Failure prediction device, failure prediction system, and failure prediction method

Abstract

A failure prediction device (2) predicts failure of an electromagnetic brake. The failure prediction device (2) comprises: a data storage unit (22) that stores a first current value which is the value of a current measured in a first region which is a period in which a current flowing through an excitation coil increases from the start of voltage application to an excitation coil, and a second current value that is the value of a current measured in a second region that is a period in which the current increases to a steady value after the first region; an estimation unit (24) that estimates a wear amount of a lining material on the basis of a first induction coefficient which is an induction coefficient in the first region of a circuit including the excitation coil and is obtained on the basis of the first current value, and a second induction coefficient which is an induction coefficient in the second region of the circuit and is obtained on the basis the second current value; and an estimation result output unit (25) that outputs the estimation result by the estimation unit (24).

Description

Technical Field

[0001] The present disclosure relates to a failure prediction device, a failure prediction system, and a failure prediction method for predicting a failure of an electromagnetic brake.

Background Art

[0002] An electromagnetic brake of the non-excitation operation type is used for a servo motor used in a robot arm for the purpose of holding the position when the motor stops. The electromagnetic brake has a lining material, an armature, an exciting coil, a spring, and a yoke. The lining material is connected to the shaft of the servo motor and rotates together with the shaft. Parts other than the lining material of the electromagnetic brake are not connected to the shaft and do not rotate.

[0003] When no voltage is applied to the exciting coil, the brake is actuated by bringing the armature into contact with the lining material by the force of the spring. In this state, when a voltage is applied to the exciting coil and a current flows through the exciting coil, the armature is attracted to the yoke by the magnetic flux generated in the magnetic circuit, and the brake is released.

[0004] When the actuation and release of the brake are repeated, the air gap between the armature and the yoke increases due to the wear of the lining material, and the current required for releasing the brake increases. When the wear amount of the lining material reaches a certain amount, the current required for releasing the brake becomes larger than the supplyable current, and the brake cannot be released.

[0005] Since the current at the time of brake release changes as the wear amount of the lining material increases, the wear amount of the lining material can be estimated based on the current at the time of brake release. By estimating the wear amount of the lining material, a failure of the electromagnetic brake can be predicted.

[0006] Patent Document 1 discloses a device for estimating the air gap of an electromagnetic actuator without using a search coil for detecting magnetic flux. This device estimates the air gap by measuring the time constant of the current flowing through the excitation coil when a constant voltage is applied to the excitation coil, and calculating the induction coefficient of the excitation coil from this time constant.

[0007] Patent Document 2 discloses a method for estimating the air gap of an electromagnetic brake based on the time-dependent change in the electromagnetic wave pattern during the period from the start of energization of the excitation coil until the current decreases. [Prior art documents] [Patent Documents]

[0008] [Patent Document 1] Japanese Patent Publication No. 2007-40975 [Patent Document 2] Japanese Patent Application Publication No. 7-187566 [Overview of the Initiative] [Problems that the invention aims to solve]

[0009] However, the conventional technology disclosed in Patent Document 1 or 2 above has the problem that the accuracy of fault prediction decreases when the resistance value of the excitation coil changes. The resistance value of the excitation coil can change due to changes in the temperature of the excitation coil caused by heat generation of the excitation coil or the influence of the external environment of the electromagnetic brake. Alternatively, the resistance value of the excitation coil can change due to differences in the length of the cables that make up the electrical circuit of the electromagnetic brake. Furthermore, if the temperature of the excitation coil is measured in order to improve the decrease in fault prediction accuracy due to temperature changes of the excitation coil, the manufacturing cost of the fault prediction configuration will increase as a temperature sensor is added to the fault prediction configuration.

[0010] This disclosure has been made in view of the above, and aims to provide a fault prediction device that can predict electromagnetic brake failures with high accuracy. [Means for solving the problem]

[0011] To solve the aforementioned problems and achieve the objective, the failure prediction device according to this disclosure is a failure prediction device that predicts failures of an electromagnetic brake. The electromagnetic brake has a lining material integrated with a shaft that applies braking force, an armature, and an excitation coil that generates an electromagnetic force to drive the armature. When no voltage is applied to the excitation coil, the armature is brought into contact with the lining material to keep the shaft stationary, and when a voltage is applied to the excitation coil, the armature is separated from the lining material to release the stationary position of the shaft. The fault prediction device according to this disclosure includes: a data storage unit that stores a first current value, which is the value of the current measured in a first region, which is the period when the current flowing through the excitation coil increases from the time a voltage is applied to the excitation coil; a second current value, which is the value of the current measured in a second region, which is the period after the first region when the current increases to a steady value; an estimation unit that estimates the amount of wear of the lining material based on a first induction coefficient in the first region of the circuit including the excitation coil, which is determined based on the first current value; and an estimation result output unit that outputs the estimated amount of wear. [Effects of the Invention]

[0012] The failure prediction device described herein has the effect of being able to predict electromagnetic brake failures with high accuracy. [Brief explanation of the drawing]

[0013] [Figure 1] This figure shows an example of the configuration of an electromagnetic brake that is the target of fault prediction by the fault prediction device according to Embodiment 1. [Figure 2] This figure shows an example configuration of the fault prediction device according to Embodiment 1. [Figure 3] A diagram illustrating the current waveform recorded by the fault prediction device according to Embodiment 1. [Figure 4]Figure showing an example of an equivalent circuit of an electromagnetic brake that is the target of failure prediction by the failure prediction device according to Embodiment 1 [Figure 5] Figure for explaining the processing in the data processing unit of the failure prediction device according to Embodiment 1 [Figure 6] Figure for explaining the relationship between the inductance ratio and the wear amount of the lining material in Embodiment 1 [Figure 7] Flowchart showing an example of the procedure of the processing executed by the failure prediction device according to Embodiment 1 [Figure 8] Figure for explaining a modification of the processing in the data processing unit of the failure prediction device according to Embodiment 1 [Figure 9] Figure for explaining the relationship between the inductance ratio and the temperature of the excitation coil in Embodiment 2 [Figure 10] Figure for explaining the calculation of the temperature of the excitation coil in Embodiment 2 [Figure 11] Figure showing a configuration example of the control circuit according to Embodiment 1 or 2 [Figure 12] Figure showing a configuration example of a dedicated hardware circuit according to Embodiment 1 or 2 [Figure 13] Figure showing a configuration example of a servo motor including an electromagnetic brake that is the target of failure prediction by the failure prediction system according to Embodiment 3 [Figure 14] Figure showing a configuration example of the failure prediction system according to Embodiment 3

Mode for Carrying Out the Invention

[0014] Hereinafter, the failure prediction device, failure prediction system, and failure prediction method according to the embodiment will be described in detail based on the drawings.

[0015] Embodiment 1. Prior to describing the fault prediction device according to Embodiment 1, the configuration of the electromagnetic brake targeted for fault prediction will be described. Figure 1 is a diagram showing an example of the configuration of an electromagnetic brake targeted for fault prediction by the fault prediction device according to Embodiment 1. The electromagnetic brake 1 shown in Figure 1 is an example of an electromagnetic brake targeted for fault prediction. The configuration of the electromagnetic brake 1 shown in Figure 1 is a basic configuration that is generally provided in an electromagnetic brake.

[0016] The electromagnetic brake 1 comprises a yoke 10, an excitation coil 11, a spring 12, a lining material 13, and an armature 14. The lining material 13 is integrated with the shaft that applies the braking force and rotates with the shaft. The other components of the electromagnetic brake 1, the yoke 10, the excitation coil 11, the spring 12, and the armature 14, are not connected to the shaft and do not rotate.

[0017] Here, the operating principle of the electromagnetic brake 1 will be explained. When no voltage is applied to the excitation coil 11, the electromagnetic brake 1 uses the force of the spring 12 to bring the armature 14 into contact with the lining material 13. The brake is activated by bringing the armature 14 into contact with the lining material 13. In this way, the electromagnetic brake 1 brings the armature 14 into contact with the lining material 13 when no voltage is applied to the excitation coil 11, thereby keeping the shaft stationary.

[0018] When the shaft is stationary, a voltage is applied to the excitation coil 11, causing current to flow through it. This generates a magnetic flux in the magnetic circuit, which consists of the yoke 10, the armature 14, and the air gap between the yoke 10 and the armature 14. The magnetic flux generated in the magnetic circuit attracts the armature 14 to the yoke 10, causing the armature 14 to separate from the lining material 13 and releasing the brake. In this way, the electromagnetic brake 1 releases the stationary position of the shaft by separating the armature 14 from the lining material 13 when a voltage is applied to the excitation coil 11.

[0019] Next, the failure mechanism of the electromagnetic brake 1 will be explained. When the electromagnetic brake 1 is repeatedly operated and released, the lining material 13 wears down, and the air gap between the yoke 10 and the armature 14 increases. As the air gap increases, the current required to release the brake also increases. When the amount of wear of the lining material 13 reaches a certain level, the current required to release the brake becomes greater than the available current, making it impossible to release the brake. Therefore, the electromagnetic brake 1 may fail due to the wear of the lining material 13, resulting in the inability to release the brake.

[0020] The failure prediction device according to Embodiment 1 predicts failure of the electromagnetic brake 1 by estimating the amount of wear of the lining material 13. By predicting failure of the electromagnetic brake 1, it is possible to prevent failures in which the brake cannot be released due to wear of the lining material 13. As the amount of wear of the lining material 13 increases, the current at the time of brake release changes, so the failure prediction device can estimate the amount of wear of the lining material 13 based on the current at the time of brake release.

[0021] Next, the configuration of the fault prediction device according to Embodiment 1 will be described. Figure 2 is a diagram showing an example of the configuration of the fault prediction device 2 according to Embodiment 1. The fault prediction device 2 includes a data acquisition unit 21, a data storage unit 22, a data processing unit 23, an estimation unit 24, and an estimation result output unit 25.

[0022] The data acquisition unit 21 acquires data used for fault prediction in the fault prediction device 2. The data storage unit 22 stores the data acquired by the data acquisition unit 21. The data processing unit 23 reads data from the data storage unit 22 and processes the read data. The estimation unit 24 estimates the amount of wear of the lining material 13 based on the data processed by the data processing unit 23. The estimation result output unit 25 outputs information indicating the amount of wear estimated by the estimation unit 24. In other words, the estimation result output unit 25 outputs the estimated amount of wear.

[0023] Next, the process performed by the fault prediction device 2 according to Embodiment 1 will be described. The data acquisition unit 21 acquires the value of the current flowing through the excitation coil 11. The data acquisition unit 21 stores the acquired current value in the data storage unit 22. By storing the current value in the data storage unit 22, the fault prediction device 2 records a current waveform that represents the change in current when a voltage is applied to the excitation coil 11.

[0024] Figure 3 is a diagram illustrating the current waveform recorded by the fault prediction device 2 according to Embodiment 1. Figure 4 is a diagram showing an example of the equivalent circuit of the electromagnetic brake 1 that is the target of fault prediction by the fault prediction device 2 according to Embodiment 1. As shown in Figure 4, the electrical circuit of the electromagnetic brake 1 can be represented as an RL series circuit having a resistor 32 and a coil 33.

[0025] Let R be the resistance of resistor 32, L be the inductance coefficient of coil 33, and E be the DC voltage applied to the electrical circuit by DC power supply 31. Then, time t i The current I(t) is the current when this condition is met. i ) is expressed by the following equation (1): t a This is the time when the voltage application began.

[0026]

number

[0027] In Figure 3, the horizontal axis represents time t [s] and the vertical axis represents current i [A]. Figure 3 shows examples of current waveforms when the electromagnetic brake 1 is functioning normally and when the electromagnetic brake 1 is malfunctioning. Normal operation of the electromagnetic brake 1 means that the lining material 13 is not worn. Malfunctioning operation of the electromagnetic brake 1 means that the lining material 13 is worn to the point where the brake cannot be released.

[0028] As shown in Figure 3, under normal conditions, the current increases with a constant time constant from the start of voltage application to the excitation coil 11. The current continues to increase monotonically, then decreases once, and then continues to increase monotonically again. From the start of voltage application until the current reaches a certain value, the armature 14 is in contact with the lining material 13, and when the current reaches that value, the armature 14 is pulled toward the yoke 10. As soon as the current reaches that value, the current decreases. Then, when the armature 14 reaches the yoke 10, the current increases again. Therefore, the current continues to increase monotonically, then decreases once, and then continues to increase monotonically again.

[0029] In the following description, the period from the time when voltage is applied to the excitation coil 11 to the time when the current flowing through the excitation coil 11 increases is referred to as the first region. After the first region, the period from the time when the current decreases once and then starts to increase again to the time when the current increases to a steady value is referred to as the second region. Since the position of the armature 14 is different in the first region and the second region, the induction coefficient in the first region and the induction coefficient in the second region are different from each other. In the following description, the induction coefficient in the first region is referred to as the first induction coefficient, and the induction coefficient in the second region is referred to as the second induction coefficient. In Embodiment 1, the induction coefficient is the induction coefficient of the electrical circuit of the electromagnetic brake 1, and is the induction coefficient of the circuit including the excitation coil 11.

[0030] The data acquisition unit 21 acquires the first current value, which is the current value measured in the first region, and the second current value, which is the current value measured in the second region. The data storage unit 22 stores the first current value and the second current value. The portion of the current waveform in Figure 3 that represents the first region represents the change in the first current value. The portion of the current waveform in Figure 3 that represents the change in the second current value represents the change in the second current value. The double arrows in Figure 3 represent the first region and the second region during abnormal conditions.

[0031] During abnormal conditions, the armature 14 is further from the yoke 10 than during normal conditions. Therefore, the first induction coefficient during abnormal conditions is smaller than that during normal conditions. Because the first induction coefficient during abnormal conditions is smaller than that during normal conditions, the change in current with respect to time in the first region is larger during abnormal conditions than during normal conditions. Thus, since the first induction coefficient during abnormal conditions is smaller than that during normal conditions, if the first induction coefficient can be estimated, a failure of the electromagnetic brake 1 can be predicted. The first induction coefficient can be calculated from equation (1). However, since the first induction coefficient is calculated by incorporating the resistance value, the failure prediction result may be affected by the temperature change of the excitation coil 11.

[0032] In Embodiment 1, the failure prediction device 2 estimates the induction coefficient ratio, which is the ratio of the first induction coefficient to the second induction coefficient. The failure prediction device 2 estimates the induction coefficient ratio by calculating the induction coefficient ratio based on the first current value and the second current value using the data processing unit 23. The estimation unit 24 estimates the amount of wear of the lining material 13 based on the induction coefficient ratio. Thus, the estimation unit 24 estimates the amount of wear of the lining material 13 based on the first induction coefficient obtained based on the first current value and the second induction coefficient obtained based on the second current value.

[0033] Thus, the failure prediction device 2 incorporates the first and second induction coefficients into the calculation for estimating the amount of wear of the lining material 13, and estimates the amount of wear of the lining material 13 based on the induction coefficient ratio. In this case, the failure prediction device 2 can eliminate the influence of the resistance value of the excitation coil 11 from the process of estimating the amount of wear of the lining material 13. Since the resistance value of the excitation coil 11 can change depending on the temperature or the length of the cable, the failure prediction device 2 can improve the accuracy of wear estimation by eliminating the influence of the resistance value of the excitation coil 11. Therefore, the failure prediction device 2 can estimate the amount of wear with higher accuracy compared to when the amount of wear is estimated from the first induction coefficient alone.

[0034] The fault prediction device 2 eliminates the influence of the resistance value of the excitation coil 11, thus eliminating the need for measures such as measuring the temperature of the excitation coil 11 to improve the accuracy of fault prediction. Since a temperature sensor is not required in the fault prediction configuration, it is possible to avoid the problem of increased manufacturing costs for the fault prediction configuration.

[0035] Figure 5 is a diagram illustrating the processing in the data processing unit 23 of the fault prediction device 2 according to Embodiment 1. Figure 5 shows the current waveform of the current flowing through the excitation coil 11. In Figure 5, the horizontal axis represents time t [s] and the vertical axis represents current i [A].

[0036] The data processing unit 23 sets an arbitrary time t1 in the first region and an arbitrary time t2 in the second region. The data processing unit 23 obtains the current value i(t1) at t1 and the current value i(t2) at t2 from the current waveform recorded in the data storage unit 22. The data processing unit 23 obtains the current value i(t1+Δt) at t1+Δt, which is Δt advanced from t1, and the current value i(t1+2Δt) at t1+2Δt, which is 2Δt advanced from t1, from the current waveform. The data processing unit 23 obtains the current value i(t2+Δt) at t2+Δt, which is Δt advanced from t2, and the current value i(t2+2Δt) at t2+2Δt, which is 2Δt advanced from t2, from the current waveform.

[0037] Δt is assumed to be a predetermined period. t1, t1+Δt, and t1+2Δt are times that fall within the first region. t2, t2+Δt, and t2+2Δt are times that fall within the second region.

[0038] Here, let t1 be the first time step, t1+Δt be the second time step, t1+2Δt be the third time step, t2 be the fourth time step, t2+Δt be the fifth time step, and t2+2Δt be the sixth time step. The data processing unit 23 calculates Δi(t1)=i(t1+Δt)-i(t1), which is the difference in the first current value between the first time step and the second time step. The data processing unit 23 calculates Δi(t1+Δt)=i(t1+2Δt)-i(t1+Δt), which is the difference in the first current value between the second time step and the third time step. The data processing unit 23 calculates Δi(t2)=i(t2+Δt)-i(t2), which is the difference in the second current value between the fourth time step and the fifth time step. The data processing unit 23 calculates Δi(t2+Δt)=i(t2+2Δt)-i(t2+Δt), which is the difference in the second current value between the fifth time step and the sixth time step.

[0039] The data processing unit 23 calculates the induction coefficient ratio L2 / L1 using equations (2) to (4) shown below. L1 represents the first induction coefficient. L2 represents the second induction coefficient. α is a number expressed by equation (2) and is derived based on the difference in the first current value. β is a number expressed by equation (3) and is derived based on the difference in the second current value.

[0040]

number

[0041]

number

[0042]

number

[0043] In this way, the data processing unit 23 calculates the induction coefficient ratio by performing a calculation that incorporates the difference in the first current value between the first time point and the second time point, the difference in the first current value between the second time point and the third time point, the difference in the second current value between the fourth time point and the fifth time point, and the difference in the second current value between the fifth time point and the sixth time point.

[0044] Figure 6 is a diagram illustrating the relationship between the induction coefficient ratio and the amount of wear of the lining material 13 in Embodiment 1. In Figure 6, the vertical axis represents the induction coefficient ratio, and the horizontal axis represents the amount of wear of the lining material 13. Each of the three points shown in Figure 6 represents an example of the measured result of the amount of wear of the lining material 13 relative to the induction coefficient ratio. In Figure 6, the value indicating the amount of wear of the lining material 13 is assumed to be a normalized value where the amount of wear of the lining material 13 when the induction coefficient ratio is "1.0" is set to "1".

[0045] Each of the three points shown in Figure 6 represents the relationship between the induction coefficient ratio and the amount of wear of the lining material 13. Here, it is assumed that the first region and the second region of the current waveform are ideal waveforms following an RL series circuit. It is also assumed that the effect of temperature changes on the excitation coil 11 only affects the resistance value of the excitation coil 11. Since the induction coefficient ratio and the amount of wear of the lining material 13 are independent of the resistance value, the relationship between the induction coefficient ratio and the amount of wear of the lining material 13 shown at each point in Figure 6 does not change even if the temperature of the excitation coil 11 changes. Therefore, by pre-setting the relationship between the induction coefficient ratio and the amount of wear of the lining material 13, the amount of wear of the lining material 13 can be estimated based on the calculated induction coefficient ratio and this relationship.

[0046] The failure prediction device 2 has a pre-set relationship between the induction coefficient ratio and the amount of wear of the lining material 13. The estimation unit 24 refers to the relationship between the induction coefficient ratio and the amount of wear of the lining material 13 and calculates the amount of wear of the lining material 13 based on the calculated induction coefficient ratio and the said relationship.

[0047] An estimation formula representing the relationship between the induction coefficient ratio and the amount of wear of the lining material 13 is created, for example, by using a regression algorithm. The estimation unit 24 can calculate the amount of wear of the lining material 13 by substituting the calculated induction coefficient ratio into the estimation formula.

[0048] In the above explanation, a regression algorithm was used to create an estimation formula representing the relationship between the induction coefficient ratio and the amount of wear of the lining material 13. However, the estimation formula representing the relationship between the induction coefficient ratio and the amount of wear of the lining material 13 may also be created, for example, by classification or other machine learning methods.

[0049] Next, the procedure for processing performed by the fault prediction device 2 will be described. Figure 7 is a flowchart showing an example of the procedure for processing performed by the fault prediction device 2 according to Embodiment 1.

[0050] In step S1, the data acquisition unit 21 acquires the value of the current flowing through the excitation coil 11. The current value acquired by the data acquisition unit 21 includes a first current value and a second current value. Step S1 corresponds to the step of acquiring the first current value and the second current value.

[0051] In step S2, the data acquisition unit 21 stores the current value acquired in step S1 in the data storage unit 22. Step S2 corresponds to the step of storing the first current value and the second current value. By storing the current value in the data storage unit 22, the fault prediction device 2 records a current waveform that represents the change in current when a voltage is applied to the excitation coil 11.

[0052] In step S3, the data processing unit 23 calculates the induction coefficient ratio. As described above, the data processing unit 23 calculates the induction coefficient ratio by incorporating the difference in the first current value between the first time point and the second time point, the difference in the first current value between the second time point and the third time point, the difference in the second current value between the fourth time point and the fifth time point, and the difference in the second current value between the fifth time point and the sixth time point.

[0053] In step S4, the estimation unit 24 estimates the amount of wear of the lining material 13. As described above, the estimation unit 24 refers to the relationship between the induction coefficient ratio and the amount of wear of the lining material 13, and calculates the amount of wear of the lining material 13 based on this relationship and the induction coefficient ratio calculated in step S3. Steps S3 and S4 correspond to steps in which the amount of wear of the lining material 13 is estimated based on the first induction coefficient and the second induction coefficient.

[0054] In step S5, the estimation result output unit 25 outputs the estimated wear amount from step S4. With this, the failure prediction device 2 completes the processing according to the procedure shown in Figure 7. The degree of deterioration of the electromagnetic brake 1 can be determined from the wear amount estimation result output by the estimation result output unit 25.

[0055] The fault prediction device 2 can record the current waveform from the time voltage application to the excitation coil 11 starts until the current reaches a steady state by continuously acquiring the value of the current flowing through the excitation coil 11. Instead of recording the current waveform from the time voltage application to the excitation coil 11 starts until the current reaches a steady state, the fault prediction device 2 may acquire and store only the first current value at the first, second, and third time points, and the second current value at the fourth, fifth, and sixth time points. The fault prediction device 2 can estimate the amount of wear of the lining material 13 if it can acquire only these first and second current values. In this case, the fault prediction device 2 can reduce the processing load for acquiring and storing data compared to when it records the current waveform from the time voltage application to the excitation coil 11 starts until the current reaches a steady state. In addition, the fault prediction device 2 can reduce the amount of data required for fault prediction of the electromagnetic brake 1.

[0056] In actual environments where current is measured, measurement errors may occur at each sampling point, which is the time when the current value is measured. To reduce the impact of measurement errors, a process to smooth the current waveform may be added to the processing performed by the data processing unit 23. Alternatively, the fault prediction device 2 may obtain an approximate curve of the current waveform for at least one of the first and second regions and calculate the induction coefficient ratio based on the approximate curve.

[0057] Figure 8 is a diagram illustrating a modified example of the processing in the data processing unit 23 of the fault prediction device 2 according to Embodiment 1. Figure 8 shows examples of approximate curves of the current waveform in the first region and approximate curves of the current waveform in the second region. In Figure 8, the horizontal axis represents time t [s] and the vertical axis represents current i [A]. In Figure 8, the dashed line between the approximate curve of the current waveform in the first region and the approximate curve of the current waveform in the second region represents the change in current value between the first and second regions. In Figure 8, each of the multiple circles represents the current value measured at each sampling point.

[0058] The data processing unit 23 obtains an approximate curve of the current waveform in the first region based on the current value measured in the first region. That is, the data processing unit 23 obtains an approximate curve representing the relationship between the first current value and time based on the first current value measured at each sampling point in the first region. The data processing unit 23 obtains an approximate curve of the current waveform in the second region based on the current value measured in the second region. That is, the data processing unit 23 obtains an approximate curve representing the relationship between the second current value and time based on the second current value measured at each sampling point in the second region.

[0059] The data processing unit 23 determines the first current value at the first, second, and third time points from the approximate curve of the current waveform in the first region. The data processing unit 23 determines the second current value at the fourth, fifth, and sixth time points from the approximate curve of the current waveform in the second region. The fault prediction device 2 calculates the induction coefficient ratio based on these first and second current values.

[0060] As described above, according to this modified example, the data processing unit 23 obtains an approximation curve representing the relationship between the first current value and time, and an approximation curve representing the relationship between the second current value and time, and calculates the induction coefficient ratio. In this case as well, the fault prediction device 2 can estimate the amount of wear of the lining material 13 based on the induction coefficient ratio. Note that the data processing unit 23 only needs to obtain at least one of the approximation curve representing the relationship between the first current value and time, and the approximation curve representing the relationship between the second current value and time, and calculate the induction coefficient ratio.

[0061] According to Embodiment 1, the estimation unit 24 estimates the amount of wear of the lining material 13 based on the first induction coefficient and the second induction coefficient. The failure prediction device 2 can eliminate the problem that the accuracy of wear estimation decreases due to changes in the resistance value of the excitation coil 11 by incorporating the first induction coefficient and the second induction coefficient into the calculation for estimating the amount of wear of the lining material 13. Since the failure prediction device 2 can estimate the amount of wear of the lining material 13 with high accuracy, it can predict failure of the electromagnetic brake 1 due to wear of the lining material 13 with high accuracy. As a result, the failure prediction device 2 has the effect of being able to predict failure of the electromagnetic brake 1 with high accuracy.

[0062] Embodiment 2. Embodiment 2 describes an example in which the induction coefficient ratio is corrected based on the temperature of the excitation coil 11. The process described in Embodiment 2 is realized by the fault prediction device 2 shown in Figure 2. In Embodiment 2, the basic operating principle of the electromagnetic brake 1 and the fault prediction principle are the same as in Embodiment 1. Embodiment 2 mainly describes processes that differ from those in Embodiment 1.

[0063] In Embodiment 1, it was assumed that the effect of temperature changes in the excitation coil 11 only affected the resistance value of the excitation coil 11. However, in reality, when the temperature of the excitation coil 11 changes, the air gap between the yoke 10 and the armature 14 may change due to thermal expansion of the components constituting the electromagnetic brake 1. The induction coefficient ratio may change due to the change in the air gap.

[0064] Figure 9 is a diagram illustrating the relationship between the induction coefficient ratio and the temperature of the excitation coil 11 in Embodiment 2. In Figure 9, the vertical axis represents the induction coefficient ratio. The horizontal axis represents the temperature of the excitation coil 11 [deg.]. Each of the five points shown in Figure 9 represents an example of the result of determining the induction coefficient ratio with respect to the temperature of the excitation coil 11. The dashed lines in Figure 9 represent an example of the relationship between the induction coefficient ratio and the temperature of the excitation coil 11.

[0065] According to the relationship shown in Figure 9, the induction coefficient ratio changes as the temperature of the excitation coil 11 changes. However, the effect of temperature change on the induction coefficient ratio is sufficiently small compared to the effect of temperature change on the resistance value of the excitation coil 11. Due to the difference between the effect of temperature change on the induction coefficient ratio and the effect of temperature change on the resistance value, it is possible that the estimation accuracy of the amount of wear of the lining material 13 will decrease.

[0066] In Embodiment 2, the data processing unit 23 corrects the induction coefficient ratio based on the temperature of the excitation coil 11. The estimation unit 24 estimates the amount of wear of the lining material 13 based on the corrected induction coefficient ratio. As a result, the failure prediction device 2 can estimate the amount of wear of the lining material 13 with high accuracy.

[0067] The data processing unit 23 calculates the temperature of the excitation coil 11 and corrects the induction coefficient ratio based on the calculated temperature. Here, an example of how to calculate the temperature of the excitation coil 11 is described.

[0068] Figure 10 is a diagram illustrating the calculation of the temperature of the excitation coil 11 in Embodiment 2. Figure 10 shows an example of the current waveform from the time when voltage is applied to the excitation coil 11 until the current flowing through the excitation coil 11 reaches a steady state. In Figure 10, the horizontal axis represents time t [s] and the vertical axis represents current i [A].

[0069] The data acquisition unit 21 shown in Figure 2 acquires the current and voltage values ​​in the steady state shown in Figure 10. The data processing unit 23 acquires the current and voltage values ​​acquired by the data acquisition unit 21 via the data storage unit 22. Based on the acquired current and voltage values, the data processing unit 23 calculates the resistance value of the electrical circuit of the electromagnetic brake 1.

[0070] Here, if we let R1 be the resistance of the electrical circuit when the temperature of the excitation coil 11 is T1, and R2 be the resistance of the electrical circuit when the temperature of the excitation coil 11 is T2, then the following equation (5) holds.

[0071]

number

[0072] Assume that T1 is a predetermined temperature, and that R1, the resistance value at temperature T1, is predetermined. Also, assume that R2 is the resistance value calculated by the data processing unit 23. The data processing unit 23 calculates T2, the temperature of the excitation coil 11, by substituting the values ​​for T1, R1, and R2 in equation (5).

[0073] The fault prediction device 2 has a pre-set relationship between the induction coefficient ratio and the temperature of the excitation coil 11. In Figure 9, the dashed line is an example of a graph representing the relationship between the induction coefficient ratio and the temperature of the excitation coil 11. The data processing unit 23 calculates the induction coefficient ratio based on the calculated temperature and this relationship. In this way, the data processing unit 23 calculates an induction coefficient ratio that takes into account the temperature change of the excitation coil 11. In other words, the data processing unit 23 calculates an induction coefficient ratio that has been corrected based on the temperature of the excitation coil 11.

[0074] The estimation unit 24 estimates the amount of wear of the lining material 13 based on the corrected induction coefficient ratio. This allows the failure prediction device 2 to estimate the amount of wear of the lining material 13 with high accuracy. Furthermore, since the data processing unit 23 calculates the temperature of the excitation coil 11, a temperature sensor for measuring the temperature of the excitation coil 11 is unnecessary. Because a temperature sensor is not required for the failure prediction configuration, it is possible to avoid the problem of increased manufacturing costs for the failure prediction configuration.

[0075] In actual environments where current is measured, measurement errors can occur at each sampling point, which is the time when the current value is measured. To reduce the impact of measurement errors, a process to smooth the current waveform may be added to the processing performed by the data processing unit 23.

[0076] In the above explanation, the temperature of the excitation coil 11 calculated by the data processing unit 23 is used to correct the induction coefficient ratio. However, the temperature information calculated by the data processing unit 23 may be used for processing other than the correction of the induction coefficient ratio.

[0077] According to Embodiment 2, the data processing unit 23 corrects the induction coefficient ratio based on the temperature of the excitation coil 11. The estimation unit 24 estimates the amount of wear of the lining material 13 based on the corrected induction coefficient ratio. As a result, the failure prediction device 2 has the effect of being able to predict the failure of the electromagnetic brake 1 with high accuracy.

[0078] Next, the hardware that implements the fault prediction device 2 according to Embodiment 1 or 2 will be described. The data processing unit 23 and estimation unit 24 of the fault prediction device 2 are implemented by a processing circuit. The processing circuit may be a circuit in which a processor executes software, or it may be a dedicated circuit.

[0079] When the processing circuit is implemented by software, the processing circuit is, for example, the control circuit shown in Figure 11. Figure 11 is a diagram showing an example configuration of the control circuit 40 according to Embodiment 1 or 2. The control circuit 40 comprises an input unit 41, a processor 42, a memory 43, and an output unit 44. The input unit 41 is an interface circuit that receives data input from outside the control circuit 40 and provides it to the processor 42. The output unit 44 is an interface circuit that sends data from the processor 42 or the memory 43 to the outside of the control circuit 40. The data acquisition unit 21 is implemented by the input unit 41. The estimation result output unit 25 is implemented by the output unit 44.

[0080] If the processing circuit is the control circuit 40 shown in Figure 11, the data processing unit 23 and the estimation unit 24 are implemented by software, firmware, or a combination of software and firmware. The software or firmware is written as a program and stored in memory 43. In the processing circuit, the processor 42 reads and executes the program stored in memory 43 to realize each function of the fault prediction device 2. In other words, the processing circuit is equipped with memory 43 for storing the program that will ultimately execute the processing of the fault prediction device 2. These programs can also be said to cause the computer to execute the procedures and methods of the fault prediction device 2.

[0081] The processor 42 is a CPU (Central Processing Unit), processing unit, arithmetic unit, microprocessor, microcomputer, processor, or DSP (Digital Signal Processor). The memory 43 includes, for example, non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), EEPROM (Registered Trademark) (Electrically Erasable Programmable Read Only Memory), magnetic disks, flexible disks, optical disks, compact disks, minidiscs, or DVDs (Digital Versatile Discs). The data storage unit 22 is implemented by the memory 43.

[0082] Figure 11 shows an example of hardware when each component is implemented using a general-purpose processor 42 and memory 43, but each component may also be implemented using dedicated hardware circuits. Figure 12 shows an example of the configuration of a dedicated hardware circuit 45 according to Embodiment 1 or 2.

[0083] The dedicated hardware circuit 45 comprises an input section 41, an output section 44, and a processing circuit 46. The processing circuit 46 is a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a circuit combining these. Each function of the fault prediction device 2 may be implemented separately by the processing circuit 46, or all functions may be implemented together by the processing circuit 46. Note that each component may be implemented by combining the control circuit 40 and the hardware circuit 45.

[0084] Embodiment 3. Embodiment 3 describes an example of a fault prediction system having a fault prediction device. In Embodiment 3, the same reference numerals are used for components identical to those in Embodiment 1, and the description mainly focuses on configurations that differ from Embodiment 1. In Embodiment 3, the basic operating principle of the electromagnetic brake 1 and the fault prediction principle are the same as in Embodiment 1.

[0085] Embodiment 3 describes an example in which the controller that controls the servo motor is equipped with a fault prediction device function. Before describing the fault prediction system according to Embodiment 3, the configuration of the servo motor including the electromagnetic brake 1 that is subject to fault prediction will be described. Figure 13 is a diagram showing an example of the configuration of a servo motor 5 including the electromagnetic brake 1 that is subject to fault prediction by the fault prediction system according to Embodiment 3.

[0086] The servo motor 5 comprises a stator 53, a rotor 52 surrounded by the stator 53 and capable of rotation, a shaft 56 located at the rotation center of the rotor 52, and bearings 54 and 55 that rotatably support the shaft 56. The shaft 56 is the axis through which the braking force is applied. The servo motor 5 also includes an encoder 51 for detecting the rotational position of the shaft 56 and an electromagnetic brake 1. The lining material 13 of the electromagnetic brake 1 is connected to the shaft 56. The lining material 13 rotates together with the shaft 56.

[0087] Figure 14 shows an example configuration of the fault prediction system 8 according to Embodiment 3. Figure 14 shows a servo motor 5, a servo amplifier 6, and a controller 7. The servo amplifier 6 and the controller 7 constitute the fault prediction system 8. The power lines of the servo motor 5, the input / output lines of the encoder 51, and the power lines of the electromagnetic brake 1 are connected to the servo amplifier 6. Note that Figure 14 illustrates the components involved in fault prediction, and other components are not shown.

[0088] The controller 7 sends a command to the servo amplifier 6 to drive the servo motor 5. The servo amplifier 6 controls the servo motor 5 based on the command from the controller 7 and the information from the encoder 51. The servo amplifier 6 is equipped with a current sensor 61 and a voltage sensor 62. The current sensor 61 detects the current flowing through the electrical circuit of the electromagnetic brake 1. The voltage sensor 62 detects the voltage applied to the terminals of the electromagnetic brake 1. The current sensor 61 outputs the detected current value to the controller 7. The voltage sensor 62 outputs the detected voltage value to the controller 7. The controller 7 obtains the current value from the current sensor 61 and the voltage value from the voltage sensor 62. The controller 7 sends a command to the servo amplifier 6 to drive the electromagnetic brake 1. The servo amplifier 6 controls the electromagnetic brake 1 based on the command from the controller 7.

[0089] The controller 7 comprises a data acquisition unit 21, a data storage unit 22, a data processing unit 23, an estimation unit 24, and an estimation result output unit 25. The data acquisition unit 21, data storage unit 22, data processing unit 23, estimation unit 24, and estimation result output unit 25 are components for realizing the functions of the fault prediction device. The data acquisition unit 21 acquires the current value sent from the current sensor 61 to the controller 7. The data acquisition unit 21 stores the acquired current value in the data storage unit 22. Alternatively, the data acquisition unit 21 acquires the current value sent from the current sensor 61 to the controller 7 and the voltage value sent from the voltage sensor 62 to the controller 7. The data acquisition unit 21 stores the acquired current value and the acquired voltage value in the data storage unit 22.

[0090] A conventional servo amplifier used for controlling the servo motor 5 can be used as the servo amplifier 6. The fault prediction system 8 can use the current sensor 61 and voltage sensor 62 mounted on the servo amplifier 6 to acquire data for fault prediction. Since the fault prediction system 8 does not require separate current and voltage sensors from the servo amplifier 6, it can have a simple configuration. Compared to the case where separate current and voltage sensors are provided for the servo amplifier 6, the cost of the fault prediction system 8 can be reduced.

[0091] Controller 7 can be realized by adding the functionality of a fault prediction device to a conventional controller that generates commands to drive the servo motor 5. The fault prediction system 8 can be realized by utilizing a conventional servo amplifier and a conventional controller. Since there is no need to add a separate device to the fault prediction system 8 from the servo amplifier 6 and controller 7, the cost of the fault prediction system 8 can be reduced compared to when a separate device is provided from the servo amplifier 6 and controller 7.

[0092] Embodiment 3 describes an example of a fault prediction system 8 comprising a servo amplifier 6 and a controller 7, but the configuration of the fault prediction system 8 can be changed as appropriate. For example, at least one of the current sensor 61 and the voltage sensor 62 may be provided outside the servo amplifier 6. Alternatively, the components of the fault prediction device, namely the data acquisition unit 21, data storage unit 22, data processing unit 23, estimation unit 24, and estimation result output unit 25, may be provided in a device outside the controller 7.

[0093] The configurations shown in each of the embodiments described above are examples of the content of this disclosure. The configurations of each embodiment can be combined with other known technologies. The configurations of each embodiment may be combined with each other as appropriate. It is possible to omit or modify parts of the configurations of each embodiment without departing from the gist of this disclosure. [Explanation of symbols]

[0094] 1 Electromagnetic brake, 2 Fault prediction device, 5 Servo motor, 6 Servo amplifier, 7 Controller, 8 Fault prediction system, 10 Yoke, 11 Excitation coil, 12 Spring, 13 Lining material, 14 Armature, 21 Data acquisition unit, 22 Data storage unit, 23 Data processing unit, 24 Estimation unit, 25 Estimation result output unit, 31 DC power supply, 32 Resistor, 33 Coil, 40 Control circuit, 41 Input unit, 42 Processor, 43 Memory, 44 Output unit, 45 Hardware circuit, 46 Processing circuit, 51 Encoder, 52 Rotor, 53 Stator, 54, 55 Bearings, 56 Shaft, 61 Current sensor, 62 Voltage sensor.

Claims

1. A fault prediction device for an electromagnetic brake, comprising a lining material integrated with a shaft that applies braking force, an armature, and an excitation coil that generates an electromagnetic force to drive the armature, wherein when no voltage is applied to the excitation coil, the armature contacts the lining material to keep the shaft stationary, and when voltage is applied to the excitation coil, the armature separates from the lining material to release the stationary state of the shaft, A data storage unit that stores a first current value, which is the value of the current measured in a first region, which is the period during which the current flowing through the excitation coil increases from the time the voltage is applied to the excitation coil, and a second current value, which is the value of the current measured in a second region, which is the period after the first region during which the current increases to a steady state. An estimation unit that estimates the amount of wear of the lining material based on the induction coefficient in the first region of the circuit including the excitation coil, which is determined based on the first current value, and the induction coefficient in the second region of the circuit, which is determined based on the second current value. An estimation result output unit that outputs the estimated wear amount, A fault prediction device characterized by comprising the following features.

2. The system includes a data processing unit that calculates an induction coefficient ratio, which is the ratio of the first induction coefficient to the second induction coefficient, based on the first current value and the second current value. The failure prediction device according to claim 1, characterized in that the estimation unit estimates the amount of wear based on the induction coefficient ratio.

3. The fault prediction device according to claim 2, characterized in that the data processing unit calculates the induction coefficient ratio by a calculation that incorporates the difference between the first current value at a first time that is included in the first region and a second time that is a predetermined period elapsed from the first time; the difference between the first current value at a third time that is a predetermined period elapsed from the second time; the difference between the second current value at a fourth time that is included in the second region and a fifth time that is a predetermined period elapsed from the fourth time; and the difference between the second current value at a sixth time that is a predetermined period elapsed from the fifth time.

4. The fault prediction device according to claim 2, characterized in that the data processing unit acquires at least one of an approximation curve representing the relationship between the first current value and time, and an approximation curve representing the relationship between the second current value and time, and calculates the induction coefficient ratio.

5. The data processing unit corrects the induction coefficient ratio based on the temperature of the excitation coil, The failure prediction device according to any one of claims 2 to 4, characterized in that the estimation unit estimates the amount of wear based on the corrected induction coefficient ratio.

6. An electromagnetic brake comprising a lining material integrated with a shaft that applies braking force, an armature, and an excitation coil that generates an electromagnetic force to drive the armature, wherein when no voltage is applied to the excitation coil, the armature contacts the lining material to keep the shaft stationary, and when voltage is applied to the excitation coil, the armature separates from the lining material to release the stationary position of the shaft, A fault prediction device for predicting the failure of the electromagnetic brake, Equipped with, The fault prediction device, A data storage unit that stores a first current value, which is the value of the current measured in a first region, which is the period during which the current flowing through the excitation coil increases from the time the voltage is applied to the excitation coil, and a second current value, which is the value of the current measured in a second region, which is the period after the first region during which the current increases to a steady state. An estimation unit that estimates the amount of wear of the lining material based on the induction coefficient in the first region of the circuit including the excitation coil, which is determined based on the first current value, and the induction coefficient in the second region of the circuit, which is determined based on the second current value. An estimation result output unit that outputs the estimated wear amount, A failure prediction system characterized by comprising the following features.

7. A failure prediction method for an electromagnetic brake having a lining material integrated with a shaft that applies braking force, an armature, and an excitation coil that generates an electromagnetic force to drive the armature, wherein when no voltage is applied to the excitation coil, the armature is brought into contact with the lining material to keep the shaft stationary, and when a voltage is applied to the excitation coil, the armature is separated from the lining material to release the stationary state of the shaft, A step of storing a first current value, which is the value of the current measured in a first region, which is the period during which the current flowing through the excitation coil increases from the time the voltage is applied to the excitation coil, and a second current value, which is the value of the current measured in a second region, which is the period after the first region during which the current increases to a steady state. A step of estimating the amount of wear of the lining material based on the induction coefficient in the first region of the circuit including the excitation coil, which is determined based on the first current value, and the induction coefficient in the second region of the circuit, which is determined based on the second current value. The steps include outputting the estimated wear amount, A failure prediction method characterized by including the following.

Images