A fault warning system and method for intelligent mold

Through the intelligent mold system combining pressure sensors and workpiece images, the production parameters are adjusted using machine learning models and the fan operating parameters are determined based on the temperature sensor, which solves the problems of insufficient production parameter adjustment and fan parameter determination in the existing technology, and improves the stamping quality and mold life.

CN118371601BActive Publication Date: 2024-11-19SU ZHOU ZHIXINDA METAL PROD CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410323867.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-05-19
Publication Date
2024-11-19
Estimated Expiration
2043-05-19

AI Technical Summary

Technical Problem

The existing intelligent temperature-controlled high-precision stamping molds fail to effectively adjust production parameters and determine fan operating parameters, resulting in low quality of stamping workpieces and insufficient service life of the mold.

Method used

The intelligent mold system is adopted, combining the pressure sensor and the workpiece image after stamping, and adjusting the production parameters through machine learning models, and determining the fan operating parameters based on the readings of the temperature sensor, triggering the early warning mechanism.

Benefits of technology

It improves the quality of stamping workpieces and the service life of the mold, realizes accurate adjustment of production parameters and reliable determination of fan parameters, timely discovers and solves mold failures, and improves production efficiency and quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118371601B_ABST
    Figure CN118371601B_ABST
Patent Text Reader

Abstract

The embodiment of the present specification provides a fault warning system for an intelligent mold, including an intelligent mold, an image acquisition device, a conveying module, and a control module; the intelligent mold includes: an upper mold base with a punch and a lower mold base with a die; the punch and the die are provided with a temperature sensor and a pressure sensor; the upper mold base and / or the lower mold base are provided with a fan; the control module is configured to: determine the morphological characteristics of the workpiece based on the image of the workpiece after stamping, the morphological characteristics of the workpiece including at least one of the degree of edge burrs, the degree of surface damage, and deformation errors; in response to the morphological characteristics of the workpiece satisfying a first preset condition, based on the first evaluation feature, determine the production parameter adjustment amount through a prediction model; and adjust the production parameters based on the production parameter adjustment amount; perform stamping production based on the adjusted production parameters and obtain a second evaluation feature; in response to the second evaluation feature satisfying the first preset condition, trigger a warning.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] Description of the case

[0002] This application is a divisional application filed for a Chinese application with a filing date of May 19, 2023, application number: 202310569557.6, and invention name: “A control system and method for an intelligent mold”. Technical Field

[0003] The present invention relates to the field of mold manufacturing, and in particular to a fault warning system and method for an intelligent mold. Background Art

[0004] Stamping dies are essential process equipment for stamping production, and they largely determine the quality of workpiece products. Intelligent dies have obvious advantages such as high technical content, long service life, wide application range, and large market space, and have great development space.

[0005] In view of this, CN103920807B discloses an intelligent temperature-controlled high-precision stamping mold, which can detect the surface temperature of the punch and the die in real time, and intelligently adjust the temperature of the punch and the die by controlling the water cooling tank through the main control unit, which can effectively prevent the mold from being oxidized by high temperature, and prevent the mold from deforming due to thermal expansion effect, thereby ensuring processing accuracy, but does not involve adjusting production parameters and determining the operating parameters of the fan.

[0006] Therefore, it is necessary to provide a fault warning system and method for an intelligent mold, which can adjust the production parameters through pressure sensors and the image of the workpiece after stamping, and then determine the fan operating parameters in combination with the temperature, so as to improve the quality of the workpiece after stamping and ensure the service life of the mold. Summary of the invention

[0007] One or more embodiments of the present specification provide a fault warning system for an intelligent mold, the system comprising: an intelligent mold, an image acquisition device, a conveying module, and a control module; the intelligent mold comprises: an upper mold base provided with a punch and a lower mold base provided with a die; the punch cooperates with the die; the punch and the die are provided with at least one temperature sensor and at least one pressure sensor; the upper mold base and / or the lower mold base are provided with at least one fan, and the at least one fan is used to cool the punch and the die; the image acquisition device is used to capture images of workpieces that have undergone stamping processing from at least one direction; the conveying module is used to convey raw materials to a workstation that matches the intelligent mold for the stamping processing, and to convey the workpiece that has undergone the stamping processing to a preset position; the control module is configured The method comprises the following steps: determining the morphological features of the workpiece based on the image of the workpiece after the stamping process, wherein the morphological features of the workpiece include at least one of the degree of edge burrs, the degree of surface damage, and the deformation error; in response to the morphological features of the workpiece satisfying the first preset condition, determining the production parameter adjustment amount through a prediction model based on the first evaluation feature; and adjusting the production parameters based on the production parameter adjustment amount; the first evaluation feature includes the morphological features of the workpiece and / or the reading of the pressure sensor before adjusting the production parameters; the prediction model is a machine learning model; performing stamping production based on the adjusted production parameters, and obtaining the second evaluation feature; the second evaluation feature includes the morphological features of the workpiece and / or the reading of the pressure sensor after adjusting the production parameters; in response to the second evaluation feature satisfying the first preset condition, triggering an early warning.

[0008] One or more embodiments of the present specification provide a fault warning method for an intelligent mold, the method comprising: determining the morphological features of the workpiece based on the image of the workpiece after stamping, the morphological features of the workpiece including at least one of the degree of edge burrs, the degree of surface damage, and deformation errors; in response to the workpiece morphological features satisfying a first preset condition, determining the production parameter adjustment amount through a prediction model based on a first evaluation feature; and adjusting the production parameters based on the production parameter adjustment amount; the first evaluation feature includes the workpiece morphological features and / or the readings of the pressure sensor before adjusting the production parameters; the prediction model is a machine learning model; stamping production is performed based on the adjusted production parameters, and a second evaluation feature is obtained; the second evaluation feature includes the workpiece morphological features and / or the readings of the pressure sensor after adjusting the production parameters; in response to the second evaluation feature satisfying the first preset condition, triggering a warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0010] Figure 1 is an exemplary schematic diagram of a control system of a smart mold according to some embodiments of this specification;

[0011] Figure 2 is an exemplary schematic diagram of a fault model according to some embodiments of this specification;

[0012] Figure 3 It is an exemplary schematic diagram of a preset algorithm shown in some embodiments of this specification. DETAILED DESCRIPTION

[0013] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0014] It should be understood that the "system", "device", "unit" and / or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0015] As shown in this specification and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0016] Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0017] As one of the most prosperous industries, the stamping industry has increased the requirements for molds with the expansion of market demand. CN103920807B can only detect the surface temperature of the punch and the die in real time, and intelligently adjust the temperature of the punch and the die by controlling the water cooling tank through the main control unit, without considering the adjustment of production parameters and the determination of the operating parameters of the cooling fan. Therefore, in some embodiments of this specification, a control method for an intelligent mold is disclosed, which can adjust the production parameters and determine the fan operating parameters in combination with the pressure sensor and the image of the workpiece after stamping, so as to improve the quality of the workpiece after stamping and ensure the service life of the mold.

[0018] Figure 1 is an exemplary schematic diagram of a control system of a smart mold according to some embodiments of the present specification.

[0019] In some embodiments, the control system 100 of the smart mold can be applied to the field of mold manufacturing and used to execute a control method of the smart mold.

[0020] like Figure 1 As shown, the control system 100 of the smart mold may include: a smart mold 110 , an image acquisition device 120 , a conveying module 130 , and a control module 140 .

[0021] The smart mold 110 may refer to a smart stamping mold that can automatically adjust parameters for stamping processing. In some embodiments, the smart mold 110 may include an upper mold base 110-1 and a lower mold base 110-2, the upper mold base 110-1 is provided with a punch 110-3, the lower mold base 110-2 is provided with a die 110-4, and the punch 110-3 cooperates with the die 110-4. At least one temperature sensor 110-5 and at least one pressure sensor 110-6 are provided on the punch 110-3 and the die 110-4, and at least one fan 110-7 is provided on the upper mold base 110-1 and / or the lower mold base 110-2. In some embodiments, the pressure sensors 110-6 on the punch 110-3 and the die 110-4 are laid at even intervals.

[0022] In some embodiments, the temperature sensor 110-5 can be used to monitor the surface temperature of the punch and the die, the pressure sensor 110-6 can be used to monitor the pressure distribution at each point of the punch and the die, and the fan 110-7 can be used to cool the punch and the die.

[0023] The image acquisition device 120 may refer to a related device having an image acquisition function, such as an industrial camera, etc. In some embodiments, the image acquisition device 120 may be used to acquire an image of a workpiece that has undergone a stamping process from at least one direction.

[0024] The conveying module 130 may refer to a module for transporting raw materials and workpieces, and the conveying module may be implemented by devices such as belt conveyors. In some embodiments, the conveying module 130 may be used to transport raw materials to a station that matches the smart mold for stamping processing, and to transfer the workpiece after stamping processing to a preset position. Raw materials refer to materials used for stamping processing, such as steel plates. The station that matches the smart mold may be a processing station corresponding to the corresponding raw materials, such as above the concave mold of the corresponding smart mold. The preset position may be a location for uniformly collecting workpieces after stamping processing, such as a workpiece storage box.

[0025] The control module 140 may refer to a module for adjusting control parameters. It may be implemented based on a control chip, etc. In some embodiments, the control module 140 may be used to adjust production parameters based on the reading of a pressure sensor and an image of a workpiece after stamping. In some embodiments, the control module 140 may also be used to determine operating parameters of at least one fan based on production parameters and readings of a temperature sensor.

[0026] For more information about the smart mold 110 , the image acquisition device 120 , the conveying module 130 , and the control module 140 , please refer to the relevant descriptions in other parts of this document.

[0027] It should be noted that the above description of a control system of an intelligent mold and its internal modules is only for the convenience of description and cannot limit this specification to the scope of the embodiments. It is understandable that for those skilled in the art, after understanding the principle of the system, it is possible to arbitrarily combine the modules or form a subsystem to connect with other modules without deviating from this principle. In some embodiments, Figure 1 The control system of the intelligent mold disclosed in the specification can be different modules in a system, or one module can realize the functions of two or more modules mentioned above. For example, each module can share a storage module, or each module can have its own storage module. Such variations are all within the scope of protection of this specification.

[0028] In some embodiments, the control method of the smart mold may be executed based on a control module of a control system of the smart mold, and the control method of the smart mold may include step 110 and step 120 .

[0029] Step 110, adjusting production parameters based on the reading of the pressure sensor and the image of the workpiece after stamping.

[0030] The reading of the pressure sensor can reflect the pressure borne by the corresponding positions of the male mold and / or the female mold. In some embodiments, each pressure sensor can send the acquired pressure value to the control module through network communication or other means.

[0031] The workpiece image may refer to an image of a product part produced by stamping. In some embodiments, the workpiece image after stamping may be collected from multiple directions by an image acquisition device.

[0032] Production parameters may refer to relevant parameters required in the stamping process, such as stamping speed, stamping pressure, stamping oil concentration, etc. In some embodiments, production parameters may also include other production-related parameters, such as stamping environment temperature or humidity, etc. In some embodiments, the initial production parameters used when the stamping process is started may be pre-set by a person skilled in the art based on production requirements.

[0033] In some embodiments, the control module can adjust the production parameters in a variety of ways. For example, the control module can adjust the production parameters from the initial production parameters to the production parameters confirmed by the table lookup based on a first preset relationship table including the reading of the pressure sensor, the image of the workpiece after stamping, and the corresponding relationship between the production parameters.

[0034] In some embodiments, the control module can determine the morphological characteristics of the workpiece based on the image of the workpiece after stamping; in response to the workpiece morphological characteristics satisfying a first preset condition, adjust the production parameters based on a first evaluation feature; the first evaluation feature includes the workpiece morphological characteristics and / or the reading of the pressure sensor before adjusting the production parameters; obtain a second evaluation feature, and trigger an early warning in response to the second evaluation feature satisfying the first preset condition, the second evaluation feature includes the workpiece morphological characteristics and / or the reading of the pressure sensor after adjusting the production parameters.

[0035] The workpiece morphological feature may refer to a feature used to characterize the shape and structure of the workpiece. In some embodiments, the workpiece morphological feature may refer to a morphological feature that may affect the quality of the workpiece, for example, it may include at least one of the degree of edge burrs, the degree of surface damage, and deformation errors. In some embodiments, the workpiece morphological feature may include a characteristic value, and the larger the characteristic value, the more serious the corresponding workpiece morphological feature exists, and the worse the workpiece quality.

[0036] The edge burr degree may refer to the degree of burrs or other protrusions on the edge of the workpiece in contact with the smart mold. In some embodiments, the control module may count the lengths of multiple protrusions on the edge of the workpiece based on the workpiece images in multiple directions, and use the sum of the lengths as the characteristic value of the edge burr degree.

[0037] The degree of surface damage may refer to the severity of the problem area on the surface of the workpiece during the stamping process. The problem area may include the surface area where there are crushing, scratches, abrasions, etc. In some embodiments, the control module may process the image of the workpiece after stamping by an image processing method such as an image segmentation algorithm to obtain the outline of the problem area in the image of the workpiece after stamping, and use the area value of the problem area framed by the outline as the characteristic value of the degree of surface damage.

[0038] The deformation error may include the error between the outer dimensions of the processed workpiece and the outer dimensions of the standard workpiece. For example, the deformation error may include the bending error of the workpiece. In some embodiments, the control module may obtain the angle of the actual bending portion of the workpiece based on the image of the workpiece after the stamping process, and calculate the angle difference between the actual bending portion and the preset standard bending portion, and use the value of the angle difference as the characteristic value of the deformation error.

[0039] In some embodiments of the present specification, by characterizing the morphological characteristics of the workpiece by the degree of edge burrs, surface damage and deformation errors, problems with the boundary shape and structure of the workpiece can be effectively and comprehensively discovered, which is conducive to the subsequent adjustment of production parameters and improving the production quality of the workpiece.

[0040] In some embodiments, the control module may determine the characteristic values ​​of the edge burr degree, the surface damage degree, and the deformation error based on the image of the workpiece after stamping, and determine the characteristic value sequence as the workpiece morphological feature.

[0041] The evaluation feature refers to a feature used to evaluate whether it is necessary to adjust the production parameters or whether it is necessary to issue an early warning. In some embodiments, the evaluation feature may include a workpiece morphological feature and / or a pressure sensor reading. In some embodiments, the evaluation feature may be represented in various forms such as a vector or a matrix. For example, the control module may use a sequence obtained by adding the pressure sensor reading to the feature sequence corresponding to the workpiece morphological feature as a representation of the evaluation feature.

[0042] The first evaluation feature may be an evaluation feature obtained based on the workpiece morphological feature and / or the reading of the pressure sensor obtained before adjusting the production parameters. Correspondingly, the second evaluation feature may be an evaluation feature obtained based on the workpiece morphological feature and / or the reading of the pressure sensor obtained after adjusting the production parameters.

[0043] The first preset condition may refer to a preset condition that the evaluation feature must meet when triggering an early warning. In some embodiments, the first preset condition may include a pressure condition and / or a morphology condition.

[0044] In some embodiments, the pressure condition may include that the reading of any pressure sensor on the punch or die is greater than the maximum pressure threshold, or the maximum pressure difference is greater than the pressure difference threshold. The maximum pressure difference refers to the maximum value of the maximum pressure difference of the punch and the maximum pressure difference of the die. The maximum pressure difference of the punch is the difference between the maximum pressure and the minimum pressure among the values ​​obtained by all pressure sensors on the punch at the same time point. The calculation of the maximum pressure difference of the die is similar.

[0045] In some embodiments, the morphological condition may include that the characteristic value of the workpiece morphological feature exceeds a preset morphological threshold. In some embodiments, if the characteristic value of any one of the edge burr degree, surface damage degree and deformation error exceeds its corresponding preset characteristic threshold, the morphological condition is considered to be met. In some embodiments, the characteristic value of the workpiece morphological feature may be the weighted sum of each characteristic value of the edge burr degree, surface damage degree and deformation error. Among them, the weight of each characteristic value can be preset in advance. If the aforementioned weighted sum exceeds the preset morphological threshold, the morphological condition is considered to be met. The preset morphological threshold and each preset characteristic threshold can be set in advance by a person skilled in the art based on experience based on production needs or set by system default.

[0046] In some embodiments, when any one of the pressure condition and the shape condition is met, it is considered that the first preset condition is met.

[0047] In some embodiments, when the workpiece morphological feature meets the first preset condition, the control module can adjust the production parameters based on the first evaluation feature in a variety of ways. For example, the control module can adjust the production parameters to the production parameters confirmed by the table lookup based on the second preset relationship table including the correspondence between the first evaluation feature and the production parameters.

[0048] In some embodiments, the control module may determine a production parameter adjustment amount based on the first evaluation feature through a prediction model; and adjust the production parameter based on the production parameter adjustment amount.

[0049] The production parameter adjustment amount may refer to the amount of change that the production parameter needs to be adjusted. For example, the production parameter adjustment amount may include a stamping speed adjustment amount, a stamping pressure adjustment amount, a stamping oil concentration adjustment amount, and the like.

[0050] A prediction model can refer to a trained machine learning model. For example, a prediction model can include a convolutional neural network or other custom model structures.

[0051] In some embodiments, the input of the prediction model may include the first evaluation feature, the production parameters before adjustment (i.e., the production parameters currently used), and the target workpiece morphological features, and the output may be the production parameter adjustment amount. The target workpiece morphological features may be manually preset based on production requirements.

[0052] In some embodiments, the prediction model can be trained based on a large number of first training samples with labels. The first training samples can be a large number of historical first evaluation features, historical pre-adjustment production parameters, and historical target workpiece morphological features corresponding to historical production that meet the historical target workpiece morphological feature requirements, and the labels of the first training samples can be historical production parameter adjustment amounts corresponding to historical pre-adjustment production parameters.

[0053] The first training sample and label can be determined based on historical production data. In some embodiments, in order to increase the amount of training data and obtain a more accurate prediction model, more training data can be generated based on the manual processing of historical data. For example, when the production parameters (stamping speed, stamping pressure, stamping oil concentration) before historical adjustment are (a1, b1, c1), on the basis of the production parameter adjustment amount being (a2, b2, c2), a workpiece that meets the corresponding target workpiece morphological feature requirements can be obtained based on the adjusted production parameters. At this time, multiple groups of training data can be manually generated, wherein each group of training data only needs to meet the requirements of stamping speed before adjustment + stamping speed adjustment amount = a1 + a2, stamping pressure before adjustment + stamping pressure adjustment amount = b1 + b2, and stamping oil concentration before adjustment + stamping oil concentration adjustment amount = c1 + c2. It should be noted that in order to ensure the accuracy of the model, the difference between the training data generated based on the manual processing of historical data and the corresponding actual historical data needs to be within a certain range, and the range can be preset by the technician.

[0054] In some embodiments, the control module may add the current production parameter and the production parameter adjustment amount to determine the adjusted production parameter.

[0055] In some embodiments of the present specification, the production parameter adjustment amount is determined by a prediction model, which can improve the accuracy and speed of the determined production parameter adjustment amount, thereby facilitating the accuracy of subsequent adjustment of the production parameters and ensuring the accuracy of the stamping process.

[0056] In some embodiments, the control module may control the smart mold to perform stamping production based on the adjusted production parameters, and obtain the second evaluation feature after the adjusted production parameters.

[0057] In some embodiments, in response to the second evaluation feature satisfying the first preset condition, the control module can directly trigger the warning. In some embodiments, the warning method can include multiple forms such as text warning, sound warning, etc. For example, the control module can send a control instruction for voice warning to the voice playback device to perform the voice warning.

[0058] In some embodiments, the control module may determine the fault type and its corresponding fault probability based on the second evaluation feature; and trigger an early warning in response to the fault probability satisfying a second preset condition.

[0059] The fault type may refer to the type of problem that may cause the second evaluation feature to meet the first preset condition, for example, inappropriate smart mold gap, inappropriate depth of the punch entering the die, punch wear, the punch or guide pin is not installed vertically, etc.

[0060] The failure probability of a certain failure type may refer to the probability that the second evaluation feature satisfies the first preset condition due to the failure type.

[0061] In some embodiments, the control module may determine the fault type and its corresponding fault probability based on the second evaluation feature in a variety of ways, such as by querying a preset relationship comparison table.

[0062] In some embodiments, the control module can also predict the fault type and its corresponding fault probability based on the fault model. For details, see Figure 2 Related instructions.

[0063] The second preset condition may refer to a condition that the failure probability corresponding to the failure type must meet when triggering an early warning. In some embodiments, the second preset condition may be that the failure probability is greater than an early warning probability threshold. The early warning probability threshold may be manually preset based on experience or set by system default.

[0064] In some embodiments of the present specification, fault types and fault probabilities are referenced to add ways to trigger early warnings, and workpiece production problems caused by the smart mold itself are taken into consideration, which is conducive to timely and accurate discovery and confirmation of problem points, thereby improving workpiece production efficiency and quality.

[0065] In some embodiments of the present specification, production parameters are adjusted more directly and efficiently based on workpiece images with the help of workpiece morphological features, and then it is determined whether an early warning is triggered based on the evaluation features after the production parameters are adjusted. This is conducive to timely discovering problems that still exist in the stamping process after the production parameters are adjusted, thereby improving production quality.

[0066] Step 120, determining operating parameters of at least one fan based on the production parameters and the reading of the temperature sensor.

[0067] In some embodiments, the operating parameters of the fan may include the number of running fans, the position of each fan, the operating power of the fan at each position, etc. The position of the fan may be represented by a spatial coordinate (x, y, z). For example, the control module may establish a spatial rectangular coordinate system based on any fixed position of the randomly selected smart mold to obtain the spatial coordinates of the location of the fan. The fixed position may refer to a part of the position where the smart mold does not need to be displaced in actual production.

[0068] In some embodiments, the control module may determine the operating parameters of at least one fan based on the production parameters and the readings of the temperature sensor in a variety of ways, such as based on vector database retrieval.

[0069] In some embodiments, the control module can determine a temperature threshold based on historical data; in response to the reading of at least one temperature sensor satisfying a third preset condition, the operating parameters of at least one fan are determined by a preset algorithm and the fan is operated based at least on production parameters, and the third preset condition is related to the temperature threshold.

[0070] The temperature threshold may refer to a maximum temperature value that may cause high-temperature oxidation and deformation of a mold (such as a punch and a die).

[0071] In some embodiments, the control module may determine the temperature threshold in a variety of ways based on historical data. For example, the temperature threshold may be manually preset based on historical data according to experience.

[0072] In some embodiments, the temperature threshold may also be determined by vector retrieval.

[0073] In some embodiments, the control module can construct a historical vector database based on historical data, wherein the historical vector database includes multiple reference vectors and historical temperature thresholds corresponding to each reference vector. Each reference vector is constructed based on the historical workpiece type, historical workpiece material, and historical mold material of historical workpieces produced in history.

[0074] In some embodiments, the control module can construct a feature vector based on the type of workpiece, workpiece material, and mold material currently being produced, and retrieve at least one reference vector whose vector distance from the feature vector is less than a distance threshold from the historical vector database as a target vector, and use the historical temperature threshold corresponding to the target vector as the current temperature threshold. The vector distance can be a Euclidean distance, etc. The distance threshold can be pre-set by a person skilled in the art based on experience or set by system default.

[0075] The third preset condition may refer to a condition for triggering the operation of the fan. For example, the third preset condition may be that the reading of the temperature sensor is greater than or equal to a temperature threshold.

[0076] In some embodiments, in response to the reading of at least one temperature sensor satisfying a third preset condition, the control module may determine an operating parameter of at least one fan through a preset algorithm and operate the corresponding fan based on the determined operating parameter.

[0077] For more information about the preset algorithm, see Figure 3 and its related description.

[0078] In some embodiments of the present specification, a temperature threshold is determined based on historical data to determine whether to run the fan, and then the fan operating parameters are determined to avoid high-temperature oxidation of the smart mold due to excessive temperature, and deformation caused by thermal expansion effects, which is beneficial to ensuring the service life of the smart mold.

[0079] In some embodiments of the present specification, adjusting production parameters through the readings of the pressure sensor and the image of the workpiece after stamping is conducive to convenient and accurate adjustment of the production parameters. At the same time, the fan operating parameters are determined in combination with the production parameters and the readings of the temperature sensor, thereby achieving the goal of improving the production quality of the workpiece while ensuring the service life of the smart mold.

[0080] Figure 2 It is a schematic diagram of a fault model according to some embodiments of the present specification.

[0081] In some embodiments, the control module determines the fault type and its corresponding fault probability 260 based on the second evaluation feature 210, which may include: the control module determines the fault type and its corresponding fault probability 260 through the fault model 250 based on the second evaluation feature 210, the adjusted production parameters 220, the operating parameters 230 of the fan, and the historical fault data 240.

[0082] In some embodiments, the historical fault data 240 may include the number of faults and fault types of the current smart mold obtained based on the historical data. For the description of the second evaluation feature 210, the adjusted production parameter 220, and the fan operation parameter 230, refer to the corresponding description above.

[0083] A fault model may refer to a trained machine learning model. For example, a fault model may include a recurrent neural network.

[0084] In some embodiments, the input of the fault model 250 may include the second evaluation feature 210 , the adjusted production parameter 220 , the operation parameter 230 of the wind turbine, and the historical fault data 240 , and the output may be the fault type and its corresponding fault probability 260 .

[0085] In some embodiments, the fault model can be trained based on a large number of second training samples with labels. A group of second training samples can be the historical second evaluation features, historical adjusted production parameters, historical fan operating parameters, and historical fault data obtained by the sample mold in a historical production. The labels of the group of second training samples can be the historical fault types and their corresponding fault probabilities that occurred in the sample mold in the historical production. In some embodiments, if the historical fault type in the label is the actual fault type that occurred in the sample mold in the historical production, the value of the fault probability of the historical fault type is 1, otherwise it is 0.

[0086] In some embodiments of the present specification, the fault model is used to make the determined fault type and its corresponding fault probability more accurate, so that timely adjustments can be made to the fault problem to ensure workpiece production quality and production efficiency.

[0087] Figure 3 It is a schematic diagram of a preset algorithm shown in some embodiments of this specification.

[0088] In some embodiments, the preset algorithm may include: randomly generating multiple candidate fan operation plans; based on each candidate fan operation plan, determining the temperature sensor reading after the candidate fan operation plan has been running for a period of time through an evaluation model; based on the output of the evaluation model, determining a candidate fan operation plan that meets preset requirements as the target fan operation plan.

[0089] The candidate fan operation scheme 310-3 may refer to a candidate scheme for determining the fan operation parameters. In some embodiments, the candidate fan operation scheme may include the operation parameters of at least one fan in the smart mold. For example, the candidate fan operation scheme may include the number of fans in operation on the smart mold, the position of each fan in operation, the operation power of each fan in operation, etc.

[0090] In some embodiments, the control module may randomly generate multiple candidate fan operation schemes. For example, the control module may randomly generate multiple candidate numbers of fan operation, candidate positions of the fan operation, and candidate operating powers based on the current fan operation parameters and in accordance with a preset adjustment method. The preset adjustment method may be pre-set by a person skilled in the art based on experience or set by system default. For example, the preset adjustment method may include adjusting the number of fan operations, adjusting the operating position, and adjusting the fan operating power according to a preset step size.

[0091] The evaluation model may refer to a trained machine learning model. For example, the evaluation model may include a recurrent neural network, etc.

[0092] In some embodiments, the input of the evaluation model 320 may include: readings 310-1 of multiple temperature sensors, temperature thresholds 310-2, candidate fan operation schemes 310-3, workpiece morphological features 310-4, and current production parameters 310-5, and the output may be a temperature sequence 330 consisting of the temperatures at multiple time points in the future under the candidate fan operation scheme 310-3. For the description of the readings 310-1 and temperature thresholds 310-2 of multiple temperature sensors, see the previous text. In some embodiments, the current production parameters 310-5 refer to the production parameters used by the smart mold when the temperature sensor acquires temperature data, which may be the initial production parameters or the adjusted production parameters. Correspondingly, the workpiece morphological features 310-4 may be the workpiece morphological features 310-4 of the workpiece produced corresponding to the current production parameters 310-5.

[0093] The temperature sequence 330 may include the temperature of multiple points of the smart mold at various time points in the future. For example, there are three temperature sensors A, B, and C, which are used to obtain the temperature data of three points of the smart mold. If the temperature at five time points (t1, t2, t3, t4, t5) within the next 10 minutes needs to be predicted, the output temperature sequence is {(a1, a2, a3, a4, a5), (b1, b2, b3, b4, b5), (c1, c2, c3, c4, c5)}.

[0094] Among them, (a1, a2, a3, a4, a5), (b1, b2, b3, b4, b5), and (c1, c2, c3, c4, c5) are the predicted readings of temperature sensors A, B, and C at five time points (t1, t2, t3, t4, and t5), respectively, and t1, t2, t3, t4, and t5 are the time points arranged in chronological order.

[0095] In some embodiments, the future period of time and the number of time points may be preset manually based on experience.

[0096] In some embodiments, the evaluation model can be trained based on a large number of third training samples with labels. A set of third training samples can include a historical fan operation scheme used by a sample mold in a historical production in a first historical time period, as well as historical readings of multiple temperature sensors in the historical production, historical temperature thresholds used in the historical production, historical current production parameters used in the historical production, and historical workpiece morphological features of the corresponding historical workpieces produced. The label of the third training sample can be a temperature sequence consisting of the temperature of the sample mold at multiple time points in the second historical time period in the historical production. The first historical time period is earlier than the second historical time period. In some embodiments, the third training samples and labels can be determined based on historical production data.

[0097] The target wind turbine operation plan may refer to a final plan for determining the operating parameters of the wind turbine.

[0098] In some embodiments, the control module may determine a candidate wind turbine operation scheme that meets preset requirements based on the output of the evaluation model as the target wind turbine operation scheme.

[0099] In some embodiments, the preset requirement 340 may include: the temperature sensor final reading 340 - 1 is lower than a preset threshold 340 - 2 , and the temperature change rate 340 - 2 is within a preset threshold range 340 - 4 .

[0100] In some embodiments, the final temperature sensor reading 340-1 may be obtained based on the temperature sequence 330 consisting of the temperatures at multiple time points in the future. For example, the final temperature sensor reading may refer to the temperature corresponding to the last time point in the temperature sequence 330. For example, the above five time points (t1, t2, t3, t4, t5), t1, t2, t3, t4, t5 are arranged in chronological order, and the final temperature sensor reading is (a5, b5, c5).

[0101] In some embodiments, in order to prevent the temperature of the mold from exceeding the temperature threshold soon after the fan stops, thereby causing damage to the mold, the preset requirement includes the final reading of the temperature sensor being lower than the preset threshold. The preset threshold can be manually set based on experience. In some embodiments, if any of the final readings of the temperature sensor is not lower than the preset threshold, it is considered that the preset requirement is not met.

[0102] The temperature change rate can refer to the ratio of the temperature change at a certain point of the mold within a period of time to the time taken for the temperature change to occur.

[0103] In some embodiments, in order to prevent a certain point of the mold from cooling too fast and causing the mold to crack, or cooling too slowly and causing the mold to deform, the preset requirement 340 includes the temperature change rate 340-3 being within a preset threshold range 340-4. In some embodiments, the preset threshold range may include a maximum temperature change rate and a minimum temperature change rate.

[0104] In some embodiments, the preset threshold range may be related to mold features, workpiece features, and historical failure data of the smart mold. The mold features may include the mold material used by the smart mold, etc. The workpiece features may include the workpiece type and workpiece material of the produced workpiece, etc. The historical failure data may include the types of failures that occurred in the smart mold and the number of times each type of failure occurred, etc. The mold features, workpiece features, historical failure data of the smart mold, etc. may be determined based on the historical production data of the smart mold.

[0105] In some embodiments, the control module can determine the preset threshold range in a variety of ways. For example, based on historical production experience, a third preset relationship table including the correspondence between mold features, workpiece features, historical fault data and threshold ranges is constructed, and the preset threshold range is determined by looking up the table.

[0106] In some embodiments, the control module can obtain the temperature change rate of each point in the future period of time based on the temperature sequence 330 composed of the temperatures of multiple time points in the future period of time. For example, for the point corresponding to the temperature sensor A, the control module can select any two temperature values ​​from (a1, a2, a3, a4, a5) and calculate the corresponding temperature change rate. In this example, the control module can calculate and obtain the maximum and minimum values ​​of the temperature change rate of the point corresponding to the temperature sensor A in the future period of time.

[0107] In some embodiments, if the maximum temperature change rate of any point exceeds the maximum temperature change rate, or the minimum temperature change rate of any point is less than the minimum temperature change rate, it is considered that the preset requirement is not met.

[0108] In some embodiments, the control module may output a candidate wind turbine operation plan corresponding to a model that meets preset requirements as a target wind turbine operation plan 350 , and operate the corresponding wind turbine based on the target wind turbine operation plan.

[0109] In some embodiments of the present specification, the final readings of all temperature sensors and the temperature change rates are required to meet the requirements at the same time as preset conditions. This not only avoids the problem that the temperature drop is too small, resulting in the mold temperature quickly exceeding the temperature threshold after the fan stops, but also avoids problems such as mold cracking and deformation caused by too fast or too slow cooling, thereby ensuring the service life of the mold.

[0110] In some embodiments of the present specification, an evaluation model is used to predict the readings of the temperature sensor after the candidate fan operation scheme has been running for a period of time, and then the candidate fan operation scheme that meets the preset requirements is selected as the target fan operation scheme, so that the determination of the fan operation parameters is more reliable, thereby improving the workpiece production quality and the mold service life.

[0111] One of one or more embodiments of the present specification provides a control device for a smart mold, the device comprising at least one processor and at least one memory; the at least one memory is used to store computer instructions; and the at least one processor is used to execute at least part of the computer instructions to implement a control method for the smart mold.

[0112] One of one or more embodiments of the present specification provides a computer-readable storage medium, the storage medium stores computer instructions, and when a computer reads the computer instructions in the storage medium, the computer executes a control method for an intelligent mold.

[0113] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements and corrections to this specification. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

[0114] At the same time, this specification uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of this specification can be appropriately combined.

[0115] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this specification, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0116] Similarly, it should be noted that in order to simplify the description disclosed in this specification and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this specification, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the subject matter of this specification are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0117] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of this specification are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0118] Each patent, patent application, patent application publication, and other materials, such as articles, books, specifications, publications, documents, etc., cited in this specification are hereby incorporated by reference in their entirety. Except for application history documents that are inconsistent with or conflicting with the contents of this specification, documents that limit the broadest scope of the claims of this specification (currently or later attached to this specification) are also excluded. It should be noted that if the descriptions, definitions, and / or use of terms in the materials attached to this specification are inconsistent or conflicting with the contents described in this specification, the descriptions, definitions, and / or use of terms in this specification shall prevail.

[0119] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A fault warning system for an intelligent mold, the system comprising: Intelligent mold, image acquisition device, conveying module, control module; The intelligent mold comprises: an upper mold base provided with a male mold and a lower mold base provided with a female mold; the male mold cooperates with the female mold; The male die and the female die are provided with at least one temperature sensor and at least one pressure sensor; At least one fan is provided on the upper die base and / or the lower die base, and the at least one fan is used to cool the male die and the female die; The image acquisition device is used to acquire an image of a workpiece that has undergone a stamping process from at least one direction; The conveying module is used to convey the raw materials to a workstation matching the smart mold for the stamping process, and to convey the workpiece after the stamping process to a preset position; The control module is configured to: Determine the workpiece morphological features based on the workpiece image after the stamping process, wherein the workpiece morphological features include at least one of the degree of edge burrs, the degree of surface damage, and deformation error; In response to the workpiece morphological feature satisfying the first preset condition, based on the first evaluation feature, the production parameter before adjustment, and the target workpiece morphological feature, the production parameter adjustment amount is determined through a prediction model; and the production parameter is adjusted based on the production parameter adjustment amount; the first evaluation feature includes the workpiece morphological feature before adjusting the production parameter and / or the reading of the pressure sensor; the prediction model is a machine learning model; the production parameters include stamping speed, stamping pressure and stamping oil concentration; Perform stamping production based on the adjusted production parameters and obtain a second evaluation feature; the second evaluation feature includes the workpiece morphological feature after adjusting the production parameters and / or the reading of the pressure sensor; In response to the second evaluation feature satisfying the first preset condition, triggering an early warning; the first preset condition is related to a pressure condition and / or a morphological condition; Determine temperature thresholds based on historical data; In response to a reading of the at least one temperature sensor satisfying a third preset condition, determining an operating parameter of the at least one fan by a preset algorithm based on the production parameter and operating the at least one fan, the third preset condition being related to the temperature threshold; The preset algorithm includes: Randomly generate multiple candidate wind turbine operation plans; Based on each candidate fan operation scheme, determining the reading of the temperature sensor after the candidate fan operation scheme has been running for a period of time through an evaluation model, wherein the evaluation model is a machine learning model; Based on the reading of the temperature sensor after the candidate fan operation scheme has been running for a period of time, the candidate fan operation scheme that meets the preset requirements is determined as the target fan operation scheme; the preset requirements include that the final reading of the temperature sensor is lower than a preset threshold and the temperature change rate is within a preset threshold range.

2. The system according to claim 1, characterized in that The control module is further configured to: Based on the second evaluation feature, determining a fault type and a corresponding fault probability; In response to the failure probability satisfying a second preset condition, triggering the early warning; The second preset condition refers to the condition that the failure probability corresponding to the failure type must meet when the warning needs to be triggered.

3. The system according to claim 1, characterized in that The preset threshold range is related to at least one of a mold feature, a workpiece feature, and historical failure data of the smart mold.

4. A fault warning method for a smart mold, executed based on the control module of the fault warning system for a smart mold according to claim 1, comprising: Determine the workpiece morphological features based on the workpiece image after stamping, wherein the workpiece morphological features include at least one of edge burr degree, surface damage degree, and deformation error; In response to the workpiece morphological feature satisfying the first preset condition, based on the first evaluation feature, the production parameter before adjustment, and the target workpiece morphological feature, the production parameter adjustment amount is determined through a prediction model; and the production parameter is adjusted based on the production parameter adjustment amount; the first evaluation feature includes the workpiece morphological feature and / or the reading of the pressure sensor before adjusting the production parameter; the prediction model is a machine learning model; the production parameters include stamping speed, stamping pressure and stamping oil concentration; Perform stamping production based on the adjusted production parameters and obtain a second evaluation feature; the second evaluation feature includes the workpiece morphological feature after adjusting the production parameters and / or the reading of the pressure sensor; In response to the second evaluation feature satisfying the first preset condition, triggering an early warning; The first preset condition is related to a pressure condition and / or a morphological condition; Determine temperature thresholds based on historical data; In response to a reading of the at least one temperature sensor satisfying a third preset condition, determining an operating parameter of the at least one fan by a preset algorithm based on the production parameter and operating the at least one fan, the third preset condition being related to the temperature threshold; The preset algorithm includes: Randomly generate multiple candidate wind turbine operation plans; Based on each candidate fan operation scheme, determining the reading of the temperature sensor after the candidate fan operation scheme has been running for a period of time through an evaluation model, wherein the evaluation model is a machine learning model; Based on the reading of the temperature sensor after the candidate fan operation scheme has been running for a period of time, the candidate fan operation scheme that meets the preset requirements is determined as the target fan operation scheme; the preset requirements include that the final reading of the temperature sensor is lower than a preset threshold and the temperature change rate is within a preset threshold range.

5. The method according to claim 4, characterized in that The method further comprises: Based on the second evaluation feature, determining a fault type and a corresponding fault probability; In response to the failure probability satisfying a second preset condition, the warning is triggered; the second preset condition refers to a condition that the failure probability corresponding to the failure type needs to satisfy when the warning needs to be triggered.

6. The method according to claim 4, characterized in that The preset threshold range is related to at least one of a mold feature, a workpiece feature, and historical failure data of the smart mold.