CN117554217A - Puncture experiment execution and puncture data acquisition and analysis method and device - Google Patents
Puncture experiment execution and puncture data acquisition and analysis method and device Download PDFInfo
- Publication number
- CN117554217A CN117554217A CN202311670696.4A CN202311670696A CN117554217A CN 117554217 A CN117554217 A CN 117554217A CN 202311670696 A CN202311670696 A CN 202311670696A CN 117554217 A CN117554217 A CN 117554217A
- Authority
- CN
- China
- Prior art keywords
- puncture
- experiment
- data
- cutter
- layers
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000002474 experimental method Methods 0.000 title claims abstract description 91
- 238000004458 analytical method Methods 0.000 title claims abstract description 22
- 238000000034 method Methods 0.000 claims abstract description 52
- 230000035515 penetration Effects 0.000 claims abstract description 30
- 238000012360 testing method Methods 0.000 claims abstract description 22
- 238000013528 artificial neural network Methods 0.000 claims abstract description 19
- 239000004973 liquid crystal related substance Substances 0.000 claims abstract description 6
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 claims description 10
- 229910052782 aluminium Inorganic materials 0.000 claims description 10
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 claims description 9
- 238000012544 monitoring process Methods 0.000 claims description 9
- 229910000831 Steel Inorganic materials 0.000 claims description 8
- 239000010959 steel Substances 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 8
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 238000006073 displacement reaction Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012935 Averaging Methods 0.000 claims description 3
- 101100194362 Schizosaccharomyces pombe (strain 972 / ATCC 24843) res1 gene Proteins 0.000 claims description 3
- 101100194363 Schizosaccharomyces pombe (strain 972 / ATCC 24843) res2 gene Proteins 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 238000011176 pooling Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000007619 statistical method Methods 0.000 claims description 2
- 238000013480 data collection Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 4
- 229920006231 aramid fiber Polymers 0.000 description 2
- 239000002131 composite material Substances 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000011151 fibre-reinforced plastic Substances 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N3/00—Investigating strength properties of solid materials by application of mechanical stress
- G01N3/30—Investigating strength properties of solid materials by application of mechanical stress by applying a single impulsive force, e.g. by falling weight
- G01N3/303—Investigating strength properties of solid materials by application of mechanical stress by applying a single impulsive force, e.g. by falling weight generated only by free-falling weight
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/28—Recognition of hand or arm movements, e.g. recognition of deaf sign language
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Chemical & Material Sciences (AREA)
- Immunology (AREA)
- Biochemistry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Analytical Chemistry (AREA)
- Pathology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Investigating Strength Of Materials By Application Of Mechanical Stress (AREA)
Abstract
The invention discloses a puncture experiment execution and puncture data acquisition and analysis method and device, comprising the following steps: firstly, fixing a puncture sample on a drop hammer testing machine for preparing a puncture experiment; the hand gesture of the experimenter is monitored and warned when the experimenter works through the neural network; after the alarm is kept in a silent state, performing a puncture experiment, collecting image information in the puncture process, and key parameters such as initial puncture kinetic energy, maximum penetration layer number, puncture peak force and the like, and displaying a single-time data collection result in a liquid crystal display screen; through the gradient experiment setting, the experiment is repeated, and the change rule of the puncture process parameters when the cutter falls down differently is obtained. The method and the device provided by the invention are used for carrying out experimental study, so that the risk in the puncture experiment process can be effectively avoided, and more comprehensive data acquisition can be obtained.
Description
Technical Field
The invention belongs to the fields of machine learning, computer vision and impact dynamics, and particularly relates to a puncture experiment execution and puncture data acquisition and analysis method and device.
Background
Currently, puncture tests typically rely on manual operations, which present a significant risk factor. Puncture experiments are dangerous experimental operation tasks, and key parameters such as initial impact kinetic energy accurate control, puncture peak force measurement, maximum penetration layer number determination and the like are generally required.
The machine vision technology combines an image processing method and a deep learning method, and can automatically identify and analyze key features in the experimental process. Through the use of machine vision technology, possible dangerous situations such as the situation that the hand gesture of staff is on the drop weight testing machine base when carrying out the experiment can be automatically detected, and real-time supervision and automatic warning can reduce the risk when the manual operation experiment. In addition, the existing puncture experiment data acquisition method has limitations, and key parameters are gradually adjusted, so that the method has important significance in comprehensively knowing the regularity of the key parameters.
Therefore, a puncture experiment execution and puncture data acquisition and analysis method is needed, the experiment safety and the data quality are improved, and more reliable support is provided for the research and the application in the field; the method changes the mode in the field of puncture experiments, so that the method is safer and highly automated, and the data acquisition and analysis are more comprehensive.
Disclosure of Invention
The invention aims to provide a puncture experiment execution and puncture data acquisition and analysis method and device, so as to solve the problems in the prior art.
In order to achieve the above object, the present invention provides a puncture experiment execution and puncture data acquisition and analysis method, including:
step one: performing puncture experiment preparation work, detecting working states of workers by adopting an image recognition neural network in the process, judging whether the preparation work is finished or not according to the working states, and performing a second step after judging the completion of the preparation work; wherein, the preparation work includes: fixing a puncture sample on a base of a drop hammer testing machine, mounting a cutter on an impactor, keeping a vertical angle between the cutter and the puncture sample, and pulling the cutter to a certain height;
step two: performing a puncture experiment and collecting puncture data, and collecting a puncture sample image after single puncture is completed; importing the puncture data and the puncture sample image into a computer and displaying the puncture data and the puncture sample image;
step three: and (3) changing puncture experiment parameters, performing multiple experiments according to the second step to obtain a plurality of groups of puncture experiment data, and analyzing the plurality of groups of puncture experiment data by adopting a statistical method to obtain a puncture data change rule.
Optionally, the first step further includes constructing and training an image recognition neural network, where the image recognition neural network is a convolutional neural network, and includes a basic convolutional network and an additional convolutional layer; the basic convolution network comprises 12 convolution layers, 4 maximum pooling layers and 1 full connection layer; the additional convolution layers comprise two residual network modules and 3 full connection layers, wherein the residual network modules comprise two convolution layers with a convolution kernel of 1x1 and a step length of 1, one convolution layer with a convolution kernel of 3x3 and a step length of 2, and one convolution layer with a convolution kernel of 1x1 and a step length of 2.
Optionally, the training the image recognition neural network in the first step includes: acquiring hand gesture images of a plurality of experimenters in preparation, wherein the hand gesture images have different sizes; dividing hand gesture images with different sizes into positive samples and negative samples based on anchor frames with different sizes; and designing a loss function, and training the image recognition neural network based on the positive sample and the negative sample and the loss function.
Optionally, the loss function is as follows:
wherein p is the predicted size of the anchor frame, t is the actual size of the target, N is the number of samples, res1 is the error of the first residual error module, res2 is the error of the second residual error module, L conf For confidence error, L loc Is a position error;
the error calculation formula of the residual error module is as follows:
where res_pred is the predicted value of each residual block and res_true is the input true value of each residual block.
Optionally, the process of collecting the puncture data in the second step includes: the force sensor is arranged between the bottom of the cutter and the impactor, the puncture sample comprises a plurality of layers, aluminum films are paved in the plurality of layers of puncture samples at fixed layer number intervals, and the aluminum films are connected with the derivative acquisition instrument; when a puncture experiment is carried out, the puncture impact force is obtained through the force sensor, and the number of penetration layers is obtained based on the aluminum film and the data acquisition instrument.
Optionally, the step three changes the puncture experiment parameters, and the process of performing multiple experiments according to the step two includes: setting the falling height range of the cutter, setting the height of the experiment interval by adopting a gradient method, carrying out multiple experiments, and carrying out repeated experiments by adopting different samples at the same height.
Optionally, the process of analyzing the plurality of sets of puncture experimental data in the third step includes: and detecting abnormal values of a plurality of groups of puncture experiment data by adopting a Z-score method, removing the abnormal values according to the deviation degree of Z score values, and averaging the puncture experiment data to obtain the corresponding relation of the initial impact kinetic energy, the puncture impact force peak value and the number of penetration layers of the cutter and different puncture sample images.
The invention also provides a puncture experiment execution and puncture data acquisition and analysis device, which is characterized by comprising:
the puncture task execution module is used for executing puncture experiments of puncture samples by vertically falling down the cutter from different heights, and comprises a drop hammer testing machine, an impactor and the cutter;
the real-time hazard monitoring module is used for carrying out real-time image acquisition and analysis on the puncture experiment area, judging whether staff operate, if so, playing warning sound, otherwise, keeping a silence state;
the puncture data acquisition module is used for acquiring the number of penetration layers, the puncture impact force and the puncture sample image in the puncture experiment process;
the puncture data display module is used for displaying the number of penetration layers in a single puncture experiment, a puncture impact force peak value and a puncture sample image in the liquid crystal screen;
the processor operation module is used for analyzing and processing the number of penetration layers and the puncture impact force acquired by the data acquisition instrument and the oscilloscope respectively to obtain the number of penetration layers and the puncture impact force peak value of the cutter;
and the storage module is used for storing puncture experiment data in the process of performing the puncture experiment.
Optionally, the puncture task execution module comprises a vertical framework of the drop hammer testing machine, wherein the vertical framework is formed by 4 standard steel pipes with the length of 210 cm, and the horizontal direction of the drop hammer testing machine is fixed by 12 steel pipes with the length of 80 cm; a clamping tool of a cutter and a force sensor for monitoring the impact force of the cutter are arranged below the impactor; and a displacement sensor is arranged at a certain distance from the bottom of the drop weight testing machine.
The invention has the technical effects that:
according to the invention, a puncture sample is fixed on a drop hammer testing machine for preparing a puncture experiment; the neural network method for real-time hazard monitoring provided by the invention realizes the monitoring and warning of the hand gesture of the experimenter during working; after the alarm is kept in a silent state, a puncture experiment is carried out, and image information, puncture initial kinetic energy, maximum penetration layer number, puncture peak force and other key parameters in the puncture process are acquired. And displaying the data acquisition result in a liquid crystal display screen. Through the gradient experiment setting, the experiment is repeated, and the change rule of the puncture process parameters when the cutter falls down differently is obtained. The method and the device provided by the invention are used by technicians and experimenters in the related field to develop experimental study, so that the risk in the puncture experiment process can be effectively avoided, and more comprehensive data acquisition can be obtained.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application, illustrate and explain the application and are not to be construed as limiting the application. In the drawings:
FIG. 1 is a schematic diagram of a drop weight tester used for performing a puncture experiment in an embodiment of the present invention;
FIG. 2 is a schematic diagram of a neural network according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of data acquisition during lancing according to an embodiment of the present invention;
FIG. 4 is a schematic diagram showing the data of the puncturing process according to the embodiment of the invention;
FIG. 5 is a graph of an analysis of the initial velocity of a lancing process tool in an embodiment of the present invention;
FIG. 6 is an analysis of the number of maximum penetration layers of a lancing process according to an embodiment of the present invention;
FIG. 7 is a graph of peak puncture force analysis during the puncture process according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of a lancing test performance and lancing data acquisition, analysis device in an embodiment of the present invention;
FIG. 9 is a flow chart of a puncture test performed in an embodiment of the present invention;
reference numerals:
101. a camera; 102. a bracket; 103. an impactor; 104. a displacement sensor; 105. a cutter; 106. and (5) a base.
Detailed Description
It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.
Example 1
The embodiment provides a method for executing a puncture experiment, which comprises the following steps:
firstly, puncture experiment preparation work is carried out. The puncture sample was fixed to the base of the drop hammer tester as shown in fig. 1. The main components of the drop hammer tester mainly comprise a camera 101, a bracket 102, an impactor 103, a displacement sensor 104, a cutter 105 and a base 106. The tool is mounted to the impactor. The knife is kept at a vertical angle with the puncture sample and pulled up to a certain height. And waiting for the alarm to be in a silent state, and preparing for the puncture experiment.
And 2, monitoring the danger in real time. And (3) carrying out real-time image acquisition on the puncture experiment area through a camera fixed above the drop hammer tester, wherein the frame rate per second is set to be between 60 and 100 images. The acquired image data is analyzed using a neural network. Fig. 2 shows a schematic structural diagram of the neural network, after an image acquired by a camera in real time is input into the neural network, whether the hand gesture and the hand gesture status of a worker exist in the image or not is detected, and if the hand gesture of an experimenter in working is detected in the image, an alarm is used for continuously playing an alarm; otherwise, the silence state will be maintained.
The neural network is a multi-layer convolutional neural network for extracting image features, and mainly comprises a basic convolutional network and additional convolutional layers. The base convolutional network contains 12 convolutional layers, 4 max pooling layers and 1 fully-connected layer. The additional convolution layer mainly comprises two residual network modules and 3 full connection layers, wherein the residual network modules respectively comprise two convolution layers with 1x1 convolution kernel and 1 step length and one convolution layer with 3x3 convolution kernel and 2 step length, and besides, one convolution layer with 1x1 convolution kernel and 2 step length is used as a residual block. Each of the additional convolution layers may be classified and regressed as an effective receptive field for network object detection.
Of neural network algorithmsThe training process is as follows: firstly, 200 images containing hand gestures of different sizes of experimenters during working are acquired through a camera fixed above a drop hammer tester. The hand gestures of different sizes are processed through anchor frames of different sizes. The size of the anchor frame is well matched with the hand gesture in the image to be used as a positive sample, and the matching effect is poor to be used as a negative sample. The positive and negative samples constitute a dataset to train the neural network. The loss function selected in the training process is:wherein p is the predicted size of the anchor frame, t is the actual size of the target, N is the number of samples, res1 is the error of the residual block 1, and res2 is the error of the residual block 2. L (L) conf For confidence error, L loc Is a position error. The calculation formula of the two residual error module errors is as follows:where res_pred is the predicted value of each residual block and res_true is the input true value of each residual block.
Step 3 puncture experiments were performed. And after the alarm is silent, executing the puncture task. And releasing the impactor and the cutter, and finishing a single experiment after the cutter is approximately free-falling and stabilized on the sample.
The embodiment also provides a puncture data acquisition and analysis method, which is specifically realized as follows:
a specific flow of puncture data collection is shown in fig. 3. And a force sensor is arranged between the bottom of the cutter and the impactor, and an oscilloscope is used for collecting the change of the impact force F in the cutter puncturing process at time t. Aluminum films having a thickness of 10 μm were laid on layers 1, 4, 8, 12, 16 of the selected samples, and each aluminum film was connected to a data acquisition instrument system. The range of the data acquisition instrument is 10V, and the acquisition frequency is 2MHz. When the cutter falls vertically to contact with the aluminum film in the sample, the circuit in the data acquisition instrument is communicated with and captures an electric signal, so that the number of layers of the cutter penetrating through the sample is recorded. After the puncturing process is finished, the camera correspondingly acquires the puncture sample image.
The result of collection of puncture data is shown in fig. 4. And importing the collected penetration layer data, puncture force data and puncture sample images into a computer. And displaying the puncture peak force, the number of penetration layers and the puncture sample image obtained in the single puncture experiment by using a liquid crystal display screen.
A gradient was set at 20 cm intervals for the drop height of the knife between 45 cm and 205 cm. Experiments were repeated 5 times for each drop height using different samples. Outliers were detected using the modified Z-score method. The Z fraction values of the 5 groups of experimental data are calculated to obtain Z1, Z2, Z3, Z4 and Z5, and the deviation degree of the Z fraction values from the average value is compared. Values that deviate more than 2 times from the standard are rejected. The calculation formula of the Z score value is as follows:where μ is the mean of the 5 data sets and σ is the standard deviation of the 5 data sets. The deviation degree calculation formula is as follows: />Where z is the value of the single set of data, mean is the mean of the 5 sets of data, std is the standard deviation of the 5 sets of data. And after removing the abnormal values, averaging 5 groups of experimental data to obtain the initial impact kinetic energy of the cutter, the puncture peak force (also called as puncture impact force peak value) and the corresponding relation between the number of sample penetration layers and different puncture sample images.
The puncture data is analyzed. An aramid fiber reinforced polymer composite (AFRP) (type FC-370) manufactured by beijing protect science limited was used as a subject. This material is typically applied as a stab-resistant layer to a flexible stab-resistant garment by means of 16 layers in a superimposed manner. The samples selected for testing were 16 layers of aramid fiber reinforced polymer composite (AFRP) 200mm long and 200mm wide. Each drop height was repeated 5 times to ensure data reliability. The cutter impact speed, the number of penetration layers and the puncture peak force obtained from the five experiments are shown in fig. 5. As the drop height increases, the lancing speed increases proportionally. The number of penetration layers and the peak penetration force of the sample correspondingly increase. Pairs using Z-scoreAnd after the abnormal values in the 5 groups of data are removed, the interpolation method is used for complementing the data. By error analysis of the 5 sets of experimental data, the standard deviation of the five sets of data did not exceed 10% of the mean, thus indicating a high reproducibility of the experiment. The average of five experimental data was taken as the cutter impact speed, number of penetration layers and penetration peak force at 9 drop heights in the dynamic penetration test. While the initial impact kinetic energy of the tool may be calculated according to the equation e=mv 2 The result is that m is the total mass of the tool and the impactor and v is the initial velocity of the tool as it contacts the sample. Therefore, the initial impact kinetic energy, the number of penetration layers and the change rule of the puncture peak force of the cutter at different falling heights are respectively shown in fig. 5, 6 and 7. By adopting the puncture experiment execution and puncture data analysis method in the embodiment, a more comprehensive puncture data rule can be obtained.
The embodiment also provides a puncture experiment execution and puncture data acquisition and analysis device, the main functional modules are shown in fig. 8, and the device comprises:
the puncture task execution module comprises a drop hammer testing machine, an impactor and a cutter used for experiments. The device is used for executing the puncture experiment task that the cutter vertically drops from different heights to puncture the sample.
The puncture task execution module has the structure that: the vertical framework of the drop weight tester is composed of 4 standard steel pipes with the length of 210 cm, and 12 steel pipes with the length of 80 cm are used for fixing in the horizontal direction. The camera is fixed on the horizontal steel pipe at top. 2 steel pipes with the length of 210 cm form a sliding rail required for the vertical movement of the impactor. The clamping tool of the cutter is arranged below the impactor, and a force sensor for monitoring the impact force of the cutter is arranged below the impactor. The force sensor is connected with the oscilloscope through a signal wire. A displacement sensor is arranged at a position 40 cm away from the bottom of the drop weight testing machine. The displacement sensor is also connected with the oscilloscope through a signal wire. A sample of 200mm x200 mm size was placed on the base of the drop weight tester. The 5 layers of 180 mm x180 mm aluminum film placed in the sample are connected into a data acquisition instrument through wires.
And the real-time hazard monitoring module is used for carrying out real-time image acquisition and analysis on the puncture experiment area. When the hand gesture of the experimenter in the puncture experiment area is monitored, the warning sound is played, and otherwise, the silence state is maintained.
And the puncture data acquisition module is used for acquiring the number of penetration layers, the puncture force and the puncture sample image in the puncture experiment process.
And the puncture data display module is used for displaying the data of the number of penetration layers, the puncture peak value force data and the puncture sample image in the liquid crystal screen in a single puncture experiment.
The processor operation module is used for analyzing and processing the puncture data (the number of penetration layers and the puncture force) acquired by the data acquisition instrument and the oscilloscope to obtain the number of penetration layers and the puncture peak force of the cutter.
And the storage module is used for storing data and images acquired by the data acquisition instrument, the oscilloscope and the camera in the process of performing the puncture experiment.
A memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the main modules in the device when the computer program is executed.
The flow of the device performing tasks is shown in fig. 9.
The foregoing is merely a preferred embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions easily contemplated by those skilled in the art within the technical scope of the present application should be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (9)
1. The puncture experiment execution and puncture data acquisition and analysis method is characterized by comprising the following steps of:
step one: performing puncture experiment preparation work, detecting working states of workers by adopting an image recognition neural network in the process, judging whether the preparation work is finished or not according to the working states, and performing a second step after judging the completion of the preparation work; wherein, the preparation work includes: fixing a puncture sample on a base of a drop hammer testing machine, mounting a cutter on an impactor, keeping a vertical angle between the cutter and the puncture sample, and pulling the cutter to a certain height;
step two: performing a puncture experiment and collecting puncture data, and collecting a puncture sample image after single puncture is completed; importing the puncture data and the puncture sample image into a computer and displaying the puncture data and the puncture sample image;
step three: and (3) changing puncture experiment parameters, performing multiple experiments according to the second step to obtain a plurality of groups of puncture experiment data, and analyzing the plurality of groups of puncture experiment data by adopting a statistical method to obtain a puncture data change rule.
2. The method for performing puncture experiment and collecting and analyzing puncture data according to claim 1, wherein,
the first step further comprises the steps of constructing and training an image recognition neural network, wherein the image recognition neural network is a convolutional neural network and comprises a basic convolutional network and an additional convolutional layer; the basic convolution network comprises 12 convolution layers, 4 maximum pooling layers and 1 full connection layer; the additional convolution layers comprise 2 residual network modules and 3 full connection layers, wherein the residual network modules comprise two convolution layers with a convolution kernel of 1x1 and a step length of 1, one convolution layer with a convolution kernel of 3x3 and a step length of 2, and one convolution layer with a convolution kernel of 1x1 and a step length of 2.
3. The method for performing puncture experiments and collecting and analyzing puncture data according to claim 2, wherein,
the training process of the image recognition neural network in the first step comprises the following steps: acquiring hand gesture images of a plurality of experimenters in preparation, wherein the hand gesture images have different sizes; dividing hand gesture images with different sizes into positive samples and negative samples based on anchor frames with different sizes; and designing a loss function, and training the image recognition neural network based on the positive sample and the negative sample and the loss function.
4. The method for performing puncture experiment and collecting and analyzing puncture data according to claim 3, wherein,
the loss function is as follows:
wherein p is the predicted size of the anchor frame, t is the actual size of the target, N is the number of samples, res1 is the error of the first residual error module, res2 is the error of the second residual error module, L conf For confidence error, L loc Is a position error;
the error calculation formula of the residual error module is as follows:
where res_pred is the predicted value of each residual block and res_true is the input true value of each residual block.
5. The method for performing puncture experiment and collecting and analyzing puncture data according to claim 1, wherein,
the process of collecting puncture data in the second step comprises the following steps: the force sensor is arranged between the bottom of the cutter and the impactor, the puncture sample comprises a plurality of layers, aluminum films are paved in the plurality of layers of puncture samples at fixed layer number intervals, and the aluminum films are connected with the derivative acquisition instrument; when a puncture experiment is carried out, the puncture impact force is obtained through the force sensor, and the number of penetration layers is obtained based on the aluminum film and the data acquisition instrument.
6. The method for performing puncture experiment and collecting and analyzing puncture data according to claim 1, wherein,
and in the third step, the puncture experiment parameters are changed, and the process of carrying out multiple experiments according to the second step comprises the following steps: setting the falling height range of the cutter, setting the height of the experiment interval by adopting a gradient method, carrying out multiple experiments, and carrying out repeated experiments by adopting different samples at the same height.
7. The method for performing puncture experiment and collecting and analyzing puncture data according to claim 1, wherein,
the process for analyzing the puncture experimental data in the third step comprises the following steps: and detecting abnormal values of a plurality of groups of puncture experiment data by adopting a Z-score method, removing the abnormal values according to the deviation degree of Z score values, and averaging the puncture experiment data to obtain the corresponding relation of the initial impact kinetic energy, the puncture impact force peak value and the number of penetration layers of the cutter and different puncture sample images.
8. An apparatus for performing the lancing test execution and lancing data acquisition and analysis method of any one of claims 1 to 7, comprising:
the puncture task execution module is used for executing puncture experiments of puncture samples by vertically falling down the cutter from different heights, and comprises a drop hammer testing machine, an impactor and the cutter;
the real-time hazard monitoring module is used for carrying out real-time image acquisition and analysis on the puncture experiment area, judging whether staff operate, if so, playing warning sound, otherwise, keeping a silence state;
the puncture data acquisition module is used for acquiring the number of penetration layers, the puncture impact force and the puncture sample image in the puncture experiment process;
the puncture data display module is used for displaying the number of penetration layers in a single puncture experiment, a puncture impact force peak value and a puncture sample image in the liquid crystal screen;
the processor operation module is used for analyzing and processing the number of penetration layers and the puncture impact force acquired by the data acquisition instrument and the oscilloscope respectively to obtain the number of penetration layers and the puncture impact force peak value of the cutter;
and the storage module is used for storing puncture experiment data in the process of performing the puncture experiment.
9. The lancing test performance and lancing data acquisition and analysis device according to claim 8, wherein,
the puncture task execution module comprises a vertical framework of the drop hammer testing machine, wherein the vertical framework is formed by 4 standard steel pipes with the length of 210 cm, and the horizontal direction of the drop hammer testing machine is fixed by 12 steel pipes with the length of 80 cm; a clamping tool of a cutter and a force sensor for monitoring the impact force of the cutter are arranged below the impactor; and a displacement sensor is arranged at a preset distance from the bottom of the drop hammer testing machine.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311670696.4A CN117554217B (en) | 2023-12-07 | Puncture experiment execution and puncture data acquisition and analysis method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311670696.4A CN117554217B (en) | 2023-12-07 | Puncture experiment execution and puncture data acquisition and analysis method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117554217A true CN117554217A (en) | 2024-02-13 |
| CN117554217B CN117554217B (en) | 2024-07-12 |
Family
ID=
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102006043912A1 (en) * | 2005-09-28 | 2007-04-19 | Hafellner, Reinhard, Dipl.-Ing. | Pendulum mechanism for testing an determining mechanical properties of a material by striking a test body |
| CN108538138A (en) * | 2018-06-21 | 2018-09-14 | 青海大学 | A kind of acquisition punctures the kinetic control system and its application method of experimental data information |
| CN109214250A (en) * | 2017-07-05 | 2019-01-15 | 中南大学 | A kind of static gesture identification method based on multiple dimensioned convolutional neural networks |
| CN111310052A (en) * | 2020-02-29 | 2020-06-19 | 平安国际智慧城市科技股份有限公司 | User portrait construction method and device and computer readable storage medium |
| CN111707566A (en) * | 2020-06-24 | 2020-09-25 | 太原理工大学 | Experimental system for researching dynamic response of structure under impact load and application method |
| CN114354394A (en) * | 2021-12-02 | 2022-04-15 | 国高材高分子材料产业创新中心有限公司 | Impact performance testing method, equipment and medium |
| CN116664524A (en) * | 2023-06-01 | 2023-08-29 | 北京理工大学 | Neural network-based anti-puncture performance detection method, system and device |
| CN117036399A (en) * | 2023-07-18 | 2023-11-10 | 中国电建集团中南勘测设计研究院有限公司 | Non-contact penetration or dynamic detection test data acquisition method and terminal equipment |
| CN117147278A (en) * | 2023-08-31 | 2023-12-01 | 广西大学 | Long-term load impact coupling loading method, system, equipment and medium |
| CN117172601A (en) * | 2023-09-06 | 2023-12-05 | 东南大学 | Non-invasive load monitoring method based on residual total convolution neural network |
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102006043912A1 (en) * | 2005-09-28 | 2007-04-19 | Hafellner, Reinhard, Dipl.-Ing. | Pendulum mechanism for testing an determining mechanical properties of a material by striking a test body |
| CN109214250A (en) * | 2017-07-05 | 2019-01-15 | 中南大学 | A kind of static gesture identification method based on multiple dimensioned convolutional neural networks |
| CN108538138A (en) * | 2018-06-21 | 2018-09-14 | 青海大学 | A kind of acquisition punctures the kinetic control system and its application method of experimental data information |
| CN111310052A (en) * | 2020-02-29 | 2020-06-19 | 平安国际智慧城市科技股份有限公司 | User portrait construction method and device and computer readable storage medium |
| CN111707566A (en) * | 2020-06-24 | 2020-09-25 | 太原理工大学 | Experimental system for researching dynamic response of structure under impact load and application method |
| CN114354394A (en) * | 2021-12-02 | 2022-04-15 | 国高材高分子材料产业创新中心有限公司 | Impact performance testing method, equipment and medium |
| CN116664524A (en) * | 2023-06-01 | 2023-08-29 | 北京理工大学 | Neural network-based anti-puncture performance detection method, system and device |
| CN117036399A (en) * | 2023-07-18 | 2023-11-10 | 中国电建集团中南勘测设计研究院有限公司 | Non-contact penetration or dynamic detection test data acquisition method and terminal equipment |
| CN117147278A (en) * | 2023-08-31 | 2023-12-01 | 广西大学 | Long-term load impact coupling loading method, system, equipment and medium |
| CN117172601A (en) * | 2023-09-06 | 2023-12-05 | 东南大学 | Non-invasive load monitoring method based on residual total convolution neural network |
Non-Patent Citations (2)
| Title |
|---|
| 刘梦真等: "小样本驱动特征分段网络的防护材料折痕检测 刘", 兵工学报, vol. 45, no. 3, 15 January 2023 (2023-01-15), pages 963 - 973 * |
| 赖鸣;兰山;黄广炎;邹浩;: "数字式高速摄像测试技术及其应用", 实验技术与管理, no. 06, 20 June 2012 (2012-06-20) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3588078B1 (en) | Inspection apparatus and inspection method for inspection target | |
| Mevel et al. | Application of a subspace-based fault detection method to industrial structures | |
| CN107014668A (en) | A kind of fatigue crack integrated monitoring based on piezoelectricity and smart coat sensor | |
| JP6795562B2 (en) | Inspection equipment and machine learning method | |
| CN114997682A (en) | Construction site safety monitoring system and method based on big data | |
| KR20200033299A (en) | System for analyzing collision and puncture resistance | |
| Santur et al. | An adaptive fault diagnosis approach using pipeline implementation for railway inspection | |
| Rosales et al. | Data driven innovations in structural health monitoring | |
| CA2034492A1 (en) | Method of monitoring the operational state of a system | |
| CN117554217B (en) | Puncture experiment execution and puncture data acquisition and analysis method and device | |
| JP6823025B2 (en) | Inspection equipment and machine learning method | |
| CN117554217A (en) | Puncture experiment execution and puncture data acquisition and analysis method and device | |
| JP7325737B2 (en) | Structure anomaly detection system | |
| CN111141472B (en) | Anti-seismic support and hanger detection method and system | |
| CN111881800A (en) | Concrete filled steel tube void defect detection method based on AR model and self-service inspection | |
| Pecho et al. | Vibration fault detection of fuel pump using recurrence quantification analysis | |
| KR101886210B1 (en) | Measuring instrument reliability evaluation apparatus and operating method thereof | |
| Boos et al. | Melt pool monitoring using fuzzy based anomaly detection in laser beam melting | |
| US11624687B2 (en) | Apparatus and method for detecting microcrack using orthogonality analysis of mode shape vector and principal plane in resonance point | |
| KR20240064037A (en) | How to identify and characterize surface defects of objects and cracks in brake discs by artificial intelligence through fatigue testing | |
| JP2021156681A (en) | Information board abnormality detection system | |
| Szeleziński et al. | Quality assessment of welded joints using the mean value distribution of amplitude spectrums calculated by the time window method | |
| CN114283392B (en) | Road target detection's difficult sample's acquisition system | |
| CN117654907B (en) | Automatic eliminating method and system for strip detector | |
| CN117056686B (en) | Alarming method and system for detecting surface defects of pressure container |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |