, the coke temperature at a street corresponding to period n is predicted at the time when the last street immediately before the street corresponding to period n (the last unloading operation) ends, or at the time when the first street corresponding to period n (the first loading operation) starts. In this case, the periods n-9 to n-1 before period n are periods in which coke temperature predictions have already been completed. For example, the coke temperature prediction performed immediately before period n is a prediction of the coke temperature at a street corresponding to period n-1. In this case, in the following description, the periods n-1 to n-9 are respectively referred to as periods n-2 to n-10, and the coke temperature at a street corresponding to period n-1 is predicted. Note that the start times of the unloading operation and the loading operation may be, for example, the start times of the processes specified as the first steps of each operation in the coke plant operation manual. Similarly, the end of the unloading operation and the charging operation may be the time when the final process specified in the operation manual of the coke plant is completed. In addition, in Fig. 4, the period from n-4 to n, including the period n, is called the current carbonization cycle, and the carbonization cycle immediately before the current carbonization cycle is called the period from n-9 to n-5, and is called the previous carbonization cycle.

[0034] <Objective variable> ΔT co (n): The change in the coke temperature in the current carbonization cycle of a certain carbonization chamber 2 from the coke temperature in the previous carbonization cycle of the carbonization chamber 2 (ΔT co (n)=T co (n)-T co (n-5) As mentioned above, the coke temperature cannot be obtained without the coke oven unloading process, so it is calculated using the carbonization cycle. The "5" in n-5 is the number of cycles and changes depending on the number of cycles (this also applies to explanatory variables).

[0035] <Explanatory variables> ΔT ro (nj):ΔTro (n - j) is calculated differently when predicting the coke temperature in the passage corresponding to the n period and when predicting the coke temperature in the passage corresponding to the n + 1 period. <<When predicting the coke temperature in the passage corresponding to the n period>> When j = 0: (Target hearth temperature in the (n - j) period) - (Actual hearth temperature in the (n - j - 5) period) (ΔT ro (n - j) = T ro (n - j) - T ro (n - j - 5)) When j = 1, 2, ··· (when j ≥ 1): (Actual hearth temperature in the (n - j) period) - (Actual hearth temperature in the (n - j - 5) period) (ΔT ro (n - j) = T ro (n - j) - T ro (n - j - 5)).

[0036] <<When predicting the coke temperature in the passage corresponding to the n + 1 period>> When j = 0, 1 (0 ≤ j ≤ 1): (Target hearth temperature in the (n - j) period) - (Actual hearth temperature in the (n - j - 5) period) (ΔT ro (n + 1 - j) = T ro (n + 1 - j) - T ro (n + 1 - j - 5)) When j = 2, 3, ··· (when j ≥ 2): (Actual hearth temperature in the (n + 1 - j) period) - (Actual hearth temperature in the (n + 1 - j - 5) period) (ΔT ro (n + 1 - j) = T ro (n + 1 - j) - T ro (n + 1 - j - 5)).

[0037] Note that since the hearth temperature is obtained every hour, the hearth temperature here is the representative value (the average value for each pass time in this embodiment) for each pass time. Also, the "5" in n-j-5 is the number of passes and is changed according to the number of passes (this also applies to other explanatory variables). Further, the period of n-j-5 is the period belonging to the previous carbonization cycle and is the period one cycle before in the carbonization cycle of the period of n-j belonging to the current carbonization cycle. When actual values are obtained during the period of n-j, actual values are used; when actual values are not obtained, target values are used.

[0038] Δt t (n-j):Δt t (n-j)'s calculation method is also ΔT ro (n-j) is different in the case of predicting the coke temperature in the pass corresponding to the period of n and the case of predicting the coke temperature in the pass corresponding to the period of n + 1, similar to n. <<When predicting the coke temperature in the pass corresponding to the period of n>> When j = 0: (planned pass time in the period of n-j)-(actual pass time in the period of n-j-5)(Δt t (n-j)=t t (n-j)-t t (n-j-5)) When j = 1, 2, ··· (when j≧1): (actual pass time in the period of n-j)-(actual pass time in the period of n-j-5)(Δt t (n-j)=t t (n-j)-t t (n-j-5)).

[0039] <<When predicting the coke temperature in the pass corresponding to the period of n + 1>> When j = 0, 1 (0≦j≦1): (planned pass time in the period of n-j)-(actual pass time in the period of n-j-5)(Δt t (n + 1-j)=t t (n + 1-j)-t t (n + 1-j-5)) For j=2, 3, ... (j≧2): (Accurate time in the period n+1-j) - (Accurate time in the period n+1-j-5) (Δt t (n+1-j)=t t (n+1-j)-t t (n+1-j-5)) ΔT ro As with (nj), Δt t In (nj), if an actual value is available for the period nj, the actual value is used, and if an actual value is not available, the planned value is used.

[0040] Δt k (n): Planned carbonization time in the current carbonization cycle of a certain carbonization chamber 2 - Actual carbonization time in the previous carbonization cycle of the carbonization chamber 2 (Δt k =t k (n)-t k (n-5) ΔS(n): Amount of coal charged in the current carbonization cycle of a certain carbonization chamber 2 - Actual amount of coal charged in the previous carbonization cycle of the carbonization chamber 2 (ΔS = S(n) - S(n-5)) ΔW: Coal moisture content in the current carbonization cycle of a certain carbonization chamber 2 - Actual coal moisture content in the previous carbonization cycle of the carbonization chamber 2 (ΔW = W(n) - W(n-5))

[0041] In equation (2), the coke temperature T co When calculating (predicting) (n), the actual value is used, otherwise the scheduled value is used. In addition, since the amount of coal charged and the amount of coal moisture in this carbonization cycle are obtained as in the last carbonization cycle (corresponding to the period n-5) in the previous carbonization cycle, a circle is marked in the n-5 column in Table 2. Furthermore, when applying the present invention, the coke temperature prediction model does not necessarily have to be in the form described above. For example, the way in which the amount of change is given may be changed, or the absolute value of the dependent variable may be predicted directly. Furthermore, the model construction method does not necessarily have to be the linear form of Equation (2). For example, a machine learning prediction model or a physical model other than Equation (2) may also be used. In this embodiment, Equation (2) corresponds to the physical quantity prediction model referred to in the present invention.

[0042] Coefficient a in Equation (2) j , b j , c, d, and e are explanatory variables, respectively. ro (nj), Δt t (nj), Δt k (n), ΔS(n), and ΔW(n). Coefficient a i , b i , c i , d i The coefficients that best fit the form of equation (2) to the past operation results of the coke oven 1 are separately obtained. For example, a set of T ro (nj), Δt t (nj), Δt k (n), ΔS(n), and ΔW(n) are used as training data to create a large number of training data, and multiple regression analysis is performed using the training data to obtain the coefficient a j , b j , c, d, and e.

[0043] Next, the processing of the target furnace temperature calculation unit 102 in the first embodiment will be described with reference to FIGS. Fig. 5 is a flowchart showing the processing of the target furnace temperature calculation unit 102. The flowchart in Fig. 5 is executed at each run time, such as when a new run time arrives. The timing to start the flowchart in Fig. 5 is the timing when the last unloading operation in each run is completed, as explained with reference to Fig. 4. However, the timing to start the flowchart in Fig. 5 may also be the timing when the first loading operation in each run begins.

[0044] In step S501, the target furnace temperature calculation unit 102 inputs data on influencing factors affecting the coke temperature for multiple time periods from the present to the past (hereinafter referred to as carbonization information data) for each coke chamber 2 via the input unit 101. This carbonization information data (time series of the above-mentioned influencing factors) may be composed of actual values ​​only, or may include schedule values ​​if necessary, such as when actual values ​​are insufficient. The carbonization information data is data that represents the coal characteristics at the time of loading, the state of the coke at the time of extrusion, the furnace core temperature during carbonization, etc., and specifically, the carbonization information data includes, for example, the following data: Coke temperature, carbonization time ·Coal loading amount, coal moisture content Furnace temperature, passage time In this embodiment, the coke temperature and carbonization time are obtained at or after the end of the coke unloading operation in each coke oven chamber 2 (as described above, the coke temperature is measured when the coke is pushed out, but the final coke temperature is obtained at or after the end of the unloading operation). The coal loading amount is obtained at or after the end of the coal loading operation in each coke oven chamber 2. The coal moisture content is obtained before the start of coal carbonization (before the start of the coal loading operation (e.g., before being transported to the coke oven 1)). As described above, the oven battery temperature is the average temperature per pass time of all combustion chambers in which thermometers 6 are installed. In addition, in the carbonization information data, actual values ​​are used for the coke temperature, coal loading amount, coal moisture content, and oven battery temperature. However, for example, if the schedule values ​​are highly reliable, scheduled values ​​may be used instead of these actual values. In addition, in the carbonization information data, both actual values ​​and scheduled values ​​are used for the carbonization time and pass time.

[0045] In step S502, the target furnace temperature calculation unit 102 generates one or more initial values ​​of the target furnace battery temperature pattern, which serve as candidates for the progression of the target furnace battery temperature over multiple consecutive future run times (hereinafter referred to as the target furnace battery temperature pattern) within predetermined constraints. The first run time of the multiple consecutive future run times is preferably the run time following the current run time in which coal loading and unloading are being performed. This is because the coke oven 1 has a large time constant (the time delay between when the input heat amount is changed and when that change affects the coke temperature). However, the first run time of the multiple consecutive future run times may be the run time two or more times after the current run time. Examples of the constraints include upper and lower limits for the furnace battery temperature and upper and lower limits for the amount of change in the furnace battery temperature. Figure 6(a) shows an example of a target furnace battery temperature pattern 601. In Figures 6(a) and 6(b), "control start" indicates the timing when control by the control device 100 begins. In Figures 6(a) and 6(b), the time indicated as "control start" is the current time. The target reactor battery temperature pattern 601 is set to a value corresponding to a plurality of future passage times. t The target furnace temperature calculation unit 102 represents the target furnace temperature that changes stepwise for each passing time t t The target furnace battery temperature for each furnace is determined by random numbers within the constraint range, and a predetermined number of (for example, 50 here) initial values ​​of target furnace battery temperature patterns are generated. Note that the initial values ​​of the predetermined number of target furnace battery temperature patterns are common to all coking chambers 2.

[0046] In step S503, the target furnace temperature calculation unit 102 calculates the right side of the coke temperature prediction model of equation (2) for each coke chamber 2 using a candidate target furnace batter temperature pattern (the initial value of the target furnace batter temperature pattern generated in step S502 or the target furnace batter temperature pattern generated in step S507 described below) and the carbonization information data (carbonization time, coal loading amount, coal moisture content, furnace batter temperature, and pass time) acquired in step S501, calculates the value of the left side of the coke temperature prediction model of equation (2), and predicts the future coke temperature of each coke chamber 2 from the calculated value. The future coke temperature of each coke chamber 2 is calculated within the range of the period indicated by the target furnace batter temperature pattern 601 (plurality of consecutive future pass times).

[0047] In step S504, the target furnace temperature calculation unit 102 converts the future coke temperature of each coke chamber 2 predicted in step S503 into a run average value, which is an average value for each run (hereinafter referred to as a predicted coke temperature value 602). Figure 6(b) shows an overview of the process for obtaining the predicted coke temperature value 602. t Since extrusion is performed in one of the ways every time, the target furnace temperature calculation unit 102 calculates the extrusion time t t For each run, a predicted value 602 of the coke temperature on the street where the extrusion is performed is calculated. Specifically, the predicted value 602 of the coke temperature is calculated by adding up the coke temperatures of each coke chamber 2 predicted in step S503 for multiple (13 in the example of FIG. 2 ) coke chambers 2 belonging to the street where the extrusion is performed, and dividing the sum by the number of coke chambers 2 belonging to the street. Note that, although an example has been described in which the average value for each run is used as a representative value of the future coke temperature for each run to express the predicted value 602 of the coke temperature, it is also possible to use, for example, the minimum value of the future coke temperature for each run. In this way, using the minimum value as a representative value of the future coke temperature for each run can more reliably prevent the coke temperature from becoming too low, resulting in so-called undercooked coke.

[0048] In step S505, the target oven temperature calculation unit 102 calculates the evaluation function J1 of Equation (3). The evaluation function J1 includes a term (the first term on the right side) that represents the difference between the predicted coke temperature 602, which is a representative value of the future coke temperature for each run, calculated in step S504, and the target coke temperature 603 for that run (see FIG. 6(b)). The second term on the right side is a term that ensures that the coke temperature satisfies a preset lower limit constraint. Because the coke temperature affects the quality of the coke, this term ensures that the coke temperature is at a minimum temperature necessary to prevent so-called underburning of the coke. The third term on the right side is a term that prevents the amount of change in the target oven battery temperature, i.e., the step-like change in the target oven battery temperature pattern 601 in FIG. 6(a), from becoming too large. This is because a sudden change in the target oven battery temperature is undesirable from the standpoint of oven operation stability, etc. The fourth term on the right-hand side is a term provided to prevent the target furnace battery temperature from becoming too high. This is because it is undesirable for the target furnace battery temperature to become too high in terms of operation stability, cost, etc. In this embodiment, the evaluation function J1 corresponds to the first evaluation function of the present invention. J1 = (target coke temperature - predicted coke temperature) + (lower limit of coke temperature) + (change in target battery temperature) + (target battery temperature) (3) In addition, to represent step S505, an example has been described in which the evaluation function is calculated using a representative value of the future coke temperature for each run. However, it is also possible to calculate the first term on the right side of the evaluation function J1 using, for example, the difference between the future coke temperature of each coke chamber 2 predicted in step S503 and the target coke temperature for each coke chamber 2.

[0049] In step S506, the target furnace temperature calculation unit 102 determines whether or not a calculation termination condition has been reached. For example, the calculation termination condition may be that a predetermined number of repeated calculations has been reached. Alternatively, the calculation termination condition may be that the value of the evaluation function J1 has converged during the repeated calculations. If it is determined that the calculation termination condition has not been reached, the process proceeds to step S507. If it is determined that the calculation termination condition has been reached, the process proceeds to step S508.

[0050] In step S507, the target furnace temperature calculation unit 102 generates one or more new target furnace battery temperature patterns as candidates for the target furnace battery temperature pattern, and then returns to step S503. It is preferable that the number of initial values ​​of the target furnace battery temperature pattern generated in step S502 is the same as the number of new target furnace battery temperature patterns generated in step S507. For example, a target furnace battery temperature pattern that reduces the evaluation function J1 is searched for based on the metaheuristic algorithm FPA (Flower Pollination Algorithm) (see, for example, Non-Patent Document 1), and 50 new target furnace battery temperature patterns are generated. While FPA is used as an example of the optimization method (algorithm for solving the optimization problem), other optimization methods such as GA (Genetic Algorithm) and PSO (Particle Swarm Optimization) may also be used. In this way, a new target furnace battery temperature pattern is generated in step S507, and the processes of steps S503 to S505 are repeated until the calculation termination condition is met in step S506.

[0051] In step S508, the target furnace temperature calculation unit 102 outputs the target furnace battery temperature pattern that minimizes the evaluation function J1 from among the candidate target furnace battery temperature patterns to the input heat amount calculation unit 103. For example, if the evaluation function J1 is obtained by multiplying each term on the right side of equation (3) by (-1), the target furnace temperature calculation unit 102 will search for the target furnace battery temperature pattern that maximizes the evaluation function J1.

[0052] As described above, the coke temperature when the furnace battery temperature is changed is predicted, and an optimal furnace battery temperature pattern that satisfies the target (minimizes the evaluation function J1) is determined. As a result, a target furnace battery temperature pattern 601 can be calculated so that the predicted coke temperature value 602 corresponds to the target coke temperature 603, as shown in Figure 8(a).

[0053] Next, the processing of the input heat amount calculation unit 103 in the first embodiment will be described with reference to FIG. Fig. 7 is a flowchart showing the processing of the input heat amount calculation unit 103. The flowchart in Fig. 7 is executed in a control cycle (here, for example, one hour cycle) of the input heat amount to the coke oven 1. In this way, the cycle for calculating the input heat amount shown in Fig. 7 is shorter than the cycle for calculating the target furnace battery temperature pattern shown in Fig. 5. In step S701, the input heat amount calculation unit 103 inputs data (hereinafter referred to as combustion information data) of influencing factors that affect the oven battery temperature for multiple consecutive periods of one hour from the present to the past for each coke chamber 2 via the input unit 101 (in the example shown in Table 1, each period is one hour). This combustion information data (time series of the above-mentioned influencing factors) may consist of only actual values, or may include scheduled values ​​if necessary, such as when actual values ​​are insufficient. The combustion information data is data that represents the amount of heat input to the coke oven and the characteristics of the coal at the time of loading, and specifically includes, for example, the oven battery temperature, the input heat amount, the total amount of coal loaded in all coke chambers 2, and the coal moisture content.

[0054] In step S702, the heat input calculation unit 103 utilizes the fact that the battery temperature prediction model can be expressed as a linear equation as in equation (1) to calculate the hourly heat input amount that minimizes the evaluation function J2 in equation (4) using generalized predictive control (GPC) (see, for example, Non-Patent Document 2). The combustion information data acquired in step S701 is used as input data for the battery temperature prediction model. The target value for the GPC is the target battery temperature, which is represented by the target battery temperature pattern calculated by the target furnace temperature calculation unit 102 as described in Figure 5. The first term on the right side of the evaluation function J2 represents the difference between the battery temperature predicted by the battery temperature prediction model (hereinafter referred to as the predicted value of the battery temperature) and the target battery temperature, which is represented by the target furnace temperature pattern calculated by the target furnace temperature calculation unit 102 as described in Figure 5. This term makes it possible to calculate the heat input amount that realizes the target battery temperature pattern calculated by the target furnace temperature calculation unit 102. The second term on the right side is a term provided to prevent the amount of change in the input heat quantity controlled by operating the regulating valve 5, in this case, from becoming too large between adjacent time periods. A sudden change in the input heat quantity means that the regulating valve 5 needs to be operated to a large extent, which is undesirable from the standpoint of stability and feasibility of furnace operation. In this embodiment, the evaluation function J2 corresponds to the second evaluation function of the present invention. J2 = (target furnace battery temperature - predicted furnace battery temperature) 2 + (Change in input heat) 2 ···(4) In GPC, the furnace battery temperature prediction model in equation (1) and the evaluation function J2 in equation (4) are written in vector form, and the vector-format furnace battery temperature prediction model is substituted into the vector-format evaluation function. The evaluation function obtained in this way is partially differentiated with respect to the input heat amount to determine the input heat amount that minimizes the evaluation function J2 in equation (4).

[0055] In step S703, the input heat amount calculation unit 103 outputs the input heat amount that minimizes the evaluation function J2 to the input heat amount setting unit 104. In response to this, the input heat amount setting unit 104 outputs the input heat amount calculated by the input heat amount calculation unit 103 to a control device of an actuator (not shown) so that the input heat amount is reflected in the coke production process. The control device of the actuator operates the adjustment valve 5 (see FIG. 2) via an actuator (not shown) to set the aperture of the adjustment valve 5 to an aperture corresponding to the input heat amount calculated by the input heat amount calculation unit 103. Note that, for example, if the evaluation function J2 is obtained by multiplying each term on the right side of equation (4) by (−1), the input heat amount calculation unit 103 will determine the input heat amount that maximizes the evaluation function J2.

[0056] As described above, the furnace battery temperature when the input heat amount is changed is predicted, and the optimal input heat amount that satisfies the target (minimizes the evaluation function J2) is determined. As a result, as shown in Figure 8(b), the input heat amount 802 per hour can be calculated so that the predicted value 801 of the furnace battery temperature becomes a value corresponding to the target furnace battery temperature pattern 601 calculated by the target furnace temperature calculation unit 102. In other words, the input heat amount 802 that realizes the target furnace battery temperature pattern 601 shown in Figure 8(a) can be calculated.

[0057] In this embodiment, the input heat quantity calculation unit 103 uses GPC to control the input heat quantity, but this is not limited to this. For example, similar to the flowchart in Figure 5, it is also possible to provide candidates for input heat quantities and search for the input heat quantity that minimizes the evaluation function J2. Alternatively, it is also possible to perform PID control to change the input heat quantity so that the actual furnace battery temperature becomes a value corresponding to the target furnace battery temperature pattern 601 calculated by the target furnace temperature calculation unit 102.

[0058] As described above, when controlling the heat input amount so that the coke temperature corresponds to the target temperature, it is possible to dynamically calculate the target furnace battery temperature (target furnace battery temperature pattern) for multiple future runs, taking into account how the coke temperature for each run will change due to the first influencing factor, including the furnace battery temperature. This makes it possible to achieve target tracking of the coke temperature and suppress variation, which is expected to have effects such as reduced production costs (e.g., reducing the amount of excess carbonization heat due to overcarbonization), stabilizing production (e.g., avoiding the risk of clogging during extrusion due to undercarbonization or overcarbonization), and stabilizing quality (suppressing quality variation due to variation in the carbonization state).

[0059] In the above-described embodiment, an example has been described in which the input heat quantity is controlled so that the coke temperature becomes a value corresponding to a predetermined target temperature, but the present invention is not limited to this. The present invention can be applied to any system that controls the input heat quantity so that a physical quantity representing the coke carbonization state becomes a value corresponding to a predetermined target value. In addition to the coke temperature, other physical quantities that represent the coke carbonization state include, for example, the temperature of the oven wall 4. The coke carbonization state indicates the degree to which coal has been carbonized (pyrolyzed) in the produced coke, and is an index of coke quality.

[0060] [Second embodiment] Next, a second embodiment will be described with reference to Fig. 9. In the second embodiment, a modified example of the coke temperature prediction model will be described. The coke temperature prediction model described in the first embodiment is a regression model based on the physical phenomena of a single coke chamber 2, but it may not be sufficient to just focus on the physical phenomena of a single coke chamber 2. This is because just focusing on the physical phenomena of a single coke chamber 2 makes it impossible to capture the trend of temperature changes in adjacent coke chambers 2 or the trend of changes in operating conditions. Also, it makes it impossible to capture nonlinearities due to changes in the thermophysical properties of coal or the way in which each type of furnace battery temperature is affected.

[0061] Therefore, as shown in FIG. 9, the models used in the target furnace temperature calculation unit 102 include a regression model 901 that predicts the coke temperature and a big data model 902 that estimates the prediction error of the regression model 901 and outputs the estimated error. The regression model 901 is expressed as, for example, equation (2), and receives operation data as input and outputs a predicted value of the coke temperature in each coke chamber. The big data model 902 is an estimation model based on machine learning, and in this embodiment, it is constructed as a nonlinear model configured using a big data machine learning technique. The big data model 902 receives operation data and the predicted value of the regression model 901 as input, estimates the prediction error of the regression model 901, and outputs the estimated error. The estimated error is multiplied by a predetermined gain K, and the result is added to the predicted value of the regression model 901 to correct the predicted value. The big data machine learning technique uses a gradient boosting decision tree (see Patent Document 2 and Non-Patent Document 8 cited in Patent Document 2). Table 3 shows the relationship between the explanatory variables (○) and the objective variable (★) of the big data model used in this embodiment. In detail, the operation data for the next set (including schedule values) and the operation data for the last five sets of data are used as explanatory variables.

[0062] [Table 3]

[0063] As described above, the coke temperature prediction model of this embodiment has a hybrid model configuration in which the prediction error of the regression model 901 is estimated and corrected using a big data machine learning method. This makes it possible to capture nonlinearities due to the tendency of temperature changes in adjacent coke chambers, the tendency of changes in operating conditions, changes in the thermophysical properties of coal, and the way in which each type of furnace battery temperature is affected. Although the example shows the big data model 902 being constructed using a gradient boosting decision tree, the big data model 902 may also be constructed using a machine learning method such as a neural network or deep learning.

[0064] [Example 1] Using past performance data, an offline simulation of the control method according to the second embodiment (hereinafter referred to as the present control method in the description of Example 1) was carried out. Figures 10(a) to (c) show the results of offline simulations of coke temperature, furnace battery temperature, and input heat amount. The dotted lines in the figures show the results of this control method. In addition, as a comparative example (hereinafter referred to as the operator operation method), the solid lines show the results of operation by operating the control valve 5 at the discretion of the operator. In the case of operator operation, it is difficult to predict the future coke temperature, and as a result, the coke temperature is affected by operational changes and disturbances, and as shown in Figure 10(a), variations occur in the coke temperature in all of the verification sections A to C. In contrast, this control method allows for proactive action that predicts the future compared to the operator operation method in all verification sections A to C (Fig. 10(b) and (c)). As shown in Fig. 10(a), the coke temperature can be set to a value corresponding to the target temperature, and variation in the coke temperature can be suppressed. This is expected to have the effects of reducing production costs, stabilizing production, and stabilizing quality.

[0065] [Third embodiment] The third embodiment is an example in which, when the target furnace temperature calculation unit 102 calculates the target furnace battery temperature (target furnace battery temperature pattern) for multiple future passage times, the feasibility of the target furnace battery temperature pattern from the perspective of controlling the input heat amount is taken into consideration. If it is necessary to increase the amount of heat input change to achieve the target furnace battery temperature pattern, the control valve 5 must be operated to a greater extent, which is undesirable from the standpoint of furnace operation stability and feasibility, and this target furnace battery temperature pattern cannot be said to be highly feasible. If the target furnace battery temperature pattern is not highly feasible, there is a concern that it will not be possible to accurately set the coke temperature to a value corresponding to the target temperature and to suppress coke temperature variation. Therefore, in the third embodiment, when the target furnace temperature calculation unit 102 calculates the target furnace battery temperature pattern, the feasibility of the target furnace battery temperature pattern is improved by providing the coke temperature prediction model with a furnace battery temperature that satisfies a predetermined evaluation of the change in input heat amount.

[0066] The third embodiment will be described below, with the differences from the first embodiment being mainly described, and descriptions of the same aspects as the first embodiment being omitted. In the first embodiment, the target furnace temperature calculation unit 102 is configured to calculate the target furnace battery temperature pattern independently and pass it to the input heat amount calculation unit 103. In contrast, in the third embodiment, the target furnace temperature calculation unit 102 is configured to calculate the target furnace battery temperature pattern in cooperation with the input heat amount calculation unit 103 and pass it to the input heat amount calculation unit 103. The functional configuration of the control device 100 according to this embodiment is different from the functional configuration shown in FIG. 1 in that the single-arrow line from the target furnace temperature calculation unit 102 to the input heat amount calculation unit 103 is changed to a double-arrow line connecting the target furnace temperature calculation unit 102 and the input heat amount calculation unit 103 (see the functional configurations shown in FIGS. 15 and 22 of the fourth and sixth embodiments described below, which are based on the third embodiment). Furthermore, in the third embodiment, some of the functions of the target furnace temperature calculation unit 102 and the input heat amount calculation unit 103 are different from those of the first embodiment. Therefore, the functional configuration of the control device 100 according to this embodiment will not be illustrated, and detailed explanations of the input unit 101, target furnace temperature calculation unit 102, input heat amount calculation unit 103, and input heat amount setting unit 104, which are the same as those of the control device 100 according to the first embodiment, will be omitted.

[0067] The process of calculating the target reactor battery temperature pattern in the third embodiment will be described with reference to FIGS. FIG. 11 is a flowchart showing the processing of the target furnace temperature calculation unit 102 and the input heat amount calculation unit 103. The flowchart of FIG. 11 is executed in a run time cycle. The timing to start the flowchart of FIG. 11 is, for example, the timing when the unloading operation for each run is completed, similar to the timing to start the flowchart of FIG. 5, but is not necessarily limited to this timing, and may be, for example, the timing when the coal loading operation for each run is started. Also, FIG. 12 is a diagram for explaining an outline of the processing in the flowchart of FIG. 11. The upper part of FIG. 12 shows an example of a target temperature 603 for the coke temperature, similar to the upper part of FIG. 6(a). In step S1101, the target furnace temperature calculation unit 102 takes in carbonization information data for a plurality of times from the present to the past via the input unit 101. Step S1101 is the same process as step S501 in FIG.

[0068] In step S1102, the input heat amount calculation unit 103 acquires combustion information data for a plurality of consecutive periods of one hour from the present to the past via the input unit 101. Step S1102 is the same process as step S701 in FIG.

[0069] In step S1103, the target furnace temperature calculation unit 102 generates one or more initial values ​​of target furnace battery temperature patterns that are candidates for the target furnace battery temperature pattern. Step S1103 is the same process as step S502 in Figure 5, and an example of a target furnace battery temperature pattern 601 is shown in the middle part of Figure 12. The target furnace temperature calculation unit 102 passes the generated initial values ​​of the target furnace battery temperature pattern to the input heat amount calculation unit 103.

[0070] In step S1104, the heat input calculation unit 103 utilizes the fact that the battery temperature prediction model can be expressed as a linear equation as in equation (1) to calculate the heat input amount that minimizes the evaluation function J2 in equation (4) using generalized predictive control (GPC). The combustion information data acquired in step S1102 is used as input data for the battery temperature prediction model. In addition, the target battery temperature expressed as a candidate target battery temperature pattern (the initial value of the target battery temperature pattern generated in step S1103 or the target battery temperature pattern generated in step S1111, described below) is used as the target value for GPC. The first term on the right-hand side of the evaluation function J2 represents the difference between the battery temperature predicted by the battery temperature prediction model (predicted value of the battery temperature) and the target battery temperature expressed as a candidate target battery temperature pattern. As described above, the cycle of calculating the input heat amount in input heat amount calculation unit 103 is executed in one-hour cycles, and therefore, as shown in the lower part of Fig. 12, it is possible to calculate input heat amount 1201 that minimizes evaluation function J2 of equation (4) every hour. The width of each bar in the lower part of Fig. 12 represents one hour, and the length represents the input heat amount. As described in the first embodiment, an example will be described in which GPC is used in input heat amount calculation unit 103, but the present invention is not limited to this.

[0071] In step S1105, the input heat amount calculation unit 103 uses the input heat amount calculated in step S1104 and the combustion information data acquired in step S1102 to calculate the right side of the furnace battery temperature prediction model in equation (1), calculates the value of the left side of the furnace battery temperature prediction model in equation (1), and calculates the predicted value of the furnace battery temperature from the calculated value. As a result, as shown in the middle part of Figure 12, it is possible to calculate the predicted value of the furnace battery temperature per hour (hereinafter referred to as the furnace battery temperature control waveform) 1202 as the furnace battery temperature corresponding to the hourly input heat amount 1201 calculated by the input heat amount calculation unit 103. The input heat amount calculation unit 103 passes the furnace battery temperature control waveform 1202 to the target furnace temperature calculation unit 102. The predicted value of the furnace battery temperature is calculated within the range of the period (multiple consecutive future time periods) indicated by the target furnace battery temperature pattern 601.

[0072] In step S1106, the target furnace temperature calculation unit 102 calculates the control waveform 1202 of the furnace battery temperature calculated in step S1105 at the time tt Since the target furnace temperature calculation unit 102 executes processing in a pass time cycle, the parameters handled here must be in pass time units. Therefore, the furnace battery temperature control waveform 1202, which is expressed as the furnace battery temperature every hour, is converted into a target furnace battery temperature pattern that changes stepwise in pass time units, like the target furnace battery temperature pattern 601. Specifically, t For each cycle, the average value of the furnace battery temperature represented by the furnace battery temperature control waveform 1202 is calculated, and each average value is used as the target furnace battery temperature for each cycle, thereby converting it into a target furnace battery temperature pattern.

[0073] In step S1107, the target furnace temperature calculation unit 102 uses the target furnace battery temperature pattern converted in step S1106 and the carbonization information data imported in step S1101 to calculate the right side of the coke temperature prediction model in equation (2), calculates the value of the left side of the coke temperature prediction model in equation (2), and predicts the future coke temperature of each coke chamber 2 from the calculated value. The future coke temperature of each coke chamber 2 is calculated within the range of the period (plurality of consecutive future passage times) indicated by the target furnace battery temperature pattern 601.

[0074] In step S1108, the target furnace temperature calculation unit 102 converts the future coke temperature of each coke chamber 2 predicted in step S1107 into a run average value (predicted value of coke temperature), which is the average value for each run. Step S1108 is the same processing as step S504 in Figure 5 (see Figure 6(b)).

[0075] In step S1109, the target furnace temperature calculation unit 102 calculates the evaluation function J1' of equation (5). The evaluation function J1' has the same first to third terms on the right-hand side as the evaluation function J1 described in the first embodiment. The fourth term on the right-hand side is a term provided to prevent the input heat amount from becoming too large. This is because an excessively large input heat amount is undesirable in terms of furnace operation stability, cost, etc. The fifth term on the right-hand side is a term provided to set the furnace battery temperature to a value corresponding to the target furnace battery temperature represented by the target furnace battery temperature pattern converted in step S1106. In this embodiment, the evaluation function J1' corresponds to the first evaluation function defined in the present invention. J1' = (target coke temperature - predicted coke temperature) + (lower limit of coke temperature) + (change in target battery temperature) + (magnitude of input heat) + (target battery temperature - predicted battery temperature) (5)

[0076] In step S1110, the target furnace temperature calculation unit 102 determines whether or not the calculation termination condition has been reached. Step S1110 is the same process as step S506 in Fig. 5. If it is determined that the calculation termination condition has not been reached, the process proceeds to step S1111. If it is determined that the calculation termination condition has been reached, the process proceeds to step S1112.

[0077] In step S1111, the target furnace temperature calculation unit 102 generates one or more new target furnace battery temperature patterns as candidates for the target furnace battery temperature pattern, and then returns to step S1104. Step S1111 is the same process as step S507 in Figure 5. The target furnace temperature calculation unit 102 passes the newly generated target furnace battery temperature pattern to the input heat amount calculation unit 103, and in response, the input heat amount calculation unit 103 executes the processes of steps S1104 and S1105. In this way, a new target furnace battery temperature pattern is generated in step S1111, and the processes of steps S1104 to S1109 are repeated until the calculation termination condition is reached in step S1110.

[0078] In step S1112, the target furnace temperature calculation unit 102 outputs the target furnace battery temperature pattern that minimizes the evaluation function J1' from among the target furnace battery temperature patterns converted in step S1106 to the input heat calculation unit 103. For example, if the evaluation function J1' is calculated by multiplying each term on the right side of equation (5) by (-1), the target furnace temperature calculation unit 102 will search for the target furnace battery temperature pattern that maximizes the evaluation function J1'.

[0079] As described above, the coke temperature when the furnace battery temperature is changed is predicted, and an optimal furnace battery temperature pattern that satisfies the target (minimizes the evaluation function J1') is determined. As a result, as shown in FIG. 8(a), a target furnace battery temperature pattern 601 can be calculated so that the predicted coke temperature value 602 corresponds to the target coke temperature 603. At this time, a predetermined evaluation of the change in input heat quantity is used, which in this embodiment is the evaluation that minimizes the evaluation function J2 of equation (4) as described above. The furnace battery temperature that satisfies the predetermined evaluation of the change in input heat quantity can be obtained, for example, as follows. First, a candidate target furnace battery temperature pattern is given the evaluation function J2 of equation (4), and the input heat quantity 1201 that minimizes the evaluation function J2 of equation (4) is calculated. Then, the input heat quantity 1201 calculated in this way is input into the furnace battery temperature prediction model shown in equation (1) to calculate the control waveform 1202 of the furnace battery temperature corresponding to the input heat quantity 1201. In this way, when calculating the target furnace battery temperature pattern 601, the furnace battery temperature control waveform 1202 corresponding to the input heat amount 1201 is used as the furnace battery temperature when the evaluation of minimizing the evaluation function J2 of equation (4) is satisfied, thereby improving the feasibility of the target furnace battery temperature pattern. In this embodiment, the input heat amount 1201 that minimizes the evaluation function J2 of equation (4) corresponds to the input heat amount that satisfies the predetermined evaluation of the amount of change in the input heat amount, as referred to in this invention. Also, in this embodiment, the furnace battery temperature control waveform 1202 corresponding to the input heat amount 1201 corresponds to the furnace battery temperature when the predetermined evaluation of the amount of change in the input heat amount is satisfied, as referred to in this invention. Also, in this embodiment, the furnace battery temperature control waveform 1202 corresponding to the input heat amount 1201 also corresponds to the predicted value of the furnace temperature when the cokemaking process is controlled with the input heat amount that satisfies the predetermined evaluation of the amount of change in the input heat amount.

[0080] Next, the input heat amount calculation unit 103 calculates the input heat amount corresponding to the target furnace battery temperature calculated by the process of Fig. 11 and reflects it in the coke production process, but this process is the same as that explained with reference to Fig. 7 in the first embodiment, so its explanation will be omitted here. Note that the input heat amount calculation unit 103 has already taken in the combustion information data in step S1102, so there is no need to perform step S701 again.

[0081] As described above, similar to the first embodiment, when controlling the heat input amount to adjust the coke temperature to a value corresponding to the target temperature, the target furnace battery temperature (target furnace battery temperature pattern) for multiple future runs can be dynamically calculated by taking into account how the coke temperature for each run varies depending on the first influencing factor, including the furnace battery temperature. This allows the coke temperature to be adjusted to a value corresponding to the target temperature and suppresses coke temperature variation, which is expected to have effects such as reduced production costs (e.g., reducing excess carbonization heat due to overcarbonization), stabilizing production (e.g., avoiding the risk of clogging during extrusion due to undercarbonization or overcarbonization), and stabilizing quality (suppressing quality variation due to variations in the carbonization state). Furthermore, by satisfying a predetermined evaluation that suppresses large changes in the heat input amount when calculating the target furnace battery temperature pattern in the target furnace temperature calculation unit 102, the feasibility of the target furnace battery temperature pattern can be improved, and the coke temperature can be adjusted to a value corresponding to the target temperature and suppressing coke temperature variation with high accuracy.

[0082] [Example 2] An offline simulation of the control method according to the third embodiment (hereinafter referred to as the present control method in the description of Example 2) was carried out using past performance data. As in Example 1 (see FIG. 10 ), the model used in the target furnace temperature calculation unit 102 was one that included the regression model 901 that predicts the coke temperature, as described in the second embodiment, and the big data model 902 that estimates the prediction error of the regression model 901 and outputs the estimated error. Figures 13(a) to (c) show the results of an offline simulation, including the coke temperature, furnace battery temperature, and input heat amount. The dotted lines in the figures represent the results of this control method. As a comparative example, the solid lines represent the results of operation performed by manipulating the control valve 5 at the discretion of the operator. Example 2 was performed under the same conditions as Example 1, and the comparative example for Example 2 is the same as the comparative example for Example 1. As in Example 1, this control method enables proactive action based on future predictions (Fig. 13(b) and (c)). As shown in Fig. 13(a), the coke temperature can be adjusted to a value corresponding to the target temperature, and the variation in the coke temperature can be suppressed. This is expected to result in reduced production costs, stable production, and stable quality. Furthermore, in Example 1, as shown in Fig. 10(c), the input heat quantity changed somewhat significantly around March 16. In contrast, in Example 2, as shown in Fig. 13(c), large changes in the input heat quantity were suppressed. Note that, although specific values ​​for the coke temperature, oven battery temperature, and input heat quantity are omitted in Figs. 10 and 13 and the vertical and horizontal scales are different between Figs. 10 and 13, it was confirmed that changes in the input heat quantity were suppressed in Example 2 compared to Example 1.

[0083] [Fourth embodiment] The fourth embodiment is an example in which the influence of a delay or an early removal from the kiln (called a return) is taken into consideration. Figure 14 illustrates the effect of a delay in unloading on coke temperature. (a) shows the time series of coke temperature changes, and (b) shows the time series of carbonization time changes. In Figure 14(a), the line parallel to the time axis with "target" next to it indicates the target coke temperature. In Figure 14(b), the line parallel to the time axis with "standard" next to it indicates the standard carbonization time. The range affected by the unloading delay indicates the loading pattern when a delay in unloading occurs due to equipment trouble or other reasons. In the range affected by the unloading delay, extra carbonization occurs due to the delay in unloading. Also, [1], ..., and [5] in the figure represent pattern 1 (coking chambers No. 1, 6, 11, 16, etc.), ..., and pattern 5 (coking chambers No. 5, 10, 15, 20, etc.), respectively. As shown in Figure 14(b), for example, assume that before the start of the first rung [1], a delay in unloading occurs due to equipment trouble or the like. In this case, as shown in the affected range of the unloading delay in the figure, the first rung [1] and the following four runs [3], [5], [2], and [4] are loaded with coal, and the carbonization time is extended by the delay in unloading (see the upward arrow in Figure 14(b)). Furthermore, as the carbonization time is extended, the coke temperature increases, as shown in Figure 14(a) (see the upward arrow in Figure 14(a)). Note that while Figure 14 describes a delay in unloading, in the case of unloading return, the carbonization time is shortened by the time of unloading return, and the coke temperature decreases. Therefore, in the fourth embodiment, when the target oven temperature calculation unit 102 predicts the coke temperature, the influence of the delay in oven discharge or return is taken into consideration so that the coke temperature can be predicted more accurately.

[0084] The fourth embodiment will be described below, with the differences from the first to third embodiments being mainly described, and descriptions of the same aspects as the first to third embodiments being omitted. In the fourth embodiment, the coke temperature prediction function of the target oven temperature calculation unit 102 has a correction function for correcting the planned carbonization time using the delay or take-back time when a delay or take-back in oven discharge occurs. The planned carbonization time is an explanatory variable of the coke temperature prediction model, and by correcting the planned carbonization time, the influence of the delay or take-back in oven discharge can be taken into account when the target oven temperature calculation unit 102 predicts the coke temperature.

[0085] The process of calculating the target furnace battery temperature pattern in the fourth embodiment will be described with reference to FIGS. First, with reference to Figs. 17 and 18, a specific example will be described in which, when a delay or return of the furnace discharge occurs, the expected dry distillation time is corrected using the delay or return time. In this embodiment, an example will be described in which there are two coke ovens 1 (Oven A and Oven B), and an extruder is shared between Oven A and Oven B. Oven A and Oven B have the same configuration, and both are divided into five-oven intervals, such as series 1 to series 5, and coal loading work is performed on each series starting from the lowest number. The order of coal loading before unloading is, for example, series 1 of Oven A, series 1 of Oven B, series 3 of Oven A, series 3 of Oven B, series 5 of Oven A, series 5 of Oven B, series 2 of Oven A, series 2 of Oven B, series 4 of Oven A, and series 4 of Oven B. It is also assumed that the control device 100 controls the input heat amount for the A furnace so that the coke temperature in the A furnace follows the target temperature.

[0086] Figure 17 shows the case where a delay in unloading occurs in furnace A. In Figures 17 and 18, a circle represents one carbonization chamber 2 (hereinafter referred to as a kiln) in furnace A, and a black circle represents one carbonization chamber 2 (hereinafter referred to as a kiln) in furnace B. In both furnace A and furnace B, each row contains 13 kilns, and the unloading and loading work is carried out in order from the lowest number in each row. Assume that the kiln discharge pitch (the time from the completion of one kiln's discharge to the completion of the next kiln's discharge) is 8 minutes. The estimated time from the start of discharge at the first kiln to the end of discharge at the last kiln in one sequence of Furnace A is 8-minute pitch × (13 kilns - 1) = 96 minutes. Therefore, when there is a time difference between the start time of discharge at the first kiln + the estimated time (96 minutes) and the actual end time of discharge at the last kiln, it can be determined that a discharge delay or return has occurred on that sequence. This time difference is the discharge delay or return time. In the example of Figure 17, Furnace A's sequence 4 [4] is operating normally, but the next sequence 1 [1] of Furnace A experiences a 30-minute discharge delay.

[0087] In this case, the discharge delay time that occurred in Furnace A is corrected by adding it to the scheduled carbonization time for the following four cases in Furnace A. In this example, a 30-minute discharge delay is detected at the end of the discharge operation for Route 1 [1], and the next Route 3 [3], Route 5 [5], Route 2 [2], and Route 4 [4] are loaded. Therefore, 30 minutes are added to the scheduled carbonization time for each of Route 3 [3], Route 5 [5], Route 2 [2], and Route 4 [4]. Also, for example, if a 30-minute discharge delay is detected at the end of the discharge operation for Route 5 [5], 30 minutes are added to the scheduled carbonization time for each of Route 2 [2], Route 4 [4], Route 1 [1], and Route 3 [3]. In this way, it is possible to determine whether there is a delay or a return in unloading for each of the A furnaces at the timing when each of the A furnaces has finished unloading.

[0088] Figure 18 shows the case where a delay occurs in unloading from furnace B. In one run of a set of furnaces A and B, the estimated time from the start of unloading the first kiln in furnace A to the end of unloading the last kiln in furnace B is determined by the operating rate (unloading schedule), e.g., 4 hours. Therefore, when there is a time difference between the start time of unloading the first kiln in the previous run of furnace A plus the estimated time (4 hours) and the start time of unloading the first kiln in furnace A, it can be determined that a delay or return occurred in unloading furnace A or furnace B before the start time of unloading the first kiln in furnace A. This time difference is the unloading delay or unloading return time. In the example of Figure 18, normal operation is observed in run 4 [4] of the set of furnaces A and B, but a 0.5-hour (30-minute) unloading delay occurs in run 1 [1].

[0089] Here, the determination at each of the start timings of unloading described in FIG. 18 cannot detect whether a delay in unloading or a return occurred in furnace A or furnace B. Therefore, the determination at each of the end timings of unloading described in FIG. 17 is also used. The determination at each end timing of unloading described in FIG. 17 detects a delay in unloading or a return that occurred in furnace A. Therefore, a delay in unloading or a return that occurred in furnace B can be detected by subtracting the result (unloading delay or unloading return time) detected by the determination at each end timing of unloading described in FIG. 17 from the result (unloading delay or unloading return time) detected by the determination at each start timing of unloading described in FIG. 18. Furthermore, by combining the determination of whether or not there was a delay in unloading or a return at each start and end timing of unloading, it becomes possible to detect a delay in unloading or a return early.

[0090] In this case, the scheduled carbonization time is corrected by adding the discharge delay time that occurred in Furnace B to the scheduled carbonization time for the current case (the case where the discharge delay was detected at the start timing of the discharge work) and the following four cases in Furnace A. In the example shown in Figure 18, a 30-minute discharge delay was detected at the start timing of the discharge work for Case 3 [3], and the current Case 3 [3], the next Case 5 [5], the next Case 2 [2], the next Case 4 [4], and the next Case 1 [1] are loaded with coal. Therefore, 30 minutes are added to the scheduled carbonization time for Case 3 [3], the next Case 5 [5], the next Case 2 [2], the next Case 4 [4], and the next Case 1 [1]. In this way, it is possible to determine whether there will be a delay or recall in unloading from furnace B at the start timing of each unloading operation from furnace A.

[0091] FIG. 15 shows the functional configuration of the control device 100 according to the fourth embodiment. The functional configuration shown in FIG. 15 differs from the functional configuration shown in FIG. 1 in that a carbonization time correction unit 105 for correcting the estimated carbonization time is added, and the single-arrow line from the target furnace temperature calculation unit 102 to the input heat amount calculation unit 103 is changed to a double-arrow line connecting the target furnace temperature calculation unit 102 and the input heat amount calculation unit 103 (see the description of the changes to the functional configuration shown in FIG. 1 in the third embodiment). Therefore, detailed descriptions of the input unit 101, target furnace temperature calculation unit 102, input heat amount calculation unit 103, and input heat amount setting unit 104, which are the same as those of the control device 100 according to the first embodiment, will be omitted. FIG. 16 is a flowchart showing the processing of the target furnace temperature calculation unit 102, input heat amount calculation unit 103, and carbonization time correction unit 105. This embodiment will be described based on the third embodiment, and the same processes as those in the flowchart of FIG. 11 will be denoted by the same reference numerals, and their descriptions will be omitted. In this embodiment, the optimal furnace battery temperature pattern is determined for Furnace A, and the flowchart in Figure 16 is executed at the end of a run for Furnace A (e.g., the end of each run of unloading for Furnace A) and the start of a run for Furnace A (e.g., the start of each run of unloading for Furnace A). The calculation flow remains the same for both the end and start of a run, and calculations are performed by adding corrections for unloading delays or take-backs only for the scheduled carbonization time to the obtained operational performance data and schedule data. For example, in the calculation at the start of Figure 17, performance data up to [4] of Run 4 is obtained, so calculations are performed using performance data prior to [4] of Run 4 and the future schedule. The future schedule reflects the unloading delay or take-back time detected at the start of [1] of Run 1.

[0092] After the target furnace temperature calculation unit 102 retrieves the carbonization information data in steps S1101 and S1102 and the input heat amount calculation unit 103 retrieves the combustion information data, the carbonization time correction unit 105 detects whether a delay in unloading or a return has occurred in step S1601. As described in FIG. 17, when detecting whether a delay in unloading or a return has occurred at the end timing of unloading work on a street including furnace A, the carbonization time correction unit 105 can determine that a delay in unloading or a return has occurred on that street when there is a time difference between (the start time of unloading work on the first kiln + the scheduled time) and (the actual end time of unloading work on the last kiln) on that street. If (the start time of unloading work on the first kiln + the scheduled time) is earlier than (the actual end time of unloading work on the last kiln), a delay in unloading has occurred in furnace A, and if it is later, a return has occurred.

[0093] 18, in the case of detecting whether or not a delay in unloading or a return has occurred at the start timing of unloading work for a certain row of furnace A, when there is a time difference between (the start time of unloading work for the first kiln in the row immediately before the row in furnace A + the scheduled time) and (the start time of unloading work for the first kiln in the row in furnace A), the carbonization time correction unit 105 can determine that a delay in unloading or a return has occurred in furnace A or furnace B before the start time of the first unloading work for the row in furnace A. If (the start time of unloading work for the first kiln in the row immediately before the row in furnace A + the scheduled time) is earlier than (the start time of unloading work for the first kiln in the row in furnace A), a delay in unloading has occurred in furnace A or furnace B, and if it is later, a return of unloading has occurred in furnace A or furnace B. In either case, a threshold value for the time difference may be set, and only when the time difference exceeds the threshold value, it may be detected that a delay in unloading or a return has occurred.

[0094] In step S1602, if the occurrence of a kiln discharge delay or kiln discharge return is detected in step S1601, the carbonization time correction unit 105 corrects the scheduled carbonization time using the time difference as the kiln discharge delay time or the kiln discharge return time, and then proceeds to step S1103. In this way, when the target furnace temperature calculation unit 102 calculates the target furnace battery temperature pattern, correction of the scheduled carbonization time is performed. When a kiln discharge delay occurs as described in Figures 17 and 18, the carbonization time correction unit 105 adds the kiln discharge delay time to the scheduled carbonization time. Also, when a kiln discharge return occurs, the carbonization time correction unit 105 subtracts the kiln discharge return time from the scheduled carbonization time. The scheduled carbonization time is an explanatory variable of the coke temperature prediction model, and by correcting the scheduled carbonization time, the influence of the kiln discharge delay or the coke discharge return can be reflected when predicting the coke temperature in step S1107.

[0095] The subsequent processing in steps S1103 to S1112 is the same as that in FIG. In the examples shown in FIGS. 17 and 18 , the presence or absence of a kiln discharge delay or return is determined at the end and start of the A furnace run in order to enable early detection of a kiln discharge delay or return. However, the determination of a kiln discharge delay or return is not limited to this timing. For example, the presence or absence of a kiln discharge delay or return may be determined at only one of the end and start of the A furnace run. Furthermore, if the B furnace is also subject to control of the input heat amount by the control device 100, the presence or absence of a kiln discharge delay or return may be determined at at least one of the end and start of the B furnace run in addition to or instead of at least one of the end and start of the A furnace run. Furthermore, although this embodiment has been described based on the third embodiment, the correction function described in this embodiment may also be applied to the first and second embodiments.

[0096] As described above, when predicting the coke temperature in the target oven temperature calculation unit 102, the influence of the delay in oven discharge or return can be taken into consideration, and the coke temperature can be predicted more accurately. By predicting the coke temperature more accurately, the coke temperature can be set to a value corresponding to the target temperature, and production costs can be reduced (reducing the amount of excess carbonization heat due to over-carbonization).

[0097] [Example 3] Using past performance data, an offline simulation was carried out for a control method (hereinafter referred to as this control method in the explanation of Example 3) that takes into account delays in unloading from the kiln (with correction of the scheduled carbonization time) as in the fourth embodiment. In addition, as a comparative method, an offline simulation was carried out for a control method that does not take into account delays in unloading from the kiln (without correction of the scheduled carbonization time). Fig. 19 is a characteristic diagram showing the results of Example 3, where (a) shows the time series change in coke temperature and (b) shows the time series change in furnace battery temperature. In the diagram, the solid line (◯) shows the results obtained by the present control method, and the dotted line (△) shows the results obtained by the comparative example. As shown in Figure 19(a), this control method can accurately adjust the coke temperature to a value corresponding to the target temperature. Also, as shown in Figure 19(b), this control method takes into account the effect of a delay in discharge from the oven, which increases the carbonization time, and as a result, it takes action to reduce the oven battery temperature, thereby reducing production costs (reducing the amount of excess carbonization heat due to over-carbonization).

[0098] [Fifth embodiment] The fifth embodiment is an example in which, in the event of a delay or return of the kiln discharge, the operator can, at his own discretion, consider shortening or extending the time required for the kiln discharge and loading work relative to the standard work time. The fifth embodiment will be explained below. Explanations of the content common to the first to fourth embodiments will be omitted, and the explanation will focus on the differences from the first to fourth embodiments.

[0099] Figure 20 illustrates the impact of operator-determined operations on coke temperature during a delay in discharge. (a) shows the time series of coke temperature changes, and (b) shows the time series of carbonization time changes. In Figure 20(a), the line parallel to the time axis labeled "target" indicates the target coke temperature, and in Figure 20(b), the line parallel to the time axis labeled "standard" indicates the standard carbonization time. As explained in Figures 14(a) and 14(b), the range affected by a discharge delay indicates the coal loading pattern when a discharge delay occurs due to equipment trouble or other reasons. In the range affected by a discharge delay, extra carbonization occurs due to the discharge delay. Also, in Figures 20(a) and (b), as in Figures 14(a) and (b), [1], ···, and [5] in the figures represent the first case (coking chamber No. 1, 6, 11, 16, etc.) and the fifth case (coking chamber No. 5, 10, 15, 20, etc.), respectively. Also, in Figure 20(b), as in Figure 14(b), assume that before the start of the first case [1], a delay in unloading due to equipment trouble or the like occurs. In this case, as shown in the range affected by the unloading delay in the figure, the first case [1] and the following four cases [3], [5], [2], and [4] are in a loaded state, and the carbonization time is extended by the amount of the unloading delay, and the coke temperature also increases (see the upward white arrows in Figures 20(a) and (b)). In such cases, to avoid production reductions in the coke oven 1, operators tend to operate the coke oven 1 (return) at their own discretion so that the unloading coal charging operation is shorter than the standard operating time. This is because, since the coke temperature has risen above the target temperature, it is determined that the coke can be properly carbonized even if the carbonization time is shortened. For example, the inventors have found that operators tend to perform return coal in the latter half of the range affected by the unloading delay so as not to affect the start of the next carbonization cycle. In the figure, the period during which return coal is performed is indicated as the return period. Note that the first [1] range after the range affected by the unloading delay is not affected by the unloading delay, so the carbonization time is not extended and is not affected by the unloading delay. According to the above-mentioned control, the input heat amount (furnace temperature) is reduced on the assumption that additional coke will remain in the coke oven 2 for the duration of the unloading delay.Therefore, if the operator decides to take back the coke at his own discretion, the carbonization time will be shortened by taking back the coke, even though the input heat amount has been reduced, and this may result in a lower coke temperature than intended (see the downward white arrows in Figures 20(a) and (b)). Note that the take-back period is not limited to the period shown in Figure 20(a) and is determined at the discretion of the operator.

[0100] FIG. 21 is a diagram illustrating a method for correcting the planned carbonization time. (a) is an example of time-series data of the corrected planned carbonization time in the method of the fourth embodiment, and (b) is an example of time-series data of the corrected planned carbonization time in the method of the present embodiment. In the fourth embodiment, as shown in FIG. 21(a), when a delay in discharge occurs, the same time is uniformly added as a discharge delay time to each planned carbonization time within the range affected by the discharge delay. Therefore, the target furnace battery temperature pattern and the input heat amount are calculated with the planned carbonization time extended. In such a case, if the operator decides to withdraw the coke, the coke temperature will decrease as described above, resulting in a so-called undercooked state and a deterioration in coke quality. Note that while the case of a delayed discharge has been described here, when a discharge withdrawal occurs, operators tend to perform operations that extend the coal loading operation beyond the standard operating time (called delayed discharge), which may result in an excessively high coke temperature.

[0101] Therefore, in the fifth embodiment, as shown in Fig. 21(b), in anticipation of an operation based on the operator's judgment when a delay or return of the kiln discharge occurs, the corrected planned carbonization time for at least one run is set to be different from the corrected planned carbonization time for at least one other run. When taking into account the tendency based on the operator's judgment as described above, the absolute value of the correction amount for the pre-correction planned carbonization time for a later run is not made larger than the absolute value of the correction amount for the pre-correction planned carbonization time for an earlier run, and the absolute value of the correction amount for the pre-correction planned carbonization time for a later run is made smaller than the absolute value of the correction amount for the pre-correction planned carbonization time for an earlier run for at least two runs. In other words, the absolute value of the correction amount for the pre-correction planned carbonization time for a later run may be the same as the absolute value of the correction amount for the pre-correction planned carbonization time for an earlier run, but the former is never smaller than the latter, and the latter may be smaller than the former.

[0102] The functional configuration of the control device for a coke production process according to the fifth embodiment is the same as the functional configuration shown in Fig. 15, but part of the carbonization time correction unit 105 is different from that of the fourth embodiment. Therefore, detailed descriptions of the input unit 101, target furnace temperature calculation unit 102, input heat amount calculation unit 103, input heat amount setting unit 104, and carbonization time correction unit 105 that are the same as those of the control device 100 according to the first to fourth embodiments will be omitted. Also, in the flowchart showing the processing of the target furnace temperature calculation unit 102 and the input heat amount calculation unit 103 according to the fifth embodiment, step S1602 in Fig. 16 is changed as follows, and in this embodiment, the carbonization time correction unit 105 performs the following processing in step S1602 in Fig. 16.

[0103] That is, in step S1602, if the occurrence of a kiln discharge delay or kiln discharge return is detected in step S1601, the carbonization time correction unit 105 calculates the time difference as the kiln discharge delay time or the kiln discharge return time. As described in the fourth embodiment, if a kiln discharge delay occurs, the carbonization time correction unit 105 adds the kiln discharge delay time to the planned carbonization time. Also, if a kiln discharge return occurs, the carbonization time correction unit 105 subtracts the kiln discharge return time from the planned carbonization time.

[0104] When the occurrence of the delay in unloading or the take-back is detected in step S1601, the target furnace temperature calculation unit 102 calculates the coke temperature T co In order to calculate (n) using equation (2), in step S1602, the dry distillation time correction unit 105 corrects the planned dry distillation time for the streets within the range of influence of the delay in unloading from the oven after the timing when the delay in unloading or return occurs (this is the same in the fourth embodiment). As described above, in the fourth embodiment, the dry distillation time correction unit 105 adds the kiln discharge delay time or subtracts the kiln discharge return time to each planned dry distillation time, so that the planned dry distillation time for each method is changed to the same time.

[0105] In contrast to this, as described above, in this embodiment, the dry distillation time correction unit 105 sets the corrected planned dry distillation time for at least one of the routes within the range affected by the delay in dry distillation after the timing at which the delay in dry distillation or the return of dry distillation occurred to be different from that for at least one other route. More specifically, the dry distillation time correction unit 105 does not make the absolute value of the correction amount for the planned dry distillation time before correction for a later route larger than the absolute value of the correction amount for the planned dry distillation time before correction for an earlier route, and sets the absolute value of the correction amount for the planned dry distillation time before correction for a later route smaller than the absolute value of the correction amount for the planned dry distillation time before correction for an earlier route for at least two routes.

[0106] For example, a positive coefficient is preset for each run to be multiplied by the kiln unloading delay time and the unloading return time to correct the expected carbonization time for each run after the timing of the unloading delay or return. In this case, the coefficient for the later run is not set to be larger than the coefficient for the earlier run, and for at least two runs, the coefficient value for the later run is set to be smaller than the coefficient value for the earlier run. In this embodiment, since unloading coal is performed every five runs, it is sufficient to preset the coefficients for up to five runs. For example, the coefficients for the last two runs within the range of influence of the unloading delay or return are set to "0.5," and the coefficients for the remaining runs are set to "1.0." Then, the dry distillation time correction unit 105 calculates the correction time by multiplying the kiln discharge delay time and the kiln discharge return time by a coefficient set for each method, and calculates the corrected dry distillation planned time by adding or subtracting the correction time to the planned dry distillation time before correction.

[0107] Here, the next steps in which the coal loading work is carried out after a delay or return of the kiln discharge will be referred to as the next steps. In the example shown in FIG. 17 described in the fourth embodiment, a delay in unloading occurs in furnace A in case 1. Therefore, the cases within the range of influence of the unloading delay in furnace A after the timing at which the unloading delay occurs are the following cases 3, 5, 2, and 4, and the order in which the unloading coal loading work is carried out is in this order. Therefore, in the above example, the last two cases within the range of influence at the timing at which the unloading delay or return occurs are the following cases 2 and 4, and the remaining cases are the following cases 3 and 5. From the above, the coefficients for the following cases 3, 5, 2, and 4 are "1.0", "1.0", "0.5", and "0.5", respectively. Therefore, if the oven discharge delay time is 30 minutes, the distillation time correction unit 105 calculates the correction times as follows: 3, 5, 2, and 4, and determines the corrected distillation time as the time obtained by adding the correction times to the distillation schedule time (before correction). The following three correction times: 30 minutes (= 30 minutes x 1.0) The following 5 correction times: 30 minutes (= 30 minutes x 1.0) The following two correction times: 15 minutes (= 30 minutes x 0.5) The following four correction times: 15 minutes (= 30 minutes x 0.5)

[0108] In the example shown in FIG. 18 described in the fourth embodiment, a delay in unloading occurs in furnace B in case 1. Therefore, the cases within the range of influence of the unloading delay in furnace A after the timing of the occurrence of the unloading delay are next case 3, next case 5, next case 2, next case 4, and next case 1, and the order of the unloading coal loading work is carried out in this order. Therefore, in the above example, the last two cases within the range of influence of unloading at the timing of the occurrence of the unloading delay or return are next case 4 and next case 1, and the remaining cases are next case 3, next case 5, and next case 2. From the above, the coefficients for next case 3, next case 5, next case 2, next case 4, and next case 1 are "1.0", "1.0", "1.0", "0.5", and "0.5", respectively. Therefore, if the oven discharge delay time is 30 minutes, the distillation time correction unit 105 calculates the correction times as follows: 3, 5, 2, 4, and 1, and determines the corrected distillation time as the time obtained by adding the correction times to the distillation schedule time (before correction). The following three correction times: 30 minutes (= 30 minutes x 1.0) The following 5 correction times: 30 minutes (= 30 minutes x 1.0) The following two correction times: 30 minutes (= 30 minutes x 1.0) The following four correction times: 15 minutes (= 30 minutes x 0.5) Correction time for the following 1: 15 minutes (= 30 minutes x 0.5)

[0109] The method for correcting the estimated carbonization time is not limited to the calculation described above. For example, the coefficients for all runs may be different values. Furthermore, the coefficient for a later run may be set to a larger value than the coefficient for an earlier run, depending on the characteristics of the coke oven 1, the operating conditions, the operating tendency determined by the operator, and the like. Furthermore, the calculation formula is not limited to the one described above. For example, the correction time may be calculated so that the correction time for a later run becomes exponentially shorter. Also, like the fourth embodiment, the present embodiment has been described based on the third embodiment, but the correction function described in this embodiment may be applied to the first and second embodiments.

[0110] As described above, when predicting the coke temperature in the target furnace temperature calculation unit 102, the influence of the operator's judgment in the event of a delay in or return from the oven discharge can be taken into consideration, and the coke temperature can be predicted more accurately. By predicting the coke temperature more accurately, it is possible to set the coke temperature to a value corresponding to the target temperature with high precision, and to further reduce production costs (reduce the amount of excess carbonization heat due to over-carbonization).

[0111] [Sixth embodiment] The sixth embodiment is an example in which the target furnace battery temperatures (target furnace battery temperature patterns) for a plurality of future passage times calculated by the target furnace temperature calculation unit 102 are corrected. Although the coke temperature prediction model described in the second embodiment improves the accuracy of coke temperature predictions, prediction errors remain. Furthermore, using the big data model 902 increases the computational load and requires a large amount of training data for machine learning. Therefore, for example, using the big data model 902 is not preferable when the goal is to reduce the computational load or when training data cannot be prepared. In such cases, using the regression model 901 (the coke temperature prediction model described in the first embodiment) results in a larger prediction error than using the big data model 902. Furthermore, in the third embodiment, a furnace battery temperature prediction model is used to calculate the target furnace battery temperature pattern, and therefore the prediction error of the furnace battery temperature prediction model also affects the target furnace battery temperature pattern. Furthermore, it is possible that the schedule values ​​used in prediction models including the coke temperature prediction model may significantly deviate from the actual values. These prediction errors of prediction models including the coke temperature prediction model and the deviation of the schedule values ​​from the actual values ​​can cause a decrease in the accuracy of the target furnace battery temperature pattern. Therefore, in the sixth embodiment, the target furnace battery temperature pattern calculated by the target furnace temperature calculation unit 102 is corrected according to the difference between the target value and the actual value of the physical quantity representing the coke carbonization state, thereby improving the feasibility of the target furnace battery temperature pattern.

[0112] The sixth embodiment will be described below. Explanation of the content common to the first to fifth embodiments will be omitted, and the explanation will be based on the third embodiment, focusing on the differences from the first to fifth embodiments. As explained in the first embodiment, the physical quantity representing the carbonization state of the coke may be, for example, the temperature of the oven wall 4 in addition to the coke temperature. However, in this explanation, the physical quantity representing the carbonization state of the coke will be the coke temperature.

[0113] FIG. 22 shows the functional configuration of the control device 100 according to the sixth embodiment. The functional configuration shown in FIG. 22 differs from the functional configuration shown in FIG. 1 in that a target furnace temperature correction unit 106 that corrects the target furnace battery temperature pattern is added, and the single-arrow line from the target furnace temperature calculation unit 102 to the input heat amount calculation unit 103 is changed to a double-arrow line connecting the target furnace temperature calculation unit 102 and the input heat amount calculation unit 103 (see the explanation of the changes to the functional configuration shown in FIG. 1 in the third embodiment). Therefore, detailed explanations of the input unit 101, target furnace temperature calculation unit 102, input heat amount calculation unit 103, and input heat amount setting unit 104 that are the same as those of the control device 100 according to the first to fifth embodiments will be omitted. FIG. 23 is a flowchart showing the processing of the target furnace temperature calculation unit 102, input heat amount calculation unit 103, and target furnace temperature correction unit 106. As described above, this embodiment will be described based on the third embodiment. Therefore, in the flowchart of Fig. 23, the same processes as those in the flowchart of Fig. 11 are denoted by the same reference numerals, and the description thereof will be omitted. Fig. 24 is a diagram for explaining the outline of the process of correcting the target reactor battery temperature pattern.

[0114] The process of correcting the target reactor battery temperature pattern in the sixth embodiment will be described with reference to FIGS. In Figure 24, "Start of control" indicates the timing when control by the control device 100 begins. The current time indicates the timing when the flowchart in Figure 23 begins. As with the flowchart in Figure 11 described in the third embodiment, the timing when the flowchart in Figure 23 begins is, for example, when the unloading operation in each run is completed, but this is not necessarily limited to this timing and may be, for example, when the loading operation in each run begins. Furthermore, the flowchart in Figure 23 is executed at a run time cycle.

[0115] The upper part of Fig. 24 shows the time series change in the coke temperature. Here, the actual values ​​of the coke temperature 2411 are shown with black circles, and the predicted values ​​are shown with white circles. The time series change in the coke temperature is shown by connecting the actual values ​​of the coke temperature with a solid line and connecting the predicted values ​​with a dashed dotted line. The upper part of Fig. 24 also shows the target temperature 2412 for the coke temperature together with the coke temperature 2411. The middle part of Figure 24 shows the time series changes in furnace battery temperature. The furnace battery temperature is obtained every hour, but here the actual value of the furnace battery temperature 2421 is shown by the solid line, and the predicted value is shown by the dashed dotted line. The middle part of Figure 24 also shows the target furnace battery temperature pattern 2422 calculated by the target furnace temperature calculation unit 102 and the corrected target furnace battery temperature pattern 2423 corrected by the target furnace temperature correction unit 106. The time series change in the input heat quantity is shown in the lower part of Fig. 24. Here, the actual value of the input heat quantity 2431 is shown by a solid line, and the future value is shown by a dashed line.

[0116] As explained in the third embodiment, by performing the processing of steps S1101 to S1111, the target battery temperature pattern 2422 that minimizes the evaluation function J1' is obtained from the target battery temperature patterns converted in step S1106. In the middle part of Figure 24, the target battery temperature pattern 2422 beyond the current time is calculated at the current time.

[0117] Step S2301 is performed after step S1110. In step S2301, the target furnace temperature correction unit 106 corrects the target furnace battery temperature pattern 2422 according to the difference between the target value (target temperature 2412) of the coke temperature and the actual value. For example, the target furnace temperature correction unit 106 calculates a weighted average of values ​​obtained by subtracting the actual value of the coke temperature for each pattern in the past carbonization cycle, starting from the current time, from the target value of the coke temperature (target temperature 2412). For example, assume that the unloading coal loading sequence is sequence 1, sequence 3, sequence 5, sequence 2, and sequence 4, and the start timing of the flowchart in FIG. 23 is the timing when the unloading operation for each sequence is completed, and the current time is the timing when the unloading operation for sequence 1 is completed (the [1] shown below the current time in the upper part of FIG. 24 indicates this). In this case, the actual values ​​of the coke temperature in the previous cases 1, 4, 2, 5, and 3 are used (in the upper part of Figure 24, to the left of [1] shown under the current time, [4], [2], [5], and [3] shown in order from left to right represent these).

[0118] If k is a variable that identifies a row and the number of rows for which the weighted average value is calculated is K (in this embodiment, K=5), the correction amount FB to be added to the target furnace battery temperature pattern 2422 is expressed, for example, by the following equations (6a) and (6b).

[0119]

number

[0120] Here, G1 and G2 are preset positive gains. k is a preset positive weighting coefficient for the kth street. co_p_k is the target coke temperature for pass k. T co_m_kis the actual value of the coke temperature in pass k. A positive correction amount FB corresponds to a low coke temperature because the target value of the coke temperature is higher than the actual value. In this case, it is considered to significantly increase the input heat amount in order to make the coke temperature reach the target temperature 2412 as quickly as possible. On the other hand, a negative correction amount FB corresponds to a high coke temperature because the target value of the coke temperature is lower than the actual value. In this case, if the coke temperature is suddenly reduced, the coke temperature may become too low, so it is considered to gradually reduce the input heat amount. In such a case, it is preferable to make the gain G1 larger than the gain G2. However, the values ​​of the gains G1 and G2, including the magnitude relationship between these values, may be determined appropriately depending on the characteristics and operating conditions of the coke oven 1.

[0121] Weighting factor w k may be appropriately determined depending on the characteristics and operating conditions of the coke oven 1. For example, when emphasis is placed on the difference between the target value and the actual value of the coke temperature at a time close to the current time, the weighting coefficient w k The weighting factor for the previous street, w k For at least two streets, the weighting factor w for the later streets must be no smaller than k The weighting factor for the previous street, w k Make it bigger than.

[0122] 24, the actual value of the coke temperature 2411 is higher than the target temperature 2412, so the correction amount FB is calculated by equation (6b). In this case, the correction amount FB has a negative value. Here, it is not desirable to change the target furnace battery temperature pattern too abruptly from the viewpoint of stable operation. Therefore, if the correction amount FB is not within the range of the preset upper and lower limit values ​​(lower limit value ≦ FB ≦ upper limit value), the target furnace temperature correction unit 106 changes the correction amount FB to the value closest to the correction amount between the upper limit value and the lower limit value. If the correction amount FB is within the range of the preset upper and lower limit values ​​(lower limit value ≦ FB ≦ upper limit value), the target furnace temperature correction unit 106 does not change the correction amount FB in this way.

[0123] The target furnace temperature correction unit 106 calculates the corrected target furnace battery temperature pattern 2423 by adding the correction amount FB obtained as described above to the target furnace battery temperature pattern 2422 calculated by performing the processing of steps S1101 to S1111. In the example shown in the upper part of Figure 24, a negative correction amount FB is calculated, so the corrected target furnace battery temperature pattern 2423 will show a lower temperature than the target furnace battery temperature pattern 2422 (see the downward hollow arrow line in the middle part of Figure 24).

[0124] In this manner, the corrected target battery temperature pattern is calculated in step S2301. Then, in step S2302, the target furnace temperature correction unit 106 outputs the corrected target furnace battery temperature pattern to the input heat amount calculation unit 103. In this embodiment, the input heat amount calculation unit 103 calculates the input heat amount using the target furnace battery temperature pattern calculated by the target furnace temperature correction unit 106, rather than the target furnace battery temperature pattern calculated by the target furnace temperature calculation unit 102, in the flowchart of Figure 7. Note that in step S1104, the input heat amount calculation unit 103 calculates the input heat amount using a candidate target furnace battery temperature pattern (the initial value of the target furnace battery temperature pattern generated in step S1103 or the target furnace battery temperature pattern generated in step S1111), rather than the target furnace temperature pattern calculated by the target furnace temperature correction unit 106.

[0125] As described above, by feeding back the actual value of the coke temperature, the target furnace battery temperature pattern calculated by the target furnace temperature calculation unit 102 can be corrected so that the difference between the actual value and the target value of the coke temperature becomes smaller, and the input heat amount can be controlled with even higher precision. This makes it possible to more accurately adjust the coke temperature to a value corresponding to the target temperature, and to further reduce production costs (reduce excess carbonization heat due to over-carbonization). Although the present embodiment has been described based on the third embodiment, the correction function described in the present embodiment may be applied to the first, second, fourth, and fifth embodiments.

[0126] Although the present invention has been described above with reference to the embodiments, the above embodiments are merely illustrative of specific examples of how the present invention can be implemented, and the technical scope of the present invention should not be construed as being limited by these embodiments. In other words, the present invention can be implemented in various forms without departing from its technical concept or main features. The control device for the coke production process to which the present invention is applied can be configured, for example, by a computer device equipped with a CPU, ROM, RAM, etc., and its functions are realized by the CPU executing a predetermined program. The present invention can also be realized by supplying software (program) that realizes control of the coke production process of the present invention to a system or device via a network or various storage media, and having the computer of the system or device read and execute the program. [Explanation of symbols]

[0127] 1. Coke oven 2. Carbonization chamber 3 Combustion chamber 4 Furnace wall 5. Regulating valve 6. Thermometer for measuring furnace temperature 7 Extrusion Ram 8. Thermometer for measuring coke temperature 9 Guide car 10. Coke 100 Control device for coke manufacturing process 101 Input section 102 Target furnace temperature calculation section 103 Input heat amount calculation section 104 Input heat setting section 105 Dry distillation time correction unit 106 Target furnace temperature correction section 200 Storage section 601 Target Battery Temperature Pattern 602 Predicted coke temperature 603 Target temperature of coke 801 Furnace temperature prediction 802 Input heat amount 901 Regression Model 902 Big Data Model 1201 Input heat quantity that minimizes the evaluation function 1202 Furnace temperature control waveform 2411 Coke temperature (actual value, predicted value) 2412 Target temperature of coke 2421 Furnace temperature (actual value, predicted value) 2422 Furnace battery temperature pattern before correction 2423 Corrected furnace battery temperature pattern 2421 Input heat amount S Coal loading amount W Coal moisture content T co Coke Temperature T ro Furnace temperature t k Dry distillation time t t Street time

Claims

1. A control device for a coke production process, in a coke oven having a plurality of carbonization chambers and a plurality of combustion chambers, for controlling an amount of heat input to the combustion chambers so that a physical quantity representing a carbonization state of coke becomes a value corresponding to a target value, a target furnace temperature calculation unit that predicts the physical quantity using a physical quantity prediction model that predicts the physical quantity based on a first influencing factor including a furnace temperature, which is the temperature of the combustion chamber, and at least one of an actual value and a schedule value of the first influencing factor, calculates a value of a first evaluation function that includes a term that represents a difference between the predicted physical quantity and a target value of the physical quantity, and calculates a target furnace temperature based on the calculated value of the first evaluation function; an input heat amount calculation unit that calculates an input heat amount according to the target furnace temperature calculated by the target furnace temperature calculation unit, The control device for a coke manufacturing process, wherein the physical quantity is a coke temperature or a furnace wall temperature.

2. 2. The control device for a coke manufacturing process as described in claim 1, wherein the input heat amount calculation unit calculates the input heat amount corresponding to the target furnace temperature using a furnace temperature prediction model that predicts the furnace temperature based on a second influence factor including the input heat amount, and at least one of an actual value and a scheduled value of the second influence factor.

3. 3. The control device for a coke manufacturing process as described in claim 2, wherein the input heat amount calculation unit calculates the input heat amount based on a second evaluation function including a term representing the difference between the oven temperature predicted by the oven temperature prediction model and the target oven temperature, using the target oven temperature calculated by the target oven temperature calculation unit as the target oven temperature included in the second evaluation function.

4. The control device for a coke manufacturing process according to claim 3 , wherein the second evaluation function further includes a term representing a change in the amount of input heat.

5. 5. The control device for a coke production process according to claim 1, wherein the target furnace temperature calculation unit generates a plurality of candidates for a future target furnace temperature, uses each of the plurality of candidates as the furnace temperature included in the first influencing factor in the physical quantity prediction model to predict the physical quantity and calculate the value of the first evaluation function, and determines the target furnace temperature from among the plurality of candidates based on the value of the first evaluation function thus calculated.

6. 6. The control device for a coke manufacturing process according to claim 1, wherein the target furnace temperature calculation unit calculates the target furnace temperature by using a furnace temperature that satisfies a predetermined evaluation regarding a change in input heat amount as the first influencing factor in the physical quantity prediction model.

7. the input heat amount calculation unit calculates an input heat amount that satisfies a predetermined evaluation regarding a change in the input heat amount as an input heat amount corresponding to a furnace temperature that has not been determined as the target furnace temperature, 7. The control device for a coke production process according to claim 6, wherein the target furnace temperature calculation unit calculates the target furnace temperature as a predicted value of the furnace temperature when the coke production process is controlled with an input heat amount that satisfies a predetermined evaluation regarding the change in the input heat amount, as the furnace temperature when the predetermined evaluation regarding the change in the input heat amount is satisfied.

8. the target furnace temperature calculation unit generates a plurality of candidates for a future target furnace temperature; the input heat amount calculation unit calculates an input heat amount based on the second evaluation function by using each of the target furnace temperature candidates generated by the target furnace temperature calculation unit as a target furnace temperature included in the second evaluation function; The control device for a coke production process according to claim 3 or 4, wherein the target furnace temperature calculation unit calculates the target furnace temperature based on a furnace temperature corresponding to the input heat amount calculated by the input heat amount calculation unit.

9. The control device for a coke manufacturing process according to any one of claims 1 to 8, wherein a cycle for calculating the input heat amount by the input heat amount calculation unit is shorter than a cycle for calculating the target furnace temperature by the target furnace temperature calculation unit.

10. The coke production process is a coke production process in which a plurality of carbonization chambers are divided into a plurality of rows, and a coke unloading and coal loading operation is performed for each row, The control device for a coke manufacturing process according to any one of claims 1 to 9, wherein the target furnace temperature is a transition of the target furnace temperature over a plurality of future passage times.

11. The control device for a coke production process according to claim 10 , wherein the target furnace temperature calculation unit executes a process of calculating the target furnace temperature at a period of a flow time.

12. 12. The control device for a coke manufacturing process according to claim 1, wherein the physical quantity prediction model comprises: a regression model that predicts the physical quantity based on the first influencing factor including a furnace temperature; and a machine learning estimation model that estimates a prediction error of the regression model and corrects the prediction error.

13. the coke production process is a coke production process in which the input heat amounts of the plurality of combustion chambers are adjusted collectively, A control device for a coke manufacturing process according to any one of claims 1 to 12, wherein the input heat amount calculation unit calculates an input heat amount to be adjusted collectively for the multiple combustion chambers as an input heat amount corresponding to the target furnace temperature calculated by the target furnace temperature calculation unit.

14. The control device for a coke manufacturing process according to any one of claims 1 to 13, wherein the furnace temperature is a furnace battery temperature that is a representative value of temperatures of the plurality of combustion chambers.

15. the physical quantity prediction model includes a scheduled dry distillation time as the first influencing factor, The control device for a coke manufacturing process according to any one of claims 1 to 14, further comprising a carbonization time correction unit that, when a delay or return of the discharge from the oven occurs, corrects the planned carbonization time using the delay or return time.

16. the target furnace temperature calculation unit executes a process of calculating the target furnace temperature at least either at the end of a pass or at the start of a pass, The control device for a coke manufacturing process according to claim 15 , wherein the carbonization time correction unit performs the correction when the target furnace temperature calculation unit calculates the target furnace temperature.

17. The control device for a coke manufacturing process according to claim 15 or 16, wherein the carbonization time correction unit sets the corrected planned carbonization time in at least one way to a time different from the corrected planned carbonization time in at least one other way.

18. The control device for a coke manufacturing process described in claim 17, wherein the distillation time correction unit does not make the absolute value of the correction amount for the planned distillation time before correction in a later time step larger than the absolute value of the correction amount for the planned distillation time before correction in an earlier time step, and makes the absolute value of the correction amount for the planned distillation time before correction in a later time step smaller than the absolute value of the correction amount for the planned distillation time before correction in an earlier time step for at least two steps.

19. a target furnace temperature correction unit that corrects the target furnace temperature calculated by the target furnace temperature calculation unit in accordance with a difference between a target value and an actual value of the physical quantity, The control device for a coke production process according to claim 1 , wherein the input heat amount calculation unit calculates an input heat amount according to the target furnace temperature corrected by the target furnace temperature correction unit.

20. 1. A method for controlling a coke production process in a coke oven having a plurality of carbonization chambers and a plurality of combustion chambers, the method comprising: controlling an amount of heat input to the combustion chambers so that a physical quantity representing a carbonization state of coke reaches a value corresponding to a target value, a target furnace temperature calculation step of predicting the physical quantity using a physical quantity prediction model that predicts the physical quantity based on a first influencing factor including a furnace temperature, which is the temperature of the combustion chamber, and at least one of an actual value and a schedule value of the first influencing factor, calculating a value of a first evaluation function that includes a term that represents a difference between the predicted physical quantity and a target value of the physical quantity, and calculating a target furnace temperature based on the calculated value of the first evaluation function; an input heat amount calculation unit step of calculating an input heat amount according to the target furnace temperature calculated in the target furnace temperature calculation step; The method for controlling a coke production process, wherein the physical quantity is a coke temperature or a furnace wall temperature.

21. A program for causing a computer to function as each part of the control device for a coke production process according to any one of claims 1 to 19.

Images