Welding repair method for crack and wear defects of turbine runner blade

Abstract

The invention relates to the technical field of water turbine runner blade welding repair, in particular to a water turbine runner blade crack and wear defect welding repair method, which comprises the following steps: extracting a first welding repair feature of welding repair blade information through a welding repair analysis model, and classifying the same welding repair category based on a clustering algorithm; the method comprises the following steps: acquiring corresponding welding repair parameters and welding repair steps, generating a time sequence of the welding repair parameters according to a time sequence, performing primary parameter optimization through a welding repair analysis model, performing secondary optimization on the welding repair parameters through a particle swarm model, and optimizing the welding repair steps. Associating the welding repair blade information, the welding repair parameters and the welding repair steps; by means of the method, the welding repair work of the runner blades with different abrasions and cracks can be classified, so that the existing work content can be optimized, the time of the welding repair work is shortened, and the economic loss of hydropower station power generation is reduced.

Description

Technical Field

[0001] The invention relates to the technical field of welding repair of turbine runner blades, in particular to a welding repair method for cracks and wear defects of turbine runner blades. Background Art

[0002] As an important facility for hydroelectric power generation, water turbines use water power to drive the runner blades of the water turbine to rotate, and then convert the mechanical energy that drives the runner blades to rotate into electrical energy through the internal structure of the water turbine, thereby achieving the purpose of converting mechanical energy into electrical energy.

[0003] Since the runner blades of the turbine are easily affected by water pressure, high-speed water flow and impurities such as mud and sand in the water during use, the runner blades are prone to cracks and wear. In order to ensure the normal operation of the turbine power generation, these damaged runner blades are usually repaired by welding.

[0004] Since the causes of cracks and wear are different, different welding repair methods need to be used for different situations during the welding repair process. For example, the material should be consistent with the damaged runner blades. At the same time, corresponding welding current, voltage, temperature and other process parameters are required according to the crack and wear positions of the runner blades. Non-destructive testing is also required after welding repair. Since the welding repair of runner blades involves too many processes and related parameters, the welding repair process will take a lot of time. During the welding repair process, the shutdown of the turbine is also likely to cause a large amount of economic losses in power generation to the hydropower plant.

[0005] In order to improve the efficiency of welding repair and reduce the economic loss of power generation, the present invention proposes a welding repair method for cracks and wear defects of turbine runner blades. Summary of the invention

[0006] The purpose of the present invention is to provide a welding repair method for cracks and wear defects of turbine runner blades. Through the welding repair blade information of the historical welding repair scheme, the corresponding first welding repair feature is extracted according to the welding repair analysis model and the historical welding repair scheme is associated; the same welding repair category is classified based on the clustering algorithm, and the corresponding welding repair parameters and welding repair steps are obtained; and a time series of welding repair parameters is generated in chronological order, and a parameter optimization is performed once through the welding repair analysis model, and the welding repair parameters are optimized twice through the particle swarm model, and the welding repair steps are optimized, and the welding repair blade information, welding repair parameters and welding repair steps are associated; through the above method, the welding repair work of runner blades with different wear and cracks can be classified, so as to optimize the existing work content, reduce the time of welding repair work, and reduce the loss of the power generation economy of the hydropower station.

[0007] To achieve the above object, the present invention provides the following technical solutions: The welding repair method for cracks and wear defects of turbine runner blades includes: Obtaining a historical welding repair scheme, and extracting welding repair blade information corresponding to the historical welding repair scheme; the welding repair blade information is image information and video information of the rotary blade that needs welding repair; Extracting a first welding repair feature of the welding repair blade information according to a welding repair analysis model; The welding repair analysis model includes a welding repair information processing unit, a welding repair feature recognition unit, a welding repair feature analysis unit and a first welding repair feature output unit; The welding repair information processing unit performs image preprocessing on the welding repair blade information; The welding repair feature recognition unit recognizes the welding repair defect feature based on the pre-processed welding repair blade information; the welding repair feature recognition unit uses yolov5 to recognize the welding repair defect feature; The welding repair feature analysis unit analyzes the global welding repair factor and the local welding repair factor according to the welding repair defect feature; Furthermore, the welding repair feature analysis unit generates the global welding repair factor according to the overall pixel ratio of the welding repair defect feature on the blade identified by yolov5; the welding repair feature analysis unit obtains the defect line of the welding repair defect feature through DeepLSD and calculates the number of pixels of the defect line to generate the local welding repair factor; The first welding repair feature output unit integrates the global welding repair factor and the local welding repair factor to generate the first welding repair feature; The historical welding repair scheme corresponding to the first welding repair feature association; Clustering is performed through the first welding repair feature, calling the historical welding repair scheme of the same welding repair category according to the clustered first welding repair feature, and obtaining corresponding welding repair parameters and welding repair steps; Further, the global welding repair factor is divided into a plurality of level labels; the level label of the global welding repair factor is used as the horizontal coordinate, and the local welding repair factor is used as the vertical coordinate to generate welding repair cluster coordinates, and the plurality of welding repair cluster coordinates are obtained by using the k-means algorithm; Marking a plurality of the welding cluster sets with welding cluster labels; calling the welding repair parameters and the welding repair steps corresponding to the historical welding repair schemes by using the same welding cluster labels, and marking the welding repair parameters and the welding repair steps with the corresponding welding cluster labels; Based on the clustered welding repair steps and the welding repair steps, an initial welding repair scheme for each welding repair category is generated; according to the time sequence of the initial welding repair scheme, the welding repair parameters are formed into a welding repair parameter sequence, and the abnormal welding repair parameters of the welding repair parameter sequence are identified according to the welding repair analysis model to perform a parameter optimization; The same welding repair parameters are used to generate the welding repair parameter sequence according to the time sequence, wherein the welding repair parameter sequence includes a plurality of welding repair parameter sequences; each welding repair parameter sequence is input into the welding repair analysis model; The welding repair analysis model also includes a welding parameter standardization unit, a welding parameter abnormality identification unit and a welding parameter abnormality marking unit; The welding parameter standardization unit divides each welding repair parameter sequence into time windows and performs standardization processing; the welding parameter abnormality identification unit identifies welding parameter abnormalities in the standardized welding repair parameter sequence; The welding parameter anomaly recognition unit includes 1 transformer layer, 4 time convolution layers, 6 LSTM layers and 1 feature fusion layer; The welding parameter abnormality marking unit marks and outputs the welding repair parameter sequence identified as the welding parameter abnormality; Performing secondary parameter optimization on the welding repair parameters of the primary parameter optimization by using a particle swarm model; Further, the speed and position of the particles are initialized with the repair time, repair cost and risk of the welding repair parameters optimized in the first parameter optimization, the minimum welding repair time is used as the target condition, the welding repair cost and the welding repair risk are used as the penalty coefficient, the welding repair parameters after the first parameter optimization are optimized, and the optimal welding repair parameters are generated; According to the abnormal welding repair parameters after the primary parameter optimization, corresponding welding repair steps are screened and eliminated; and the welding repair steps after elimination are optimized by using the particle swarm model; Further, the welding repair step is digitized, and the digitized repair time, repair cost and risk of the welding repair step are used as the speed and position of the particle to initialize the welding repair step, and the minimum welding repair time is used as the target condition, and the welding repair cost and welding repair risk are used as the penalty coefficient to optimize the welding repair step to generate the optimal welding repair step; Associating the first welding repair feature, the optimized welding repair step and the welding repair parameter optimized by the secondary parameter to form a welding repair solution network for welding repair work calling; Furthermore, the first welding repair feature, the optimized welding repair step and the welding repair parameters optimized by the secondary parameters generate an initialized welding associated weight, and the welding associated weight is reduced or enhanced according to the call result.

[0008] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention obtains the first welding repair feature of each welding repair blade information in the historical welding repair scheme and the abnormal data in the welding process through the welding repair analysis model. The method can divide the existing multiple cracks and wear defects based on historical data to obtain the welding treatment process for the same type of cracks and wear defects. At the same time, it can also screen out some welding repair parameters that are more obviously not conducive to reducing the welding repair time, which is convenient for the subsequent optimization process of the welding scheme, and provides a stable data basis for the optimization of the subsequent welding parameters, reduces the data interference between the welding repair parameters and steps of different wear and crack problems, improves the accuracy of the optimization of the welding repair scheme, and is also convenient for improving the efficiency of the welding repair scheme.

[0009] 2. The present invention can obtain various categories of defects and cracks based on the identified first welding repair features and cluster these feature data through a clustering method, and associate corresponding welding repair parameters and welding repair steps; this method can quickly provide corresponding welding repair parameters and welding repair steps based on existing data features, and provide welding repair data corresponding to different problem situations for subsequent welding repair time optimization, so as to facilitate the optimization and classification of subsequent welding repair plans and improve the efficiency of welding repair.

[0010] 3. The present invention uses a particle swarm algorithm to take time data as the target, cost data and repair risk data as penalty coefficients, to optimize welding repair parameters and welding repair steps, and obtain optimal welding repair parameters and welding repair steps; this method can not only optimize from a time perspective to improve the welding repair efficiency of different cracks and defect problems; it can also be considered from a cost perspective and the risk of the repair process, while ensuring the welding repair efficiency. Reduce costs and improve the safety of welding repair personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 is a flow chart of the method of the present invention; Figure 2 A virtual device diagram of a welding repair information processing unit, a welding repair feature recognition unit, a welding repair feature analysis unit and a first welding repair feature output unit of a welding repair analysis model of the present invention; Figure 3 A virtual device diagram of a welding parameter standardization unit, a welding parameter abnormality identification unit and a welding parameter abnormality marking unit of the welding repair analysis model of the present invention; Figure 4 It is a defect line diagram of the crack of the runner blade of DeepLSD of the present invention; Figure 5 A network structure diagram for identifying welding repair parameter sequences for the welding repair analysis model of the present invention. DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0013] The runner blades of a water turbine are an important structure for water turbine power generation. The wear and cracks caused during the use of the runner blades have a significant impact on the power generation efficiency and power generation safety of hydropower. Therefore, once the above problems occur, the runner blades need to be welded and repaired. Since different welding repair schemes are required for different problems, the welding repair takes a long time, and the water turbine needs to be shut down for repair during the repair process, which greatly reduces the economic loss of power generation. For this reason, the present invention provides a welding repair method for cracks and wear defects of water turbine runner blades, referring to Figure 1 As shown, the technical solution is as follows: Obtaining a historical welding repair scheme, and extracting welding repair blade information corresponding to the historical welding repair scheme; Extracting a first welding repair feature of the welded repair blade information according to the welding repair analysis model, and associating the corresponding historical welding repair scheme based on the first welding repair feature; Clustering is performed through the first welding repair feature, calling the historical welding repair scheme of the same welding repair category according to the clustered first welding repair feature, and obtaining corresponding welding repair parameters and welding repair steps; Based on the clustered welding repair steps and the welding repair steps, an initial welding repair scheme for each welding repair category is generated; according to the time sequence of the initial welding repair scheme, the welding repair parameters are formed into a welding repair parameter sequence, and the abnormal welding repair parameters of the welding repair parameter sequence are identified according to the welding repair analysis model for primary parameter optimization; the welding repair parameters of the primary parameter optimization are secondary optimized by a particle swarm model; According to the abnormal welding repair parameters after the primary parameter optimization, corresponding welding repair steps are screened and eliminated; and the welding repair steps after elimination are optimized by using the particle swarm model; The first welding repair feature, the optimized welding repair step and the welding repair parameters optimized by the secondary parameters are associated to form a welding repair solution network for welding repair work calls.

[0014] Based on the welding repair blade information of different wear and crack problems in the historical welding repair scheme, the first welding repair features of different blade wear and cracks are extracted through the welding repair analysis model, and clustering is performed through the extracted first welding repair features. Through this method, the existing various wear and crack problems can be classified in the first stage, and the welding repair parameters and welding repair steps corresponding to the same crack and wear problems can be called based on the existing various problems for further optimization, providing a more accurate data basis for the subsequent optimization of welding repair work, so as to achieve the purpose of improving welding repair efficiency and reducing economic losses in power generation; Based on the clustered welding repair steps and welding repair parameters, the welding repair parameters are formed into a time series in chronological order, and the abnormal data is extracted for primary optimization and the corresponding welding repair steps are screened out according to the abnormal data; the welding repair parameters and welding repair steps after the primary optimization are secondary optimized through the particle swarm model. Through the primary optimization, data different from other welding repair parameters can be excluded. At the same time, the possible misjudgment problem caused by technical improvement or equipment improvement is also taken into account. Therefore, feature learning based on time series can exclude some accidentally abnormal welding repair parameters and retain the normal welding repair parameters that have been retained for a long time, so as to minimize the possibility of misjudgment; At the same time, based on the particle swarm algorithm, various welding repair parameters and welding repair steps can be comprehensively optimized in terms of efficiency, etc., to extract the optimal welding repair parameters and welding repair steps under the same crack and wear problems, and generate a welding repair network in combination with the corresponding welding repair blade information. This not only makes it convenient for welding personnel to call welding repair plans to maximize welding repair efficiency and reduce economic losses in power generation, but also enables actual optimization work based on the actual repair work.

[0015] Embodiment 1 For the purpose of specific explanation, the welding repair process of the same crack problem and wear problem is summarized as follows: Obtain a historical welding repair scheme, and extract the welding repair blade information corresponding to the historical welding repair scheme; the welding repair blade information is the image information and video information of the rotating blade that needs welding repair; if it is video information, extract the image information in the video frame by frame as input data; Based on the image data, we can more clearly understand the crack and wear problems of the existing runner blades, which is convenient for subsequent data training, data analysis and data feature storage, as well as subsequent archiving and problem visualization. It can ensure the accuracy of the information analysis of the welding repair blade problem to the greatest extent, so as to provide accurate problem data to facilitate the optimization process of the subsequent welding repair work and improve the efficiency of the welding repair work. Extracting a first welding repair feature of the welding repair blade information according to a welding repair analysis model; The welding repair analysis model includes a welding repair information processing unit, a welding repair feature recognition unit, a welding repair feature analysis unit and a first welding repair feature output unit. Figure 2 As shown; The welding repair information processing unit performs image preprocessing on the welding repair blade information; the preprocessing includes Gaussian filtering denoising, Laplace operator edge processing and image data normalization; The welding repair feature recognition unit recognizes the welding repair defect feature based on the pre-processed welding repair blade information; the welding repair feature recognition unit uses yolov5 to recognize the welding repair defect feature; The welding repair feature analysis unit analyzes the global welding repair factor and the local welding repair factor according to the welding repair defect feature; in order to ensure that the image of the runner blade is relatively complete as a whole and each image can be unified, each image only contains one runner blade and the resolution of each image is consistent after image cutting and other operations; Furthermore, the welding repair feature analysis unit generates the global welding repair factor according to the overall pixel ratio of the welding repair defect feature identified by yolov5 on the blade; the global welding repair factor is calculated as follows: ; in, is the global welding repair factor, is the total number of pixels of the yolov5 candidate box, is the total number of pixels in the entire image; The welding repair feature analysis unit obtains the defect line of the welding repair defect feature through DeepLSD and calculates the number of pixels of the defect line to generate the local welding repair factor. Figure 4 As shown; the number of pixels of the defective line is the total number of pixels of the image data through which the defective line generated after identification passes; Based on yolov5, the crack and wear data in the image data and video data of the runner blade can be feature extracted. At the same time, the feature extraction frame based on yolov5 can also determine the pixel range of the crack and wear data, so as to obtain the overall characteristics of the crack and wear data on the entire runner blade; based on DeepLSD, the direction data in the crack and wear data can be obtained. Through the line features in these image data, the local data of these problem data can be extracted. Combining the overall data and the local data can more fully obtain the recognition results, improve the recognition accuracy of the crack and wear data, and facilitate the subsequent data induction and summary; The first welding repair feature output unit integrates the global welding repair factor and the local welding repair factor to generate the first welding repair feature; The model can be used to extract unified data features from the welding repair blade information provided in the historical welding repair scheme. At the same time, based on the extracted problem data features, it can also provide a data basis for the subsequent classification of data with the same problem, so as to achieve targeted optimization for different cracks and wear problems, improve the accuracy of welding repair work optimization, and realize a more efficient and accurate welding repair optimization process. The historical welding repair scheme corresponding to the first welding repair feature association; Clustering is performed through the first welding repair feature, calling the historical welding repair scheme of the same welding repair category according to the clustered first welding repair feature, and obtaining corresponding welding repair parameters and welding repair steps; Furthermore, the global welding repair factor is divided into multiple level labels; the level label of the global welding repair factor is used as the horizontal coordinate, and the local welding repair factor is used as the vertical coordinate to generate welding repair cluster coordinates, and the multiple welding repair cluster coordinates are obtained by k-means algorithm. In this embodiment, the level label The calculation formula is: ; The calculation expression of is not unique, and can also be expressed based on expert experience and other calculation formulas; Marking a plurality of the welding cluster sets as welding cluster labels; calling the welding repair parameters (for example, but not limited to welding current, welding voltage, welding speed, etc.) and the welding repair steps (for example, but not limited to preheating, segmented welding, post-weld heat treatment, non-destructive testing, etc.) corresponding to the historical welding repair scheme through the same welding cluster label, and marking the welding repair parameters and the welding repair steps with the corresponding welding cluster label; By converting all the data, the features recognized by yolov5 can be quantified, and combined with DeepLSD to form corresponding data coordinates. Through these data coordinates, multiple categories can be obtained based on the clustering algorithm to summarize the same crack problems and wear problems. This method can generate corresponding clustering labels for all coordinate data, which is convenient for obtaining the corresponding category recognition results from the data feature dimensions of the blade, so as to obtain the corresponding welding repair parameters and welding repair processes for subsequent summarization, providing a solid data foundation for improving the efficiency of welding repair.

[0016] In this embodiment, the recognition results of cracks and wear of runner blades by several commonly used target detection algorithms are compared, and multiple detections are performed using the same image data set. The highest recognition accuracy is selected as the final result. The recognition results are shown in Table 1: Table 1 Recognition accuracy of various target detection algorithms for cracks and wear of runner blades Object Detection Algorithms Recognition accuracy yolov5 92.84% R-FCN 92.61% Fast R-CNN 92.06% Faster R-CNN 92.67% From the results in Table 1, it can be seen that yolov5 has the highest recognition accuracy for cracks and wear of runner blades; Embodiment 2 For the purpose of specific explanation, the optimization process of the welding repair solution is described in combination with the following contents: Based on the clustered welding repair steps and the welding repair steps, an initial welding repair scheme for each welding repair category is generated; according to the time sequence of the initial welding repair scheme, the welding repair parameters are formed into a welding repair parameter sequence, and the abnormal welding repair parameters of the welding repair parameter sequence are identified according to the welding repair analysis model to perform a parameter optimization; The same welding repair parameters are used to generate the welding repair parameter sequence according to the time sequence, wherein the welding repair parameter sequence includes a plurality of welding repair parameter sequences; each welding repair parameter sequence is input into the welding repair analysis model; The welding repair analysis model also includes a welding parameter standardization unit, a welding parameter abnormality identification unit and a welding parameter abnormality marking unit. Figure 3 As shown; The welding parameter standardization unit divides each welding repair parameter sequence into time windows and performs standardization processing; the time window is to set the time window length according to expert experience to divide the data, and the standardization processing adopts Z-score for standardization; The welding parameter abnormality identification unit identifies welding parameter abnormalities for the standardized welding repair parameter sequence; Reference Figure 5 As shown, the welding parameter abnormality recognition unit includes 1 transformer layer, 4 time convolution layers, 6 LSTM layers and 1 feature fusion layer; the transformer layer corresponds to Figure 5 The converter in LSTM layer corresponds to Figure 5 The long short-term memory layer in The transformer layer and the time convolution layer can strengthen the extraction process of abnormal features. At the same time, the LSTM layer can also ensure learning based on the time features of time series data to reduce the abnormal recognition errors caused by welding equipment upgrades or technology upgrades. The welding processing parameters are screened from different dimensions of abnormal features to ensure the accuracy of the screened data, and also ensure the subsequent data optimization work to reduce the efficiency of welding repair work; In this embodiment, some welding repair parameter sequences containing anomalies are generated and annotated in combination with expert data processing, forming a total of 5 welding repair parameter sequences. Each welding repair parameter sequence has not been trained with data to simulate data anomaly recognition in a real environment. The anomaly recognition results of the welding repair analysis model are shown in Table 2: Table 2 Recognition accuracy and precision of welding repair analysis model for 5 welding repair parameter sequences Welding repair parameter sequence Anomaly recognition accuracy Anomaly recognition accuracy Sequence 1 90.17% 92.12% Sequence 2 89.88% 91.53% Sequence 3 90.52% 91.67% Sequence 4 90.03% 92.66% Sequence 5 89.79% 91.85% From the results in Table 1, it can be seen that the accuracy of abnormal recognition is around 90%, the precision of abnormal recognition is between 91% and 93%, and the recognition results of welding repair parameter sequences are good; The welding parameter abnormality marking unit marks and outputs the welding repair parameter sequence identified as the welding parameter abnormality; the abnormal welding repair parameter is a short-term obvious data rise or obvious data drop in the time series, and these abnormal data will be output; the output result is the data label of the data segment after the time window is divided, for example, the data label of normal data is 0, and the data label of abnormal data is 1; by arranging the welding repair parameters according to the time sequence and extracting the abnormal characteristics of the welding repair parameters through the welding repair analysis model, it is possible to screen some welding repair parameters with obvious abnormalities based on the same crack problem and wear problem, and at the same time, it is also possible to reduce the abnormal identification error caused by the upgrade of welding equipment or technology, and improve the accuracy of these abnormal parameters, so as to reduce the interference of these abnormal data on the particle swarm model optimization process during the optimization process, improve the accuracy of data optimization, and facilitate more accurate reduction of welding repair time and reduce the downtime of turbine welding repair project; in order to ensure the accuracy of data screening, after output, the marked abnormal data will be secondary identified by relevant technical personnel, so as to exclude data that is beneficial to the optimization of welding scheme but identified as abnormal; Performing secondary parameter optimization on the welding repair parameters of the primary parameter optimization by using a particle swarm model; Further, the speed and position of the particle are initialized with the repair time, repair cost and danger of the welding repair parameters optimized in the first parameter optimization, the minimum welding repair time is used as the target condition, the welding repair cost and the welding repair danger are used as the penalty coefficient, the welding repair parameters after the first parameter optimization are optimized to generate the optimal welding repair parameters; specifically, the corresponding welding repair parameter weights are generated according to the repair time, repair cost and danger corresponding to the welding repair parameters, and the speed and position of the particle are initialized based on the welding repair parameter weights, wherein the repair time generates the corresponding weight according to the length of the repair time, the repair cost generates the corresponding weight according to the size of the cost investment, and the danger generates the corresponding weight according to the size of the danger degree, and the setting of the weight value is not unique, and can be set based on existing data combined with expert experience; Taking the minimum welding repair time as the target condition, the welding repair parameters used for the crack and wear problems of the same type of runner blades can be guaranteed to achieve the minimum welding repair time. At the same time, the cost problem and the danger of welding repair are also taken into consideration. Therefore, taking these two parameters as penalty coefficients can ensure the safe implementation of the welding repair process while minimizing the welding repair time and cost investment to ensure the maximization of the turbine power generation economy. According to the abnormal welding repair parameters after the primary parameter optimization, corresponding welding repair steps are screened and eliminated; and the welding repair steps after elimination are optimized by using the particle swarm model; Furthermore, the welding repair step is digitized, and the repair time, repair cost and danger of the digitized welding repair step are used as the speed and position of the particle for initialization. The welding repair step is optimized with the minimum welding repair time as the target condition, and the welding repair cost and welding repair danger as the penalty coefficient to generate the optimal welding repair step. The content of the welding step digitization is to generate corresponding data labels according to the steps, such as preheating is 1, segmented welding is 2, etc. The welding repair step also initializes the speed and position of the particle with the welding repair step weights of the repair time, repair cost and danger. The welding repair step weight generation method is the same as the welding repair parameter weight. After the welding steps are quantified, the same optimization process as the welding repair parameters is adopted, with the minimum welding repair time as the goal and the cost and danger as the penalty coefficients. The same optimization process can ensure that the welding repair steps are consistent with the welding repair parameters in the optimization data analysis dimension, ensure the accuracy of the data in the optimization dimension, and also obtain the best welding optimization steps based on the same optimization goal, which can not only improve the efficiency of welding repair from the aspect of welding repair parameters, but also improve the efficiency of welding repair work, and reduce the downtime of turbines during welding repair process; Associating the first welding repair feature, the optimized welding repair step and the welding repair parameter optimized by the secondary parameter to form a welding repair solution network for welding repair work calling; Further, the first welding repair feature, the optimized welding repair step and the welding repair parameter optimized by the secondary parameters generate an initialization welding association weight, and the welding association weight is reduced or enhanced according to the call result; Through the association weights, corresponding data can be fed back for the results of welding repair in actual welding repair projects to understand whether the optimization plan can achieve the purpose of complete welding repair and reduce welding repair time. By reducing or increasing the welding association weights, the optimized welding repair parameters and the optimization results of the welding repair plan can be understood, and subsequent adjustments and re-optimization of the welding repair plan can be made.

[0017] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A welding repair method for cracks and wear defects of turbine runner blades, characterized in that: include: Obtaining a historical welding repair scheme, and extracting welding repair blade information corresponding to the historical welding repair scheme; Extracting a first welding repair feature of the welded repair blade information according to the welding repair analysis model, and associating the corresponding historical welding repair scheme based on the first welding repair feature; Clustering is performed through the first welding repair feature, calling the historical welding repair scheme of the same welding repair category according to the clustered first welding repair feature, and obtaining corresponding welding repair parameters and welding repair steps; Based on the clustered welding repair steps and the welding repair steps, an initial welding repair scheme for each welding repair category is generated; according to the time sequence of the initial welding repair scheme, the welding repair parameters are formed into a welding repair parameter sequence, and the abnormal welding repair parameters of the welding repair parameter sequence are identified according to the welding repair analysis model to perform a parameter optimization; Performing secondary parameter optimization on the welding repair parameters of the primary parameter optimization by using a particle swarm model; According to the abnormal welding repair parameters after the primary parameter optimization, corresponding welding repair steps are screened and eliminated; and the welding repair steps after elimination are optimized by using the particle swarm model; The first welding repair feature, the optimized welding repair step and the welding repair parameters optimized by the secondary parameters are associated to form a welding repair solution network for welding repair work calls.

2. The welding repair method for cracks and wear defects of turbine runner blades according to claim 1 is characterized in that: The welding repair blade information is the image information and video information of the runner blade that needs welding repair.

3. The welding repair method for cracks and wear defects of turbine runner blades according to claim 1, characterized in that: The first welding repair feature of the welding repair blade information extracted according to the welding repair analysis model includes: The welding repair analysis model includes a welding repair information processing unit, a welding repair feature recognition unit, a welding repair feature analysis unit and a first welding repair feature output unit; The welding repair information processing unit performs image preprocessing on the welding repair blade information; The welding repair feature recognition unit identifies welding repair defect features based on the preprocessed welding repair blade information; the welding repair feature analysis unit analyzes global welding repair factors and local welding repair factors according to the welding repair defect features; The first welding repair feature output unit integrates the global welding repair factor and the local welding repair factor to generate the first welding repair feature.

4. The welding repair method for cracks and wear defects of turbine runner blades according to claim 3 is characterized in that: The welding repair feature recognition unit uses yolov5 to identify the welding repair defect feature; the welding repair feature analysis unit generates the global welding repair factor based on the overall pixel ratio of the welding repair defect feature identified by yolov5 on the blade; the welding repair feature analysis unit obtains the defect line of the welding repair defect feature through DeepLSD and calculates the number of pixels of the defect line to generate the local welding repair factor.

5. The welding repair method for cracks and wear defects of turbine runner blades according to claim 1, characterized in that: Clustering is performed through the first welding repair feature, calling the historical welding repair scheme of the same welding repair category according to the clustered first welding repair feature, and obtaining corresponding welding repair parameters and welding repair steps, including: The global welding repair factor is divided into multiple level labels; the level label of the global welding repair factor is used as the horizontal coordinate and the local welding repair factor is used as the vertical coordinate to generate welding repair clustering coordinates, and the multiple welding repair clustering coordinates are obtained by k-means algorithm to obtain multiple welding clustering sets; The plurality of welding cluster sets are marked with welding cluster labels; the welding repair parameters and the welding repair steps corresponding to the historical welding repair scheme are called by the same welding cluster label, and the welding repair parameters and the welding repair steps are marked with the corresponding welding cluster labels.

6. The welding repair method for cracks and wear defects of turbine runner blades according to claim 1, characterized in that: According to the time sequence of the initial welding repair scheme, the welding repair parameters are formed into a welding repair parameter sequence, and the abnormal welding repair parameters of the welding repair parameter sequence are identified according to the welding repair analysis model to perform a parameter optimization, which includes: The same welding repair parameters are used to generate the welding repair parameter sequence according to the time sequence, wherein the welding repair parameter sequence includes a plurality of welding repair parameter sequences; each welding repair parameter sequence is input into the welding repair analysis model; The welding repair analysis model also includes a welding parameter standardization unit, a welding parameter abnormality identification unit and a welding parameter abnormality marking unit; The welding parameter standardization unit divides each welding repair parameter sequence into time windows and performs standardization processing; the welding parameter abnormality identification unit identifies welding parameter abnormalities in the standardized welding repair parameter sequence; and the welding parameter abnormality marking unit marks and outputs the welding repair parameter sequence identified as the welding parameter abnormality.

7. The welding repair method for cracks and wear defects of turbine runner blades according to claim 6, characterized in that: The welding parameter anomaly recognition unit includes 1 transformer layer, 4 time convolution layers, 6 LSTM layers and 1 feature fusion layer.

8. The welding repair method for cracks and wear defects of turbine runner blades according to claim 1, characterized in that: The secondary parameter optimization of the welding repair parameters optimized by the primary parameter optimization through the particle swarm model includes: initializing the speed and position of the particles with the repair time, repair cost and hazard of the welding repair parameters optimized by the primary parameter optimization, taking the minimum welding repair time as the target condition, taking the welding repair cost and the welding repair hazard as the penalty coefficients, optimizing the welding repair parameters after the primary parameter optimization, and generating the optimal welding repair parameters.

9. The welding repair method for cracks and wear defects of turbine runner blades according to claim 1, characterized in that: The optimization of the welding repair step after elimination by using the particle swarm model includes: digitizing the welding repair step, initializing the repair time, repair cost and hazard of the digitized welding repair step as the speed and position of the particle, taking the minimum welding repair time as the target condition, taking the welding repair cost and the welding repair hazard as the penalty coefficient, optimizing the welding repair step, and generating the optimal welding repair step.

10. The welding repair method for cracks and wear defects of turbine runner blades according to claim 1, characterized in that: Associating the first welding repair feature, the optimized welding repair step and the welding repair parameters optimized by the secondary parameters includes: generating an initialized welding association weight using the first welding repair feature, the optimized welding repair step and the welding repair parameters optimized by the secondary parameters, and enhancing or reducing the welding association weight according to the calling result.

Images