Plunger pump liquid leakage detection method and system
Through the combination of thermal imaging monitoring array and ultrasonic detection, the problem of accurate and low efficiency of liquid leakage detection of plunger pump is solved, and efficient and accurate leakage detection effect is achieved.
Patent Information
- Application Number
- CN202411375827.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-30
- Publication Date
- 2025-01-07
AI Technical Summary
In the prior art, the accuracy and efficiency of liquid leakage detection of plunger pumps are low, making it difficult to meet the needs of modern industrial production for efficient and accurate detection.
Thermal imaging monitoring array is used for fixed-point monitoring, and temperature distribution data is obtained from different angles through multiple thermal imagers, registration and fusion and three-dimensional reconstruction are carried out, temperature variation deviation proofreading channels are constructed, standard three-dimensional temperature distribution maps are generated, and the twin model is performed. Mesh traversal comparison, determine the set of abnormal areas, and finally perform leakage detection with ultrasound.
The accuracy and efficiency of liquid leakage detection of plunger pump is improved, the leakage area is determined through thermal imaging, and then ultrasonic waves are used for accurate detection, which significantly improves the accuracy and efficiency of detection.
Smart Images
Figure CN119268960A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent detection technology, and in particular to a plunger pump liquid leakage detection method and system. Background Art
[0002] In the field of industrial production, plunger pumps, as a key fluid transmission equipment, are widely used in many industries such as petrochemicals, pharmaceuticals, and food. However, due to long-term operation, material aging, improper operation and other reasons, plunger pumps often have problems with liquid leakage. Liquid leakage not only leads to waste of resources and environmental pollution, but also may pose a serious threat to the safe operation of equipment. Therefore, it is of great significance to conduct timely and accurate liquid leakage detection of plunger pumps. Traditional plunger pump liquid leakage detection methods mostly rely on manual inspections and simple sensor monitoring. These methods often have shortcomings such as slow response speed and low accuracy, and it is difficult to meet the needs of modern industrial production for efficient and accurate detection. Summary of the invention
[0003] The embodiments of the present application provide a plunger pump liquid leakage detection method and system, which solve the technical problems of low accuracy and efficiency in plunger pump liquid leakage detection in the prior art.
[0004] In view of the above problems, an embodiment of the present application provides a plunger pump liquid leakage detection method and system.
[0005] A first aspect of an embodiment of the present application provides a plunger pump liquid leakage detection method, the method comprising:
[0006] The target plunger pump is monitored at a fixed point by a thermal imaging monitoring array to obtain a temperature distribution data set of the target plunger pump, wherein the thermal imaging monitoring array includes N thermal imagers, and each thermal imager has a different monitoring angle, and N is an integer greater than 1; based on a predetermined fusion strategy, the temperature distribution data set is aligned and fused according to the monitoring angle, and the fusion result is three-dimensionally reconstructed to generate a target three-dimensional temperature distribution map; a temperature change deviation correction channel is constructed, wherein the temperature change deviation correction channel includes a plurality of temperature change deviation correction units, and each temperature change deviation correction unit corresponds to a marked area; based on the marked area, the target three-dimensional temperature distribution map is mapped and corrected by the plurality of temperature change deviation correction units to generate a standard three-dimensional temperature distribution map; a simulated three-dimensional temperature distribution map of the target plunger pump twin model is obtained, a grid traversal comparison is performed on the standard three-dimensional temperature distribution map and the simulated three-dimensional temperature distribution map, and a set of abnormal areas is determined based on the comparison results; ultrasonic leakage detection is performed with the set of abnormal areas as the detection target to obtain a liquid leakage detection result.
[0007] A second aspect of the embodiments of the present application provides a plunger pump liquid leakage detection system, the system comprising:
[0008] A monitoring module, the monitoring module is used to perform fixed-point monitoring of a target plunger pump through a thermal imaging monitoring array to obtain a temperature distribution data set of the target plunger pump, wherein the thermal imaging monitoring array includes N thermal imagers, and each thermal imager has a different monitoring angle, and N is an integer greater than 1; a three-dimensional reconstruction module, the three-dimensional reconstruction module is used to perform registration and fusion of the temperature distribution data set according to the monitoring angle based on a predetermined fusion strategy, and perform three-dimensional reconstruction of the fusion result to generate a target three-dimensional temperature distribution map; a channel construction module, the channel construction module is used to construct a temperature change deviation correction channel, wherein the temperature change deviation correction channel includes multiple temperature change deviation corrections A pair of units, and each temperature change deviation correction unit corresponds to a marked area; a correction module, the correction module is used to map and correct the target three-dimensional temperature distribution map through the multiple temperature change deviation correction units based on the marked area, and generate a standard three-dimensional temperature distribution map; a comparison module, the comparison module is used to obtain a simulated three-dimensional temperature distribution map of the target plunger pump twin model, perform grid traversal comparison on the standard three-dimensional temperature distribution map and the simulated three-dimensional temperature distribution map, and determine an abnormal area set based on the comparison result; a detection module, the detection module is used to take the abnormal area set as the detection target, perform ultrasonic leakage detection, and obtain a liquid leakage detection result.
[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0010] First, a thermal imaging monitoring array is used to perform fixed-point monitoring on the target plunger pump to obtain a temperature distribution data set of the target plunger pump, wherein the thermal imaging monitoring array includes N thermal imagers, and each thermal imager has a different monitoring angle, and N is an integer greater than 1. Then, according to a predetermined fusion strategy, the temperature distribution data at different angles are aligned and fused, and three-dimensional reconstruction is performed to generate a target three-dimensional temperature distribution map. Next, a temperature change deviation correction channel is constructed, which includes multiple temperature change deviation correction units, and each unit corresponds to a marked area. Using these marked areas, the target three-dimensional temperature distribution map is mapped and corrected through multiple temperature change deviation correction units to generate a standard three-dimensional temperature distribution map. Then, the simulated three-dimensional temperature distribution map of the target plunger pump twin model is obtained, and the standard three-dimensional temperature distribution map and the simulated three-dimensional temperature distribution map are grid traversed and compared, and the abnormal area set is determined based on the comparison results. Finally, these abnormal areas are used as detection targets, ultrasonic leakage detection is performed, and liquid leakage detection results are obtained. The invention solves the technical problem of low accuracy and efficiency of plunger pump liquid leakage detection in the prior art, and achieves the technical effect of determining the leakage area by thermal imaging and then using ultrasound for precise detection, thereby improving the detection accuracy and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0012] Figure 1 A schematic flow chart of a plunger pump liquid leakage detection method provided in an embodiment of the present application;
[0013] Figure 2 A schematic structural diagram of a plunger pump liquid leakage detection system provided in an embodiment of the present application.
[0014] Explanation of the accompanying drawings: monitoring module 11, three-dimensional reconstruction module 12, channel construction module 13, proofreading module 14, comparison module 15, detection module 16. DETAILED DESCRIPTION
[0015] The embodiments of the present application solve the technical problem of low accuracy and efficiency in plunger pump liquid leakage detection in the prior art by providing a plunger pump liquid leakage detection method and system.
[0016] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0017] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or are inherent to these processes, methods, products or devices.
[0018] Embodiment 1
[0019] like Figure 1 As shown, an embodiment of the present application provides a plunger pump liquid leakage detection method, wherein the method comprises:
[0020] The target plunger pump is monitored at a fixed point through a thermal imaging monitoring array to obtain a temperature distribution data set of the target plunger pump, wherein the thermal imaging monitoring array includes N thermal imagers, and each thermal imager has a different monitoring angle, and N is an integer greater than 1.
[0021] The target plunger pump is monitored at a fixed point through a thermal imaging monitoring array to obtain a temperature distribution data set of the target plunger pump. The thermal imaging monitoring array consists of N thermal imagers, each of which captures the thermal radiation of the plunger pump from a different monitoring angle, thereby obtaining temperature information at different angles. Select a suitable location at the target plunger pump and deploy N thermal imagers, ensuring that each thermal imager has a different monitoring angle so that it can cover all parts of the plunger pump. Start all thermal imagers to capture the thermal radiation of the plunger pump, extract temperature information from the image data captured by each thermal imager, and integrate the temperature data obtained by all thermal imagers into a complete temperature distribution data set.
[0022] Based on a predetermined fusion strategy, the temperature distribution data set is registered and fused according to the monitoring angle, and the fusion result is three-dimensionally reconstructed to generate a target three-dimensional temperature distribution map.
[0023] According to the predetermined fusion strategy, the temperature distribution data sets obtained by the thermal imagers at different monitoring angles are registered and fused to obtain a temperature distribution data set containing all the information. Three-dimensional reconstruction is performed on the fused basis to generate the target three-dimensional temperature distribution map. Optionally, the depth information of the thermal imager is obtained, and the two-dimensional temperature distribution data and the depth information are combined to generate three-dimensional point cloud data, where each point represents a position on the plunger pump and carries the corresponding temperature information. The point cloud data is converted into a three-dimensional mesh, and each vertex or face element of the mesh carries temperature information. Graphics rendering technology is used to convert the three-dimensional mesh into a visual three-dimensional temperature distribution map.
[0024] Furthermore, the configuration of the predetermined fusion strategy includes:
[0025] The data distortion interference factors under multiple monitoring angles are collected respectively to determine a set of distortion interference factors; interference coefficients are evaluated on the set of distortion interference factors by a thermal imaging expert group to obtain a set of interference influence coefficients; the angle data credibility is reversely set based on the set of interference influence coefficients, and a fusion strategy of overlapping areas is configured according to the angle data credibility to generate the predetermined fusion strategy.
[0026] In order to accurately fuse the temperature distribution data sets obtained from different monitoring angles, it is necessary to analyze the possible sources of data distortion at each monitoring angle, such as reflection, occlusion, and thermal radiation interference. By comparing the actual temperature data with the theoretical value, historical data, or simulation data, the influence of each distortion source on the data is quantitatively or qualitatively evaluated, and the corresponding distortion interference factors are collected, and the distortion interference factors of all monitoring angles are integrated into a set of distortion interference factors. A thermal imaging expert group is composed of experts in the field of thermal imaging with rich experience and professional knowledge. The thermal imaging expert group can independently evaluate the data distortion at each angle and give the corresponding interference influence coefficient. The interference influence coefficient reflects the influence of the data distortion at this angle on the overall temperature distribution data. The interference influence coefficients given by the expert group are summarized and sorted to form an interference influence coefficient set. Based on the interference influence coefficient set, the data credibility at each monitoring angle is set in reverse, that is, the higher the interference influence coefficient of the monitoring angle, the lower its data credibility; conversely, the higher the data credibility. According to the angle data credibility, a suitable fusion strategy is configured for the area where data overlaps at different monitoring angles. For example, for areas with high data credibility, a higher weight can be given; for areas with low data credibility, it may be necessary to take measures such as smoothing or removing outliers. In summary, a predetermined fusion strategy is obtained. For overlapping areas, temperature distribution data is obtained based on credibility weighted calculation, and for non-overlapping areas, temperature distribution data can be directly obtained.
[0027] Furthermore, the fusion result is reconstructed in three dimensions to generate a target three-dimensional temperature distribution map, including:
[0028] In a visual simulation platform, three-dimensional simulation modeling is performed on the target plunger pump to construct a three-dimensional geometric model of the plunger pump; the fusion result is mapped and distributed to the three-dimensional geometric model of the plunger pump to obtain a target three-dimensional temperature distribution map.
[0029] Preferably, a three-dimensional simulation modeling of the target plunger pump is performed in a visual simulation platform to construct a three-dimensional geometric model of the plunger pump. In the visual simulation platform, mapping parameters are set to map the fusion results to the three-dimensional geometric model of the plunger pump. Among them, the mapping parameters include a color mapping scheme (for mapping temperature values to different colors), a transparency setting (for displaying the internal temperature distribution), etc. According to the set mapping parameters, the mapping operation is performed. At this time, the visual simulation platform will display the temperature data in the fusion result on the three-dimensional geometric model of the plunger pump according to the corresponding color mapping scheme to form a target three-dimensional temperature distribution diagram.
[0030] A temperature variation deviation correction channel is constructed, wherein the temperature variation deviation correction channel includes a plurality of temperature variation deviation correction units, and each temperature variation deviation correction unit corresponds to a marking area.
[0031] The temperature variation deviation calibration channel includes multiple temperature variation deviation calibration units, each of which can collect temperature data of the corresponding marked area and compare it with the expected or reference temperature value. Since different areas of the plunger pump differ in structure, function, material, and degree of contact with the liquid, these differences will cause them to respond differently to changes in ambient temperature; therefore, when calibrating, temperature variation deviation calibration is performed according to the divided areas, which can improve the precision and accuracy of temperature data calibration.
[0032] Furthermore, a temperature change deviation calibration channel is constructed, including:
[0033] Multidimensional test conditions are configured based on temperature variation interference factors and preset adjustment ranges, wherein the temperature variation interference factors include at least ambient temperature, ambient humidity and hydraulic oil temperature; a temperature variation test is performed on the target plunger pump according to the multidimensional test conditions to obtain a plurality of sample temperature variation distribution diagrams; regional clustering is performed according to the plurality of sample temperature variation distribution diagrams to obtain a plurality of marked regions; temperature variation deviation analysis is performed on the plurality of marked regions in turn to construct a plurality of temperature variation deviation correction units; mapping and fusion are performed on the plurality of marked regions and the plurality of temperature variation deviation correction units based on a decision tree to establish the temperature variation deviation correction channel.
[0034] The temperature variation interference factor is a parameter that affects the temperature change of the plunger pump. The temperature variation interference factor includes at least the ambient temperature, ambient humidity and hydraulic oil temperature. The preset adjustment range refers to presetting one or more adjustment ranges for each temperature variation interference factor to simulate different working conditions and environmental changes. Different temperature variation interference factors and adjustment ranges are combined to form multidimensional test conditions. According to the configured multidimensional test conditions, the test environment is adjusted, including temperature, humidity and hydraulic oil temperature. Under each test condition, the target plunger pump is subjected to a temperature variation test, the temperature distribution data is recorded, and multiple sample temperature variation distribution maps are generated. Multiple sample temperature variation distribution maps are regionally clustered to obtain multiple marked areas with similar temperature characteristics. For each marked area, its temperature change under different test conditions is analyzed, and the temperature variation deviation is identified and quantified. Based on the analysis results, a temperature variation deviation correction unit is constructed for each marked area. The input of the decision tree is the feature of the marked area, and the output should be the identifier of the corresponding temperature variation deviation correction unit. After training, it can be used to map the new marked area to the corresponding temperature variation deviation correction unit. Through the decision tree, multiple temperature variation deviation correction units are mapped and fused with the corresponding marked areas. Each correction unit is responsible for correcting the temperature variation deviation of its corresponding area. The mapped and fused correction units and marked areas are integrated into a unified temperature variation deviation correction channel. The temperature variation deviation correction channel can monitor the temperature distribution of the target plunger pump in real time and automatically correct it when a temperature variation deviation is detected.
[0035] Furthermore, regional clustering is performed according to the temperature change distribution diagrams of the multiple samples to obtain multiple marked regions, including:
[0036] The temperature change distribution diagrams of the multiple samples are divided into regions according to a preset window to obtain multiple temperature change distribution region sets; a first temperature change region distribution set is selected, and a first temperature change region and a second temperature change region are randomly obtained, wherein the first temperature change region and the second temperature change region are any two adjacent regions in the first temperature change region distribution set; a temperature deviation between the first temperature change region and the second temperature change region is calculated, and if the temperature deviation meets a predetermined temperature difference threshold, the first temperature change region and the second temperature change region are fused and clustered to generate a first updated temperature change region; based on the first updated temperature change region, the first temperature change region distribution set is iteratively analyzed in turn to obtain a first region clustering result, and added to multiple region clustering results; regions whose intersection region frequency is greater than a frequency threshold in the multiple region clustering results are selected for fusion, and multiple labeled regions are generated based on the fusion result.
[0037] A preset window, i.e., the size of the area division, is set to divide the temperature change distribution area in the sample temperature change distribution map. Multiple sample temperature change distribution maps are divided into regions according to the preset window to obtain multiple temperature change distribution area sets. One is selected from multiple temperature change distribution area sets as the first temperature change area distribution set, and two adjacent areas are randomly selected in the first temperature change area distribution set as the first temperature change area and the second temperature change area. The average temperature of the first temperature change area and the second temperature change area is calculated, and the temperature deviation between them is calculated. If the calculated temperature deviation is less than or equal to the predetermined temperature difference threshold, it is considered that the two areas have similar temperature characteristics, and they are fused and clustered to generate a new, larger area, i.e., the first updated temperature change area. Repeat the above process, and iteratively analyze all areas in the first temperature change area distribution set until there are no areas that can be fused. After iterative analysis, the first regional clustering result of the first temperature change area distribution set is obtained. The above iterative analysis is performed on all temperature change area distribution sets to obtain multiple regional clustering results. Compare these regional clustering results to find the intersection area, that is, the area that exists in different regional clustering results. A frequency threshold is set to determine which intersection regions are stable and representative. Regions whose intersection frequency is greater than the frequency threshold are selected for fusion to generate multiple final marked regions, and the temperature change amplitude is the same in each marked region by default.
[0038] Furthermore, performing temperature variation deviation analysis on the plurality of marked areas in sequence includes:
[0039] A first marked area is selected, and a first sample training data set is obtained with the first marked area as a constraint, wherein the first sample training data includes a sample temperature change interference factor, a sample temperature distribution diagram, and a sample temperature correction diagram; the first sample training data set is divided into a first training set and a first verification set according to a preset division strategy; a first temperature change deviation correction unit is built based on a feedforward neural network, and the first training set and the first verification set are used to perform supervised learning and feedback verification on the first temperature change deviation correction unit, respectively; after the expected convergence constraint is met, a trained first temperature change deviation correction unit is obtained and added to the multiple temperature change deviation correction units.
[0040] One of the multiple marked areas is randomly selected as the first marked area, and the first marked area is used as a constraint to screen out data related to the area from all sample data, including sample temperature change interference factors (such as ambient temperature, ambient humidity, hydraulic oil temperature, etc.), sample temperature distribution diagrams, and corresponding sample temperature correction diagrams to form a first sample training data set. According to experience, a preset partitioning strategy is set, such as 80% of the data is used for training and 20% of the data is used for verification. According to the preset partitioning strategy, the first sample training data set is divided into a first training set and a first verification set. The first training set is used to train the temperature change deviation correction unit, and the first verification set is used to verify the performance of the correction unit during the training process. A feedforward neural network (such as a multi-layer perceptron MLP) is used as the basic architecture to build the first temperature change deviation correction unit. The first temperature change deviation correction unit can accept the temperature change interference factor as input and output a predicted temperature correction diagram. The first training set is used to supervise the learning of the first temperature change deviation correction unit. During the training process, the actual sample temperature correction diagram is used as the target output, and the network parameters are adjusted by the back propagation algorithm to minimize the error between the predicted output and the target output. At the same time, the first validation set is used to perform feedback verification on the proofreading unit to evaluate its generalization ability and performance. Repeat the above supervised learning and feedback verification process until the expected convergence constraints are met. When the first temperature variation deviation proofreading unit meets the convergence constraints, it is considered that it has been trained and has good performance. The trained first temperature variation deviation proofreading unit is added to multiple temperature variation deviation proofreading units.
[0041] Based on the marked area, the target three-dimensional temperature distribution map is mapped and proofread by the multiple temperature change deviation proofreading units to generate a standard three-dimensional temperature distribution map.
[0042] The target three-dimensional temperature distribution map is mapped and proofread through multiple temperature variation deviation proofreading units, that is, the temperature data of the corresponding area in the target three-dimensional temperature distribution map is input into the corresponding temperature variation deviation proofreading unit, and each temperature variation deviation proofreading unit corrects the input temperature data to generate corrected temperature data. The corrected temperature data output by all temperature variation deviation proofreading units are fused according to the structure of the original target three-dimensional temperature distribution map to form a standard three-dimensional temperature data set.
[0043] Furthermore, mapping and calibrating the target three-dimensional temperature distribution map by the multiple temperature change deviation calibration units includes:
[0044] The temperature change interference factor is collected in real time through a multi-dimensional sensor array to obtain a real-time interference factor; the target three-dimensional temperature distribution map is divided based on the marked area, and a matching temperature change deviation correction unit is obtained according to the mapping of the division result; the real-time interference factor and the division result are input into the matching temperature change deviation correction unit for deviation correction, and the standard three-dimensional temperature distribution map is obtained after fusion.
[0045] Preferably, by deploying a multi-dimensional sensor array, including temperature sensors, humidity sensors, pressure sensors, etc., the interference factors that affect the temperature distribution in the environment are collected in real time to obtain real-time interference factors. The target three-dimensional temperature distribution map is divided according to the marked areas to obtain divided areas. According to the divided areas and the mapping relationship, the corresponding temperature change deviation correction unit is found for each area. The real-time interference factor data and the division results are input into the matching temperature change deviation correction unit, and the temperature change deviation correction unit will correct the temperature data of the corresponding area in the target three-dimensional temperature distribution map to eliminate or reduce the influence of the temperature change interference factor. The corrected temperature data output by all correction units are fused according to the structure of the original target three-dimensional temperature distribution map to obtain a standard three-dimensional temperature distribution map.
[0046] A simulated three-dimensional temperature distribution map of the target plunger pump twin model is obtained, a grid traversal comparison is performed on the standard three-dimensional temperature distribution map and the simulated three-dimensional temperature distribution map, and a set of abnormal areas is determined based on the comparison results.
[0047] Construct a twin model of the target plunger pump, and use the twin model of the target plunger pump for simulation to generate a simulated three-dimensional temperature distribution map. The simulated three-dimensional temperature distribution map can accurately reflect the temperature distribution of the target plunger pump at the same time node. Define grids of the same size and position on the simulated three-dimensional temperature distribution map and the standard three-dimensional temperature distribution map, and traverse the temperature data in each grid. For each grid, compare the temperature value on the simulated distribution map with the temperature value on the standard distribution map. Set a temperature difference threshold based on actual needs. If the difference between the temperature value of a grid on the simulated distribution map and the temperature value on the standard distribution map exceeds this threshold, the grid is considered abnormal, and all abnormal grids are marked on the simulated distribution map to form an abnormal area set.
[0048] Furthermore, a simulated three-dimensional temperature distribution diagram of the target plunger pump twin model is obtained, including:
[0049] Based on the digital twin technology, a target plunger pump twin model is constructed; the target plunger pump twin model is operated synchronously with the target plunger pump, and the simulated three-dimensional temperature distribution map is obtained based on the thermal imaging fixed-point monitoring node extraction.
[0050] Preferably, based on digital twin technology, a target plunger pump twin model is constructed using modeling software. Positions corresponding to thermal imaging monitoring points on the physical plunger pump are set in the twin model, and the two models are run synchronously to extract a simulated three-dimensional temperature distribution map.
[0051] Furthermore, performing a grid traversal comparison on the standard three-dimensional temperature distribution map and the simulated three-dimensional temperature distribution map, and determining a set of abnormal regions based on the comparison result, includes:
[0052] Meshing the standard three-dimensional temperature distribution map and the simulated three-dimensional temperature distribution map based on a preset mesh step size, and performing a temperature deviation traversal comparison according to the meshing results;
[0053] The grid area where the extracted temperature deviation is greater than a predetermined temperature difference tolerance threshold is set as an abnormal grid area, wherein the predetermined temperature difference tolerance threshold is set based on the thermal imaging acquisition accuracy and the simulation operation accuracy;
[0054] Adjacent abnormal grid regions are connected by fitting to obtain the abnormal region set.
[0055] According to the actual situation and the complexity of the plunger pump, a suitable grid step size is preset, and the preset grid step size is used to grid the standard three-dimensional temperature distribution map and the simulated three-dimensional temperature distribution map. According to the divided grids, each grid in the two distribution maps is traversed to calculate the temperature deviation of the corresponding grids in the standard distribution map and the simulated distribution map. Based on the thermal imaging acquisition accuracy and the simulation operation accuracy, a predetermined temperature difference tolerance threshold is set to determine the abnormal area. For grids with temperature deviations greater than the predetermined temperature difference tolerance threshold, they are set as abnormal grid areas. Check all abnormal grid areas and connect adjacent abnormal grids to form an abnormal area set.
[0056] Taking the abnormal area set as the detection target, ultrasonic leakage detection is performed to obtain liquid leakage detection results.
[0057] According to the characteristics and working environment of the target plunger pump, a suitable ultrasonic probe and parameter settings are selected, and the abnormal area set is taken as the detection target. Ultrasonic leak detection is performed to identify liquid leaks and obtain liquid leak detection results. The liquid leak detection results include the leakage location, size and other information of the target plunger pump.
[0058] In summary, the embodiments of the present application have at least the following technical effects:
[0059] First, a thermal imaging monitoring array is used to perform fixed-point monitoring on the target plunger pump to obtain a temperature distribution data set of the target plunger pump, wherein the thermal imaging monitoring array includes N thermal imagers, and each thermal imager has a different monitoring angle, and N is an integer greater than 1. Then, according to a predetermined fusion strategy, the temperature distribution data at different angles are aligned and fused, and three-dimensional reconstruction is performed to generate a target three-dimensional temperature distribution map. Next, a temperature change deviation correction channel is constructed, which includes multiple temperature change deviation correction units, and each unit corresponds to a marked area. Using these marked areas, the target three-dimensional temperature distribution map is mapped and corrected through multiple temperature change deviation correction units to generate a standard three-dimensional temperature distribution map. Then, the simulated three-dimensional temperature distribution map of the target plunger pump twin model is obtained, and the standard three-dimensional temperature distribution map and the simulated three-dimensional temperature distribution map are grid traversed and compared, and the abnormal area set is determined based on the comparison results. Finally, these abnormal areas are used as detection targets, ultrasonic leakage detection is performed, and liquid leakage detection results are obtained. The invention solves the technical problem of low accuracy and efficiency of plunger pump liquid leakage detection in the prior art, and achieves the technical effect of determining the leakage area by thermal imaging and then using ultrasound for precise detection, thereby improving the detection accuracy and efficiency.
[0060] Embodiment 2
[0061] Based on the same inventive concept as the plunger pump liquid leakage detection method in the aforementioned embodiment, Figure 2 As shown, the present application provides a plunger pump liquid leakage detection system, and the system and method embodiments in the present application are based on the same inventive concept. The system includes:
[0062] A monitoring module 11, the monitoring module 11 is used to perform fixed-point monitoring of a target plunger pump through a thermal imaging monitoring array to obtain a temperature distribution data set of the target plunger pump, wherein the thermal imaging monitoring array includes N thermal imagers, and each thermal imager has a different monitoring angle, and N is an integer greater than 1; a three-dimensional reconstruction module 12, the three-dimensional reconstruction module 12 is used to perform registration and fusion of the temperature distribution data set according to the monitoring angle based on a predetermined fusion strategy, and perform three-dimensional reconstruction of the fusion result to generate a target three-dimensional temperature distribution map; a channel construction module 13, the channel construction module 13 is used to construct a temperature change deviation correction channel, wherein the temperature change deviation correction channel includes multiple temperature change deviation corrections Units, and each temperature change deviation correction unit corresponds to a marked area; a correction module 14, the correction module 14 is used to map and calibrate the target three-dimensional temperature distribution map through the multiple temperature change deviation correction units based on the marked area, and generate a standard three-dimensional temperature distribution map; a comparison module 15, the comparison module 15 is used to obtain a simulated three-dimensional temperature distribution map of the target plunger pump twin model, perform grid traversal comparison on the standard three-dimensional temperature distribution map and the simulated three-dimensional temperature distribution map, and determine the abnormal area set based on the comparison result; a detection module 16, the detection module 16 is used to perform ultrasonic leakage detection with the abnormal area set as the detection target, and obtain a liquid leakage detection result.
[0063] Furthermore, the three-dimensional reconstruction module 12 is used to perform the following method:
[0064] The data distortion interference factors under multiple monitoring angles are collected respectively to determine a set of distortion interference factors; interference coefficients are evaluated on the set of distortion interference factors by a thermal imaging expert group to obtain a set of interference influence coefficients; the angle data credibility is reversely set based on the set of interference influence coefficients, and a fusion strategy of overlapping areas is configured according to the angle data credibility to generate the predetermined fusion strategy.
[0065] Furthermore, the three-dimensional reconstruction module 12 is used to perform the following method:
[0066] In a visual simulation platform, three-dimensional simulation modeling is performed on the target plunger pump to construct a three-dimensional geometric model of the plunger pump; the fusion result is mapped and distributed to the three-dimensional geometric model of the plunger pump to obtain a target three-dimensional temperature distribution map.
[0067] Furthermore, the channel construction module 13 is used to execute the following method:
[0068] Multidimensional test conditions are configured based on temperature variation interference factors and preset adjustment ranges, wherein the temperature variation interference factors include at least ambient temperature, ambient humidity and hydraulic oil temperature; a temperature variation test is performed on the target plunger pump according to the multidimensional test conditions to obtain a plurality of sample temperature variation distribution diagrams; regional clustering is performed according to the plurality of sample temperature variation distribution diagrams to obtain a plurality of marked regions; temperature variation deviation analysis is performed on the plurality of marked regions in turn to construct a plurality of temperature variation deviation correction units; mapping and fusion are performed on the plurality of marked regions and the plurality of temperature variation deviation correction units based on a decision tree to establish the temperature variation deviation correction channel.
[0069] Furthermore, the channel construction module 13 is used to execute the following method:
[0070] The temperature change distribution diagrams of the multiple samples are divided into regions according to a preset window to obtain multiple temperature change distribution region sets; a first temperature change region distribution set is selected, and a first temperature change region and a second temperature change region are randomly obtained, wherein the first temperature change region and the second temperature change region are any two adjacent regions in the first temperature change region distribution set; a temperature deviation between the first temperature change region and the second temperature change region is calculated, and if the temperature deviation meets a predetermined temperature difference threshold, the first temperature change region and the second temperature change region are fused and clustered to generate a first updated temperature change region; based on the first updated temperature change region, the first temperature change region distribution set is iteratively analyzed in turn to obtain a first region clustering result, and added to multiple region clustering results; regions whose intersection region frequency is greater than a frequency threshold in the multiple region clustering results are selected for fusion, and multiple labeled regions are generated based on the fusion result.
[0071] Furthermore, the channel construction module 13 is used to execute the following method:
[0072] A first marked area is selected, and a first sample training data set is obtained with the first marked area as a constraint, wherein the first sample training data includes a sample temperature change interference factor, a sample temperature distribution diagram, and a sample temperature correction diagram; the first sample training data set is divided into a first training set and a first verification set according to a preset division strategy; a first temperature change deviation correction unit is built based on a feedforward neural network, and the first training set and the first verification set are used to perform supervised learning and feedback verification on the first temperature change deviation correction unit, respectively; after the expected convergence constraint is met, a trained first temperature change deviation correction unit is obtained and added to the multiple temperature change deviation correction units.
[0073] Furthermore, the proofreading module 14 is used to perform the following method:
[0074] The temperature change interference factor is collected in real time through a multi-dimensional sensor array to obtain a real-time interference factor; the target three-dimensional temperature distribution map is divided based on the marked area, and a matching temperature change deviation correction unit is obtained according to the mapping of the division result; the real-time interference factor and the division result are input into the matching temperature change deviation correction unit for deviation correction, and the standard three-dimensional temperature distribution map is obtained after fusion.
[0075] Furthermore, the comparison module 15 is used to perform the following method:
[0076] Based on the digital twin technology, a target plunger pump twin model is constructed; the target plunger pump twin model is operated synchronously with the target plunger pump, and the simulated three-dimensional temperature distribution map is obtained based on the thermal imaging fixed-point monitoring node extraction.
[0077] Furthermore, the comparison module 15 is used to perform the following method:
[0078] The standard three-dimensional temperature distribution map and the simulated three-dimensional temperature distribution map are gridded based on a preset grid step size, and a temperature deviation traversal comparison is performed according to the grid division result; grid areas with temperature deviations greater than a predetermined temperature difference tolerance threshold are extracted and set as abnormal grid areas, wherein the predetermined temperature difference tolerance threshold is set based on the thermal imaging acquisition accuracy and the simulation operation accuracy; adjacent abnormal grid areas are fitted and connected to obtain the abnormal area set.
[0079] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0080] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.
[0081] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.
Claims
1. A plunger pump liquid leakage detection method, characterized in that: The method comprises: Performing fixed-point monitoring on a target plunger pump through a thermal imaging monitoring array to obtain a temperature distribution data set of the target plunger pump, wherein the thermal imaging monitoring array includes N thermal imagers, and each thermal imager has a different monitoring angle, and N is an integer greater than 1; Based on a predetermined fusion strategy, the temperature distribution data set is registered and fused according to the monitoring angle, and the fusion result is three-dimensionally reconstructed to generate a target three-dimensional temperature distribution map; Constructing a temperature variation deviation correction channel, wherein the temperature variation deviation correction channel includes a plurality of temperature variation deviation correction units, and each temperature variation deviation correction unit corresponds to a marking area; Based on the marked area, mapping and calibrating the target three-dimensional temperature distribution map by the multiple temperature change deviation calibration units to generate a standard three-dimensional temperature distribution map; Obtain a simulated three-dimensional temperature distribution map of the target plunger pump twin model, perform a grid traversal comparison on the standard three-dimensional temperature distribution map and the simulated three-dimensional temperature distribution map, and determine a set of abnormal regions based on the comparison results; Taking the abnormal area set as the detection target, ultrasonic leakage detection is performed to obtain liquid leakage detection results.
2. The method according to claim 1, characterized in that Configure the scheduled convergence policy, including: Collect data distortion interference factors at multiple monitoring angles respectively, and determine a set of distortion interference factors; The thermal imaging expert group respectively evaluates the interference coefficients of the distortion interference factor set to obtain an interference influence coefficient set; The angle data credibility is reversely set based on the interference influence coefficient set, and the fusion strategy of the overlapping area is configured according to the angle data credibility to generate the predetermined fusion strategy.
3. The method according to claim 1, characterized in that Perform 3D reconstruction on the fusion results to generate the target 3D temperature distribution map, including: In a visual simulation platform, three-dimensional simulation modeling is performed on the target plunger pump to construct a three-dimensional geometric model of the plunger pump; The fusion result is mapped and distributed to the three-dimensional geometric model of the plunger pump to obtain a target three-dimensional temperature distribution map.
4. The method according to claim 1, characterized in that: Construct a temperature change deviation calibration channel, including: Configuring multi-dimensional test conditions based on temperature variation interference factors and preset adjustment ranges, wherein the temperature variation interference factors at least include ambient temperature, ambient humidity, and hydraulic oil temperature; Performing a temperature variation test on the target plunger pump according to the multi-dimensional test conditions to obtain a plurality of sample temperature variation distribution diagrams; Performing regional clustering according to the temperature change distribution diagrams of the multiple samples to obtain multiple marked regions; Performing temperature variation deviation analysis on the plurality of marked areas in sequence to construct a plurality of temperature variation deviation calibration units; The plurality of marking regions and the plurality of temperature variation deviation correction units are mapped and fused based on a decision tree to establish the temperature variation deviation correction channel.
5. The method according to claim 4, characterized in that Performing regional clustering according to the temperature change distribution diagrams of the multiple samples to obtain multiple marked regions, including: Dividing the temperature change distribution diagrams of the plurality of samples into regions according to a preset window to obtain a plurality of temperature change distribution region sets; Selecting a first temperature change region distribution set, and randomly acquiring a first temperature change region and a second temperature change region, wherein the first temperature change region and the second temperature change region are any two adjacent regions in the first temperature change region distribution set; Calculating a temperature deviation between the first temperature change region and the second temperature change region, and if the temperature deviation meets a predetermined temperature difference threshold, fusing and clustering the first temperature change region and the second temperature change region to generate a first updated temperature change region; Based on the first updated temperature change region, iteratively analyzing the first temperature change region distribution set in turn to obtain a first region clustering result, and adding it to the multiple region clustering results; Regions whose intersection region frequency is greater than a frequency threshold in the plurality of region clustering results are selected for fusion, and a plurality of labeled regions are generated based on the fusion results.
6. The method according to claim 4, characterized in that Performing temperature variation deviation analysis on the multiple marked areas in sequence includes: Selecting a first marked area, and using the first marked area as a constraint to obtain a first sample training data set, wherein the first sample training data includes a sample temperature variation interference factor, a sample temperature distribution map, and a sample temperature correction map; Dividing the first sample training data set into a first training set and a first validation set according to a preset division strategy; Building a first temperature variation deviation correction unit based on a feedforward neural network, and using the first training set and the first verification set to perform supervised learning and feedback verification on the first temperature variation deviation correction unit respectively; After the expected convergence constraint is met, a first temperature variation deviation correction unit that has been trained is obtained and added to the multiple temperature variation deviation correction units.
7. The method according to claim 4, characterized in that Mapping and calibrating the target three-dimensional temperature distribution map by using the multiple temperature change deviation calibration units includes: The temperature variation interference factor is collected in real time by a multi-dimensional sensor array to obtain a real-time interference factor; Dividing the target three-dimensional temperature distribution map based on the marked area, and obtaining a matching temperature change deviation correction unit according to the division result mapping; The real-time interference factor and the division result are input into the matching temperature change deviation correction unit for deviation correction, and the standard three-dimensional temperature distribution map is obtained after fusion.
8. The method according to claim 1, characterized in that: Get the simulated 3D temperature distribution map of the target plunger pump twin model, including: Based on digital twin technology, a twin model of the target plunger pump is constructed; The target plunger pump twin model is operated synchronously with the target plunger pump, and the simulated three-dimensional temperature distribution map is obtained based on the thermal imaging fixed-point monitoring node extraction.
9. The method according to claim 1, characterized in that: Performing a grid traversal comparison on the standard three-dimensional temperature distribution map and the simulated three-dimensional temperature distribution map, and determining an abnormal area set based on the comparison result, including: Meshing the standard three-dimensional temperature distribution map and the simulated three-dimensional temperature distribution map based on a preset mesh step size, and performing a temperature deviation traversal comparison according to the meshing results; The grid area where the extracted temperature deviation is greater than a predetermined temperature difference tolerance threshold is set as an abnormal grid area, wherein the predetermined temperature difference tolerance threshold is set based on the thermal imaging acquisition accuracy and the simulation operation accuracy; Adjacent abnormal grid regions are connected by fitting to obtain the abnormal region set.
10. A plunger pump liquid leakage detection system, characterized in that: A method for detecting liquid leakage of a plunger pump according to any one of claims 1 to 9, the system comprising: A monitoring module, the monitoring module is used to perform fixed-point monitoring on a target plunger pump through a thermal imaging monitoring array to obtain a temperature distribution data set of the target plunger pump, wherein the thermal imaging monitoring array includes N thermal imagers, and each thermal imager has a different monitoring angle, and N is an integer greater than 1; A three-dimensional reconstruction module, the three-dimensional reconstruction module is used to perform registration and fusion on the temperature distribution data set according to the monitoring angle based on a predetermined fusion strategy, and perform three-dimensional reconstruction on the fusion result to generate a target three-dimensional temperature distribution map; A channel construction module, wherein the channel construction module is used to construct a temperature variation deviation correction channel, wherein the temperature variation deviation correction channel includes a plurality of temperature variation deviation correction units, and each temperature variation deviation correction unit corresponds to a marking area; A calibration module, the calibration module being used to perform mapping calibration on the target three-dimensional temperature distribution map through the plurality of temperature variation deviation calibration units based on the marked area to generate a standard three-dimensional temperature distribution map; A comparison module, the comparison module is used to obtain a simulated three-dimensional temperature distribution map of the target plunger pump twin model, perform a grid traversal comparison on the standard three-dimensional temperature distribution map and the simulated three-dimensional temperature distribution map, and determine an abnormal area set based on the comparison result; A detection module is used to perform ultrasonic leakage detection with the abnormal area set as the detection target to obtain a liquid leakage detection result.