A water jet impact sensor and method of use thereof

Abstract

The application belongs to the technical field of sensors and particularly relates to a water jet impact sensor and a use method thereof. An LED lamp indication area is arranged on a target plate surface, a pressure sensing unit with a fence type grid node is adopted, cylindrical weights with different masses and different round bottom areas are used to drop and impact the target plate from different heights, the water jet impact target plate is simulated under different pressures, the node pressure information, the node response coordinates and the drop height data of the drop from different heights are used as training data input values, the impact pressure is theoretically calculated, the theoretical calculation impact area is used as an output value, a multi-objective fast non-dominant sorting genetic algorithm model is trained, a database is established for predicting the jet area and the effective jet pressure mean value generated by the impact, prediction values are provided for water jet impact tests at different distances, wavelet threshold processing data in a real test process is checked, and the impact area and the impact pressure mean value in different areas are given.

Description

Technical fields:

[0001] This invention belongs to the field of sensor technology and relates to a water jet impact sensor and its usage method. Based on a non-dominated sorting genetic algorithm, it can accurately give the pressure value corresponding to the water jet area. Background technology:

[0002] With the development of electronic technology, integrated circuits, and computer technology, more and more new types of sensors have emerged on the market. Among them, pressure sensors are devices or apparatuses that can sense pressure signals and convert them into usable output electrical signals according to certain rules. They typically consist of a pressure-sensitive element and a signal processing unit, and are widely used in various industrial automation environments, including water conservancy and hydropower, railway transportation, intelligent buildings, production automation, aerospace, military, petrochemicals, oil wells, power, shipbuilding, machine tools, pipelines, and many other industries. Based on different test pressure types, pressure sensors can be divided into gauge pressure sensors, differential pressure sensors, and absolute pressure sensors. In the field of water jet impact force measurement, whether it's air-to-water jets or submerged water jets, there is no pressure sensor specifically designed for their operating environment, offering high accuracy and waterproofing. Existing pressure sensors are mostly strain gauge sensors, injected with waterproof adhesive and connected to a target plate. They cannot accurately measure the water jet impact area, and the detection module can only provide a maximum monitoring value, unable to provide the effective monitoring pressure value corresponding to the effective area of ​​the jet impacting the target plate. For example, Chinese Patent 202211186042 discloses a combined sensor including a detection sensor, a pressure sensor, a waterproof and breathable membrane, a housing, and an outer shell connected to the housing. The housing forms a receiving cavity and an air vent. The detection sensor is disposed in the receiving cavity. A separator is disposed in the outer shell, which divides the inner cavity of the outer shell into a first receiving cavity and a second receiving cavity that are isolated from each other. The air vent connects the first cavity and the receiving cavity. The first cavity forms a first opening that communicates with the outside, and the second cavity forms a second opening that communicates with the outside. The waterproof and breathable membrane is configured to prevent water from the outside from entering the receiving cavity through the first opening. The pressure sensor is at least partially disposed in the second cavity, and at least part of the pressure sensor disposed in the second cavity is wrapped with a waterproof adhesive layer to isolate it from the outside.Chinese Patent 202210484894 discloses a lifting-type pressure sensor with a waterproof structure, including a lifting mechanism. The top of the lifting mechanism has a first protective mechanism, and the outer wall of the first protective mechanism has a second protective mechanism. The lifting mechanism includes a first support rod, and the outer wall of the first support rod is movably sleeved with a second support rod. The first protective mechanism includes a base plate, the bottom of which is fixedly equipped with a spherical hinge seat for connecting to the first support rod. The top of the base plate is fixedly equipped with a support column, and the outer wall of the support column is fixedly equipped with multiple inflatable airbags arranged in an I-shape. The second protective mechanism... The mechanism includes a waterproof housing, inside which a pressure sensor is fixedly installed. A control board for connecting to a support column is fixedly connected to the bottom of the pressure sensor, and a connector for connecting to the waterproof housing is fixedly installed on the outer wall of the control board. A pressure pad is fitted onto the top of the waterproof housing, and multiple pressure rods for connecting to the pressure sensor are fixedly installed in a ring at equal intervals at the bottom of the pressure pad. Multiple second pressure plates for compressing the inflatable airbag are arranged in a ring at equal intervals on one side of the inner wall of the waterproof housing. A waterproof rubber pad for connecting to the support column is fixedly installed at the bottom of the waterproof housing. Chinese Patent 202221405074 discloses a multi-purpose pressure sensor, comprising a housing with an internal cavity and a pressure inlet hole at the top, the pressure inlet hole being connected to the cavity. The sensor is characterized by a substrate fixedly mounted at the bottom of the housing, on which a first pressure-sensitive chip and a second pressure-sensitive chip are mounted. The substrate also has a first reference hole and a second reference hole. The first reference hole is connected to the low-pressure cavity of the first pressure-sensitive chip, and the second reference hole is connected to the low-pressure cavity of the second pressure-sensitive chip. The high-pressure cavities of both the first and second pressure-sensitive chips are connected to the pressure inlet hole. However, in the process of measuring the impact force of a water jet, the same test is often repeated multiple times, resulting in significant differences in the values ​​measured by strain gauge sensors. Therefore, it is necessary to develop and design a water jet impact sensor and its usage method, which can provide pressure feedback based on the water jet impact area, offering positive social and economic benefits. Summary of the Invention:

[0003] The purpose of this invention is to overcome the shortcomings of the existing technology and to develop and design a water jet impact sensor and its usage method to effectively monitor the magnitude of the impact force during the water jet impact test.

[0004] The main structure of the water jet impact sensor involved in this invention includes a sensor substrate and a target plate disposed thereon, as well as an LED indicator area and a pressure sensing unit with a grid-like node disposed on the surface of the target plate.

[0005] The target plate involved in this invention is made of transparent acrylic; the monitoring nodes of the fence-type grid nodes are arranged in a rectangular grid to form a fence, and correspond to the coordinate information of the impact area.

[0006] The water jet impact sensor involved in this invention is applied in a water jet laboratory for water jet impact force testing (including air-water jet impact testing and submerged water jet impact testing). When the water jet generated by the water jet impacts the target plate surface, the LEDs in the LED indicator area impacted by the water jet light up. The location of the water jet impact on the target plate is determined by the area where the LEDs light up. Based on the node coordinate information corresponding to the impact surface of the target plate, the area and pressure value of the water jet impact are obtained after wavelet thresholding. The specific process is as follows:

[0007] The first step is to filter the monitoring point data based on the coordinates of the response nodes. Based on the MATLAB coordinate search method, the geometric coordinates of each response node are unique. When the pressure sensing unit has a data response, the response nodes are grouped into a subset according to their adjacent geometric coordinates or adjacent values ​​with a difference of less than 0.5. The coordinate subsets are created sequentially from the outside to the inside according to the MATLAB coordinate method.

[0008] The second step is to use wavelet threshold denoising to process the values ​​of each subset of monitoring points, and determine whether the difference between the monitoring point values ​​is less than 0.5, so as to remove sharp points.

[0009] The third step is to connect the values ​​of each subset of monitoring points sequentially to form multiple closed regions;

[0010] The fourth step is to input the coordinate information generated by the impact, the monitoring values, the impact target distance, and the initial jet pressure into the prediction model.

[0011] Step 5: Based on the prediction model, give the maximum area of ​​impact and the average impact force.

[0012] Compared with existing technologies, this invention sets up an LED indicator area on the target plate surface, employs a pressure sensing unit with a fence-like grid node, and simulates water jet impacts on the target plate by dropping cylindrical weights of different masses and different bottom areas from different heights. The node pressure information, node response coordinates, and drop height data from different heights are used as training data inputs to theoretically calculate the impact pressure. The theoretically calculated impact area is then used as the output value and fed into a multi-objective fast non-dominated sorting genetic algorithm model for training. This establishes a database for predicting the jet region and the average effective jet pressure generated by the impact, providing predicted values ​​for water jet impact tests at different distances. The wavelet threshold processing data from the actual test process is verified, and the impact area and average impact pressure for different regions are given. Attached image description:

[0013] Figure 1This is a schematic diagram of the main structure of the water jet impact sensor involved in this invention.

[0014] Figure 2 This is a schematic diagram of the air-water jet impact test involved in the present invention.

[0015] Figure 3 This is a schematic diagram of the submerged water jet impact test involved in the present invention.

[0016] Figure 4 This is a schematic diagram of drop simulation calibration of the water jet impact sensor involved in the present invention.

[0017] Figure 5 This is a schematic diagram of the drop simulation data processing flow of the water jet impact sensor involved in this invention.

[0018] Figure 6 This is a schematic diagram of the operation process of the water jet impact sensor involved in this invention. Detailed implementation method:

[0019] The present invention will be further described below with reference to the embodiments and accompanying drawings.

[0020] Example 1:

[0021] The main structure of the water jet impact sensor involved in this embodiment includes a sensor base 1, a target plate 2, a pressure sensing unit 3, a yellow LED light area 4, a green LED light area 5, and a red LED light area 6. The surface of the sensor base 1 is provided with a circular target plate 2, and the inside is provided with a pressure sensing unit 3 with a grid-like node. The surface of the target plate 2 is provided with a yellow LED light area 4, a green LED light area 5, and a red LED light area 6 from the center to the circumference.

[0022] In this embodiment, the yellow LED light area 4, the green LED light area 5, and the red LED light area 6 each occupy 1 / 3 of the area of ​​the target plate 2, and the three together constitute the LED light indicator area.

[0023] The water jet impact sensor involved in this embodiment is as follows: Figure 2 During the air-water jet impact test, after the water jet impacts the surface of the target plate 2, the impact angle of the water jet is adjusted according to the display in the LED indicator area to ensure the accuracy of the measurement results. Specifically, when the water jet impacts the center area of ​​the target plate, the yellow light illuminates; when the water jet impacts the edge area of ​​the target plate, the red light illuminates; and when the water jet impacts the area between the center area and the edge area, the green light illuminates.

[0024] The water jet impact sensor involved in this embodiment is as follows: Figure 3During the submerged water jet impact test shown, a high-pressure water jet is ejected from the nozzle, passes through water to a set depth, and reaches the surface of the target plate 2. Due to the presence of water at a set depth, the high-pressure water jet will deflect. The jet direction of the high-pressure water jet is corrected according to the display of the LED indicator area.

[0025] The calibration process for the water jet impact sensor involved in this embodiment includes the following steps:

[0026] Step 1: Conduct drop simulation impact tests using cylindrical weights of different masses (0.5kg, 1kg, 1.5kg, 2kg, etc.) and different base areas. Drop the impact sensor from different heights to obtain a series of nodal information, including nodal pressure, nodal coordinates, drop height, and theoretical impact force at the time of drop. Figure 4 The nodes A1, A2, A3...A11, A12, A13 shown are based on their coordinates. Taking the innermost node information as an example, that is... Figure 4 In the machine learning boundary 1, the circle of effective contact node information is based on the MATLAB coordinate method. The coordinates of the response node are unique. By traversing and searching the coordinates of the response nodes, according to the principle of adjacent coordinates, data with monitoring pressure information less than or equal to one are divided into a subset, and so on, several subsets are divided.

[0027] The second step is to use wavelet threshold denoising to process the pressure values ​​of each subset of monitoring, remove bad coordinate points including sharp corners, and determine whether the pressure difference between monitoring points in the same subset is less than 0.5. If it is less than 0.5, continue to the third step. This is because during the weight drop simulation test, there are noise points with large data deviations in the monitoring nodes, and the monitoring node positions are far off or beyond the drop area. Otherwise, return to the second step.

[0028] The third step involves wavelet threshold analysis to connect the coordinates of monitoring points in the same subset into a closed region, forming several closed regions from the inside out, and providing the average pressure value corresponding to each closed region.

[0029] The fourth step involves inputting the monitoring node coordinates, node pressure values, and drop height, and the actual weight bottom area and theoretical impact force of the weight as output values ​​into the training model. After wavelet thresholding, this data is used as the model training set. The model employs the NSGA-II fast non-dominated sorting genetic algorithm with an elite strategy, a multi-objective genetic algorithm.

[0030] During the drop impact sensor process, the impact range is confirmed based on a neural network algorithm, and the impact boundary is determined based on monitoring module data. Based on the actual impact situation, the boundary of the drop impact is a regular circle. The effective boundary and the effective data nodes cannot completely overlap. When a weight with a set area S falls onto the surface of the impact sensor, the data monitoring nodes respond as A1, A2, A3...A10, A11, A12. At this time, the data monitoring nodes are connected in sequence to form a region, which differs from the boundary area formed by the weight falling onto the sensor surface. Wavelet threshold noise reduction is performed on the data of each monitoring node, and a threshold is set to ensure that the difference between the outermost numerical values ​​is less than 0.5, so that the outermost monitoring data points ultimately form a closed region after filtering.

[0031] The water jet impact sensor involved in this embodiment operates as follows:

[0032] First, a mass of 3.5 kg and a cross-sectional area of ​​9π cm² are considered. 2 The weight was dropped from a height of 10cm onto the pressure sensing unit 3. A series of coordinate points were monitored through the grid nodes of the pressure sensing unit 3. Each grid node of the pressure sensing unit 3 has corresponding geometric coordinates. The coordinate values ​​of the monitored nodes after the drop are as follows: A35 (3.51kg), A34 (3.45kg), A30 (3.98kg), A32 (3.45kg), A33 (3.22kg), A31 (3.44kg), A29 (3.85kg), A27 (3.19kg), A25 (2.84kg), A24 (2.68kg). kg), A23 (2.56kg), A22 (2.51kg), A21 (2.68kg), A20 (2.41kg), A19 (2.20kg), A18 (2.05kg), A17 (1.68kg), A16 (2.08kg), A15 (1.45kg), A14 (1.12kg), A13 (1.38kg), A12 (0.95kg), A10 (1.01kg), A09 (0.85kg), A07 (0.82kg), A03 (0.70kg), A01 (0.52kg);

[0033] Then, using MATLAB coordinate search, based on the principle of adjacent coordinate values ​​of each response node, data with monitoring pressure information less than or equal to one are divided into a subset. This is repeated to divide the data into several subsets. Based on the set error standard, the corresponding coordinate values ​​of the response nodes are divided into the following four subsets: M1(A35, A34, A30, A32, A33, A31, A29, A27), M2(A25, A24, A23, A21, A22, A21, A20, A18, A19), M3(A17, A16, A15, A14, A13), M4(A12, A11, A10, A08, A09, A07, A03, A01).

[0034] Secondly, noise points in the subset are removed using wavelet thresholding in MATLAB. By setting the threshold in wavelet filtering, noise points are filtered out, ensuring that the values ​​of all monitoring nodes in the subset are within a set range. The coordinates of the monitoring points in the same subset are then connected sequentially to form a closed region, from the inside out, resulting in four closed regions S1(8.9πcm). 2 S2(7.1πcm) 2 S3(6.2πcm) 2 ) and S4(4πcm 2 The average impact forces F1 (3.45 kg), F1 (2.74 kg), F1 (1.24 kg), and F1 (0.85 kg) corresponding to the closed area are given.

[0035] Next, the monitoring node coordinates, node pressure values, and drop height processed by wavelet thresholding are used as input values, and the actual weight bottom area and theoretical weight impact force are used as output values. These are then fed into the training model and, after wavelet thresholding, are used as the model training set data. The training model uses the NSGA-II fast non-dominated sorting genetic algorithm with elite strategy and a multi-objective genetic algorithm.

[0036] In sequence, each has a mass of 3 kg and a cross-sectional area of ​​9π cm². 2 Weights with a mass of 2 kg and a cross-sectional area of ​​6π cm² were dropped from 15 cm, 20 cm, and 25 cm respectively. 2 The weights were dropped from 10cm and 15cm to the pressure sensing unit 3, respectively. The above steps were repeated to form a training database with 100 subsets. The coordinates of the monitoring node, the node pressure value, and the drop height were used as input values, and the actual bottom area of ​​the weight and the theoretical impact force of the weight were used as output values, which were then fed into the training model.

[0037] A material with a mass of 5 kg and a cross-sectional area of ​​10π cm² is used. 2The weight is dropped from 10cm to the pressure sensing unit 3. The values ​​detected by the grid nodes are fed into the prediction model. The values ​​output by the prediction model are compared with the real data (real bottom area, drop impact force). If there is a deviation in the data, the error is controlled to be minimized by adjusting the number of population iterations.

[0038] The predictive model was applied to the water jet impact test. When the water jet impacted the target plate 2, the jet angle was adjusted according to the color of the light displayed in the LED indicator area, so that the water jet hit the target center perpendicularly. Wavelet threshold noise reduction was performed on the response node data of the water jet impact, and the region was divided using the MATLAB coordinate method.

[0039] The first step is to filter the monitoring point data based on the coordinates of the response nodes. Data with a difference of less than 0.5 are grouped into a subset. Subsets are created sequentially from the outside to the inside using the MATLAB coordinate method.

[0040] The second step is to use wavelet threshold denoising to process the values ​​of each subset of monitoring points, remove sharp points, and determine whether the difference between the values ​​of the outermost monitoring points is less than 0.5. If not, repeat the above operation.

[0041] The third step is to connect the values ​​of each subset of monitoring points sequentially to form multiple closed area regions.

[0042] The fourth step is to input the coordinate information and monitoring values ​​generated by the water jet impact into the prediction model;

[0043] Step 5: Based on the prediction model, provide the area and average impact force of each impact subset.

Claims

1. A water jet impact sensor, characterized in that, The main structure includes a sensor substrate and a target plate mounted on it, as well as an LED indicator area and a pressure sensing unit with a grid-like node on the surface of the target plate. The specific process for using it is as follows: The first step is to filter the monitoring point data based on the coordinates of the response nodes. Based on the MATLAB coordinate search method, the geometric coordinates of each response node are unique. When the pressure sensing unit has a data response, the response nodes are grouped into a subset according to their adjacent geometric coordinates or adjacent values ​​with a difference of less than 0.

5. The coordinate subsets are created sequentially from the outside to the inside according to the MATLAB coordinate method. The second step is to use wavelet threshold denoising to process the values ​​of each subset of monitoring points, and determine whether the difference between the monitoring point values ​​is less than 0.5, so as to remove sharp points. The third step is to connect the values ​​of each subset of monitoring points sequentially to form multiple closed regions; The fourth step is to input the coordinate information generated by the impact, the monitoring values, the impact target distance, and the initial jet pressure into the prediction model. Step 5: Based on the prediction model, give the maximum area of ​​impact and the average impact force.

2. The water jet impact sensor according to claim 1, characterized in that, The surface of the sensor substrate is provided with a circular target plate, and the inside is provided with a pressure sensing unit with a grid-like node; the surface of the target plate is provided with yellow LED light area, green LED light area and red LED light area from the center to the circumference.

3. The water jet impact sensor according to claim 2, characterized in that, The yellow LED light area, green LED light area, and red LED light area each occupy 1 / 3 of the target plate area, and the three together form the LED light indicator area.

4. The water jet impact sensor according to any one of claims 1-3, characterized in that, The target plate is made of transparent acrylic; the monitoring nodes of the fence-style grid nodes are arranged in a rectangular grid to form a fence, and correspond to the coordinate information of the impact area.

5. The water jet impact sensor according to claim 4, characterized in that, In the water jet laboratory, it is used for water jet impact force testing. When the water jet generated by the water jet impacts the target plate surface, the LED lights in the LED indicator area that is impacted by the water jet light up. The location of the water jet impact on the target plate is determined by the area where the LED lights are lit. Based on the node coordinate information corresponding to the impact surface of the target plate, the area and pressure value of the water jet impact are obtained after wavelet thresholding.

6. The water jet impact sensor according to any one of claims 1-3, characterized in that, During the air-water jet impact test, after the water jet impacts the surface of the target plate, the impact angle of the water jet is adjusted according to the display in the LED indicator area to ensure the accuracy of the measurement results. Specifically, when the water jet impacts the center area of ​​the target plate, the yellow light illuminates; when the water jet impacts the edge area of ​​the target plate, the red light illuminates; and when the water jet impacts the area between the center and the edge area, the green light illuminates.

7. The water jet impact sensor according to any one of claims 1-3, characterized in that, During the submerged water jet impact test, a high-pressure water jet is ejected from the nozzle, passes through water at a set depth, and reaches the surface of the target plate. Due to the presence of water at the set depth, the high-pressure water jet will deflect. The jet direction of the high-pressure water jet is corrected according to the display in the LED indicator area.

8. The water jet impact sensor according to claim 4, characterized in that, The calibration process includes the following steps: The first step is to conduct a drop simulation impact test using cylindrical weights of different masses and bottom areas. The impact sensor is dropped from different heights to obtain a series of node information, including node pressure, node coordinates, drop height, and theoretical impact force at the time of drop. Based on the node coordinates, using the MATLAB coordinate method, the coordinates of the response node are unique. By traversing and searching the response node coordinates, according to the principle of adjacent coordinates, data with monitoring pressure information less than or equal to one are divided into a subset, and so on, several subsets are divided. The second step is to use wavelet threshold denoising to process the pressure values ​​of each subset of monitoring, remove bad coordinate points including sharp corners, and determine whether the pressure difference between monitoring points in the same subset is less than 0.

5. If it is less than 0.5, continue to the third step. This is because during the weight drop simulation test, there are noise points with large data deviations in the monitoring nodes, and the monitoring node positions are far off or beyond the drop area. Otherwise, return to the second step. The third step involves wavelet threshold analysis to connect the coordinates of monitoring points in the same subset into a closed region, forming several closed regions from the inside out, and providing the average pressure value corresponding to each closed region. The fourth step involves inputting the monitoring node coordinates, node pressure values, and drop height, and the actual weight bottom area and theoretical impact force of the weight as output values ​​into the training model. After wavelet thresholding, this data is used as the model training set. The model employs the NSGA-II fast non-dominated sorting genetic algorithm with an elite strategy, a multi-objective genetic algorithm.

9. The water jet impact sensor according to claim 8, characterized in that, At runtime: First, a mass of 3.5 kg and a cross-sectional area of ​​9π cm² are considered. 2 The weight is dropped from 10cm to the pressure sensing unit. A series of coordinate points are monitored through the grid nodes of the pressure sensing unit. Each grid node of the pressure sensing unit has a corresponding geometric coordinate, and the coordinate values ​​of the response node after the drop are obtained. Then, using MATLAB coordinate search, based on the principle of adjacent coordinate values ​​of each response node, data with monitoring pressure information less than or equal to one are divided into a subset, and so on, to divide into several subsets. Based on the set error standard, the corresponding coordinate values ​​of the response nodes are divided into several subsets. Secondly, noise in the subset is removed by using MATLAB wavelet thresholding. By setting the threshold in wavelet filtering, noise is filtered out, so that the values ​​of the monitoring nodes in the entire subset are within the set range. The data coordinates of the monitoring points in the same subset are connected sequentially to form a closed region. From the inside to the outside, several closed regions are formed, and the average impact force corresponding to the closed region is given. Next, the monitoring node coordinates, node pressure values, and drop height processed by wavelet thresholding are used as input values, and the actual weight bottom area and theoretical weight impact force are used as output values. These are then fed into the training model and, after wavelet thresholding, are used as the model training set data. The training model uses the NSGA-II fast non-dominated sorting genetic algorithm with elite strategy and a multi-objective genetic algorithm. In sequence, each has a mass of 3 kg and a cross-sectional area of ​​9π cm². 2 Weights with a mass of 2 kg and a cross-sectional area of ​​6π cm² were dropped from 15 cm, 20 cm, and 25 cm respectively. 2 The weights were dropped from 10cm and 15cm to the pressure sensing unit, respectively. The above steps were repeated to form a training database with 100 subsets. The coordinates of the monitoring node, the node pressure value, and the drop height were used as input values, and the actual bottom area of ​​the weight and the theoretical impact force of the weight were used as output values, which were then fed into the training model. A material with a mass of 5 kg and a cross-sectional area of ​​10π cm² is used. 2 The weight is dropped from 10cm to the pressure sensing unit, and the value detected by the grid node is fed into the prediction model. The value output by the prediction model is compared with the real data. If there is a deviation in the data, the error is controlled to be minimized by adjusting the number of population iterations. The predictive model was applied to a water jet impact test. When the water jet impacted the target plate, the jet angle was adjusted according to the color of the light displayed in the LED indicator area, so that the water jet hit the target center perpendicularly. Wavelet threshold noise reduction was performed on the response node data of the water jet impact, and the region was divided using the MATLAB coordinate method. The first step is to filter the monitoring point data based on the coordinates of the response nodes. Data with a difference of less than 0.5 are grouped into a subset. Subsets are created sequentially from the outside to the inside using the MATLAB coordinate method. The second step is to use wavelet threshold denoising to process the values ​​of each subset of monitoring points, remove sharp points, and determine whether the difference between the values ​​of the outermost monitoring points is less than 0.

5. If not, repeat the above operation. The third step is to connect the values ​​of each subset of monitoring points sequentially to form multiple closed area regions. The fourth step is to input the coordinate information and monitoring values ​​generated by the water jet impact into the prediction model; Step 5: Based on the prediction model, provide the area and average impact force of each impact subset.

Images