JP2020085395A - Automatic slag removal device and automatic slag removal program - Google Patents

Abstract

To automatically securely remove slag generated in a molten metal at the inside of a melting furnace.SOLUTION: An automatic slag removal device comprises: a photographed image data acquisition part of acquiring photographed image data by a photographing camera that has photographed the whole of the surface of a molten metal; a slag decision part of performing the decision of the presence of the generation of slag and the specification of the position of slag generation regarding images shown by photographed image data based on a learning-finished model beforehand learned about the decision of the generation of slag shown by photographed image data; a removal content decision part of, in the case it is decided that slag is generated, deciding a removal position in which the slag generated at any position in a melting furnace shall be removed, deciding an arm to be used for removal, and also, deciding the control contents of the arm; and a slag removal practicing part of practicing the removal treatment of the slab by performing the electronic control of the arm based on the decided removal contents.SELECTED DRAWING: Figure 1

Description

The present invention relates to an automatic slag removing device and an automatic slag removing program for automatically removing slag generated in molten metal in a melting furnace.

The slag generated during the casting process is an extra substance composed of a mineral component or the like different from the intended metal, and therefore may lead to product defects if not properly removed. The work of removing the slag from the high-temperature molten metal has conventionally been performed manually by a worker. The work of removing a heavy slag in a high temperature environment is a very heavy work and is a dangerous work, so that mechanization and automation have been required.

For example, Patent Document 1 discloses a slag removing device that removes slag from the liquid surface of the molten metal by running a filter that filters the slag along the liquid surface of the molten metal based on a preset traveling route. ..

JP, 2007-32868, A

The slag removing device disclosed in Patent Document 1 is configured to remove the slag by running the filter over the entire liquid surface. However, the removing operation is performed on such a pre-programmed path. In the case of the configuration, there is a problem that the device does not have a function to cope with the case where the slag is not completely removed by a series of removing operations and the slag remains on the liquid surface.

The present invention has been made in view of the above problems, and an object thereof is to provide an automatic slag removing device and an automatic slag removing program for automatically and reliably removing slag generated in molten metal in a melting furnace. To do.

The automatic slag removing device according to the present invention is an automatic slag removing device for automatically removing slag generated in molten metal in a melting furnace, so that the entire molten metal surface of the molten metal in the melting furnace can be seen. At least a photographing camera for photographing from a predetermined angle, at least one arm for removing slag in the melting furnace by electronic control, and a control unit for performing various controls including electronic control of the arm are provided. , The control unit has already learned about a captured image data acquisition unit that acquires captured image data captured by the capturing camera that captures the entire surface of the molten metal of molten metal, and has previously learned about determining whether or not slag has occurred on the surface of the molten metal. Based on the model, a slag determination unit for determining the presence or absence of slag and specifying the slag generation position for the image represented by the captured image data, and when it is determined that there is slag in the slag determination unit, in the melting furnace Of which position to remove the slag, which is to remove the slag, determines the arm to be used for removal, and determines the control content of the arm, and the removal content determination unit. And a slag removal execution unit that performs electronic control of the arm based on the removal content determined in (1) to perform a slag removal process.

In the automatic slag removing device according to the present invention, the learned model uses a neural network, and each region when the image represented by the captured image data is divided into a plurality of predetermined regions is a slag region. It is characterized in that the learning of the neural network is performed so that at least two kinds of determination can be made for at least two types, that is, the molten metal surface area and the molten metal surface area.

Further, in the automatic slag removal device according to the present invention, the removal content determination unit determines the slag removal position based on the information of the slag occurrence position, and the position, size, and shape of the slag present at the removal position. Based on the above, the arm to be used is determined, and the position at which the arm is inserted and the removal operation to be executed by the arm are determined.

An automatic slag removal program according to the present invention is an automatic slag removal program for causing a computer to realize control for automatically removing slag generated in molten metal in a melting furnace, and in the computer, a photographing camera is used. Based on a learned model that has been learned in advance to determine the presence or absence of slag on the surface of the molten metal, a captured image data acquisition function that acquires captured image data that captures the entire surface of the molten metal melt, A slag determination function that determines the presence or absence of slag and specifies the slag generation position for the image shown, and if it is determined that there is slag in the slag determination function, removes the slag that occurred in any position in the melting furnace. And a removal content determining function for determining the arm to be used for slag removal from at least one or more arms to be controlled, and determining the control content of the arm, and the removal content. And a slag removal execution function of performing electronic control of the arm and executing a slag removal process based on the removal content determined by the determination function.

According to the present invention, the presence or absence of slag is determined based on the captured image data showing the entire molten metal surface of the molten metal, and when it is determined that there is slag, the slag removal processing is executed. By surely determining the slag portion appearing in the captured image data and executing the slag removal, it is possible to reliably remove the slag without omission.

It is a block diagram showing composition of an automatic slag removal device concerning the present invention. It is explanatory drawing showing the concept of the learned model used in an automatic slag removal apparatus. It is an explanatory view showing the concept of slag determination using photographed image data. It is explanatory drawing showing the example of the arm used in an automatic slag removal device. It is a flowchart figure showing the flow of the slag removal process in an automatic slag removal apparatus.

Hereinafter, an example of the automatic slag removing apparatus according to the first embodiment will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of an automatic slag removing apparatus 100 according to the present invention. As shown in FIG. 1, the automatic slag removing apparatus 100 according to the present invention includes at least a photographing camera 10, an arm 20, a control unit 30, and a storage unit 40. The automatic slag removing apparatus 100 according to the present invention is used by being applied to equipment that needs to melt metal by a melting furnace to remove slag from molten metal, such as casting equipment. Here, the melting furnace means a furnace for melting the metal, and includes all furnaces used for melting the metal. That is, the term "melting furnace" is used as an expression that includes all of various furnaces such as a casting furnace, a cupola, a reverberatory furnace, a converter, and an electric furnace. Further, in this example, it is expressed as slag, but it is synonymous with terms such as slag and slag. In addition, although an automatic slag removing device is described in this example, an automatic slag removing program for causing a computer to realize the control content of the control unit 30 described later may be used.

The photographing camera 10 has a function of photographing the entire molten metal surface of the molten metal in the melting furnace from a predetermined angle. The entire surface of the molten metal of the molten metal in the melting furnace is installed at a position and angle where it can be photographed. For example, the entire surface of the molten metal is installed at a position where an image can be taken from directly above in the vertical direction. The photographing camera 10 may be a camera device for photographing a still image or a camera device for photographing a moving image, but in the case of a moving image, data for each frame can be used as photographed image data. It is assumed that it has various shooting formats.

The arm 20 is an operation target operated by electronic control in order to remove slag in the melting furnace. Various shapes of the arm 20 can be adopted. Further, in the present example, the tip portion directly charged into the melting furnace is mainly represented as the arm 20, but a robot arm capable of performing complicated operations in order to cause the arm 20 to perform various operations. The location of the robot arm is indispensable, and the location of the robot arm needs to be appropriately designed according to the control content to be executed by the arm 20. That is, in the present example, the arm 20 is used in a narrow sense to refer to the tip portion that is directly charged into the melting furnace. It also includes the robot arm location for executing complicated operations. Further, the number of arms 20 does not have to be one, and a plurality of same arms 20 may be prepared, or a plurality of types of arms 20 having different shapes may be prepared and used properly according to the situation.

The control unit 30 determines whether or not the slag removal is necessary based on the captured image data captured by the capturing camera 10, determines the slag removal content when it is necessary, and electronically controls the arm 20 based on the determined content. It has a function of controlling and executing control for executing the slag removal processing. Then, as shown in FIG. 1, the control unit 30 includes at least a captured image data acquisition unit 31, a slag determination unit 32, a removal content determination unit 33, and a slag removal execution unit 34.

The captured image data acquisition unit 31 has a function of acquiring captured image data captured by the camera 10 that captures the entire surface of the molten metal. Regarding the timing of shooting by the shooting camera 10, the shooting camera 10 may perform shooting by the control signal from the shooting image data acquisition unit 31 to acquire the shot image data. The captured image data acquisition unit 31 may acquire captured image data captured (continuously captured in the case of a moving image) at set predetermined intervals.

The slag determination unit 32 determines the presence or absence of slag and specifies the slag occurrence position for the image represented by the captured image data based on a learned model that has been learned in advance to determine whether or not slag occurs on the surface of the molten metal. Have the function to do. The slag determination unit 32 determines whether or not a slag has occurred in the image indicated by the captured image data acquired by the captured image data acquisition unit 31. Then, for the determination of the slag, a learned model obtained by previously learning the neural network by machine learning is used. Any learned model may be used as long as it is possible to determine whether or not slag has occurred. It may be divided and the presence or absence of slag may be determined for each area. The slag determination unit 32 also identifies the slag occurrence position. Any method may be used as long as the slag occurrence position can be specified.For example, the slug area in the captured image data is determined from the determination result of the presence or absence of the slag for each area obtained by dividing the captured image data into a plurality of predetermined areas. Alternatively, the position where the slag occurs in the captured image data may be specified. If the slag determination unit 32 determines that there is slag, slag removal processing is required. When it is determined that no slag has occurred, the subsequent slag removal processing is unnecessary.

When the slag determination unit 32 determines that there is slag, the removal content determination unit 33 determines the removal position that is the position in the melting furnace at which the slag that has been removed is removed, and the arm 20 used for the removal is determined. It has a function of determining and controlling the content of control of the arm 20. When the slag determination unit 32 determines that there is slag, slag removal processing is required, and thus the removal content determination unit 33 determines the removal content. As the content of removal, first, it is necessary to determine the removal position, which is the position in the melting furnace where the slag generated should be removed. Determining the removal position is to specify the coordinate information of the slag generation position in the melting furnace accurately and which slag to remove when the slag is generated at two or more different locations separated from each other. It means that both the removal target identification for determining the slag of the removal target and the removal target identification are performed. Although the slag to be removed may be determined in any way, for example, a deciding method in which the slag having the largest area is preferentially removed can be considered. Next, as the removal content, the arm used for the removal is determined. If there is only one arm, there is no choice, but if a plurality of types of arms are to be installed, it is necessary to decide which arm to use. Any method may be used to determine the arm, but it is optimal depending on, for example, the area of the slag, the position of the slag (whether it is near the center, or near the wall of the furnace). The arm may be determined. Finally, the content of control of the arm is determined as the content of removal. The control content of the arm is the control content of what position the arm selected for the slag to be removed is inserted and what operation is performed. Achieves efficient slag removal by determining the optimum control content.

The slag removal execution unit 34 has a function of performing electronic control of the arm 20 and executing a slag removal process based on the removal content determined by the removal content determination unit 33. The slag removal executing unit 34 executes the slag removal process. After the slag removal execution unit 34 has executed the slag removal processing, the captured image data acquisition unit 31 acquires captured image data, and the slag determination unit 32 determines whether the slag has been completely removed. When the slag remains, the removal content determination unit 33 and the slag removal execution unit 34 perform the slag removal process again.

The storage unit 40 has a function of storing various information such as information used in the control unit 30 and information obtained as a result of processing by the control unit 30. Further, the learned model 41 used in the slag determination unit 32 may be stored in the storage unit 40.

FIG. 2 is an explanatory diagram showing the concept of the learned model used in the automatic slag removing apparatus. As shown in FIG. 2, the learned model is obtained by learning the neural network in advance, and the photographed image data showing the entire surface of the molten metal is input to the learned model, and the photographed image data shows it. It is determined whether or not slag has occurred in the learned model for the image, and the image is output. The output at that time may be an output that specifies a slag area in the captured image data, or may be a determination result of the presence or absence of a slag for each predetermined area. Learning of a learned model for such slag determination processing is performed by, for example, using a plurality of sets of captured image data and correct data representing whether or not slag exists for the captured image data, as one set. Neural network learning is performed using the set learning data. The difference between the judgment result output by the neural network and the correct answer data is evaluated by the loss function, and the weight and bias of the neural network are updated based on the error backpropagation method or the like so that the loss by the loss function is reduced. By repeating this based on a plurality of sets of learning data, a learned model that can realize highly accurate slag determination is obtained. Further, in the case where the image shown by the photographed image data is configured to make a determination for each predetermined region and output the determination result for each region, the structure of the neural network is set as such and correct data is set for each region. It is preferable to provide the correct answer data.

In addition, the learned model may be learned as a learned model specialized for the equipment to be applied (for example, a learned model for slag determination of cast iron), and is versatile enough to support various molten metals. The learning may be performed as a high-learned model. When using a learned model that is specialized for the equipment to be applied, it is necessary to set the captured image data used for learning only for the metal that can be melted in that equipment. Needs to widely learn the captured image data used for learning by using various metal data.

Further, the learned model may be one that has been learned so that not only the slag region and the molten metal region but also the debris region can be determined. A slag material is used in the process of removing slag from molten metal. When the slag material is put into the melting furnace, it reacts with the slag and slag is generated on the surface of the molten metal. However, when the slag material is put into the molten metal in a state where the slag component is small, what the slag material is melted may be generated on the surface of the molten metal. It is also necessary to remove such slag material. Since the slag and the slag material differ in how they appear in the captured image data, multiple sets of learning data are prepared to include the three patterns of the slag area, the slag material area, and the molten metal surface area when learning the learned model. By performing the learning in this way, it is possible to obtain a learned model that can determine the three patterns of the slag area, the slag material area, and the molten metal surface area.

FIG. 3 is an explanatory diagram showing the concept of slag determination using captured image data. In FIG. 3, the circles represent the molten metal surface of the melting furnace composed of a cylindrical container, and the grids represent the regions that are the judgment units of the learned model. In the image represented by the captured image data of FIG. 3A, the range indicated by A and the range indicated by B are the places where the slag has occurred. When the slag determination is performed on the photographed image data of FIG. 3A using the learned model, the area of the grid where A and A overlap as much as possible is determined to be the slag area. In the image represented by the photographed image data of FIG. 3B, the range indicated by C is the place where the slag has occurred, and the range indicated by D is the place where the slag material remains. When the slag determination is performed on the photographed image data of FIG. 3(b) by the learned model, the area of the lattice in which c is slightly overlapped is determined to be the slag area, and the area of the lattice in which d is slightly overlapped is removed. It will be judged as a region.

FIG. 4 is an explanatory view showing an example of an arm used in the automatic slag removing device. FIG. 4A shows arms branched into three branches and extended in different directions by 120° in the same plane. FIG. 4B shows an arm having a circularly processed portion (or a portion where R is formed by bending) at the tip. For example, if the slag is generated as shown in FIG. 3(a), the three-pronged arm shown in FIG. 4(a) may be suitable for the slag in the range indicated by (a). An arm having an annular portion shown in FIG. 4B may be suitable for the slag in the range shown. Since the slag in the range indicated by b is closer to the wall surface of the furnace, it can be said that there is a possibility that scooping the slag at the annular portion rather than the three-pronged arm can remove the slag more efficiently.

Next, the flow of slag removal processing in the automatic slag removal device will be described. FIG. 5 is a flowchart showing the flow of slag removal processing in the automatic slag removal device. In FIG. 5, the slag removal processing in the automatic slag removal apparatus is started by the captured image data being acquired by the control unit 30 of the automatic slag removal apparatus 100 (step S101). Next, the control unit 30 of the automatic slag removing apparatus 100 performs slug determination on the image indicated by the acquired captured image data (step S102). When it is determined that the slag exists (S102-Y), the slag removal content is determined (step S103). Then, slag removal is executed based on the determined slag removal content (step S104). After executing the slag removal, the process always returns to step S101. That is, the captured image data is acquired again (step S101), and the slag determination is executed (step S102). When it is determined that there is no slag in the slag determination process (S102-N), it is determined that there are no more removal targets, and the slag removal process ends.

As described above, according to the automatic slag removing apparatus according to the present invention, it is possible to obtain image data captured by a camera that captures the entire molten metal surface of molten metal, and determine whether or not slag is generated on the molten metal surface in advance. Based on the learned model learned, it determines the presence or absence of slag occurrence in the image indicated by the captured image data and identifies the slag occurrence position, and when it is determined that slag exists, it occurs at any position in the melting furnace The removal position, which is to remove the slag, is determined, the arm used for the removal is determined, and the control content of the arm is determined. Based on the determined removal content, the electronic control of the arm is performed and the slag removal is performed. Since the removal process is executed, it is highly accurate to judge whether the slag has occurred in the captured image data by the learned model and to execute the appropriate slag removal process when the slag has occurred. Is possible. As a result, it is possible to prevent slag from remaining in the molten metal, so that it is possible to improve the quality of metal products and improve the product yield. Further, since reliable slag removal can be automated, it is not necessary for a worker to perform dangerous work, and safety can be improved.

In the above-described embodiment, the control unit 30 of the automatic slag removal device 100 executes various processes including the slag determination process, and the learned model 41 stored in the storage unit 40 of the automatic slag removal device 100 is slugged by the control unit 30. Although the description has been given as being used in the determination unit 32, the present invention is not limited to this. For example, the learned model 41 may be stored in a server device connectable from the automatic slag removal device 100 via a communication network, and the slag determination process may be executed in the server device. That is, the captured image data acquired by the captured image data acquisition unit 31 is transmitted to the server device via the communication network, the server device is provided with the same function as the slag determination unit 32, and the slag determination process is executed in the server device. Even with the configuration in which the slag determination result in the server device is received by the automatic slag removing device 100 via the communication network, it is possible to realize the automatic slag removing device 100 having the same effect as that of the above embodiment.

100 automatic slag removal device 10 shooting camera 20, 20a, 20b arm 30 control unit 31 captured image data acquisition unit 32 slug determination unit 33 removal content determination unit 34 slag removal execution unit 40 storage unit 41 learned model

Claims (4)

An automatic slag removing device for automatically removing slag generated in molten metal in a melting furnace,
A photographing camera for photographing from a predetermined angle so that the entire molten metal surface of the molten metal in the melting furnace is photographed,
At least one or more arms for removing slag in the melting furnace by electronic control;
And a control unit for performing various controls including electronic control of the arm,
The control unit is
A captured image data acquisition unit for acquiring captured image data captured by the capturing camera, which shows the entire surface of the molten metal.
Based on a learned model learned in advance to determine the presence or absence of slag on the surface of the molten metal, a slag determination unit for determining the presence or absence of slag and specifying the slag occurrence position for the image represented by the captured image data. ,
When it is determined that slag is present in the slag determination unit, a removal position that determines which position in the melting furnace to remove the slag is determined, the arm used for removal is determined, and the arm A removal content determination unit that determines the control content of
An automatic slag removal device comprising: a slag removal execution unit that performs electronic control of the arm and executes a slag removal process based on the removal content determined by the removal content determination unit. The learned model uses a neural network, and whether each area when the image represented by the captured image data is divided into a predetermined plurality of areas is a slag area or a molten metal surface area. The automatic slag removal device according to claim 1, wherein learning of the neural network is performed so that at least two types of the above can be determined. The removal content determination unit, based on the information of the slag occurrence position, determines the slag removal position, determines the position of the slag present at the removal position, the size, the arm to be used based on the shape, etc., The automatic slag removing device according to claim 1 or 2, which determines a position to insert the arm and a removing operation to be executed by the arm. An automatic slag removal program for causing a computer to perform control for automatically removing slag generated in molten metal in a melting furnace,
On the computer,
A captured image data acquisition function that acquires captured image data that captures the entire surface of the molten metal with a capturing camera,
Based on a learned model learned in advance about determining the presence or absence of slag on the surface of the molten metal, a slag determination function for determining the presence or absence of slag generation for the image represented by the captured image data and specifying the slag generation position. ,
When it is determined by the slag determination function that slag is present, a removal position that determines which position in the melting furnace to remove the slag is determined, and the slag is selected from at least one or more arms to be controlled. A removal content determination function for determining an arm to be used for removal and determining control content of the arm;
An automatic slag removal program that realizes a slag removal execution function that performs electronic control of the arm and executes a slag removal process based on the removal content determined by the removal content determination function.

Images