JP7421211B2 - Pouring status estimation system - Google Patents

Description

The present invention relates to a system for estimating the amount of molten metal poured in a molten metal pouring operation.

2. Description of the Related Art Conventionally, there has been an automatic pouring device that controls the flow rate of pouring molten metal in a pouring operation of pouring a molten material into a mold. For example, Patent Document 1 discloses that the input voltage applied to a servo motor that tilts a ladle and the weight of molten metal in the ladle are measured, and the pouring flow rate is estimated using an exponential decay type observer based on an extended Kalman filter. However, an automatic pouring device that controls the pouring flow rate has been disclosed.

Japanese Patent Application Publication No. 2008-290148

On the other hand, in manual pouring work, there has been a large difference in the quality of the castings produced by the pouring work between an operator who is proficient in pouring work and an operator who is not. Since the quality of the product of pouring work depends on the pouring conditions during the work, there was a need to accurately estimate the pouring conditions and control the amount of poured metal based on the estimated pouring conditions. .

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide an estimation system that makes it possible to accurately estimate the pouring state in manual pouring work.

According to the present invention, there is provided a system for estimating a pouring state in a pouring operation in which a ladle is tilted to pour a molten material into a mold, and the system includes an angle acquisition section, a weight acquisition section, and a ladle internal shape acquisition section. and a calculation unit, the angle acquisition unit acquires the tilt angle of the ladle in the pouring operation, the weight acquisition unit acquires the weight of the ladle in the pouring operation, and the angle acquisition unit acquires the tilt angle of the ladle in the pouring operation. The ladle internal shape acquisition unit acquires data regarding the internal shape of the ladle, and the calculation unit calculates the amount of water from the ladle based on the data regarding the tilt angle, the weight, and the internal shape. An estimation system is provided that calculates an estimated value of the amount of hot water poured.

With this configuration, the estimated value of the amount of poured molten metal is calculated based on the tilt angle and weight of the ladle obtained in real time during pouring work, and data about the internal shape of the ladle, which eliminates manual manual work. It becomes possible to accurately estimate the pouring state during pouring work.

Below, various embodiments of the present invention will be illustrated. The embodiments shown below can be combined with each other. Furthermore, each feature independently constitutes the invention.

Preferably, the calculation unit calculates the amount of change in the volume of a portion of the melt located above the lower end of the tap outlet of the ladle, with respect to changes in the vertical height and the tilt angle of the portion, respectively. and calculate the estimated value of the amount of molten metal poured.
Preferably, the apparatus further includes a storage section, and the ladle internal shape acquisition section acquires data about the internal shape of the ladle stored in the storage section.
Preferably, the ladle further includes a measuring device that measures data about the internal shape of the ladle, and the ladle internal shape acquisition section acquires data about the internal shape of the ladle measured by the measuring device.
Preferably, the storage unit stores data regarding the internal shape of the ladle measured by the measuring device.
Preferably, the apparatus further includes an imaging section capable of detecting infrared rays, and the imaging section images the position where the melt is poured into the mold during the pouring operation.
Preferably, the imaging unit is arranged in front of the ladle in the tilting direction of the ladle and on the side of the ladle in a direction perpendicular to the tilting direction in plan view.
Preferably, the evaluation unit further includes an evaluation unit, and the evaluation unit compares the reference value, which is the estimated value in the pouring operation as a standard, with the estimated value in the pouring operation as the evaluation target, and determines the evaluation target. Evaluate the pouring work as follows.
Preferably, the evaluation unit performs the evaluation by calculating a correlation coefficient between the reference value and the estimated value.
Preferably, the evaluation unit performs the evaluation by calculating a Manhattan distance between the reference value and the estimated value using a dynamic time warping method.
Preferably, the evaluation unit performs the evaluation by calculating a Manhattan distance between the reference value and the estimated value.
Preferably, the evaluation unit performs the evaluation by calculating a Euclidean distance between the reference value and the estimated value.

According to another aspect of the present invention, there is provided an estimation method comprising an angle acquisition step, a weight acquisition step, a ladle internal shape acquisition step, and a calculation step, wherein in the angle acquisition step, the The tilt angle of the ladle is acquired, the weight acquisition step acquires the weight of the ladle in the pouring operation, and the ladle internal shape acquisition step acquires data about the internal shape of the ladle. In the calculation step, an estimation method is provided, in which an estimated value of the amount of molten metal poured from the ladle is calculated based on data about the tilt angle, the weight, and the internal shape.

1 is an overall perspective view of an estimation system P according to a first embodiment of the present invention. It is a functional block diagram showing the functions of the estimation system P according to the first embodiment. It is a figure explaining the measurement of the injection position of a melt. FIG. 4A is a cross-sectional view of the ladle 1 showing parameters in generating the pouring flow rate model. FIG. 4B is a perspective view of the ladle 1 showing parameters in generating the pouring flow rate model. FIG. 5A is a diagram illustrating an analysis of the internal shape of the ladle 1 when the tilt angle is θ1. FIG. 5B is a diagram illustrating an analysis of the internal shape of the ladle 1 when the tilt angle is θ2. It is a graph showing measurement results and estimation results in Examples. It is a graph (f) in FIG. 6. It is a graph which shows the measurement result of the injection position of the melt in an Example. It is a functional block diagram showing the functions of an estimation system P according to a second embodiment. It is a figure which shows an example of the evaluation result in the evaluation part 11e.

<1. Embodiment>
(1.1. Overall composition)
FIG. 1 shows an overall diagram of a pouring amount estimation system P according to an embodiment. The estimation system P includes a ladle 1, a mold 2, an attitude angle sensor 3, a load cell 4, an imaging unit 5, and an information processing device 10. Each configuration will be explained in detail below. Note that in the present disclosure, "pouring amount" is a general term for the amount of melt flowing out of the ladle during pouring work, such as pouring weight, pouring volume, or pouring flow rate.

The ladle 1 is formed, for example, in a cylindrical shape with a bottom, and is configured to be able to accommodate a melt therein. At the front of the ladle 1, a tap hole 1a is provided which is a spout for pouring the molten material. The ladle 1 further includes a handle 1b for tilting the ladle 1 when pouring molten material, and a handle 1c for hooking a hooked wire 9 suspended from above. The operator tilts the ladle 1 to a desired angle by operating the handle 1b.

The mold 2 is formed into a rectangular parallelepiped shape, for example. An injection port 2a for injecting the melt is formed on the upper surface of the mold 2. The molten material poured from the injection port 2a flows through a flow path formed inside the mold 2 and solidifies, thereby forming a desired shaped object inside the mold 2.

The attitude angle sensor 3 is attached to the rear of the ladle 1 (that is, on the opposite side of the tap hole 1a). A shielding member 6 is provided between the attitude angle sensor 3 and the ladle 1 to shield heat from the molten material housed inside the ladle 1. The attitude angle sensor 3 measures the tilting angle θ and the tilting angular velocity ω of the ladle 1.

The load cell 4 is provided above the ladle 1 and measures the weight of the ladle 1. The weight of the ladle 1 also includes the weight of the melt contained therein. A shielding material 6 is provided between the load cell 4 and the ladle 1 to shield heat from the molten material housed inside the ladle 1.

The imaging unit 5 images the injection position, which is the position where the molten material is injected into the mold 2 during pouring work. Since the molten material emits light at a very high temperature, it is preferable that the imaging unit 5 uses a thermography capable of detecting infrared rays.

More preferably, the imaging unit 5 is disposed in front of the ladle 1 in the tilting direction and on the side of the ladle 1 in a direction perpendicular to the tilting direction in plan view. By arranging it in this way, it becomes possible to more accurately image the injection position of the melt.

The information processing device 10 calculates an estimated value of the amount of molten material poured from the ladle 1 based on data received from the attitude angle sensor 3 and the load cell 4. The configuration and processing contents of the information processing device 10 will be described in detail later.

(1.2. Configuration of information processing device 10)
The configuration of the information processing device 10 will be described with reference to FIG. 2. The information processing device 10 is realized by, for example, a PC (Personal Computer), and includes a control section 11 and a storage section 12.

(1.2.1. Control unit 11)
The control unit 11 is implemented by, for example, a CPU (Central Processing Unit), a microprocessor, a DSP (Digital Signal Processor), etc., and controls the overall operation of the information processing device 10. The control section 11 includes an angle acquisition section 11a, a weight acquisition section 11b, a ladle internal shape acquisition section 11c, and a calculation section 11d.

The angle acquisition unit 11a acquires the tilting angle θ and the tilting angular velocity ω of the ladle 1 measured by the attitude angle sensor 3. The weight acquisition unit 11b acquires the weight of the ladle 1 measured by the load cell 4. The ladle internal shape acquisition unit 11c acquires data about the internal shape of the ladle 1 stored in the storage unit 12.

The calculation unit 11d generates a pouring flow rate model regarding the pouring state of the molten material based on the data acquired by the angle acquisition unit 11a, the weight acquisition unit 11b, and the ladle internal shape acquisition unit 11c. The calculation unit 11d further calculates an estimated value of the pouring amount of the molten material based on the generated pouring flow rate model. Furthermore, the calculation unit 11d calculates the injection position of the melt based on the data captured by the imaging unit 5.

(1.2.2. Storage unit 12)
The storage unit 12 is realized by, for example, a RAM (Random Access Memory) or a DRAM (Dynamic Random Access Memory), and is used as a work area or the like when the control unit 11 executes processing based on various programs.

The storage unit 12 may further include a nonvolatile memory such as a ROM (Read Only Memory) or a HDD (Hard Disk Drive), and stores programs and various data used in the processing of the control unit 11. As an example, the storage unit 12 stores data about the internal shape of the ladle 1 measured by a measuring device 13 including a 3D scanner or the like.

(1.3. Calculation of injection position)
Calculation of the injection position will be described with reference to FIG. 3. The calculation unit 11d performs edge processing on the image data acquired from the imaging unit 5, and determines the position Pn of the end point of the melt near the ladle 1 at the injection port 2a and the end point of the end point far from the ladle 1. Calculate position Pf. Then, the injection position Pp of the melt relative to the center of the injection port 2a is calculated using the following equation (1).

Here, Cn is the position of the end point of the injection port 2a closer to the ladle 1, and Cf is the position of the end point of the injection port 2a farther from the ladle 1. Since the molten material emits light at a very high temperature, it is preferable to use a thermography capable of detecting infrared rays as the imaging unit 5. When using a thermography as the imaging unit 5, the temperature display range of the thermography is set at room temperature, and the temperature difference between the sand and air that makes up the mold 2 causes a difference in gradation between pixels of the image data obtained. By performing edge processing, the positions Cn and Cf of the end points of the injection port 2a can be detected.

Since the end points obtained from the thermography in this way are pixel data, the injection position Xp of the melt relative to the center of the injection port 2a as an actual dimension is calculated using the following equation (2).

Here, D means the diameter of the injection port 2a. In this way, by executing equations (1) and (2) for each frame of image data acquired by the imaging unit 5, the injection position of the melt flowing out from the ladle 1 is calculated. Although FIG. 3 describes image data taken from the side of the ladle 1 as an example, similar processing is performed for image data taken from the front of the ladle 1. Note that as long as the injection position Xp of the melt can be obtained, an imaging device other than a thermography may be used as the imaging unit 5.

(1.4. Generation of pouring flow rate model)
With reference to FIGS. 4 and 5, the generation process of the pouring flow rate model will be described. Regarding the flow rate q(t) of the melt flowing out of the ladle 1, the following equations (3) and (4) are known.

Here, as shown in FIGS. 4A and 4B, Vr is the volume of the liquid (molten material) inside the ladle 1 above the lower end R of the tap 1a, and Vs is the volume of the liquid (molten material) inside the ladle 1 below the lower end R of the tap 1a. 1 Volume of liquid inside, h is height of liquid at tap 1a, Lf is width of tap 1a, hb is depth from liquid surface in ladle 1, c is flow coefficient, t is time, each means.

In equation (3), since the volume Vs depends on the tilt angle θ of the ladle 1, the following equation (5) holds true.

Furthermore, since the volume Vr depends on the tilting angle θ of the ladle 1 and the height h of the liquid, the following equation (6) holds true.

By substituting Equation (5) and Equation (6) into Equation (3), the following Equation (7) can be obtained as a pouring flow rate model.

In the pouring flow rate model shown by equation (7), it is necessary to calculate the amount of change X(θ,h) for the change in height h and Y(θ,h) for the change in tilt angle θ for the volume Vr. . Therefore, the calculation unit 11d calculates X(θ,h) and Y(θ,h) based on data regarding the internal shape of the ladle 1.

Specifically, as shown in FIG. 5A, the calculation unit 11d reads the internal shape data of the ladle 1 measured by the measuring device 13 using 3D CAD, and uses the API (Application Programming Interface) function of the 3D CAD. Using , the amount of volume change dV/dh in the ladle 1 with respect to the change in height h when the tilt angle is a predetermined value (θ=θ1 in the example of FIG. 5A) is calculated for each predetermined height Δh (for example, 10 cm). Calculate from bottom to top.

Further, as shown in FIG. 5B, the calculation unit 11d changes the value of the tilting angle θ (θ=θ2 in the example of FIG. 5B), and changes the volume change dV/in the ladle 1 with respect to the change in height h. dh is calculated sequentially from the bottom for each predetermined height Δh.

In this way, the ladle 1 responds to changes in height h with respect to θ at predetermined angles (for example, every 1 degree) in the range that the tilting angle θ can take (for example, 0°≦θ≦90°). X(θ, h) and Y(θ, h) are calculated by determining the volume change amount dV/dh within. Further, the calculation unit 11d calculates Vs for each tilt angle θ.

(1.5. Calculation of estimated value of pouring amount)
The calculation unit 11d performs an extended Kalman filter using the above-mentioned pouring flow rate model (Equation (7)), the tilting angle θ detected by the attitude angle sensor 3, the tilting angular velocity ω, and the weight of the ladle 1 measured by the load cell 4. By configuring this, an estimated value of the pouring amount of molten material that actually flows out from the ladle 1 is calculated.

When the response delay of the load cell 4 is expressed in a first-order delay system with respect to the weight _WL of the melted material flowing out from the ladle 1 detected by the load cell 4, the following formulas (8) and (9) hold true.

Here, Wo is the weight of the melt actually flowing out from the ladle 1, T _L is a time constant indicating the response characteristics of the load cell 4, and ρ is the density of the melt. The weight W _L of the melt that has flowed out from the ladle 1 detected by the load cell 4 is affected by the response delay of the load cell 4 relative to the weight Wo of the melt that has actually flowed out from the ladle 1 .

Using these parameters, an extended Kalman filter is constructed as shown in equations (10) and (11) below.

Here, the pouring flow rate model calculated as the equation (7) is adopted in the equation (11-2), and Ts means the sampling time. The height h of the liquid has the condition of the following equation (12).

θb in equation (12) is the tapping boundary angle (the ladle tilting angle at which the liquid flows out from the ladle), and using the tilting angle θ of ladle 1 and the volume Vs, it can be obtained as shown in equation (13) below. It will be done.

In equation (13), Wcini means the weight of the liquid in the ladle measured by the load cell 4 before the start of pouring. Furthermore, the Jacobian matrices A(x) and C(x) of the extended Kalman filter are set as the following equations (14) and (15).

Using the Kalman filter constructed in this way, the height h of the liquid at the outlet 1a, the pouring flow rate q, the pouring weight Wo excluding the load cell response delay, and the pouring weight W with the load cell response delay Estimate _L in real time. Note that the flow coefficient and liquid density may be identified using a well-known identification method (for example, see International Publication 2014/174977).

(1.6. Examples)
Embodiments of the present invention will be described with reference to FIGS. 6 to 8. In FIG. 6, graph (a) shows the tilting angular velocity ω of the ladle 1 measured by the attitude angle sensor 3, graph (b) shows the tilting angle θ of the ladle 1 measured by the attitude angle sensor 3, and graph (c) is the estimated liquid height h at the outlet 1a of the ladle 1, graph (d) is the estimated pouring (volume) flow rate q, graph (e) is the estimated pouring mass flow rate, graph ( f) is the estimated pouring weight.

As shown in an enlarged view in FIG. 7, the graph (f) in FIG. 6 includes the poured metal weight Wxp measured by the load cell 4, the estimated data W _L of the poured metal weight including the load cell response delay, and the load cell response delay Wxp. Estimated data Wo of pouring weight excluding .

As can be seen from the graph (f) in FIG. 6 and FIG. 7, the pouring weight Wxp, which is the data obtained from the load cell 4, contains noise and a delay in the response of the load cell, and this data is differentiated with respect to time. It can be seen that the pouring mass flow rate obtained by this method is not clear. As described above, in manual pouring work, the worker manually adjusts the heavy ladle and pours the metal, so noise is generated in the outputs of the attitude angle sensor 3 and the load cell 4, making it difficult to analyze the data.

On the other hand, the estimated value data using the extended Kalman filter in this embodiment (i.e., W _L and Wo in graph (f)) suppresses noise and response delay of the load cell, so the pouring (volume) flow rate is You can clearly understand the weight of poured metal.

In FIG. 8, graph (a) is the pouring (volume) flow rate, and corresponds to graph (d) in FIG. Graph (b) in Figure 8 shows the longitudinal direction of the molten metal flowing out of the ladle on the upper surface of the mold, which was estimated using equations (1) and (2) from the temperature distribution image of the thermograph installed on the side of the ladle. Graph (c) shows the injection position of the molten metal flowing out of the ladle on the upper surface of the mold, which is estimated by equations (1) and (2) from the temperature distribution image of the thermograph installed in front of the ladle. This is the injection position in the left and right direction.

In graphs (b) and (c), the dashed line indicates the edge of the sprue cup, and the * mark is the injection position of the outflowing molten metal. Further, graphs (b) and (c) show the injection position with the center of the sprue cup set at 0. As can be seen from FIG. 8, in this example, the pouring position is near the center of the sprue cup.

As described above, the pouring amount estimation system P in this embodiment includes the angle acquisition section 11a, the weight acquisition section 11b, the ladle internal shape acquisition section 11c, and the calculation section 11d. The angle acquisition unit 11a acquires the tilt angle θ of the ladle 1 during pouring work. The weight acquisition unit 11b acquires the weight of the ladle 1 during pouring work. The ladle internal shape acquisition unit 11c acquires data regarding the internal shape of the ladle 1. The calculation unit 11d calculates an estimated value of the flow rate of the molten material poured from the ladle, based on data regarding the tilt angle θ, the weight of the ladle 1, and the internal shape.

With such a configuration, it becomes possible to accurately estimate the pouring state in the manual pouring device.

Furthermore, the pouring amount estimation system P may further include a measuring device such as a 3D scanner in order to measure data regarding the internal shape of the ladle 1. With this configuration, the ladle shape data measured by a measuring device such as a 3D scanner is used to calculate the pouring flow rate model from ladle 1, and it is possible to prevent the ladle design data from being obtained. It is also possible to calculate an estimated value of the amount of molten metal poured into a molten metal pouring device (for example, a ladle or the like that has undergone repair work).

<2. Second embodiment>
A second embodiment of the present invention will be described with reference to FIGS. 9 and 10. As shown in FIG. 9, the pouring amount estimation system P according to the second embodiment further includes an evaluation section 11e in addition to the configuration of the first embodiment. Hereinafter, differences from the first embodiment will be mainly explained.

The evaluation unit 11e evaluates the pouring operation to be evaluated by comparing the estimated data of the pouring amount calculated by the calculation unit 11d with a reference value that is an estimated value in the pouring operation as a reference. In this way, by comparing the good pouring condition and clearly showing the evaluation results to the operator, it is possible to suppress variations in quality of the casting caused by the pouring operation and maintain it in a good condition.

Figure 10 shows an example of evaluating the pouring work. The horizontal axis shows the correlation coefficient, and the vertical axis shows the coordinate axis with the sign of the Manhattan distance reversed using the dynamic time warping method. The graph in FIG. 10 shows the evaluation results of three times of pouring work by an expert and six times of pouring work by an inexperienced person. Specifically, results T1 to T3 correspond to the evaluation results of the first to third operations by an expert in pouring work, and results N1 to N6 correspond to the evaluation results of the first to sixth operations by an inexperienced pouring worker. Respond to work evaluation results.

Here, when the evaluation result is placed at the upper right of the graph in FIG. 10 (that is, when it approaches results T1 to T3), it indicates that pouring similar to the pouring flow rate data of good casting quality is being performed. In other words, the closer the direction is to the arrow D1 or the arrow D2, the more similar the molten metal pouring data is to good pouring flow rate data.

Since the evaluation results T1 to T3 of the pouring work by experts are arranged in the upper right corner with little variation, it can be confirmed that the pouring work, which results in good casting quality, is performed stably. In addition, in the evaluation results N1 to N6 of the pouring work by inexperienced people, the number of times the pouring work increases is gradually approaching the upper right, and it can be confirmed that the pouring work is getting closer to being good.

In addition, when a worker who is not familiar with pouring work acquires pouring skills, the first step is to determine the size of the entire pouring flow rate so that the Manhattan distance using the dynamic time stretching method approaches zero. It is preferable to train the pouring operation so that it is equivalent to the data, and then finely adjust the pouring flow rate and train the pouring operation so that the correlation coefficient approaches 1. By training workers based on quantitative evaluation in this way, it becomes possible to have them efficiently learn the pouring work that produces quality castings.

<3. Other embodiments>
Although the embodiments have been described above, the technical scope of the present application is not limited to the above embodiments.

For example, in the embodiment described above, data regarding the internal shape of the ladle 1 is stored in the storage unit 12, but the data is not limited to this example. For example, it may be stored in an external storage medium such as a USB (Universal Serial Bus) memory, or it may be stored in another server such as a cloud server.

Further, in the above embodiment, the calculation unit 11d performs calculation using an extended Kalman filter, but the invention is not limited to this example, and the technical idea of the present disclosure can be applied even if other estimation methods are used. be able to.

Although various embodiments of the invention have been described, these are presented as examples and are not intended to limit the scope of the invention. The new embodiment can be implemented in various other forms, and various omissions, substitutions, and changes can be made without departing from the gist of the invention. The embodiment and its modifications are included within the scope and gist of the invention, and are included within the scope of the invention described in the claims and its equivalents.

1: Ladle 1a: Tap spout 1b: Handle 1c: Handle 2: Mold 2a: Filling port 3: Attitude angle sensor 4: Load cell 5: Imaging section 6: Shielding material 9: Wire with hook 10: Information processing device 11: Control Section 11a: Angle acquisition section 11b: Weight acquisition section 11c: Ladle internal shape acquisition section 11d: Calculation section 11e: Evaluation section 12: Storage section 13: Measuring device

Claims (13)

A system for estimating a pouring state in a pouring operation in which a ladle is tilted to pour a molten material into a mold, the system comprising:
Comprising an angle acquisition section, a weight acquisition section, a ladle internal shape acquisition section, and a calculation section,
The angle acquisition unit acquires a tilting angle and a tilting angular velocity of the ladle in the pouring operation,
The weight acquisition unit acquires the weight of the ladle in the pouring operation,
The ladle internal shape acquisition unit acquires data about the internal shape of the ladle,
The calculation unit calculates a pouring flow rate model representing a pouring state of the molten material based on the tilting angle, the weight, and the data about the internal shape , and the calculated pouring flow rate model and the An estimation system that calculates an estimated value of the amount of molten metal poured from the ladle by configuring an extended Kalman filter using a tilt angle, the tilt angular velocity, and the weight . The estimation system according to claim 1,
The calculation unit calculates the amount of change in the volume of a portion of the melt located above the lower end of the tap outlet of the ladle, with respect to changes in the vertical height and the tilt angle of the portion, respectively, An estimation system that calculates an estimated value of the amount of poured molten metal. The estimation system according to claim 1 or 2,
Further equipped with a storage section,
The ladle internal shape acquisition unit is an estimation system that acquires data about the internal shape of the ladle stored in the storage unit. The estimation system according to claim 3,
further comprising a measuring device that measures data about the internal shape of the ladle,
The ladle internal shape acquisition unit is an estimation system that acquires data about the internal shape of the ladle measured by the measuring device. The estimation system according to claim 4,
The storage unit is an estimation system that stores data regarding the internal shape of the ladle measured by the measuring device. The estimation system according to any one of claims 1 to 5,
It is further equipped with an imaging section capable of detecting infrared rays,
The imaging unit is an estimation system that images the position where the molten material is poured into the mold during the pouring operation. The estimation system according to claim 6,
In the estimation system, the imaging unit is disposed in front of the ladle in the tilting direction of the ladle and on the side of the ladle in a direction perpendicular to the tilting direction in plan view. The estimation system according to any one of claims 1 to 7,
Additionally equipped with an evaluation department,
The evaluation unit compares the reference value, which is the estimated value in the pouring work as a standard, with the estimated value in the pouring work as the evaluation target, and evaluates the pouring work as the target for evaluation. , Estimation System. The estimation system according to claim 8,
The estimation system is configured such that the evaluation unit performs the evaluation by calculating a correlation coefficient between the reference value and the estimated value. The estimation system according to claim 8,
In the estimation system, the evaluation unit performs the evaluation by calculating a Manhattan distance between the reference value and the estimated value using a dynamic time warping method. The estimation system according to claim 8,
The estimation system is configured such that the evaluation unit performs the evaluation by calculating a Manhattan distance between the reference value and the estimated value. The estimation system according to claim 8,
The estimation system is configured such that the evaluation unit performs the evaluation by calculating a Euclidean distance between the reference value and the estimated value. A method for estimating a pouring state in a pouring operation in which a ladle is tilted to pour a molten material into a mold, the method comprising:
The method includes an angle acquisition step, a weight acquisition step, a ladle internal shape acquisition step, and a calculation step,
In the angle acquisition step, a tilting angle and a tilting angular velocity of the ladle in the pouring operation are acquired;
In the weight obtaining step, obtaining the weight of the ladle in the pouring operation,
In the ladle internal shape obtaining step, data regarding the internal shape of the ladle is obtained;
In the calculation step, a pouring flow rate model representing the pouring state of the molten material is calculated based on the tilting angle, the weight, and the data regarding the internal shape , and the calculated pouring flow rate model and the An estimation method that calculates an estimated value of the amount of molten metal poured from the ladle by configuring an extended Kalman filter using a tilt angle, the tilt angular velocity, and the weight .