JP7753828B2 - Concentration estimation device, concentration estimation method, and program - Google Patents

Description

The present invention relates to a concentration estimation device, a concentration estimation method, and a program.

In order to achieve both energy saving and pollution prevention in a heating furnace, it is important to control the oxygen (O ₂ ) concentration in the heating furnace, and for this purpose, various methods are used to measure the oxygen concentration. For example, the oxygen concentration in a heating furnace is measured using a zirconia-type oxygen concentration meter (see Non-Patent Document 1).

"Instrumentation Tips | Zirconia Oxygen Analyzer," Internet <URL: https://www.m-system.co.jp/mstoday/plan/mame/b_sensor/9407/index.html>

Generally, to achieve both energy conservation and pollution prevention, heating furnaces need to be operated at low oxygen concentrations, but when using a zirconia oxygen analyzer, they must be operated at a certain oxygen concentration. This is because if the oxygen concentration becomes too low, incomplete combustion occurs, and the zirconia element undergoes an oxidation reaction using carbon monoxide, a combustible gas, and unburned combustible fuel as a catalyst, which can result in an erroneous measurement where the oxygen concentration measurement value becomes close to zero.

One embodiment of the present invention has been developed in light of the above points, and aims to estimate the oxygen concentration inside a heating furnace.

In order to achieve the above object, one embodiment of the concentration estimation device is an oxygen concentration estimation device that estimates the oxygen concentration in a heating furnace, and includes: a model creation unit that creates a model for estimating the oxygen concentration in the heating furnace using, as learning data, an oxygen concentration measurement value y(t) in the heating furnace at each time t within a predetermined past period and a predetermined n (n is a predetermined integer of 1 or more) physical quantity measurement values x ₁ (t), x ₂ (t), ..., x _n (t) related to the heating furnace; and an oxygen concentration estimation unit that calculates an estimate of the oxygen concentration in the heating furnace using the model and the n physical quantity measurement values x ₁ (t'), x ₂ (t'), ..., x _n (t') at a time t' for which the oxygen concentration is to be estimated.

The oxygen concentration inside the heating furnace can be estimated.

1 is a diagram showing an example of the overall configuration of a heating furnace control system according to an embodiment of the present invention; 3A and 3B are diagrams schematically showing the relationship between the excess air ratio and heat loss and the relationship between the excess air ratio and thermal efficiency. FIG. 10 is a diagram schematically illustrating an example of installation of an oxygen concentration meter and peripheral sensors in a heating furnace. FIG. 4 is a diagram schematically showing the relationship between the air-fuel ratio and the oxygen concentration measurement value. FIG. 1 is a diagram illustrating an example of a hardware configuration of an oxygen concentration estimation device according to an embodiment of the present invention. FIG. 2 is a diagram illustrating an example of a functional configuration of the oxygen concentration estimation device according to the present embodiment. 4 is a flowchart illustrating an example of an oxygen concentration estimation process according to the first embodiment. 10 is a flowchart illustrating an example of an oxygen concentration estimation process according to a second embodiment.

One embodiment of the present invention will be described below. The following describes a heating furnace control system 1 that includes an oxygen concentration estimation device 10 that estimates the oxygen concentration in a heating furnace.

<Overall configuration of heating furnace control system 1>
An example of the overall configuration of a heating furnace control system 1 according to this embodiment is shown in Fig. 1. As shown in Fig. 1, the heating furnace control system 1 according to this embodiment includes an oxygen concentration estimation device 10, a control device 20, an adjustment device 30, a heating furnace 40, an oxygen concentration meter 50, and a peripheral sensor 60.

The heating furnace 40 is, for example, a facility or device used in various plants (petrochemical plants, various manufacturing plants, etc.), which consumes fuel and oxygen to heat a heated object (e.g., a fluid such as oil) through furnace combustion. However, the heating furnace 40 is not limited to these and may also be, for example, a boiler or incinerator.

The oxygen concentration meter 50 is a device that measures the oxygen concentration inside the heating furnace 40. In the following, it is assumed that a laser oxygen concentration meter ₅₀₁ and a zirconia oxygen concentration meter ₅₀₂ are installed in the heating furnace 40 as the oxygen concentration meter 50. However, the zirconia oxygen concentration meter ₅₀₂ does not have to be installed. In the following, it is also assumed that the oxygen concentration meter 50 (the laser oxygen concentration meter ₅₀₁ and the zirconia oxygen concentration meter ₅₀₂ ) will be removed from the heating furnace 40 after the performance data described below has been created and saved.

The oxygen concentration meter 50 also transmits the measurement value (hereinafter also referred to as the oxygen concentration measurement value) measured at each time t to the oxygen concentration estimation device 10. Hereinafter, the oxygen concentration measurement value measured by the laser oxygen concentration meter _50-1 at time t is defined _{as y1} (t), and the oxygen concentration measurement value measured by the zirconia oxygen concentration meter _50-2 is defined as _y2 (t).

The laser oximeter ₅₀₁ is an oximeter that uses laser absorption spectroscopy. The laser oximeter ₅₀₁ has a light-emitting unit that emits laser light and a light-receiving unit that receives the laser light, and measures the oxygen concentration from the amount of laser light lost between the light-emitting unit and the light-receiving unit.

The peripheral sensors 60 are various devices that measure various physical quantities around or in the vicinity of the heating furnace 40 and various physical quantities of equipment related to the heating furnace 40. The peripheral sensors 60 transmit measurement values (hereinafter also referred to as physical quantity measurement values) measured at each time t to the oxygen concentration estimation device 10. Hereinafter, the total number of peripheral sensors 60 is n, and when the peripheral sensors 60 are not distinguished from one another, they are referred to as "peripheral sensors 60." When the peripheral sensors 60 are distinguished from one another, they are referred to as "peripheral sensors 60 ₁ ,""peripheral sensors 60 ₂ ," ..., "peripheral sensors 60 _n ." Furthermore, for i = 1, 2, ..., n, the physical quantity measurement value measured by the peripheral sensor 60 _i at time t is denoted as x _i (t). Examples of physical quantities measured by the peripheral sensors 60 include the temperature inside the heating furnace 40, the temperature of the heated object, the pressure inside the heating furnace 40, the fuel flow rate, the heated object flow rate, and the exhaust gas flow rate. However, these physical quantities are merely examples, and the physical quantities measured by the surrounding sensor 60 are not limited to these. For example, the outside temperature, wind speed, etc. may be measured as the physical quantity.

At each time t for which actual data is to be created and saved, the oxygen concentration estimation device 10 receives oxygen concentration measurement values _y1 (t) and _y2 (t) from each oxygen concentration meter 50, and also receives physical quantity measurement values _x1 (t), _x2 (t), ..., _xn (t) from each peripheral sensor 60, and creates and saves actual data from these measurement values.

Furthermore, the oxygen concentration estimation device 10 receives physical quantity measurement values x ₁ (t), x ₂ (t), . . . , x _n (t) from the surrounding sensors 60 at each time t for which the oxygen concentration is to be estimated, and calculates an oxygen concentration estimation value at the time t from these measurement values.

and output it to the control device 20. In the following text of the specification, the symbol "^" indicating that it is an estimated value will be written immediately after y, and the estimated oxygen concentration at time t will be represented as "y^(t)."

In other words, the oxygen concentration estimation device 10 functions as a software sensor that calculates an oxygen concentration estimation value y^(t) from physical quantity measurement values _x1 (t), _x2 (t), ..., _xn (t) at each time t for which the oxygen concentration is to be estimated. Hereinafter, the model for estimating the oxygen concentration will be referred to as the oxygen concentration estimation model and represented by f. In addition, in the following, the time t for which the oxygen concentration is to be estimated will be represented as "t'" to distinguish it from the time t for which actual data is to be created and saved.

The control device 20 calculates the optimal operation amount for the adjustment device 30 based on the oxygen concentration estimation value y^(t') received from the oxygen concentration estimation device 10, and outputs a control command including this operation amount to the adjustment device 30. The adjustment device 30 is an air damper (or air valve) that adjusts the amount of air (oxygen) flowing into the heating furnace 40, and a fuel valve that adjusts the amount of fuel flowing into the heating furnace 40. The operation amount of the air damper is its opening/closing angle (or opening/closing amount), while the operation amount of the air valve is its opening/closing angle (or opening/closing amount).

The control device 20 may output a control command to both the air damper and the fuel valve, or to one of them (i.e., either the air damper or the fuel valve). In particular, for example, if the fuel flow rate is constant while the heating furnace 40 is operating, the control device 20 may output a control command only to the air damper.

Here, the heat loss and thermal efficiency within the heating furnace 40 fluctuate depending on the amount of fuel and air flowing into the heating furnace 40, which in turn causes the oxygen concentration within the heating furnace 40 to fluctuate. Therefore, the control device 20 uses known control technology to calculate the manipulated variable for the adjustment device 30 so that the heat loss and thermal efficiency within the heating furnace 40 are optimized, and outputs this manipulated variable to the adjustment device 30. There are various known control technologies such as this, including the technology described in Japanese Patent No. 6135831, for example.

In general, the relationship between excess air ratio and heat loss, and the relationship between excess air ratio and thermal efficiency, are shown in Figure 2. In Figure 2, the vertical axis represents heat loss or thermal efficiency, and the horizontal axis represents excess air ratio. The excess air ratio is the ratio of the amount of air actually flowing into the heating furnace 40 to the theoretical amount of air required for combustion per unit of fuel. In Figure 2, line 1001 represents heat loss due to excess air, and curve 1002 represents heat loss due to incomplete combustion. As shown by line 1001, the greater the excess air ratio is above 1, the greater the heat loss, as heat escapes and heats the excess air. On the other hand, as shown by curve 1002, a small excess air ratio causes incomplete combustion, resulting in greater heat loss due to CO generation, and when it exceeds a certain threshold, smoke is generated. Also in Figure 2, curve 2001 represents the thermal efficiency of the heating furnace 40. As shown by curve 2001, thermal efficiency is greatest in region D, which includes the excess air ratio, where heat loss due to excess air and heat loss due to incomplete combustion are approximately the same, and decreases as the excess air ratio moves away from region D. Therefore, control device 20 calculates the manipulated variable for adjustment device 30 so that heat loss and thermal efficiency are within region D, and outputs a control command including that manipulated variable to adjustment device 30.

Note that the overall configuration of the heating furnace control system 1 shown in Figure 1 is an example and is not limited to this. For example, it may include various facilities, equipment, devices, etc. that are not shown.

<Example of installation of oxygen concentration meter 50 and peripheral sensor 60 in heating furnace 40>
An example of the installation of the oxygen concentration meter 50 and the peripheral sensor 60 relative to the heating furnace 40 is shown in Fig. 3. In the example shown in Fig. 3, a laser oxygen concentration meter ₅₀₁ and a zirconia oxygen concentration meter ₅₀₂ are installed. In addition, in the example shown in Fig. 3, an exhaust gas flow meter ₆₀₁ , a heated material flow meter ₆₀₂ , a fuel supply flow meter ₆₀₃ , an in-furnace pressure meter ₆₀₄ , an in-furnace thermometer ₆₀₅ , an in-furnace thermometer ₆₀₆ , and a heated material thermometer ₆₀₇ are installed as peripheral sensors 60.

It goes without saying that the installation example of the oxygen concentration meter 50 and peripheral sensor 60 shown in Figure 3 is just one example and is not limited to this.

<Relationship between air-fuel ratio and oxygen concentration measurement value>
The air-fuel ratio is the ratio of air to fuel (air/fuel). The relationship between the air-fuel ratio and the measured oxygen concentration is shown in FIG. 4. In FIG. 4, the vertical axis represents the air-fuel ratio or oxygen concentration, and the horizontal axis represents time. As shown in FIG. 4, when the air-fuel ratio falls below a certain value (i.e., when the amount of oxygen decreases), the measured value of the zirconia oxygen concentration meter ₅₀₂ approaches zero, resulting in an erroneous measurement. This is because when the oxygen concentration becomes too low, incomplete combustion occurs, resulting in an oxidation reaction in the zirconia element, catalyzed by carbon monoxide and unburned combustible fuel. On the other hand, the laser oxygen concentration meter ₅₀₁ does not experience such erroneous measurement.

Therefore, in this embodiment, when the measurement value of the zirconia oxygen analyzer ₅₀₂ is near 0 (that is, when | _y2 (t)|<ε for a certain small value ε>0), the measurement value of the laser oxygen analyzer ₅₀₁ is used as training data (correct data) to create the oxygen concentration estimation model f. Note that when the measurement value of the zirconia oxygen analyzer ₅₀₂ is not near 0, either _y1 (t) or _y2 (t) may be used as training data, or the average value of _y1 (t) and _y2 (t) may be used as training data.

<Hardware configuration of oxygen concentration estimation device 10>
An example of the hardware configuration of the oxygen concentration estimation device 10 according to this embodiment is shown in Fig. 5. As shown in Fig. 5, the oxygen concentration estimation device 10 according to this embodiment includes an input device 101, a display device 102, an external I/F 103, a communication I/F 104, a processor 105, and a memory device 106. These pieces of hardware are connected to each other via a bus 107 so as to be able to communicate with each other.

The input device 101 is, for example, a keyboard, a mouse, a touch panel, various physical buttons, etc. The display device 102 is, for example, a display, a display panel, etc. Note that the oxygen concentration estimation device 10 does not necessarily have to have at least one of the input device 101 and the display device 102, for example.

The external I/F 103 is an interface with external devices such as a recording medium 103a. Examples of recording media 103a include a CD (Compact Disc), a DVD (Digital Versatile Disk), an SD memory card (Secure Digital memory card), and a USB (Universal Serial Bus) memory card.

The communication I/F 104 is an interface for connecting the oxygen concentration estimation device 10 to a communication network. The processor 105 is, for example, a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), or other computing device. The memory device 106 is, for example, a SSD (Solid State Drive), RAM (Random Access Memory), ROM (Read Only Memory), flash memory, or other storage device.

Note that the hardware configuration shown in FIG. 5 is an example, and the oxygen concentration estimation device 10 may have other hardware configurations. For example, the oxygen concentration estimation device 10 may have multiple processors 105 or multiple memory devices 106, or may have various types of hardware other than the hardware shown in the figure.

<Functional configuration of oxygen concentration estimation device 10>
An example of the functional configuration of the oxygen concentration estimation device 10 according to this embodiment is shown in Fig. 6. As shown in Fig. 6, the oxygen concentration estimation device 10 according to this embodiment includes a performance data creation unit 201, a model creation processing unit 202, and an oxygen concentration estimation unit 203. These units are realized, for example, by a processor 105 executing one or more programs installed in the oxygen concentration estimation device 10. The oxygen concentration estimation device 10 according to this embodiment also includes a performance data storage unit 204. The storage unit 204 is realized, for example, by the memory device 106, but may also be realized by a database server connected via a communications network.

At each time t for which performance data is to be created and stored, the performance data creation unit 201 receives oxygen concentration measurement values _y1 (t) and _y2 (t) from each oxygen concentration meter 50 and physical quantity measurement values _x1 (t), _x2 (t), ..., _xn (t) from each peripheral sensor 60, creates performance data {y(t), _x1 (t), _x2 (t), ..., _xn (t)} from these measurement values, and stores it in the performance data storage unit 204. Here, for a certain small value ε>0, at time t where | _y2 (t)|<ε, y(t)= _y1 (t). On the other hand, at other times t, y(t) may be either y(t) = _y1 (t) or y(t) = _y2 (t), or y(t) = ( _y1 (t) + _y2 (t))/2, or y(t) = min( _y1 (t), _y2 (t)) or y(t) = max( _y1 (t), y2(t ₎ ), etc. Note that y(t) is training data.

In the following, it is assumed that the performance data is z(t):=(y(t), _x1 (t), _x2 (t), ..., _xn (t)), and that the performance data set {z(t)| _t∈T1 } is stored in the performance data storage unit 204. Here, _T1 is a set of times for which performance data is to be created and saved. For example, when performance data is to be created and saved from times _t1 to _t2 , it is expressed as _T1 = {t| _t1 ≦t≦ _t2 }.

The model creation processing unit 202 creates an oxygen concentration estimation model f from the actual data set {z(t)|t∈T ₁ }. Here, the model creation processing unit 202 includes a data acquisition unit 211 and a model creation unit 212. The data acquisition unit 211 acquires the actual data set {z(t)|t∈T ₂ } used to create the oxygen concentration estimation model f from the actual data storage unit 204. Here, T ₂ is a set of times of the actual data used to create the oxygen concentration estimation model f, and T ₂ ⊆ _{T 1.} The model creation unit 212 creates the oxygen concentration estimation model f using the actual data set {z(t)|t∈T ₂ } acquired by the data acquisition unit 211. Note that the oxygen concentration estimation model f is, for example, a statistical model or a machine learning model, and has a parameter θ to be learned.

The oxygen concentration estimation unit 203 receives physical quantity measurements _x1 (t'), _x2 (t'), ..., xn(t') from the surrounding sensors 60 at each time t' for which the oxygen concentration is to be estimated, and calculates an oxygen concentration estimation value y _^ (t') from these physical quantity measurements _x1 (t'), _x2 (t'), ..., _xn (t') and the oxygen concentration estimation model f. That is, the oxygen concentration estimation unit 203 calculates the oxygen concentration estimation value y^(t') at the time t' using the physical quantity measurements _x1 (t'), _x2 (t'), ..., _xn (t') as explanatory variables and the oxygen concentration as a response variable, using y^(t') = f _{( x1} (t'), _x2 (t'), ..., xn(t')). The oxygen concentration estimation value y^(t') is output to the control device 20.

<Oxygen Concentration Estimation Process (Example 1)>
The oxygen concentration estimation process in the first embodiment will be described below with reference to Fig. 7. In the following, it is assumed that a performance data set {z(t)|tεT ₁ } for a certain past period T ₁ is stored in the performance data storage unit 204. Steps S101 and S102 in Fig. 7 are model creation processes that are performed in advance. Meanwhile, steps S103 and S104 are repeatedly performed for each time t' (i.e., the current time t') at which the oxygen concentration is to be estimated.

The data acquisition unit 211 of the model creation processing unit 202 acquires the actual data set {z(t)|t∈T2} used to create the oxygen concentration estimation model f from the actual _data storage unit 204 (step S101). Here, the data acquisition unit 211 may acquire the actual data set {z(t)| _t∈T1 } itself stored in the actual data storage unit 204 as the actual data set {z(t)| _t∈T2 }, or may acquire an actual data set composed of some of the actual data as the actual data set {z(t)| _t∈T2 }.

The model creation unit 212 of the model creation processing unit 202 creates an oxygen concentration estimation model f using the actual data set {z(t)| _t∈T2 } acquired in step S101 (step S102). That is, the model creation unit 212 uses the actual data set {z(t)| _t∈T2 } as a learning data set and learns the parameter θ so as to reduce the error between the oxygen concentration estimated value y^(t)=f( _x1 (t), _x2 (t), ..., _xn (t)) at time t and y(t). Here, the oxygen concentration estimation model f can be a linear model such as a multiple regression model, or a nonlinear model such as support vector regression (SVR) or a neural network. Generally, the relationship between the oxygen concentration and the furnace temperature, etc. changes continuously in the heating furnace 40, and the characteristics of each physical quantity in the heating furnace 40 change depending on the oxygen concentration. Therefore, a linear model often results in low estimation accuracy. For this reason, it is particularly preferable to employ a nonlinear model as the oxygen concentration estimation model f.

The oxygen concentration estimation unit 203 receives the physical quantity measurement values x ₁ (t'), x ₂ (t'), . . . , x _n (t') at the current time t' for which oxygen concentrations are to be estimated (step S103).

The oxygen concentration estimation unit 203 calculates an oxygen concentration estimate y^(t') at the current time t' from the physical quantity measurements _x1 (t'), _x2 (t'), ..., _xn (t') received in step S103 and the oxygen concentration estimation model f (step S104). That is, the oxygen concentration estimation unit 203 calculates the oxygen concentration estimate y^(t') at the current time t' using y^(t') = f _{( x1} (t'), _x2 (t'), ..., xn(t')). The oxygen concentration estimate y^(t') is output to the control device 20, which controls the manipulated variable for the adjustment device 30. In this way, the oxygen concentration in the heating furnace 40 is controlled.

<Oxygen concentration estimation process (Example 2)>
The oxygen concentration estimation process in the second embodiment will be described below with reference to Fig. 8. In the following, it is assumed that a performance data set {z(t)|tεT ₁ } for a certain past period T ₁ is stored in the performance data storage unit 204. Note that steps S201 to S204 in Fig. 8 are repeatedly performed for each time t' (i.e., the current time t') at which the oxygen concentration is to be estimated.

The oxygen concentration estimation unit 203 receives physical quantity measurement values x ₁ (t'), x ₂ (t'), . . . , x _n (t') at the current time t' for which oxygen concentrations are to be estimated (step S201).

The data acquisition unit 211 of the model creation processing unit 202 acquires the actual data set {z(t)| _t∈T2 } used to create the oxygen concentration estimation model f from the actual data storage unit 204 (step S202). Here, the data acquisition unit 211 acquires a predetermined number of actual data from the actual data storage unit 204 in order of proximity to the physical quantity measurement values _x1 (t'), _x2 (t'), ..., _xn (t') received in step S201 above, and defines the actual data set consisting of these acquired actual data as the actual data set {z(t)| _t∈T2 }.

Specifically, the distance between the physical quantity measurement values x ₁ (t'), x ₂ (t'), ..., x _n (t') and the performance data z(t) = (y(t), x ₁ (t), x ₂ (t), ..., x _n (t)) is defined as d(t', t) = ((x ₁ (t') - x ₁ (t)) ² + (x ₂ (t') - x ₂ (t)) ² + ... + (x _n (t') - x _n (t)) ² ) ^1/2 . N (e.g., 100) pieces of z(t) are then obtained from the performance data storage unit 204 in order of smallest distance, and a performance data set consisting of these obtained performance data z(t) is defined as a performance data set {z(t)|t∈T ₂ }. Although the Euclidean distance is used as the distance d in the above, a distance other than the Euclidean distance may be used.

The model creation unit 212 of the model creation processing unit 202 creates an oxygen concentration estimation model f using the actual data set {z(t)|t∈T ₂ } acquired in step S202 (step S203). That is, the model creation unit 212 uses the actual data set {z(t)|t∈T ₂ } as a learning data set and learns the parameter θ so as to reduce the error between the oxygen concentration estimated value y^(t)=f(x ₁ (t), x ₂ (t), ..., x _n (t)) at time t and y(t). Here, as the oxygen concentration estimation model f, for example, a local linear model such as JIT-PLS (Just-in-time PLS) or LW-PLS (Locally-Weighted PLS) can be adopted. Generally, the relationship between the oxygen concentration and the temperature inside the heating furnace 40 changes continuously, and the characteristics of each physical quantity inside the heating furnace 40 change depending on the oxygen concentration, so that a (global) linear model often results in low estimation accuracy. For this reason, it is preferable to adopt a linear model (local linear model) that uses local learning data as the oxygen concentration estimation model f.

The oxygen concentration estimation unit 203 calculates an oxygen concentration estimate y^(t') at the current time t' from the physical quantity measurement values _x1 (t'), _x2 (t'), ..., _xn (t') received in step S201 and the oxygen concentration estimation model f created in step S203 (step S204). That is, the oxygen concentration estimation unit 203 calculates the oxygen concentration estimate y^(t') at the current time t' using y^(t') = f _{( x1} (t'), _x2 (t'), ..., xn(t')). The oxygen concentration estimate y^(t') is output to the control device 20, which controls the manipulated variable for the adjustment device 30. In this way, the oxygen concentration in the heating furnace 40 is controlled.

<Modification>
In the above embodiment, the oxygen concentration meter 50 is removed after the performance data is created and saved, but it is also possible to remove, for example, only the laser oxygen concentration meter ₅₀₁ and not the zirconia oxygen concentration meter _502. In this case, the oxygen concentration measurement value _y2 (t') of the zirconia oxygen concentration meter ₅₀₂ may also be used as an explanatory variable.

Specifically, the actual data is defined as z(t):=(y(t)= _y1 (t), _y2 (t), _x1 (t), _x2 (t), ..., _xn (t)), and an oxygen concentration estimation model f is created using y(t) as the objective variable and _y2 (t), _x1 (t), _x2 (t), ..., _xn (t) as explanatory variables. When calculating the oxygen concentration estimate, the oxygen concentration estimate y^(t') at the current time t' is calculated using y^(t')=f _{( y2} (t'), _x1 (t'), _x2 (t'), ..., xn(t')). In this case, for example, a model g such that y^(t) = g(x ₁ (t), x ₂ (t), ..., x _n (t)) is created in advance, and if |y ₂ (t')| < ε, y^(t') = g(x ₁ (t'), x ₂ (t'), ..., x _n (t')) is calculated, and y^(t') = f(y ₂ (t'), x ₁ (t'), x ₂ (t'), ..., x _n (t')) is calculated only if |y ₂ (t')| ≥ ε.

<Summary>
As described above, in the heating furnace control system 1 according to this embodiment, an oxygen concentration estimation model f is created using training data measured by the laser oxygen analyzer ₅₀₁ , which can stably measure oxygen concentrations even at low oxygen concentrations, and the oxygen concentration in the heating furnace 40 is then estimated using this oxygen concentration estimation model f. As a result, instead of the zirconia oxygen analyzer ₅₀₂ , which may erroneously measure oxygen concentrations at low oxygen concentrations, the oxygen concentration estimation model f can accurately estimate the oxygen concentration even at low oxygen concentrations, making it possible to achieve optimal control of the heating furnace 40 that achieves both energy conservation and pollution prevention.

Moreover, after the oxygen concentration estimation model f is created, the laser oxygen analyzer ₅₀₁ can be removed from the heating furnace 40, so that it is possible to install the laser oxygen analyzer ₅₀₁ in another heating furnace 40 and create an oxygen concentration estimation model f in the same way. Therefore, there is no need to prepare a laser oxygen analyzer ₅₀₁ , which is generally expensive, for each heating furnace 40, and the oxygen concentration estimation model f can be created relatively inexpensively.

The present invention is not limited to the specifically disclosed embodiments above, and various modifications, alterations, and combinations with known technologies are possible without departing from the scope of the claims.

REFERENCE SIGNS LIST 1 Heating furnace control system 10 Oxygen concentration estimation device 20 Control device 30 Adjustment device 40 Heating furnace 50 Oxygen concentration meter 60 Periphery sensor 101 Input device 102 Display device 103 External I/F
103a Recording medium 104 Communication I/F
105 Processor 106 Memory device 107 Bus 201 Performance data creation unit 202 Model creation processing unit 203 Oxygen concentration estimation unit 204 Performance data storage unit 211 Data acquisition unit 212 Model creation unit

Claims (5)

An oxygen concentration estimation device for estimating an oxygen concentration in a heating furnace,
a model creation unit that creates a model for estimating the oxygen concentration in the heating furnace using, as learning data, an oxygen concentration measurement value y(t) in the heating furnace at each time t within a predetermined period in the past and a predetermined number n (n is a predetermined integer of 1 or more) of physical quantity measurement values x ₁ (t), x ₂ (t), ..., x _n (t) related to the heating furnace;
an oxygen concentration estimation unit that calculates an estimate of the oxygen concentration in the heating furnace using the n physical quantity measurement values x ₁ (t'), x ₂ (t'), ..., x _n (t') at time t', which is a target of oxygen concentration estimation, and the model;
and
the heating furnace includes a laser oxygen concentration meter that is installed so as to be removable after the learning data is created;
The oxygen concentration measurement value y(t) is a measurement value y ₁ (t) of the oxygen concentration inside the heating furnace measured by the laser oxygen analyzer . The heating furnace is further equipped with a zirconia oxygen concentration meter,
The oxygen concentration in the heating furnace measured by the zirconia oxygen concentration meter is defined as y ₂ (t),
If |y ₂ (t)|<ε, for some predetermined value ε>0, then y(t)=y ₁ (t);
_2. The concentration estimation device according to claim 1, wherein, when | _y2 (t)|≧ε, any of y(t)= _y1 (t), y(t)= _y2 (t), y(t)=( _y1 (t)+ _y2 (t))/2, y(t)=min( _y1 (t), _y2 (t)), or y( t )=max( _y1 (t), y2(t)). The model creation unit
calculating distances d(t', t) between the n physical quantity measurement values _x1 (t'), _x2 (t'), ..., _xn (t') at time t' for which the oxygen concentration is to be estimated and the n physical quantity measurement values _x1 (t), _x2 (t), ..., _xn (t) at each time t within a predetermined period in the past;
3. The concentration estimation device according to claim 1, wherein the model is created using the oxygen concentration measurement values y(t) and the n physical quantity measurement values x ₁ (t), x ₂ (t), ..., x _n (t) corresponding to a predetermined number of t's in ascending order of the distance d(t ', t ) as learning data. An oxygen concentration estimation device for estimating the oxygen concentration in a heating furnace,
a model creation step of creating a model for estimating the oxygen concentration in the heating furnace using, as learning data, the oxygen concentration measurement value y(t) in the heating furnace at each time t within a predetermined period in the past and a predetermined number n (n is a predetermined integer of 1 or more) of physical quantity measurement values x ₁ (t), x ₂ (t), ..., x _n (t) related to the heating furnace;
an oxygen concentration estimation procedure for calculating an estimated value of the oxygen concentration in the heating furnace using the n physical quantity measurement values x ₁ (t'), x ₂ (t'), ..., x _n (t') at time t' for which the oxygen concentration is to be estimated and the model;
Run
the heating furnace includes a laser oxygen concentration meter that is installed so as to be removable after the learning data is created;
The concentration estimation method, wherein the oxygen concentration measurement value y(t) is a measurement value y ₁ (t) of the oxygen concentration inside the heating furnace measured by the laser oxygen analyzer . An oxygen concentration estimation device that estimates the oxygen concentration in a heating furnace,
a model creation step of creating a model for estimating the oxygen concentration in the heating furnace using, as learning data, the oxygen concentration measurement value y(t) in the heating furnace at each time t within a predetermined period in the past and a predetermined number n (n is a predetermined integer of 1 or more) of physical quantity measurement values x ₁ (t), x ₂ (t), ..., x _n (t) related to the heating furnace;
an oxygen concentration estimation procedure for calculating an estimated value of the oxygen concentration in the heating furnace using the n physical quantity measurement values x ₁ (t'), x ₂ (t'), ..., x _n (t') at time t' for which the oxygen concentration is to be estimated and the model;
Execute
the heating furnace includes a laser oxygen concentration meter that is installed so as to be removable after the learning data is created;
The oxygen concentration measurement value y(t) is a measurement value y ₁ (t) of the oxygen concentration inside the heating furnace measured by the laser oxygen analyzer .