Stress field intelligent monitoring device and method based on array sensor

Abstract

The invention relates to an array sensor-based stress field intelligent monitoring device and method, the device comprises a strain gauge, the strain gauge is connected with a detection box, the detection box comprises a box body, and a communication module, a power supply module, a processing module, a storage module and a data acquisition module which are arranged at the bottom of the box body, and the box body is also internally provided with a storage module; the communication module is used for being wirelessly connected with a mobile phone and a computer mobile terminal to realize wireless data transmission; the power supply module is used for supplying power to the detection device; the control module is used for receiving data of the strain gauges and performing real-time alarm when the damage level exceeds a threshold value through data processing; the storage module is used for realizing data storage; the data acquisition module is used for acquiring strain data and converting the strain data into stress data. The stress detection device is simple in structure, comprehensive in function and capable of working independently, and real-time monitoring, intelligent data processing, data storage, remote transmission and danger early warning of a local dangerous area are achieved through a single monitoring device.

Description

Technical Field

[0001] This invention relates to stress field detection and monitoring of ship hull structures, and more specifically, to an intelligent stress field monitoring device and method based on array sensors. Background Technology

[0002] Ships are subjected to complex alternating loads from wind, waves, and currents during their voyages at sea. These alternating loads can cause deformation, fatigue damage, and even fracture of the hull structure. Therefore, real-time monitoring of the stress state of the hull structure is crucial for ensuring the safe operation of ships. Existing stress detection devices are typically complex in structure and have numerous circuits, making long-distance monitoring and real-time data processing difficult. Existing devices suffer from complex structures, chaotic wiring, bottlenecks in long-distance monitoring technology, and low efficiency in damage level identification. Summary of the Invention The technical problem to be solved by the present invention is to provide a stress field intelligent monitoring device and method based on array sensors, which can realize real-time monitoring of local dangerous areas, intelligent data processing, data storage and remote transmission, and danger warning through a single monitoring device, and can also form a parallel system through multiple sets of devices to build a three-dimensional health monitoring network for the whole ship, providing support for the digital twin of the whole ship.

[0003] The technical solution adopted by the present invention to solve its technical problem is: to construct a stress field intelligent monitoring device based on array sensors, including strain gauges, the strain gauges being connected to a detection box, the detection box including a box body and a communication module, a power supply module, a processing module, a storage module and a data acquisition module disposed at the bottom of the box body, and a storage module being disposed inside the box body.

[0004] According to the above scheme, the communication module is composed of a wireless transparent transmission board and an AI chip connected by wires, and is used to wirelessly connect with mobile phones and computer mobile terminals to realize wireless data transmission. The power supply module is a lithium battery or a solar panel, used to power the detection device; The control module consists of an AI chip and a buzzer connected by wires. It is used to receive data from the strain gauge and, through data processing, to issue a real-time alarm when the damage level exceeds the threshold. The storage module stores the data read by the AI ​​chip onto the SD card, thus enabling data storage. The data acquisition module consists of a wireless force acquisition plate and strain gauges connected by wires, and is used to acquire strain data and convert it into stress data.

[0005] According to the above scheme, the storage module includes a fixed pulley, a winding wheel, and a conductive slip ring. The fixed pulley includes two fixedly arranged on the side of the housing, one above the other. The winding wheel is arranged at the lower top of the housing. A steel bar is arranged in the middle of the housing. The conductive slip ring is fixedly arranged on the steel bar. The strain gauge is connected to the wire. The wire passes through the fixed pulley, the winding wheel, and the conductive slip ring in sequence. The storage module is used to store excess wire.

[0006] According to the above scheme, the bottom of the winding wheel is provided with a through hole, and the wire passes through the through hole and the conductive slip ring in sequence before being connected to the wireless force acquisition board for data acquisition.

[0007] According to the above scheme, the box body is made of rigid plastic, and a magnetic locating piece is glued to the bottom of the box body for fixing. The box body is arranged in the shape of a cuboid.

[0008] According to the above scheme, a recessed switch connected to the control module is provided on the side of the enclosure, and an indicator light for indicating the detection status is provided on the side of the enclosure, and the indicator light is located outside the recessed switch. The strain gauge is connected to the first interface and the second interface in sequence via wires. The first interface and the second interface are connected by male and female connectors.

[0009] This invention also provides a method for intelligent monitoring of stress fields based on array sensors, comprising the following steps: S1. Check the performance of the instrument and ensure that the wire connections are normal; S2. Attach one end of the required strain gauge to the target and connect the other end to the connector to form an array sensor. S3. Pick up the test box and open the recessed switch from the side, and observe whether the indicator light lights up; S4. After the detection box is powered on, use a mobile terminal to pair with the detection box via wireless connection. After successful pairing, the mobile terminal will be able to obtain the detection data. S5. After a certain period of time, the strain gauge undergoes deformation. The array sensor transmits the data to the wireless force acquisition board to complete the stress-strain conversion. The data is then transmitted to the AI ​​chip, which processes the data to determine the damage level and segmented area. This data is then compared to a pre-set damage threshold to determine whether an alarm is needed. S6. After the test is completed, remove the strain gauge from the second interface, collect and process it, and turn off the recessed switch of the test box, waiting for the next test. Alternatively, it can be continuously powered on for long-term monitoring and forecasting. The test results can be displayed through a mobile terminal or the data can be read from the SD card.

[0010] According to the above scheme, the array sensor consists of multiple strain gauges. When the stress state of the object changes, the strain gauges are subjected to strain pressure and change accordingly. The data is processed by an AI chip.

[0011] According to the above scheme, the method of data processing using AI chips includes the following steps: S1. The strain gauge changes are collected and extracted using a wireless force acquisition plate; S2. The wireless acquisition board sends the processed data into the AI ​​chip for secondary processing. The S3 and AI chip integrate the Res-Mask-R-CNN algorithm to test the data and complete the damage type classification and damage region segmentation. S4. The AI ​​chip compares the predicted damage level with a threshold. If the threshold is exceeded, the alarm device will be controlled to trigger an alarm. The S5 and AI chips remotely transmit data to the mobile terminal and store it on the SD card via transmission and storage devices, respectively.

[0012] The stress field intelligent monitoring device and method based on array sensors of the present invention have the following beneficial effects: 1. This invention simplifies the structure of the detection device by using only one detection box to simultaneously house wires, collect data, identify damage, store data, and transmit data, thus integrating the originally cumbersome device and saving production costs. Multiple stress detection interfaces can be carried in one detection box body. Based on different structural forms, the sensor array range can be expanded and the form can be changed, making it suitable for complex ship structures, curved surfaces, etc.

[0013] 2. The type of strain gauge can be selected according to user needs. The position and number of strain gauges can be further optimized based on actual requirements, adding more strain gauges to capture more subtle stress changes. The entire testing chamber has a relatively compact and convenient structure, which is very conducive to its promotion and use. 3. The AI ​​chip of this invention employs deep learning technology, constructing a Res-Mask-R-CNN model based on the ResNet-101 backbone network for identifying, classifying, and segmenting damage regions from stress cloud maps of hull materials. ResNet-101 is used to extract image features, and a feature pyramid network is constructed to fuse features at different scales. The Mask-R-CNN network is used for damage type classification and damage region segmentation. 4. The multiple independent stress monitoring devices of the present invention can operate in parallel through communication protocols to construct a three-dimensional monitoring network and realize the overall health monitoring of the entire ship. Attached Figure Description

[0014] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings: Figure 1 This is a schematic diagram of the structure of the intelligent stress field monitoring device and method based on array sensors of the present invention; Figure 2 This is a schematic diagram of the storage module of the present invention; Figure 3 This is a structural framework diagram of the Res-Mask-R-CNN algorithm of this invention; Figure 4 This is a network structure diagram of the feature pyramid network of the present invention; Figure 5 This is an example diagram of an array sensor composed of multiple strain gauges according to the present invention; Figure 6 This is a flowchart illustrating the workflow of the intelligent stress field monitoring device and measurement method based on array sensors of the present invention. Figure 7 This is a three-dimensional detection network diagram of the entire ship according to the present invention; Figure 8 This is a first schematic diagram showing the connection of the strain gauge of the present invention to the first interface and the second interface respectively; Figure 9 This is a second schematic diagram showing the connection between the strain gauge of the present invention and the first interface and the second interface respectively; Figure 10 This is a schematic diagram of the connection structure between the recessed switch and the indicator light of the present invention; In the diagram: 1. Strain gauge, 2. First interface, 3. Second interface, 4. Winding wheel, 5. AI chip, 6. Wireless force acquisition board, 7. SD card, 8. Fixed pulley, 9. Through hole, 10. Conductive slip ring, 11. Hole, 12. Housing, 19. Lithium battery, 20. Wireless transmission board, 21. Buzzer, 22. Data acquisition module, 23. Recessed switch, 24. Indicator light, 25. Steel bar. Detailed Implementation

[0015] To provide a clearer understanding of the technical features, objectives, and effects of the present invention, specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0016] like Figure 1-7As shown, the intelligent stress field monitoring device based on array sensors of the present invention includes strain gauges. Strain gauge 1 is connected sequentially to a first interface 2 and a second interface 3 via wires. The first interface 2 and the second interface 3 are connected by a male-female connector. The strain gauge 1 is connected to a detection box, which includes a box body 12 and a communication module, a power supply module, a processing module, a storage module, and a data acquisition module 22 located at the bottom of the box. A storage module is also provided inside the box. The number of strain gauges 1 is selected according to the needs of the detected stress field. Based on different structural forms, the sensor array range can be expanded and its form can be changed, making it suitable for complex ship structures, curved surfaces, etc.

[0017] The communication module consists of a wireless transmission board 20 and an AI chip 5 connected by wires, used for wireless connection with mobile terminals such as mobile phones and computers to achieve wireless data transmission; the power supply module is a lithium battery 19 or a solar panel, used to power the detection device; the control module consists of an AI chip 5 and a buzzer 21 connected by wires, used to receive data from the strain gauge 1 and, through data processing, to issue a real-time alarm when the damage level exceeds the threshold; the storage module stores the data read by the AI ​​chip 5 into an SD card 7; the data acquisition module 22 consists of a wireless force acquisition board 6 and a strain gauge 1 connected by wires, used to acquire strain data and convert it into stress data.

[0018] The storage module includes a fixed pulley 8, a winding wheel 4, and a conductive slip ring 10. The fixed pulley 8 consists of two parts, one above the other, fixedly mounted on the side of the housing. The winding wheel 4 is located at the bottom top of the housing. A steel bar is positioned in the middle of the housing, and the conductive slip ring 10 is fixedly mounted on the steel bar. The strain gauge 1 is connected to the wire, which passes sequentially around the fixed pulley 8, the winding wheel 4, and the conductive slip ring 10. The winding wheel 4 and the conductive slip ring 10 cooperate to prevent the wire from tangling. The storage module is used to store excess wire. A through hole 9 is provided at the bottom of the winding wheel 4. The wire passes through the through hole 9 and the conductive slip ring 10 sequentially before connecting to the wireless force acquisition board 6 for data acquisition. Holes 11 are provided on the steel bar, through which the conductive slip ring 10 passes for fixed connection. The steel bar 25 is an L-shaped steel bar.

[0019] The housing 12 is made of rigid plastic. A magnetic locating piece 22 is glued to the bottom of the housing 12 for fixing. The housing 12 is rectangular in shape. A recessed switch 23 connected to the control module is provided on the side of the housing 12. An indicator light 24 for indicating the detection status is also provided on the side of the housing 12. The indicator light 24 is located outside the recessed switch 23. The method for real-time monitoring of strain gauges includes the following steps: S1. Check the performance of each instrument and ensure that the wire connections are normal. S2. Attach one end of the required strain gauge to the target and connect the other end to the connector to form an array sensor. S3. Pick up the test box and open the recessed switch from the side, and observe whether the indicator light lights up; S4. After the detection box is powered on, use a mobile terminal to pair with the detection box via wireless connection. After successful pairing, the mobile terminal will be able to obtain the detection data. S5. After a certain period of time, the strain gauge will deform. The array sensor will transmit the data to the wireless force acquisition board to complete the stress-strain conversion. The data will be transmitted to the AI ​​chip 5. The AI ​​chip 5 will process the data to obtain the damage level and segmentation area. Then, it will be compared with the pre-set damage threshold to determine whether an alarm is needed. S6. After the test is completed, remove the strain gauge from the second connector, collect and process it, and turn off the recessed switch of the test box for the next test. Alternatively, it can be powered on continuously for long-term monitoring and forecasting. The test results can be displayed on a mobile terminal or the data can be read from the SD card.

[0020] This invention also provides a post-processing method for intelligent stress field monitoring based on an array sensor. The array sensor consists of multiple strain gauges 1. When the stress state of an object changes, the strain gauges 1 are subjected to strain pressure and change accordingly. Data processing is performed by an AI chip 5, specifically including the following steps: S1. The strain gauge changes are collected and extracted using a wireless force acquisition plate; S2. The wireless acquisition board sends the processed data into the AI ​​chip 5 for secondary processing. The S3 and AI chip 5 are combined with the Res-Mask-R-CNN algorithm to test the data and complete the damage type classification and damage region segmentation. S4 and AI chip 5 will compare the predicted damage level with a threshold. If the threshold is exceeded, the alarm device will be controlled to trigger an alarm. The S5 and AI chip 5 remotely transmit data to the mobile terminal and store it on the SD card via transmission and storage devices, respectively.

[0021] like Figure 6 As shown, multiple sets of devices form a parallel system, constructing a three-dimensional health monitoring network for the entire ship, providing support for the ship's digital twin. For example... Figure 3 As shown, the AI ​​chip 5 research employs deep learning technology to construct a Res-Mask-R-CNN model based on the ResNet-101 backbone network. The stress cloud map of the hull material is input into ResNet-101 to extract its image features, and a Feature Pyramid Network (FPN) is constructed to fuse features at different scales. The Mask-R-CNN network is used for damage type classification and damage region segmentation. Among these: like Figure 4 As shown, the network backbone consists of ResNet-101 + FPN: ResNet-101 is divided into multiple stages. The diagram shows Stages 1-4. The core formula is: F(x) = H(x). x, each Stage outputs feature maps of different scales, denoted as C2-C5.

[0022] FPN fuses features from different levels through top-down sampling and lateral connections: P5 = Conv3 × 3(Conv1 × 1(C5)) P4 =Conv3×3(Conv1×1(C4)+Upsample(P5)) P3 =Conv3×3(Conv1×1(C3)+Upsample(P4)) P2 =Conv3×3(Conv1×1(C2)+Upsample(P3)) P6 = MaxPooling(P5) (step size is 2) RPN generates candidate boxes (Anchors) by sliding them across the feature map Pi of each layer of FPN, and filters out possible regions (Rol) containing damage. RolAlign accurately extracts Rol features using bilinear interpolation to generate Rol feature maps. The Rol feature map is input into the detection head, which simultaneously performs damage type classification and bounding box refinement, and finally completes damage type classification and damage region segmentation.

[0023] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A stress field intelligent monitoring device based on an array sensor, comprising strain gauges connected to a detection chamber, characterized in that, The testing box includes a box body and a communication module, a power supply module, a processing module, a storage module and a data acquisition module located at the bottom of the box body. The box body also includes a storage module.

2. The intelligent stress field monitoring device based on array sensors according to claim 1, characterized in that, The communication module consists of a wireless transparent transmission board and an AI chip connected by wires, and is used to wirelessly connect with mobile terminals such as mobile phones and computers to realize wireless data transmission. The power supply module is a lithium battery or a solar panel, used to power the detection device; The control module consists of an AI chip and a buzzer connected by wires. It is used to receive data from the strain gauge and, through data processing, to issue a real-time alarm when the damage level exceeds the threshold. The storage module stores the data read by the AI ​​chip onto the SD card, thus enabling data storage. The data acquisition module consists of a wireless force acquisition plate and strain gauges connected by wires, and is used to acquire strain data and convert it into stress data.

3. The intelligent stress field monitoring device based on array sensors according to claim 1, characterized in that, The storage module includes a fixed pulley, a winding wheel, and a conductive slip ring. The fixed pulley comprises two parts, one above the other, fixedly mounted on the side of the housing. The winding wheel is located at the lower top of the housing. A steel bar is provided in the middle of the housing. The conductive slip ring is fixedly mounted on the steel bar. The strain gauge is connected to a wire, and the wire passes sequentially around the fixed pulley, the winding wheel, and the conductive slip ring. The storage module is used to store excess wire.

4. The intelligent stress field monitoring device and method based on array sensors according to claim 3, characterized in that, The bottom of the winding wheel is provided with a through hole. The wire passes through the through hole and the conductive slip ring in sequence and is then connected to the wireless force acquisition board for data acquisition.

5. The intelligent stress field monitoring device based on array sensors according to claim 4, characterized in that, The box is made of rigid plastic, and a magnetic clasp is glued to the bottom of the box for fixing. The box is rectangular in shape.

6. The intelligent stress field monitoring device based on array sensors according to claim 1, characterized in that, The side of the enclosure is equipped with a recessed switch connected to the control module, and the side of the enclosure is equipped with an indicator light for indicating the detection status, which is located outside the recessed switch.

7. The intelligent stress field monitoring device based on array sensors according to claim 1, characterized in that, The strain gauge is connected to the first interface and the second interface in sequence via wires. The first interface and the second interface are connected by male and female connectors.

8. A method for intelligent monitoring of stress field based on array sensors according to claim 1, characterized in that, Includes the following steps: S1. Check the performance of the instrument and ensure that the wire connections are normal; S2. Attach one end of the required strain gauge to the target and connect the other end to the connector to form an array sensor. S3. Pick up the test box and open the recessed switch from the side, and observe whether the indicator light lights up; S4. After the detection box is powered on, use a mobile terminal to pair with the detection box via wireless connection. After successful pairing, the mobile terminal will be able to obtain the detection data. S5. After a certain period of time, the strain gauge will deform. The array sensor will transmit the data to the wireless force acquisition board to complete the stress-strain conversion. The data will be transmitted to the AI ​​chip, which will process the data to determine the damage level and segmentation area. Then, it will be compared with the pre-set damage threshold to determine whether an alarm is needed. S6. After the test is completed, remove the strain gauge from the second interface, collect and process it, and turn off the recessed switch of the test box, waiting for the next test. Alternatively, it can be continuously powered on for long-term monitoring and forecasting. The test results can be displayed through a mobile terminal or the data can be read from the SD card.

9. The intelligent stress field monitoring method based on array sensors according to claim 8, characterized in that, The array sensor consists of multiple strain gauges. When the stress state of an object changes, the strain gauges are subjected to strain pressure and change accordingly. The data is processed by an AI chip.

10. The intelligent stress field monitoring method based on array sensors according to claim 9, characterized in that, The method of data processing using AI chips includes the following steps: S1. The strain gauge changes are collected and extracted using a wireless force acquisition plate; S2. The wireless acquisition board sends the processed data into the AI ​​chip for secondary processing. The S3 and AI chip integrate the Res-Mask-R-CNN algorithm to test the data and complete the damage type classification and damage region segmentation. S4. The AI ​​chip compares the predicted damage level with a threshold. If the threshold is exceeded, the alarm device will be controlled to trigger an alarm. The S5 and AI chips remotely transmit data to the mobile terminal and store it on the SD card via transmission and storage devices, respectively.

Images