Auxiliary device and computer-readable recording medium

Abstract

An auxiliary device according to the present disclosure comprises: a levelness acquisition unit that acquires the levelness of a frame of an industrial machine before leveling; a target value setting unit that sets a target value of the levelness of the frame of the industrial machine after leveling; an adjustment value estimation unit that estimates, using a first model, an adjustment value pertaining to adjustment during leveling, on the basis of the levelness of the frame of the industrial machine before leveling and the target value of the levelness of the frame of the industrial machine after predetermined leveling; a levelness estimation unit that estimates, using a second model, the levelness after leveling, on the basis of the levelness of the frame of the industrial machine before leveling and the adjustment value estimated by the adjustment value estimation unit; a reliability calculation unit that calculates the reliability of the adjustment value estimated by the adjustment value estimation unit, on the basis of the target value of the levelness of the frame of the industrial machine after the predetermined leveling and the levelness of the frame of the industrial machine after leveling estimated by the levelness estimation unit; and an output unit that outputs the adjustment value estimated by the adjustment value estimation unit and the reliability calculated by the reliability calculation unit.

Description

Auxiliary device and computer-readable recording medium

[0001] The present disclosure relates to an auxiliary device and a computer-readable recording medium.

[0002] When installing industrial machinery such as processing machines or semiconductor manufacturing equipment at a manufacturing site, it is necessary to ensure that the platform of the industrial machinery is level. This work is generally called leveling or horizontal alignment. This leveling work is performed, for example, by adjusting adjustment mechanisms such as leveling bolts attached to each of the multiple legs of the platform (see, for example, Patent Document 1).

[0003] Japanese Patent Application Laid-Open No. 2006-291982

[0004] Machine leveling is a meticulous task, and its accuracy directly affects the quality of the final product, making it extremely important. However, such adjustments are often performed by experienced workers, who are scarce and expensive to work with, based on their own experience. While machine learning-based inference devices are sometimes used to replace the implicit knowledge of these workers, the expected accuracy (reliability) of the inference results is usually unknown. If adjustments based on the inference results do not fall within the tolerances, corrections must be made, but this work requires the help of experienced workers, resulting in duplicate work and being extremely inefficient. Therefore, there is a need for technology that can assist in more efficient leveling work on-site.

[0005] The auxiliary device according to the present disclosure creates a first model in which the "levelness before leveling" and the "levelness after leveling" are explanatory variables (inputs) and the "adjustment value for leveling adjustment" is the objective variable (output), and a second model in which the "levelness before leveling" and the "adjustment value for leveling adjustment" are explanatory variables (inputs) and the "levelness after leveling" is the objective variable (output). The "adjustment value for leveling adjustment" output by the first model is then input into the second model, and the value obtained by comparing the "levelness after leveling" output by the second model with the "levelness after leveling" input into the first model is used as the reliability of the inference result, and a judgment is made based on that reliability, thereby solving the above problem.

[0006] One aspect of the present disclosure includes a level degree acquisition unit that acquires the level of the base of the industrial machine before leveling, a target value setting unit that sets a target value for the level of the base of the industrial machine after leveling, an adjustment value estimation unit that estimates an adjustment value for the adjustment in leveling based on the level of the base of the industrial machine before leveling acquired by the level degree acquisition unit and the target value for the level of the base of the industrial machine after leveling set by the target value setting unit using a first model that has been trained with the level of the base of the industrial machine before leveling and the target value for the level of the base of the industrial machine after leveling as explanatory variables and an adjustment value for the adjustment in leveling as a response variable, and and an adjustment value related to the above-mentioned adjustment estimation unit that estimates the level after leveling based on the level of the industrial machinery pedestal before leveling acquired by the levelness acquisition unit and the adjustment value estimated by the adjustment value estimation unit using a second model that has been trained using an adjustment value related to the above-mentioned adjustment estimation unit as an explanatory variable and a target variable that is the level of the industrial machinery pedestal after leveling; a reliability calculation unit that calculates the reliability of the adjustment value estimated by the adjustment value estimation unit based on the target value of the level of the industrial machinery pedestal after leveling set by the target value setting unit and the level of the industrial machinery pedestal after leveling estimated by the levelness estimation unit; and an output unit that outputs the adjustment value estimated by the adjustment value estimation unit and the reliability calculated by the reliability calculation unit.

[0007] FIG. 1 is a hardware configuration diagram of a control device equipped with an auxiliary device according to a first embodiment. FIG. 2 is a block diagram showing the auxiliary device and functions of the auxiliary device according to the first embodiment. FIG. 3 is a schematic configuration diagram of an injection molding machine. FIG. 4 is a screen diagram showing an example of an input screen for levelness. FIG. 5 is a screen diagram showing an example of an input screen for target levelness. FIG. 6 is a screen diagram showing an example of a display screen for adjustment value estimation results. FIG. 7 is a block diagram showing the functions of a control device and auxiliary device according to a second embodiment. FIG. 8 is a block diagram showing the functions of a control device and auxiliary device according to a third embodiment. FIG. 9 is a screen diagram showing an example of an input screen for a target levelness range.

[0008] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In the following description, components having the same or similar functions will be denoted by the same reference numerals. Duplicate descriptions of those components may be omitted.

[0009] In this application, "based on XX" means "based on at least XX," and includes cases where it is based on other elements in addition to XX. Furthermore, "based on XX" is not limited to cases where XX is used directly, but also includes cases where it is based on XX that has been calculated or processed. "XX" is any element (for example, any information).

[0010] [First Embodiment] Fig. 1 is a schematic hardware configuration diagram showing the main parts of a control device having the functions of an auxiliary device according to a first embodiment of the present disclosure. The auxiliary device of the present disclosure can be implemented on a control device that controls, for example, industrial machinery. The auxiliary device of the present disclosure can also be implemented on a computer such as a personal computer operated by a user, a cell computer, a fog computer 6, or a cloud server 7. In this embodiment, an example is shown in which the auxiliary device is implemented on a personal computer operated by a user.

[0011] The CPU 11 provided in the auxiliary device 1 according to this embodiment is a processor that controls the entire auxiliary device 1. The CPU 11 reads a system program stored in the ROM 12 via the bus 22 and controls the entire auxiliary device 1 in accordance with the system program. The RAM 13 temporarily stores temporary calculation data, display data, and various data input from outside.

[0012] The nonvolatile memory 14 is configured, for example, by a battery-backed memory (not shown) or an SSD (Solid State Drive), and maintains its stored state even when the auxiliary device 1 is powered off. The nonvolatile memory 14 stores programs and data read from an external device 72 via the interface 15, and programs and data input via the input device 71. The data stored in the nonvolatile memory 14 may be expanded into the RAM 13 during execution / use. In addition, various system programs, such as known analysis programs, are written in the ROM 12 in advance.

[0013] The interface 15 is an interface for connecting the CPU 11 of the auxiliary device 1 to an external device 72 such as a USB memory, a CompactFlash (registered trademark), or an SD card. For example, programs and various data can be read from the external device 72. In addition, programs and various data edited within the auxiliary device 1 can be stored in the external device 72.

[0014] The display device 70 displays various data loaded into the memory, data obtained as a result of executing programs, etc., output via the interface 18. An input device 71, which is comprised of a keyboard, pointing device, etc., passes instructions, data, etc., based on operations by an operator to the CPU 11 via the interface 19.

[0015] The interface 20 is an interface for connecting the CPU 11 of the auxiliary device 1 to a wired or wireless network 5. The network 5 may communicate using technologies such as serial communication such as RS-485, Ethernet (registered trademark), optical communication, wireless LAN, Wi-Fi (registered trademark), Bluetooth (registered trademark), etc. Computers such as a fog computer 6 and a cloud server 7 are connected to the network 5, and data is exchanged between the network 5 and the auxiliary device 1.

[0016] Interface 21 is an interface for connecting CPU 11 and machine learning device 100. Machine learning device 100 includes processor 101 that controls the entire machine learning device 100, ROM 102 that stores system programs and the like, RAM 103 that provides temporary storage for each process related to machine learning, and non-volatile memory 104 that stores models and the like. Machine learning device 100 can observe each piece of information that can be acquired by auxiliary device 1 via interface 21. In addition, auxiliary device 1 acquires processing results output from machine learning device 100 via interface 21, and stores or displays the acquired results, or transmits them to another device via network 5 or the like.

[0017] 2 is a schematic block diagram illustrating functions provided in the auxiliary device 1 according to the first embodiment of the present disclosure. The auxiliary device 1 has functions provided in a general computer in addition to the functions illustrated in FIG. 2. Each function provided in the auxiliary device 1 according to this embodiment is realized by the CPU 11 of the auxiliary device 1 and the processor 101 of the machine learning device 100 shown in FIG. 1 executing a system program and controlling the operation of each part of the auxiliary device 1 and the machine learning device 100.

[0018] The auxiliary device 1 of this embodiment includes a levelness acquisition unit 110, a target value setting unit 120, an adjustment value estimation unit 130, a levelness estimation unit 140, a reliability calculation unit 150, and an output unit 160. In addition, the RAM 13 to the nonvolatile memory 14 of the auxiliary device 1 are provided in advance with a first model storage unit 210 that stores a first model, which is a model for estimating an adjustment value of the adjustment mechanism during leveling based on the levelness before leveling and the target levelness after leveling, and a second model storage unit 220 that stores a second model that estimates the levelness after leveling based on the levelness before leveling and the adjustment value of the adjustment unit during leveling.

[0019] The levelness acquisition unit 110 acquires the levelness of the industrial machine platform before leveling work is performed. The acquired levelness is then output to the adjustment value estimation unit 130 and the levelness estimation unit 140. The levelness of the platform will be described using FIG. 3 . FIG. 3 is a diagram showing a schematic configuration of an injection molding machine as an example of the industrial machine 2. The industrial machine 2, which is an injection molding machine, is mainly composed of a mold clamping unit 401 and an injection unit 402. The mold clamping unit 401 and the injection unit 402 are attached to a platform 410. The platform 410 has multiple legs 412 and is supported by the legs 412 in contact with the floor surface. In the example of FIG. 3 , legs 412e to 412h (not shown) are attached to the rear of the legs 412a to 412d in the figure, making up eight legs that support the platform 410. Additionally, multiple spirit levels 414 are attached to the platform 410. In the example of FIG. 3, levels 414e to 414h (not shown) are provided behind the levels 414a to 414d with the clamping unit 401 and injection unit 402 sandwiched therebetween.

[0020] The spirit level 414 measures, for example, the horizontality 1 in one axial direction. The spirit level 414 may indicate the horizontality from a predetermined negative value to a predetermined positive value, with 0 as the center. In this case, 0 generally indicates horizontality. It is desirable that at least some of the spirit levels 414 measure the horizontality in an axial direction different from that of the other spirit levels 414. The spirit levels 414 may be configured to visually display the measurement results. They may also be configured to output the measurement results to the auxiliary device 1 as an electrical signal or the like.

[0021] Each leg 412 is equipped with an adjustment mechanism, such as a leveling bolt, and the height of the platform 410 can be adjusted by adjusting the adjustment mechanism. The leveling operation, which adjusts the horizontality of the platform 410, is performed by adjusting the adjustment mechanism of each leg 412 attached to the bottom of the platform 410 based on a predetermined adjustment value a. For example, in the case of an industrial machine 2 equipped with eight adjustable legs 412a to 412h as illustrated in FIG. 3, the adjustment mechanisms of each leg 412a to 412h are adjusted based on a predetermined set of adjustment values ​​(a1, a2, ..., a8). If this adjustment value indicates an adjustment amount from the current height, for example, the adjustment mechanism of each leg 412 is adjusted by the specified adjustment value from the current position, so that a negative adjustment value lowers the platform, or a positive adjustment value raises the platform 410. After performing this adjustment operation, the platform 410 can be said to be level when the measurement results of all levels 414 simultaneously indicate approximately horizontal.

[0022] The worker who installs the industrial machine 2 needs to adjust each leg 412 so that the measurement results of all the levels 414 indicate approximately horizontal at the same time. When performing this work, the worker must adjust the set of horizontal degrees indicated by the respective levels 414 (l1, l2, ..., l i) into the auxiliary device 1. For example, when performing adjustment work on an industrial machine 2 equipped with eight spirit levels 414a to 414h as shown in FIG. 3 , a set of levels (l1, l2, ..., l8) measured by each of the spirit levels 414a to 414h is input to the auxiliary device 1. When this input is performed, the level degree acquisition unit 110 may, for example, display a screen on the display device 70 for inputting the level degrees measured by each of the spirit levels 414, prompting the worker to enter the measurement results of the spirit levels 414. FIG. 4 is a screen diagram showing an example of a level degree input screen displayed on the display device 70 by the level degree acquisition unit 110. The worker looks at the measurement results of each spirit level 414 attached to the stand 410 of the industrial machine 2 and sequentially enters them into the input field 112. Then, by inputting the measurement results of all the spirit levels 414 and pressing the OK button 114, the level degree acquisition unit 110 acquires the level degrees input by the worker. In addition, if the spirit level 414 has the function of outputting the measurement results as an electrical signal, etc., the levelness acquisition unit 110 may acquire the levelness as a measurement result by the spirit level 414 via a specified communication path, etc.

[0023] Returning to FIG. 2 , the target value setting unit 120 sets target values ​​for the levelness of the platform during the leveling work. These target values ​​for levelness become a set of levels to be measured after the adjustment work using the respective spirit levels 414 attached to the platform 410 of the industrial machine 2. The target value setting unit 120 then outputs the set target values ​​for levelness to the adjustment value estimation unit 130. For example, the target value setting unit 120 may display a screen on the display device 70 for inputting target values ​​for levelness indicated by the respective spirit levels 414 after the adjustment work, thereby prompting the operator to input the target values ​​for levelness. FIG. 5 is a screen diagram showing an example of an input screen for target values ​​for levelness displayed on the display device 70 by the target value setting unit 120. The operator inputs the target values ​​for levelness to be indicated by each spirit level 414 in the target value input field 122. Then, by pressing an OK button 124, the target value setting unit 120 sets the values ​​input by the operator as the target values ​​for levelness. In general, the adjustment work aims to make all the horizontal degrees indicated by the level 414 equal to values ​​indicating horizontality. Therefore, the target value setting unit 120 may set all the target values ​​to values ​​indicating horizontality (for example, 0) without receiving input from the operator.

[0024] The adjustment value estimation unit 130 estimates an adjustment value for the legs 412 of the industrial machine 2 in the leveling work based on the current level before the leveling work and the target value for level after the leveling work. The first model storage unit 210 stores a first model that has been trained in advance and that uses the level before leveling and the target value for level after leveling as explanatory variables (inputs) and the adjustment value of the legs 412 of the industrial machine 2 in the leveling work as a target variable (output). The adjustment value estimation unit 130 inputs the current level of the industrial machine 2 acquired by the level degree acquisition unit 110 and the target value for level after leveling set by the target value setting unit 120 to the first model read from the first model storage unit 210, and estimates the output as the adjustment value for the legs 412 of the industrial machine 2 in the leveling work. The adjustment value estimation unit 130 outputs the estimated adjustment value to the level degree estimation unit 140 and the output unit 160.

[0025] Examples of the first model stored in the first model storage unit 210 include a convolutional neural network, a recurrent neural network, a multilayer perceptron, and a regression model. Deep learning models such as a convolutional neural network are expected to improve estimation accuracy. On the other hand, a regression model has a simple configuration and reduces the load on the CPU 11 and the processor 101. This first model may be obtained by extracting the horizontality of each level 414 before the leveling operation, the adjustment values ​​for the adjustment mechanisms of each leg 412 during the leveling operation, and the resulting horizontality indicated by each level 414 after the leveling operation from data previously recorded during leveling operations performed on the industrial machine 2 and data obtained through additional experiments, and then performing a learning process until sufficient accuracy is obtained.

[0026] The level degree estimation unit 140 estimates the level degree after the leveling work has been performed based on the current level degree before the leveling work and the adjustment value of the legs 412 of the industrial machine 2 during the leveling work. The second model storage unit 220 stores a second model that has been trained in advance and that uses the level degree before leveling and the adjustment value of the legs 412 of the industrial machine 2 during the leveling work as explanatory variables (inputs) and the level degree after leveling as a target variable (output). The level degree estimation unit 140 inputs the current level degree of the industrial machine 2 acquired by the level degree acquisition unit 110 and the adjustment value estimated by the adjustment value estimation unit 130 into the second model read from the second model storage unit 220, and uses the output as an estimation result of the level degree after the leveling work has been performed. The level degree estimation unit 140 outputs the estimated level degree after leveling to the reliability calculation unit 150.

[0027] The second model stored in the second model storage unit 220 may be, like the first model, a convolutional neural network, a recurrent neural network, a multilayer perceptron, a regression model, etc. This second model may be obtained by extracting the horizontality of each level 414 before the leveling work, the adjustment values ​​of the adjustment mechanisms of each leg 412 in the leveling work, and the horizontality indicated by each level 414 after the leveling work from data recorded in advance during leveling work performed in the industrial machine 2 and data obtained through additional experiments, and then performing a learning process until sufficient accuracy is obtained.

[0028] The reliability calculation unit 150 compares the degree of level after leveling estimated by the level degree estimation unit 140 with the target value of the degree of level after leveling set by the target value setting unit 120. Then, based on the comparison result, calculates the reliability c of the adjustment value of the legs 412 of the industrial machine 2 in the leveling work estimated by the adjustment value estimation unit 130. For example, the reliability calculation unit 150 calculates the Euclidean distance or Manhattan distance when the degree of level after leveling of each level 414 estimated by the level degree estimation unit 140 and the target value of the degree of level after leveling of each level 414 set by the target value setting unit 120 are regarded as vectors, and calculates the reliability c using a function (for example, c = 1 / (1 + d) where d is the distance between them) such that the reliability c is high when the calculation result is close to 0 and the reliability c decreases as the distance increases. The reliability calculation unit 150 outputs the calculated reliability c to the output unit 160 .

[0029] The output unit 160 outputs the adjustment value of the leg 412 of the industrial machine 2 in the leveling work estimated by the adjustment value estimation unit 130 and the reliability of the adjustment value calculated by the reliability calculation unit 150. The output unit 160 may output and display the same on the display device 70, for example. Alternatively, the output unit 160 may output and transmit the same to the fog computer 6 or the cloud server 7 via the network 5.

[0030] Fig. 6 is a screen diagram showing an example in which the output unit 160 displays and outputs the adjustment values ​​of the legs 412 of the industrial machine 2 and the reliability of the adjustment values ​​on the display device 70. In the example of Fig. 6, the display screen for the adjustment value estimation results has a display field 162 for the adjustment values ​​of the adjustment mechanisms of the legs 412 estimated by the adjustment value estimation unit 130 and a display field 164 for the reliability of the estimated adjustment values.

[0031] The auxiliary device 1 according to this embodiment, which is configured as described above, can estimate the reliability of the leveling adjustment value estimated by the first model using the second model. If the estimated reliability is high, it is expected that the industrial machine 2 can be leveled by adjusting the adjustment mechanism of the legs 412 of the industrial machine 2 according to the adjustment value. On the other hand, if the estimated reliability is low, the adjustment value estimated by the first model cannot be trusted. Therefore, to avoid duplicate work, it is possible to avoid redoing the work by entrusting the work to an experienced worker from the beginning.

[0032] Second Embodiment An auxiliary device according to a second embodiment will be described below. The auxiliary device 1 according to this embodiment has the same hardware configuration as the auxiliary device 1 according to the first embodiment.

[0033] 7 is a schematic block diagram illustrating functions provided in the auxiliary device 1 according to the second embodiment of the present disclosure. The auxiliary device 1 has functions provided in a general computer in addition to the functions illustrated in FIG. 7. Each function provided in the auxiliary device 1 according to this embodiment is realized by the CPU 11 of the auxiliary device 1 and the processor 101 of the machine learning device 100 shown in FIG. 1 executing a system program and controlling the operation of each part of the auxiliary device 1 and the machine learning device 100.

[0034] The auxiliary device 1 of this embodiment further includes a height acquisition unit 115 in addition to the levelness acquisition unit 110, target value setting unit 120, adjustment value estimation unit 130, levelness estimation unit 140, reliability calculation unit 150, and output unit 160. Also, the RAM 13 to the nonvolatile memory 14 of the auxiliary device 1 are provided in advance with a first model storage unit 210 that stores a first model, which is a model for estimating the adjustment value of the adjustment mechanism in leveling based on the levelness before leveling, the height of the industrial machine 2 before leveling, and the target levelness after leveling, and a second model storage unit 220 that stores a second model that estimates the levelness after leveling based on the levelness before leveling, the height of the industrial machine 2 before leveling, and the adjustment value of the adjustment mechanism in leveling.

[0035] The functions of the levelness acquisition unit 110, the target value setting unit 120, and the output unit 160 according to this embodiment are similar to those of the first embodiment.

[0036] The height acquisition unit 115 acquires the height of the industrial machine 2 at the attachment positions of the legs 412 of the pedestal 410 before the leveling work is performed. Then, the acquired height at the attachment positions of the legs 412 is output to the adjustment value estimation unit 130 and the levelness estimation unit 140. The height acquired by the height acquisition unit 115 may be the distance between the bottom surface of the pedestal 410 and the floor contact surface of the legs 412 at the attachment positions of each of the legs 412 attached to the pedestal 410. Furthermore, for example, if the adjustment mechanism of the legs 412 is equipped with a gauge or the like that measures the height based on a predetermined standard, the measurement value of the gauge may be used as the height at the attachment positions of the legs 412.

[0037] The adjustment value estimation unit 130 according to this embodiment estimates the adjustment value of the legs 412 of the industrial machine 2 in the leveling work, based on the current level before the leveling work, the current height of the legs 412 of the stand 410 of the industrial machine 2 at the attachment positions of the legs 412 before the leveling work, and the target value of the level after the leveling work. The first model storage unit 210 stores a first model that has been trained in advance, in which the level before leveling, the height of the stand 410 at the attachment positions of the legs 412 before leveling, and the target value of the level after leveling are used as explanatory variables (inputs), and the adjustment value of the legs 412 of the industrial machine 2 in the leveling work is used as a target variable (output). The adjustment value estimation unit 130 inputs the current level of the industrial machine 2 acquired by the level degree acquisition unit 110, the current height of the industrial machine 2 at the attachment position of the legs 412 of the stand 410 acquired by the height acquisition unit 115, and the target value of level after leveling set by the target value setting unit 120 to the first model read out from the first model storage unit 210, and estimates the output as the adjustment value of the legs 412 of the industrial machine 2 in the leveling work. The adjustment value estimation unit 130 outputs the estimated adjustment value to the level degree estimation unit 140 and the output unit 160.

[0038] As in the first embodiment, the first model stored in the first model storage unit 210 can be obtained by extracting the horizontality of each spirit level 414 before the leveling work, the height of the stand 410 at the mounting position of each leg 412 before leveling, the adjustment values ​​for the adjustment mechanisms of each leg 412 during the leveling work, and the horizontality indicated by each spirit level 414 after the leveling work from data recorded in advance during leveling work performed in the industrial machine 2 in advance and data obtained through additional experiments, and then performing a learning process until sufficient accuracy is obtained.

[0039] The level estimation unit 140 according to this embodiment estimates the level after the leveling work has been performed based on the current level before the leveling work, the current height of the legs 412 of the stand 410 of the industrial machine 2 at the attachment position before the leveling work, and the adjustment value of the legs 412 of the industrial machine 2 during the leveling work. The second model storage unit 220 stores a second model that has been trained in advance, with the level before leveling, the height of the legs 412 of the stand 410 of the industrial machine 2 at the attachment position before leveling, and the adjustment value of the legs 412 of the industrial machine 2 during the leveling work as explanatory variables (input), and the level after leveling as a target variable (output). The level degree estimation unit 140 inputs the current level of the industrial machine 2 acquired by the level degree acquisition unit 110, the current height of the industrial machine 2 at the attachment position of the legs 412 of the stand 410 acquired by the height acquisition unit 115, and the adjustment value estimated by the adjustment value estimation unit 130 into the second model read out from the second model storage unit 220, and outputs the result as an estimation result of the level after the leveling work has been performed. The level degree estimation unit 140 outputs the estimated level after leveling to the reliability calculation unit 150.

[0040] The auxiliary device 1 according to this embodiment having the above configuration can estimate the reliability of the adjustment value in leveling estimated by the first model using the second model. When estimating the adjustment value and the level, by using the height of the mounting position of the legs 412 of the stand 410 of the industrial machine 2 in addition to the level before leveling, it becomes possible to indirectly consider the influence of the material and strength of the floor surface on which the industrial machine 2 is installed on the level of the stand 410 before adjustment.

[0041] Third Embodiment An auxiliary device according to a third embodiment will be described below. The auxiliary device 1 according to this embodiment has the same hardware configuration as the auxiliary device 1 according to the first embodiment.

[0042] 8 is a schematic block diagram illustrating functions provided in the auxiliary device 1 according to the third embodiment of the present disclosure. The auxiliary device 1 has functions provided in a general computer in addition to the functions illustrated in FIG. 8. Each function provided in the auxiliary device 1 according to this embodiment is realized by the CPU 11 of the auxiliary device 1 and the processor 101 of the machine learning device 100 shown in FIG. 1 executing a system program and controlling the operation of each part of the auxiliary device 1 and the machine learning device 100.

[0043] Similar to the auxiliary device 1 according to the first embodiment, the auxiliary device 1 according to the present embodiment includes a levelness acquisition unit 110, a target value setting unit 120, an adjustment value estimation unit 130, a levelness estimation unit 140, a reliability calculation unit 150, and an output unit 160. In addition, the RAM 13 to the nonvolatile memory 14 of the auxiliary device 1 are provided in advance with a first model storage unit 210 that stores a first model, which is a model for estimating an adjustment value of the adjustment mechanism during leveling based on the levelness before leveling and the target levelness after leveling, and a second model storage unit 220 that stores a second model that estimates the levelness after leveling based on the levelness before leveling and the adjustment value of the adjustment unit during leveling.

[0044] The functions of the horizontality acquisition unit 110, the horizontality estimation unit 140, and the output unit 160 according to this embodiment are similar to those of the first embodiment.

[0045] The target value setting unit 120 according to this embodiment acquires a range of target values ​​for the level of the platform during leveling work. This range of target values ​​for level becomes a set of ranges of target values ​​for each of the levels 414 attached to the platform 410 of the industrial machine 2 to be measured after the adjustment work. Next, the target value setting unit 120 creates a set of predetermined sets of levels within the acquired range of target values ​​for level according to a predetermined rule. For example, suppose that a range of target values ​​for level indicated by each of the eight levels 414a to 414h is given as −0.1 to +0.1. In this case, suppose that a rule has been set in advance for creating the set of sets of level: "For each level, the target value for level is allowed to be in increments of 0.1 within the set range, and all combination patterns of all possible values ​​for the target value of level are created as a set." In this case, the target value setting unit 120 sets (target value of the level of the level 414a, target value of the level of the level 414b, ..., target value of the level of the level 414h) (-0.1, -0.1, ..., -0.1), (target value of the level of the level 414a, target value of the level of the level 414b, ..., target value of the level of the level 414h) (0 ... , (target value of level of level of level 414a, target value of level of level of level 414b, ..., target value of level of level of level 414h,) (-0.1, -0.1, ..., -0.1), ..., (target value of level of level of level of level 414a, target value of level of level of level of level 414b, ..., target value of level of level of level of level 414h,) (+0.1, +0.1, ..., +0.1). Then, target value setting unit 120 sets these sets of target values ​​of levelness as target values, and then instructs adjustment value estimation unit 130 to estimate an adjustment value.

[0046] The target value setting unit 120 may, for example, display on the display device 70 a screen for inputting the range of target values ​​for the level of each level 414, and prompt the worker to input the range of target values ​​for the level of the level 414. Fig. 9 is a screen diagram showing an example of an input screen for the range of target values ​​for level displayed on the display device 70 by the target value setting unit 120. The worker inputs the range of target values ​​for the level indicated by each level 414 into a target value input field 126. Then, by pressing an OK button 128, the target value setting unit 120 acquires the range of target values ​​for level input input by the worker.

[0047] The reliability calculation unit 150 according to this embodiment compares the levelness after leveling estimated by the levelness estimation unit 140 with the target value of the levelness after leveling set by the target value setting unit 120. Then, based on the comparison result, it calculates the reliability c of the adjustment value of the leg 412 of the industrial machine 2 in the leveling work estimated by the adjustment value estimation unit 130. The reliability calculation unit 150 calculates the reliability c for each set of target values ​​set by the target value setting unit 120. Then, of the reliabilities calculated for all sets of target values ​​for leveling, it selects the one with the highest reliability. Then, it outputs an adjustment value estimated based on the set of target values ​​for leveling with the highest reliability to the output unit 160.

[0048] The auxiliary device 1 according to this embodiment, which is configured as described above, exploratoryally estimates the adjustment value within the range of target values ​​for levelness set by the target value setting unit 120. Then, it outputs the adjustment value with the highest reliability. This makes it possible to optimize the adjustment value so that it has the highest reliability within the allowable range of levelness after leveling, which is expected to reduce the risk of having to redo the work.

[0049] Other Embodiments In the above-described embodiment, an example has been shown in which the adjustment value of the adjustment mechanism for leveling is an adjustment amount from the current position. However, the height of the industrial machine 2 at the attachment position of the legs 412 of the pedestal 410 before the leveling work is performed can also be used as the adjustment value of the adjustment mechanism for leveling. This height may be the distance between the bottom surface of the pedestal 410 and the floor contact surface of the legs 412 at the attachment position of each leg 412 attached to the pedestal 410. For example, if the adjustment mechanism of the legs 412 is equipped with a gauge or the like that measures height based on a predetermined standard, the measurement value of the gauge may be used as the height at the attachment position of the legs 412. This can be achieved by using the above-described values ​​as the adjustment values ​​of the adjustment mechanism for leveling used in advance in learning the first and second models.

[0050] Although the embodiments of the present disclosure have been described in detail above, the present disclosure is not limited to the individual embodiments described above. Various additions, substitutions, modifications, partial deletions, etc. are possible in these embodiments without departing from the gist of the invention or the idea and intent of the present disclosure derived from the content described in the claims and their equivalents. For example, in the above-described embodiments, the order of each operation and the order of each process are shown as examples and are not limited to these. The same applies when numerical values ​​or mathematical expressions are used in the description of the above-described embodiments.

[0051] The following are supplementary notes related to embodiments of the present disclosure. (Supplementary Note 1) An auxiliary device (1) according to one aspect of the present disclosure includes a levelness acquisition unit (110) that acquires a levelness of a platform of an industrial machine (2) before leveling, a target value setting unit (120) that sets a target value for the levelness of the platform of the industrial machine (2) after leveling, an adjustment value estimation unit (130) that estimates an adjustment value for the adjustment in leveling based on the levelness of the platform of the industrial machine (2) before leveling acquired by the levelness acquisition unit (110) and the target value for the levelness of the platform of the industrial machine (2) after leveling set by the target value setting unit (120) using a first model trained with the levelness of the platform of the industrial machine (2) before leveling and the target value for the levelness of the platform of the industrial machine (2) after leveling as explanatory variables and an adjustment value for the adjustment in leveling as a response variable, and The system includes: a levelness estimation unit (140) that estimates the levelness after leveling based on the levelness of the base of the industrial machine (2) before leveling acquired by the levelness acquisition unit (110) and the adjustment value estimated by the adjustment value estimation unit (130) using a second model that has been trained using the adjustment value acquired as an explanatory variable and the levelness of the base of the industrial machine (2) after leveling as a target variable; a reliability calculation unit (150) that calculates the reliability of the adjustment value estimated by the adjustment value estimation unit (130) based on the target value of the levelness of the base of the industrial machine (2) after leveling set by the target value setting unit (120) and the levelness of the base of the industrial machine (2) after leveling estimated by the levelness estimation unit (110); and an output unit (160) that outputs the adjustment value estimated by the adjustment value estimation unit (130) and the reliability calculated by the reliability calculation unit (150).

[0052] (Supplementary Note 2) An auxiliary device (1) according to another aspect of the present disclosure further includes a height acquisition unit (115) that acquires the height of the stand of the industrial machine (2) before leveling, and the first model is learned using the levelness of the stand of the industrial machine (2) before leveling, the stand height of the industrial machine (2) before leveling, and a target value of the levelness of the stand of the industrial machine (2) after leveling as explanatory variables, and an adjustment value related to the adjustment in leveling as a target variable, and the adjustment value estimation unit (130) uses the first model to estimate the levelness of the stand of the industrial machine (2) before leveling acquired by the levelness acquisition unit (110), the stand height of the industrial machine (2) before leveling acquired by the height acquisition unit, and the target value of the stand of the industrial machine (2) after leveling set by the target value setting unit (120). The level degree estimation unit (140) estimates an adjustment value for adjustment during leveling based on a target value for the level of the base of the industrial machine (2), and the second model is learned using the level of the base of the industrial machine (2) before leveling, the base height of the industrial machine (2) before leveling, and the adjustment value for adjustment during leveling as explanatory variables, and the level of the base of the industrial machine (2) after leveling as a target variable, and the level degree estimation unit (140) uses the second model to estimate the level after leveling based on the level of the base of the industrial machine (2) before leveling acquired by the level degree acquisition unit (110), the base height of the industrial machine (2) before leveling acquired by the height acquisition unit (115), and the adjustment value estimated by the adjustment value estimation unit (130).

[0053] (Supplementary Note 3) The target value setting unit (120) provided in the auxiliary device (1) according to another aspect of the present disclosure acquires a range of target values ​​for the levelness of the platform of the industrial machine (2) after leveling, and creates a set of multiple target values ​​for the levelness within the range of target values ​​for the levelness of the platform of the industrial machine (2) after leveling based on a predetermined rule, the adjustment value estimation unit (130) estimates adjustment values ​​for all of the multiple target values ​​for the levelness created by the target value setting unit (120), the levelness estimation unit (140) estimates the levelness after leveling for all the adjustment values ​​estimated by the adjustment value estimation unit (130), the reliability calculation unit (150) calculates reliability for all the adjustment values ​​estimated by the adjustment value estimation unit (130), and the output unit (160) outputs the adjustment value with the highest reliability among the adjustment values ​​estimated by the adjustment value estimation unit (130).

[0054] (Supplementary Note 4) A computer-readable recording medium according to one aspect of the present disclosure includes a levelness acquisition unit (110) that acquires a levelness of a platform of an industrial machine (2) before leveling, a target value setting unit (120) that sets a target value of the levelness of the platform of the industrial machine (2) after leveling, and an adjustment value estimation unit (130) that estimates an adjustment value related to the adjustment in leveling based on the levelness of the platform of the industrial machine (2) before leveling acquired by the levelness acquisition unit (110) and the target value of the levelness of the platform of the industrial machine (2) after leveling set by the target value setting unit (120) using a first model that has been trained with the levelness of the platform of the industrial machine (2) before leveling and the target value of the levelness of the platform of the industrial machine (2) after leveling as explanatory variables and an adjustment value related to the adjustment in leveling as a response variable. The computer is recorded with a program that causes the computer to operate as: a levelness estimation unit (140) that estimates the levelness after leveling based on the levelness of the base of the industrial machine (2) before leveling acquired by the levelness acquisition unit (110) and the adjustment value estimated by the adjustment value estimation unit (130) using a second model that has been trained with the levelness of the base of the industrial machine (2) after leveling as an explanatory variable and the levelness of the base of the industrial machine (2) after leveling as a target variable; a reliability calculation unit (150) that calculates the reliability of the adjustment value estimated by the adjustment value estimation unit (130) based on the target value of the levelness of the base of the industrial machine (2) after leveling set by the target value setting unit (120) and the levelness of the base of the industrial machine (2) after leveling estimated by the levelness estimation unit (140); and an output unit (160) that outputs the adjustment value estimated by the adjustment value estimation unit (130) and the reliability calculated by the reliability calculation unit (150).

[0055] REFERENCE SIGNS LIST 1 Auxiliary device 2 Industrial machine 5 Network 6 Fog computer 7 Cloud server 11 CPU 12 ROM 13 RAM 14 Non-volatile memory 15, 18, 19, 20, 21 Interface 22 Bus 70 Display device 71 Input device 72 External device 100 Machine learning device 101 Processor 102 ROM 103 RAM 104 Non-volatile memory 110 Levelness acquisition unit 115 Height acquisition unit 120 Target value setting unit 130 Adjustment value estimation unit 140 Levelness estimation unit 150 Reliability calculation unit 160 Output unit 210 First model storage unit 220 Second model storage unit

Claims

1. A level degree acquisition unit that acquires the level of the industrial machinery platform before leveling; a target value setting unit that sets a target value for the level of the industrial machinery platform after leveling; an adjustment value estimation unit that estimates an adjustment value for adjustment during leveling based on the level of the industrial machinery platform before leveling acquired by the level degree acquisition unit and the target value for the level of the industrial machinery platform after leveling set by the target value setting unit, using a first model trained with the level of the industrial machinery platform before leveling and the target value for the level of the industrial machinery platform after leveling as explanatory variables and an adjustment value for adjustment during leveling as a dependent variable; a level degree estimation unit that estimates the level after leveling based on the level of the industrial machinery platform before leveling acquired by the level degree acquisition unit and the adjustment value estimated by the adjustment value estimation unit, using a second model trained with the level of the industrial machinery platform before leveling and the adjustment value for adjustment during leveling as explanatory variables and an an adjustment value estimating unit that outputs the adjustment value estimated by the adjustment value estimating unit and the reliability calculated by the reliability calculating unit, based on a target value for the level of the base of the industrial machine after leveling set by the target value setting unit and the level of the base of the industrial machine after leveling estimated by the level estimating unit; and an output unit that outputs the adjustment value estimated by the adjustment value estimating unit and the reliability calculated by the reliability calculating unit.

2. The system further comprises a height acquisition unit that acquires the height of the industrial machinery stand before leveling, wherein the first model is trained using the levelness of the industrial machinery stand before leveling, the height of the industrial machinery stand before leveling, and a target value for the levelness of the industrial machinery stand after leveling as explanatory variables, and an adjustment value related to adjustment in leveling as a dependent variable; the adjustment value estimation unit uses the first model to estimate an adjustment value related to adjustment in leveling based on the levelness of the industrial machinery stand before leveling acquired by the levelness acquisition unit, the height of the industrial machinery stand before leveling acquired by the height acquisition unit, and the target value for the levelness of the industrial machinery stand after leveling set by the target value setting unit; and the second model is trained using the levelness of the industrial machinery stand before leveling, the height of the industrial machinery stand before leveling, and an adjustment value related to adjustment in leveling as explanatory variables, and a levelness of the industrial machinery stand after leveling as a dependent variable; 2. The auxiliary device according to claim 1, wherein the level estimation unit uses the second model to estimate the level after leveling based on the level of the base of the industrial machine before leveling acquired by the level acquisition unit, the height of the base of the industrial machine before leveling acquired by the height acquisition unit, and the adjustment value estimated by the adjustment value estimation unit.

3. The auxiliary device described in claim 2, wherein the target value setting unit obtains a range of target values ​​for the levelness of the industrial machinery platform after leveling, and creates a set of multiple target values ​​for levelness within the range of target values ​​for the levelness of the industrial machinery platform after leveling based on a predetermined rule; the adjustment value estimation unit estimates adjustment values ​​for all of the multiple target values ​​for levelness created by the target value setting unit; the levelness estimation unit estimates the levelness after leveling for all adjustment values ​​estimated by the adjustment value estimation unit; the reliability calculation unit calculates reliability for all adjustment values ​​estimated by the adjustment value estimation unit; and the output unit outputs the adjustment value estimated by the adjustment value estimation unit that has the highest reliability.

4. A levelness acquisition unit that acquires the levelness of the industrial machinery platform before leveling; A target value setting unit that sets a target value for the levelness of the industrial machinery platform after leveling; An adjustment value estimation unit that estimates an adjustment value for adjustment during leveling based on the levelness of the industrial machinery platform before leveling acquired by the levelness acquisition unit and the target value for the levelness of the industrial machinery platform after leveling set by the target value setting unit, using a first model trained with the levelness of the industrial machinery platform before leveling and the target value for the levelness of the industrial machinery platform after leveling as explanatory variables and an adjustment value for adjustment during leveling as a dependent variable; A levelness estimation unit that estimates the levelness after leveling based on the levelness of the industrial machinery platform before leveling acquired by the levelness acquisition unit and the adjustment value estimated by the adjustment value estimation unit, using a second model trained with the levelness of the industrial machinery platform before leveling and the adjustment value for adjustment during leveling as explanatory variables and an a reliability calculation unit that calculates the reliability of the adjustment value estimated by the adjustment value estimation unit based on the target value of the level of the base of the industrial machine after leveling set by the target value setting unit and the level of the base of the industrial machine after leveling estimated by the level estimation unit; and an output unit that outputs the adjustment value estimated by the adjustment value estimation unit and the reliability calculated by the reliability calculation unit.

Images