Deformable monitoring architecture based on flexible sensor

Abstract

The invention relates to the technical field of flexible electronics, and discloses a deformable monitoring architecture based on a flexible sensor, which comprises a flexible sensor array and a multi-layer heterostructure consisting of a piezoresistive layer, a capacitance layer and a temperature compensation layer, the distributed signal processing module is electrically connected with the flexible sensor array; the self-adaptive topology network is configured to dynamically reconstruct communication topology among the sensor nodes according to the real-time deformation quantity of the monitored target; and the cloud analysis platform is used for receiving and processing the data flow from the distributed signal processing module through an encrypted communication protocol. Coupling sensing of deformation, strain and temperature is achieved through the multi-layer heterogeneous sensor array, the piezoresistive layer material keeps linear response under 200% tensile strain, the distance between electrodes of the capacitor layer is adjusted in a self-adaptive mode, the temperature compensation precision is + / -0.1 DEG C, the deformation monitoring resolution reaches the 0.5 mm level, and the temperature change interference resistance is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of flexible electronics technology, specifically to a deformable monitoring architecture based on flexible sensors. Background Technology

[0002] With the continuous development of flexible electronics technology, various fields have put forward higher requirements for the accuracy, range and adaptability of object state monitoring. In particular, in some complex working conditions and special scenarios, it is necessary to be able to obtain information on the deformation, strain and temperature of objects in real time and accurately.

[0003] In existing technologies in related fields, some traditional monitoring systems use a single type of sensor to monitor objects. Some conventional strain monitoring devices only use resistance strain gauges to measure the strain of objects. Their working principle is based on the change in resistance of metal wires or semiconductors when subjected to force and deformation. The strain magnitude is calculated by measuring the resistance value. These devices have a simple structure and low cost, and can meet basic strain monitoring needs to a certain extent, but their function is relatively limited and cannot simultaneously acquire other key environmental parameters such as temperature. There are also some temperature monitoring devices, which are mostly based on thermistors, thermocouples, and other components. They use the characteristics of their resistance or thermoelectric potential changing with temperature to measure temperature. Although they have a certain accuracy in temperature measurement, they cannot monitor the deformation and strain information of objects.

[0004] The above-mentioned technologies have at least the following technical problems: traditional single-sensor monitoring methods cannot achieve coupled sensing of deformation, strain and temperature, and the monitoring accuracy is limited in complex environments. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a deformable monitoring architecture based on flexible sensors, which solves the problem that traditional single-sensor monitoring methods cannot achieve coupled sensing of deformation, strain, and temperature, and have limited monitoring accuracy in complex environments.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a deformable monitoring architecture based on flexible sensors, comprising:

[0007] A flexible sensor array, consisting of a multilayer heterogeneous structure comprising a piezoresistive layer, a capacitive layer, and a temperature compensation layer;

[0008] A distributed signal processing module is electrically connected to the flexible sensor array;

[0009] An adaptive topology network is configured to dynamically reconstruct the communication topology between sensor nodes based on the real-time deformation of the monitored target.

[0010] The cloud-based analytics platform receives and processes data streams from the distributed signal processing module via an encrypted communication protocol.

[0011] Preferably, the piezoresistive layer is composed of carbon nanotubes and graphene in a 3:7 mass ratio, with a thickness of 50 μm to 200 μm, and has the characteristic of maintaining a linear resistance response under 200% tensile strain.

[0012] The capacitor layer adopts a serpentine interdigitated electrode configuration, and the spacing between adjacent electrodes is configured to adaptively adjust with deformation within the range of 0.1 mm to 2 mm;

[0013] The temperature compensation layer integrates a micro thermopile array, with a temperature compensation resolution of ±0.1℃.

[0014] Preferably, the adaptive topology network includes:

[0015] The dynamic routing module uses a long short-term memory neural network to establish a deformation trend prediction model.

[0016] The channel quality assessment unit monitors the signal-to-noise ratio and transmission delay of the communication channel in real time. The effective communication channel must have a signal-to-noise ratio greater than 15dB and a transmission delay less than 50ms.

[0017] The path reconfiguration controller automatically switches to a multi-hop communication mode with at least three relay nodes when the deformation rate exceeds 3% / second.

[0018] Preferred self-calibration system:

[0019] The reference sensor array is arranged in a mirror-symmetric relationship with the working sensor array in terms of spatial arrangement;

[0020] The environmental parameter acquisition module simultaneously acquires temperature, humidity, and air pressure parameters;

[0021] The calibration execution unit uses a differential signal processing algorithm to complete the sensor zero-point drift compensation operation within 10 milliseconds.

[0022] Preferably, the data fusion method specifically includes:

[0023] The spatiotemporal feature extraction module applies wavelet transform to decompose the signal in the time domain and uses a convolutional neural network to extract spatial features in the spatial domain.

[0024] A multimodal data alignment unit, based on a dynamic time warping algorithm, achieves spatiotemporal synchronization of multi-sensor data;

[0025] The anomaly detection engine identifies deformation anomalies of 0.5mm or larger using the isolated forest algorithm.

[0026] Preferably, the power management system includes:

[0027] The flexible photovoltaic conversion layer has a photoelectric conversion efficiency of over 18% and can withstand more than 1,000 bending cycles.

[0028] Miniature supercapacitor array with an energy density of no less than 15Wh / kg;

[0029] The dynamic power regulation module controls the system power consumption to below 50μW in static monitoring mode.

[0030] Preferably, the mounting structure includes:

[0031] The magnetic fixing assembly integrates a Hall sensor to achieve a positioning accuracy of ±0.5mm;

[0032] The shape-adaptive substrate is made of a shape memory polymer material with a glass transition temperature of 60°C.

[0033] The biodegradable connecting components are made of polylactic acid, and their degradation time is controlled by molecular weight within the range of 6 to 24 months.

[0034] Preferably, the routing algorithm in the dynamic routing module includes:

[0035] The deformation prediction submodule takes historical deformation data sequences, environmental parameter sets, and material constitutive relations as input parameters.

[0036] The path evaluation submodule constructs an evaluation matrix that includes latency weights, energy consumption coefficients, and reliability indicators.

[0037] The path optimization submodule uses an improved ant colony algorithm to generate the optimal communication path within 50ms.

[0038] Preferably, in the self-calibration system, the calibration process includes:

[0039] During the benchmark establishment phase, a sensor benchmark curve is generated under standard environmental conditions of 25℃±0.5℃.

[0040] During the periodic calibration phase, online differential calibration is performed at 30-second intervals.

[0041] During the emergency calibration phase, a real-time calibration procedure is triggered when the rate of change in ambient temperature exceeds 1°C / minute.

[0042] Preferably, a method for using a deformable monitoring architecture based on flexible sensors includes the following steps:

[0043] S1. Sensor array deployment: The flexible sensor array is attached to the surface of the object being monitored by a magnetic fixing component. The Hall sensor is used to achieve positioning accuracy of ±0.5mm. According to the curvature distribution of the target area, the gradient sensitivity arrangement mode is selected. The sensor density in the high curvature area is increased to 3-5 times that in the low curvature area. The shape adaptive substrate is activated and heated to the glass transition temperature (60℃) to conformally attach it to the surface being monitored.

[0044] S2. Multimodal data acquisition and preprocessing: Synchronously acquire strain signals of the piezoresistive layer and the capacitive layer, as well as environmental parameters of the temperature compensation layer. The sampling frequency is not less than 1kHz. The distributed signal processing module executes the environmental compensation algorithm to eliminate temperature drift (resolution ±0.1℃) and electromagnetic interference. The original signal is segmented using a sliding window mechanism, with each window length set to 200ms and an overlap rate of 50%.

[0045] S3. Adaptive Topology Reconstruction: The dynamic routing module predicts the deformation trend for the next 5 sampling periods based on the LSTM neural network. When the deformation rate exceeds 3% / second, the path reconstruction controller starts the multi-hop communication mode and establishes a redundant transmission path through at least three relay nodes. The channel quality assessment unit monitors the signal-to-noise ratio and transmission delay in real time. If the signal-to-noise ratio is lower than 15dB or the delay exceeds 50ms, topology switching is triggered.

[0046] S4. Cloud-based collaborative analysis and feedback: After receiving the encrypted data stream, the cloud analysis platform performs spatiotemporal feature extraction: In the time domain, Daubechies wavelet transform is applied to decompose the signal to 6 scales, and in the spatial domain, a convolutional neural network is used to extract the spatial distribution features of the strain field. The anomaly detection engine is optimized by improving the ant colony algorithm to identify 0.5mm-level deformation anomalies and generate a warning report with a confidence score of >95%.

[0047] S5. System self-maintenance and calibration: In static monitoring mode, the dynamic power adjustment module controls the system power consumption to below 50μW and performs periodic calibration every 30 seconds. It compensates for sensor zero drift through differential signal processing algorithm. When the ambient temperature change rate is >1℃ / minute, the emergency calibration program is activated and real-time compensation is completed within 10ms.

[0048] S6. Sustainable Operation and Management: The flexible photovoltaic conversion layer maintains the basic power supply of the system under indoor lighting conditions, with a photoelectric conversion efficiency of ≥18%. The micro supercapacitor array provides 72 hours of continuous operation under continuous cloudy and rainy conditions, with an energy density of ≥15Wh / kg. The biodegradable connection components gradually degrade within 6-24 months according to the preset molecular weight, achieving environmentally friendly disassembly.

[0049] This invention provides a deformable monitoring architecture based on flexible sensors. It possesses the following...

[0050] Beneficial effects:

[0051] 1. This invention achieves coupled sensing of deformation, strain, and temperature through a multilayer heterogeneous sensor array. The piezoresistive layer material maintains a linear response under 200% tensile strain, the electrode spacing of the capacitor layer is adaptively adjustable, the temperature compensation accuracy is ±0.1℃, the deformation monitoring resolution reaches 0.5mm level, and the resistance to temperature change interference is significantly improved.

[0052] 2. The encrypted communication protocol of this invention ensures data transmission security and prevents information leakage or tampering. The multi-hop communication mode and redundant path design of the adaptive topology network maintain network topology stability in highly dynamic deformation scenarios.

[0053] 3. The self-calibration system of this invention completes zero-point drift compensation within 10 milliseconds through a reference sensor array, an environmental parameter acquisition module, and a calibration execution unit. It also has a benchmark establishment, periodic calibration, and emergency calibration process to ensure measurement accuracy. The data fusion method accurately extracts features, synchronizes data, and identifies minute deformation anomalies through spatiotemporal feature extraction, multimodal data alignment, and an anomaly detection engine. It has high detection sensitivity and a low false alarm rate.

[0054] 4. The multi-layer structure and characteristics of each layer of the sensor array of this invention give it excellent environmental adaptability and anti-interference ability. The temperature compensation layer compensates for environmental temperature interference in real time; the electrode spacing of the capacitor layer is adaptively adjusted to adapt to different degrees of deformation; the overall structural design can fit into various complex surfaces for monitoring. Attached Figure Description

[0055] Figure 1 This is a flowchart of the overall system architecture of the present invention;

[0056] Figure 2 This is a schematic diagram of the sensor array operation of the present invention;

[0057] Figure 3 This is a flowchart of the adaptive network routing process of the present invention;

[0058] Figure 4 This is a schematic flowchart of the self-calibration system of the present invention;

[0059] Figure 5 This is a schematic diagram of the anomaly detection process of the present invention;

[0060] Figure 6 This is a schematic diagram of the power management process of the present invention. Detailed Implementation

[0061] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] Please see the appendix Figure 1 This invention provides a deformable monitoring architecture based on flexible sensors, including a flexible sensor array and a multilayer heterogeneous structure composed of a piezoresistive layer, a capacitor layer and a temperature compensation layer.

[0063] A distributed signal processing module is electrically connected to a flexible sensor array;

[0064] An adaptive topology network is configured to dynamically reconstruct the communication topology between sensor nodes based on the real-time deformation of the monitored target.

[0065] The cloud-based analytics platform receives and processes data streams from the distributed signal processing module via encrypted communication protocols.

[0066] Specifically, the multi-layer heterogeneous structure design of the flexible sensor array, combining a piezoresistive layer, a capacitive layer, and a temperature compensation layer, can simultaneously sense deformation, strain, and temperature changes, improving monitoring accuracy and reliability. The adaptive topology network ensures stable data transmission even when the target deforms or the environment changes by dynamically reconstructing the communication path, making it suitable for complex and ever-changing monitoring scenarios.

[0067] The distributed signal processing module works in collaboration with the cloud analysis platform to achieve local preprocessing and in-depth cloud analysis of data, significantly improving data processing efficiency while reducing transmission latency. Encrypted communication protocols ensure data security during transmission, preventing information leakage or tampering and meeting privacy protection requirements. Intelligent feedback and early warning: The cloud analysis platform extracts spatiotemporal features through wavelet transform, quickly identifies abnormal deformations, and generates high-confidence early warnings, providing timely feedback to users. This enables high-precision, real-time monitoring of target deformation. Multi-layer heterogeneous sensor arrays and dynamic routing technology effectively solve the problem of insufficient adaptability of traditional monitoring systems in complex environments.

[0068] Please see the appendix Figure 2 The piezoresistive layer is composed of carbon nanotubes and graphene in a 3:7 mass ratio, with a thickness of 50 μm to 200 μm, and has the characteristic of maintaining a linear resistance response under 200% tensile strain.

[0069] The capacitor layer adopts a serpentine interdigitated electrode configuration, and the spacing between adjacent electrodes is configured to adaptively adjust with deformation within the range of 0.1 mm to 2 mm;

[0070] The temperature compensation layer integrates a micro thermopile array, with a temperature compensation resolution of ±0.1℃.

[0071] Specifically, the piezoresistive layer is composed of carbon nanotubes and graphene in a 3:7 mass ratio, which combines high conductivity and excellent mechanical properties, ensuring that it can maintain a linear resistance response under 200% tensile strain. This improves the measurement accuracy and stability of the sensor under extreme deformation conditions. The thickness design of 50μm to 200μm ensures both flexibility and stretchability while maintaining sufficient structural strength, making it suitable for bonding and monitoring of various complex surfaces.

[0072] Capacitor layer: The serpentine interdigitated electrode configuration increases the sensing area between electrodes, improves the sensitivity to capacitance changes, and the spacing between adjacent electrodes can be adaptively adjusted within the range of 0.1mm to 2mm, ensuring that strain signals can still be accurately captured under different degrees of deformation, thus enhancing the dynamic response capability of the sensor.

[0073] Temperature compensation layer: Integrated micro thermopile array with temperature compensation resolution up to ±0.1℃, which can monitor and compensate for the interference of ambient temperature changes on sensor signals in real time, improving the accuracy and reliability of monitoring data.

[0074] Through the synergistic effect of multi-layered heterogeneous structures, this flexible sensor array not only achieves high-sensitivity and wide-range deformation monitoring, but also possesses excellent environmental adaptability and anti-interference capabilities, providing reliable technical support for accurate monitoring under complex working conditions.

[0075] Please see the appendix Figure 3 Adaptive topology networks include:

[0076] The dynamic routing module uses a long short-term memory neural network to establish a deformation trend prediction model.

[0077] The channel quality assessment unit monitors the signal-to-noise ratio and transmission delay of the communication channel in real time. The effective communication channel must have a signal-to-noise ratio greater than 15dB and a transmission delay less than 50ms.

[0078] The path reconfiguration controller automatically switches to a multi-hop communication mode with at least three relay nodes when the deformation rate exceeds 3% / second.

[0079] Specifically, the dynamic routing module uses a long short-term memory neural network (LSTM) to build a deformation trend prediction model, which can predict future deformation trends based on historical data, realize proactive optimization of routing paths, avoid potential communication bottlenecks in advance through prediction, significantly improve network response speed, and ensure the real-time transmission of key monitoring data.

[0080] The channel quality assessment unit monitors both signal-to-noise ratio and transmission delay in real time, dynamically selects the optimal communication channel, effectively resists environmental electromagnetic interference and signal attenuation, and provides data support for path selection by quantitatively assessing communication quality, avoiding data packet loss or delay problems caused by channel degradation.

[0081] The path reconfiguration controller automatically switches to multi-hop communication mode (≥3 relay nodes) under extreme conditions where the deformation rate exceeds 3% / second. Through redundant path design, it ensures the continuity of data transmission and maintains network topology stability even in high dynamic deformation scenarios.

[0082] Please see the appendix Figure 4 Self-calibration system:

[0083] The reference sensor array is arranged in a mirror-symmetric relationship with the working sensor array in terms of spatial arrangement;

[0084] The environmental parameter acquisition module simultaneously acquires temperature, humidity, and air pressure parameters;

[0085] The calibration execution unit uses a differential signal processing algorithm to complete the sensor zero-point drift compensation operation within 10 milliseconds.

[0086] Specifically, the reference sensor array adopts a spatial layout design that is mirror-symmetrical to the working sensor array. By synchronously comparing the output signals of the working sensor and the reference sensor, the target deformation signal and environmental interference are effectively isolated.

[0087] The environmental parameter acquisition module uses a multi-dimensional environmental sensing network (temperature / humidity / air pressure) to achieve full-parameter environmental modeling. It employs timestamp synchronization technology to ensure the alignment of environmental data with sensor data and establishes an environmental interference feature library to provide a data foundation for intelligent compensation.

[0088] The calibration execution unit features an innovative differential signal processing algorithm for real-time calibration, a 10ms-level fast compensation response, and an adaptive compensation strategy that dynamically adjusts the calibration intensity based on the rate of environmental change.

[0089] Please see the appendix Figure 1 The data fusion methods specifically include:

[0090] The spatiotemporal feature extraction module applies wavelet transform to decompose the signal in the time domain and uses a convolutional neural network to extract spatial features in the spatial domain.

[0091] A multimodal data alignment unit, based on a dynamic time warping algorithm, achieves spatiotemporal synchronization of multi-sensor data;

[0092] The anomaly detection engine identifies deformation anomalies of 0.5mm or larger using the isolated forest algorithm.

[0093] Specifically, the spatiotemporal feature extraction module uses wavelet transform for multi-scale signal decomposition in the time domain to accurately extract feature components of different dynamic characteristics such as transient impact and gradual deformation. In the spatial domain, it uses a deep separable convolutional neural network to capture the spatial distribution pattern of the strain field.

[0094] The multimodal data alignment unit, based on the dynamic time warping (DTW) algorithm, breaks through the limitations of traditional fixed delay compensation, intelligently corrects the spatiotemporal misalignment between multiple sensors, and achieves μs-level time synchronization and registration.

[0095] The anomaly detection engine employs an improved isolated forest algorithm, integrates prior knowledge of materials mechanics to construct a discrimination model, and achieves early identification of minute anomalies down to 0.5mm. The detection sensitivity is 5 times higher than that of traditional methods. It also integrates an adaptive threshold mechanism to keep the false alarm rate below 0.1%.

[0096] Please see the appendix Figure 6 The power management system includes:

[0097] The flexible photovoltaic conversion layer has a photoelectric conversion efficiency of over 18% and can withstand more than 1,000 bending cycles.

[0098] Miniature supercapacitor array with an energy density of no less than 15Wh / kg;

[0099] The dynamic power regulation module controls the system power consumption to below 50μW in static monitoring mode.

[0100] Specifically, the miniature supercapacitor array features a high energy density design of 15Wh / kg, reducing its volume by 60% compared to traditional lithium batteries. Its million-cycle lifespan perfectly matches the long-term usage requirements of flexible devices, and its instantaneous high-current discharge capability (100C rate) meets the peak power consumption requirements for sudden data transmission.

[0101] The installation structure includes:

[0102] The magnetic fixing assembly integrates a Hall sensor to achieve a positioning accuracy of ±0.5mm;

[0103] The shape-adaptive substrate is made of a shape memory polymer material with a glass transition temperature of 60°C.

[0104] The biodegradable connecting components are made of polylactic acid, and their degradation time is controlled by molecular weight within the range of 6 to 24 months.

[0105] Specifically, the integrated Hall sensor achieves a positioning accuracy of ±0.5mm through magnetic field sensing, ensuring precise contact between the sensor array and the surface of the monitored object. It is especially suitable for curved or complex geometric surfaces, and features dynamic conformal bonding. It uses a shape memory polymer material with a glass transition temperature (Tg) of 60℃. After heating to Tg, the substrate softens and can adapt to changes in the curvature of the monitored surface. After cooling, it regains rigidity and forms a stable conformal contact interface.

[0106] The routing algorithms in the dynamic routing module include:

[0107] The deformation prediction submodule takes historical deformation data sequences, environmental parameter sets, and material constitutive relations as input parameters.

[0108] The path evaluation submodule constructs an evaluation matrix that includes latency weights, energy consumption coefficients, and reliability indicators.

[0109] The path optimization submodule uses an improved ant colony algorithm to generate the optimal communication path within 50ms.

[0110] Specifically, the deformation prediction submodule integrates three-dimensional input features: historical deformation data, environmental parameters, and material constitutive relations. Based on an LSTM-based spatiotemporal coupled prediction model, it achieves a deformation trend prediction accuracy of over 92% and enables advanced prediction capabilities for 5-10 sampling periods, providing a decision window for route optimization.

[0111] The path evaluation submodule pioneered a multi-objective evaluation system for latency, energy consumption, and reliability, along with a dynamic weight adjustment mechanism. It intelligently balances real-time transmission performance with system endurance based on business needs, constructs a quantitative evaluation matrix, and achieves objective and standardized evaluation of communication paths.

[0112] The path optimization submodule improves the ant colony algorithm by integrating deformation prediction results, and completes the optimal path calculation within 50ms, meeting the real-time requirements of highly dynamic scenarios.

[0113] Please see the appendix Figure 4 The self-calibration system's calibration process includes:

[0114] During the benchmark establishment phase, a sensor benchmark curve is generated under standard environmental conditions of 25℃±0.5℃.

[0115] During the periodic calibration phase, online differential calibration is performed at 30-second intervals.

[0116] During the emergency calibration phase, a real-time calibration procedure is triggered when the rate of change in ambient temperature exceeds 1°C / minute.

[0117] Specifically, in the benchmark establishment stage, a sensor benchmark curve is established in a high-precision constant temperature environment of 25℃±0.5℃. A full-range reference feature matrix is ​​generated using a multi-point calibration method to provide a standard reference system for subsequent calibration, with a benchmark accuracy of ±0.05%FS.

[0118] During the periodic calibration phase, online differential calibration is automatically performed every 30 seconds, using sliding window dynamic baseline correction technology to control long-term drift within ±0.1%.

[0119] During the emergency calibration phase, the system intelligently monitors the rate of change of ambient temperature (>1℃ / minute) and triggers a millisecond-level real-time calibration response mechanism to ensure that the measurement accuracy remains ±0.2% even under sudden changes in the environment.

[0120] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A deformable monitoring architecture based on flexible sensors, characterized in that, include: A flexible sensor array, consisting of a multilayer heterogeneous structure comprising a piezoresistive layer, a capacitive layer, and a temperature compensation layer; A distributed signal processing module is electrically connected to the flexible sensor array; An adaptive topology network is configured to dynamically reconstruct the communication topology between sensor nodes based on the real-time deformation of the monitored target. The cloud-based analytics platform receives and processes data streams from the distributed signal processing module via an encrypted communication protocol.

2. The deformable monitoring architecture based on a flexible sensor according to claim 1, characterized in that, The piezoresistive layer is composed of carbon nanotubes and graphene in a 3:7 mass ratio, with a thickness of 50 μm to 200 μm, and has the characteristic of maintaining a linear resistance response under 200% tensile strain. The capacitor layer adopts a serpentine interdigitated electrode configuration, and the spacing between adjacent electrodes is configured to adaptively adjust with deformation within the range of 0.1 mm to 2 mm; The temperature compensation layer integrates a micro thermopile array, with a temperature compensation resolution of ±0.1℃.

3. The deformable monitoring architecture based on a flexible sensor according to claim 1, characterized in that, The adaptive topology network includes: The dynamic routing module uses a long short-term memory neural network to establish a deformation trend prediction model. The channel quality assessment unit monitors the signal-to-noise ratio and transmission delay of the communication channel in real time. The effective communication channel must have a signal-to-noise ratio greater than 15dB and a transmission delay less than 50ms. The path reconfiguration controller automatically switches to a multi-hop communication mode with at least three relay nodes when the deformation rate exceeds 3% / second.

4. The deformable monitoring architecture based on a flexible sensor according to claim 1, characterized in that, It also includes a self-calibration system: The reference sensor array is arranged in a mirror-symmetric relationship with the working sensor array in terms of spatial arrangement; The environmental parameter acquisition module simultaneously acquires temperature, humidity, and air pressure parameters; The calibration execution unit uses a differential signal processing algorithm to complete the sensor zero-point drift compensation operation within 10 milliseconds.

5. The deformable monitoring architecture based on a flexible sensor according to claim 1, characterized in that, The data fusion method specifically includes: The spatiotemporal feature extraction module applies wavelet transform to decompose the signal in the time domain and uses a convolutional neural network to extract spatial features in the spatial domain. A multimodal data alignment unit performs spatiotemporal synchronization of multi-sensor data based on a dynamic time warping algorithm; The anomaly detection engine identifies deformation anomalies of 0.5mm or larger using the isolated forest algorithm.

6. The deformable monitoring architecture based on a flexible sensor according to claim 1, characterized in that, The power management system includes: The flexible photovoltaic conversion layer has a photoelectric conversion efficiency of over 18% and can withstand more than 1,000 bending cycles. Miniature supercapacitor array with an energy density of no less than 15Wh / kg; The dynamic power regulation module controls the system power consumption to below 50μW in static monitoring mode.

7. The deformable monitoring architecture based on a flexible sensor according to claim 6, characterized in that, The installation structure includes: Magnetic fixing assembly with integrated Hall sensor for ±0.5mm positioning accuracy; The shape-adaptive substrate is made of a shape memory polymer material with a glass transition temperature of 60°C. The biodegradable connecting components are made of polylactic acid, and their degradation time is controlled by molecular weight within the range of 6 to 24 months.

8. The deformable monitoring architecture based on a flexible sensor according to claim 3, characterized in that, The routing algorithm in the dynamic routing module includes: The deformation prediction submodule takes historical deformation data sequences, environmental parameter sets, and material constitutive relations as input parameters. The path evaluation submodule constructs an evaluation matrix that includes latency weights, energy consumption coefficients, and reliability indicators. The path optimization submodule uses an improved ant colony algorithm to generate the optimal communication path within 50ms.

9. The deformable monitoring architecture based on a flexible sensor according to claim 4, characterized in that, The self-calibration system, wherein the calibration process includes: During the benchmark establishment phase, a sensor benchmark curve is generated under standard environmental conditions of 25℃±0.5℃. During the periodic calibration phase, online differential calibration is performed at 30-second intervals. During the emergency calibration phase, a real-time calibration procedure is triggered when the rate of change in ambient temperature exceeds 1°C / minute.

10. A method for using a deformable monitoring architecture based on flexible sensors, characterized in that, The deformable monitoring architecture based on a flexible sensor, as described in any one of claims 1-9, comprises the following steps: S1. Sensor array deployment: The flexible sensor array is attached to the surface of the object being monitored by a magnetic fixing component. The Hall sensor is used for positioning with an accuracy of ±0.5mm. According to the curvature distribution of the target area, the gradient sensitivity arrangement mode is selected. The sensor density in the high curvature area is increased to 3-5 times that in the low curvature area. The shape adaptive substrate is activated and heated to the glass transition temperature to conformally fit with the surface being monitored. S2. Multimodal data acquisition and preprocessing: Synchronously acquire strain signals of the piezoresistive layer and the capacitive layer, as well as environmental parameters of the temperature compensation layer. The sampling frequency is not less than 1kHz. The environmental compensation algorithm is executed through the distributed signal processing module to eliminate temperature drift and electromagnetic interference. The original signal is segmented using a sliding window mechanism, with each window length set to 200ms and an overlap rate of 50%. S3. Adaptive Topology Reconstruction: The dynamic routing module predicts the deformation trend for the next 5 sampling periods based on the LSTM neural network. When the deformation rate exceeds 3% / second, the path reconstruction controller starts the multi-hop communication mode and establishes a redundant transmission path through at least three relay nodes. The channel quality assessment unit monitors the signal-to-noise ratio and transmission delay in real time. If the signal-to-noise ratio is lower than 15dB or the delay exceeds 50ms, topology switching is triggered. S4. Cloud-based collaborative analysis and feedback: After receiving the encrypted data stream, the cloud analysis platform performs spatiotemporal feature extraction: In the time domain, Daubechies wavelet transform is applied to decompose the signal to 6 scales, and in the spatial domain, a convolutional neural network is used to extract the spatial distribution features of the strain field. The anomaly detection engine is optimized by improving the ant colony algorithm to identify 0.5mm-level deformation anomalies and generate a warning report with a confidence score of >95%. S5. System self-maintenance and calibration: In static monitoring mode, the dynamic power adjustment module controls the system power consumption to below 50μW and performs periodic calibration every 30 seconds. It compensates for sensor zero drift through differential signal processing algorithm. When the ambient temperature change rate is >1℃ / minute, the emergency calibration program is activated and real-time compensation is completed within 10ms. S6. Sustainable Operation and Management: The flexible photovoltaic conversion layer maintains the basic power supply of the system under indoor lighting conditions, with a photoelectric conversion efficiency of ≥18%. The micro supercapacitor array provides 72 hours of continuous operation under continuous cloudy and rainy conditions, with an energy density of ≥15Wh / kg. The biodegradable connection components gradually degrade within 6-24 months according to the preset molecular weight, allowing for environmentally friendly disassembly.

Images