A method and system for quality inspection of the middle section structure of a multistage pump

Abstract

The present invention relates to a method and system for quality inspection of the middle section structure of a multistage pump, belonging to the field of quality inspection. During the operation of the middle section structure of the multistage pump, vibration signal monitoring is carried out, and the operation time parameters and water quality parameters of the middle section structure are collected for predicting the fouling of the flow passage; the flow passage image inside the middle section structure is collected by an industrial endoscope, and the flow passage image is respectively input into a plurality of fouling parameter classifiers to identify and obtain the actual fouling parameters; the guide vane deformation is predicted, the guide vane image inside the middle section structure is collected by an industrial endoscope, and the guide vane image is respectively input into a plurality of guide vane parameter classifiers to identify and obtain the actual guide vane deformation parameters; the quality inspection result is obtained by combining the actual fouling and guide vane deformation parameters, solving the technical problems that it is difficult to accurately and efficiently detect the fouling of the flow passage and the deformation of the guide vane in the middle section structure of the multistage pump, and thus it is impossible to control the overall performance of the multistage pump, resulting in difficulties in quality assessment and maintenance.

Description

Technical Field

[0001] The present invention relates to the field of quality inspection, and particularly to a method and system for quality inspection of the middle section structure of a multistage pump. Background Art

[0002] As a key device for fluid transportation, multistage pumps are widely used in various industrial fields. The design and operating state of the middle section structure, which is the core component, have a crucial impact on the overall performance of the multistage pump. The middle section structure mainly includes two parts: the flow passage and the guide vane. Among them, the flow passage is the main channel for fluid in the multistage pump. Taking water as an example, its structural design and material selection directly determine the smoothness and efficiency of the water flow. However, during long-term use, due to the deposition of impurities and chemical reactions in the water flow, scaling often occurs on the inner wall of the flow passage. These scales not only reduce the efficiency of the water flow but also change the flow state of the water flow, resulting in vibrations during the operation of the multistage pump. The vibrations not only affect the stability and lifespan of the multistage pump but may also cause damage to surrounding equipment and building structures. On the other hand, the guide vane is an important component for realizing the energy conversion of the water flow in the multistage pump, and its structural design and deformation state directly affect the hydraulic performance and efficiency of the multistage pump. Under the long-term water flow scouring and stress action, the guide vane often deforms due to changes in the water flow. This deformation not only changes the hydraulic characteristics of the guide vane but may also cause hydraulic imbalance and vibration problems of the multistage pump. Therefore, real-time monitoring and evaluation of the deformation state of the guide vane are of great significance for ensuring the stable operation of the multistage pump. Summary of the Invention

[0003] Aiming at the technical problems in the prior art that it is difficult to accurately and efficiently detect the scaling of the flow passage and the deformation of the guide vane in the middle section structure of the multistage pump, and thus it is impossible to control the overall performance of the multistage pump, resulting in difficulties in quality assessment and maintenance, the present invention provides a method and system for quality inspection of the middle section structure of a multistage pump to solve these problems.

[0004] The technical solution of the present invention to solve the above technical problems is as follows:

[0005] In a first aspect, the present invention provides a method for quality inspection of a multi-stage pump middle section structure, the method comprising: monitoring vibration signals of the middle section structure of a target multi-stage pump during operation to obtain vibration parameters, and collecting the operation time parameters and water quality parameters of the middle section structure to perform prediction of flow channel scaling and obtain predicted scaling parameters; collecting flow channel images inside the middle section structure through an industrial endoscope, and inputting the flow channel images into a plurality of scaling parameter classifiers respectively according to the predicted scaling parameters to identify and obtain actual scaling parameters, wherein each scaling parameter classifier includes a plurality of scaling parameter classification paths; performing prediction of guide vane deformation according to the actual scaling parameters to obtain predicted guide vane deformation parameters, collecting guide vane images inside the middle section structure through an industrial endoscope, and inputting the guide vane images into a plurality of guide vane parameter classifiers respectively according to the predicted guide vane deformation parameters to identify and obtain actual guide vane deformation parameters; combining the actual scaling parameters and the actual guide vane deformation parameters to obtain the quality inspection result of the middle section structure.

[0006] In a second aspect, the present invention provides a quality inspection system for a multi-stage pump middle section structure, the system comprising: a vibration monitoring module for monitoring vibration signals of the middle section structure of a target multi-stage pump during operation to obtain vibration parameters, and collecting the operation time parameters and water quality parameters of the middle section structure to perform prediction of flow channel scaling and obtain predicted scaling parameters; an image acquisition module for collecting flow channel images inside the middle section structure through an industrial endoscope, and inputting the flow channel images into a plurality of scaling parameter classifiers respectively according to the predicted scaling parameters to identify and obtain actual scaling parameters, wherein each scaling parameter classifier includes a plurality of scaling parameter classification paths; a classification and recognition module for performing prediction of guide vane deformation according to the actual scaling parameters to obtain predicted guide vane deformation parameters, collecting guide vane images inside the middle section structure through an industrial endoscope, and inputting the guide vane images into a plurality of guide vane parameter classifiers respectively according to the predicted guide vane deformation parameters to identify and obtain actual guide vane deformation parameters; and a result acquisition module for combining the actual scaling parameters and the actual guide vane deformation parameters to obtain the quality inspection result of the middle section structure.

[0007] The beneficial effects of the present invention are as follows: By monitoring vibration signals, collecting water quality parameters and operation time parameters, and combining industrial endoscope image recognition and deep learning prediction technologies, it is possible to accurately and efficiently detect the flow channel scaling and guide vane deformation conditions of the multi-stage pump middle section structure, thereby realizing a comprehensive evaluation of the quality of the middle section structure and providing strong support for the maintenance and servicing of multi-stage pumps. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 It is a flow schematic diagram of a method for quality inspection of a multi-stage pump middle section structure provided by the present invention.

[0009] Figure 2Structural schematic diagram of a multi-stage pump middle section structure quality detection system provided by the present invention.

[0010] Explanation of reference numerals: vibration monitoring module 11, image acquisition module 12, classification and recognition module 13, result acquisition module 14. Detailed implementation manners

[0011] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present invention.

[0012] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality of" means two or more, unless otherwise specifically defined.

[0013] In the description of the present invention, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in the present invention is not necessarily construed as being more preferred or more advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the present invention. In the following description, details are set forth for purposes of explanation. It should be understood that those of ordinary skill in the art can recognize that the present invention can be practiced without these specific details. In other instances, well-known structures and processes are not described in detail to avoid unnecessary details from obscuring the description of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

[0014] Embodiment 1:

[0015] This scheme is based on the research of the middle section structure of the multi-stage pump. The middle section structure of the multi-stage pump is equipped with a flow channel inside, which is connected to the inlet and outlet of the guide vane impeller. It receives the high-speed water flow thrown out by the previous impeller, first decelerates it, and then arranges the direction of the water flow to stabilize the water flow, and then smoothly transmits it to the next impeller to continue pressurizing. In layman's terms, it can be understood that the middle section is like a courier sorting center, and each level of impeller is like a courier. Each level of impeller accelerates the water flow. The middle section flow channel receives the high-speed water flow thrown out by the previous impeller, but the accelerated water will fly around under the action of the impeller. Therefore, the high-speed water flow needs to be "calmed down", that is, decelerated, so that the water flow speed is reduced, and then subsequent pressurization is carried out, that is, the acceleration of the next level of impeller. In this process, the direction of the water flow needs to be arranged to stabilize the water flow state, and the direction of the water flow is aligned with the inlet of the next level of impeller to avoid energy loss caused by inconsistent directions during water flow transmission and pressurization. Then the water flow is smoothly transmitted to the next level of impeller so that the next level of impeller can continue to pressurize the water flow. Based on the above working principles and actual impacts, the quality inspection of this plan is implemented to achieve a comprehensive assessment of the quality of the mid-section structure and provide strong support for the maintenance and servicing of multi-stage pumps.

[0016] like Figure 1 As shown, an embodiment of the present invention provides a method for detecting the quality of a multi-stage pump mid-section structure, the method comprising:

[0017] S10: monitoring vibration signals of the middle section structure of the target multistage pump during operation to obtain vibration parameters, collecting operation time parameters and water quality parameters of the middle section structure, predicting flow channel scaling, and obtaining predicted scaling parameters.

[0018] For example, vibration sensors are arranged to continuously monitor the vibration signals of a multi-stage pump during operation. These sensors can capture tiny vibration changes in the pump body and convert these vibration signals into analyzable vibration parameters, such as vibration frequency, amplitude, etc. These parameters can indirectly reflect the operating status of the internal flow path and component wear of the pump, and are important indicators for evaluating the operating status of the pump.

[0019] At the same time, water quality monitoring instruments are used to regularly collect various parameters of the water quality in the middle section of the multi-stage pump, such as water hardness, pH value, oxygen content, and impurity content. These water quality parameters are important bases for evaluating scaling tendency, and different water quality conditions will directly affect the speed and type of scaling.

[0020] In addition, the running time parameters of the mid-section structure are collected, that is, the cumulative running time of the multi-stage pump is recorded, which helps to understand the use history and potential aging problems of the pump, and provides time parameters for scale deposition.

[0021] After obtaining the vibration parameters, water quality parameters, and operation time parameters, a prediction model based on deep learning is used for predicting the fouling of the flow channel. This model is trained and optimized through a large amount of historical maintenance data of multi-stage pumps of the same type, and can accurately identify the complex relationships between various parameters and the fouling dimensions (such as fouling area and thickness). By inputting the currently monitored parameters into the model, the predicted fouling parameters can be obtained, so as to understand the fouling situation of the flow channel in advance and provide a scientific basis for subsequent maintenance decisions. The specific fouling parameters refer to the fouling dimensions, that is, the fouling area and thickness. In addition, in other possible cases, it is also possible to predict the fouling type, fouling distribution, fouling density, and fouling growth rate, which are not involved here for the time being. For example, if the prediction results show that the fouling area and thickness of a certain section of the flow channel have reached the level that affects the pump performance, then the cleaning work can be arranged in time to avoid the adverse impact of fouling on the pump performance and ensure the stable operation of the multi-stage pump.

[0022] S20: Through an industrial endoscope, collect the flow channel images inside the middle section structure, and input the flow channel images into multiple fouling parameter classifiers respectively according to the predicted fouling parameters, and identify and obtain the actual fouling parameters. Among them, each fouling parameter classifier includes multiple fouling parameter classification paths.

[0023] Optionally, use an industrial endoscope device to penetrate into the interior of the middle section structure of the multi-stage pump to collect clear images of the flow channel. These images intuitively show the fouling situation on the inner wall of the flow channel and provide information for subsequent analysis. The industrial endoscope device is a high-precision instrument for industrial inspection. It clearly presents the internal structure, defects, corrosion, fouling, etc. of the device through optical or electronic imaging technology, helps to accurately judge the internal condition of the device, timely discover potential problems, and provides strong support for the maintenance and overhaul of the device.

[0024] After obtaining the flow channel images through the industrial endoscope, select the fouling parameter classifiers corresponding to the fouling parameter levels close to the predicted fouling parameters obtained previously through vibration signals, water quality, and operation time parameters. These classifiers are obtained by training a large amount of sample data in advance and can accurately identify the characteristics of different fouling parameter levels. Among them, each fouling parameter classifier contains multiple fouling parameter classification paths inside. These paths are constructed based on deep learning algorithms and can perform integrated binary classification on the input flow channel images. That is to say, they will judge whether the fouling situation in the image belongs to the fouling parameter level corresponding to this classifier. Through this method, the fouling parameter classification results output by multiple classifiers are obtained.

[0025] To obtain the most accurate actual scaling parameters, calculate the probability of each scaling parameter level appearing in multiple classification results. Specifically, it is to count the proportion of classification results judged as "yes", so as to obtain the probability distribution of different scaling parameter levels. Finally, select the scaling parameter level with the highest probability as the actual scaling parameter.

[0026] In summary, this method not only improves the detection efficiency, but also greatly improves the accuracy. Through the integrated binary classification of multiple classifiers, the characteristics of the flow channel scaling can be captured more comprehensively, avoiding the misjudgment situation that may exist in a single classifier. At the same time, since the classification model has been fully trained with a large amount of sample data, its convergence efficiency is high, and it can quickly give accurate detection results in practical applications. Furthermore, it can more efficiently monitor and maintain the scaling situation of the flow channel in the middle section structure of the multistage pump to ensure the stable operation of the multistage pump.

[0027] S30: According to the actual scaling parameters, predict the guide vane deformation, obtain the predicted guide vane deformation parameters. Through an industrial endoscope, collect the guide vane images inside the middle section structure. According to the predicted guide vane deformation parameters, input the guide vane images into multiple guide vane parameter classifiers respectively to identify and obtain the actual guide vane deformation parameters.

[0028] S40: Combine the actual scaling parameters and the actual guide vane deformation parameters to obtain the quality inspection result of the middle section structure.

[0029] Specifically, predict the deformation of the guide vane based on the actual scaling parameters. Since scaling will change the flow channel structure, thereby affecting the water flow state, and then causing the deformation of the guide vane. Therefore, based on the correlation between the structural parameters and the guide vane deformation, the predicted guide vane deformation parameters are obtained through model prediction. Next, use an industrial endoscope to collect clear images of the guide vanes inside the middle section structure, and these images visually show the actual shape of the guide vanes.

[0030] In order to accurately identify the deformation parameters of the guide vane, select the guide vane parameter classifier corresponding to the guide vane deformation parameter level closest to the predicted guide vane deformation parameter. Each guide vane parameter classifier contains multiple classification paths based on deep learning algorithms, which can perform integrated binary classification on the input guide vane images to judge whether they belong to the guide vane deformation parameter level corresponding to the classifier. Calculate the probability distribution of each guide vane deformation parameter level through the comprehensive judgment of multiple classifiers, and select the level with the highest probability as the actual guide vane deformation parameter. This step is similar to the acquisition of the actual scaling parameters.

[0031] Finally, combine the actual scaling parameters and the actual guide vane deformation parameters to comprehensively evaluate the quality status of the middle section structure. For example, if the actual scaling parameters indicate severe scaling in the flow channel and the actual guide vane deformation parameters show significant deformation of the guide vanes, then it can be judged that there are problems with the quality of the middle section structure and it needs to be cleaned and maintained in a timely manner. This process not only improves the detection efficiency but also ensures the accuracy of the detection results, providing a strong guarantee for the stable operation of the multistage pump.

[0032] In a preferred embodiment, monitor the vibration signal of the middle section structure of the target multistage pump during operation to obtain vibration parameters, and collect the operation time parameters and water quality parameters of the middle section structure to perform flow channel scaling prediction and obtain predicted scaling parameters, including: monitor the vibration signal of the middle section structure of the target multistage pump during operation to obtain vibration parameters; during the operation of the target multistage pump, regularly monitor the water quality parameters to obtain a set of water quality parameters and calculate to obtain the water quality parameters; collect the cumulative operation time of the middle section structure to obtain the operation time parameters; according to the vibration parameters, water quality parameters and operation time parameters, perform flow channel scaling prediction to obtain predicted scaling parameters.

[0033] Preferably, use a high-precision vibration sensor to monitor the vibration signal of the multistage pump in real time during operation, capture the minute vibrations generated by the pump body due to changes in the internal flow state or component wear, and thus obtain vibration parameters, which can reflect the operation stability of the pump and the health status of the internal structure.

[0034] At the same time, regularly collect water quality parameters during the operation of the target multistage pump, including water hardness, pH value, oxygen content, impurity content, etc., to form a set of water quality parameters. By comprehensively analyzing these water quality parameters, the influence of water quality on the scaling tendency of the internal flow channel of the pump can be understood. For example, water with high hardness is more likely to form mineral deposits on the inner wall of the flow channel, thus accelerating the scaling process.

[0035] In addition, record the cumulative operation time of the multistage pump as an important indicator for evaluating the aging degree and operation history of the pump. The longer the operation time, the higher the wear and scaling risks of the internal components of the pump usually are.

[0036] Using a prediction model, vibration parameters, water quality parameters, and operating time parameters are used as inputs to predict the fouling of the flow channel. Through the training of a large amount of historical data, the prediction model can accurately identify the complex relationships between various parameters and the fouling situation, and thus output the predicted fouling parameters. These predicted fouling parameters not only include the possibility and degree of fouling, but may also include information such as the type and distribution of fouling, providing a scientific basis for subsequent maintenance and cleaning work. However, in this solution, the fouling parameter is specifically the fouling size, that is, the fouling area and thickness. If the prediction result shows that a certain section of the flow channel may be severely fouled in the near future, a cleaning plan can be arranged in advance to avoid the adverse impact of fouling on the pump performance.

[0037] In a preferred embodiment, according to the vibration parameters, water quality parameters, and operating time parameters, the fouling of the flow channel is predicted to obtain the predicted fouling parameters, including: collecting a set of sample vibration parameters, a set of sample water quality parameters, and a set of sample operating time parameters according to the historical maintenance data of the middle section structure of the same type of multi-stage pump, and collecting the fouling size of the flow channel fouling, and obtaining a set of sample fouling parameters by labeling; constructing a flow channel fouling predictor based on deep learning; using the set of sample vibration parameters, the set of sample water quality parameters, the set of sample operating time parameters, and the set of sample fouling parameters as supervised training data and test data to perform supervised training and testing on the flow channel fouling predictor, and completing the training and testing after passing the accuracy test; inputting the vibration parameters, water quality parameters, and operating time parameters into the flow channel fouling predictor, and predicting and outputting to obtain the predicted fouling parameters.

[0038] Specifically, historical maintenance data of the middle section structure of the same type of multi-stage pump is collected, which includes a set of sample vibration parameters, a set of sample water quality parameters, and a set of sample operating time parameters. At the same time, the actual fouling size of the flow channel fouling of these pumps during maintenance is collected, and a set of sample fouling parameters is obtained by labeling based on this. These sample data form the basis for the training of the prediction model.

[0039] Subsequently, a flow channel fouling predictor is constructed using deep learning technology. This predictor can learn and identify the complex relationships between vibration parameters, water quality parameters, operating time parameters, and fouling parameters. To train this predictor, the collected sample data is divided into supervised training data and test data, and the predictor is subjected to supervised training and testing. During the training process, the predictor continuously adjusts its internal parameters to minimize the error between the prediction result and the actual fouling parameters. After multiple iterative trainings, when the accuracy of the predictor on the test data reaches a satisfactory level, the training and testing process is completed.

[0040] In actual operation, the vibration parameters, water quality parameters, and operation time parameters real-time monitored during the operation of the target multistage pump are input into the trained runner fouling predictor. The predictor quickly calculates the predicted fouling parameters based on these input parameters, including the possible size information of the fouling. For example, if the prediction result shows that a certain section of the runner may have large-scale fouling in the future, a maintenance plan can be arranged in advance to clean or repair the pump to avoid the adverse impact of fouling on the pump performance. This method not only improves the prediction accuracy but also greatly enhances the efficiency and reliability of the multistage pump maintenance.

[0041] In a preferred embodiment, through an industrial endoscope, the runner images inside the middle section structure are collected. According to the predicted fouling parameters, the runner images are respectively input into multiple fouling parameter classifiers to identify the actual fouling parameters, including: through the industrial endoscope, collecting the runner images inside the middle section structure; according to the predicted fouling parameters, screening to obtain N fouling parameter levels closest to the predicted fouling parameters, where N is a positive integer; calculating to obtain N classification path numbers according to the error margins between the N fouling parameter levels and the predicted fouling parameters, selecting the N fouling parameter classifiers corresponding to the N fouling parameter levels, inputting the runner images into the N fouling parameter classifiers respectively for the fouling parameter classification paths with N classification path numbers, and outputting to obtain N fouling parameter classification result sets, where each fouling parameter classifier includes multiple fouling parameter classification paths, and each fouling parameter classification result includes yes or no; calculating the proportion of the fouling parameter classification results that are yes according to the N fouling parameter classification result sets to obtain N fouling parameter level probabilities; outputting the fouling parameter level with the largest fouling parameter level probability to obtain the actual fouling parameters.

[0042] Specifically, an industrial endoscope is used to penetrate into the interior of the middle section structure of the multistage pump to collect clear images of the runner. These images visually show the fouling condition of the runner inner wall. Since there may be certain errors in fouling prediction, multiple fouling parameter levels are preset, and the corresponding fouling parameter classifiers are respectively constructed. Among them, each classifier contains multiple fouling parameter classification paths inside, and these paths are constructed based on deep learning algorithms and can perform binary classification judgments on the input runner images.

[0043] After obtaining the predicted scaling parameters, screen out the N scaling parameter levels (e.g., N is set to 5) that are closest to the predicted scaling parameters as the candidate actual scaling parameter levels. Then, calculate a classification path number based on the error margin between each candidate scaling parameter level and the predicted scaling parameter. Now, an example is given to illustrate this calculation process. The error margin between the scaling parameter level and the predicted scaling parameter is the absolute difference divided by the predicted scaling parameter. If the error margin is 10%, then 1 - 10% = 90% is used, and then 90% is multiplied by the number of scaling parameter classification paths in the corresponding scaling parameter classifier (e.g., it is 10), so the classification path number is 9. Subsequently, input the flow channel images into these N scaling parameter classifiers respectively, and perform binary classification on the randomly selected scaling parameter classification paths according to the calculated classification path number. Each classification path will output a scaling parameter classification result, that is, to judge whether the scaling parameter in the flow channel image is the scaling parameter level corresponding to this classifier. In this way, a set of N scaling parameter classification results is obtained.

[0044] To determine the most likely actual scaling parameter level, calculate the proportion of each scaling parameter level that is judged as "yes" in the N classification results, that is, the scaling parameter level probability. For example, if a certain scaling parameter level is judged as "yes" 5 times in 9 classification paths, then its scaling parameter level probability is 5 / 9. Finally, output the scaling parameter level with the largest scaling parameter level probability as the actual scaling parameter. This method not only improves the detection and recognition efficiency but also reduces the waste of computing power by selecting the classifiers corresponding to the close scaling parameter levels, calculating the configured classification path number according to the error margin for binary classification, and selecting the most likely scaling parameter based on probability.

[0045] For example, assume that the predicted scaling parameter of a certain section of the flow channel is medium, but there may be a certain error in the prediction. Then select 5 scaling parameter classifiers corresponding to the 5 scaling parameter levels (such as slight, mild, moderate, severe, very severe) that are closest to the medium level. Determine the number of classification paths to be used in each classifier by calculating the error margin, and input the flow channel images into these classification paths for binary classification. Finally, calculate the probability of each scaling parameter level based on the classification results, and select the scaling parameter level with the largest probability as the actual scaling parameter. This can more accurately understand the scaling situation of the flow channel and provide strong support for subsequent maintenance and cleaning work.

[0046] In a preferred embodiment, according to the error margins between the N fouling parameter levels and the predicted fouling parameter, N numbers of classification paths are calculated. The N fouling parameter classifiers corresponding to the N fouling parameter levels are selected. The flow channel images are respectively input into the fouling parameter classification paths with the N numbers of classification paths in the N fouling parameter classifiers, and N sets of fouling parameter classification results are output, including: calculating the error margins between the N fouling parameter levels and the predicted fouling parameter respectively, multiplying the results of subtracting each of the N error margins from 1 by the number of fouling parameter classification paths in each fouling parameter classifier and taking the integer to obtain N numbers of classification paths; selecting the N fouling parameter classifiers corresponding to the N fouling parameter levels, wherein multiple fouling parameter classifiers corresponding to multiple fouling parameter levels are pre-trained, each fouling parameter classifier includes multiple fouling parameter classification paths constructed and trained based on a convolutional neural network, the training data includes sample flow channel images and sample fouling parameter classification results, and each sample fouling parameter classification result includes a binary classification result indicating whether the fouling parameter in the sample flow channel image is the fouling parameter level corresponding to the fouling parameter classifier; according to the N numbers of classification paths, randomly select the fouling parameter classification paths with the N numbers of classification paths in the N fouling parameter classifiers respectively, input the flow channel images, and output N sets of fouling parameter classification results.

[0047] Further, for the predicted fouling parameter, the N fouling parameter levels closest to it are screened out, and the error margins between these N fouling parameter levels and the predicted fouling parameter are calculated respectively. Specifically, the results are multiplied by the preset number of fouling parameter classification paths in each fouling parameter classifier by the way of subtracting each error margin from 1, and the integer is taken to obtain N numbers of classification paths. These fouling parameter classifiers are pre-trained with a large amount of sample data. Each classifier is constructed based on a convolutional neural network and contains multiple fouling parameter classification paths. The training data includes sample flow channel images and corresponding sample fouling parameter classification results, where each sample fouling parameter classification result clearly labels whether the fouling parameter in the sample flow channel image is the fouling parameter level corresponding to the classifier, that is, binary classification is performed.

[0048] After obtaining the number of N classification paths, for each fouling parameter level, randomly select the corresponding number of fouling parameter classification paths in its corresponding fouling parameter classifier. Then, input the collected runner images into these selected classification paths, and each classification path will output a fouling parameter classification result based on the image features, that is, to determine whether the fouling parameter in the runner image is the fouling parameter level corresponding to this classifier. Finally, summarize the output results of all classification paths to form N sets of fouling parameter classification results. These result sets provide an important basis for determining the actual fouling parameter subsequently. For example, if a certain fouling parameter level is judged as "yes" in multiple classification paths, then this fouling parameter level may be the actual fouling parameter. Through this method, the fouling situation in the runner can be identified more accurately, providing strong support for the maintenance and cleaning work of the multistage pump.

[0049] In a preferred embodiment, according to the actual fouling parameter, conduct a guide vane deformation prediction to obtain a predicted guide vane deformation parameter, including: collecting a sample fouling parameter set based on the historical maintenance data of the middle section structure of the same type of multistage pump, and collecting the deformation angles of the guide vanes under different sample fouling parameters, which are marked as guide vane deformation parameters to obtain a sample guide vane deformation parameter set; based on deep learning, construct a guide vane deformation predictor; use the sample fouling parameter set and the sample guide vane deformation parameter set as supervised training data and test data to conduct supervised training and testing on the guide vane deformation predictor, and complete the training and testing after the accuracy test is qualified; input the actual fouling parameter into the guide vane deformation predictor, and predict and output to obtain the predicted guide vane deformation parameter.

[0050] Exemplarily, historical maintenance data of the middle section structure of the same type of multistage pump is collected, and these data contain a sample fouling parameter set, that is, the fouling situations of different pumps under different operating conditions. At the same time, collect the deformation angles of the guide vanes of these pumps under the corresponding fouling parameters, and these angles are marked as guide vane deformation parameters, thus forming a sample guide vane deformation parameter set. These sample data provide a basis for constructing a prediction model.

[0051] Subsequently, use deep learning technology to construct a guide vane deformation predictor, which can learn and identify the complex relationship between the fouling parameter and the guide vane deformation parameter. To train this predictor, use the collected sample fouling parameter set and sample guide vane deformation parameter set as supervised training data and test data. During the training process, the predictor continuously adjusts its internal parameters to minimize the error between the prediction result and the true guide vane deformation parameter. After multiple iterative trainings, when the accuracy of the predictor on the test data reaches a satisfactory level, the training and testing process is completed.

[0052] After that, the actually detected scaling parameters are input into the trained guide vane deformation predictor. The predictor quickly calculates the predicted guide vane deformation parameters based on these input parameters, which usually include information such as the deformation angle and deformation direction of the guide vane. For example, if the prediction result shows that the guide vane of a certain pump will undergo a large-angle deformation under specific scaling parameters, a maintenance plan can be arranged in advance to make necessary adjustments or repairs to the pump to avoid the adverse effects of guide vane deformation on the pump performance. This method not only improves the prediction accuracy but also greatly enhances the efficiency and reliability of the maintenance of multi-stage pumps.

[0053] In a preferred embodiment, through an industrial endoscope, guide vane images inside the middle section structure are collected. According to the predicted guide vane deformation parameters, the guide vane images are respectively input into multiple guide vane parameter classifiers to identify and obtain the actual guide vane deformation parameters, including: collecting guide vane images inside the middle section structure through an industrial endoscope; selecting M guide vane deformation parameter levels that are closest to the predicted guide vane deformation parameters, where M is a positive integer; selecting the M guide vane parameter classifiers corresponding to the M guide vane deformation parameter levels, where multiple guide vane parameter classifiers corresponding to multiple guide vane deformation parameter levels are pre-trained, and each guide vane parameter classifier includes multiple guide vane parameter classification paths constructed and trained based on a convolutional neural network. The training data includes sample guide vane images and sample guide vane parameter classification results, and each sample guide vane parameter classification result includes a binary classification result on whether the guide vane deformation parameter in the sample guide vane image is the guide vane deformation parameter level corresponding to the guide vane parameter classifier; inputting the guide vane images into multiple guide vane parameter classification paths in the M guide vane parameter classifiers to obtain M sets of guide vane parameter classification results, where each guide vane parameter classification result includes yes or no; respectively calculating the proportion of the guide vane parameter classification results that are yes in the M sets of guide vane parameter classification results to obtain M guide vane deformation parameter level probabilities; outputting the guide vane deformation parameter level with the largest guide vane deformation parameter level probability as the actual guide vane deformation parameter.

[0054] Optionally, based on the same principle as the above scaling parameter classifier, an industrial endoscope is used to penetrate deep into the middle section structure of the multi-stage pump to collect clear images of the guide vane, and these images intuitively show the actual shape of the guide vane.

[0055] Similarly, due to certain possible errors in predicting the guide vane deformation parameters, in order to more accurately identify the actual guide vane deformation parameters, select the M guide vane deformation parameter levels (M is a positive integer and can be the same as N in the fouling parameter classification) that are closest to the predicted guide vane deformation parameters as candidates. Then, according to the error magnitude between each candidate guide vane deformation parameter level and the predicted guide vane deformation parameter, calculate a classification path number. This calculation process is similar to the fouling parameter classification: for example, the error magnitude between the guide vane deformation parameter level and the predicted guide vane deformation parameter is the absolute difference divided by the predicted guide vane deformation parameter. If the error magnitude is a certain percentage (such as 10%), then use 1 minus this percentage (such as 1 - 10% = 90%), and then multiply the result by the number of guide vane parameter classification paths in the corresponding guide vane parameter classifier (this number is preset, for example, 10), so as to obtain the classification path number.

[0056] Subsequently, select the M guide vane parameter classifiers corresponding to the M guide vane deformation parameter levels. These classifiers are pre-trained with a large amount of sample data. Each classifier is constructed based on a convolutional neural network and contains multiple guide vane parameter classification paths. The training data includes sample guide vane images and corresponding sample guide vane parameter classification results, where each sample guide vane parameter classification result clearly labels whether the guide vane deformation parameter in the sample guide vane image is the guide vane deformation parameter level corresponding to this classifier, that is, perform binary classification. Then, input the collected guide vane images into multiple guide vane parameter classification paths in the M guide vane parameter classifiers respectively. Each classification path will output a guide vane parameter classification result according to the image features, that is, judge whether the guide vane deformation parameter in the guide vane image is the guide vane deformation parameter level corresponding to this classifier. In this way, a set of M guide vane parameter classification results is obtained.

[0057] In order to determine the most likely actual guide vane deformation parameter level, calculate the proportion of the guide vane parameter classification results that are "yes" in the set of M guide vane parameter classification results respectively, that is, the guide vane deformation parameter level probability. For example, if the proportion of a certain guide vane deformation parameter level being judged as "yes" is the highest in multiple classification paths, then its corresponding guide vane deformation parameter level probability is the largest. Finally, output the guide vane deformation parameter level with the largest guide vane deformation parameter level probability as the actual guide vane deformation parameter. This method not only improves the detection and recognition efficiency but also reduces the waste of computing power by selecting the classifiers corresponding to the close guide vane deformation parameter levels, calculating the configured classification path number according to the error magnitude for binary classification, and selecting the most likely guide vane deformation parameter based on probability.

[0058] For example, assume that the predicted deformation parameter of a certain pump guide vane is medium - degree bending, but there may be a certain error in the prediction. Then, three guide vane parameter classifiers corresponding to the three guide vane deformation parameter levels closest to medium - degree bending (such as slight bending, medium bending, and severe bending) are selected. The number of classification paths to be used within each classifier is determined by calculating the error margin, and the guide vane image is input into these classification paths for binary classification. Finally, the probability of each guide vane deformation parameter level is calculated based on the classification results, and the guide vane deformation parameter level with the highest probability is selected as the actual guide vane deformation parameter. In this way, the deformation situation of the guide vane can be understood more accurately, providing strong support for subsequent maintenance and adjustment work.

[0059] In a preferred embodiment, by combining the actual scaling parameter and the actual guide vane deformation parameter, a quality inspection result of the middle section structure is obtained, including: integrating the actual scaling parameter and the actual guide vane deformation parameter as the quality inspection result of the middle section structure of the target multistage pump.

[0060] Specifically, integrating the actual scaling parameter and the actual guide vane deformation parameter to form the quality inspection result of the middle section structure of the target multistage pump. This result not only reflects the scaling situation of the flow channel but also reveals the deformation degree of the guide vane, providing comprehensive and accurate data support for the maintenance and servicing of the multistage pump. For example, if the actual scaling parameter shows severe scaling in the flow channel and the actual guide vane deformation parameter indicates significant deformation of the guide vane, then it can be judged that there is a problem with the quality of the middle section structure of the multistage pump, and it is necessary to clean and maintain it in a timely manner to avoid adverse effects on the performance and lifespan of the pump.

[0061] A method for quality inspection of the middle section structure of a multistage pump provided by an embodiment of the present invention has at least the following technical effects:

[0062] 1. By fusing multi - dimensional data such as vibration parameters, water quality parameters, and operation time parameters, a deep - learning - based flow channel scaling predictor is constructed, realizing accurate prediction of the flow channel scaling situation. The fusion of such multi - dimensional parameters not only improves the accuracy of prediction but also makes the prediction results more comprehensive and reliable. At the same time, based on the predicted scaling parameter, further prediction of the guide vane deformation is carried out, forming a complete prediction chain from scaling to guide vane deformation, providing a more in - depth and detailed analysis for the quality inspection of the middle section structure of the multistage pump.

[0063] 2. When identifying the actual scaling parameter and the actual guide vane deformation parameter, an intelligent classification and recognition technology is adopted. By screening multiple parameter levels closest to the predicted parameters and calculating the corresponding number of classification paths, the image is input into multiple classifiers for binary classification judgment. Finally, the most likely parameter level is determined through probability analysis. This method combining intelligent classification and recognition with probability analysis effectively improves the accuracy and efficiency of recognition, reducing the possibility of misjudgment and missed judgment.

[0064] 3. By combining industrial endoscope technology with deep learning, the intuitive observation and accurate analysis of the internal conditions of the middle section structure of a multistage pump are realized. Clear images of the flow channel and guide vanes are collected by the endoscope, providing rich training data and test data for the deep learning model. At the same time, the processing results of the deep learning model provide strong support for guiding the further observation and analysis of the endoscope. This organic combination makes the quality inspection of the middle section structure of the multistage pump more intuitive, accurate, and efficient.

[0065] Embodiment 2:

[0066] As Figure 2 shown, based on the same inventive concept as the method for quality inspection of the middle section structure of a multistage pump provided in Embodiment 1, the embodiment of the present invention further provides a quality inspection system for the middle section structure of a multistage pump, and the system includes:

[0067] A vibration monitoring module 11, configured to monitor vibration signals of the middle section structure of the target multistage pump during operation, obtain vibration parameters, collect the operation time parameter and water quality parameter of the middle section structure, perform flow channel fouling prediction, and obtain predicted fouling parameters.

[0068] An image acquisition module 12, configured to collect flow channel images inside the middle section structure through an industrial endoscope, and input the flow channel images into a plurality of fouling parameter classifiers respectively according to the predicted fouling parameters to identify and obtain actual fouling parameters, wherein each fouling parameter classifier includes a plurality of fouling parameter classification paths.

[0069] A classification and recognition module 13, configured to perform guide vane deformation prediction according to the actual fouling parameters to obtain predicted guide vane deformation parameters, collect guide vane images inside the middle section structure through an industrial endoscope, and input the guide vane images into a plurality of guide vane parameter classifiers respectively according to the predicted guide vane deformation parameters to identify and obtain actual guide vane deformation parameters.

[0070] A result acquisition module 14, configured to combine the actual fouling parameters and actual guide vane deformation parameters to obtain the quality inspection result of the middle section structure.

[0071] Furthermore, the vibration monitoring module 11 is further configured to perform the following steps:

[0072] Monitor vibration signals of the middle section structure of the target multistage pump during operation to obtain vibration parameters; during the operation of the target multistage pump, regularly monitor water quality parameters to obtain a water quality parameter set, calculate to obtain water quality parameters; collect the cumulative operation time of the middle section structure to obtain an operation time parameter; perform flow channel fouling prediction according to the vibration parameters, water quality parameters, and operation time parameters to obtain predicted fouling parameters.

[0073] Furthermore, the vibration monitoring module 11 is further configured to perform the following steps:

[0074] According to the historical maintenance data of the middle section structure of multi-stage pumps of the same model, collect a set of sample vibration parameters, a set of sample water quality parameters, a set of sample operation time parameters, and collect the scaling size of the flow channel scaling, and label to obtain a set of sample scaling parameters; Based on deep learning, construct a flow channel scaling predictor; Use the set of sample vibration parameters, the set of sample water quality parameters, the set of sample operation time parameters and the set of sample scaling parameters as supervised training data and test data to perform supervised training and testing on the flow channel scaling predictor, and complete the training and testing after the accuracy test is qualified; Input the vibration parameters, water quality parameters and operation time parameters into the flow channel scaling predictor, and predict and output to obtain predicted scaling parameters.

[0075] Furthermore, the image acquisition module 12 is further configured to perform the following steps:

[0076] Collect the flow channel image inside the middle section structure through an industrial endoscope; According to the predicted scaling parameters, screen and obtain N scaling parameter levels that are closest to the predicted scaling parameters, where N is a positive integer; According to the error margins between the N scaling parameter levels and the predicted scaling parameters, calculate and obtain the number of N classification paths, select the N scaling parameter classifiers corresponding to the N scaling parameter levels, and input the flow channel images into the N classification paths of the N scaling parameter classifiers for scaling parameter classification respectively, and output to obtain a set of N scaling parameter classification results, where each scaling parameter classifier includes multiple scaling parameter classification paths, and each scaling parameter classification result includes yes or no; According to the set of N scaling parameter classification results, calculate the proportion of the scaling parameter classification results that are yes to obtain the probabilities of the N scaling parameter levels; Output the scaling parameter level with the largest scaling parameter level probability to obtain the actual scaling parameters.

[0077] Furthermore, the image acquisition module 12 is further configured to perform the following steps:

[0078] Calculate the error margins between N fouling parameter levels and the predicted fouling parameter respectively, subtract each of the N error margins from 1, multiply the result by the number of fouling parameter classification paths within each fouling parameter classifier, and round up to obtain N numbers of classification paths; select the N fouling parameter classifiers corresponding to the N fouling parameter levels, where multiple fouling parameter classifiers corresponding to multiple fouling parameter levels are pre-trained, each fouling parameter classifier includes multiple fouling parameter classification paths constructed and trained based on a convolutional neural network, the training data includes sample runner images and sample fouling parameter classification results, and each sample fouling parameter classification result includes a binary classification result indicating whether the fouling parameter in the sample runner image is the fouling parameter level corresponding to the fouling parameter classifier; according to the N numbers of classification paths, randomly select N numbers of fouling parameter classification paths in the N fouling parameter classifiers respectively, input the runner image, and output to obtain N sets of fouling parameter classification results.

[0079] Furthermore, the classification and recognition module 13 is further configured to perform the following steps:

[0080] Collect a set of sample fouling parameters according to the historical maintenance data of the middle section structure of multi-stage pumps of the same model, and collect the deformation angles of the guide vanes under different sample fouling parameters, which are marked as guide vane deformation parameters to obtain a set of sample guide vane deformation parameters; construct a guide vane deformation predictor based on deep learning; use the set of sample fouling parameters and the set of sample guide vane deformation parameters as supervised training data and test data to perform supervised training and testing on the guide vane deformation predictor, and complete the training and testing after passing the accuracy test; input the actual fouling parameter into the guide vane deformation predictor, and predict and output to obtain the predicted guide vane deformation parameter.

[0081] Furthermore, the classification and recognition module 13 is further configured to perform the following steps:

[0082] Collect the guide vane images inside the middle section structure through an industrial endoscope; select M guide vane deformation parameter levels that are closest to the predicted guide vane deformation parameters, where M is a positive integer; select M guide vane parameter classifiers corresponding to the M guide vane deformation parameter levels. Among them, multiple guide vane parameter classifiers corresponding to multiple guide vane deformation parameter levels are pre-trained. Each guide vane parameter classifier includes multiple guide vane parameter classification paths constructed and trained based on a convolutional neural network. The training data includes sample guide vane images and sample guide vane parameter classification results. Each sample guide vane parameter classification result includes a binary classification result indicating whether the guide vane deformation parameter in the sample guide vane image is the guide vane deformation parameter level corresponding to the guide vane parameter classifier; input the guide vane images into the multiple guide vane parameter classification paths in the M guide vane parameter classifiers respectively to obtain M sets of guide vane parameter classification results. Among them, each guide vane parameter classification result includes "yes" or "no"; calculate the proportions of the guide vane parameter classification results that are "yes" in the M sets of guide vane parameter classification results respectively to obtain M guide vane deformation parameter level probabilities; output the guide vane deformation parameter level with the largest guide vane deformation parameter level probability as the actual guide vane deformation parameter.

[0083] Furthermore, the result acquisition module 14 is further configured to perform the following steps:

[0084] Integrate the actual fouling parameters and the actual guide vane deformation parameters as the quality inspection result of the middle section structure of the target multistage pump.

[0085] Through the foregoing detailed description of a method for inspecting the quality of the middle section structure of a multistage pump in this specification, those skilled in the art can clearly know a system for inspecting the quality of the middle section structure of a multistage pump in this embodiment. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple. For related parts, refer to the description in the method section.

[0086] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for detecting the quality of a multistage pump mid-section structure, characterized in that: The method comprises: Monitor the vibration signal of the middle section structure of the target multi-stage pump during operation to obtain vibration parameters, collect the operation time parameters and water quality parameters of the middle section structure, predict the flow channel scaling, and obtain the predicted scaling parameters; Using an industrial endoscope, collecting a flow channel image in the middle section structure; According to the predicted scaling parameter, N scaling parameter levels closest to the predicted scaling parameter are screened and obtained, wherein N is a positive integer; According to the error amplitudes between N scaling parameter levels and the predicted scaling parameters, the number of N classification paths is calculated, N scaling parameter classifiers corresponding to the N scaling parameter levels are selected, the flow channel images are respectively input into the scaling parameter classification paths of the N classification paths in the N scaling parameter classifiers, and N scaling parameter classification result sets are obtained as output, wherein each scaling parameter classifier includes a plurality of scaling parameter classification paths, and each scaling parameter classification result includes yes or no; According to the N scaling parameter classification result sets, the proportion of scaling parameter classification results that are yes is calculated to obtain N scaling parameter level probabilities; Output the scaling parameter level with the highest scaling parameter level probability to obtain the actual scaling parameter; According to the actual scaling parameters, guide vane deformation prediction is performed to obtain predicted guide vane deformation parameters; Using an industrial endoscope, collecting an image of the guide vanes in the middle section structure; Selecting M guide vane deformation parameter levels closest to the predicted guide vane deformation parameter, where M is a positive integer; Selecting M guide vane parameter classifiers corresponding to the M guide vane deformation parameter levels, wherein a plurality of guide vane parameter classifiers corresponding to a plurality of guide vane deformation parameter levels are pre-trained, each guide vane parameter classifier includes a plurality of guide vane parameter classification paths constructed and trained based on a convolutional neural network, the training data includes a sample guide vane image and a sample guide vane parameter classification result, and each sample guide vane parameter classification result includes a binary classification result of whether a guide vane deformation parameter in the sample guide vane image is a guide vane deformation parameter level corresponding to the guide vane parameter classifier; Inputting the guide vane images into a plurality of guide vane parameter classification paths in the M guide vane parameter classifiers respectively to obtain M guide vane parameter classification result sets, wherein each guide vane parameter classification result includes yes or no; Calculate the proportion of the guide vane parameter classification results that are "yes" in the M guide vane parameter classification result sets respectively, and obtain the M guide vane deformation parameter level probabilities; Output the guide vane deformation parameter level with the highest probability as the actual guide vane deformation parameter; The quality inspection result of the mid-section structure is obtained by combining the actual scaling parameters and the actual guide vane deformation parameters.

2. The method for detecting the quality of the middle section structure of a multistage pump according to claim 1, characterized in that: The vibration signal of the middle section structure of the target multi-stage pump is monitored during operation to obtain vibration parameters. The operation time parameters and water quality parameters of the middle section structure are collected to predict flow channel scaling and obtain predicted scaling parameters, including: Monitor the vibration signal of the middle section structure of the target multi-stage pump during operation to obtain vibration parameters; During the operation of the target multi-stage pump, water quality parameters are regularly monitored, a water quality parameter set is obtained, and water quality parameters are calculated; Collect the accumulated running time of the middle section structure and obtain the running time parameters; According to the vibration parameters, water quality parameters and operation time parameters, flow channel scaling prediction is performed to obtain predicted scaling parameters.

3. The method for detecting the quality of the middle section structure of a multistage pump according to claim 2, characterized in that: According to the vibration parameters, water quality parameters and operation time parameters, flow channel scaling prediction is performed to obtain predicted scaling parameters, including: According to the historical maintenance data of the middle section structure of the same model multi-stage pump, collect the sample vibration parameter set, sample water quality parameter set, sample running time parameter set, and collect the scaling size of the flow channel scaling, and mark the sample scaling parameter set; Based on deep learning, a flow channel scaling predictor is constructed; Using the sample vibration parameter set, the sample water quality parameter set, the sample running time parameter set and the sample scaling parameter set as supervised training data and test data, the flow channel scaling predictor is supervised trained and tested, and the training and testing are completed after the accuracy test is qualified; The vibration parameter, water quality parameter and operation time parameter are input into the flow channel scaling predictor, and the predicted scaling parameter is obtained by prediction output.

4. The method for detecting the quality of the middle section structure of a multistage pump according to claim 1, characterized in that: According to the error amplitude between N scaling parameter levels and the predicted scaling parameter, the number of N classification paths is calculated, N scaling parameter classifiers corresponding to the N scaling parameter levels are selected, the flow channel images are respectively input into the scaling parameter classification paths of the N classification paths in the N scaling parameter classifiers, and N scaling parameter classification result sets are obtained by output, including: Calculate the error margins between N scaling parameter levels and the predicted scaling parameter respectively, and use 1 minus the N error margins, multiply by the number of scaling parameter classification paths in each scaling parameter classifier and round up to obtain the number of N classification paths; Selecting N scaling parameter classifiers corresponding to the N scaling parameter levels, wherein a plurality of scaling parameter classifiers corresponding to a plurality of scaling parameter levels are pre-trained, each scaling parameter classifier includes a plurality of scaling parameter classification paths constructed and trained based on a convolutional neural network, the training data includes a sample flow channel image and a sample scaling parameter classification result, and each sample scaling parameter classification result includes a binary classification result of whether a scaling parameter in the sample flow channel image is a scaling parameter level corresponding to the scaling parameter classifier; According to the number of N classification paths, N scaling parameter classification paths of the number of classification paths are randomly selected in the N scaling parameter classifiers, the flow channel image is input, and N scaling parameter classification result sets are obtained as output.

5. The method for detecting the quality of the middle section structure of a multistage pump according to claim 1, characterized in that: According to the actual scaling parameters, guide vane deformation prediction is performed to obtain predicted guide vane deformation parameters, including: According to the historical maintenance data of the middle section structure of the same model multi-stage pump, a set of sample scaling parameters is collected, and the deformation angles of the guide vanes under different sample scaling parameters are collected and marked as guide vane deformation parameters to obtain a set of sample guide vane deformation parameters; Based on deep learning, a guide vane deformation predictor is constructed; Using the sample scaling parameter set and the sample guide vane deformation parameter set as supervised training data and test data, supervised training and testing are performed on the guide vane deformation predictor, and the training and testing are completed after the accuracy test is qualified; The actual fouling parameters are input into the guide vane deformation predictor, and the prediction output is used to obtain the predicted guide vane deformation parameters.

6. The method for detecting the quality of the middle section structure of a multistage pump according to claim 1, characterized in that: In combination with the actual scaling parameters and the actual guide vane deformation parameters, the quality inspection results of the mid-section structure are obtained, including: The actual scaling parameters and the actual guide vane deformation parameters are integrated as the quality inspection result of the middle section structure of the target multi-stage pump.

7. A multi-stage pump mid-section structure quality detection system, characterized in that: A method for detecting the quality of a multistage pump mid-section structure according to any one of claims 1 to 6, the system comprising: The vibration monitoring module is used to monitor the vibration signal of the middle section structure of the target multi-stage pump during operation, obtain vibration parameters, collect the operation time parameters and water quality parameters of the middle section structure, predict the flow channel scaling, and obtain the predicted scaling parameters; An image acquisition module is used to acquire the flow channel image in the middle section structure through an industrial endoscope, and input the flow channel image into a plurality of scaling parameter classifiers respectively according to the predicted scaling parameters to identify and obtain the actual scaling parameters, wherein each scaling parameter classifier includes a plurality of scaling parameter classification paths; a classification and recognition module, configured to predict guide vane deformation according to the actual scaling parameters, obtain predicted guide vane deformation parameters, collect guide vane images in the middle section structure through an industrial endoscope, input the guide vane images into a plurality of guide vane parameter classifiers according to the predicted guide vane deformation parameters, and identify and obtain actual guide vane deformation parameters; The result acquisition module is used to obtain the quality detection result of the mid-section structure by combining the actual scaling parameter and the actual guide vane deformation parameter.

Images