Lifetime estimation method, lifetime estimation device, and computer program

Abstract

To provide a life estimation method that can accumulate physical quantity data indicating the state of a predetermined part constituting an industrial machine and calculate a failure timing or failure probability of the predetermined part.SOLUTION: The life estimation method acquires physical quantity data indicating the state of a predetermined part constituting an industrial machine, stores the acquired physical quantity data and time data indicating the time of acquisition of the physical quantity data in association with each other, calculates a function for estimating changes over time in parameter values correlated with the life of the predetermined part based on the acquired physical quantity data and time data, and calculates a failure time or failure probability of the predetermined part using the calculated function.SELECTED DRAWING: Figure 6

Description

[Technical field]

[0001] The present invention relates to a lifespan estimation method, a lifespan estimation device, and a computer program. [Background technology]

[0002] An injection molding machine includes an injection device that melts and injects molding material, and a mold clamping device. The injection device includes a heating cylinder with a nozzle at its tip, and a screw that is arranged within the heating cylinder so as to be rotatable in the circumferential and axial directions. The screw is driven in the rotational and axial directions by a drive mechanism. The drive mechanism includes a ball screw that converts the rotational drive force of an injection servo motor into axial drive force for the screw and transmits it (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2019-166702 A Summary of the Invention [Problem to be solved by the invention]

[0004] By the way, the time when the ball screw should be inspected is displayed based on the operating conditions and operating time of the injection molding machine. However, even if the inspection time is reached, the damage to the ball screw varies from injection molding machine to injection molding machine. As a result, there are few cases where the injection molding machine is inspected even when the inspection time is reached.

[0005] An object of the present invention is to provide a lifespan estimation method, a lifespan estimation device, and a computer program that can accumulate physical quantity data indicating the state of specific parts that constitute industrial machinery and calculate the failure time or failure probability of the specific parts. [Means for solving the problem]

[0006] A life estimation method according to one embodiment of the present invention acquires physical quantity data indicating a state of a specific part that constitutes an industrial machine, stores the acquired physical quantity data and time data indicating when the physical quantity data was acquired in correspondence with each other, calculates a function for estimating a change over time in a parameter value that is correlated with the life of the specific part based on the acquired physical quantity data and the time data, and calculates a failure time or failure probability of the specific part using the calculated function.

[0007] A life estimation device according to one embodiment of the present invention includes an acquisition unit that acquires physical quantity data indicating a state of a specific part that constitutes an industrial machine, a memory unit that corresponds and stores the acquired physical quantity data and time data indicating the time when the physical quantity data was acquired, and a calculation unit that calculates a function for estimating a change over time in a parameter value correlated with the life of the specific part based on the acquired physical quantity data and the time data, and calculates a failure time or failure probability of the specific part using the calculated function.

[0008] A computer program according to one aspect of the present invention causes a computer to execute a process of acquiring physical quantity data indicating a state of a specific part of an industrial machine, storing the acquired physical quantity data and time data indicating a time when the physical quantity data was acquired in association with each other, calculating a function for estimating a change over time in a parameter value correlated with a life of the specific part based on the acquired physical quantity data and the time data, and calculating a failure time or failure probability of the specific part using the calculated function. Effect of the Invention

[0009] According to the above, physical quantity data indicating the state of a predetermined part constituting an industrial machine can be accumulated, and a failure time or failure probability of the predetermined part can be calculated. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a schematic diagram showing a configuration example of an injection molding machine according to a first embodiment. [Diagram 2]1 is a cross-sectional view showing an example of the configuration of a drive device of an injection molding machine according to a first embodiment. [Diagram 3] 5 is a flowchart showing a processing procedure of a calculation unit according to the first embodiment. [Figure 4] FIG. 11 is an explanatory diagram showing a method of calculating a function for estimating a change over time in a vibration acceleration peak value that is correlated with the life of a ball screw. [Diagram 5] FIG. 11 is an explanatory diagram showing a method for calculating the failure probability and failure time of a ball screw over its life. [Figure 6] FIG. 13 is a schematic diagram showing an example of a display screen for an estimation result. [Figure 7] 10 is a flowchart showing a processing procedure of a calculation unit according to the second embodiment. [Figure 8] FIG. 11 is a schematic diagram showing a configuration example of a control device according to a third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Specific examples of an injection molding machine (life estimation device), a life estimation method, and a computer program according to embodiments of the present invention will be described below with reference to the drawings. At least some of the embodiments described below may be combined in any manner. Note that the present invention is not limited to these examples, but is defined by the claims, and is intended to include all modifications within the meaning and scope equivalent to the claims.

[0012] 1 is a schematic diagram showing a configuration example of an injection molding machine 1 according to embodiment 1. The injection molding machine 1 according to embodiment 1 includes a clamping device 2 that clamps a mold 21, an injection device 3 that melts and injects a molding material, and a control device 4. The control device 4 functions as a life estimation device according to embodiment 1.

[0013] The mold clamping device 2 includes a fixed platen 22 fixed on a bed 20, a mold clamping housing 23 slidably provided on the bed 20, and a movable platen 24 that similarly slides on the bed 20. The fixed platen 22 and the mold clamping housing 23 are connected by a plurality of, for example, four tie bars 25, 25, .... The movable platen 24 is configured to be slidable between the fixed platen 22 and the mold clamping housing 23. A mold clamping mechanism 26 is provided between the mold clamping housing 23 and the movable platen 24. The mold clamping mechanism 26 is configured, for example, by a toggle mechanism. The mold clamping mechanism 26 may be configured as a direct pressure type mold clamping mechanism, that is, by a mold clamping cylinder. The fixed platen 22 and the movable platen 24 are provided with a fixed mold 28 and a movable mold 27, respectively, and the mold 21 is opened and closed when the mold clamping mechanism 26 is driven.

[0014] The injection device 3 is provided on a base 30. The injection device 3 includes a heating cylinder 31 having a nozzle 31a at its tip, and a screw 32 arranged within the heating cylinder 31 so as to be rotatable in the circumferential and axial directions. A heater for melting the molding material is provided inside or on the outer periphery of the heating cylinder 31. The screw 32 is driven by a drive device 5 in the rotational and axial directions.

[0015] A hopper 33 into which molding material is poured is provided near the rear end of the heating cylinder 31. The injection molding machine 1 also includes a nozzle touch device 34 that moves the injection unit 3 in the front-rear direction (the left-right direction in FIG. 1). When the nozzle touch device 34 is driven, the injection unit 3 moves forward so that the nozzle 31a of the heating cylinder 31 touches the contact portion of the fixed platen 22.

[0016] The control device 4 is a computer that controls the operations of the clamping device 2 and the injection device 3, and includes a processor (arithmetic unit) 41, a storage unit 42, an operation unit 43, an acquisition unit 44, and a display unit 45 as a hardware configuration. The control device 4 may be a server device connected to a network. The control device 4 may be configured to perform distributed processing by using multiple computers, may be realized by multiple virtual machines provided in one server, or may be realized by using a cloud server.

[0017] The processor 41 includes arithmetic circuits such as a CPU (Central Processing Unit), a multi-core CPU, a GPU (Graphics Processing Unit), a GPGPU (General-purpose computing on graphics processing units), a TPU (Tensor Processing Unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), and an NPU (Neural Processing Unit), internal storage devices such as a ROM (Read Only Memory) and a RAM (Random Access Memory), an I / O terminal, a timer, etc. The processor 41 implements the control method according to the first embodiment by executing a computer program (program product) 42a stored in a storage unit 42 described later.

[0018] The storage unit 42 includes a hard disk, an EEPROM (Electrically Erasable Programmable Read Only Memory), The storage unit 42 is a non-volatile memory such as a ROM, a flash memory, etc. The storage unit 42 stores a computer program 42a for causing a computer to execute a process for calculating the failure time and failure probability of the ball screw 51.

[0019] The computer program 42a according to the first embodiment may be recorded in a computer-readable manner on the recording medium 6. The storage unit 42 stores the computer program 42a read from the recording medium 6 by a reading device. The recording medium 6 is a semiconductor memory such as a flash memory. The recording medium 6 may also be an optical disk such as a CD (Compact Disc)-ROM, a DVD (Digital Versatile Disc)-ROM, or a BD (Blu-ray (registered trademark) Disc). The recording medium 6 may also be a magnetic disk such as a flexible disk or a hard disk, a magnetic optical disk, or the like. The computer program 42a according to the first embodiment may also be downloaded from an external server connected to a communication network and stored in the storage unit 42.

[0020] The operation unit 43 is an input device such as a touch panel, soft keys, hard keys, a keyboard, and a mouse. The acquisition unit 44 performs AD conversion on a rotation angle signal output from an encoder 50d (described later) and an acceleration signal output from the acceleration sensor 5a, and acquires rotation angle data and acceleration data. The display unit 45 is a liquid crystal panel, an organic EL display, electronic paper, a plasma display, etc. The display unit 45 displays various information according to image data provided from the processor 41.

[0021] In the injection molding machine 1, set values ​​are set to determine molding conditions such as injection start time, resin temperature in the mold, nozzle temperature, cylinder temperature, hopper temperature, mold clamping force, injection speed, injection acceleration, injection peak pressure, injection stroke, etc. In addition, set values ​​are set to determine molding conditions such as cylinder tip resin pressure, back pressure switching pressure, holding pressure switching speed, holding pressure switching position, holding pressure completion position, cushion position, metering back pressure, metering torque, etc. In addition, set values ​​are set to determine molding conditions such as metering completion position, screw 32 retreat speed, cycle time, mold closing time, injection time, holding pressure time, metering time, mold opening time, etc. In addition, the injection molding machine 1, to which these set values ​​are set, operates according to the set values.

[0022] 2 is a cross-sectional view showing a configuration example of the drive unit 5 of the injection molding machine 1 according to the first embodiment. The drive unit 5 includes an injection servomotor 50 and a ball screw 51 for driving the screw 32 in the axial direction. The injection servomotor 50 is provided with an encoder 50d that detects a rotation angle and outputs a rotation angle signal indicating the rotation angle to the control device 4. The control device 4 controls the rotation of the injection servomotor 50 based on the rotation angle signal output from the encoder 50d. A small pulley 50a is provided on the output shaft of the injection servomotor 50.

[0023] The ball screw 51 includes a ball screw shaft 51a and a nut 51b screwed onto the ball screw shaft 51a. A base end of the ball screw shaft 51a is rotatably supported via a bearing 52a on a first plate 52 having a hole and a bearing seat. A large pulley 50c is provided on the base end of the ball screw shaft 51a. The small pulley 50a and the large pulley 50c are connected by a timing belt 50b, and the rotational force of the small pulley 50a is reduced and transmitted to the large pulley 50c, causing the ball screw shaft 51a to rotate. Hereinafter, the direction toward the base end of the ball screw shaft 51a (right side in FIG. 1) will be referred to as the backward direction, and the opposite direction (left side in FIG. 1) will be referred to as the forward direction. The forward and backward directions will be collectively referred to as the forward and backward directions. When the injection servo motor 50 is driven to rotate the large pulley 50c and the ball screw shaft 51a, the nut 51b moves in the forward or backward direction depending on the direction of rotation.

[0024] A load cell 53 is provided on the surface of the nut 51b facing the forward direction, and the surface of the load cell 53 facing the forward direction is fixed to a second plate 54. A plurality of through holes are formed in the second plate 54, and a guide shaft 55 is inserted into the through holes. The second plate 54 moves in the forward and backward directions while being guided by the guide shaft 55. A third plate 56 is provided on the forward direction side of the first plate 52, and one end and the other end of the guide shaft 55 are supported by the first plate 52 and the third plate 56, respectively. When the ball screw shaft 51a rotates, the nut 51b, the load cell 53, and the second plate 54 move integrally in the forward and backward directions along the guide shaft 55.

[0025] The second plate 54 is provided with a hole and a bearing seat, and the output shaft 58 is supported in the hole via a bearing 54a. The output shaft 58 is provided with a plasticizing pulley 57. The plasticizing pulley 57 is connected to a pulley attached to a motor for rotating the screw 32 (not shown) via a timing belt (not shown). The center of rotation of the output shaft 58 coincides with the center of rotation of the ball screw shaft 51a. The output shaft 58 is formed with a recess into which the tip of the ball screw shaft 51a enters when the nut 51b moves forward and backward. In addition, one end of the screw 32 is fixed to the output shaft 58 so that the central axes coincide. The third plate 56 is formed with a through hole through which the screw 32 is inserted. One end of the heating cylinder 31 is fixed to the third plate 56 so that the screw 32 inserted through the through hole of the third plate 56 can move axially inside the heating cylinder 31.

[0026] Furthermore, an acceleration sensor 5a that detects vibration of the nut 51b is attached to the nut 51b of the ball screw 51. The acceleration sensor 5a outputs a detected acceleration signal to the control device 4. The processor 41 of the control device 4 acquires the acceleration signal output from the acceleration sensor 5a by AD conversion in the acquisition unit 44 into acceleration data. The control device 4 can estimate the life of the ball screw 51 based on the acceleration data.

[0027] The molding process cycle is outlined below, and the control device 4 performs a process of sequentially moving the forward and backward movement range of the nut 51b in the repeated molding process cycle. During injection molding, the well-known mold closing process, mold clamping process, injection unit advancement process, injection process, measurement process, injection unit retreat process, mold opening process, and ejection process are performed in this order.

[0028] 3 is a flowchart showing a processing procedure of the processor 41 according to the embodiment 1. The processor 41 acquires acceleration data indicating the state of the ball screw 51 from the acceleration sensor 5a via the acquisition unit 44 (step S111).

[0029] Then, the calculation unit associates the acquired acceleration data with time data indicating the time when the acceleration data was acquired, and stores the data in the storage unit 42 (step S112). The time data is obtained from the clock unit.

[0030] Next, the calculation unit calculates a first function for estimating a change over time in the acceleration peak value correlated with the life of the ball screw 51, based on the acceleration data and the time data stored in the memory unit 42 (step S113). The acceleration peak value is an example of a parameter value correlated with the life of the ball screw 51.

[0031] Fig. 4 is an explanatory diagram showing a method of calculating a function for estimating the change over time of the vibration acceleration peak value correlated with the life of the ball screw 51. The horizontal axis of the graph shown in Fig. 4 indicates time, and the vertical axis indicates the acceleration peak value. The time on the horizontal axis corresponds to the operating time of the ball screw 51, and the acceleration peak value corresponds to the degree of failure of the ball screw 51. The processor 41 uses a maximum likelihood estimation method to find a first function that indicates the change over time of the acceleration peak value, based on the acceleration data and time data accumulated in the memory unit 42 over a predetermined period of time.

[0032] For example, the first function is expressed by the following equation. P = a × e bt +c…(1) however, P: Peak acceleration value t: time a, b, c: coefficients

[0033] The processor 41 obtains the first function by calculating the coefficients a, b, and c that minimize the mean square error.

[0034] Next, as shown by the dashed line in FIG. 4, the processor 41 calculates a second function for estimating the change over time of a value obtained by adding a standard deviation (predetermined deviation) of a normal distribution to the acceleration peak value calculated using the first function (step S114). Specifically, the processor 41 calculates the standard deviation at each of a plurality of time points based on the acceleration peak value (average value) calculated using the first function at a plurality of time points and the acceleration data and time data as sample data stored in the storage unit 42. Then, a value is obtained by adding the standard deviation at each time point to each of the acceleration peak values ​​(average value) at the respective time points. The processor 41 then calculates a second function indicating the change over time of the calculated value by the maximum likelihood estimation method. The second function is also expressed using an exponential function as shown in the above formula (1).

[0035] Furthermore, the processor 41 calculates a third function for estimating a change over time of a value obtained by subtracting the standard deviation of the normal distribution from the meter values ​​at a plurality of points in time (step S115). The calculation method of the third function is the same as the calculation method of the second function.

[0036] Next, the processor 41 calculates the failure time and failure probability of the ball screw 51 using the first function (step S116). Similarly, the processor 41 calculates the failure time and failure probability of the ball screw 51 using the second function (step S117). In addition, the processor 41 calculates the failure time and failure probability of the ball screw 51 using the third function (step S118).

[0037] 5 is an explanatory diagram showing a method of calculating the failure probability and failure time of the life of a ball screw 51. For example, as shown in FIG. 5A, the processor 41 calculates the time point at which the acceleration peak value calculated using a first function reaches a predetermined failure determination threshold as the failure time at which the failure probability becomes 50% (first probability). The failure determination threshold may be stored in the storage unit 42 in advance, or the processor 41 may be configured to receive it from the operator via the operation unit 43. Furthermore, the processor 41 calculates the time when the acceleration peak value calculated using the second function reaches a predetermined failure determination threshold as the failure time when the failure probability becomes 16% (second probability). Furthermore, the processor 41 calculates the time when the acceleration peak value calculated using the third function reaches a predetermined failure determination threshold as the failure time when the failure probability becomes 84% ​​(third probability).

[0038] In the above example, the failure time is estimated from the intersection points of the first function, the second function, and the third function with the failure determination threshold. However, as shown in FIG. 5B, the failure probability at a specific estimated reference time point may be calculated. The processor 41 uses the first function to obtain the acceleration peak value (average value) at a specific estimated reference time point. Then, the processor 41 calculates the variance or standard deviation at the estimated reference time point based on the average value of the acceleration peak values ​​and the acceleration data and time data as samples stored in the storage unit 42, and calculates the probability that the acceleration peak value becomes the failure determination threshold at the estimated reference time point. In other words, it is only necessary to assume that the acceleration peak value at the estimated reference time point follows a normal distribution and to obtain the probability that the acceleration peak value becomes the failure determination threshold.

[0039] Next, the processor 41 displays the actual measurement value graph 45a and the estimated value graphs 45b, 45c, 45d, etc. on the display unit 45 (step S119), and ends the process.

[0040] 6 is a schematic diagram showing an example of a display screen of the estimation results. Processor 41 displays the horizontal and vertical axes of the graph on display unit 45, and displays graph lines of actual measurement value graph 45a and estimated value graphs 45b, 45c, and 45d. The horizontal axis indicates the operating time corresponding to the elapsed time, and the vertical axis indicates the failure degree corresponding to the acceleration peak value. The actual measurement value graph 45a is a graph showing the actual measurement value of the change over time of the acceleration peak value based on the acceleration data and time data stored in the memory unit 42. The estimated value graphs 45b, 45c, and 45d are graphs showing the change over time of the acceleration peak value obtained by the first function, the second function, and the third function, respectively. The processor 41 also displays a threshold line image 45f indicating the failure determination threshold on the display unit 45. The processor 41 also displays a normal distribution image 45e indicating the elapsed time from the present time and the failure probability of the ball screw 51.

[0041] The operator of the injection molding machine 1 can visually recognize the current actual state of the ball screw 51, the time of failure, and the failure probability from the actual measurement value graph 45a and the estimated value graphs 45b, 45c, and 45d displayed on the display unit 45.

[0042] As described above, according to the injection molding machine 1 of the first embodiment, the acceleration data indicating the state of the ball screw 51 of the injection molding machine 1 can be accumulated, and the failure time and failure probability of the ball screw 51 can be calculated.

[0043] Specifically, a first function that indicates the relationship between time and the acceleration peak value is obtained using the maximum likelihood estimation method, and the point in time at which the acceleration peak value calculated using the first function reaches a predetermined failure determination threshold can be calculated as the failure time with a 50% failure probability. Furthermore, the time point at which the acceleration peak value calculated using the second function reaches a predetermined failure determination threshold can be calculated as the failure time with a failure probability of 16%. Furthermore, the time point at which the acceleration peak value calculated using the third function reaches a predetermined failure determination threshold can be calculated as the failure time with a failure probability of 84%.

[0044] In addition, the current state of the ball screw 51 can be displayed using the actual measurement value graph 45a and the estimated value graphs 45b, 45c, and 45d.

[0045] In the first embodiment, an example has been described in which the failure time and failure probability of the ball screw 51 are estimated based on the vibration of the ball screw 51, but physical quantity data indicating the current or torque of the injection servo motor 50 may also be used. The processor 41 calculates a function indicating the change over time in the peak value of the current or torque based on the physical quantity data indicating the current or torque of the injection servo motor 50, and can calculate the failure time and failure probability of the ball screw 51 in the same manner as in the first embodiment. Also, the configuration may be such that the failure time and failure probability of the ball screw 51 are calculated based on the concentration of iron powder contained in the lubricating oil of the ball screw 51. Further, in the first embodiment, an example has been described in which the failure time and failure probability of the ball screw 51 is calculated, but the failure time and failure probability of the mold clamping unit 2 and the injection unit 3 may be calculated. Furthermore, the apparatus may be configured to acquire physical quantity data indicating the state of a specific part constituting the industrial machine, and calculate the failure time and failure probability of the specific part.

[0046] Furthermore, in the first embodiment, the second and third functions that estimate the value obtained by adding and subtracting the standard deviation from the average value of the acceleration peak values ​​are exemplified, but the present invention is not limited to these. A function that estimates the value obtained by adding and subtracting an arbitrary predetermined deviation from the average value of the acceleration peak values ​​may be calculated, and the failure time and failure probability may be estimated using the function.

[0047] (Embodiment 2) The injection molding machine 1 according to the second embodiment differs from the first embodiment in that the failure time and failure probability are calculated for each part of the ball screw shaft 51a. Since the other configurations of the injection molding machine 1 are similar to those of the injection molding machine 1 according to the first embodiment, the same reference numerals are used for the similar parts and detailed description is omitted.

[0048] 7 is a flowchart showing the processing procedure of the calculation unit according to the second embodiment. The processor 41 acquires acceleration data indicating the state of the ball screw 51 from the acceleration sensor 5a via the acquisition unit 44 (step S211). The processor 41 also acquires angle data, in other words, position data indicating the position of the nut 51b, from the encoder 50d via the acquisition unit 44 (step S211). The rotation angle of the injection servo motor 50 corresponds to the position of the nut 51b on the ball screw shaft 51a.

[0049] Then, the calculation unit stores the acquired acceleration data, position data, and time data indicating the time when the acceleration data and the position data were acquired in association with each other in the storage unit 42 (step S212). The time data is obtained from the clock unit.

[0050] Next, the calculation unit calculates a first function for estimating the change over time in the acceleration peak value that is correlated with the life of the ball screw 51, for each part of the ball screw shaft 51a, based on the acceleration data, position data, and time data stored in the memory unit 42 (step S213).

[0051] Next, the processor 41 calculates a second function for estimating a change over time in a value obtained by adding the standard deviation of a normal distribution to the acceleration peak value calculated using the first function, for each portion of the ball screw shaft 51a (step S214). Furthermore, the processor 41 calculates a third function for estimating a change over time of a value obtained by subtracting the standard deviation of a normal distribution from the meter values ​​at a plurality of points in time, for each portion of the ball screw shaft 51a (step S215). The methods for calculating the first to third functions are the same as those in the first embodiment.

[0052] Next, the processor 41 calculates the failure time and failure probability of the ball screw 51 for each part of the ball screw shaft 51a using the first function (step S216). Similarly, the processor 41 calculates the failure time and failure probability of the ball screw 51 for each part of the ball screw shaft 51a using the second function (step S217). In addition, the processor 41 calculates the failure time and failure probability of the ball screw 51 for each part of the ball screw shaft 51a using the third function (step S218). The method of calculating the failure time and failure probability using the first to third functions is the same as in the first embodiment.

[0053] Then, the processor 41 displays the actual measurement value graph 45a and the estimated value graphs 45b, 45c, 45d, etc. on the display unit 45 (step S219), and ends the process.

[0054] According to the injection molding machine 1 of this embodiment 2, acceleration data indicating the state of the ball screw 51 of the injection molding machine 1 can be accumulated, and the failure time and failure probability of the ball screw 51 can be calculated for each portion of the ball screw shaft 51a.

[0055] (Embodiment 3) The injection molding machine 1 according to the third embodiment differs from the first embodiment in that the failure time and failure probability of the ball screw shaft 51a are calculated by machine learning. Since the other configurations of the injection molding machine 1 are similar to those of the injection molding machine 1 according to the first embodiment, the same reference numerals are used for the similar parts and detailed description will be omitted.

[0056] 8 is a schematic diagram showing a configuration example of the control device 4 according to the third embodiment. The processor 41 of the control device 4 according to the third embodiment includes a learning processing unit 41a as a functional unit. The storage unit 42 stores a learning model 42b for estimating the failure time and failure probability of the ball screw 51. The learning processing unit 41a may be realized by software or may be configured by hardware. A part of the learning processing unit 41a may be configured by hardware.

[0057] The processor 41 reads out the learning model 42b and the computer program 42a from the storage unit 42 and executes them.

[0058] The learning model 42b is a neural network that outputs a failure probability when the peak value of the vibration acceleration and the elapsed time, which is the operating time of the ball screw 51, are input. The learning model 42b includes an input layer, a hidden layer, and an output layer. The input layer has a plurality of nodes to which the peak value of the vibration acceleration and the elapsed time are input. The hidden layer includes a plurality of intermediate layers each having a plurality of nodes, and the nodes of the intermediate layer on the input side are connected to the nodes of the input layer. The output layer has nodes that output the failure probability. Each node of the output layer is connected to a node of the intermediate layer on the output side.

[0059] The learning model 42b can be generated by machine learning using training data including the peak value and elapsed time of the vibration acceleration and the failure probability. For example, the learning processing unit 41a of the processor 41 optimizes the weighting coefficient of the learning model 42b by an error backpropagation method, an error gradient descent method, or the like using the training data, thereby performing machine learning on the learning model 42b.

[0060] In addition, the learning model 42b may be configured to have nodes corresponding to multiple elapsed times in the output layer, and when a peak value of vibration acceleration is input, a failure probability is output from the node corresponding to each elapsed time.

[0061] The processor 41 can calculate the failure probability of the ball screw 51 by inputting the peak value of the vibration acceleration detected at the present time and an arbitrary elapsed time into the learning model 42b. The processor 41 may use the relationship between the elapsed time and the failure probability calculated using the learning model 42b to display the actual measurement value graph 45a and the estimated value graphs 45b, 45c, and 45d on the display unit 45 in the same manner as in the first embodiment.

[0062] According to the injection molding machine 1 of this embodiment 3, as in embodiment 1, the failure time and failure probability of the ball screw 51 can be calculated and the actual measurement value graph 45a and estimated value graphs 45b, 45c, and 45d can be displayed on the display unit 45.

[0063] In addition, in the third embodiment, the learning model 42b using a neural network has been described, but the model may be configured to estimate the failure time and failure probability using other known machine learning models such as a Support Vector Machine (SVM) or a Bayesian network. [Explanation of symbols]

[0064] 1 injection molding machine 2 Mold clamping device 3 Injection device 4. Control device 5. Drive unit 5a Acceleration sensor 6 Recording media 31 Heating Cylinder 32 Screw 50 Injection servo motor 50d Encoder 51 Ball screw 41 Processors 51 Ball screw 51a Ball screw shaft 51b Nut

Claims

1. Acquire physical quantity data indicating the state of a specific part constituting the industrial machine; storing the acquired physical quantity data and time data indicating the time when the physical quantity data was acquired in association with each other; calculating a function for estimating a change over time in a parameter value correlated with a life of the predetermined portion based on the acquired physical quantity data and the acquired time data; The calculated function is used to calculate the failure time or failure probability of the predetermined portion. Lifespan estimation method.

2. The time when the parameter value calculated using the function reaches a failure determination threshold is calculated as the failure time. The method for estimating a lifetime according to claim 1 .

3. The time point when the parameter value calculated using the function calculated by the maximum likelihood estimation method reaches the failure determination threshold is calculated as the failure time when the failure probability is 50%. The method for estimating a lifetime according to claim 2 .

4. calculating a first function for estimating a change in the parameter value over time by a maximum likelihood estimation method; calculating a second function for estimating a change over time of a value obtained by adding a predetermined deviation to the parameter value calculated using the first function; calculating a third function for estimating a change over time in a value obtained by subtracting the predetermined deviation from the parameter value calculated using the first function; calculating a time point at which the parameter value calculated using the first function reaches the failure determination threshold as a failure time when the failure probability is a first probability; calculating a time point at which the parameter value calculated using the second function reaches the failure determination threshold as a failure time when the failure probability is a second probability; A time point when the parameter value calculated using the third function reaches the failure determination threshold is calculated as a failure time when the failure probability is a third probability. The method for estimating a lifetime according to claim 2 .

5. a measured value graph showing a change over time of the parameter value based on the physical quantity data and the time data, an estimated value graph showing the first function, the second function, and the third function, and the fault determination threshold value are displayed; The method for estimating a lifetime according to claim 4.

6. the industrial machine is a molding machine having a ball screw, Calculate the time or probability of failure of the ball screw The method for estimating a lifetime according to any one of claims 1 to 5.

7. The physical quantity data is data indicating a vibration acceleration of the ball screw, and a current or torque of a motor that drives the ball screw, and the parameter value is a peak value of the vibration acceleration, the current, or the torque. The method for estimating a lifetime according to claim 6.

8. The molding machine includes a screw that is provided in a cylinder having a nozzle at a tip end and that can be driven in a rotational direction and an axial direction, and the ball screw has a rotatably provided ball screw shaft and a nut that is screwed onto the ball screw shaft and moves forward and backward as the ball screw shaft rotates, and the screw is driven in the axial direction by the forward and backward movement of the nut. acquiring position data indicating a position of the nut relative to the ball screw and the physical quantity data indicating a state of the ball screw at the position; storing the acquired position data, the physical quantity data, and the time data indicating the time when the physical quantity data was acquired in association with each other; The function for estimating the change over time of the parameter value correlated with the life of the ball screw is calculated for each portion of the ball screw shaft based on the acquired position data, the physical quantity data, and the time data. The method for estimating a lifetime according to claim 7.

9. an acquisition unit that acquires physical quantity data indicating a state of a predetermined part of the industrial machine; a storage unit that stores the acquired physical quantity data and time data indicating the acquisition time of the physical quantity data in association with each other; a calculation unit that calculates a function for estimating a change over time of a parameter value correlated with a life of the predetermined portion based on the acquired physical quantity data and the acquired time data, and calculates a failure time or a failure probability of the predetermined portion using the calculated function; A lifespan estimation device comprising:

10. Acquire physical quantity data indicating the state of a specific part constituting the industrial machine; storing the acquired physical quantity data and time data indicating the time when the physical quantity data was acquired in association with each other; calculating a function for estimating a change over time in a parameter value correlated with a life of the predetermined portion based on the acquired physical quantity data and the acquired time data; The calculated function is used to calculate the failure time or failure probability of the predetermined portion. A computer program for causing a computer to execute a process.

Images