A mandrel repair method
Through the combination of image recognition technology and repair devices, the problem of cracks in the mandrel during processing is solved, and the efficient repair and qualification rate of the mandrel is achieved, ensuring that the surface after repair is flat and the material is fully filled.
Patent Information
- Application Number
- CN202411203376.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-30
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-08-30
AI Technical Summary
Existing mandrels are prone to cracks during processing, resulting in a decrease in structural stability and a decrease in pass rate.
Image recognition technology is used to determine and identify the crack position, and the repair material is injected into the repair device to fill the cracks, and combined with grinding, scraper and spraying methods, to ensure that the repair material is fully filled and the surface is flat.
It effectively reduces cracks on the mandrel, improves the pass rate of the mandrel, and ensures that the repaired surface is flat and the material is fully filled through grinding and scraper devices.
Smart Images

Figure CN118720632B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of automobile parts, in particular to a mandrel repairing method. Background Art
[0002] The mandrel is a mechanical part that is usually used to connect rotating parts and maintain the coaxiality of the inner hole and outer circle of the workpiece in large equipment.
[0003] In the prior art, the mandrel is generally processed from a rough blank by lathes, milling machines, drilling machines, grinders, CNC machine tools and other processing equipment. When the mandrel is processed by the processing equipment, errors are prone to occur, resulting in cracks on the mandrel.
[0004] When cracks appear on the mandrel, it is easy to reduce the structural stability of the mandrel, which will cause the quality inspection of the mandrel to fail and reduce the qualified rate of the mandrel. Summary of the invention
[0005] In order to improve the qualified rate of the mandrel and reduce cracks on the mandrel, the present invention provides a mandrel repairing method.
[0006] In a first aspect, the present invention provides a mandrel repair method, which adopts the following technical solution:
[0007] A mandrel repair method, comprising:
[0008] acquiring a mandrel image of the mandrel;
[0009] Judging whether there is a crack feature from a preset crack feature recognition model according to the mandrel image;
[0010] When crack features exist, identifying the crack location from a preset crack location identification model based on the crack features;
[0011] Matching a repair volume from a preset repair volume identification model based on crack characteristics;
[0012] The preset repair device is controlled according to the repair volume to inject the preset repair material into the crack position.
[0013] By adopting the above technical solution, when cracks exist on the mandrel, the repair material is injected into the cracks through the repair device, so that the cracks are filled with the repair material, thereby reducing the cracks on the mandrel and improving the qualified rate of the mandrel.
[0014] Optionally, a polishing method after repair is also included, and the polishing method includes:
[0015] Acquire a repair image after the repair material is injected according to the crack position;
[0016] Matching the polishing range from a preset polishing range recognition model according to the repair image;
[0017] Matching a grinding stroke from a preset grinding stroke matching model according to the grinding range;
[0018] The preset grinding device is controlled according to the grinding stroke to grind the mandrel to make the surface of the crack position flat.
[0019] By adopting the above technical solution, after the repair material is injected through the repair device, it is easy for the repair material to cool too quickly, resulting in pits on the surface of the repair material. At this time, the surface of the mandrel is polished by the grinding device, so that the repair material protruding from the surface of the mandrel is polished flat.
[0020] Optionally, a filling method after polishing is also included, and the filling method includes:
[0021] Acquire a polished image after polishing;
[0022] identifying a chip region from a preset chip region identification model according to the grinding image;
[0023] Controlling a preset cleaning device to clean the cracks according to the debris area, and matching the filling volume from a preset filling volume matching model according to the polishing image;
[0024] Matching the filling position from a preset filling position matching model according to the polishing image;
[0025] Controlling a preset repairing device to inject the repairing material into the filling position according to the filling position, and matching the scraper stroke from a preset scraper stroke matching model according to the filling position and the polishing image;
[0026] The preset scraper device is controlled according to the scraper stroke to fill the repair material into the crack from the filling position.
[0027] By adopting the above technical solution, after the repair material is injected through the repair device, it is easy for some areas of the crack to not be filled in place due to the repair material cooling too quickly. The repair material is scraped towards the crack by the scraper device so that the repair material fills the crack while keeping the surface of the core shaft flat.
[0028] Optionally, the filling method further includes:
[0029] When the scraper device fills the repair material into the crack, a material image of the repair material is obtained;
[0030] Matching the material volume from a preset material volume recognition model according to the material image;
[0031] Matching the replenishment rate from a preset replenishment rate matching model according to the material volume and the polishing image;
[0032] Matching the material position from a preset material position recognition model according to the material image;
[0033] The material replenishing device preset on the scraper device is controlled according to the replenishing rate to spray the mist-like repairing material to the material position.
[0034] By adopting the above technical solution, when the repair material is scraped towards the crack by the scraper device, the repair material is easily lost, resulting in insufficient filling of some areas of the crack. The repair material is replenished in real time by the material replenishment device, thereby improving the convenience of crack repair.
[0035] Optionally, the filling method further includes:
[0036] Acquire a scraper image of the scraper device;
[0037] Matching the scraper position from a preset scraper position recognition model according to the scraper image;
[0038] According to the scraper position and the polishing image, the repair position of the crack under the scraper device is matched from the preset repair position recognition model;
[0039] According to the repair position and the material position, an adjustment path for moving the repair material from the material position to the repair position is matched from a preset adjustment path matching model;
[0040] Matching the blowing parameters from a preset blowing parameter matching model according to the adjustment path;
[0041] The blowing device preset on the scraper device is controlled to blow air toward the repair material according to the blowing parameters.
[0042] By adopting the above technical solution, the repair material is blown by the blowing device, thereby driving the repair material to move in the extension direction of the crack, and then the repair material enters the crack, thereby improving the convenience of crack repair.
[0043] Optionally, also include:
[0044] When crack features exist, the crack length is identified from a preset crack length identification model based on the crack features;
[0045] When the crack length exceeds a preset extension range, a repair path is matched from a preset repair path matching model based on crack characteristics;
[0046] Matching the injection rate curve from a preset injection rate matching model based on crack characteristics and repair path;
[0047] According to the injection rate curve, the preset repair device is controlled to inject the repair material into the crack position along the repair path.
[0048] By adopting the above technical solution, when the crack is too long, the repair device is controlled to inject in the extension direction of the crack, and the injection rate is controlled according to the size of the crack, thereby reducing the waste of repair materials and improving the convenience of crack repair.
[0049] Optionally, a preheating method before repairing is also included, and the preheating method includes:
[0050] Obtaining the mandrel temperature of the mandrel;
[0051] Matching the heating time from a preset heating time matching model according to the core shaft temperature and the preset repair temperature;
[0052] Matching a heating stroke from a preset heating stroke matching model based on crack characteristics;
[0053] The preset heating device is controlled according to the heating time to heat the crack according to the heating stroke.
[0054] By adopting the above technical solution, the crack is preheated by a heating device before injecting the repair material, thereby reducing the rapid cooling of the repair material after contacting the mandrel surface, resulting in the repair material solidifying before completely entering the crack.
[0055] Optionally, a spraying method after repair is also included, and the spraying method includes:
[0056] Based on the crack characteristics, the spray stroke is matched from the preset spray stroke matching model, and the injection temperature after the repair material is injected is obtained according to the crack position;
[0057] Matching a spray rate curve from a preset spray rate matching model according to the spray stroke, injection temperature and preset cooling temperature;
[0058] According to the spray rate curve, the preset spray device is controlled to spray water toward the crack according to the spray stroke to cool the crack.
[0059] By adopting the above technical solution, after the repair material is injected into the crack, the temperature is high, the texture is soft, and it is easy to deform. Water is sprayed on the high-temperature repair material through the spray device, thereby accelerating the cooling speed of the repair material and quickly increasing the hardness of the repair material.
[0060] Optionally, also include:
[0061] When crack features exist, judging whether the crack penetrates the mandrel based on the crack features from a preset penetration judgment model;
[0062] When the crack penetrates the mandrel, a penetration region and a penetration area corresponding to the penetration region are identified from a preset penetration identification model based on crack characteristics;
[0063] When the penetration area falls within the preset closed area range, a blocking area is matched from a preset blocking area matching model according to the penetration area;
[0064] Controlling the preset cutting device to cut out corresponding blocking materials according to the blocking area;
[0065] The preset plugging mechanical arm is controlled according to the penetration area to fix the plugging material to the penetration.
[0066] By adopting the above technical solution, when the crack penetrates the core shaft, injecting the repair material into the crack easily causes the repair material to leak out from the other end of the crack, and one end of the crack is closed by the sealing material, thereby improving the convenience of crack repair.
[0067] In a second aspect, the present application provides a mandrel, which adopts the following technical solution:
[0068] A core shaft comprises a body, one end of the body is provided with a positioning groove for installation and positioning, the other end of the body is provided with a clamping groove for clamping, and the body is provided with reinforcing ribs for enhancing structural strength.
[0069] By adopting the above technical solution, one end of the core shaft is fixed by the positioning groove, the other end of the core shaft is clamped and fixed by the clamping groove, and the overall structural strength of the core shaft is strengthened by the reinforcing ribs, thereby improving the stability of the core shaft in use.
[0070] In summary, the present application includes at least one of the following beneficial technical effects:
[0071] 1. When there are cracks on the mandrel, the repair material is injected into the crack through the repair device, so that the crack is filled with the repair material, thereby reducing the cracks on the mandrel and improving the qualified rate of the mandrel;
[0072] 2. After the repair material is injected through the repair device, it is easy to cause pits on the surface of the repair material due to the repair material cooling too quickly. At this time, the surface of the mandrel is polished by the grinding device, so that the repair material protruding from the surface of the mandrel is polished flat;
[0073] 3. After the repair material is injected through the repair device, it is easy for some areas of the crack to not be filled in place due to the repair material cooling too quickly. The repair material is scraped toward the crack through the scraper device so that the repair material fills the crack while keeping the surface of the mandrel flat. BRIEF DESCRIPTION OF THE DRAWINGS
[0074] Figure 1 It is a schematic diagram of the structure of a mandrel.
[0075] Figure 2 It is a process of mandrel repair method Figure 1 .
[0076] Figure 3 It is a flow chart of the grinding method.
[0077] Figure 4 This is the flow of the filling method Figure 1 .
[0078] Figure 5 This is the flow of the filling method Figure 2 .
[0079] Figure 6 This is the flow of the filling method Figure 3 .
[0080] Figure 7 It is a process of mandrel repair method Figure 2 .
[0081] Figure 8 is a flow chart of the preheating method.
[0082] Fig. 9 It is a flow chart of the spraying method.
[0083] Fig.10 It is a process of mandrel repair method Figure 3 .
[0084] The parts indicated by the numerical symbols in the above drawings are as follows: 1. body; 2. positioning groove; 3. snap-fit groove; 4. reinforcing rib. DETAILED DESCRIPTION
[0085] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the following Figure 1-10 It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0086] Reference Figure 1 The embodiment of the present application discloses a mandrel, including a body 1, one end of the body 1 is coaxially provided with a plurality of positioning grooves 2 for installation and positioning, the positioning groove 2 is provided with the axis of the body 1 as the axis, and the positioning groove 2 is achieved by filling one of the plurality of evenly provided grooves to achieve the positioning effect. The other end of the body 1 is provided with a snap-in groove 3 for snap-in, the snap-in groove 3 is symmetrical about the axis of the body 1, and the other end of the body 1 is snap-in to the component to be installed through the snap-in groove 3. The body 1 is integrally provided with reinforcing ribs 4 for enhancing the structural strength, and a plurality of reinforcing ribs 4 are staggeredly provided, and the parts of the body 1 with lower structural strength are protected by the reinforcing ribs 4.
[0087] Reference Figure 2 Based on the same inventive concept, an embodiment of the present invention provides a mandrel repair method, comprising the following steps:
[0088] Step 100: Acquire a mandrel image of the mandrel.
[0089] The mandrel image refers to a picture of the mandrel surface, which displays the condition of the mandrel surface. The mandrel image can be obtained through a camera. The method for obtaining the mandrel image is selected by the staff according to the actual situation and will not be elaborated here.
[0090] Step 101: judging whether crack features exist from a preset crack feature recognition model according to the mandrel image.
[0091] The crack feature refers to the crack on the surface of the mandrel. The crack feature recognition model refers to a neural network model obtained through sample training in advance. The crack feature recognition model can determine whether there is a crack feature in the image.
[0092] Step 102: When crack features exist, the crack position is identified from a preset crack position identification model based on the crack features.
[0093] The presence of crack features indicates that the mandrel needs to be repaired to remove the crack. The crack location recognition model refers to a neural network model obtained through sample training in advance. The crack location recognition model can identify the location of the crack from the image, where the crack location refers to the coordinates of the center of the crack. The crack location is the result obtained by the crack location recognition model.
[0094] Step 103: Matching a repair volume from a preset repair volume identification model based on crack characteristics.
[0095] Repair material refers to the material used to repair cracks. The repair material is generally a molten alloy liquid. The repair material is selected by the staff according to the actual situation and will not be described here. The repair volume recognition model refers to a neural network model obtained through sample training in advance. The repair volume recognition model can identify the range of cracks in the image and match the volume of repair material required to fill the cracks. The repair volume is the result obtained by matching the repair volume recognition model.
[0096] Step 104: Control a preset repair device according to the repair volume to inject a preset repair material into the crack position.
[0097] The repair device refers to a device used to inject repair material to fill the crack repair mandrel. The repair device and its control method are selected by the staff according to actual conditions and will not be described in detail here.
[0098] Reference Figure 3 , the grinding method comprises the following steps:
[0099] Step 200: Acquire a repair image after the repair material is injected according to the crack position.
[0100] The repair image refers to the image of the mandrel after the repair device has injected the repair material into the crack. The repair image shows the repair status of the crack after the repair. The repair image can be obtained through a camera. The method for obtaining the repair image is selected by the staff according to the actual situation and will not be elaborated here.
[0101] Step 201: matching a polishing range from a preset polishing range recognition model according to the repaired image.
[0102] The grinding range recognition model refers to a neural network model obtained through sample training in advance. The grinding range recognition model can match the range beyond the mandrel surface according to the state of the crack in the image. The grinding range is the result obtained from the grinding range recognition model.
[0103] Step 202: Matching a grinding stroke from a preset grinding stroke matching model according to the grinding range.
[0104] The grinding device refers to a device used to grind the mandrel to keep the surface of the mandrel flat. The grinding device can use a grinding wheel. The grinding device and its control method are selected by the staff according to the actual situation, and will not be described in detail here. The grinding stroke matching model refers to a neural network model obtained through sample training in advance. The grinding stroke matching model can match the path of removing the grinding range through the grinding device. The grinding stroke is the result obtained from the grinding stroke matching model.
[0105] Step 203: Control a preset grinding device to grind the mandrel according to the grinding stroke to make the surface of the crack position smooth.
[0106] The portion of the repair material that exceeds the mandrel surface after solidification is removed by a grinding device, so that the mandrel as a whole meets the processing specifications.
[0107] Reference Figure 4 , the filling method includes the following steps:
[0108] Step 300: Acquire a polishing image after polishing.
[0109] The grinding image refers to a picture of the core shaft after grinding by the grinding device. The grinding image shows the state of the crack after grinding. The grinding image can be obtained through a camera. The method of obtaining the grinding image is selected by the staff according to the actual situation and will not be elaborated here.
[0110] Step 301: identifying a debris area from a preset debris area identification model according to a polishing image.
[0111] The debris region recognition model refers to a neural network model obtained through sample training in advance. The debris region recognition model can recognize the region where debris exists from the image. The debris region is the result recognized by the debris region recognition model.
[0112] Step 302: Control a preset cleaning device to clean cracks according to the debris area, and match a filling volume from a preset filling volume matching model according to the polishing image.
[0113] The cleaning device refers to a device used to clean debris to facilitate further repair of the mandrel. The cleaning device can be a water spray device. The cleaning device and its control method are selected by the staff according to the actual situation and will not be described here. The filling volume matching model refers to a neural network model obtained through sample training in advance. The filling volume matching model can match the volume of the repair material to be filled from the crack state shown in the polishing image. The filling volume is the result obtained from the filling volume matching model.
[0114] Step 303: matching the filling position from a preset filling position matching model according to the polishing image.
[0115] The filling position matching model refers to a neural network model obtained through sample training in advance. The filling position matching model can match the position suitable for injection repair material according to the area of the crack that needs to be filled in the picture. The filling position is the result obtained from the matching of the filling position matching model.
[0116] Step 304: Control a preset repair device to inject the repair material into the filling position according to the filling position, and match the scraper stroke from a preset scraper stroke matching model according to the filling position and the polishing image.
[0117] The scraper device refers to a device used to drive the repair material to move so that the repair material can fill the crack evenly. The scraper device and its control method are selected by the staff according to the actual situation and will not be described in detail here. The scraper stroke matching model refers to a neural network model obtained through sample training in advance. The scraper stroke matching model can match the path of scraping the repair material to the crack to fill the crack through the scraper device. The scraper stroke is the result matched from the scraper stroke matching model.
[0118] Step 305: Control a preset scraper device according to the scraper stroke to fill the repair material into the crack from the filling position.
[0119] The repair material is scraped toward the crack by a scraper device, thereby repairing the crack while keeping the mandrel surface flat.
[0120] Reference Figure 5 , the filling method further comprises the following steps:
[0121] Step 306: When the scraper device fills the repair material into the crack, a material image of the repair material is obtained.
[0122] The material image refers to the image of the repair material on the scraper device. The position of the repair material on the scraper device is displayed based on the material image. The material image can be obtained through a camera set on the scraper device. The method of obtaining the material image is selected by the staff according to the actual situation and will not be elaborated here.
[0123] Step 307: Match the material volume from a preset material volume recognition model according to the material image.
[0124] The material volume recognition model refers to a neural network model obtained through sample training in advance. The material volume recognition model can recognize the remaining volume of the repair material from the image. The material volume is the result recognized by the material volume recognition model.
[0125] Step 308: Match the replenishment rate from a preset replenishment rate matching model according to the material volume and the polishing image.
[0126] The material replenishing device refers to a device used to replenish the repair material. When the repair material is driven by a scraper to move according to the extension method of the crack, the repair material is easily lost, resulting in insufficient repair material to completely fill the crack. The material repairing device replenishes the repair material in a mist form to maintain sufficient repair material.
[0127] The replenishment rate matching model refers to a neural network model obtained through sample training in advance. The replenishment rate matching model can identify the volume required for repair from the polishing image and compare it with the remaining volume of the repair material, so as to obtain the replenishment speed of replenishing the volume of the repair material to a sufficient volume through the material replenishment device. The replenishment rate is the result obtained from the matching of the replenishment rate matching model.
[0128] Step 309: Match the material position from a preset material position recognition model according to the material image.
[0129] The material position recognition model refers to a neural network model obtained through sample training in advance. The material position recognition model can recognize the coordinate position of the material on the scraper device from the material image. The material position is the result recognized by the material position recognition model.
[0130] Step 310: Control the material replenishing device preset on the scraper device to spray the mist repair material to the material position according to the replenishing rate.
[0131] The repair material is replenished by the material replenishing device according to the material position, thereby ensuring that the repair material is sufficient to fill the crack.
[0132] Reference Figure 6, the filling method further comprises the following steps:
[0133] Step 311: Acquire a scraper image of the scraper device.
[0134] The scraper image refers to a picture of the scraper device, which shows the spatial relationship of the scraper device. The scraper image can be obtained through a camera. The method for obtaining the scraper image is selected by the staff according to actual conditions and will not be elaborated here.
[0135] Step 312: Match the scraper position from a preset scraper position recognition model according to the scraper image.
[0136] The scraper position recognition model refers to a neural network model obtained through sample training in advance. The scraper position recognition model can match the coordinate position of the scraper device from the image. The scraper position is the result recognized by the scraper position recognition model.
[0137] Step 313: Match the repair position of the crack under the scraper device from a preset repair position recognition model according to the scraper position and the polishing image.
[0138] The repair position recognition model refers to a neural network model obtained through sample training in advance. The repair position recognition model can compare the position relationship of the crack in the polishing image with the scraper position to identify the coordinates where the scraper device and the crack intersect. The repair position is the result identified by the repair position recognition model.
[0139] Step 314: Matching an adjustment path for moving the repair material from the material position to the repair position from a preset adjustment path matching model according to the repair position and the material position.
[0140] The adjustment path matching model refers to a neural network model obtained through sample training in advance. The adjustment path matching model can match a path for moving the repair material from the material position to the repair position. The adjustment path is the result matched from the adjustment path matching model.
[0141] Step 315: Match the blowing parameters from the preset blowing parameter matching model according to the adjustment path.
[0142] The blowing device refers to a device that moves the repair material by blowing air at it. The blowing device is generally set on the scraper device. The blowing device and its control method are selected by the staff according to the actual situation and will not be described in detail here. The blowing parameter matching model refers to a neural network model obtained through sample training in advance. The blowing parameter matching model can match the blowing force and blowing direction required by the blowing device to drive the repair material to move along the adjustment path. The blowing parameters are the data set formed by the combination of the blowing force and blowing direction matched from the blowing parameter matching model.
[0143] Step 316: Controlling the blowing device preset on the scraper device to blow air toward the repair material according to the blowing parameters.
[0144] The position of the repair material is driven by the air blowing device to always correspond to the position of the crack, thereby reducing the situation where the repair material deviates from the crack during the process of being driven by the scraper, resulting in incomplete crack filling.
[0145] Reference Figure 7 , a mandrel repair method also includes the following steps:
[0146] Step 400: When crack features exist, the crack length is identified from a preset crack length identification model based on the crack features.
[0147] The crack length recognition model refers to a neural network model obtained through sample training in advance. The crack length recognition model can match the extension length of the crack from the image. The crack length is the result recognized from the crack length recognition model.
[0148] Step 401: When the crack length exceeds a preset extension range, a repair path is matched from a preset repair path matching model based on crack characteristics.
[0149] When the crack is too long, it is not easy to completely fill the crack by injecting the repair material in one place. The extension range refers to the crack extension length interval that can completely fill the crack by injecting the repair material in one place. The extension range is selected by the staff according to the actual situation and will not be described here. The crack length exceeds the extension range, which means that the crack cannot be completely filled by injecting the repair material in one place.
[0150] The repair path matching model refers to a neural network model obtained through sample training in advance. The repair path matching model can match the crack filling path corresponding to the direction of the crack shown in the mandrel image. The repair path is the result obtained from the repair path matching model.
[0151] Step 402: Matching an injection rate curve from a preset injection rate matching model based on crack characteristics and repair paths.
[0152] The injection rate matching model refers to a neural network model obtained through sample training in advance. The injection rate matching model can match the injection rate of the corresponding repair material according to the size of the crack to which the repair device moves according to the repair path. The injection rate curve is the injection rate change curve corresponding to the repair path matched by the injection rate matching model.
[0153] Step 403: Control a preset repair device according to the injection rate curve to inject repair material into the crack position along the repair path.
[0154] The repair device is injected according to the direction and size of the crack, thereby reducing the waste of repair materials and improving the utilization rate of repair materials.
[0155] Reference Figure 8 , the preheating method comprises the following steps:
[0156] Step 500: Obtain the mandrel temperature of the mandrel.
[0157] The core shaft temperature refers to the surface temperature of the core shaft. The core shaft temperature can be obtained through remote sensing testing with an infrared tester. The method for obtaining the core shaft temperature is selected by the staff based on actual conditions and will not be elaborated here.
[0158] Step 501: Match the heating time from a preset heating time matching model according to the core shaft temperature and the preset repair temperature.
[0159] The heating device refers to a device used to heat the surface of the mandrel to reduce the cooling rate of the repair material. The heating device can be an electric heating wire. The heating device and its control method are selected by the staff according to the actual situation and are not described in detail here. The repair temperature refers to the temperature of the repair material when it is used in the repair device. The repair temperature is selected by the staff according to the actual situation and is not described in detail here.
[0160] The heating time matching model refers to a neural network model obtained through sample training in advance. The heating time matching model can match the time required for the heating device to heat the mandrel from the mandrel temperature to the repair temperature. The heating time is the result matched from the heating time matching model.
[0161] Step 502: Match a heating stroke from a preset heating stroke matching model based on the crack characteristics.
[0162] The heating stroke matching model refers to a neural network model obtained through sample training in advance. The heating stroke matching model can match the path of uniformly heating the crack area of the mandrel according to the direction of the crack shown in the mandrel image. The heating stroke is the result matched from the heating stroke matching model.
[0163] Step 503: Control a preset heating device according to the heating time to heat the crack according to the heating stroke.
[0164] The crack area on the mandrel is heated by a heating device, thereby reducing the rapid cooling and solidification of the repair material when it contacts the mandrel surface, resulting in incomplete repair.
[0165] Reference Fig. 9 , the spraying method comprises the following steps:
[0166] Step 600: Match the spray stroke from a preset spray stroke matching model based on the crack characteristics, and obtain the injection temperature after the repair material is injected according to the crack position.
[0167] The spray device refers to a device used to spray water onto the core shaft to cool the repair material and thereby increase the hardness of the repair material. The spray device and its control method are selected by the staff according to actual conditions and will not be described in detail here.
[0168] The spray stroke matching model refers to a neural network model obtained through sample training in advance. The spray stroke matching model can match the path of spraying to the crack area of the mandrel to uniformly cool it according to the direction of the crack shown in the mandrel image. The spray stroke is the result matched from the spray stroke matching model.
[0169] Step 601: Matching a spray rate curve from a preset spray rate matching model according to the spray stroke, injection temperature and preset cooling temperature.
[0170] The cooling temperature refers to the temperature at which the repair material cools down, solidifies and hardens. The cooling temperature is selected by the staff according to the actual situation and will not be described here. The spray rate matching model refers to a neural network model obtained through sample training in advance. The spray rate matching model can reduce the temperature of the crack reached by the spray device according to the spray stroke to the spray rate required for the cooling temperature. The spray rate curve is the spray rate change curve corresponding to the spray stroke matched from the spray rate matching model.
[0171] Step 602: According to the spray rate curve, a preset spray device is controlled to spray water toward the crack according to the spray stroke to cool the crack.
[0172] The repair material is relatively soft at high temperatures and is easily deformed by stress. The cooling rate of the repair material is accelerated by the spray device, so that the repair material at the crack can harden quickly and reduce the stress deformation of the high-temperature repair material.
[0173] Reference Fig.10 , a mandrel repair method also includes the following steps:
[0174] Step 700: When crack features exist, judging whether the crack penetrates the mandrel based on the crack features from a preset penetration judgment model.
[0175] The penetration judgment model refers to a neural network model obtained through sample training in advance. The penetration judgment model can judge whether the crack penetrates the core shaft from the core shaft image.
[0176] Step 701: When a crack penetrates the core shaft, a penetration region and a penetration area corresponding to the penetration region are identified from a preset penetration identification model based on crack characteristics.
[0177] The crack penetrates the mandrel, which means that when the repair material is injected into the crack from one end of the crack, the repair material is easy to leak out from the other end of the crack. The penetration recognition model refers to a neural network model obtained through sample training in advance. The penetration recognition model can identify the contours and areas of the two penetrations of the crack from the mandrel image. The penetration area and penetration area are the results identified from the penetration recognition model.
[0178] Step 702: When the penetration opening area falls within the preset closed area range, a blocking area is matched from a preset blocking area matching model according to the penetration opening area.
[0179] The closed area range refers to the range of penetration areas used to select the closed penetration based on the penetration area. Generally, a smaller penetration is selected. The closed area range is selected by the staff based on actual conditions and will not be elaborated here.
[0180] The plugging material refers to the material used to plug the penetration. The plugging material can be heat-resistant cloth. The plugging material is selected by the staff according to the actual situation and will not be described here. The plugging area matching model refers to a neural network model obtained through sample training in advance. The plugging area matching model can match the contour of the plugging material that closes the penetration according to the area of the penetration. The plugging area is the result matched from the plugging area matching model.
[0181] Step 703: Control a preset cutting device to cut out corresponding blocking materials according to the blocking area.
[0182] The cutting device refers to a device used to cut the whole piece of blocking material into blocking materials corresponding to the blocking area. The cutting device and its control method are selected by the staff according to actual conditions and will not be described in detail here.
[0183] Step 704: Control a preset plugging robot arm to fix the plugging material onto the penetration according to the penetration area.
[0184] The plugging mechanical arm refers to a mechanical arm used to fix the plugging material on the penetration to close the penetration. The plugging mechanical arm and its control method are selected by the staff according to the actual situation and are not described in detail here. The plugging material can be attached to the surface of the mandrel by a sticky material and removed by the mechanical arm after the repair is completed. The fixing method and removal method of the plugging material are selected by the staff according to the actual situation and are not described in detail here.
[0185] The plugging material is fixed to one of the penetration openings of the crack on the core shaft by a plugging mechanical arm, thereby reducing the leakage of the repair material and improving the convenience of crack repair.
[0186] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.
Claims
1. A mandrel repair method, characterized in that: include: acquiring a mandrel image of the mandrel; Judging whether there is a crack feature from a preset crack feature recognition model according to the mandrel image; When crack features exist, identifying the crack location from a preset crack location identification model based on the crack features; Matching a repair volume from a preset repair volume identification model based on crack characteristics; Controlling a preset repair device according to the repair volume to inject a preset repair material into the crack position; Also included is a grinding method after repairing, the grinding method comprising: Acquire a repair image after the repair material is injected according to the crack position; Matching the polishing range from a preset polishing range recognition model according to the repair image; Matching a grinding stroke from a preset grinding stroke matching model according to the grinding range; Controlling a preset grinding device to grind the mandrel according to the grinding stroke to make the surface of the crack position smooth; Also included is a filling method after grinding, the filling method comprising: Acquire a polished image after polishing; identifying a chip region from a preset chip region identification model according to the grinding image; Controlling a preset cleaning device to clean the cracks according to the debris area, and matching the filling volume from a preset filling volume matching model according to the polishing image; Matching the filling position from a preset filling position matching model according to the polishing image; Controlling a preset repairing device to inject the repairing material into the filling position according to the filling position, and matching the scraper stroke from a preset scraper stroke matching model according to the filling position and the polishing image; Controlling the preset scraper device according to the scraper stroke to fill the repair material into the crack from the filling position; The filling method further comprises: When the scraper device fills the repair material into the crack, a material image of the repair material is obtained; Matching the material volume from a preset material volume recognition model according to the material image; Matching the replenishment rate from a preset replenishment rate matching model according to the material volume and the polishing image; Matching the material position from a preset material position recognition model according to the material image; The material replenishing device preset on the scraper device is controlled according to the replenishing rate to spray the mist-like repairing material to the material position.
2. A mandrel repair method according to claim 1, characterized in that: The filling method further comprises: Acquire a scraper image of the scraper device; Matching the scraper position from a preset scraper position recognition model according to the scraper image; According to the scraper position and the polishing image, the repair position of the crack under the scraper device is matched from the preset repair position recognition model; According to the repair position and the material position, an adjustment path for moving the repair material from the material position to the repair position is matched from a preset adjustment path matching model; Matching the blowing parameters from a preset blowing parameter matching model according to the adjustment path; The blowing device preset on the scraper device is controlled to blow air toward the repair material according to the blowing parameters.
3. A mandrel repair method according to claim 1, characterized in that: Also includes: When crack features exist, the crack length is identified from a preset crack length identification model based on the crack features; When the crack length exceeds a preset extension range, a repair path is matched from a preset repair path matching model based on crack characteristics; Matching the injection rate curve from a preset injection rate matching model based on crack characteristics and repair path; According to the injection rate curve, the preset repair device is controlled to inject the repair material into the crack position along the repair path.
4. A mandrel repair method according to claim 1, characterized in that: It also includes a preheating method before repairing, the preheating method comprising: Obtaining the mandrel temperature of the mandrel; Matching the heating time from a preset heating time matching model according to the core shaft temperature and the preset repair temperature; Matching a heating stroke from a preset heating stroke matching model based on crack characteristics; The preset heating device is controlled according to the heating time to heat the crack according to the heating stroke.
5. A mandrel repair method according to claim 1, characterized in that: Also included is a spraying method after repair, the spraying method comprising: Based on the crack characteristics, the spray stroke is matched from the preset spray stroke matching model, and the injection temperature after the repair material is injected is obtained according to the crack position; Matching a spray rate curve from a preset spray rate matching model according to the spray stroke, injection temperature and preset cooling temperature; According to the spray rate curve, the preset spray device is controlled to spray water toward the crack according to the spray stroke to cool the crack.
6. A mandrel repair method according to claim 1, characterized in that: Also includes: When crack features exist, judging whether the crack penetrates the mandrel based on the crack features from a preset penetration judgment model; When the crack penetrates the mandrel, a penetration region and a penetration area corresponding to the penetration region are identified from a preset penetration identification model based on crack characteristics; When the penetration area falls within the preset closed area range, a blocking area is matched from a preset blocking area matching model according to the penetration area; Controlling the preset cutting device to cut out corresponding blocking materials according to the blocking area; The preset plugging mechanical arm is controlled according to the penetration area to fix the plugging material to the penetration.
