A method and system for optimizing process data of a spray-wrapped polyethylene thermal insulation pipe

Abstract

The application provides a spraying and winding polyethylene heat preservation pipe process data optimization method and system, comprising the following steps: placing a processing pipe into a processing device, setting a processing area, acquiring real-time processing images of the processing pipe during spraying and winding work in a spraying area and a winding area, extracting ROI processing images, performing image splicing to acquire real-time spliced processing images, acquiring definition data, performing a judgment operation based on the definition data, if the difference is less than 0, a proportional controller acquires optimal control gain based on root locus optimization of a processing transfer function, adjusting based on the optimal control gain, reacquiring real-time processing images and performing relevant steps, and if the difference is greater than or equal to 0, acquiring spraying evaluation data and winding evaluation data based on the real-time spliced processing images, and performing real-time optimization and adjustment on the processing device. The method can acquire more optimal process data through clearer images, and improve the accuracy of subsequent processing processes.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of pipeline processing, more specifically, relates to a spraying and winding polyethylene heat preservation pipe process data optimization method and system. BACKGROUND

[0002] The spraying and winding polyethylene heat preservation pipe is widely used in central heating, oil and gas transportation, chemical medium transportation and other scenes due to its good heat preservation performance, corrosion resistance and long service life. Whether it is a heating pipe network in a northern city in winter or a long-distance oil and gas transportation pipeline, the polyethylene heat preservation pipe plays an important role in reducing heat or medium loss and ensuring safe and stable transportation. Its quality directly affects the energy transportation efficiency and pipe network operation cost.

[0003] At present, the production process of the spraying and winding polyethylene heat preservation pipe mainly relies on fixed image shooting and collection and fixed parameter setting. In the spraying and winding stages, the image deviation generated during processing often leads to poor precision in the subsequent process. SUMMARY

[0004] To solve the above technical problems, the present application provides a spraying and winding polyethylene heat preservation pipe process data optimization method and system to solve the technical problem of low precision in the traditional heat preservation pipe processing process due to image deviation generated during processing in the prior art.

[0005] The purpose and effect of the spraying and winding polyethylene heat preservation pipe process data optimization method and system of the present application are achieved by the following specific technical means:

[0006] A spraying and winding polyethylene heat preservation pipe process data optimization method comprises the following steps:

[0007] S1: Put the processing pipe into the processing device, and the processing device processes the processing pipe according to the corresponding processing process;

[0008] S2: Set the processing area, which is divided into a spraying area and a winding area;

[0009] S3: Based on the processing device, obtain the real-time processing image of the processing pipe during spraying and winding in the spraying area and the winding area, extract the ROI processing image from the real-time processing image, and perform image splicing operation on the ROI processing image to obtain the real-time splicing processing image;

[0010] S4: Calculate the definition of the real-time splicing processing image based on the gradient method, obtain the definition data, and subtract the definition data from the preset definition threshold, take the obtained difference value as feedback data and feed it back to the proportional controller, and perform judgment operation;

[0011] S41: if the difference value is less than 0, the proportional controller obtains the optimal control gain based on the root locus optimization of the machining transfer function, the optimal control gain including the optimal exposure time and the optimal machining pipe machining rotating speed, the proportional controller adjusts according to the optimal control gain, and step S3 is executed;

[0012] S5: when the difference value is greater than or equal to 0, spraying evaluation data and winding evaluation data are obtained based on the real-time spliced machining image, and the process data of the machining device is optimized and adjusted in real time based on the spraying evaluation data and the winding evaluation data.

[0013] As a further scheme of the application, if the difference value is less than 0, the proportional controller obtains the optimal control gain based on the root locus optimization of the machining transfer function, the optimal control gain including the optimal exposure time and the optimal machining pipe machining rotating speed, the proportional controller adjusts according to the optimal control gain, and includes:

[0014] The exposure time and the machining pipe machining rotating speed are obtained, the exposure time and the machining pipe machining rotating speed are introduced into the proportional controller as inputs, a basic model is established for each input and a coupling relationship is analyzed, a relative gain matrix is obtained, the optimal control structure is confirmed according to the relative gain matrix, a feedforward compensator is added to offset the coupling effect on the optimal control structure, and a multi-input model is established based on the optimal control structure;

[0015] The machining transfer function is established based on the multi-input model, the root locus graph of the machining transfer function is drawn, and a plurality of machining poles are obtained, a dominant machining pole is obtained based on the plurality of machining poles, and the dominant machining pole represents the machining pole that is closest to the imaginary axis and has no zero point near it in the root locus graph;

[0016] The machining transfer function is simplified based on the dominant machining pole, and the machining steady-state error of the simplified machining transfer function is calculated, the optimal control gain is obtained according to the machining steady-state error, the optimal control gain including the optimal exposure time and the optimal machining pipe machining rotating speed, and the proportional controller adjusts according to the optimal control gain.

[0017] As a further scheme of the application, the method further includes:

[0018] The blur threshold and the brightness threshold are obtained based on the preset sharpness threshold, and the blur and the brightness of the real-time spliced machining image are obtained by the proportional controller;

[0019] If the blur is greater than the blur threshold and the brightness is greater than the brightness threshold, the proportional controller reduces the machining pipe rotating speed;

[0020] If the blur is less than the blur threshold and the brightness is less than the brightness threshold, the proportional controller increases the exposure time;

[0021] If the blur is greater than the blur threshold and the brightness is less than the brightness threshold, the proportional controller reduces the rotating speed of the processing pipe while increasing the exposure time.

[0022] As a further scheme of the present application, the real-time processing image based on the spraying area and the winding area of the processing pipe during the spraying work and the winding work is obtained, the ROI processing image is extracted from the real-time processing image, and the image splicing operation is performed on the ROI processing image to obtain the real-time spliced processing image, comprising:

[0023] The processing pipe during the spraying work and the winding work in the real-time processing image is dynamically positioned by a lightweight model, and an initial ROI candidate box is generated;

[0024] The position and size of the ROI candidate box in each real-time processing image are continuously updated based on a target tracking algorithm;

[0025] The ROI processing image in each real-time processing image is dynamically extracted based on the ROI candidate box, and each ROI processing image is aligned and registered;

[0026] The overlapping parts between every four ROI processing images after registration and alignment are obtained, the overlapping parts are eliminated based on hybrid fusion, and the four ROI processing images are spliced to generate a real-time spliced processing image.

[0027] As a further scheme of the present application, the real-time spliced processing image is obtained by obtaining the overlapping parts between every four ROI processing images after registration and alignment, eliminating the overlapping parts based on hybrid fusion, and splicing the four ROI processing images, comprising:

[0028] The distance assignment weight of the pixel position and the overlapping boundary in each ROI processing image is obtained, and each ROI processing image is smoothly transitioned based on the distance assignment weight;

[0029] Based on the image pyramid, the high-frequency part, the medium-frequency part and the low-frequency part of each ROI processing image after smooth transition are decomposed, the low-frequency part is represented as the most blurred layer, the high-frequency part is represented as the most clear layer, and the medium-frequency part is represented as the medium clear layer, the high-frequency part, the medium-frequency part and the low-frequency part are layered fused to obtain the real-time spliced processing image to be optimized;

[0030] An energy function is constructed, the seam path with the smallest difference of the real-time spliced processing image to be optimized is searched based on the energy function, and the ghosting in the real-time spliced processing image to be optimized is eliminated according to the seam path with the smallest difference in the layered fusion;

[0031] The real-time spliced processing image to be optimized is globally optimized to generate a real-time spliced processing image.

[0032] As a further scheme of the present application, the gradient-based method calculates the sharpness of the real-time splicing processing image to obtain sharpness data, including:

[0033] The Sobel operator with a 5*5 kernel is used to calculate the X-axis and Y-axis sharp gradient amplitudes of a 5*5 window in each real-time splicing processing image, and the sharp gradient amplitudes are calculated based on the sum of squares, and the X-axis and Y-axis sharp gradient amplitudes of the 5*5 window in each real-time splicing processing image are normalized.

[0034] The normalization formula is represented as:

[0035] ;

[0036] Wherein, represents the X-axis and Y-axis sharp gradient amplitudes of the normalized 5*5 window, represents the sharp gradient amplitude mean value in the first window, represents the sharp gradient amplitude standard deviation in the first window, represents the X-axis and Y-axis sharp gradient amplitudes of the 5*5 window before normalization.

[0037] The process key area is calibrated, and the process key area represents an 80% area of the processing pipe spraying center width and an area of ±10mm on both sides of the winding joint;

[0038] The sharp gradient amplitude summation calculation is performed based on the process key area to obtain a sharp gradient amplitude sum, and the sharp gradient amplitude sum is linearly mapped to a 0-100% interval to obtain sharpness data.

[0039] As a further scheme of the present application, the sharp gradient amplitude summation calculation is performed based on the process key area to obtain a sharp gradient amplitude sum, including:

[0040] After the sharp gradient amplitude summation calculation, the processing blur length is obtained based on the exposure time and the processing pipe processing speed, and the sharp gradient amplitude sum is noise-optimized based on a noise suppression function.

[0041] The noise suppression function can be represented as:

[0042] ;

[0043] Wherein, represents the sharp gradient amplitude sum after noise optimization, represents the sharp gradient amplitude sum, represents the processing blur length.

[0044] As a further scheme of the present application, the real-time spliced processing image-based spraying evaluation data and winding evaluation data are obtained, comprising:

[0045] The real-time spliced processing image includes a real-time spliced processing image of the processing pipe performing spraying work and a real-time spliced processing image of the processing pipe performing winding work.

[0046] The real-time spliced processing image of the processing pipe performing spraying work is used to obtain processing pipe spraying features, the processing pipe spraying features including defect features and thickness features, the real-time spliced processing image of the processing pipe performing winding work is used to obtain processing pipe winding features, the processing pipe winding features being represented as winding angle deviation, and the spraying evaluation data and the winding evaluation data are obtained according to the processing pipe spraying features and the processing pipe winding features.

[0047] As a further scheme of the present application, the spraying evaluation data and the winding evaluation data are obtained according to the processing pipe spraying features and the processing pipe winding features, comprising:

[0048] A scoring model is established, the processing pipe spraying features and the processing pipe winding features are introduced into the scoring model, and the spraying evaluation data and the winding evaluation data are obtained through the scoring model.

[0049] The spraying evaluation data is represented as a spraying processing score between 0 and 100, and the winding evaluation data is represented as a winding processing score between 0 and 100.

[0050] A spraying and winding polyethylene heat preservation pipe process data optimization system, comprising:

[0051] A processing device, comprising a sensor module, an image module and a processing module.

[0052] The sensor module is used to obtain process data of the processing device.

[0053] The processing module is used to process the processing pipe according to corresponding processing process steps.

[0054] The image module is used to capture and collect the processing pipe in the processing process, and obtain real-time processing images of the processing pipe performing spraying work and winding work in spraying and winding areas.

[0055] The processing module is used to extract ROI processing images of the real-time processing images, and obtain real-time spliced processing images.

[0056] The proportional controller is used to obtain optimal control gain based on root locus optimization of the processing transfer function, and adjust according to the optimal control gain.

[0057] An adjusting module is used to acquire spraying evaluation data and winding evaluation data, and the process data of the processing device can be optimized and adjusted in real time based on the spraying evaluation data and the winding evaluation data.

[0058] Compared with the prior art, the present application has the following beneficial effects:

[0059] Firstly, the processing pipe is placed into the processing device through step S1, the processing area is set through step S2, and then the real-time processing image of the processing pipe during spraying work and winding work in the spraying area and the winding area is acquired through step S3. The real-time processing image is subjected to ROI processing image extraction, and a real-time spliced processing image is acquired. Then, the clarity calculation is performed through step S4 to acquire clarity data, and the clarity data is subtracted from the preset clarity threshold. The difference value is judged to determine whether step S41 of adjusting the control gain needs to be performed. If the difference value is less than 0, step S41 is performed to acquire the best control gain by optimizing the root locus of the processing transfer function, and the proportional controller is adjusted based on the best control gain to improve the clarity of the acquired real-time processing image, so as to optimize and adjust the process data of the subsequent processing device. If the difference value is greater than or equal to 0, step S5 is performed to acquire spraying evaluation data and winding evaluation data based on the real-time spliced processing image, and the process data of the processing device is optimized and adjusted in real time according to the spraying evaluation data and the winding evaluation data, thereby improving the precision of the subsequent processing process and solving the problem of low precision caused by image deviation during processing in the traditional heat preservation pipe processing process. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 is a step flow chart of the spraying and winding polyethylene heat preservation pipe process data optimization method of the present application;

[0061] Figure 2 is a schematic diagram of the spraying and winding polyethylene heat preservation pipe process data optimization system of the present application;

[0062] Figure 3 is a schematic diagram of the processing device in the spraying and winding polyethylene heat preservation pipe process data optimization system of the present application. DETAILED DESCRIPTION

[0063] The embodiments of the present application will be further described in detail below in combination with the drawings and examples. The following examples are used to illustrate the technical solutions of the present application, but cannot be used to limit the protection scope of the present application.

[0064] Example 1: as shown in the accompanying drawings Figure 1 , Figure 2 , Figure 3 ,

[0065] The application provides a spraying and winding polyethylene heat preservation pipe process data optimization method, which is suitable for polyethylene heat preservation pipe processing and comprises the following steps.

[0066] In step S1, the processing pipe is placed into a processing device, and the processing device processes the processing pipe according to a corresponding processing process.

[0067] In step S2, a processing area is set, and the processing area is divided into a spraying area and a winding area.

[0068] Specifically, the spraying area is used for performing a related process flow of spraying work, and the winding area is used for performing a related process flow of winding work.

[0069] In step S3, real-time processing images of the processing pipe during spraying work and winding work in the spraying area and the winding area are obtained based on the processing device, ROI processing images are extracted from the real-time processing images, and image splicing operations are performed on the ROI processing images to obtain real-time spliced processing images.

[0070] In this embodiment, step S3 comprises:

[0071] In step S3-1, a lightweight model is used to dynamically position the processing pipe during spraying work and winding work in the real-time processing images, and an initial ROI candidate box is generated.

[0072] Specifically, a lightweight model such as YOLO or MobileNet-SSD can be used to dynamically position the processing pipe during spraying work and winding work in the real-time processing images, and an initial ROI candidate box is generated.

[0073] In step S3-2, a target tracking algorithm is used to continuously update the position and size of the ROI candidate box in each real-time processing image.

[0074] In step S3-3, the ROI candidate box is used to dynamically extract the ROI processing image in each real-time processing image, and each ROI processing image is aligned.

[0075] It should be noted that the core principle of dynamic ROI extraction is to capture the dynamically changing region of interest in the real-time processing image, such as the processing pipe during spraying work and winding work, by using real-time detection and tracking technology, and to realize accurate alignment and seamless fusion by using the spatio-temporal correlation between multiple frames.

[0076] Specifically, first, a lightweight target detection model such as YOLO is used to locate the ROI candidate frame, and then a tracking algorithm is used to continuously update the position and shape thereof to ensure time consistency. Subsequently, the ROI processing image in each real-time processing image is dynamically extracted through the ROI candidate frame, and the ROI processing image is geometrically registered based on feature matching and motion model prediction to eliminate the differences in perspective and displacement.

[0077] In step S3-4, the overlapping parts between each four registered ROI processing images are obtained, the overlapping parts are eliminated based on hybrid fusion, and the four ROI processing images are spliced to generate a real-time spliced processing image.

[0078] In this embodiment, step S3-4 includes:

[0079] In step S3-4-1, a distance assignment weight of a pixel position in each ROI processing image and an overlapping boundary is obtained, and each ROI processing image is smoothly transitioned based on the distance assignment weight.

[0080] In this embodiment, a fade-in and fade-out method is used to smoothly transition each ROI processing image, and a weight is dynamically assigned according to the distance of the overlapping region pixel and the boundary, such as a linear weight gradually changing from 0 to 1 to smoothly transition the splicing gap. However, when a linear weight is used, the linear weight in a complex texture region may cause blurring, and therefore a gradient sensitive weight can be introduced. If the gradient value of a certain pixel is higher than a threshold value, the weight of the pixel can be increased to retain edge details.

[0081] In step S3-4-2, each ROI processing image after the smooth transition is decomposed into a high-frequency part, a medium-frequency part and a low-frequency part based on an image pyramid. The low-frequency part represents the most blurred layer, the high-frequency part represents the most clear layer, and the medium-frequency part represents the medium clarity layer. The high-frequency part, the medium-frequency part and the low-frequency part are hierarchically fused to obtain a real-time spliced processing image to be optimized.

[0082] In this embodiment, the high-frequency, low-frequency and medium-frequency image information is decomposed by the image pyramid, and is fused respectively to avoid blurring and color difference. For the high-frequency part, i.e., the most clear layer, the pixel with the largest gradient is selected for hierarchical fusion to retain texture details. For the medium-frequency part, i.e., the medium clarity layer, a weight mixing method is selected for hierarchical fusion. For the low-frequency part, i.e., the most blurred layer, an average fusion method is selected for hierarchical fusion to eliminate light difference.

[0083] In step S3-4-3, an energy function is constructed, and a seam path with the smallest difference is searched based on the energy function. The ghosting in the real-time spliced processing image to be optimized is eliminated according to the seam path with the smallest difference in the hierarchical fusion.

[0084] In this embodiment, by constructing an energy function, the maximum flow algorithm is used to find the joint path with the smallest difference in the splicing area, and the ghosting is eliminated.

[0085] It can be understood that for the overlapping part between the multiple ROI processed images, a hybrid strategy based on spatial or temporal weight is adopted, for example, when using fade-in and fade-out, the weight is dynamically assigned according to the distance between the pixel position and the boundary of the overlap, the weight at the boundary is 0.5, and the weight gradually changes to 1 or 0 away from the boundary, to achieve smooth transition, and multi-band fusion is performed by decomposing the image through a Laplacian pyramid, in the low-frequency part (i.e. the most blurred layer), the fusion is performed in a hierarchical fusion manner using an average fusion method to eliminate light differences, in the high-frequency part (i.e. the most clear layer), the maximum gradient value is directly retained to maintain the clarity of details, and in the medium-frequency part (i.e. the medium clarity layer), for the small misalignment caused by motion blur or registration error, the optimal joint finding technology is used, and the graph cut algorithm is used to construct an energy function, which includes color difference, gradient continuity and other constraints, to find the path with the smallest difference in the splicing overlap area as the joint, thereby avoiding ghosting or breaking phenomenon.

[0086] In step S3-4-4, the real-time splicing processed image to be optimized is globally optimized to generate a real-time splicing processed image.

[0087] In this embodiment, after the fusion splicing is completed, it is necessary to ensure the geometric and lighting consistency of the splicing result with the global canvas, which can be realized by solving the Poisson equation through Poisson fusion to combine the gradient field of the ROI processed image with the boundary conditions of the target canvas, while retaining the details of the original ROI processed image, the color transition at the splicing boundary is forced to be natural. For the ROI processed image frequently updated in a dynamic scene, an incremental canvas strategy is adopted, only the changed area is re-rendered through dirty rectangle marking or motion vector detection, instead of full canvas refreshing, to reduce the calculation load. In addition, a lighting equalization mechanism can also be introduced to perform histogram matching or brightness correction on each ROI, so that the lighting conditions of the ROI are consistent with those of the global canvas, avoiding the abrupt change of light and shade caused by environmental changes.

[0088] Specifically, the solution of the Poisson equation can be expressed as:

[0089] ;

[0090] wherein, is a Laplacian operator, represents the gradient field of the ROI processed image, represents the solution of the Poisson equation, which is discretized into a linear equation system when solving the Poisson equation, and is solved by using a preconditioned conjugate gradient method or a multigrid method.

[0091] Specifically, the lightness equalization mechanism is represented as extracting the ROI processed image and the reference region RGB histogram, mapping through the cumulative distribution function, matching the ROI processed image histogram to the reference histogram, limiting the contrast of the lightness uneven ROI processed image, and self-adapting the histogram equalization.

[0092] In step S4, the gradient method is used to calculate the definition of the real-time spliced processed image, obtain the definition data, and subtract the definition threshold from the definition data, and the obtained difference value is fed back to the proportional controller as feedback data, and a judgment operation is performed.

[0093] In this embodiment, step S4 includes:

[0094] In step S4-1, the Sobel operator with a 5x5 kernel is used to calculate the X-axis and Y-axis definition gradient amplitudes of the 5x5 window in each real-time spliced processed image, and the definition gradient amplitudes are calculated based on the sum of squares. The X-axis and Y-axis definition gradient amplitudes of the 5x5 window in each real-time spliced processed image are normalized.

[0095] The normalization formula is represented as:

[0096] ;

[0097] Wherein, represents the normalized X-axis and Y-axis definition gradient amplitudes of the 5x5 window, represents the definition gradient amplitude mean value in the first window, represents the definition gradient amplitude standard deviation in the first window, represents the definition gradient amplitude of the 5x5 window before normalization.

[0098] It can be understood that the 3x3 kernel is prone to noise interference when detecting small textures, such as polyethylene film movement, and has insufficient response to large size features, such as sprayed particle distribution, so the horizontal and vertical gradients are expanded to 5x5 kernel to enhance the sensitivity to medium frequency features.

[0099] For example, the horizontal kernel can be represented as:

[0100] ;

[0101] Wherein, represents the horizontal kernel, which increases the center weight to enhance the edge contrast, and smooths the far-end weight to suppress noise, and when performing horizontal kernel calculation, the horizontal kernel is optimized by using separate convolution, which decomposes the 5x5 kernel into the product of 5x1 and 1x5 vectors, thereby reducing the number of calculations.

[0102] It can be understood that, in the calculation of the clear gradient amplitude, the calculation by the sum of squares can avoid the open operation of the traditional gradient module length, and can preserve the relative intensity.

[0103] In step S4-2, a process key area is calibrated, which is represented as an area of 80% of the processing tube spraying center width and an area of ±10 mm on both sides of the winding joint.

[0104] It can be understood that, only using the clear gradient amplitude of the process key area can exclude the interference of the tube end.

[0105] In step S4-3, the clear gradient amplitude is calculated by summing based on the process key area, the clear gradient amplitude sum is obtained, the clear gradient amplitude sum is linearly mapped to the interval of 0-100%, and the clarity data is obtained.

[0106] Further, after the clear gradient amplitude summing calculation, the processing blur length is obtained based on the exposure time and the processing tube processing speed, and the clear gradient amplitude sum is noise optimized based on a noise suppression function.

[0107] The noise suppression function can be represented as:

[0108] ;

[0109] Wherein, represents the clear gradient amplitude sum after noise optimization, represents the clear gradient amplitude sum, represents the processing blur length.

[0110] It should be noted that the gradient field quantization calculation performs high-sensitivity gradient analysis on the real-time splicing processing image by an extended 5x5 kernel Sobel operator, captures the micro-texture changes of the sprayed coating and the winding fiber, and the core principle is to use a weighted gradient kernel to enhance the response capability to medium-frequency features such as polyethylene film fiber direction and sprayed particle distribution, to replace the traditional module length operation by calculating the gradient energy by sum of squares to improve the calculation efficiency, and to eliminate the light interference by normalization, to suppress the periodic texture interference by the noise suppression function, to ensure that the quantization result is strongly associated with the true process quality.

[0111] In step S41, if the difference is less than 0, the proportional controller obtains the best control gain based on the root locus optimization of the processing transfer function, the best control gain includes the best exposure time and the best processing tube processing speed, the proportional controller adjusts according to the best control gain, and step S3 is executed.

[0112] In this embodiment, step S41 includes:

[0113] Step S41-1, the exposure time and the machining pipe machining speed are obtained, the exposure time and the machining pipe machining speed are introduced into the proportional controller as inputs, a basic model is established for each input and a coupling relationship is analyzed, a relative gain matrix is obtained, the best control structure is confirmed according to the relative gain matrix, a feedforward compensator is added to offset the coupling effect on the best control structure, and a multi-input model is established based on the best control structure.

[0114] Step S41-2, a machining transfer function is established based on the multi-input model, a root locus diagram of the machining transfer function is drawn, and a plurality of machining poles are obtained, a dominant machining pole is obtained based on the plurality of machining poles, and the dominant machining pole represents a machining pole that is closest to the imaginary axis and has no zero point near the imaginary axis in the root locus diagram.

[0115] Step S41-3, the machining transfer function is simplified based on the dominant machining pole, and the machining steady-state error of the simplified machining transfer function is calculated, the best control gain is obtained according to the machining steady-state error, the best control gain includes the best exposure time and the best machining pipe machining speed, and the proportional controller is adjusted according to the best control gain.

[0116] In this embodiment, the machining transfer function can be represented as:

[0117] ;

[0118] wherein, , , , is a machining pole, is a gain;

[0119] Based on the above machining poles, a root locus diagram is drawn, the moving track of the machining poles when the gain K changes is analyzed, the possible pole distribution is determined, and it is assumed that , , , The four points are the four machining poles of the machining transfer function, and the dominant machining pole is obtained according to the four poles, and the dominant machining pole is the pole that is closest to the imaginary axis and has no zero point near the imaginary axis, so that and are the dominant machining poles, , can be ignored, the machining transfer function is simplified based on the dominant machining pole, the original closed-loop transfer function four-order system is simplified into a two-order system, and the best control gain is calculated according to the steady-state error of the two-order closed-loop transfer function, and the proportional controller is adjusted based on the best control gain.

[0120] In this embodiment, the specific way of adjusting the control gain is:

[0121] The blur threshold and the brightness threshold are obtained based on a preset definition threshold, and the blur and brightness of the real-time spliced processing image are obtained by a proportional controller;

[0122] If the blur is greater than the blur threshold and the brightness is greater than the brightness threshold, the proportional controller reduces the processing pipe rotating speed;

[0123] If the blur is less than the blur threshold and the brightness is less than the brightness threshold, the proportional controller increases the exposure time;

[0124] If the blur is greater than the blur threshold and the brightness is less than the brightness threshold, the proportional controller reduces the processing pipe rotating speed and increases the exposure time.

[0125] In step S5, when the difference is greater than or equal to 0, the spraying evaluation data and the winding evaluation data are obtained based on the real-time spliced processing image, and the process data of the processing device is optimized and adjusted in real time based on the spraying evaluation data and the winding evaluation data.

[0126] Specifically, the real-time spliced processing image includes a real-time spliced processing image of the processing pipe performing spraying work and a real-time spliced processing image of the processing pipe performing winding work.

[0127] Specifically, the real-time spliced processing image of the processing pipe performing spraying work is used to obtain the processing pipe spraying feature, and the processing pipe spraying feature includes a defect feature and a thickness feature. The real-time spliced processing image of the processing pipe performing winding work is used to obtain the processing pipe winding feature, and the processing pipe winding feature is represented as a winding angle deviation. The spraying evaluation data and the winding evaluation data are obtained according to the processing pipe spraying feature and the processing pipe winding feature.

[0128] Further, a scoring model is established, the processing pipe spraying feature and the processing pipe winding feature are input into the scoring model, and the spraying evaluation data and the winding evaluation data are obtained by the scoring model. The spraying evaluation data is represented as a spraying processing score, which is between 0 and 100. The winding evaluation data is represented as a winding processing score, which is between 0 and 100.

[0129] Further, the scoring model can be represented as:

[0130] ;

[0131] ;

[0132] wherein, is represented as the thickness feature, is represented as the number of defect features, The processing pipe winding feature is represented, the process data is adjusted in real time based on the built-in process data, the spraying evaluation data and the winding evaluation data in the adjustment module, and the precision of the process processing is improved.

[0133] A spraying winding polyethylene heat preservation pipe process data optimization system is suitable for coating winding polyethylene heat preservation pipe processing, comprising:

[0134] The processing device comprises a sensor module, an image module and a processing module;

[0135] The sensor module is used to acquire the process data of the processing device;

[0136] The processing module processes the processing pipe according to the corresponding processing process steps;

[0137] The image module is used to shoot and collect the processing pipe in the processing process, and acquire real-time processing images;

[0138] The processing module is used to acquire real-time local processing images, real-time splicing processing images and definition data, and can perform judgment operation based on the definition data;

[0139] The proportional controller can acquire control gain based on the root locus optimization of the processing transfer function, and can adjust the control;

[0140] The adjustment module is used to acquire the spraying evaluation data and the winding evaluation data, and can adjust the process data of the processing device in real time based on the spraying evaluation data and the winding evaluation data.

[0141] The specific use mode and effect of the embodiment one are as follows:

[0142] Firstly, the processing pipe is put into the processing device through step S1, the processing area is set through step S2, and then the real-time processing image of the processing pipe in the spraying area and the winding area during the spraying work and the winding work is acquired through step S3. The real-time processing image is extracted as an ROI processing image, and a real-time spliced processing image is acquired. Then, the clarity calculation is performed through step S4, the clarity data is acquired, and the clarity data is subtracted from the preset clarity threshold. The difference value is judged to determine whether the control gain needs to be adjusted through step S41. If the difference value is less than 0, step S41 is executed, the best control gain is acquired by optimizing the root locus of the processing transfer function, and the proportion controller is adjusted based on the best control gain to improve the clarity of the acquired real-time processing image, so as to optimize and adjust the process data of the subsequent processing device. If the difference value is greater than or equal to 0, step S5 is executed, the spraying evaluation data and the winding evaluation data are acquired based on the real-time spliced processing image, and the process data of the processing device is optimized and adjusted in real time according to the spraying evaluation data and the winding evaluation data, so as to improve the precision of the subsequent processing process, and solve the problem of low precision caused by image deviation during processing in the traditional heat preservation pipe processing process.

[0143] The above embodiments can be implemented in whole or in part by software, hardware (such as a circuit), firmware, or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the flow or function described in the embodiments of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by limited (for example, infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid state disk.

[0144] It should be understood that the term "and / or" in this document is merely used to describe associated objects, and can represent three relationships, for example, A and / or B can represent three cases of A alone, A and B, and B alone, where A and B can be singular or plural. In addition, the character " / " in this document generally represents an "or" relationship between the associated objects before and after it, but it can also represent an "and / or" relationship. The specific meaning can be understood according to the context before and after it.

[0145] It should be understood that the size of the sequence number of each process described above in the embodiments of the present application does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0146] The above-described embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can still be modified, or some technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for optimizing the process data of a spray-wrapped polyethylene insulating pipe, characterized in that, The method comprises the following steps: S1: placing the processing pipe into the processing device, which processes the processing pipe according to the corresponding processing process; S2: setting a processing area, which is divided into a spraying area and a winding area; S3: acquiring real-time processing images of the processing pipe in the spraying area and the winding area during spraying and winding, extracting ROI processing images from the real-time processing images, and performing image splicing on the ROI processing images to obtain real-time spliced processing images; S4: calculating the sharpness of the real-time spliced processing images based on the gradient method, obtaining sharpness data, and subtracting the sharpness data from the preset sharpness threshold, taking the difference as feedback data to the proportional controller, and performing a judgment operation; S41: if the difference is less than 0, the proportional controller obtains the best control gain based on the root locus optimization of the processing transfer function, the best control gain includes the best exposure time and the best processing pipe processing speed, the proportional controller adjusts according to the best control gain, and executes step S3; S5: when the difference is greater than or equal to 0, the spraying evaluation data and the winding evaluation data are obtained based on the real-time spliced processing images, and the process data of the processing device is optimized and adjusted in real time based on the spraying evaluation data and the winding evaluation data.

2. A preferred method of process data for spray and wound polyethylene pipe, as claimed in claim 1, wherein If the difference is less than 0, the proportional controller obtains the best control gain based on the root locus optimization of the processing transfer function, the best control gain includes the best exposure time and the best processing pipe processing speed, the proportional controller adjusts according to the best control gain, which includes: obtaining the exposure time and the processing pipe processing speed, inputting the exposure time and the processing pipe processing speed into the proportional controller as inputs, establishing a basic model for each input and analyzing the coupling relationship, obtaining a relative gain matrix, confirming the best control structure according to the relative gain matrix, and adding a feedforward compensator to offset the coupling effect on the best control structure, and establishing a multi-input model based on the best control structure; establishing a processing transfer function based on the multi-input model, drawing a root locus diagram of the processing transfer function, and obtaining a plurality of processing poles, obtaining a dominant processing pole based on the plurality of processing poles, the dominant processing pole representing the processing pole closest to the virtual axis and having no zero point near it in the root locus diagram; simplifying the processing transfer function based on the dominant processing pole, calculating the processing steady-state error of the simplified processing transfer function, and obtaining the best control gain according to the processing steady-state error, the best control gain including the best exposure time and the best processing pipe processing speed, the proportional controller adjusting according to the best control gain.

3. A process data optimization method for spray and wound polyethylene insulating pipe according to claim 2, characterized in that, The method further comprises: obtaining the blur threshold and the brightness threshold based on the preset sharpness threshold, obtaining the blur and brightness of the real-time spliced processing image through the proportional controller; if the blur is greater than the blur threshold and the brightness is greater than the brightness threshold, the proportional controller reduces the processing pipe speed; if the blur is less than the blur threshold and the brightness is less than the brightness threshold, the proportional controller increases the exposure time; If the blur degree is greater than the blur degree threshold and the brightness is less than the brightness threshold, the proportional controller reduces the rotating speed of the processing pipe while increasing the exposure time.

4. A preferred method of process data for spray and wound polyethylene pipe, as claimed in claim 1, wherein The real-time processing image of the processing pipe during the spraying and winding work in the spraying area and the winding area is acquired based on the processing device, the ROI processing image is extracted from the real-time processing image, and the image splicing operation is performed on the ROI processing image to acquire the real-time spliced processing image, including: The processing pipe during the spraying and winding work in the real-time processing image is dynamically positioned by the lightweight model, and an initial ROI candidate box is generated; The position and size of the ROI candidate box in each real-time processing image are continuously updated based on a target tracking algorithm; The ROI processing image in each real-time processing image is dynamically extracted based on the ROI candidate box, and each ROI processing image is aligned by registration; Every four ROI processing images after registration and alignment are overlapped, the overlapped part is eliminated based on hybrid fusion, and the four ROI processing images are spliced to generate a real-time spliced processing image.

5. A process data optimization method for spray and wound polyethylene insulating pipe according to claim 4, characterized in that, The real-time spliced processing image is generated by acquiring the overlapped part between every four ROI processing images after registration and alignment, eliminating the overlapped part based on hybrid fusion, and splicing the four ROI processing images. The distance distribution weight of the pixel position and the overlap boundary in each ROI processing image is acquired, and each ROI processing image is smoothly transitioned based on the distance distribution weight; Each ROI processing image after smooth transition is decomposed into a high-frequency part, a medium-frequency part, and a low-frequency part based on an image pyramid, the low-frequency part is represented as the most blurred layer, the high-frequency part is represented as the most clear layer, and the medium-frequency part is represented as the medium clarity layer, the high-frequency part, the medium-frequency part, and the low-frequency part are layered fused to obtain a real-time spliced processing image to be optimized; An energy function is constructed, the energy function is used to find a seam path with the smallest difference in the real-time spliced processing image to be optimized, and the ghosting in the real-time spliced processing image to be optimized is eliminated according to the seam path with the smallest difference in the layered fusion; The real-time spliced processing image is globally optimized to generate a real-time spliced processing image.

6. A preferred method of process data for spray and wound polyethylene pipe, as claimed in claim 1, wherein The clarity of the real-time spliced processing image is calculated based on the gradient method to obtain clarity data, including: The X-axis and Y-axis clear gradient amplitudes of the 5*5 window in each real-time spliced processing image are calculated based on the Sobel operator with a 5*5 kernel, the clear gradient amplitudes are calculated based on the sum of squares, and the X-axis and Y-axis clear gradient amplitudes of the 5*5 window in each real-time spliced processing image are normalized. The normalization formula is represented as: ； wherein, is the clear gradient magnitude of the X-axis and Y-axis of the normalized 5x5 window, is the clear gradient magnitude of the X-axis and Y-axis of the normalized 5x5 window, is the clear gradient magnitude of the X-axis and Y-axis of the normalized 5x5 window, is the clear gradient magnitude of the X-axis and Y-axis of the normalized 5x5 window, is the clear gradient magnitude of the X-axis and Y-axis of the normalized 5x5 window, is the clear gradient magnitude of the X-axis and Y-axis of the normalized 5x5 window; The process key area is calibrated, the process key area is represented as the 80% area of the processing pipe spraying center width and the area of ±10mm on both sides of the winding seam; The clear gradient amplitude sum is calculated based on the process key area to obtain the clear gradient amplitude sum, the clear gradient amplitude sum is linearly mapped to the 0-100% interval to obtain the clarity data.

7. A preferred method of process data for spray and wound polyethylene pipe, as claimed in claim 6, wherein The clear gradient amplitude summation calculation based on the process key area obtains a clear gradient amplitude sum, including: After the clear gradient amplitude summation calculation, the processing blur length is obtained based on the exposure time and the processing tube processing speed, and the clear gradient amplitude sum is optimized based on the noise suppression function; The noise suppression function can be represented as: ； wherein, denotes the clear gradient amplitude after noise optimization and denotes the clear gradient amplitude and denotes the processing blur length.

8. A preferred method of process data for spray and wound polyethylene pipe, as claimed in claim 1, wherein The real-time spliced processing image based on the real-time spliced processing image obtains spraying evaluation data and winding evaluation data, including: The real-time spliced processing image includes a real-time spliced processing image of the processing tube performing spraying work and a real-time spliced processing image of the processing tube performing winding work; Based on the real-time spliced processing image of the processing tube performing spraying work, the processing tube spraying feature is obtained, and the processing tube spraying feature includes a defect feature and a thickness feature. Based on the real-time spliced processing image of the processing tube performing winding work, the processing tube winding feature is obtained, and the processing tube winding feature is represented as a winding angle deviation. The spraying evaluation data and the winding evaluation data are obtained according to the processing tube spraying feature and the processing tube winding feature.

9. A preferred method of process data for spray and wound polyethylene pipe, according to claim 8, wherein, The spraying evaluation data and the winding evaluation data are obtained according to the processing tube spraying feature and the processing tube winding feature, including: A scoring model is established, the processing tube spraying feature and the processing tube winding feature are imported into the scoring model, and the spraying evaluation data and the winding evaluation data are obtained through the scoring model; The spraying evaluation data is represented as a spraying processing score, which is between 0 and 100, and the winding evaluation data is represented as a winding processing score, which is between 0 and 100.

10. A spray and wound polyethylene insulating pipe process data optimization system, characterized by, Including: The processing device includes a sensor module, an image module, and a processing module; The sensor module is used to obtain process data of the processing device; The processing module processes the processing tube according to the corresponding processing process step; The image module is used to capture and collect the processing tube in the processing process to obtain real-time processing images of the processing tube performing spraying work and winding work in the spraying area and the winding area; The processing module is used to extract ROI processing images from the real-time processing images and obtain real-time spliced processing images; The proportional controller can obtain the best control gain based on the root locus optimization of the processing transfer function, and can be adjusted according to the best control gain; The adjustment module is used to obtain spraying evaluation data and winding evaluation data, and can perform real-time optimization adjustment on the process data of the processing device based on the spraying evaluation data and the winding evaluation data.

Images