A hardware stamping forming process data analysis method and system

Abstract

The present application relates to stamping process monitoring technical field, especially to a kind of hardware stamping forming process data analysis method and system, wherein, method includes: the maximum pressure value of each bending direction in each stroke is collected, and arranged as pressure sequence;Adjacent two numerical ratio in the pressure sequence is used to construct the ratio change sequence of each adjacent direction pair;Proportion deviation determination is executed based on the ratio change sequence, deviation amplitude and deviation direction are determined, effective deviation depth is determined in combination with hardness difference evolution speed, the number of remaining developable strokes is determined according to the ratio of the effective deviation depth and the hardness difference evolution speed, combined with historical database grading early warning, output stop early warning signal or repair early warning signal.The present application can identify abnormal changes of proportion relationship between directions in advance when each bending direction is not out of tolerance, and early warning opportunity is adaptively adjusted with the degradation process of each direction pair, which improves stamping quality early warning efficiency.

Description

Technical Field

[0001] This invention relates to the field of stamping process monitoring technology, and in particular to a data analysis method and system for metal stamping forming process. Background Technology

[0002] In a multi-bending, one-step forming metal stamping production line, the press operates continuously, forming several bending directions of the product simultaneously or sequentially during each stroke. To monitor the forming quality, force sensors and displacement sensors are deployed on the forming components in each bending direction to collect the maximum pressure value and pressure change curve in each direction in real time. Whether the maximum pressure value in each direction reaches the preset upper limit alarm value is used as the basis for early warning.

[0003] Chinese Patent Publication No. CN122252482A discloses a fault prediction method and system for a liquid-cooled plate stamping production line, which includes collecting stamping process data through a multi-channel sensor network, calculating the wear accumulation index to assess the life decay trend, and outputting maintenance instructions in a graded manner.

[0004] Therefore, the following problems still exist in the existing technology: In existing technologies, for multi-bending one-time forming stamping scenarios, the absolute value of pressure in a single direction is used as the monitoring basis. This does not consider identifying abnormal changes in the proportional relationship between directions when all directions are within tolerance, leading to a problem of delayed early warning. Furthermore, the early warning timing in existing technologies relies on fixed thresholds and does not consider the differences in the degradation process of each direction, resulting in problems such as early intervention when degradation is slow or insufficient response when degradation is rapid. Summary of the Invention

[0005] To address this, the present invention provides a data analysis method and system for the metal stamping process, which overcomes the problem that in the prior art, under multi-bending one-time forming stamping scenarios, the absolute value of pressure in a single direction is used as the monitoring basis, without considering the identification of abnormal changes in the proportional relationship between directions when all directions are within tolerance, resulting in delayed early warning; and the problem that the early warning timing in the prior art depends on a fixed threshold, without considering the differences in the degradation process of each direction, resulting in early intervention when degradation is slow or insufficient response when degradation is rapid.

[0006] To achieve the above objectives, the present invention provides a data analysis method for metal stamping and forming processes, comprising: Collect the maximum pressure values ​​in each bending direction during each stroke and arrange them into a pressure sequence according to the order of bending. Based on the ratio of two adjacent values ​​in the pressure sequence, a ratio change sequence for each pair of adjacent directions is constructed. Based on the ratio change sequence and the maximum pressure value, a proportional deviation determination is performed to determine the adjacent direction pairs and the starting stroke that are in a proportional deviation state. Based on the span combination of the bending direction and the pressure sequence, the span ratio and the check value are determined respectively, and the span consistency is checked based on the deviation between the span ratio and the check value. In response to the cross-position consistency check being passed, the deviation magnitude and deviation direction of the adjacent direction pair are determined based on the average difference of the ratio change sequence before and after the initial stroke. The pressure rise rate of the bending direction of the adjacent direction pair after the initial stroke is obtained. Based on the pressure rise rate, the deviation range and the historical normal fluctuation range, the hardness difference evolution rate and the effective deviation depth are determined respectively to determine the remaining number of developable strokes. The actual number of usable strokes is determined based on the remaining number of developable strokes and the number of safety margin strokes. Based on the comparison between the actual available stroke count and the warning stroke count, it is determined whether the current condition is at the empirical failure boundary, so as to output a differentiated warning signal. The warning stroke count includes the empirical life stroke count and the safety buffer stroke count.

[0007] Furthermore, the process of constructing the ratio change sequence of each adjacent direction pair includes: Obtain the maximum pressure value data pair of any adjacent direction in the pressure sequence to determine the ratio of adjacent directions; The ratio of adjacent directions in the same adjacent direction pair during several consecutive strokes is arranged in the order of stroke sequence to construct the ratio change sequence.

[0008] Furthermore, the process for determining deviations in the execution ratio includes: Determine the direction of deviation of each value in the ratio change sequence relative to the earlier average level in order to perform a ratio deviation determination; Determine whether the maximum pressure value in each bending direction has reached its respective preset upper limit alarm value in order to perform overpressure judgment; If the ratio change sequence satisfies the ratio deviation judgment and each bending direction does not satisfy the overpressure judgment, then the adjacent direction pair is determined to be in the ratio deviation state.

[0009] Furthermore, the process of determining the deviation magnitude and direction of the adjacent direction pair includes: The sequence segment of the most recent strokes before the starting stroke in the ratio change sequence is determined as the reference sequence segment, and the sequence segment from the starting stroke to the current stroke is determined as the deviation sequence segment. The deviation magnitude and the deviation direction are determined based on the difference between the average value of the deviation sequence segment and the average value of the reference sequence segment, and the deviation direction.

[0010] Furthermore, the rate of hardness difference evolution is determined based on the difference in the pressure rise rate between the two bending directions aligned in the adjacent directions; The historical normal fluctuation range is determined based on the fluctuation range of the ratio change sequence before the initial stroke.

[0011] Furthermore, it also includes verifying the hardening rate difference based on the stated proportional deviation state, and determining the deviation magnitude and the deviation direction after the verification is passed, wherein the hardening rate difference verification is as follows: Determine the degree of increase of the maximum pressure value in the two bending directions of the adjacent direction pair before and after the initial stroke; In response to the fact that the degree of ascent of the two bending directions after the initial stroke is greater than the degree of ascent before the initial stroke, and the difference between the degree of ascent of the two bending directions exceeds a preset allowable difference value, the proportional deviation state is confirmed to have passed the hardening rate difference verification.

[0012] Furthermore, the process of determining the safety margin stroke number based on completed deviation cases in the historical database includes: Filter completed deviation cases from the historical database that have the same deviation direction as the stated deviation direction; Determine the number of strokes taken from the start of the proportional deviation state to the occurrence of angular deviation in the corresponding bending direction in each deviation case; The safety margin stroke number is determined based on the lower limit of the stroke number in each of the aforementioned deviation cases.

[0013] Furthermore, the actual number of available strokes is determined based on the smaller of the remaining developable strokes and the safety margin strokes.

[0014] Furthermore, the process of determining the empirical life stroke count includes: Determine the number of strokes that occur from the start of the proportional deviation state to the failure in each evolution record; The empirical life stroke count is determined based on the lower limit of the stroke count.

[0015] To achieve the above objectives, the present invention also provides a data analysis system for the metal stamping process, comprising: The pressure acquisition module is used to collect the maximum pressure value in each bending direction during each stroke and arrange them into a pressure sequence according to the order of bending. A sequence construction module is used to construct a sequence of ratio changes for each pair of adjacent directions based on the ratio of two adjacent values ​​in the pressure sequence. The deviation determination module is used to perform proportional deviation determination based on the ratio change sequence and the maximum pressure value, and to determine the adjacent direction pair and the starting stroke that are in the proportional deviation state. The consistency verification module determines the span ratio and verification value based on the span combination in the bending direction and the pressure sequence, and performs span consistency verification based on the deviation between the span ratio and the verification value. The deviation analysis module is used to determine the deviation magnitude and direction of the adjacent direction pair in response to the cross-position consistency check passing, based on the average difference of the ratio change sequence before and after the starting stroke. The stroke calculation module is used to obtain the pressure rise rate of the bending direction of the adjacent direction pair after the initial stroke. Based on the pressure rise rate, the deviation range and the historical normal fluctuation range, the hardness difference evolution rate and the effective deviation depth are determined respectively to determine the remaining number of developable strokes, and the actual number of usable strokes is determined based on the remaining number of developable strokes and the safety margin number of strokes. The early warning output module compares the actual available stroke count with the early warning stroke count to determine whether the current condition is at the empirical failure boundary, and outputs a differentiated early warning signal. The early warning stroke count includes the empirical life stroke count and the safety buffer stroke count.

[0016] The beneficial effects of each functional module in the above system executing the corresponding steps of the above method are the same as those of the above method, and will not be repeated here.

[0017] Compared with existing technologies, this invention transforms the monitoring target from the absolute pressure value of each bending direction to the ratio of the maximum pressure value of adjacent bending directions. Based on the ratio change sequence, it performs proportional deviation judgment. When the maximum pressure value in each bending direction is lower than the upper limit alarm value, abnormal changes in the proportional relationship between directions can be identified. Furthermore, it determines the deviation magnitude and direction based on the difference in average values ​​before and after the initial stroke, determines the effective deviation depth and the remaining number of developable strokes by combining the hardness difference evolution rate, and performs graded early warning by combining the safety margin stroke count and empirical life stroke count in the historical database. This invention achieves the effect of adaptively adjusting the early warning timing according to the degradation process of each direction, improving the quality early warning efficiency of multi-bending stamping production and reducing scrap losses caused by product angle deviations.

[0018] Furthermore, this invention addresses the blind spots in existing monitoring systems that rely on absolute pressure values. In reality, during multi-bend one-time forming, the maximum pressure values ​​in each bending direction are affected by material hardening, often resulting in a synchronous and slow increase. While no single direction may reach the upper limit alarm value, the proportional relationship of deformation resistance between directions has undergone a continuous shift, which is an early indication of product angle deviation. This invention constructs a sequence of ratio changes for each pair of adjacent directions based on the ratio of two adjacent values. This ratio separates common factors such as material batches and thickness fluctuations from the absolute values, allowing the changes in the relative relationship between directions to be presented independently. Furthermore, the remaining number of developable strokes is determined by the ratio of the effective deviation depth to the rate of hardness difference evolution, converting the abstract degree of deviation into a directly manageable number of remaining strokes. This achieves the effect of continuously providing differentiated early warnings while each pair of directions remains within a controllable degradation range.

[0019] Furthermore, to address the issue that proportional deviation may originate from abrupt changes in a single direction or sensor noise, this invention provides an executable verification path. In actual operating conditions, instantaneous sensor disturbances can cause isolated jumps in the ratio, and issuing warnings based on this would result in false alarms. This invention verifies and confirms the physical cause of the proportional deviation by using the difference in hardening rate. Subsequent analysis is only performed when the degree of rise in both bending directions after the initial stroke is greater than the degree of rise before the initial stroke, and the difference in the degree of rise exceeds a preset allowable difference value. This achieves the effect of eliminating false deviations and providing warnings based on reliable physical causes.

[0020] Furthermore, addressing the difficulty in distinguishing between measurement anomalies and material hardening, this invention provides a verification path that eliminates the need for process calibration. In actual operating conditions, measurement anomalies in force sensors can cause false shifts in ratios, which can be confused with true shifts caused by material hardening. This invention determines the span ratio by the ratio of two values ​​separated by one value in the pressure sequence, and determines the ratios of two adjacent directions by dividing the intermediate value between the two values ​​by the ratio of the two values. A verification value is determined from the two adjacent direction ratios. A span consistency verification is performed based on the deviation between the span ratio and the verification value. This deviation is independent of the degree of material hardening and remains within the consistency tolerance when the measurement is accurate, only exceeding it when there is an anomaly in the measurement process. This allows for the differentiation between measurement anomalies and changes caused by material hardening, achieving self-identification of measurement anomalies.

[0021] Furthermore, this invention recognizes the differences in degradation processes among different directional pairs. In practical applications, applying the same threshold to all directional pairs results in early intervention for slowly degrading pairs, but insufficient response for rapidly degrading pairs. This invention determines the actual number of usable strokes by combining the currently measured and calculated remaining developable strokes with the historically statistical safety margin strokes, and uses the empirical lifetime strokes recorded in historical evolution records as the failure boundary verification. This achieves the effect of mutual verification between physical calculations and statistical laws, and adaptive adjustment of the early warning timing according to the degradation process of each directional pair.

[0022] Furthermore, this invention addresses the issue of maintenance efficiency after an early warning. In actual maintenance, multi-bending production lines have numerous inspection items, and blindly checking them consumes significant downtime. This invention identifies the bending direction indicated by the deviation as the recommended priority for inspection and outputs it along with the maintenance early warning signal, thereby narrowing the scope of inspection and reducing downtime.

[0023] Furthermore, addressing the difficulty in locating the hardening source when several adjacent directional pairs simultaneously exhibit proportional deviation, this invention provides an executable location path. In actual operating conditions, abnormal hardening in a single bending direction can propagate along the bending sequence to two adjacent directional pairs, leading to blind troubleshooting and wasted downtime. This invention compares the pressure rise rate of a common bending direction with that of another bending direction for adjacent directional pairs simultaneously exhibiting proportional deviation. The common bending direction with the faster hardening process is identified as the hardening source direction and output along with the warning signal, achieving the effect of locking down the source of the anomaly and narrowing the investigation scope when several directional pairs deviate simultaneously.

[0024] Furthermore, this invention considers the interference from mold changes and material batch switching comparison values. In reality, batch switching can cause a step change in the maximum pressure value in each bending direction. If the values ​​before and after the step change are mixed in the statistics, the early average level will be contaminated by the step change, and the batch switching itself may be misjudged as a proportional deviation. This invention uses the stroke number corresponding to the batch switching as the segment boundary to perform a segmented reset on the ratio change sequence, so that the early average level only reflects the normal proportional level within the same batch, achieving the effect of eliminating batch switching interference and ensuring the purity of the statistical caliber for proportional deviation judgment.

[0025] Furthermore, this invention establishes a closed loop between early warning and inspection. In practical applications, the statistical caliber of the historical database gradually becomes distorted with mold changes and material supply fluctuations. This distorted statistical caliber causes the early warning lead time to deviate from the actual degradation process. This invention writes the inspection results back to the historical database after the inspection is completed, using the actual inspection results to correct the failure markers and statistical caliber of the evolution record. This achieves a self-reinforcing effect of historical statistical caliber adapting to the actual operation of the production line. Attached Figure Description

[0026] Figure 1 This is a flowchart of the data analysis method for the metal stamping process according to an embodiment of the present invention; Figure 2 A flowchart for constructing the ratio change sequence of each adjacent direction pair in an embodiment of the present invention; Figure 3 A flowchart for determining the proportional deviation state of the adjacent direction pair in an embodiment of the present invention; Figure 4 This is a structural block diagram of the hardware stamping process data analysis system according to an embodiment of the present invention. Detailed Implementation

[0027] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0028] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0029] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0030] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0031] Please see Figure 1 The diagram shows a flowchart of a data analysis method for the metal stamping process according to an embodiment of the present invention. The data analysis method for the metal stamping process according to an embodiment of the present invention includes: Step S1: Collect the maximum pressure value in each bending direction during each stroke, and arrange them into a pressure sequence according to the order of bending. Step S2: Based on the ratio of two adjacent values ​​in the pressure sequence, construct a ratio change sequence for each pair of adjacent directions; Step S3: Based on the ratio change sequence and the maximum pressure value, perform a proportional deviation determination to determine the adjacent direction pairs in the proportional deviation state and their starting strokes. Step S4: Based on the span combination in the bending direction and combined with the pressure sequence, determine the span ratio and the check value respectively, and perform span consistency verification based on the deviation between the span ratio and the check value. Step S5: In response to the cross-position consistency check being passed, the deviation magnitude and deviation direction of the adjacent direction pair are determined based on the average difference of the ratio change sequence before and after the starting stroke. Step S6: Obtain the pressure rise rate of the two bending directions of the adjacent direction pair after the initial stroke; determine the hardness difference evolution rate and effective deviation depth based on the pressure rise rate, the deviation range and the historical normal fluctuation range. Step S7: Determine the remaining number of developable strokes according to the ratio of the effective deviation depth to the hardness difference evolution rate; Step S8: Determine the number of safety margin strokes based on the completed deviation cases in the historical database; Step S9: Determine the actual number of usable strokes based on the remaining number of developable strokes and the number of safety margin strokes, and determine the number of empirical life strokes based on the evolution records in the historical database. Step S10: Based on the comparison between the actual available stroke count and the warning stroke count, determine whether the current condition is at the empirical failure boundary, and output a differentiated warning signal. The warning stroke count includes the empirical life stroke count and the safety buffer stroke count. The output of differentiated early warning signals includes: in response to the actual available number of strokes being lower than the empirical life stroke number, determining that the current experience has failed, and outputting a shutdown early warning signal; otherwise, in response to the actual available number of strokes being lower than the safety buffer stroke number, determining that the current situation is at the empirical failure boundary, and outputting a maintenance early warning signal.

[0032] Specifically, force sensors are deployed on the forming components in each bending direction. Each force sensor is connected to a processing device, which in turn communicates with a historical database. The processing device receives the maximum pressure value measured by each force sensor and executes steps S1 to S10. In response to a query from the processing device, the historical database returns completed deviation cases and evolution records.

[0033] Specifically, the force sensor refers to a sensing device deployed on the forming component in each bending direction, which measures the maximum pressure value borne in the corresponding bending direction according to the stroke, and its measuring point is located on the force path of the corresponding forming component.

[0034] In this embodiment, each force sensor outputs the maximum pressure value corresponding to the bending direction according to the stroke, and the processing device aligns the maximum pressure values ​​of each bending direction according to the stroke number, so that the maximum pressure values ​​under the same stroke number correspond to the same product.

[0035] It is understood that the force sensor can be a piezoelectric force sensor or a strain gauge force sensor, as long as it can measure the maximum pressure value that the corresponding bending direction can withstand according to the stroke.

[0036] Specifically, the processing device refers to the device that receives data measured by each force sensor, executes steps S1 to S10, and outputs a warning signal. It is connected to each force sensor and the historical database respectively.

[0037] In this embodiment, the processing device maintains the ratio change sequence of each adjacent direction pair according to the stroke number. After the data of each stroke is written, the proportional deviation judgment and overpressure judgment are re-executed. When the judgment is passed, the subsequent deviation analysis and stroke calculation are performed.

[0038] Understandably, the processing equipment can be an industrial computer or a programmable logic controller, as long as it can process the data of each channel according to the stroke and output early warning signals.

[0039] Specifically, the aforementioned proportional deviation state refers to a state in which the maximum pressure value in each bending direction has not reached its respective upper limit alarm value, while the ratio relationship between adjacent directions has shifted trendwise. This is an early manifestation of inconsistent hardening processes in different directions during multi-bending forming.

[0040] Specifically, the hardness difference evolution rate refers to the rate at which the hardening degree difference between the two bending directions of the adjacent direction pair increases with the number of strokes, which is determined based on the pressure rise rate of the two bending directions.

[0041] Specifically, the effective deviation depth refers to the portion of the deviation amplitude after removing inherent fluctuations, which represents the true degree of hardening difference.

[0042] Specifically, the remaining number of developable strokes refers to the number of strokes remaining from the current deviation state to the point where the product exhibits a visible angular deviation, assuming the current rate of hardness difference evolution remains unchanged.

[0043] Specifically, the empirical life stroke count refers to the lower limit of the stroke count in the evolution record of each previous proportional deviation state to failure in the adjacent direction, which serves as the basis for determining whether the empirical failure boundary has been exceeded.

[0044] Specifically, in step S1, the maximum pressure value in each bending direction during each stroke is collected and arranged into a pressure sequence according to the order of bending.

[0045] In this embodiment, the pressure sequence refers to the sequence formed by arranging the maximum pressure values ​​in each bending direction in a single stroke according to the order of bending, which represents the relative magnitude and sequential relationship of the deformation resistance in each bending direction.

[0046] Please see Figure 2 As shown, it is a flowchart of the ratio change sequence of each adjacent direction pair in an embodiment of the present invention.

[0047] Specifically, in step S2, the process of constructing a ratio change sequence for each adjacent direction pair based on the ratio of two adjacent values ​​in the pressure sequence includes: Step S21: Obtain the maximum pressure value data pair of any adjacent direction in the pressure sequence to determine the ratio of adjacent directions; Step S22: Arrange the ratios of adjacent directions in the same adjacent direction pair in a series of consecutive strokes according to the stroke order to construct the ratio change sequence.

[0048] Specifically, the adjacent direction pair refers to two bending directions that are adjacent in the bending sequence. The ratio of the adjacent directions reflects the relative magnitude of the deformation resistance of the two bending directions in the adjacent direction pair. Each of the adjacent direction pairs covers all adjacent positions in the pressure sequence.

[0049] Specifically, the ratio change sequence refers to the sequence formed by arranging the ratios of the adjacent directions of the same pair of adjacent directions in a series of strokes according to the stroke order, which shows the evolution trajectory of the proportional relationship of the adjacent direction pairs with the stroke.

[0050] It is understandable that by constructing the ratio change sequence based on the ratio of two adjacent values, common mode factors such as material batches and material thickness fluctuations acting on each bending direction are separated from the absolute values, so that the changes in the relative relationship between directions can be presented independently, providing a data basis for the determination of proportional deviation that is not affected by absolute load fluctuations.

[0051] It is understood that the thickness and hardness fluctuations of the same batch of materials have the same effect on the maximum pressure value in each bending direction. The ratio of adjacent directions cancels out the synchronous fluctuations mentioned above, and the remaining changes in the ratio change sequence reflect the evolution of the relative differences in each bending direction.

[0052] Specifically, in step S3, a proportional deviation determination is performed based on the ratio change sequence and the maximum pressure value to determine the adjacent direction pairs in the proportional deviation state and their starting strokes.

[0053] Please see Figure 3 As shown, it is a flowchart of an embodiment of the present invention for determining the adjacent direction pair in the state of the proportional deviation.

[0054] Specifically, in step S3, the process of determining the proportional deviation based on the ratio change sequence and the maximum pressure value includes: Step S31: Determine the offset direction of each value in the ratio change sequence relative to the early average level, in order to perform a ratio deviation determination; Step S32: Determine whether the maximum pressure value in each bending direction has reached its respective preset upper limit alarm value in order to perform overpressure judgment; Step S33: If the ratio change sequence satisfies the ratio deviation determination and each bending direction does not satisfy the overpressure determination, then the adjacent direction pair is determined to be in the ratio deviation state.

[0055] In this embodiment, the ratio deviation determination is as follows: in response to the fact that the offset direction of a series of consecutive values ​​in the ratio change sequence is the same, it is confirmed that the ratio change sequence satisfies the ratio deviation determination; the early average level is the average value of each value in the ratio change sequence that is located before the series of consecutive values, which reflects the ratio level of the adjacent direction pair under normal working conditions; the offset direction of each value is the same, that is, each value is on the same side of the early average level.

[0056] Specifically, the overpressure determination is as follows: in response to the maximum pressure value in any bending direction reaching the upper limit alarm value of the corresponding bending direction, it is confirmed that the corresponding bending direction meets the overpressure determination.

[0057] In this embodiment, the number of consecutive numerical values ​​is determined based on the fluctuation period of the ratio relationship of the adjacent direction pairs under normal operating conditions; isolated jump values ​​do not constitute the same offset direction because the offset direction is not consistent with the preceding and following values, thereby filtering out misjudgments caused by instantaneous disturbances of the sensor.

[0058] In this embodiment, the proportional deviation determination is executed sequentially according to the stroke. After the maximum pressure value of each stroke is written into the ratio change sequence, the processing device re-executes the proportional deviation determination and the overpressure determination.

[0059] It is understood that the ratio deviation judgment is based on the premise that each bending direction does not meet the overpressure judgment. At this time, any bending direction is within the normal range when viewed alone. Conventional single-direction threshold monitoring will not output any warning. However, the ratio change sequence of adjacent direction pairs has shown a trend of deviation. The ratio deviation judgment is to identify the ratio deviation state within the above-mentioned monitoring blank interval, so that abnormal changes in the ratio relationship between directions are detected when each bending direction is within the tolerance range.

[0060] Specifically, the offset direction refers to the direction in which a single value in the ratio change sequence deviates from the early average level. When the value is greater than the early average level, the offset direction is upward, and when the value is less than the early average level, the offset direction is downward.

[0061] In this embodiment, the early average level is determined based on the ratio of adjacent directions of several strokes preceding the consecutive values. The number of values ​​participating in the averaging is the same as the number of consecutive values, so that the statistical caliber of the average level corresponds to the deviation judgment.

[0062] In this embodiment, the early average level is recalculated with each stroke data update, so that the early average level always reflects the normal proportional level of the adjacent direction to the recent.

[0063] It is understood that the early average level can be refreshed stroke by stroke using a moving average method, or it can be refreshed periodically using a fixed segment method, as long as it can reflect the proportional level of the adjacent direction pair under normal operating conditions.

[0064] Specifically, the upper limit alarm value refers to the pressure alarm upper limit corresponding to each of the bending directions, which serves as the comparison basis for the overpressure determination. The upper limit alarm value for each of the bending directions is set separately.

[0065] In this embodiment, the upper limit alarm value is determined based on the allowable load of the forming component in the corresponding bending direction and the pressure required by the product's process, and is written into the processing equipment after being calibrated during the production line debugging stage.

[0066] It is understood that the upper limit alarm values ​​for each bending direction can be different or partially the same, as long as they can identify the abnormal overpressure state in the corresponding bending direction.

[0067] Specifically, in step S34, the adjacent direction pairs that are in the proportional deviation state are obtained, and the latest stroke corresponding to the ratio change sequence when the proportional deviation judgment is first met is determined as the starting stroke.

[0068] Specifically, in step S4, based on the span combination of the bending direction and combined with the pressure sequence, the span ratio and the check value are determined respectively, so as to perform span consistency verification based on the deviation between the span ratio and the check value.

[0069] In this embodiment, a span combination refers to a combination of two bending directions that are separated by one bending direction in the bending sequence. The span combination corresponds to two pairs of adjacent directions. Taking the first bending direction, the second bending direction, and the third bending direction arranged in the bending sequence as an example, the first bending direction and the third bending direction constitute the span combination. The two pairs of adjacent directions corresponding to the span combination are the pairs of adjacent directions formed by the first bending direction and the second bending direction, and the pairs of adjacent directions formed by the second bending direction and the third bending direction. That is, the span combination is split into two pairs of adjacent directions that are connected end to end along the bending sequence, and the second bending direction is the common bending direction of the two pairs of adjacent directions.

[0070] In this embodiment, the span ratio is determined based on the ratio of the two maximum pressure values ​​corresponding to the span combination of the pressure sequence, and the ratio of the common bending direction to the two adjacent directions determined by the two directions corresponding to the span combination is obtained. A check value is determined based on the two adjacent direction ratios, and a span consistency check is performed according to the deviation between the span ratio and the check value.

[0071] Specifically, the cross-position consistency verification refers to: determining the cross-position ratio based on the ratio of the maximum pressure values ​​in the two bending directions of the cross-position combination, i.e., the ratio of the maximum pressure values ​​in the first bending direction to the third bending direction; determining a verification value based on the ratio of the adjacent directions of the two adjacent direction pairs corresponding to the cross-position combination, wherein the verification value is the value obtained by multiplying the ratios of the adjacent directions of the two adjacent direction pairs; comparing the cross-position ratio with the verification value, and determining whether the measurement process is abnormal based on whether the deviation between the two exceeds a preset consistency tolerance.

[0072] Specifically, material hardening in each bending direction only changes the magnitude of the maximum pressure value in each direction. Provided the measurement is accurate, regardless of how the maximum pressure values ​​change, the deviation between the span ratio and the verification value remains within the consistency tolerance. If the deviation exceeds the consistency tolerance, a measurement anomaly is detected in one of the three force sensors. Therefore, the span consistency verification distinguishes between anomalies in the measurement process and changes caused by material hardening.

[0073] In this embodiment, the consistency tolerance is determined based on the accuracy level of each force sensor and is written into the processing device after calibration during the production line debugging phase; after the maximum pressure value of each stroke is written, the processing device performs the cross-position consistency check while performing the proportional deviation judgment.

[0074] In this embodiment, in response to the deviation not exceeding the consistency tolerance, the cross-position consistency check is confirmed to be passed, and the processing device performs the determination of the deviation magnitude and the deviation direction.

[0075] In this embodiment, in response to the deviation exceeding the consistency tolerance, it is determined that the cross-position consistency check has failed. The processing device outputs the sensor abnormality prompt signal, prompting the inspection of the corresponding three force sensors, and the subsequent analysis of the adjacent direction pair is suspended.

[0076] In one specific embodiment of the present invention, the production line includes five bending directions, and the span combinations with an interval of one bending direction include three groups: the first bending direction and the third bending direction, the second bending direction and the fourth bending direction, and the third bending direction and the fifth bending direction. The processing device performs a span consistency check on each of the three groups of span combinations. In response to the span consistency check failing for any span combination, the sensor abnormality prompt signal points to the three force sensors corresponding to the span combination.

[0077] In another specific embodiment of the present invention, the production line includes six bending directions, and the span combinations with an interval of one bending direction include four groups: the first bending direction and the third bending direction, the second bending direction and the fourth bending direction, the third bending direction and the fifth bending direction, and the fourth bending direction and the sixth bending direction. The processing equipment performs span consistency verification on each of the four groups of span combinations. In response to the span consistency verification of any span combination failing, the sensor abnormality prompt signal points to the three force sensors corresponding to the span combination.

[0078] It is understood that the deviation can be determined by the absolute value of the difference between the span ratio and the verification value, or by the relative deviation between the span ratio and the verification value, as long as it can identify the abnormality in the measurement process.

[0079] Specifically, in step S5, in response to the successful cross-position consistency check, the process of determining the deviation magnitude and direction of the adjacent direction pair based on the average difference of the ratio change sequence before and after the initial stroke includes: Step S51: Determine the sequence segment of the most recent several strokes before the starting stroke in the ratio change sequence as the reference sequence segment, and determine the sequence segment from the starting stroke to the current stroke as the deviation sequence segment; Step S52: Determine the deviation magnitude based on the difference between the average value of the deviation sequence segment and the average value of the reference sequence segment; Step S53: Determine the deviation direction based on the deviation of the average value of the deviation sequence segment relative to the average value of the reference sequence segment.

[0080] Specifically, the deviation magnitude refers to the difference between the average value of the deviation sequence segment and the average value of the reference sequence segment, and the deviation direction refers to the direction in which the average value of the deviation sequence segment deviates from the average value of the reference sequence segment, pointing towards the bending direction in which the deformation resistance increases faster in the adjacent direction.

[0081] Specifically, the stroke number refers to the sequential number of each stroke of the press, and the processing equipment aligns the data of each channel and each judgment result according to the stroke number.

[0082] It is understood that the value of the reference sequence segment can be truncated according to a fixed number of strokes, or it can be truncated according to a statistical window of the early average level, as long as it can reflect the normal proportion level before the proportion deviates from the state.

[0083] In this embodiment, the starting stroke is the stroke number of the latest stroke corresponding to the ratio change sequence when the ratio deviation judgment is first met.

[0084] In this embodiment, step S5 further includes determining the deviation magnitude and the deviation direction in response to the proportional deviation state after hardening rate difference verification, wherein the hardening rate difference verification is as follows: Determine the degree of increase of the maximum pressure value in the two bending directions of the adjacent direction pair before and after the initial stroke; In response to the fact that the degree of ascent of the two bending directions after the initial stroke is greater than the degree of ascent before the initial stroke, and the difference between the degree of ascent of the two bending directions exceeds a preset allowable difference value, the proportional deviation state is confirmed to have passed the hardening rate difference verification.

[0085] In this embodiment, the degree of rise is determined based on the pressure rise rate. For each bending direction, the maximum pressure values ​​of the most recent several strokes before the initial stroke are extracted and subjected to least-squares linear fitting according to the stroke number. The resulting fitting slope is used as the degree of rise of the bending direction before the initial stroke. Similarly, the maximum pressure values ​​of each stroke from the initial stroke to the current stroke are extracted and subjected to least-squares linear fitting according to the stroke number. The resulting fitting slope is used as the degree of rise of the bending direction after the initial stroke. The number of the most recent several strokes before the initial stroke must be consistent with the number of strokes from the initial stroke to the current stroke.

[0086] The preset allowable difference value is adaptively determined based on the historical fluctuation range of the difference in pressure rise rates between the two bending directions within the normal operating condition segment prior to the initial stroke. Specifically, within the normal operating condition segment prior to the initial stroke, the pressure rise rates in the two bending directions are calculated separately using a sliding window of a set length, and the difference between the two is obtained to obtain a rate difference sequence. The range of the rate difference sequence is used as the preset allowable difference value. The set length sliding window is set according to the number of strokes from the initial stroke to the current stroke.

[0087] Understandably, the hardening rate difference verification confirms that the proportional deviation originates from the differentiation of the actual hardening process between the die and the material rather than sensor noise by quantifying the changes in the rate of pressure rise in each direction before and after the initial stroke. Its function is to filter out false offsets, avoid instantaneous disturbances triggering false warnings, ensure that the remaining stroke calculation is based on reliable physical degradation, and improve the accuracy of the warning.

[0088] In this embodiment, if the hardening rate difference verification fails, the processing device outputs a false ratio deviation warning signal, indicating that the current ratio deviation is caused by non-hardening factors, and suspends the execution of subsequent analysis steps.

[0089] Specifically, in step S6, the pressure rise rate of the two bending directions of the adjacent direction pair after the initial stroke is obtained, and the hardness difference evolution rate and effective deviation depth are determined according to the pressure rise rate, the deviation range and the historical normal fluctuation range.

[0090] Specifically, the hardness difference evolution rate is the difference in the pressure rise rate in the two bending directions of the adjacent direction pair; the historical normal fluctuation amplitude is determined based on the fluctuation range of the ratio change sequence before the initial stroke.

[0091] In this embodiment, the process of determining the effective deviation depth includes: subtracting the historical normal fluctuation amplitude of the adjacent direction pair from the deviation amplitude, and the difference obtained is the effective deviation depth, that is, the part remaining after removing the inherent fluctuation from the overall offset of the adjacent direction pair.

[0092] Specifically, in step S7, the remaining number of developable strokes is determined according to the ratio of the effective deviation depth to the hardness difference evolution rate.

[0093] It is understood that the remaining number of developable strokes is determined by the effective deviation depth and the hardness difference evolution rate. The effective deviation depth reflects the current offset, and the hardness difference evolution rate reflects the rate of offset expansion. The ratio of the two converts the abstract degree of offset into the number of remaining strokes that can be directly used for production scheduling. This allows the early warning advance to be refreshed successively with the degradation process of the adjacent direction, avoiding premature intervention when the fixed early warning threshold is slow or insufficient response when the degradation is rapid.

[0094] In this embodiment, the pressure rise rate refers to the rate at which the maximum pressure value in the bending direction increases with the number of strokes, reflecting the speed of the material hardening process in the bending direction.

[0095] In this embodiment, the pressure rise rate is determined based on the maximum pressure value of several strokes after the initial stroke, and is calculated as the ratio of the increase in maximum pressure value with stroke number to the corresponding stroke interval.

[0096] It is understood that the pressure rise rate can be determined by the difference between adjacent strokes or by the fitting slope of several strokes, as long as it allows for comparison of the hardening process speed in the two bending directions.

[0097] Specifically, in step S8, the process of determining the safety margin stroke number based on completed deviation cases in the historical database includes: Step S81: Filter completed deviation cases from the historical database that have the same deviation direction as the deviation direction; Step S82: Determine the number of strokes that occur from the start of the proportional deviation state to the occurrence of angular deviation in the corresponding bending direction in each deviation case; Step S83: Determine the safety margin stroke number based on the lower limit of the stroke number in each of the deviation cases.

[0098] Specifically, the safety margin stroke count refers to the statistical lower limit of the remaining stroke space presented by similar completed deviation cases in the historical database. It estimates the number of strokes remaining before the current state exceeds the angle tolerance from a historical statistical perspective.

[0099] Specifically, the historical database refers to a database that stores and manages completed deviation cases and evolution records for each of the adjacent direction pairs, and each of the completed deviation cases and each of the evolution records is archived and stored according to the adjacent direction pairs.

[0100] In this embodiment, the historical database is deployed on the production line server, and the processing device is communicatively connected to the historical database. After each deviation process is completed, the corresponding case data and evolution records are sent back to the historical database.

[0101] It is understood that the historical database can be deployed on a local server or a cloud server, as long as it can retrieve completed deviation cases and evolution records in the adjacent directions.

[0102] Specifically, in step S9, the actual number of usable strokes is determined based on the remaining number of developable strokes and the number of safety margin strokes, and the empirical life strokes are determined based on the evolution records in the historical database.

[0103] Specifically, the actual number of available strokes is the smaller of the remaining developable strokes and the safety margin strokes; the actual number of available strokes is updated sequentially with each stroke data update.

[0104] Specifically, the actual available stroke count refers to the more conservative remaining stroke count between the current measured estimate and the historical statistics. It is obtained by cross-checking the remaining developable stroke count with the safety margin stroke count to avoid the estimation bias from a single source from directly entering the early warning decision.

[0105] It is understandable that the actual available stroke count is refreshed with each stroke data update. When degradation accelerates, the actual available stroke count converges faster, and when degradation slows down, the actual available stroke count stabilizes accordingly. The early warning lead always follows the actual degradation process of the adjacent direction pair.

[0106] Specifically, in step S9, the process of determining the empirical life stroke number based on the evolution records in the historical database includes: Step S91: Determine the number of strokes in each evolution record from the start of the proportional deviation state to the failure. Step S92: Determine the empirical life stroke count based on the lower limit of the stroke count.

[0107] Specifically, the evolution record refers to the complete process record of a single stroke from the start of the proportional deviation state to failure in the adjacent direction, which includes the ratio change data of each stroke and the stroke number at the time of failure.

[0108] In this embodiment, the evolution record is archived to the historical database after each deviation process is completed, and failure is defined as the occurrence of an angle deviation in the corresponding bending direction.

[0109] It is understood that the evolution record may include the maximum pressure value of each stroke, or it may only include the stroke number and the angle deviation result, as long as it can extract the number of strokes from the start of the proportional deviation state to the failure.

[0110] Specifically, the angle deviation refers to the bending angle of the product exceeding the range allowed by the process. The hardening degree of the two bending directions of the adjacent direction pair is continuously unbalanced, and the difference in springback between the two directions causes the bending angle of the product to gradually deviate from the process requirements, ultimately resulting in the angle deviation.

[0111] In this embodiment, the out-of-tolerance angle is confirmed by sampling the bending angle of the product, and the sampling results are sent back to the historical database as a failure marker for the corresponding evolution record.

[0112] Specifically, in step S10, the actual available stroke count and the warning stroke count are compared to determine whether the current condition is at the empirical failure boundary, so as to output a differentiated warning signal. The warning stroke count includes the empirical life stroke count and the safety buffer stroke count.

[0113] Specifically, the output of differentiated early warning signals includes: in response to the actual available number of strokes being lower than the empirical life stroke count, determining that the current experience has failed, and outputting a shutdown early warning signal; otherwise, in response to the actual available number of strokes being lower than the safety buffer stroke count, determining that the current situation is at the empirical failure boundary, and outputting a maintenance early warning signal.

[0114] In this embodiment, the processing device outputs the shutdown warning signal or the maintenance warning signal, and simultaneously outputs the remaining number of developable strokes and the actual number of available strokes for the adjacent direction pair for production line scheduling reference.

[0115] Specifically, the process of outputting the maintenance warning signal for the adjacent direction pair includes: determining the bending direction pointed to by the deviation direction as the bending direction recommended for priority inspection; and outputting the maintenance warning signal and the bending direction recommended for priority inspection.

[0116] Understandably, by identifying the bending direction pointed to by the deviation direction as the recommended bending direction for priority inspection and outputting it along with the maintenance warning signal, the direction most likely to cause angle deviation can be directly locked from the many inspection items on the multi-bending production line during subsequent maintenance, thus narrowing the scope of investigation and shortening the production time occupied by downtime inspection.

[0117] Specifically, the shutdown warning signal refers to the warning information that instructs the press to stop running and prompts for inspection of the adjacent direction pair, and the maintenance warning signal refers to the warning information that instructs for scheduling inspections according to the actual available strokes. Both are output in stages according to the degree of degradation of the adjacent direction pair.

[0118] In this embodiment, in response to the output of the shutdown warning signal, the press stops running, and during subsequent maintenance, the forming parts of the two bending directions of the adjacent direction pair can be inspected and replaced; in response to the output of the maintenance warning signal, subsequent maintenance can be scheduled before the actual available strokes are exhausted.

[0119] In this embodiment, the inspection window is arranged before the actual available stroke count is exhausted, and the safety buffer stroke count is the reserved operation margin for the execution of the inspection window.

[0120] Specifically, the number of safety buffer strokes refers to the number of buffer strokes reserved for mold changing and maintenance operations, which serves as the basis for comparison in triggering the maintenance warning signal.

[0121] In this embodiment, the number of safety buffer strokes is determined based on the number of strokes required for mold changing and inspection operations, and can be calibrated during the production line commissioning phase.

[0122] It is understood that the number of safety buffer strokes can be set according to a fixed number of strokes or according to the ratio of the number of strokes with the empirical lifespan. As long as it can provide an operating margin for mold changing and maintenance, it will not be elaborated here.

[0123] Please see Figure 4 As shown, it is a structural block diagram of the hardware stamping process data analysis system according to an embodiment of the present invention.

[0124] In this embodiment, the hardware stamping process data analysis system includes a pressure acquisition module, a sequence construction module, a deviation judgment module, a consistency verification module, a deviation analysis module, a stroke calculation module, and an early warning output module.

[0125] Specifically, the hardware stamping process data analysis system of this invention includes: The pressure acquisition module is used to collect the maximum pressure value in each bending direction during each stroke and arrange them into a pressure sequence according to the order of bending. A sequence construction module is used to construct a sequence of ratio changes for each pair of adjacent directions based on the ratio of two adjacent values ​​in the pressure sequence. The deviation determination module is used to perform proportional deviation determination based on the ratio change sequence and the maximum pressure value, and to determine the adjacent direction pair and the starting stroke that are in the proportional deviation state. The consistency verification module determines the span ratio and verification value based on the span combination in the bending direction and the pressure sequence, and performs span consistency verification based on the deviation between the span ratio and the verification value. The deviation analysis module is used to determine the deviation magnitude and direction of the adjacent direction pair in response to the cross-position consistency check passing, based on the average difference of the ratio change sequence before and after the starting stroke. The stroke calculation module is used to obtain the pressure rise rate of the bending direction of the adjacent direction pair after the initial stroke. Based on the pressure rise rate, the deviation range and the historical normal fluctuation range, the hardness difference evolution rate and the effective deviation depth are determined respectively to determine the remaining number of developable strokes, and the actual number of usable strokes is determined based on the remaining number of developable strokes and the safety margin number of strokes. The early warning output module compares the actual available stroke count with the early warning stroke count to determine whether the current condition is at the empirical failure boundary, and outputs a differentiated early warning signal. The early warning stroke count includes the empirical life stroke count and the safety buffer stroke count.

[0126] It should be noted that the above-mentioned functional modules involved in this application are only a logical division of the functions implemented according to the present invention. In practical applications, the above-mentioned functional modules can be integrated into the same processing device or can be distributed in several processing devices that communicate with each other. This will not be elaborated here.

[0127] Specifically, the deviation determination module refers to the functional module that performs the proportional deviation determination and the overpressure determination. After each stroke's data is written, the determination is re-executed, and the consistency verification module is triggered when the determination passes.

[0128] Specifically, the consistency verification module refers to the functional module that performs the cross-position consistency verification. When the verification passes, it triggers the deviation analysis module; when the verification fails, it outputs the sensor abnormality prompt signal.

[0129] Specifically, the stroke estimation module refers to the functional module that queries the historical database, determines the number of safety margin strokes and the number of empirical life strokes, and outputs the actual usable stroke number after cross-validating the current measured estimation with the historical statistics.

[0130] It is understood that the above functional modules can be implemented by hardware circuits or by the processor executing programs in memory, as long as they can perform the corresponding steps of the above methods. This will not be elaborated further here.

[0131] Specifically, the production line includes several bending directions, and each bending direction forms several adjacent direction pairs according to the bending sequence. The processing equipment performs the proportional deviation judgment and subsequent stroke calculation on each adjacent direction pair in parallel. The ratio change sequence, the actual available stroke number, and the warning signal of each adjacent direction pair are maintained independently.

[0132] In this embodiment, in response to two adjacent pairs of directions being in the proportional deviation state within the same time period, the processing device compares the pressure rise rate of the common bending direction of the two pairs of adjacent directions with that of another bending direction. In response to the pressure rise rate of the common bending direction being higher than that of the corresponding other bending direction, the common bending direction is determined as the hardening source direction and output along with the maintenance warning signal.

[0133] In one specific embodiment of the present invention, the production line includes five bending directions, which are arranged in the order of bending to form four pairs of adjacent directions; in response to the adjacent direction pairs formed by the second bending direction and the first bending direction, and the adjacent direction pairs formed by the second bending direction and the third bending direction being in the proportional deviation state, the processing equipment compares the pressure rise rate of the first bending direction, the second bending direction, and the third bending direction, and in response to the pressure rise rate of the second bending direction being higher than that of the first bending direction and the third bending direction, the second bending direction is determined as the hardening source direction.

[0134] It is understood that the direction of the hardening source can be determined by comparing the pressure rise rate or by sorting the effective deviation depth of each of the adjacent direction pairs. It is only necessary to be able to locate the bending direction that causes the deviation when several of the adjacent direction pairs are simultaneously in the proportional deviation state. This will not be elaborated further here.

[0135] Specifically, the processing equipment responds to the mold change signal and material batch switching signal of the production line and performs segmented reset on the ratio change sequence: taking the stroke number corresponding to the batch switching as the segment boundary, the ratios of the adjacent directions after the segment boundary are re-accumulated, the early average level is determined based on the values ​​within the same segment, and the values ​​before the segment boundary are not included in the calculation of the early average level.

[0136] In this embodiment, the mold change signal is output by the mold change detection unit of the press, and the material batch switching signal is output by the production line control system when the material batch number changes; in response to the segment reset, the counting of the consecutive values ​​restarts from the segment boundary, and the step change caused by the batch switching is not mixed with the values ​​before the segment.

[0137] It is understood that the segmented reset can be executed in response to the mold change signal or the material batch switching signal, as long as the statistical range of the early average level does not cross the batch boundary.

[0138] Specifically, the shutdown warning signal and the maintenance warning signal each correspond to an inspection operation; in response to the completion of the corresponding inspection operation, the processing equipment writes the inspection results back to the historical database. The inspection results include whether the angle of the corresponding bending direction exceeds the tolerance and the actual hardening state. Based on the written-back inspection results, the historical database corrects the failure markers of the corresponding evolution records and updates the statistical caliber of the safety margin stroke count and the empirical life stroke count.

[0139] In this embodiment, the inspection results are stored in association with the actual number of available strokes output during the early warning; in the subsequent screening of completed deviation cases, completed deviation cases from the same material batch are given priority, so that the statistical caliber of the safety margin strokes and the empirical life strokes is continuously corrected according to the actual operation of the production line.

[0140] It is understood that the inspection results can be written back when the inspection work is completed, or they can be written back uniformly after the batch production is completed, as long as the statistical scope of the historical database can be corrected according to the actual inspection results.

[0141] Specifically, the embodiments of the present invention convert the absolute pressure values ​​in each bending direction into the ratio relationship between adjacent directions. When no deviation is exceeded in any direction, the trend of the proportional relationship can be identified. The remaining number of development strokes drives the early warning timing, thereby achieving the effect of adaptive adjustment of the early warning lead as each direction degrades.

[0142] It is understood that the embodiments of the present invention are applicable to online quality early warning of multi-bending one-time forming hardware stamping production lines, and can also be extended to scenarios with pressure monitoring requirements in several forming directions, such as multi-station continuous stamping and progressive die stamping. It is only necessary to ensure that the maximum pressure value of each forming direction can be collected according to the stroke, which will not be elaborated here.

[0143] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A data analysis method for metal stamping and forming process, characterized in that, include: Collect the maximum pressure values ​​in each bending direction during each stroke and arrange them into a pressure sequence according to the order of bending. Based on the ratio of two adjacent values ​​in the pressure sequence, a ratio change sequence for each pair of adjacent directions is constructed. Based on the ratio change sequence and the maximum pressure value, a proportional deviation determination is performed to determine the adjacent direction pairs and the starting stroke that are in a proportional deviation state. Based on the span combination in the bending direction and combined with the pressure sequence, the span ratio and the check value are determined respectively, and the span consistency is checked based on the deviation between the span ratio and the check value. In response to the cross-position consistency check being passed, the deviation magnitude and deviation direction of the adjacent direction pair are determined based on the average difference of the ratio change sequence before and after the initial stroke. The pressure rise rate of the bending direction of the adjacent direction pair after the initial stroke is obtained. Based on the pressure rise rate, the deviation range and the historical normal fluctuation range, the hardness difference evolution rate and the effective deviation depth are determined respectively to determine the remaining number of developable strokes. The actual number of usable strokes is determined based on the remaining number of developable strokes and the number of safety margin strokes. Based on the comparison between the actual available stroke count and the warning stroke count, it is determined whether the current condition is at the empirical failure boundary, so as to output a differentiated warning signal. The warning stroke count includes the empirical life stroke count and the safety buffer stroke count.

2. The data analysis method for metal stamping forming process according to claim 1, characterized in that, The process of constructing the ratio change sequence of each adjacent direction pair includes: Obtain the maximum pressure value data pair of any adjacent direction in the pressure sequence to determine the ratio of adjacent directions; The ratio of adjacent directions in the same adjacent direction pair during several consecutive strokes is arranged in the order of stroke sequence to construct the ratio change sequence.

3. The data analysis method for metal stamping forming process according to claim 1, characterized in that, The process for determining deviations in the execution ratio includes: Determine the direction of deviation of each value in the ratio change sequence relative to the earlier average level in order to perform a ratio deviation determination; Determine whether the maximum pressure value in each bending direction has reached its respective preset upper limit alarm value in order to perform overpressure judgment; If the ratio change sequence satisfies the ratio deviation judgment and each bending direction does not satisfy the overpressure judgment, then the adjacent direction pair is determined to be in the ratio deviation state.

4. The data analysis method for metal stamping forming process according to claim 1, characterized in that, The process of determining the deviation magnitude and direction of the adjacent direction pair includes: The sequence segment of the most recent strokes before the starting stroke in the ratio change sequence is determined as the reference sequence segment, and the sequence segment from the starting stroke to the current stroke is determined as the deviation sequence segment. The deviation magnitude and the deviation direction are determined based on the difference between the average value of the deviation sequence segment and the average value of the reference sequence segment, and the deviation direction.

5. The data analysis method for metal stamping forming process according to claim 1, characterized in that, The rate of hardness difference evolution is determined based on the difference in the pressure rise rate between the two bending directions centered in the adjacent directions; The historical normal fluctuation range is determined based on the fluctuation range of the ratio change sequence before the initial stroke.

6. The data analysis method for metal stamping forming process according to claim 4, characterized in that, It also includes verifying the hardening rate difference based on the stated proportional deviation, and determining the deviation magnitude and the deviation direction after the verification is passed, wherein the hardening rate difference verification is as follows: Determine the degree of increase of the maximum pressure value in the two bending directions of the adjacent direction pair before and after the initial stroke; In response to the fact that the degree of ascent of the two bending directions after the initial stroke is greater than the degree of ascent before the initial stroke, and the difference between the degree of ascent of the two bending directions exceeds a preset allowable difference value, the proportional deviation state is confirmed to have passed the hardening rate difference verification.

7. The data analysis method for metal stamping forming process according to claim 1, characterized in that, The process of determining the safety margin stroke number based on completed deviation cases in the historical database includes: Filter completed deviation cases from the historical database that have the same deviation direction as the stated deviation direction; Determine the number of strokes taken from the start of the proportional deviation state to the occurrence of angular deviation in the corresponding bending direction in each deviation case; The safety margin stroke number is determined based on the lower limit of the stroke number in each of the aforementioned deviation cases.

8. The data analysis method for metal stamping forming process according to claim 7, characterized in that, The actual number of available strokes is determined based on the smaller of the remaining developable strokes and the safety margin strokes.

9. The data analysis method for metal stamping forming process according to claim 1, characterized in that, The process of determining the empirical life stroke count includes: Determine the number of strokes that occur from the start of the proportional deviation state to the failure in each evolution record; The empirical life stroke count is determined based on the lower limit of the stroke count.

10. A data analysis system for metal stamping and forming process, which is applied to the data analysis method for metal stamping and forming process according to any one of claims 1-9, characterized in that, include: The pressure acquisition module is used to collect the maximum pressure value in each bending direction during each stroke and arrange them into a pressure sequence according to the order of bending. A sequence construction module is used to construct a sequence of ratio changes for each pair of adjacent directions based on the ratio of two adjacent values ​​in the pressure sequence. The deviation determination module is used to perform proportional deviation determination based on the ratio change sequence and the maximum pressure value, and to determine the adjacent direction pair and the starting stroke that are in the proportional deviation state. The consistency verification module determines the span ratio and verification value based on the span combination in the bending direction and the pressure sequence, and performs span consistency verification based on the deviation between the span ratio and the verification value. The deviation analysis module is used to determine the deviation magnitude and direction of the adjacent direction pair in response to the cross-position consistency check passing, based on the average difference of the ratio change sequence before and after the starting stroke. The stroke calculation module is used to obtain the pressure rise rate of the bending direction of the adjacent direction pair after the initial stroke. Based on the pressure rise rate, the deviation range and the historical normal fluctuation range, the hardness difference evolution rate and the effective deviation depth are determined respectively to determine the remaining number of developable strokes, and the actual number of usable strokes is determined based on the remaining number of developable strokes and the safety margin number of strokes. The early warning output module compares the actual available stroke count with the early warning stroke count to determine whether the current condition is at the empirical failure boundary, and outputs a differentiated early warning signal. The early warning stroke count includes the empirical life stroke count and the safety buffer stroke count.

Images