Intelligent forging system and its diagnosis method

Abstract

The intelligent forging system conceived according to the present invention may include: a punch block, which is provided with at least one punch; a die block, which is provided with at least one die to correspond to the punch; a pressure sensor, which is provided on the die or the die assembly provided with the die to be able to sense the load applied by the punch; and a control unit, which compares the measured pressure signal applied by the pressure sensor with a normal range input in advance, and outputs an abnormal signal if it exceeds the range.

Description

Technical Field

[0001] The present invention relates to an intelligent forging system and a diagnostic method thereof, and more specifically, to an intelligent forging system and a precise diagnostic method thereof, which utilizes abnormal sound signals of a mold and temperature-corrected pressure values ​​to perform multi-step cold forging of a workpiece such as a bolt, a nut, or various pins through a series of steps and can accurately determine the abnormality and life of the mold. Background Art

[0002] Generally, forging is one of the metalworking methods that forms a material into a target shape by applying a compressive load. For this purpose, a commonly used forging machine has a die or die block for supporting a workpiece such as a material to be processed, and a punch block equipped with a tool such as a punch, and has a structure that uses hydraulic pressure or an electric motor to move the punch back and forth toward the die block side so that the workpiece is formed in a shape corresponding to the shape of the die. In the multi-stage forging machine of the forging machine, multiple dies are installed in each die block, and the die has slightly different specifications according to the arrangement order. The die corresponding to the final formed object is arranged at the end. The die block is provided with a transmission device for transferring the workpiece, and the workpiece can be transferred to each die in sequence through the transmission device, and the die is formed in stages, and finally formed into the target shape based on the last die. Summary of the invention

[0003]

Technical issues

[0004] However, in the existing multi-stage forging machine, multiple dies are installed on a die block. During the long forming process, fatigue failure may occur due to the spread of damage and cracks in the tool steel and hard materials used as the die materials due to adhesion, which may damage the die.

[0005] Even if the phenomenon of mold damage due to the above-mentioned adhesion and fatigue failure occurs in any one of the multiple molds, it will have a fatal adverse effect on the subsequent processes due to the characteristics of multi-stage forging. In particular, if defects occur in fasteners such as bolts, nuts, or fixing pins, it will also be fatal to the safety of vehicles, buildings, structures, etc. fixed by these defective fasteners.

[0006] The present invention is proposed to solve various problems including the technical problems mentioned above, and its purpose is to provide an intelligent forging system and its precision diagnosis method. By installing pressure sensors and temperature sensors on each mold assembly, and taking the abnormal sound signal of the mold and the pressure value of the corrected temperature value as the reference, the abnormality of each mold can be sensed in real time and more accurately, and the life of the mold can be sensed by the change of the consistency rate between the abnormal sound signal or the corrected pressure value and the reference value. By accumulating deviations, it can be sensed whether the deformation energy borne by the workpiece in the multi-stage forging process is excessive, so as to prevent defective products from the source, and by sensing the signal difference of the pressure sensor between the mold block and the punch block, even the abnormality of the punch block can be sensed. However, the technical problems described above are only exemplary, and the scope of the present invention is not limited to this.

[0007]

Technical solution

[0008] In order to solve the above-mentioned technical problems, the intelligent forging system conceived by the present invention may include: a punch block, which is provided with at least one punch; a die block, which is provided with at least one die to correspond to the punch; a pressure sensor, which is provided on the die or the die assembly provided with the die to be able to sense the load applied by the punch; and a control unit, which compares the measured pressure signal applied by the pressure sensor with a normal range input in advance, and outputs an abnormal signal if it exceeds the range.

[0009] In the intelligent forging system, in order to enable the workpiece to be transferred to the second die after the first forging and formed in the first die, and to be transferred to the third die after the second forging and formed in the second die, the punch block is a cold multi-stage forging punch block with a plurality of punches arranged in a row, the die block is a cold multi-stage forging die block arranged side by side so that the plurality of dies correspond to the punches, and the pressure sensor may include: a first pressure sensor, which is arranged in the first die or the first die assembly; a second pressure sensor, which is arranged in the second die or the second die assembly; and a third pressure sensor, which is arranged in the third die or the third die assembly, and the control unit uses the measured pressure signal of the pressure sensor to determine whether the die has an abnormality or predict the life of the die.

[0010] In the intelligent forging system, the control unit calculates the deviation rate or degree of deviation between the measured pressure signal measured by the first pressure sensor and the expected reference waveform, calculates the deviation rate or degree of deviation between the measured pressure signal measured by the second pressure sensor and the expected reference waveform, calculates the deviation rate or degree of deviation between the measured pressure signal measured by the third pressure sensor and the expected reference waveform, and calculates the accumulated total value of their deviation rates or degrees of deviation, and determines whether the accumulated total value is within a normal range. If the accumulated total value is within the normal range, a normal accumulation judgment signal is output; if the measured total value exceeds the normal range, an abnormal accumulation judgment signal may be output.

[0011] In the intelligent forging system, the mold assembly may include: a mold, which includes a carbide cutter head with a cavity formed inside for inserting a workpiece to correspond to the punch; a support for supporting the bottom of the carbide cutter head; and a sleeve surrounding the outer diameter surface of the carbide cutter head and the support; and a spacer, which, in order to transmit the pressure of the mold, contacts the bottom of the support of the mold and the bottom of the sleeve on the top, and has a first vertical groove unit formed on one side, which is formed in a direction perpendicular to the pressure direction of the mold, and the pressure sensor will be inserted into the first vertical groove unit of the spacer of the mold assembly.

[0012] The intelligent forging system may further include a temperature sensor, which is disposed near the pressure sensor and is used to measure the temperature when the pressure sensor measures the pressure, so as to compensate for the temperature deviation of the pressure signal measured by the pressure sensor.

[0013] In the intelligent forging system, the control unit corrects the measured pressure signal applied from the pressure sensor with a temperature correction pressure signal based on the measured temperature signal measured by the temperature sensor, and compares the temperature correction pressure signal with a pre-input normal range, and outputs an abnormal signal if it exceeds the range.

[0014] In the intelligent forging system, the control unit may include a die life prediction unit, which uses the SN bending curve of the die material and the upper limit value of the Nth stress of the temperature-corrected pressure value as a reference to predict the life of the die, or calculates the consistency rate or consistency between the temperature-compensated pressure signal and the expected reference waveform, and predicts the life of the die by measuring the change in the consistency rate or the change in the consistency.

[0015] The intelligent forging system may further include an AE sensor (acoustic emission sensor), which is at least arranged on the mold or a mold assembly provided with the mold, so as to measure abnormal sound generated in the mold when measuring the load of the pressure sensor.

[0016] In the intelligent forging system, when measuring the load of the pressure sensor, the control unit compares the die sound signal measured by the AE sensor with a pre-input normal range, and outputs an abnormal signal if the range is exceeded.

[0017] The intelligent forging system may further include an external sound sensor, which is arranged at a certain distance from the AE sensor to compensate the mold sound signal measured by the AE sensor through external sound.

[0018] In the intelligent forging system, the spacer is further formed with a second vertical groove portion, and further includes a temperature sensor, which is arranged near the pressure sensor and measures the temperature when the pressure sensor measures pressure to compensate for the temperature deviation of the pressure signal measured by the pressure sensor. The temperature sensor can be a contact or non-contact temperature sensor inserted in the second groove unit of the spacer and converts the heat energy transferred to the spacer into electrical energy.

[0019] In the intelligent forging system, the spacer may further be formed with a third vertical groove portion, and further include an AE sensor, which is at least arranged on the mold or the mold assembly provided with the mold, so that when measuring the load of the pressure sensor, it can measure the abnormal sound generated in the mold. The AE sensor can be a microphone sensor inserted in the third groove unit of the spacer and used to convert the sound energy transmitted to the spacer into electrical energy.

[0020] The intelligent forging system may further include a punch side pressure sensor, which is arranged on the punch or the punch assembly provided with the punch, and the control unit determines whether there is an abnormality in the punch or the punch assembly through the pressure signal measured by the punch side pressure sensor.

[0021] In the intelligent forging system, the control unit accumulates at least the pressure signal data measured by the first pressure sensor, the second pressure sensor and the third pressure sensor when forming the workpiece, and based on the accumulated data and using the machine learning method of artificial intelligence, it can determine whether there is an abnormality in the first mold, the second mold and the third mold or predict the life of the mold.

[0022]

Beneficial Effects

[0023] According to some embodiments of the present invention structured as described above, by installing a pressure sensor on the mold or mold assembly, it is possible to sense mold abnormalities in real time and sense the mold life based on the change in the consistency rate between the measured pressure and the reference value. Furthermore, the measured pressure can be corrected by adding a temperature sensor and an AE sensor to increase the sensing efficiency. Of course, the scope of the present invention is not limited to these effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a stereoscopic appearance diagram of an intelligent forging system according to some embodiments of the present invention.

[0025] Figure 2 yes Figure 1 Cross-sectional view of the intelligent forging system.

[0026] Figure 3 yes Figure 2 An enlarged cross-sectional view of the die block of the intelligent forging system.

[0027] Figure 4 yes Figure 3 A three-dimensional view of the die assembly of the intelligent forging system.

[0028] Figure 5 yes Figure 4 A partial three-dimensional cross-sectional view of the die assembly of the intelligent forging system.

[0029] Figure 6 yes Figure 2 Block diagram of the control unit of the intelligent forging system.

[0030] Figure 7 is a flow chart of a precision diagnosis method for an intelligent forging system according to some embodiments of the present invention.

[0031] Figure 8 It is a curve diagram of pressure values ​​measured according to time in the temperature compensation pressure signal calculation step of the precision diagnosis method of the intelligent forging system according to some embodiments of the present invention.

[0032] Fig. 9 yes Figure 7 A curve diagram of a pressure correction value for offsetting a temperature value from a pressure value measured according to time in a temperature compensation pressure signal calculation step of a precision diagnosis method for an intelligent forging system.

[0033] Fig.10 yes Figure 7In the die life prediction step of the precision diagnosis method of the intelligent forging system, a curve chart of predicting the die life is provided using the SN bending curve of the die material based on the upper limit value of the Nth stress of the temperature-corrected pressure value.

[0034] Fig.11 yes Figure 1 FIG. 1 is a graph showing an example of a measured pressure signal sensed by a die side pressure sensor of an intelligent forging system.

[0035] Figure 12 to Figure 14 It is a diagram showing the detection state of the measured pressure signal when a defect occurs in the "phenomenon" photograph, comparing the case where the pressure sensor is set in the "punch part" (left curve graph) and the case where the pressure sensor is set in the "mold part" of the present invention (right curve graph).

[0036] Fig.15 is a cross-sectional view of an intelligent forging system according to another embodiment of the present invention.

[0037] Fig.16 is a cross-sectional view of an intelligent forging system according to yet another embodiment of the present invention.

[0038] [Description of Reference Numerals]

[0039] 1: Workpiece 10: Punching block

[0040] P: Punch P1: First Punch

[0041] P2: Second punch P3: Third punch

[0042] 20: Mold block M: Mold

[0043] M1: First mold M2: Second mold

[0044] M3: The third mold 30: Mold assembly

[0045] 31: Carbide tip 32: Support

[0046] 33: sleeve 34: spacer

[0047] 35: Ejector pin G1: First vertical slot unit

[0048] G2: Second slot unit G3: Third slot unit

[0049] S1: Pressure sensor S11: First pressure sensor

[0050] S12: Second pressure sensor S13: Third pressure sensor

[0051] S2: Temperature sensor S21: First temperature sensor

[0052] S22: Second temperature sensor S23: Third temperature sensor

[0053] S3: AE sensor S31: First AE sensor

[0054] S32: Second AE sensor S33: Third AE sensor

[0055] S4: External sound sensor S5: Punch side pressure sensor

[0056] 40: control unit 41: expected waveform storage unit

[0057] 42: Mould abnormal sound identification unit 43: Temperature compensation pressure signal calculation unit

[0058] 44: Compensation pressure determination unit 45: Normal determination signal output unit

[0059] 46: Abnormal judgment signal output unit 47: Mold life prediction unit

[0060] 48: Deviation accumulation unit 49: Accumulation compensation pressure determination unit

[0061] 50: Normal accumulation signal output unit 51: Abnormal accumulation signal output unit

[0062] 100: Intelligent forging system DETAILED DESCRIPTION

[0063] Hereinafter, several preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0064] The embodiments of the present invention are provided for more complete description to those skilled in the art. The following embodiments may be modified into various forms, and the scope of the present invention is not limited to these embodiments. On the contrary, these embodiments can more fully and completely disclose the present invention and fully convey the idea of ​​the present invention to those skilled in the art. In addition, for ease of description and accurate description, the thickness or size of each layer in the drawings will be exaggerated.

[0065] The terms used in this specification are only used to illustrate specific embodiments and are not intended to limit the present invention. As used in this specification, unless otherwise clearly stated in the article, a singular expression may include a plural expression. In addition, the "comprise" and / or "comprising" used in this specification are specifically used to indicate the existence of the described shape, number, step, action, component, element and / or its combination, and are not used to exclude the existence or increase of one or more other shapes, numbers, steps, actions, components, elements and / or their combinations.

[0066] Embodiments of the present invention are described below with reference to the accompanying drawings which schematically illustrate preferred embodiments of the present invention. In the accompanying drawings, variations in shape shown may be predicted, for example, based on manufacturing techniques and / or tolerances. Therefore, embodiments of the present invention should not be construed as being limited to the specific shapes within the ranges shown in this specification, and should include variations in shape resulting from, for example, manufacturing processes.

[0067] Figure 1 is a perspective view of the appearance of an intelligent forging system 100 according to some embodiments of the present invention. Figure 2 yes Figure 1 A cross-sectional view of the intelligent forging system 100 is shown.

[0068] First, if Figure 1 and Figure 2 As shown, the intelligent forging system 100 according to some embodiments of the present invention may include: a punch block 10, which is provided with at least one punch P; a die block 20, which is provided with at least one die M to correspond to the punch P; a pressure sensor S1, which is provided on the die M or a die assembly 30 provided with the die M to be able to sense the load applied by the punch M; and a control unit 40, which compares the measured pressure signal applied by the pressure sensor S1 with a pre-input normal range, and outputs an abnormal signal if it exceeds the range.

[0069] In the intelligent forging system 100, the control unit 40 basically measures the load applied to the die M or the die assembly 30 based on the measured value of the pressure sensor S1, and determines whether the die M is abnormal by continuously monitoring the load (measuring the pressure signal), thereby predicting the life of the die M. This will be described in detail later.

[0070] In addition, in a variation of this embodiment, the intelligent forging system 100 may further include an additional sensor in addition to the pressure sensor S1. The measurement values ​​of the pressure sensor S1 and the additional sensor can further improve the ability to determine whether the mold M has an abnormality and the ability to predict its life.

[0071] For example, in some embodiments, the intelligent forging system 100 further includes a temperature sensor S2, which is disposed near the pressure sensor S1 and measures the temperature when the pressure sensor S1 measures the pressure to compensate for the temperature deviation of the pressure signal measured by the pressure sensor S1. The control unit 40 corrects the temperature compensated pressure signal input from the pressure sensor S1 with a temperature correction pressure signal based on the measured temperature signal measured by the temperature sensor S2, and compares the temperature correction pressure signal with a pre-input normal range, and outputs an abnormal signal if it exceeds the range.

[0072] As another example, in some embodiments, the intelligent forging system 100 may further include an AE sensor S3 (acoustic emission sensor), which is arranged in at least one of the mold M, the mold assembly 30 provided with the mold M, the punch block 10, and any combination thereof, so as to be able to measure the abnormal sound generated in the mold M when measuring the load of the pressure sensor S1. Optionally, the intelligent forging system 100 may further include an external sound sensor S4, which is arranged at a certain distance from the AE sensor S3 to compensate the mold sound signal measured by the AE sensor S3 through external sound. When measuring the load of the pressure sensor S1, the control unit 40 may compare the mold sound signal measured by the AE sensor S3 with the normal range input in advance, and output an abnormal signal if it exceeds the range.

[0073] As another example, in some embodiments, the intelligent forging system 100 may include a temperature sensor S2 and an AE sensor S3 in addition to the pressure sensor S1. When measuring the load of the pressure sensor S1, the control unit 40 may compare the mold sound signal measured by the AE sensor S3 with the normal range input in advance, and output an abnormal signal if it exceeds the range. In addition, based on the measured temperature signal measured by the temperature sensor S2, the temperature compensated pressure signal input from the pressure sensor S1 is corrected with a temperature-corrected pressure signal, and the temperature-corrected pressure signal is compared with the normal range input in advance, and an abnormal signal is output if it exceeds the range.

[0074] In the above-mentioned embodiment, the intelligent forging system 100 can be a device for performing multi-stage cold forging of a workpiece such as a bolt, a nut, or various pins through a series of steps. However, the present invention is not limited to multi-stage cold forging, and can be applied to all forging devices having the punch block 10 and the die block 20.

[0075] More specifically, for example, in order to enable the workpiece 1 to be transferred to the second die M2 after the first forging in the first die M1 and to be transferred to the third die M3 after the second forging in the second die M2, the punch block 10 is a cold multi-stage forging type punch block with a plurality of punches P1, P2 and P3 arranged in a row, and the die block 20 can be a cold multi-stage forging type die block arranged side by side so that the plurality of dies M1, M2 and M3 correspond to the punches P1, P2 and P3. Hereinafter, the die assembly 30 provided with the first die M1 is referred to as the first die assembly 30, the die assembly 30 provided with the second die M2 is referred to as the second die assembly 30, and the die assembly 30 provided with the third die M3 is referred to as the third die assembly 30.

[0076] For example, the pressure sensor S1 is disposed in any at least one of the first to third molds M1, M2 and M3 or any at least one of the first to third mold assemblies 30, and the AE sensor S3 can be disposed in any at least one of the first to third molds M1, M2 and M3 or any at least one of the first to third mold assemblies 30 provided with the pressure sensor S1. In addition, the temperature sensor S2 can also be disposed in any at least one of the first to third molds M1, M2 and M3 or any at least one of the first to third mold assemblies 30 provided with the pressure sensor S1.

[0077] However, the technical concept of the present invention is applicable to all forging devices with two or more stages. Although three stages are shown in the drawings, the technical concept of the present invention is also applicable to forging devices with two or more stages.

[0078] In addition, for example, Figure 1 and Figure 2 As shown, the pressure sensor S1 of the intelligent forging system 100 according to some embodiments of the present invention may include a first pressure sensor S11 arranged in the first mold M1 or the first mold assembly 30, a second pressure sensor S12 arranged in the second mold M2 or the second mold assembly 30, and a third pressure sensor S13 arranged in the third mold M3 or the third mold assembly 30.

[0079] In addition, for example, Figure 1 and Figure 2As shown, the temperature sensor S2 of the intelligent forging system 100 according to some embodiments of the present invention may include a first temperature sensor S21 arranged in the first mold M1 or the first mold assembly 30, a second temperature sensor S22 arranged in the second mold M2 or the second mold assembly 30, and a third temperature sensor S23 arranged in the third mold or the third mold assembly 30.

[0080] In addition, for example, Figure 1 and Figure 2 As shown, the AE sensor S3 of the intelligent forging system 100 according to some embodiments of the present invention may include a first AE sensor S31 arranged in the first mold M1 or the first mold assembly 30, a second AE sensor S32 arranged in the second mold M2 or the second mold assembly 30, and a third AE sensor S33 arranged in the third mold M3 or the third mold assembly 30.

[0081] However, the technical concept of the present invention is applicable to all devices provided with at least two sensors. The drawings illustrate a case where three sensors are provided as an example, but the technical concept of the present invention is also applicable to a forging device provided with two or more sensors.

[0082] Figure 3 It is enlarged and illustrated Figure 2 An enlarged cross-sectional view of the die block 20 of the intelligent forging system 100, Figure 4 yes Figure 3 A three-dimensional view of the die assembly 30 of the intelligent forging system 100, Figure 5 yes Figure 4 A partial three-dimensional cross-sectional view of the die assembly 30 of the intelligent forging system 100.

[0083] More specifically, for example, Figures 1 to 5 As shown, the mold assembly 30 may include: a mold M and a spacer 34, wherein the mold M includes a carbide bit 31 having a cavity formed therein for inserting a workpiece 1 to correspond to the punch; a support 32 for supporting the bottom of the carbide bit 31; and a sleeve 33 surrounding the outer diameter surface of the carbide bit 31 and the support 32. In order to transmit the pressure of the mold M, the spacer 34 contacts the bottom of the support 32 of the mold M and the bottom of the sleeve 33, and a first vertical groove unit G1, a second groove unit G2, and a third groove unit G3 in a direction perpendicular to the pressure direction of the mold M are formed on one side.

[0084] The mold M1, the support 32 and the spacer 34 are provided with Figure 2 and Figure 3The through hole through which the ejector pin 35 passes is formed, so that the workpiece 1 can be ejected to the outside of the first mold M1 after the first molding.

[0085] In addition, for example, the pressure sensor S1 is used to be inserted into the first vertical groove unit G1 of the spacer 34, and a piezoelectric element that converts the pressure transmitted to the spacer 34 into an electric current may be used.

[0086] In addition, for example, the temperature sensor S2 is used to be inserted into the second slot unit G2 of the spacer 34, and may be a contact or non-contact temperature sensor that converts thermal energy transferred to the spacer 34 into electrical energy.

[0087] In addition, for example, the AE sensor S3 is used to be inserted into the third slot unit G3 of the spacer 34, and can be a microphone sensor that converts the sound energy transmitted to the spacer 34 into electrical energy.

[0088] Therefore, when the punch P forms the workpiece 1, the forming load is transmitted to the first mold M1, and then transmitted to the spacer 34 through the support 32. At this time, the AE sensor S3 can convert abnormal sound energy such as breaking sound or crushing sound transmitted to the spacer 34 into electrical energy form and apply it to the control unit 40. In order to compensate the mold sound signal measured by the AE sensor S3 through external sound, the external sound sensor S4 can convert external sound energy into electrical energy form and apply it to the control unit 40. At the same time, the first pressure sensor S11 can convert the load acting on the spacer 34 into an electrical signal, that is, convert it into current and apply it to the control unit 40.

[0089] At this time, the first temperature sensor S21 can convert the temperature value indicating the ambient temperature when measuring the pressure value of the first pressure sensor S31 into an electrical signal, that is, into a current and apply it to the control unit 40 at the same time.

[0090] Therefore, when measuring the load of the pressure sensor, the control unit 40 can compare the mold sound signal measured by the AE sensor S3 with the normal range input in advance, and output an abnormal signal if it exceeds the range. Based on the pressure value of the temperature compensated pressure signal applied by the pressure sensor S1, the temperature corrected pressure signal that offsets the temperature value of the measured temperature signal measured by the temperature sensor S2 can be calculated.

[0091] When the pressure sensor S1, the temperature sensor S2 and / or the AE sensor S3 are directly arranged in the mold M, it may be easy to measure the load, but it may be difficult to arrange the sensor. For example, in the case where a groove is formed in the mold M, it may induce the mold to break, etc., and in the case where the pressure sensor S1, the temperature sensor S2 and / or the AE sensor S3 are attached to the mold M, the load on the mold M may cause damage to the sensor. However, as described above, when the pressure sensor S1, the temperature sensor S2 and / or the AE sensor S3 are arranged in the mold assembly 30, for example, in the spacer 34, there is an advantage that the load applied to the mold M can be measured almost unchanged and the mold M does not need to be deformed.

[0092] Figure 6 yes Figure 2 FIG. 4 is a block diagram of a control unit 40 of an intelligent forging system 100 .

[0093] like Figures 1 to 6 As shown, the control unit 40 of the intelligent forging system 100 according to some embodiments of the present invention may include: an expected waveform storage unit 41, which is used to store information of an expected reference waveform for a normal expected correction pressure signal, an expected upper limit waveform allowed to be higher than the expected reference waveform, and an expected lower limit waveform allowed to be lower than the expected reference waveform; a mold abnormal sound discrimination unit 42, which compares the compensated sound signal obtained by attenuating the external sound signal measured by the external sound sensor from the mold sound signal measured by the AE sensor with a pre-input normal range, and outputs an abnormal signal if the range is exceeded; a temperature compensated pressure signal calculation unit 43, which calculates the pressure signal from the pressure sensor. The temperature-corrected pressure signal is a pressure signal applied by S1 to offset the temperature value of the measured temperature signal measured by the temperature sensor S2; a compensation pressure judgment unit 44, which judges whether the temperature-compensated pressure signal belongs to the normal range between the expected upper limit waveform and the expected lower limit waveform; a normal judgment signal output unit 45, which judges that the temperature-compensated pressure signal belongs to the normal range between the expected upper limit waveform and the expected lower limit waveform, and outputs a normal judgment signal; and an abnormal judgment signal output unit 46, which judges that the temperature-compensated pressure signal exceeds the normal range between the expected upper limit waveform and the expected lower limit waveform, and outputs an abnormal judgment signal.

[0094] Among them, the compensation pressure judgment unit 44 judges normal and abnormal or predicts life by using the expected waveform storage unit 41. In addition, it can also judge the normal and abnormal or predict life of load, sound, temperature and other data by using deep learning to repeatedly learn and guide the artificial intelligence unit of specific data. By using the finite element method to calculate the load of each part of the mold, the life based on fatigue damage can be predicted. By using feedback, the setting conditions or design conditions on site can be improved.

[0095] For example, the appropriateness of the initial forging stroke setting using the sensing value can be determined, and the state of the process or equipment can be diagnosed by analyzing the changes in the sensing value during the process. For example, the heat generated during a long-term continuous process can be analyzed, which causes the temperature of the equipment to rise and thus affects the process. The data that changes in the early stage of production and even during the process can also be considered, thereby realizing the design of the forging process and the manufacture of the tools for the forging process.

[0096] Therefore, the control unit 40 of the present invention can perform the following steps: store information of an expected reference waveform for a normal expected correction pressure signal, an expected upper limit waveform allowed to be higher than the expected reference waveform, and an expected lower limit waveform allowed to be lower than the expected reference waveform; compare the compensated sound signal obtained by attenuating the external sound signal measured by the external sound sensor S4 from the mold sound signal measured by the AE sensor S3 with a normal range input in advance, and output an abnormal signal if the range is exceeded; calculate the temperature-corrected pressure signal obtained by offsetting the temperature value of the measured temperature signal measured by the temperature sensor S2 from the pressure value applied by the pressure sensor S1 to the temperature-compensated pressure signal, and judge whether the temperature-compensated pressure signal belongs to the normal range between the expected upper limit waveform and the expected lower limit waveform; and output a normal discrimination signal if it is judged that the temperature-compensated pressure signal belongs to the normal range between the expected upper limit waveform and the expected lower limit waveform, or output an abnormal discrimination signal if it is judged that the temperature-compensated pressure signal exceeds the normal range between the expected upper limit waveform and the expected lower limit waveform.

[0097] Therefore, by installing a plurality of the pressure sensors S11, S12 and S13, the temperature sensors S21, S22 and S23, the AE sensors S31, S32 and S33 and the external sound sensor S4 on each of the mold assemblies 30, the abnormality of each of the molds M1, M2 and M3 can be sensed in real time based on the standard sound value or the temperature-corrected pressure value. If any at least one of the molds M1, M2 and M3 has an abnormality, the user or operator will be promptly informed, so that if impurities stick to the mold or the mold is damaged, subsequent measures can be taken immediately before the subsequent process is affected.

[0098] In addition, for example, Figure 6 As shown, the control unit 40 may further include a mold life prediction unit 47, which uses the SN bending curve of the mold material to predict the life of the mold based on the upper limit value of the Nth stress of the temperature-corrected pressure value, or calculates the consistency rate or consistency between the temperature-compensated pressure signal and the expected reference waveform, and predicts the life of the mold by measuring the change in the consistency rate or the change in the consistency.

[0099] Therefore, for example, based on the SN bending curve of the mold material, the mold life can be predicted very accurately using the temperature-corrected pressure value that is almost similar to the load acting on the actual mold regardless of temperature and the upper limit value of the Nth-order stress as a reference.

[0100] In addition, for example, if the consistency rate between the actual measured value and the expected reference value is above 99% on the first day but decreases by 1% to 98% after the first day and the inconsistency rate at the time of failure decreases by 1% per day to 80%, it can be predicted that the failure rate will be high after the next 20 days, so corresponding measures such as preparing the mold can be taken in advance by predicting the life of the mold after 20 days.

[0101] Therefore, by considering the change in the consistency ratio between the measured pressure and the reference value, the life of each of the molds M1, M2, and M3 can be predicted in advance, and subsequent measures can be taken.

[0102] In addition, if Figures 1 to 6 As shown, the control unit 40 may further include: a deviation accumulation unit 48, which calculates the deviation rate or deviation degree between the temperature compensated pressure signal calculated by the first pressure sensor S11 and the first temperature sensor S21 and the expected reference waveform, calculates the deviation rate or deviation degree between the temperature compensated pressure signal calculated by the second pressure sensor S12 and the second temperature sensor S22 and the expected reference waveform, calculates the deviation rate or deviation degree between the temperature compensated pressure signal calculated by the third pressure sensor S13 and the third temperature sensor S23 and the expected reference waveform, and calculates the accumulated total value of the deviation rates or the deviation degrees; an accumulated compensation pressure determination unit 49, which determines whether the accumulated total value is within a normal range; a normal accumulation signal output unit 50, which outputs a normal accumulation determination signal if it determines that the accumulated total value belongs to the normal range; and an abnormal accumulation signal output unit 51, which outputs an abnormal accumulation determination signal if it determines that the measured total value exceeds the normal range.

[0103] Therefore, the control unit 40 of the present invention can perform the following series of steps: respectively calculate the deviation rate or deviation degree of the temperature compensated pressure signal measured by the first pressure sensor S11, the second pressure sensor S12 and the third pressure sensor S13 and the expected reference waveform, and calculate the accumulated total value of their deviation rate or the deviation degree, and judge whether the accumulated total value is within the normal range, if the accumulated total value belongs to the normal range, output a normal accumulation judgment signal, or if the measured total value exceeds the normal range, output an abnormal accumulation judgment signal.

[0104] Therefore, by accumulating the deviations between the actual values ​​and ideal values ​​generated in each die, it is possible to sense whether the workpiece is subjected to excessive deformation energy during the multi-stage forging process, thereby preventing defective products from the source.

[0105] If the total amount of deformation energy borne by the workpiece 1 during the multi-stage forging process exceeds the normal value, the workpiece 1 cannot withstand the defective products that may cause internal and external cracks or damage or be easily damaged by various impacts, etc. The present invention can prevent defects in the mold and even the workpiece.

[0106] In addition, if Figure 1 and Figure 2 As shown, the intelligent forging system 100 according to some embodiments of the present invention may further include a punch side pressure sensor S5, which is disposed on the punch P or the punch assembly on which the punch P is disposed. Figure 1 and Figure 2 Although only one punch side pressure sensor S5 is shown in the figure, the number of punch side pressure sensors S5 may have various variations in the modified examples of this embodiment. For example, in order to correspond to each punch P, the punch side pressure sensors S5 may also be arranged on the punch P or the punch assembly according to the number of punches P.

[0107] Therefore, the control unit 40 can compare the temperature compensated pressure signal applied by the punch side pressure sensor S5 with the normal range input in advance, and output an abnormal signal if it exceeds the range. The punch side pressure sensor S5 is mainly used to determine whether there is an abnormality in the punch P or the punch assembly. Therefore, while judging whether there is an abnormality in the punch P or the punch assembly through the punch side pressure sensor S5, it is also possible to judge whether there is an abnormality in the mold M or the mold assembly 30 through the mold side pressure sensor S1 set on the mold M or the mold assembly 30.

[0108] In some embodiments, the control unit 40 can even sense an abnormality of the mold block by sensing the signal difference between the pressure sensors S1 and S5 between the mold block 20 and the punch block 10 .

[0109] Therefore, for example, if the signal difference between the pressure sensors S1 and S5 between the mold block 20 and the punch block 10 exceeds the normal value, it is determined that one of the two has an abnormal phenomenon or uneven pressure, so that the user or operator can take subsequent measures.

[0110] Although not shown in the figure, by providing a temperature sensor near the punch side pressure sensor S5, a temperature-compensated pressure value can also be calculated.

[0111] In addition, in the aforementioned embodiment, any at least one of the temperature sensor S2, the AE sensor S3, the external sound sensor S4 and a combination thereof may be omitted in the intelligent forging system 100.

[0112] For example, Fig.15 As shown, if the temperature sensor S2, the AE sensor S3 and the sound sensor S4 are all omitted, the control unit 40 can determine whether the forging system 100 is normal through the information of the pressure sensor S1. For example, even without temperature compensation, the control unit 40 can determine whether the measured pressure signal belongs to the normal range between the expected upper limit waveform and the expected lower limit waveform. If the measured pressure signal belongs to the normal range between the expected upper limit waveform and the expected lower limit waveform, a normal judgment signal is output, and if the measured pressure signal exceeds the normal range between the expected upper limit waveform and the expected lower limit waveform, an abnormal judgment signal is output.

[0113] Further, if the forging system 100 is a multi-stage forging die, the control unit 40 calculates the deviation rate or deviation degree between the measured pressure signal measured by the first pressure sensor S11 and the expected reference waveform, calculates the deviation rate or deviation degree between the measured pressure signal measured by the second pressure sensor S12 and the expected reference waveform, calculates the deviation rate or deviation degree between the measured pressure signal measured by the third pressure sensor S13 and the expected reference waveform, and calculates the accumulated total value of their deviation rates or deviation degrees, and determines whether the accumulated total value is within the normal range. If the accumulated total value belongs to the normal range, a normal accumulation judgment signal is output; if the measured total value exceeds the normal range, an abnormal accumulation judgment signal is output.

[0114] As another example, Fig.16As shown, if the AE sensor S3 and the sound sensor S4 do not exist, the control unit 40 calculates the temperature-corrected pressure signal by offsetting the temperature value of the measured temperature signal measured by the temperature sensor S2 with the pressure value applied from the pressure sensor S1 to the measured pressure signal, and determines whether the temperature-compensated pressure signal belongs to the normal range between the expected upper limit waveform and the expected lower limit waveform. If the temperature-compensated pressure signal belongs to the normal range between the expected upper limit waveform and the expected lower limit waveform, a normal judgment signal is output; if the temperature-compensated pressure signal exceeds the normal range between the expected upper limit waveform and the expected lower limit waveform, an abnormal judgment signal is output.

[0115] Further, if the forging system 100 is a multi-stage forging die, the control unit 40 calculates the deviation rate or deviation degree between the temperature-compensated pressure signal calculated using the first pressure sensor S11 and the first temperature sensor S21 and the expected reference waveform, calculates the deviation rate or deviation degree between the temperature-compensated pressure signal calculated using the second pressure sensor S12 and the second temperature sensor S22 and the expected reference waveform, and calculates the accumulated total value of their deviation rates or deviation degrees, and determines whether the accumulated total value is within the normal range. If the accumulated total value belongs to the normal range, a normal accumulation judgment signal is output; if the measured total value exceeds the normal range, an abnormal accumulation judgment signal is output.

[0116] In the embodiment described above, if the forging system 100 is used, in the multi-stage forging die, the control unit 40 can monitor the status of the first to third dies M1, M2 and M3 in real time and individually using the sensors provided on each stage, and further, continuously accumulate the measurement data, and use the accumulated data to determine whether the first to third dies M1, M2 and M3 are abnormal or to predict the life of the first to third dies M1, M2 and M3. As described above, the abnormality of the first to third dies M1, M2 and M3 or the prediction of the life can be determined by comparing with the reference value, by using the SN bending curve estimation, or by using the machine learning method of artificial intelligence. For example, in Fig.15 In the example, the accumulated data is the data of the measured pressure signals of the first to third pressure sensors S11, S12 and S13. Fig.16 The accumulated data is the accumulated data of the pressure signals measured by the first to third pressure sensors S11, S12 and S13 and the temperature signals measured by the first to third temperature sensors S21, S22 and S23. Figure 2The accumulated data may be accumulated data for measured pressure signals of the first to third pressure sensors S11, S12 and S13, measured temperature signals of the first to third temperature sensors S21, S22 and S23, and measured sound wave signals of the first to third AE sensors S31, S32 and S33.

[0117] Figure 7 is a flow chart of a precision diagnosis method of the intelligent forging system 100 according to some embodiments of the present invention.

[0118] like Figures 1 to 7 As shown, according to the precision diagnosis method of the intelligent forging system 100 of some embodiments of the present invention, in the diagnosis process of the first mold M1, it may include: an expected waveform storage step S110 for storing information of an expected reference waveform of a normal expected correction signal, an expected upper limit waveform allowed to be higher than the expected reference waveform, and an expected lower limit waveform allowed to be lower than the expected reference waveform; a mold abnormal sound judgment step S120 for comparing the compensated sound signal obtained by attenuating the external sound signal measured by the external sound sensor S4 from the mold sound signal measured by the first AE sensor S31 with a pre-input normal range, and outputting an abnormal signal if the range is exceeded; a temperature compensation pressure signal calculation step 130 for calculating the temperature compensation pressure signal obtained by offsetting the temperature value of the measured temperature signal measured by the first temperature sensor S21 from the pressure value applied by the first pressure sensor S11 to the temperature compensation pressure signal; and judging the temperature compensation pressure signal. A pressure determination step S140 is provided to determine whether the temperature-compensated pressure signal belongs to a normal range between the expected upper limit waveform and the expected lower limit waveform; a determination step S150 is provided to determine whether it belongs to a normal range; a normal determination signal output step S160 is provided to output a normal determination signal if the temperature-compensated pressure signal belongs to a normal range between the expected upper limit waveform and the expected lower limit waveform; an abnormal determination signal output step S170 is provided to output an abnormal determination signal if the temperature-compensated pressure signal exceeds the normal range between the expected upper limit waveform and the expected lower limit waveform; and a mold life prediction step S180 is provided to predict the life of the first mold M1 by using the SN bending curve of the first mold M1 and taking the upper limit value of the N-th stress of the temperature-corrected pressure value as a reference, or calculating the consistency rate or degree of consistency between the temperature-compensated pressure signal and the expected reference waveform, and measuring the change in the consistency rate or the change in the degree of consistency, thereby predicting the life of the first mold M1.

[0119] In addition, for example, in the diagnosis process of the second mold M2, it may include: an expected waveform storage step S210 for storing information on an expected reference waveform of a normal expected correction signal, an expected upper limit waveform allowed to be higher than the expected reference waveform, and an expected lower limit waveform allowed to be lower than the expected reference waveform; a mold abnormal sound discrimination step S220 for comparing a compensated sound signal obtained by attenuating the external sound signal measured by the external sound sensor S4 from the mold sound signal measured by the second AE sensor S32 with a pre-input normal range, and outputting an abnormal signal if the range is exceeded; a temperature compensated pressure signal calculation step 230 for calculating the temperature compensated pressure signal obtained by offsetting the temperature value of the measured temperature signal measured by the second temperature sensor S22 with the pressure value applied from the second pressure sensor S12 to the temperature compensated pressure signal; and judging whether the temperature compensated pressure signal belongs to the expected upper limit waveform. The step S240 of determining the pressure within the normal range between the upper limit waveform and the expected lower limit waveform; the step S250 of determining whether it belongs to the normal range; the step S260 of outputting a normal determination signal if the temperature compensated pressure signal belongs to the normal range between the expected upper limit waveform and the expected lower limit waveform; the step S270 of outputting an abnormal determination signal if the temperature compensated pressure signal exceeds the normal range between the expected upper limit waveform and the expected lower limit waveform; and the step S280 of predicting the mold life of the second mold M2 by using the SN bending curve of the second mold M2 and taking the upper limit value of the Nth stress of the temperature-corrected pressure value as the reference to predict the life of the second mold M2, or calculating the consistency rate or degree of consistency between the temperature compensated pressure signal and the expected reference waveform, and measuring the change in the consistency rate or the change in the degree of consistency, and then predicting the life of the second mold M2.

[0120] In addition, for example, in the diagnosis process of the third mold M3, it may include: an expected waveform storage step S310 for storing information on an expected reference waveform of a normal expected correction signal, an expected upper limit waveform allowed to be higher than the expected reference waveform, and an expected lower limit waveform allowed to be lower than the expected reference waveform; a mold abnormal sound discrimination step S320 for comparing a compensated sound signal obtained by attenuating the external sound signal measured by the external sound sensor S4 from the mold sound signal measured by the third AE sensor S33 with a pre-input normal range, and outputting an abnormal signal if the range is exceeded; a temperature compensated pressure signal calculation step 330 for calculating the temperature compensated pressure signal obtained by offsetting the temperature value of the measured temperature signal measured by the third temperature sensor S23 from the pressure value applied by the third pressure sensor S13 to the temperature compensated pressure signal; and judging whether the temperature compensated pressure signal belongs to the expected A pressure determination step S340 of measuring the normal range between the upper limit waveform and the expected lower limit waveform; a determination step S350 of whether it belongs to the normal range; a normal determination signal output step S360 of outputting a normal determination signal if the temperature compensated pressure signal belongs to the normal range between the expected upper limit waveform and the expected lower limit waveform; an abnormal determination signal output step S370 of outputting an abnormal determination signal if the temperature compensated pressure signal exceeds the normal range between the expected upper limit waveform and the expected lower limit waveform; and a mold life prediction step S380 of predicting the life of the third mold M3 by using the SN bending curve of the third mold M3 and taking the upper limit value of the N-th stress of the temperature-corrected pressure value as a reference, or calculating the consistency rate or consistency degree between the temperature compensated pressure signal and the expected reference waveform, and measuring the change in the consistency rate or the change in the consistency degree, thereby predicting the life of the third mold M3.

[0121] Next, the precision diagnosis method of the intelligent forging system 100 according to some embodiments of the present invention may include: calculating the deviation rate or deviation degree between the temperature compensated pressure signal calculated by the first pressure sensor S11 and the first temperature sensor S21 and the expected reference waveform, calculating the deviation rate or deviation degree between the temperature compensated pressure signal calculated by the second pressure sensor S12 and the second temperature sensor S22 and the expected reference waveform, calculating the deviation rate or deviation degree between the temperature compensated pressure signal calculated by the third pressure sensor S13 and the third temperature sensor S23 and the expected reference waveform, and calculating the deviation accumulation step S410 of the accumulated total value of these deviation rates or the deviation degrees; judging the accumulated compensation pressure whether the accumulated total value is within the normal range S420; judging whether the accumulated total value is within the normal range S430, and outputting a normal accumulation judgment signal if the accumulated total value belongs to the normal range S440; and outputting an abnormal accumulation signal output step S450 of an abnormal accumulation judgment signal if the measured total value exceeds the normal range.

[0122] The deviation rate or the degree of deviation may be aggregated based on the number of sensors, and by comparing the theoretical aggregate value of the intermediate step or the final step with the actual aggregate value, it may be determined step by step whether the accumulated amount of deformation energy is within a normal range.

[0123] For example, if the object 1 is within the normal range and the deformation is slightly insufficient in the first and second times, then even if the deformation is slightly beyond the normal range in the third time, it can be judged that no defective products have occurred. On the contrary, if the deformation of the object 1 is within the normal range in the first, second, and third times, but excessive energy is accumulated in each step, resulting in the final total deformation energy exceeding the total normal range, then it can be predicted that the probability of producing defective products is high.

[0124] In addition, as described above, any at least one of the temperature sensor S2, the AE sensor S3, the external sound sensor S4 and their combination may be omitted, and the expected waveform storage step may be a step of storing information of an expected reference waveform for a normal expected pressure signal, an expected upper limit waveform allowed to be higher than the expected reference waveform, and an expected lower limit waveform allowed to be lower than the expected reference waveform.

[0125] Figure 8 1 is a graph of pressure values ​​measured according to time in the temperature-compensated pressure signal calculation step S130 of the precision diagnosis method of the intelligent forging system according to some embodiments of the present invention.

[0126] from Figure 8It can be seen that since the pressure sensor may have temperature dependence, in the temperature compensated pressure signal calculation step S130, the pressure value (blue) measured according to time is an irregular fluctuation shape, which has the overall characteristic of rising rapidly at the beginning and then becoming gentle with the passage of time.

[0127] Fig. 9 yes Figure 7 A graph of a pressure correction value for offsetting a temperature value from a pressure value measured according to time in the temperature compensated pressure signal calculation step S130 of the precision diagnosis method for an intelligent forging system.

[0128] Then, from Fig. 9 As can be seen from the upper curve graph, in the temperature-compensated pressure signal calculation step S130, the temperature value (red) measured according to time is generally similar to the pressure value in form, with a characteristic of rising rapidly at the beginning and then becoming gentle with the passage of time.

[0129] Therefore, if Fig. 9 As shown in the curve graph below, Figure 8 The pressure value (blue) is offset Fig. 9 From the temperature value (red) and the trend chart after compensation, it can be seen that it changes in a fluctuating manner within a certain range, that is, between the upper limit line and the lower limit line.

[0130] Therefore, by offsetting the temperature value from these measured pressure values, the actual load value can be determined very accurately independently of the temperature and can be monitored in real time, based on which the actual service life can be predicted very accurately.

[0131] Fig.10 yes Figure 7 In the die life prediction step of the precision diagnosis method of the intelligent forging system, a curve diagram is provided for predicting the die life by using the SN bending curve of the die material based on the upper limit value of the Nth stress of the temperature-corrected pressure value.

[0132] That is, Fig.10 As shown, even when using the SN bending curve of the existing mold material, the life can be accurately predicted based on the temperature correction pressure value and the upper limit value of the actual N-th stress.

[0133] Therefore, the current usage of the mold can be predicted in real time and very accurately based on the temperature-corrected pressure value, and the mold installation date can be predicted based on this, so that product defects and production interruptions caused by mold damage can be prevented in advance.

[0134] Existing multi-stage forging machines have the problem of carbide cutter heads being damaged or broken due to the adhesion of impurities in the mold. However, the intelligent forging system according to some embodiments of the present invention, for example, goes through a total of 5 steps. Therefore, the maximum expected load acting on each mold is completely different from the actual pressure measurement curve. Numerical interpretation is performed based on the shape of each step. Since the pressure distribution is different, a theoretically normal curve can be extracted through repeated experiments and numerical interpretation. The present invention can timely confirm the abnormal status of each mold or the workpiece by real-time monitoring the actual pressure values ​​or their accumulated total values ​​that exceed the normal range of the normal curve.

[0135] The peak values ​​of the load curves may vary depending on the actual shape of the mold or the specifications of the workpiece, and the numerical interpretation may also vary depending on the numerical interpretation method or the numerical value.

[0136] Fig.11 yes Figure 1 FIG. 1 is a graph showing an example of a measured pressure signal sensed by the die side pressure sensor S1 of the intelligent forging system 100 .

[0137] For example, Fig.11 As shown, the measured pressure signal (WW) sensed by each of the mold-side pressure sensors S1 has a pattern of gradually increasing over time until reaching a peak and then gradually decreasing.

[0138] However, the above-mentioned mode may have various forms depending on the type of mold, the use environment, the specifications of the workpiece, etc. Fig.11 The overall trend is shown to help illustrate.

[0139] like Fig.11 As shown, the control unit 40 of the intelligent forging system 100 according to some embodiments of the present invention can perform the following series of steps: storing information of an expected reference waveform W1 of a normal expected pressure signal (expected correction pressure signal), an expected upper limit waveform W2 that is allowed to be higher than the expected reference waveform (expected correction reference waveform), and an expected lower limit waveform W3 that is allowed to be lower than the expected reference waveform, and judging whether the measured pressure signal WW belongs to a normal range between the expected upper limit waveform W2 and the expected lower limit waveform W3, and outputting a normal judgment signal if the measured pressure signal WW belongs to the normal range A between the expected upper limit waveform W2 and the expected lower limit waveform W3, or outputting an abnormal judgment signal if the measured pressure signal WW exceeds the normal range A between the expected upper limit waveform W2 and the expected lower limit waveform W3.

[0140] Figure 12 to Figure 14It is a diagram comparing the detection states of the measured pressure signals when a defect occurs in the "phenomenon" photograph, in which the pressure sensor is installed in the "punch part (punch or punch assembly)" (left curve graph) and in which the pressure sensor is installed in the "mold part (mold or mold assembly)" of the present invention (right curve graph).

[0141] For example, from Fig.12 As can be seen from the "phenomenon" photo, the actual occurrence of the appearance defect phenomenon (the left dotted line portion) caused by the surface groove of the raw material or the mold is as shown in the left curve diagram. When the pressure sensor is set in the "punch part", since the measured pressure signal (the center line of the M shape as a whole) belongs to the normal range represented by the wider area around it, the defective phenomenon is mistakenly judged as "undetected". On the contrary, with regard to the present invention, since the pressure sensor S1 is set in the "mold part", the measured pressure signal (the center line of the M shape as a whole) exceeds the normal range represented by the wider area around it at the middle peak part (the right dotted line portion), and the defective phenomenon is accurately judged as "detected".

[0142] In addition, for example, from Fig.13 As can be seen from the photo of the "phenomenon", as an example of the defective shape of the product tilted into the die and forged in the actual process, as shown in the left curve diagram, when the pressure sensor is set in the "punch part", since the measured pressure signal (the center line of the M shape as a whole) belongs to the normal range represented by the wider area around it, the defective phenomenon is mistakenly judged as "undetected". On the contrary, with regard to the present invention, since the pressure sensor S1 is set in the "die part", the measured pressure signal (the center line of the M shape as a whole) exceeds the normal range represented by the wider area around it at the middle peak part (the dotted line part on the right), and the defective phenomenon is accurately judged as "detected".

[0143] In addition, for example, Fig.14 Also as shown in the "phenomenon" photo, as another example of the defective shape (the left dotted line portion) in which the product enters the die at an angle and is forged in the actual process, as shown in the left curve graph, when the pressure sensor is set at the "punch part", since the measured pressure signal (the center line of the M shape as a whole) belongs to the normal range represented by the wider area around it, the defective phenomenon is mistakenly judged as "undetected". On the contrary, with regard to the present invention, since the pressure sensor S1 is set in the "die part", the measured pressure signal (the center line of the M shape as a whole) exceeds the normal range represented by the wider area around it at the rear end part (the right dotted line portion), and the defective phenomenon is accurately judged as "detected".

[0144] Therefore, the present invention is different from the general stamping that completes the forging process of one product in one cycle. Multiple dies and punch groups are inserted corresponding to each other, and a multi-stage forging process is performed by performing a continuous process on one product. Therefore, the installation position of the sensor should be different from the stamping die, and each die can be monitored, so that the abnormality of an individual die can be determined in real time. For the existing situation where the pressure sensor is only installed on the stamping block, the signal generated by each stage may be weakened or overlapped, which may lead to a decrease in accuracy.

[0145] According to the present invention, in the multi-stage forging process, the pressure sensor, temperature sensor, and AE sensor are respectively arranged on the spacer, housing, wedge, etc. corresponding to the individual molds, so that the accuracy and sensitivity of the sensor can be improved, and the cost can be saved while the field operation can be smoothly carried out. In addition, although it is shown in the figure, an optical sensor, a magnetic sensor, etc. can be further arranged.

[0146] The present invention has been described above with reference to the embodiments shown in the accompanying drawings, but this is only exemplary, and a person skilled in the art with general knowledge in the art will understand that various modifications can be made based on this and other equivalent embodiments can be implemented. Therefore, the true technical protection scope of the present invention should be determined according to the technical ideas of the attached claims.

Claims

1. An intelligent forging system, comprising: A punching block, which is provided with at least one punch; A die block, which is provided with at least one die to correspond to the punch; a pressure sensor disposed on the die or a die assembly provided with the die so as to be able to sense a load applied by the punch; a control unit that compares the measured pressure signal applied by the pressure sensor with a pre-input normal range and outputs an abnormal signal if the range is exceeded; as well as a temperature sensor disposed near the pressure sensor and used to measure the temperature of the pressure sensor when measuring pressure, so as to compensate for the temperature deviation of the pressure signal measured by the pressure sensor; Wherein, the mold assembly comprises: A die, comprising a cemented carbide cutter head having a cavity formed therein for inserting a workpiece to correspond to the punch; a support for supporting the bottom of the cemented carbide cutter head; and a sleeve surrounding the outer diameter surface of the cemented carbide cutter head and the support; and A spacer, in order to transmit the pressure of the mold, has its upper surface in contact with the lower surface of the support of the mold and the lower surface of the sleeve, and has a first vertical groove unit formed on one side, which is perpendicular to the pressure direction of the mold. The pressure sensor is inserted into the first vertical groove unit of the spacer of the mold assembly; wherein the control unit corrects the measured pressure signal applied from the pressure sensor with a temperature-corrected pressure signal based on the measured temperature signal measured by the temperature sensor, and compares the temperature-corrected pressure signal with a normal range input in advance, and outputs an abnormal signal if the pressure signal exceeds the range; The control unit includes a mold life prediction unit, which uses the S-N bending curve of the mold material and the upper limit value of the Nth stress of the temperature-corrected pressure signal as a reference to predict the life of the mold, or calculates the consistency rate or consistency between the temperature-corrected pressure signal and the expected reference waveform of the normal expected pressure signal, and predicts the life of the mold by measuring the change in the consistency rate or the change in the consistency.

2. The intelligent forging system according to claim 1, wherein: In order to enable the workpiece to be transferred to the second die after the first forging in the first die and to be transferred to the third die after the second forging in the second die, the punch block is a cold multi-stage forging type punch block with a plurality of punches arranged in a row, and the die block is a cold multi-stage forging type die block arranged side by side so that the plurality of dies correspond to the punches. The pressure sensor comprises: A first pressure sensor, which is disposed on the first mold or the first mold assembly; A second pressure sensor is disposed on the second mold or the second mold assembly; and A third pressure sensor is disposed on the third mold or the third mold assembly. The control unit determines whether the mold has an abnormality or predicts the life of the mold by using the measured pressure signal of the pressure sensor.

3. The intelligent forging system according to claim 2, wherein: The control unit calculates the deviation rate or degree of deviation between the measured pressure signal measured by the first pressure sensor and an expected reference waveform of a normal expected pressure signal, calculates the deviation rate or degree of deviation between the measured pressure signal measured by the second pressure sensor and the expected reference waveform, calculates the deviation rate or degree of deviation between the measured pressure signal measured by the third pressure sensor and the expected reference waveform, and calculates the accumulated total value of their deviation rates or degrees of deviation, and determines whether the accumulated total value is within a normal range, outputs a normal accumulation judgment signal if the accumulated total value is within the normal range, and outputs an abnormal accumulation judgment signal if the accumulated total value exceeds the normal range.

4. The intelligent forging system according to claim 1, wherein: The system further includes an AE sensor which is at least disposed on the mold or a mold assembly having the mold disposed thereon, so as to be able to measure an abnormal sound generated in the mold when measuring the load of the pressure sensor.

5. The intelligent forging system according to claim 4, wherein: When measuring the load of the pressure sensor, the control unit compares the mold sound signal measured by the AE sensor with a normal range input in advance, and outputs an abnormal signal if the range is exceeded.

6. The intelligent forging system according to claim 4, wherein: The invention further comprises an external sound sensor which is arranged at a certain distance from the AE sensor so as to compensate the mold sound signal measured by the AE sensor by external sound.

7. The intelligent forging system according to claim 1, wherein: The spacer is further formed with a second vertical groove unit, and further includes a temperature sensor, which is arranged near the pressure sensor and measures the temperature when the pressure sensor measures pressure to compensate for the temperature deviation of the pressure signal measured by the pressure sensor. The temperature sensor is inserted in the second vertical groove unit of the spacer, and is a contact or non-contact temperature sensor that converts heat energy transferred to the spacer into electrical energy.

8. The intelligent forging system according to claim 1, wherein: The spacer further forms a third vertical groove unit, and further includes an AE sensor, which is at least arranged on the mold or the mold assembly provided with the mold, so that when measuring the load of the pressure sensor, it can measure the abnormal sound generated in the mold. The AE sensor is inserted into the third vertical groove unit of the spacer, and is a microphone sensor used to convert the sound energy transmitted to the spacer into electrical energy.

9. The intelligent forging system according to claim 1, wherein: The invention further comprises a punch side pressure sensor, which is arranged on the punch or a punch assembly provided with the punch, The control unit determines whether there is an abnormality in the punch or the punch assembly based on a pressure signal measured by the punch-side pressure sensor.

10. The intelligent forging system according to claim 2, wherein: The control unit accumulates at least the pressure signal data measured by the first pressure sensor, the second pressure sensor and the third pressure sensor when forming the workpiece, and based on the accumulated data and using the machine learning method of artificial intelligence, determines whether there is any abnormality in the first mold, the second mold and the third mold or predicts the life of the mold.

Images