JP2008188646A - Method for detecting abnormality in plastic working, working system, and ae detector - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To correctly judge the generation of abnormality in plastic working. <P>SOLUTION: In the working system provided with a plastic working device 10, an AE sensor 21 and an AE signal processing section 22, the AE signal processing section includes: an A/D converter 33; a frequency analysis unit 34 analyzing the frequency characteristics of digital AE signals, and calculating the powers to frequencies per unit time; a memory unit 38 memorizing data obtained regarding the powers to frequencies per unit time upon normal time by time series as standard data; and a difference calculation unit 35 calculating power differences with the standard data in each frequency of the digital AE signals detected per unit time, adding the power differences in each frequency, and calculating power differences per unit time, so as to calculate power differences by time series. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a plastic working abnormality detection method for detecting an abnormality in a plastic working apparatus such as a press machine, a working system, and an AE detection apparatus having an AE (Acoustic Emission) sensor.

  In a press machine or the like, a processing abnormality may occur due to wear of the mold or damage to a part of the mold. Conventionally, such processing abnormality has been found by inspecting a processed product visually. However, processing with a press machine is performed at high speed, and since a considerable amount of time elapses between the occurrence of an abnormality and the detection of an abnormality in the processed product, many processed products become defective during that time. was there. Therefore, it has been desired to make it possible to find a processing abnormality during processing of a plastic processing apparatus such as a press machine.

  In Patent Document 1, an AE sensor is provided in a plastic processing apparatus, and an elastic wave generated in a structure of the plastic processing apparatus is detected by the AE sensor. It is described that when an abnormality is determined, an elastic wave is divided into a plurality of time portions corresponding to a plurality of elastic waves from the start of machining each time, and evaluation is performed with the elastic waves for each time portion.

  There are various types of AE sensors depending on the target frequency, but the AE sensor used in the plastic apparatus is intended for an ultrasonic region of several tens of kHz to several hundreds of MHz, and is composed of a piezoelectric element or the like. .

  Hereinafter, processing in a press machine will be described as an example. However, the present invention is not limited to this, and can be applied to any plastic processing apparatus.

  For AE sensors, NDIS No. Since the grinding machines using the AE sensor in 2106-79, 2106-91 and the like are widely known as described in Patent Document 1 and the like, the detailed explanation of the AE sensor and the press machine is here. Only the portions directly related to the present invention will be described.

JP 2004-358487 A

  An output signal (AE signal) from an AE sensor provided in the plastic working apparatus changes according to an elapsed time from the start of machining, and a plurality of portions (peak portions) where the signal becomes large are generated. Therefore, as described in Patent Document 1, it seems effective to divide the AE signal into a plurality of time parts corresponding to a plurality of elastic waves and compare the parts for each part. There was a problem that it was difficult to determine the occurrence of an abnormality.

  An object of the present invention is to solve such a problem and to more accurately determine the occurrence of an abnormality.

  In order to achieve the above object, the present invention performs frequency analysis for each unit time of the output signal of the AE sensor when normal plastic working is performed in time series, and calculates the power for the frequency for each unit time in time series. Is calculated and stored as reference data, the frequency analysis of the output signal of the AE sensor when plastic processing is performed is performed in the same manner, and the power difference from the reference data at each frequency is calculated for each unit time. By integrating the power difference in frequency and calculating the integrated power difference per unit time, the integrated power difference is calculated in time series. Then, an abnormality in the plastic working is detected from the time series change of the integrated power difference.

  As described in Patent Document 1, the output signal of the AE sensor provided in the plastic working apparatus changes when there is an abnormality, but the difference is sufficiently clear if the voltage signal output from the AE sensor remains as it is. I can't say that. On the other hand, as in the present invention, the frequency analysis for each unit time of the output signal of the AE sensor is performed in time series, the power difference with the reference data at each frequency is calculated for each unit time, and the frequency at each frequency is calculated. It was found that when the integrated power difference per unit time was calculated by combining the power difference and the time series change of the integrated power difference was obtained, abnormality in plastic working could be detected clearly.

  An AE detection apparatus according to the present invention is an AE detection apparatus that detects an abnormality in a plastic working apparatus, and includes an AE sensor and an AE signal processing unit that processes an AE signal output from the AE sensor. The processing unit is normal in an A / D converter that converts an AE signal into a digital signal, a frequency analysis unit that analyzes a frequency characteristic of the digital AE signal and calculates power with respect to a frequency per unit time, and a plastic processing apparatus. A power difference between a storage unit that stores, as reference data, data obtained in time series for power for a frequency per unit time when processing is performed, and reference data at each frequency of a digital AE signal detected per unit time A difference calculator that calculates the integrated power difference in time series by calculating the integrated power difference per unit time by combining the power differences at each frequency , Characterized in that it comprises a.

  Furthermore, the processing system of the present invention has a plastic processing device and the AE detection device described above, and can detect an abnormality in the plastic processing device.

  There are various methods for determining the occurrence of an abnormality, depending on the object of plastic processing and the plastic processing apparatus, and it is desirable to set appropriately according to the conditions. For example, the total amount of time series change of the integrated power difference is a predetermined value or more. Sometimes it is determined that an abnormality has occurred.

  In addition, it has been confirmed that the time series change pattern of the integrated power difference differs depending on the type of abnormality in a certain plastic processing object or plastic processing apparatus, and the time series change of the integrated power difference according to the type of abnormality is confirmed. The pattern can be stored as an abnormality response pattern, and the time series change of the calculated integrated power difference can be compared with this to determine the type of abnormality from the similar abnormality response pattern.

  According to the present invention, the occurrence of abnormality in plastic processing can be clearly detected during processing, so the quality control of plastic processing is improved and the generation of defective processed products can be reduced.

  FIG. 1 is a diagram showing an overall configuration of a plastic working system according to an embodiment of the present invention. Reference numeral 10 indicates a plastic working apparatus, and here, a press machine is shown as the plastic working apparatus. The press machine 10 includes a press bed 11, a lower mold 12 provided on the press bed 11, an upper moving unit 14, an upper mold 15 attached to the upper moving unit 14, and the upper moving unit 14. It has a stripper 17 provided so as to be movable up and down, a pressing part 18 provided on the stripper 17, a spring 19 for pressurizing the stripper 17 downward, and a processing machine control device 20. The lower mold 12 is provided with a hole 13 corresponding to a female mold, and the upper mold 15 is provided with a projection (pin) 16 corresponding to a male mold. In addition, a guide for aligning the lower mold 12 and the upper mold 15 is provided, but the illustration is omitted.

  When a thin workpiece W to be processed is placed on the lower mold 12 and the upper moving portion 14 is moved from the top dead center toward the bottom dead center, the pressing portion 18 first contacts the surface of the workpiece W. And hold the work W. When the upper moving part 14 moves further downward, the spring 19 contracts in the stripper 17 and moves relative to the upper moving part 14, and the pressing part 18 holds the work W as it is. When the protrusion 16 of the upper mold 15 comes into contact with the surface of the workpiece W and moves further downward, the protrusion 16 shears the workpiece W and enters the hole 13 of the lower mold 12, and the upper movement portion 14 reaches bottom dead center. Thereafter, the upper moving part 14 rises toward the top dead center. In this way, the press work of the workpiece W is performed, and thereafter the same operation is repeated by replacing the workpiece W. The processing machine control device 20 controls the upper moving unit 14 to move from the top dead center to the bottom dead center, and further from the bottom dead center to the top dead center, and stop in accordance with the operator's switch operation.

  In the plastic working system of the present embodiment, an AE sensor 21 provided on the press bed 11 of the press machine 10 and an AE signal processing device 22 for processing an output signal of the AE sensor 21 are further provided. The AE signal processing device 22 receives the operation start signal from the processing machine control device 20 and performs processing. Further, when the AE signal processing device 22 detects a processing abnormality, the AE signal processing device 22 sends the information to the processing machine control device 20, and the processing machine control device 20 stops its operation and displays this information on a display unit such as an operation panel.

  The AE sensor 21 may be any type, but in order to relatively increase the detection sensitivity of the AE wave generated characteristically in the press working, the sensitivity in the low frequency region is not detected so that the mechanical vibration in the low frequency region is not detected. It is desirable to reduce this. In the embodiment, the AE sensor 21 is provided on the press bed 11. However, any position may be used as long as an elastic wave can be detected, and a position where an elastic wave in a desired frequency region can be detected with high sensitivity is appropriately selected. .

  FIG. 2 is a diagram illustrating a configuration of the AE signal processing device 22. As shown in the figure, the AE signal processing device 22 passes a signal in the frequency range targeted for the AE wave, for example, in the range of 20 kHz to 500 kHz, from the analog AE signal output from the AE sensor 21, and signals having other frequencies. A band-pass filter 31 that cuts off the signal, a variable gain amplifier 32 that amplifies the signal that has passed through the band-pass filter 31 to a predetermined intensity, and an A / D that converts an analog signal output from the variable gain amplifier 32 into a digital AE signal. Predetermined processing for the frequency characteristics of the converter 33, the FFT (Fast Fourier Transformer) 34 that performs frequency analysis on the digital AE signal output from the A / D converter 33, and the AE signal output from the FFT 34 A spectrum subtraction method (SSM) 35 for performing (subtraction processing and integration processing here), and SSM 35 And the contact detection signal is output when the output is greater than or equal to a predetermined value, and the determination unit 36 for determining the type of abnormality from the output pattern of the SSM 35, the FFT 34, the SSM 35, and the determination unit 36 An AE processing control unit 37 for controlling the memory and a memory 38. The FFT 34, the SSM 35, the determination unit 36, and the AE processing control unit 37 are configured by a digital signal processor (DSP) 39, and the memory 38 is an operation memory for operating the DSP. The DSP 39 outputs an abnormality detection signal to the processing machine control device 15 and receives an operation start signal from the processing machine control device 15. Note that the FFT 34, the SSM 35, and the determination unit 36 can be realized by a DSP, and the AE processing control unit 37 can be realized by a microcomputer or the like. Since frequency analysis processing using a DSP is widely known, further explanation is omitted.

  Next, processing in the A / D converter 33 and the FFT 34 will be described with reference to FIGS.

  As shown in FIG. 3, the A / D converter 33 samples the analog AE signal output from the variable gain amplifier 32 every 1 μs (at 1 MHz) and converts it into a digital signal (AE data signal). The FFT 34 performs an FFT (Fast Fourier Transform) process on the AE data signal every 256 μs unit time, that is, every 256 sampling data, and calculates the frequency characteristics (power relative to the frequency) of the AE data signal per unit time. To do. Thereby, the frequency characteristic (power with respect to the frequency) of the AE data signal is calculated for each unit time in a time series of 256 μs from the start of the operation of the press. FIG. 4 shows this time-series frequency characteristic.

  In the present invention, the time-series frequency characteristics when normal pressing is performed are stored as reference data, and the time-series frequency characteristics are obtained by the same processing during normal processing, for each unit time, The difference from the reference data for each frequency is obtained. In other words, the power difference from the reference data for each frequency in each unit time is obtained in time series. Then, the square root of the sum of squares of the power difference is obtained every unit time. Thereby, the integrated power difference for each unit time is obtained in time series. Here, the square root of the sum of squares of the power difference is obtained. However, the present invention is not limited to this, and any calculation method that can express the power difference per unit time may be used.

  FIG. 5 is a diagram illustrating an example of a time-series integrated power difference obtained by performing the above-described processing in the plastic working system of the embodiment, where the horizontal axis represents elapsed time and the vertical axis represents integrated power. (A) is normal processing data when the press processing is performed under the same conditions immediately after obtaining the reference data, and (B) is the press machine operated in a state in which the workpiece W is detached from the lower mold 12. (C) is data when foreign matter is mixed in the press surface due to residue rise, etc. (D) is data when a part of the pin (projection) 16 of the upper mold 15 is broken. It is data.

  In the case of (A), the integrated power difference is small over the entire elapsed time (one machining time). In the case of (B), the accumulated power difference is large over the entire elapsed time, but it can be seen that a large accumulated power difference is generated particularly in the portion of the long elapsed time. In the case of (C), the accumulated power difference is large over the entire elapsed time as in the case of (B), but in the portion of the long elapsed time, the larger accumulated power difference is generated as in (B). Absent. In the case of (D), a larger integrated power difference is generated in the intermediate elapsed part than in the case of (B), but in the part of the long elapsed time, the integrated power difference is not as large as in the case of (B).

  As shown in FIG. 5, the time-series integrated power difference pattern varies depending on the type of abnormality, so that the type of abnormality can be specified from the time-series integrated power difference pattern.

  FIG. 6 is a flowchart illustrating the processing of the AE signal processing device 22 according to the embodiment.

  Here, it is assumed that a time-series integrated power difference pattern corresponding to the type of abnormality is obtained in advance and stored in the memory 38.

  In step 101, processing is performed once in a state where normal processing can be performed, and the frequency characteristics of the AE sensor signal in the normal processing are obtained in time series by the FF.

  In step 102, the AE processing control unit 37 stores the time-series frequency characteristics obtained in step 101 as reference data.

  The above steps 101 and 102 are the preparation stage, and normal press working is performed after step 103.

  In step 103, for each processing, the time series frequency characteristics of the AE sensor signal in the processing are obtained by the FF.

  In step 104, the SSM 35 obtains a difference from the reference data for each frequency obtained in step 103 for each unit time. In other words, the power difference from the reference data for each frequency in each unit time is obtained in time series.

  In step 105, the SSM 35 obtains the square root of the sum of squares of the power difference every unit time. Thereby, the integrated power difference for each unit time is obtained in time series.

  In step 106, the determination unit 36 determines the occurrence of abnormality from the time-series integrated power difference obtained in step 105, and stores the time-series accumulated power difference pattern obtained in step 105 in the memory 38. The type of abnormality is determined by comparison with the integrated power difference pattern corresponding to the type. There are various methods for determining the occurrence of an abnormality depending on the plastic processing object and the plastic processing apparatus, and it is desirable to set appropriately according to the conditions, but here, the total amount of time series change of the integrated power difference is greater than or equal to a predetermined value It is determined that an abnormality has occurred.

  In step 107, it is determined whether or not an abnormality has occurred in step 106. If no abnormality has occurred, the process returns to step 103. If an abnormality has occurred, the process proceeds to step 108 to cause the processing machine control device 20 to generate an abnormality. And send information on the type of anomaly. In response to this, the processing machine control device 20 prevents the press machine from operating, and notifies the operator of the occurrence of the abnormality and the type of abnormality by means of a display or sound output provided on the operation unit.

  The present invention can be applied to any configuration as long as the AE sensor is provided in the plastic working apparatus.

FIG. 1 is a diagram showing an overall configuration of a machining system according to an embodiment of the present invention. FIG. 2 is a diagram illustrating a configuration of the AE signal processing device. FIG. 4 is a diagram for explaining the sampling and processing cycle of the AE signal. FIG. 4 is a diagram showing the frequency characteristics for each unit time in time series. FIG. 5 is a diagram illustrating examples of time-series integrated power differences corresponding to various abnormal states. FIG. 6 is a flowchart illustrating processing of the AE signal processing device in the machining system according to the embodiment.

Explanation of symbols

12 Lower mold 15 Upper mold 20 Processing machine control device 21 AE sensor 22 AE signal processing device 31 Band pass filter 33 A / D converter 34 Frequency analysis unit (FFT)
35 Spectrum Subtraction Method (SSM)
36 AE Processing Control Unit W Workpiece

Claims (9)

A plastic working abnormality detecting method for detecting an abnormality in plastic working from an output signal of an AE sensor disposed in a plastic working apparatus,
When normal plastic working is performed, frequency analysis for each unit time of the output signal of the AE sensor is performed in time series, and power with respect to frequency is calculated in time series for each unit time,
Stores the time-series change of power with respect to the frequency at the time of normal plastic working as reference data,
When plastic processing is performed, frequency analysis per unit time of the output signal of the AE sensor is performed in time series, and power for the frequency per unit time is calculated in time series,
Calculate the power difference with the reference data at each frequency per unit time, calculate the integrated power difference per unit time by combining the power difference at each frequency, and calculate the integrated power difference in time series,
An abnormality detection method for plastic working, wherein abnormality in plastic working is detected from a time series change of the integrated power difference.   The plastic working abnormality detection method according to claim 1, wherein when the total amount of time series change of the integrated power difference is equal to or greater than a predetermined value, abnormality is determined. Depending on the type of abnormality, memorize the time series change pattern of accumulated power difference,
The plastic working abnormality detection method according to claim 1, wherein a type of abnormality in plastic working is detected from a time series change of the calculated integrated power difference. A plastic working device;
An AE sensor provided in the plastic working apparatus;
An AE signal processing unit that processes an AE signal output from the AE sensor;
The AE signal processing unit
An A / D converter for converting the AE signal into a digital signal;
A frequency analysis unit that analyzes the frequency characteristics of the digital AE signal and calculates the power for the frequency per unit time;
A storage unit that stores, as reference data, data obtained by time-series power with respect to frequency per unit time when normal processing is performed in the plastic processing apparatus;
By calculating the power difference from the reference data at each frequency of the digital AE signal detected every unit time, and calculating the integrated power difference per unit time by combining the power difference at each frequency, the integrated power in time series A machining system comprising: a difference calculation unit that calculates a difference.   5. The processing system according to claim 4, wherein the AE signal processing unit includes a determination unit that determines whether or not a total amount of time series change of the integrated power difference is equal to or greater than a predetermined value and outputs an abnormality occurrence signal. According to the type of abnormality, the storage unit stores a time series change pattern of accumulated power difference as an abnormality corresponding pattern,
The processing according to claim 5, wherein the determination unit of the AE signal processing unit determines a type of abnormality from a similar abnormality correspondence pattern by comparing a time-series change pattern of an integrated power difference and the abnormality correspondence pattern. system. An AE detection device for detecting an abnormality in a plastic working device,
An AE sensor provided in the plastic working apparatus;
An AE signal processing unit for processing an AE signal output from the AE sensor,
The AE signal processing unit
An A / D converter for converting the AE signal into a digital signal;
A frequency analysis unit that analyzes the frequency characteristics of the digital AE signal and calculates the power for the frequency per unit time;
A storage unit that stores, as reference data, data obtained by time-series power with respect to frequency per unit time when normal processing is performed in the plastic processing apparatus;
By calculating the power difference from the reference data at each frequency of the digital AE signal detected every unit time, and calculating the integrated power difference per unit time by combining the power difference at each frequency, the integrated power in time series An AE detection apparatus comprising: a difference calculation unit that calculates a difference.   The AE detection apparatus according to claim 7, wherein the AE signal processing unit includes a determination unit that determines whether or not a total amount of time series change of the integrated power difference is equal to or greater than a predetermined value and outputs an abnormality occurrence signal. According to the type of abnormality, the storage unit stores a time series change pattern of accumulated power difference as an abnormality corresponding pattern,
The AE according to claim 8, wherein the determination unit of the AE signal processing unit determines a type of abnormality from a similar abnormality correspondence pattern by comparing a time series change pattern of an integrated power difference and the abnormality correspondence pattern. Detection device.

Images