Pump jet flow field testing device and flow field missing compensation method

Abstract

The invention discloses a pump jet flow field testing device and a flow field missing compensation method. The pump jet flow field testing device comprises a high-frequency double-pulse laser, an image processing system, a camera synchronizer, a high-speed camera, a cavitation bubble water cylinder, a reflecting prism, a pump jet model and a transparent conduit. The high-frequency double-pulse laser is placed at the rear end of the pump spraying model in the cavitation bubble water cylinder, and laser emitted by the high-frequency double-pulse laser irradiates the pump spraying model through the cavitation bubble water cylinder; installing a reflecting prism and a high-speed camera on a laser plane of the pump spraying model; the data assimilation used by the image processing system and based on the generative adversarial network model effectively solve the problems of complex structure shielding, particle concentration loss and flow field information loss caused by wall reflection. According to the invention, the problem that related experimental devices for testing the internal flow field of the pump-jet propeller are not perfect enough is solved, the pump-jet missing flow field is effectively compensated, and the method is of great significance for the pump-jet propeller to know the form evolution law of the internal flow field from the experimental mechanism and to forecast the noise of the subsequent pump-jet propeller and calculate and verify the flow field.

Description

Technical Field

[0001] The invention belongs to the field of macro PIV experimental technology measurement, and in particular relates to a pump jet flow field testing device and a flow field loss compensation method. Background Art

[0002] As a new type of marine equipment propulsion mechanism, pump-jet propulsor (PJP) has attracted extensive attention due to its high efficiency, low noise and high critical speed. However, due to the complex coupling disturbance of the stator and rotor of the pump-jet propulsor, a highly complex near-wall turbulence is formed, which makes the internal flow field have obvious unsteady, nonlinear and multi-scale three-dimensional characteristics. In order to explore the intrinsic evolution mechanism of the changes in macroscopic phenomena such as pump-jet force, noise and cavitation, it is necessary to have a deep understanding of the fine flow field inside the pump-jet. Non-contact measurement methods such as laser Doppler velocimetry (LDV), particle image velocimetry (PIV) and laser induced fluorescence velocimetry (LIF) provide high-precision flow field data for the mechanism research of complex structures. Among them, PIV is widely used in flow field measurement due to its unique flow field visualization ability. PIV continuously shoots the uniformly scattered tracer particles illuminated by the double-pulse laser in the flow field in a very short time, and finally obtains the flow field data through the cross-correlation algorithm. In the actual measurement process, the flow field data is missing due to the obstruction of the structure, the uneven distribution of the tracer particles and the reflection of the wall. For example, during the test measurement of the internal flow field of the pump-jet, the pump-jet flow field is missing due to the limitations of the cavitation water cylinder conditions and the interference of structures such as blades and hubs on the laser. Therefore, the lack of flow field data is one of the common bottlenecks in experimental fluid mechanics. In addition, due to the complexity of the pump-jet propulsor itself, the publicly published literature at home and abroad is very limited. The current model test research focuses on the overall macroscopic performance of the pump-jet propulsor (such as thrust, torque, power, efficiency, etc.). In addition, no organization has yet measured the internal flow field characteristics of the pump-jet propulsor. The lack of internal flow field testing methods and capabilities will make it difficult to obtain the fine flow information and flow evolution characteristics inside the pump-jet, and it is even more impossible to explore the internal mechanism of changes in its macroscopic phenomena such as force, noise, and cavitation.

[0003] In summary, the current experiments on the flow field measurement of pump-jet propulsion are not perfect, and the measurement scheme is not mature, which makes it impossible to effectively compensate for the missing flow field of the pump-jet. Therefore, it is urgent to form a complete test system and missing flow field compensation method. Summary of the invention

[0004] The object of the present invention is to provide a pump jet flow field testing device and a flow field loss compensation method.

[0005] The purpose of the present invention is achieved through the following technical solutions:

[0006] A method for compensating for flow field loss in a pump jet flow field test, the specific steps are as follows:

[0007] Step 1: The acquired experimental and numerical simulation data are respectively passed through 10 convolutional layers to collect the spatial characteristic information of the flow field in the pump jet;

[0008] Step 2: Use intrinsic orthogonal decomposition to achieve flow field price reduction and obtain different modal information of the flow field;

[0009] Step 3: Establish the mapping relationship between the observation space and the state space through the Voronoi operator, fuse the modal parameters with the spatial characteristic information, and realize the assimilation of the pump jet field test data and the numerical simulation data;

[0010] Step 4: Input the flow field after data variation assimilation into the neural network, and generate a network including deformation convolution, ATN attention transfer layer, convolution layer and deconvolution layer;

[0011] Step 5: The discriminator network, i.e., the D network, distinguishes the flow field information generated by the generator network from the refined flow field calculated by CFD, so that the missing area generated by the generator G is close to the real flow field;

[0012] Step 6: Optimize the missing compensation weight of the G network; feed the loss function value calculated by the D network back to the G network to promote the back propagation of the network, repeat steps 4-5, and optimize the missing compensation effect of the flow field;

[0013] Step 7: Obtain the final pump-jet experiment and the accurate compensation flow field for numerical missing values.

[0014] Furthermore, the step 2 is specifically as follows:

[0015] The size of the flow field information obtained through convolution is f x ×f y , the flow field information is flattened according to the spatial dimension to ensure that the flattened flow field data does not lose the original flow field data structure. The flattened flow field feature snapshot matrix is ​​as follows:

[0016]

[0017] Among them, p∈[1,f x ],q∈[1,f y ],x(r p,q ,t k ) represents the physical quantity of the flow field. In the flow field inside the pump jet, the velocity along the three-dimensional space direction is mainly selected, that is, v x 、v y 、v z ;

[0018] In addition, in order to reduce the amount of data processing for flow field dimension reduction, the steady average component in the flow field space is subtracted, and only the part that deviates from the mean in the pump jet is modeled. The formula is as follows:

[0019]

[0020] A set of orthogonal bases are selected to achieve the best approximation to the original pump jet flow field data. The orthogonal base is expressed as:

[0021]

[0022] Among them, n represents the order of the orthogonal basis, that is, the order of the mode; in the process of pump jet flow field compensation, the first 60 modes are selected as the main flow field information of the flow field.

[0023] Furthermore, the step 3 is specifically as follows:

[0024] Construct a Voronoi inverse operator to map the observation space to the state space and define an observation field As a set of observation points of the pump jet flow field test data at a given location, the location of the point is defined as {(i t,k ,j t,k )}, which is expressed as follows:

[0025] (i t,k ,j t,k )∈[1,...,N y ]×[1,...,N x ]

[0026] Where k represents the number of observation points during the pump spray test, and the Voronoi operator unit R t,k Satisfies the following formula:

[0027]

[0028] Where d(·) represents the Euclidean distance, which is calculated as follows:

[0029]

[0030] Therefore, the observation space is divided into multiple Voronoi operator units and the following formula is satisfied:

[0031]

[0032] This yields:

[0033]

[0034] Substitute the obtained observation space into:

[0035]

[0036] Ultimately, the assimilation and fusion of pump-jet test data and simulation data is achieved.

[0037] Furthermore, the generating network in step 4 is designed with six layers of convolution in the encoding part, and the flow field features after data fusion are downsampled respectively to obtain the feature information of the low-dimensional space of the flow field in the pump jet; in the process of downsampling in the first two layers, deformed convolution is embedded respectively, and the sampling position of the convolution kernel is adaptively changed; the features obtained from adjacent convolution layers are input into the ATN module together; at the same time, the ATN module calculates the internal and external cosine similarity of the missing flow field area in the input feature map; the data expression of the working principle of ATN is as follows:

[0038]

[0039]

[0040] in, represents the cosine similarity of the feature map of the lth layer, p i Indicates the extraction of flow field feature values ​​from outside the i-th Mask in the feature map, p j The flow field feature value extracted from the jth Mask; ||p i || 2 and ||p j || 2 They represent the flow field characteristic values ​​p i and p j The L2 norm of Convert to attention weight After that, the missing flow field of the adjacent low-level features The missing area weight can be obtained through attention weighting, guiding the mutual fusion between high-level and low-level features, so as to achieve compensation for the missing flow field while making up for the information loss caused by convolution pooling.

[0041] Furthermore, the step 5 introduces an adversarial loss function in the network training process, as follows:

[0042]

[0043] Among them, E(*) represents the expected value of the distribution function, p d (x) represents the flow field distribution in the pump jet simulated by CFD calculation, p d (z) represents the flow field distribution generated by the generator G, and D(*) represents the output of the discriminator, whose output value ranges from 0 to 1.

[0044] A pump jet flow field testing device, characterized in that it includes a high-frequency double-pulse laser, an image processing system, a camera synchronizer, a high-speed camera, a cavitation water cylinder, a reflective prism, a pump jet model, and a transparent catheter;

[0045] The high-frequency double-pulse laser is placed at the rear end of the pump-jet model in the cavitation water cylinder, and the laser emitted by the high-frequency double-pulse laser irradiates the pump-jet model through the cavitation water cylinder; the pump-jet model is fixed inside the cavitation water cylinder through a transparent catheter, and a mutually coupled stator and rotor are installed inside the pump-jet model; a reflective prism is installed on the laser plane of the pump-jet model, and a high-speed camera is strengthened to capture image information and collect particle images; the high-speed camera is connected to an image processing system through a camera synchronizer, and the camera synchronizer is used to control the synchronous collection of the high-speed camera.

[0046] Furthermore, one side of the cavitation water cylinder is made of transparent glass, and the two high-speed cameras are placed on the transparent glass side of the cavitation water cylinder.

[0047] A computer-readable storage medium stores a computer program / instruction, which, when executed by a processor, implements the steps of a method for compensating for flow field loss in a pump jet flow field test.

[0048] A computer program product includes a computer program / instruction, which, when executed by a processor, implements the steps of a pump jet flow field test flow field missing compensation method.

[0049] An electronic device, comprising:

[0050] Memory, used to store computer programs;

[0051] A processor is used to execute the computer program to implement the instruction tracking method described in the pump jet flow field test flow field loss compensation method.

[0052] The beneficial effects of the present invention are:

[0053] The purpose of the present invention is to solve the problem that the relevant experiments of pump-jet propulsion internal flow field testing are not perfect and lack the exploration of internal fine flow field. At the same time, the present invention overcomes the problem of missing flow field data due to the obstruction of structures, uneven distribution of tracer particles and wall reflection, and proposes a pump-jet flow field testing device and flow field missing compensation method, which provides technical support for the understanding of the evolution law of internal flow field morphology of pump-jet propulsion from the experimental mechanism and the subsequent noise prediction and flow field calculation verification of pump-jet propulsion. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is an overall schematic diagram of the present invention;

[0055] Figure 2 It is a side view structural diagram of the pump spray of the present invention;

[0056] Figure 3 It is a diagram of the internal composition of the pump spray of the present invention;

[0057] Figure 4 is a flow chart of the pump jet missing flow field compensation algorithm of the present invention;

[0058] Figure 5 It is a compensation result diagram of the pump jet missing flow field of the present invention. DETAILED DESCRIPTION

[0059] The present invention is further described below in conjunction with the accompanying drawings.

[0060] according to Figures 1 to 3 The present invention discloses a pump jet flow field testing device, comprising: a high-frequency double-pulse laser 1, an image processing system 2, a camera synchronizer 3, a high-speed camera 4, a cavitation water cylinder 5, a reflecting prism 6, a pump jet model 7, and a transparent catheter 8.

[0061] The high-frequency double-pulse laser 1 is placed at the rear end of the pump-jet model 7 in the cavitation water cylinder 5. The high-frequency double-pulse laser 1 emits a laser with a wavelength of 527nm and a pulse frequency of 0.2-10kHz, which irradiates the pump-jet model 7 through the cavitation water cylinder 5. The pump-jet model 7 is internally installed with a mutually coupled stator 9 and a rotor 10, and the stator 9 is made of 6061-T6 aluminum alloy to ensure structural strength. The rotor 10 and the transparent conduit 8 are both made of transparent polymethyl methacrylate (PMMA, organic glass). When the laser passes through the pump-jet model 7, due to the mutual coupling of the stator 9 and the rotor 10 and the shielding of the structure, the flow field structure is inevitably missing. In order to capture the rich flow field information of the flow field in the pump-jet, a reflective prism 6 is installed on the laser plane of the pump-jet model 7 to enhance the image information capture of the high-speed camera 4. One side of the cavitation water cylinder 5 is made of transparent glass, which is convenient for the high-speed camera 4 to shoot the flow field. The two high-speed cameras 4 are Phantom VEO 640 A CMOS high-speed camera with a maximum resolution of 2560×1600 pixels and a maximum full-frame shooting frequency of 1400 fps is placed on the transparent glass side of the cavitation water cylinder 5 to collect particle images; the camera synchronizer 3 is externally connected to two high-speed cameras 4 and an image processing system 2 to control the high-speed cameras 4 to collect synchronously.

[0062] according to Figure 4The generative adversarial network model used in the image processing system 2 is mainly composed of a pump jet flow field data assimilation network, a generation network and a discriminator network, which is mainly used to compensate for the missing flow field of the pump jet. The pump jet flow field data assimilation network assimilates the flow field information processed by experimental measurement with the flow field information simulated by computational fluid dynamics under the same working conditions. The pump jet flow field data assimilation network is composed of a Voronoi operator, 10 convolutional layers and a proper orthogonal decomposition (POD) layer, and its algorithm steps are as follows:

[0063] Step 1: The acquired experimental and numerical simulation data are passed through 10 convolutional layers respectively to collect the spatial characteristic information of the flow field in the pump jet.

[0064] Step 2: Use intrinsic orthogonal decomposition to reduce the flow field and obtain different modal information of the flow field. The size of the flow field information obtained by convolution is f x ×f y , the flow field information is flattened according to the spatial dimension to ensure that the flattened flow field data does not lose the original flow field data structure. The flattened flow field feature snapshot matrix is ​​shown in formula (1):

[0065]

[0066] Among them, p∈[1,f x ],q∈[1,f y ],x(r p,q ,t k ) represents the physical quantity of the flow field. In the flow field inside the pump jet, the velocity along the three-dimensional space direction is mainly selected, that is, v x 、v y 、v z In addition, in order to reduce the amount of data processing for flow field dimension reduction, the steady average component in the flow field space is usually subtracted, and only the part that deviates from the mean in the pump jet is modeled, as shown in formula (2):

[0067]

[0068] A set of orthogonal bases are selected to achieve the best approximation to the original pump jet flow field data. The orthogonal base is expressed as:

[0069]

[0070] Where n represents the order of the orthogonal basis, that is, the order of the mode. In the process of pump jet flow field compensation, the first 60 modes are selected as the main flow field information of the flow field.

[0071] Step 3: Use the Voronoi operator to establish the mapping relationship between the observation space and the state space, fuse the modal parameters with the spatial feature information, and realize the assimilation of the pump jet field test data and numerical simulation data. Map the observation space to the state space by constructing a Voronoi inverse operator. Define an observation field As a set of observation points of the pump jet flow field test data at a given location, the location of the point is defined as {(i t,k ,j t,k )}, its expression is:

[0072] (i t,k ,j t,k )∈[1,...,N y ]×[1,...,N x ](4) Where k represents the number of observation points during the pump spray test, and the Voronoi operator unit R t,k Satisfies the following formula:

[0073] (5)Where d(·) represents the Euclidean distance, which is calculated as follows:

[0074]

[0075] Therefore, the observation space is divided into multiple Voronoi operator units and the following formula is satisfied:

[0076]

[0077] This yields:

[0078]

[0079] Substitute the obtained observation space into:

[0080]

[0081] Ultimately, the assimilation and fusion of pump-jet test data and simulation data is achieved.

[0082] Step 4: Input the flow field after data variation assimilation into the neural network, where the generated network consists of deformed convolution, ATN (attention transfer) layer, convolution layer and deconvolution layer. In the encoding part of the generated network, six layers of convolution are designed to sample the flow field features after data fusion, so as to obtain the feature information of the low-dimensional space of the flow field in the pump jet. In addition, considering that the shape of the pump jet rotor changes with the position during the spatial sampling of the data, and the traditional convolution is limited by the size and position of the convolution kernel, it is impossible to learn the features of objects with different deformations at different positions. Therefore, in the process of downsampling in the first two layers, deformed convolution is embedded respectively, and the sampling position of the convolution kernel is adaptively changed to adapt to the change of the rotor shape. The features obtained from adjacent convolution layers are input into the ATN module together. At the same time, the ATN module calculates the internal and external cosine similarity of the missing flow field area in the input feature map. The data expression of the working principle of ATN is as follows:

[0083]

[0084] in, represents the cosine similarity of the feature map of the lth layer, p i Indicates the extraction of flow field feature values ​​from outside the i-th Mask in the feature map, p j The flow field feature value extracted from the jth Mask. ||p i || 2 and ||p j || 2 They represent the flow field characteristic values ​​p i and p j The L2 norm of Convert to attention weight After that, the missing flow field of the adjacent low-level features The missing area weight can be obtained through attention weighting, guiding the mutual fusion between high-level and low-level features, so as to achieve compensation for the missing flow field while making up for the information loss caused by convolution pooling.

[0085] Step 5: The discriminator network (D network) guides the missing area generated by G to be as close to the real flow field as possible by distinguishing the flow field information generated by the generator network and the refined flow field calculated by CFD. The adversarial loss function is introduced in the training process as follows:

[0086]

[0087] Among them, E(*) represents the expected value of the distribution function, p d (x) represents the flow field distribution in the pump jet simulated by CFD calculation, p d(z) represents the flow field distribution generated by the generator G, and D(*) represents the output of the discriminator, whose output value ranges from 0 to 1.

[0088] Step 6: Optimize the missing compensation weight of the G network. Feed the loss function value calculated by the D network back to the G network to promote the back propagation of the network, and repeat steps 4-5 to optimize the missing compensation effect of the flow field.

[0089] Step 7: Get the final pump-jet experiment and the accurate compensation flow field of numerical missing. Compared with the standard flow field calculated by CFD, the root mean square error is less than 3%. Figure 5 The figure shows the compensation result of the missing flow field of the pump jet. Figure 5 (a) shows the pump-jet flow field after data assimilation, and (c) shows the pump-jet missing flow field after missing data. Figure 4 The algorithm flow in can realize the compensation for the missing flow field of the pump jet. Figure 5 (b) It can be seen from the compensation results that after being processed by the method flow of the present invention, the missing pump jet field can be effectively compensated.

[0090] The principle or use method of the present invention is as follows: When conducting the test, turn on the high-frequency double-pulse laser 1 and the high-speed camera 4. The high-frequency double-pulse laser 1 emits a laser with a wavelength of 527nm and a pulse frequency of 0.2 to 10kHz, which passes through the pump-jet model 7 and irradiates the flow field to be measured inside the pump-jet. The laser excites the tracer particles in the cavitation water cylinder 5 and emits reflected light. However, due to the mutual coupling between the rotor 10 and the stator 9, the flow field information is missing. At the same time, in order to make the flow field information more comprehensive, a reflective prism 6 is placed behind the pump-jet model 7. The two beams of light are captured by the high-speed camera 4 respectively. The acquired image information is transmitted to the image processing system 2 via Ethernet via the camera synchronizer 3. Finally, the self-developed pump-jet flow field missing compensation algorithm is used to achieve the fine measurement of the flow field inside the pump-jet.

[0091] The above is only a preferred specific implementation of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, shall make equivalent substitutions or changes according to the technical scheme and inventive concept of the present invention, which shall be covered within the protection scope of the present invention. The above is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for compensating for flow field loss in a pump jet flow field test, characterized in that: The specific steps are as follows: Step 1: The acquired experimental and numerical simulation data are respectively passed through 10 convolutional layers to collect the spatial characteristic information of the flow field in the pump jet; Step 2: Use intrinsic orthogonal decomposition to achieve flow field price reduction and obtain different modal information of the flow field; Step 3: Establish the mapping relationship between the observation space and the state space through the Voronoi operator, fuse the modal parameters with the spatial characteristic information, and realize the assimilation of the pump jet field test data and the numerical simulation data; Step 4: Input the flow field after data variation assimilation into the neural network, and generate a network including deformation convolution, ATN attention transfer layer, convolution layer and deconvolution layer; Step 5: The discriminator network, i.e., the D network, distinguishes the flow field information generated by the generator network from the refined flow field calculated by CFD, so that the missing area generated by the generator G is close to the real flow field; Step 6: Optimize the missing compensation weight of the G network; feed the loss function value calculated by the D network back to the G network to promote the back propagation of the network, repeat steps 4-5, and optimize the missing compensation effect of the flow field; Step 7: Obtain the final pump-jet experiment and the accurate compensation flow field for numerical missing values.

2. A pump jet flow field test flow field missing compensation method according to claim 1, characterized in that: The step 2 is specifically as follows: The size of the flow field information obtained through convolution is f x ×f y , the flow field information is flattened according to the spatial dimension to ensure that the flattened flow field data does not lose the original flow field data structure. The flattened flow field feature snapshot matrix is ​​as follows: Among them, p∈[1,f x ],q∈[1,f y ],x(r p,q ,t k ) represents the physical quantity of the flow field. In the flow field inside the pump jet, the velocity along the three-dimensional space direction is mainly selected, that is, v x 、v y 、v z ; In addition, in order to reduce the amount of data processing for flow field dimension reduction, the steady average component in the flow field space is subtracted, and only the part that deviates from the mean in the pump jet is modeled. The formula is as follows: A set of orthogonal bases are selected to achieve the best approximation to the original pump jet flow field data. The orthogonal base is expressed as: Among them, n represents the order of the orthogonal basis, that is, the order of the mode; in the process of pump jet flow field compensation, the first 60 modes are selected as the main flow field information of the flow field.

3. A pump jet flow field test flow field missing compensation method according to claim 1, characterized in that: The step 3 is specifically as follows: Construct a Voronoi inverse operator to map the observation space to the state space and define an observation field As a set of observation points of the pump jet flow field test data at a given location, the location of the point is defined as {(i t,k ,j t,k )}, which is expressed as follows: (i t,k ,j t,k )∈[1,...,N y ]×[1,...,N x ] Where k represents the number of observation points during the pump spray test, and the Voronoi operator unit R t,k Satisfies the following formula: Where d(·) represents the Euclidean distance, which is calculated as follows: Therefore, the observation space is divided into multiple Voronoi operator units and the following formula is satisfied: This yields: Substitute the obtained observation space into: Ultimately, the assimilation and fusion of pump-jet test data and simulation data is achieved.

4. A pump jet flow field test flow field missing compensation method according to claim 1, characterized in that: The generation network in step 4 is designed with six layers of convolution in the encoding part, and the flow field features after data fusion are downsampled respectively to obtain the feature information of the low-dimensional space of the flow field in the pump jet; in the process of downsampling in the first two layers, deformed convolution is embedded respectively, and the sampling position of the convolution kernel is adaptively changed; the features obtained from adjacent convolution layers are input into the ATN module together; at the same time, the ATN module calculates the internal and external cosine similarity of the missing flow field area in the input feature map; the data expression of the working principle of ATN is as follows: in, represents the cosine similarity of the feature map of the lth layer, p i Indicates the extraction of flow field feature values ​​from outside the i-th Mask in the feature map, p j The flow field feature value extracted from the jth Mask; ||p i ||2 and ||p j ||2 respectively represent the flow field characteristic value p i and p j The L2 norm of Convert to attention weight After that, the missing flow field of the adjacent low-level features The missing area weight can be obtained through weighted attention, guiding the mutual fusion between high-level and low-level features, so as to compensate for the missing flow field while making up for the information loss caused by convolution pooling.

5. A pump jet flow field test flow field missing compensation method according to claim 1, characterized in that: The step 5 introduces the adversarial loss function in the network training process, as follows: Among them, E(*) represents the expected value of the distribution function, p d (x) represents the flow field distribution in the pump jet simulated by CFD calculation, p d (z) represents the flow field distribution generated by the generator G, and D(*) represents the output of the discriminator, whose output value ranges from 0 to 1.

6. A pump jet flow field test device used in a pump jet flow field test flow field loss compensation method according to claims 1-5, characterized in that: It includes a high-frequency double-pulse laser (1), an image processing system (2), a camera synchronizer (3), a high-speed camera (4), a cavitation water cylinder (5), a reflective prism (6), a pump-jet model (7), and a transparent catheter (8); The high-frequency double-pulse laser (1) is placed at the rear end of a pump-jet model (7) in a cavitation water cylinder (5), and the laser light emitted by the high-frequency double-pulse laser (1) irradiates the pump-jet model (7) via the cavitation water cylinder (5); the pump-jet model (7) is fixed inside the cavitation water cylinder (5) via a transparent catheter (8), and a mutually coupled stator (9) and rotor (10) are installed inside the pump-jet model (7); a reflective prism (6) is installed on the laser plane of the pump-jet model (7), and a high-speed camera (4) is strengthened to capture image information and collect particle images; the high-speed camera (4) is connected to an image processing system (2) via a camera synchronizer (3), and the camera synchronizer (3) is used to control the high-speed camera (4) to collect synchronously.

7. A pump jet flow field testing device according to claim 6, characterized in that: One side of the cavitation water cylinder (5) is made of transparent glass, and the two high-speed cameras (4) are placed on the transparent glass side of the cavitation water cylinder (5).

8. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer program product comprising a computer program / instructions, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

10. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the instruction tracing method according to any one of claims 1 to 5.

Images