Temperature monitoring system based on thermal response cholesteric liquid crystal elastomer, temperature monitoring method and application

Abstract

The invention discloses a temperature monitoring system based on a thermal response cholesteric liquid crystal elastomer, a temperature monitoring method and application. The temperature monitoring system comprises a cholesteric liquid crystal elastomer film, imaging equipment and a processor, or comprises a plurality of cholesteric liquid crystal elastomer films, imaging equipment, infrared imaging equipment and a processor. The optical color change characteristic of the cholesteric liquid crystal elastomer film is combined with the machine vision technology, the temperature value (a color-based temperature detection model) or a temperature distribution diagram (a color array-based temperature mapping model) of a monitored object is monitored through the color change of the cholesteric liquid crystal elastomer film, and experimental verification shows that the cholesteric liquid crystal elastomer film can be used for monitoring the temperature of the monitored object. The monitoring method can realize high-precision temperature monitoring in a range of 50-70 DEG C.

Description

Technical Field

[0001] The present invention relates to the technical field of temperature monitoring, and in particular to a temperature monitoring system based on a thermally responsive cholesteric liquid crystal elastomer and a temperature monitoring method and application. Background Art

[0002] With the acceleration of modern industrialization, temperature monitoring plays an increasingly important role in many fields, especially in electronic equipment, manufacturing, chemical and medical industries. Real-time monitoring of temperature changes can effectively prevent equipment failure and material aging.

[0003] Existing temperature measurement equipment, such as infrared temperature measurement equipment and temperature sensors, can achieve temperature detection, but they often have problems such as high cost, complex installation, and the equipment itself may cause safety hazards. In addition, various traditional thermometers cannot intuitively display temperature changes, and the response speed is slow when the ambient temperature changes rapidly.

[0004] In recent years, temperature-responsive materials (such as liquid crystal elastomers, shape memory polymers, etc.) have gradually become a hot topic in temperature monitoring research. Some physical or chemical properties of these materials will change significantly with temperature changes, and can be used for temperature detection. However, in the existing technology, the application of temperature-responsive materials is still subject to some limitations, especially in the problem of how to effectively associate the color change of the material with the ambient temperature. There is no mature method to achieve high-precision temperature monitoring. Summary of the invention

[0005] Based on this, the present invention provides a temperature monitoring system and a temperature monitoring method and application based on a thermally responsive cholesteric liquid crystal elastomer, which realizes visualization of the temperature change feedback of the monitored object through the color change of the cholesteric liquid crystal elastomer film.

[0006] In a first aspect, the present invention provides a temperature monitoring system based on a thermally responsive cholesteric liquid crystal elastomer, comprising: Cholesteric liquid crystal elastomer film, used to characterize the temperature change of the monitored object based on its own color change; An imaging device for capturing the color change of the cholesteric liquid crystal elastomer film and forming an optical image; The processor is used to receive the optical image and input the optical image into a trained color-based temperature detection model for processing to output a temperature value under corresponding conditions.

[0007] In a second aspect, the present invention provides a temperature monitoring method of a temperature monitoring system based on a thermally responsive cholesteric liquid crystal elastomer, comprising the following steps: Step S11, placing a cholesteric liquid crystal elastomer film on the surface of the object to be monitored; Step S12, using an imaging device to capture the color change of the cholesteric liquid crystal elastomer film and form an optical image; Step S13, receiving the optical image through a processor, and inputting the optical image into a trained color-based temperature detection model for processing to output a temperature value under corresponding conditions.

[0008] Furthermore, in step S13, the construction of the color-based temperature detection model includes the following steps: Step S131, after placing the red cholesteric liquid crystal elastic film that has been uniaxially stretched and maintained in a stretched state in an oven for baking until the dynamic covalent bonds are completely exchanged, the cholesteric liquid crystal elastic film is taken out of the oven for cooling to room temperature, and cut into blue sample films; Step S132, placing the blue sample film on a hot stage, capturing the color change of the blue sample film by an imaging device within a preset time to generate an optical image, and preprocessing the optical image to form an RGB image, and simultaneously recording the corresponding temperature conditions to form a first temperature label file; Step S133, using a convolutional neural network architecture to establish a color-based temperature detection model, and inputting the first temperature label file into the color-based temperature detection model for iterative training; Step S134, continuously optimizing the parameters of the color-based temperature detection model through the loss function, evaluating the performance of the color-based temperature detection model through the validation set, and adjusting the hyperparameters of the color-based temperature detection model according to the evaluation results to achieve the optimal color-based temperature detection model.

[0009] Further, in step S134, the loss function is a mean square error loss function, and in the optimal color-based temperature detection model, the learning rate is 0.0005, the batch size is 16, and the training rounds are 100.

[0010] In a third aspect, the present invention provides a temperature monitoring system based on a thermally responsive cholesteric liquid crystal elastomer, comprising: A plurality of cholesteric liquid crystal elastomer films, wherein the plurality of cholesteric liquid crystal elastomer films are pasted on black oil paper in a pixel array manner, and are used to characterize the temperature change of the monitored object according to the color change of the cholesteric liquid crystal elastomer films; An imaging device for capturing the color change of the cholesteric liquid crystal elastomer film and forming an optical image; Infrared imaging equipment, used to capture the color change of the cholesteric liquid crystal elastomer film and form an infrared temperature image under conditions corresponding to the optical image; The processor is used to receive the optical image and the infrared temperature image, and input the optical image into a trained temperature mapping model based on a color array for processing, so as to output a temperature distribution map under corresponding conditions.

[0011] In a fourth aspect, the present invention provides a temperature monitoring method of a temperature monitoring system based on a thermally responsive cholesteric liquid crystal elastomer, comprising the following steps: Step S21, after pasting a plurality of cholesteric liquid crystal elastomer films on black oil paper in a pixel array manner to form an array film, the array film is placed on the surface of the object to be monitored; Step S22, using an imaging device to capture the color change of the cholesteric liquid crystal elastomer film and form an optical image; Step S23, using an infrared imaging device to capture the color change of the cholesteric liquid crystal elastomer film and form an infrared temperature image under conditions corresponding to the optical image; Step S24, receiving the optical image and the infrared temperature image through a processor, and inputting the optical image and the infrared temperature image into a trained color array-based temperature mapping model for processing, so as to output a temperature distribution map under corresponding conditions.

[0012] Furthermore, in step S24, the construction of the temperature mapping model based on the color array includes the following steps: Step S241, after placing the red cholesteric liquid crystal elastic film that has been uniaxially stretched and maintained in a stretched state in an oven for baking until the dynamic covalent bonds are completely exchanged, the cholesteric liquid crystal elastic film is taken out of the oven for cooling to room temperature, and the green sample film cut into pieces is pasted on black oil paper in a pixel array manner to form an array film; Step S242, placing the array film on a glass plate, heating different positions with an open flame source, capturing the color change of the heated area of ​​the array film with an imaging device within a preset time to generate an optical image, and preprocessing the optical image to form an RGB image; Step S243, using an infrared imaging device to capture the color change of the array film and form an infrared temperature image under conditions corresponding to the optical image; Step S244, using a generative adversarial neural network architecture to establish a temperature mapping model based on a color array, and inputting a second temperature label file formed by the RGB image and the infrared temperature image into the temperature mapping model based on the color array for iterative training; Step S245, continuously optimizing the parameters of the temperature mapping model based on the color array through the loss function, evaluating the performance of the temperature mapping model based on the color array through the validation set, and adjusting the hyperparameters of the temperature mapping model based on the color array according to the evaluation results to achieve the optimal temperature mapping model based on the color array.

[0013] Further, in step S245, the loss function is a binary cross entropy loss function, and in the optimal color array-based temperature mapping model, the generator learning rate is 0.00005, the discriminator learning rate is 0.000005, the batch size is 16, and the training rounds are 100.

[0014] Furthermore, the method for preparing the cholesteric liquid crystal elastomer film comprises the following steps: Step S31, adding a thermally responsive liquid crystal monomer, a chiral dopant, a cross-linking agent, a photoinitiator, a dynamic covalent bond, and a catalyst into an organic solvent in a preset mass ratio for dissolution to form a precursor solution, wherein the mass ratio of the thermally responsive liquid crystal monomer, the chiral dopant, the cross-linking agent, the photoinitiator, the dynamic covalent bond, and the catalyst is: 1107 mg: 46 mg: 270 mg: 10 mg: 136 mg: 3 mg; Step S32, pouring the precursor solution into a mold and letting it stand, and after the organic solvent evaporates and the cholesteric phase is formed, placing it under ultraviolet light for curing to obtain a completely cross-linked cholesteric liquid crystal elastomer film; Step S33, uniaxially stretching the cholesteric liquid crystal elastomer film and maintaining the stretched state at a preset temperature for a preset time to complete dynamic covalent bond exchange, so that the initially red cholesteric liquid crystal elastomer film is oriented along the stretching direction and changes to a corresponding color.

[0015] In a fifth aspect, the present invention provides an application of a temperature monitoring system based on a thermally responsive cholesteric liquid crystal elastomer in monitoring the surface temperature of an electronic device.

[0016] Compared with the prior art, the present invention has the following beneficial effects: First, the preparation process of the cholesteric liquid crystal elastomer film used in the temperature monitoring system of the present invention is simple, the reaction conditions are mild, and it is easy to industrialize the production.

[0017] Secondly, the present invention realizes the visualization of the temperature change feedback of the monitored object through the color change of the cholesteric liquid crystal elastomer film, which greatly enhances the usability and intuitiveness of the monitoring system.

[0018] Third, compared with traditional infrared temperature monitoring equipment, temperature sensors and thermometers, the temperature monitoring method of the present invention has the advantages of low cost, simple structure and high adaptability, and can be cut into various shapes as needed. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 Schematic diagram of preliminary temperature detection using CLCE film in Example 1.

[0020] Figure 2 It is a schematic diagram of the process of realizing the temperature detection function based on CLCE and machine vision technology in Example 1 and Comparative Example 1, and a schematic diagram of the network architecture of the color-based temperature detection model (CTPM).

[0021] Figure 3 The loss curve (left) and training result evaluation curve (right) of the color-based temperature detection model (CTPM) obtained in Example 1.

[0022] Figure 4 is the temperature detection result of the CLCE film obtained in Example 1 combined with the color-based temperature detection model (CTPM), where T t Represents the actual temperature value under the corresponding conditions, T p Represents the detected temperature value output by the model (scale bar is 2 cm).

[0023] Figure 5 Schematic diagram of the change of the blue CLCE film obtained in Comparative Example 1 when heated (scale bar is 2 cm).

[0024] Figure 6 The loss curve (left) and training result evaluation curve (right) of the color-based temperature detection model (CTPM) obtained in comparative example 1.

[0025] Figure 7 Schematic diagram of the process for preparing the CLCE array in Example 2.

[0026] Figure 8 Schematic diagram of the network structure of the color array temperature mapping model (CATMM) obtained in Example 2.

[0027] Fig. 9 Loss curves of the generator (left) and discriminator (right) during model training in Example 2.

[0028] Fig.10 This is a schematic diagram of the process of simulating the use of a CLCE array to monitor the temperature of a local high-temperature object surface in Example 2.

[0029] Fig.11 The temperature monitoring result obtained in Example 2 (scale bar is 2 cm).

[0030] Fig.12 The results of temperature monitoring at different locations under high temperature in Example 2 (scale bar is 2 cm). DETAILED DESCRIPTION

[0031] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0032] In a first aspect, the present invention provides a temperature monitoring system based on a thermally responsive cholesteric liquid crystal elastomer, comprising: Cholesteric liquid crystal elastomer film, used to characterize the temperature change of the monitored object based on its own color change; An imaging device for capturing the color change of the cholesteric liquid crystal elastomer film and forming an optical image; The processor is used to receive the optical image and input the optical image into a trained color-based temperature detection model for processing to output a temperature value under corresponding conditions.

[0033] In a second aspect, the present invention provides a temperature monitoring method of a temperature monitoring system based on a thermally responsive cholesteric liquid crystal elastomer, comprising the following steps: Step S11, placing a cholesteric liquid crystal elastomer film on the surface of the object to be monitored; Step S12, using an imaging device to capture the color change of the cholesteric liquid crystal elastomer film and form an optical image; Step S13, receiving the optical image through a processor, and inputting the optical image into a trained color-based temperature detection model for processing to output a temperature value under corresponding conditions.

[0034] Furthermore, in step S13, the construction of the color-based temperature detection model includes the following steps: Step S131, after placing the red cholesteric liquid crystal elastic film that has been uniaxially stretched and maintained in a stretched state in an oven for baking until the dynamic covalent bonds are completely exchanged, the cholesteric liquid crystal elastic film is taken out of the oven for cooling to room temperature, and cut into blue sample films; Step S132, placing the blue sample film on a hot stage, capturing the color change of the blue sample film by an imaging device within a preset time to generate an optical image, and preprocessing the optical image to form an RGB image, and simultaneously recording the corresponding temperature conditions to form a first temperature label file; Specifically, in this step, preprocessing includes rotation, cropping, and scaling to increase data diversity and improve the robustness of the model.

[0035] Step S133, using a convolutional neural network architecture to establish a color-based temperature detection model, and inputting the first temperature label file into the color-based temperature detection model for iterative training; Specifically, before iterative training, the model needs to be initialized and parameters set.

[0036] Step S134, continuously optimizing the parameters of the color-based temperature detection model through the loss function, evaluating the performance of the color-based temperature detection model through the validation set, and adjusting the hyperparameters of the color-based temperature detection model according to the evaluation results to achieve the optimal color-based temperature detection model.

[0037] Further, in step S134, the loss function is a mean square error loss function, and in the optimal color-based temperature detection model (CTPM), the learning rate is 0.0005, the batch size is 16, and the training rounds are 100.

[0038] In a third aspect, the present invention provides a temperature monitoring system based on a thermally responsive cholesteric liquid crystal elastomer, comprising: A plurality of cholesteric liquid crystal elastomer films, wherein the plurality of cholesteric liquid crystal elastomer films are pasted on black oil paper in a pixel array manner, and are used to characterize the temperature change of the monitored object according to the color change of the cholesteric liquid crystal elastomer films; An imaging device for capturing the color change of the cholesteric liquid crystal elastomer film and forming an optical image; Infrared imaging equipment, used to capture the color change of the cholesteric liquid crystal elastomer film and form an infrared temperature image under conditions corresponding to the optical image; The processor is used to receive the optical image and the infrared temperature image, and input the optical image into a trained temperature mapping model based on a color array for processing, so as to output a temperature distribution map under corresponding conditions.

[0039] In a fourth aspect, the present invention provides a temperature monitoring method of a temperature monitoring system based on a thermally responsive cholesteric liquid crystal elastomer, comprising the following steps: Step S21, after pasting a plurality of cholesteric liquid crystal elastomer films on black oil paper in a pixel array manner to form an array film, the array film is placed on the surface of the object to be monitored; Step S22, using an imaging device to capture the color change of the cholesteric liquid crystal elastomer film and form an optical image; Step S23, using an infrared imaging device to capture the color change of the cholesteric liquid crystal elastomer film and form an infrared temperature image under conditions corresponding to the optical image; Step S24, receiving the optical image and the infrared temperature image through a processor, and inputting the optical image and the infrared temperature image into a trained color array-based temperature mapping model for processing, so as to output a temperature distribution map under corresponding conditions.

[0040] Furthermore, in step S24, the construction of the temperature mapping model based on the color array includes the following steps: Step S241, after placing the red cholesteric liquid crystal elastic film that has been uniaxially stretched and maintained in a stretched state in an oven for baking until the dynamic covalent bonds are completely exchanged, the cholesteric liquid crystal elastic film is taken out of the oven for cooling to room temperature, and the green sample film cut into pieces is pasted on black oil paper in a pixel array manner to form an array film; Step S242, placing the array film on a glass plate, heating different positions with an open flame source, capturing the color change of the heated area of ​​the array film with an imaging device within a preset time to generate an optical image, and preprocessing the optical image to form an RGB image; Specifically, in this step, preprocessing includes rotation, cropping, and scaling to increase data diversity and improve the robustness of the model.

[0041] Step S243, using an infrared imaging device to capture the color change of the array film and form an infrared temperature image under conditions corresponding to the optical image; Step S244, using a generative adversarial neural network architecture to establish a temperature mapping model based on a color array, and inputting a second temperature label file formed by the RGB image and the infrared temperature image into the temperature mapping model based on the color array for iterative training; Specifically, before iterative training, the model needs to be initialized and parameters set.

[0042] Step S245, continuously optimizing the parameters of the temperature mapping model based on the color array through the loss function, evaluating the performance of the temperature mapping model based on the color array through the validation set, and adjusting the hyperparameters of the temperature mapping model based on the color array according to the evaluation results to achieve the optimal temperature mapping model based on the color array.

[0043] Further, in step S245, the loss function is a binary cross entropy loss function, and in the optimal color array-based temperature mapping model (CATMM, including a generator and a discriminator), the generator learning rate is 0.00005, the discriminator learning rate is 0.000005, the batch size is 16, and the training rounds are 100.

[0044] Furthermore, the method for preparing the cholesteric liquid crystal elastomer film comprises the following steps: Step S31, adding a thermally responsive liquid crystal monomer, a chiral dopant, a cross-linking agent, a photoinitiator, a dynamic covalent bond, and a catalyst into an organic solvent in a preset mass ratio for dissolution to form a precursor solution, wherein the mass ratio of the thermally responsive liquid crystal monomer, the chiral dopant, the cross-linking agent, the photoinitiator, the dynamic covalent bond, and the catalyst is: 1107 mg: 46 mg: 270 mg: 10 mg: 136 mg: 3 mg; Specifically, in this step, the thermally responsive liquid crystal monomer is RM257, the chiral dopant is LC756, the cross-linking agent is pentaerythritol tetrakis(3-mercaptopropionate) (PETMP) and 2,2′-(1,2-ethylenedioxy)bis(ethylenethiol) (EDDET), the photoinitiator is benzoin dimethyl ether (HHMP), the dynamic covalent bond is a heat exchange borate bond (BDB), the catalyst is di-n-propylamine (DPA), and it is diluted in toluene at a ratio of 1:50 when used.

[0045] Furthermore, the organic solvent is toluene, the concentration of the thermally responsive liquid crystal monomer and the chiral dopant in the precursor solution is 50%-60%, the dissolution temperature is about 80°C, the stirring speed during dissolution is 300-400r / min, and the stirring time is 10-15min.

[0046] Step S32, pouring the precursor solution into a mold and letting it stand, after the organic solvent evaporates and the cholesteric phase is formed, placing it under ultraviolet light for curing to obtain a completely cross-linked cholesteric liquid crystal elastomer film.

[0047] Specifically, in this step, the size of the cavity in the mold is 20 mm × 40 mm × 0.5 mm, the ambient temperature for the volatilization of the organic solvent and the formation of the cholesteric phase is room temperature, the time is 20-24 hours, and the conditions for UV curing are: UV band 365 nm, UV power 20-30 mW / cm 2 , lighting time 10-15min.

[0048] Step S33, uniaxially stretching the cholesteric liquid crystal elastomer film and maintaining the stretched state at a preset temperature for a preset time to complete dynamic covalent bond exchange, so that the initially red cholesteric liquid crystal elastomer film is oriented along the stretching direction and changes to a corresponding color.

[0049] Specifically, in this step, the preset temperature condition is 70-75°C, and the preset time is 2-6h.

[0050] Furthermore, the cholesteric elastomer film oriented along the stretching direction is manually cut into required shapes and sizes, or combined for later use.

[0051] In a fifth aspect, the present invention provides an application of a temperature monitoring system based on a thermally responsive cholesteric liquid crystal elastomer in monitoring the surface temperature of an electronic device.

[0052] Compared with the prior art, the present invention has the following beneficial effects: First, the preparation process of the cholesteric liquid crystal elastomer film used in the temperature monitoring system of the present invention is simple, the reaction conditions are mild, and it is easy to industrialize the production.

[0053] Secondly, the present invention realizes the visualization of the temperature change feedback of the monitored object through the color change of the cholesteric liquid crystal elastomer film, which greatly enhances the usability and intuitiveness of the monitoring system.

[0054] Third, compared with traditional infrared temperature monitoring equipment, temperature sensors and thermometers, the temperature monitoring method of the present invention has the advantages of low cost, simple structure and high adaptability, and can be cut into various shapes as needed.

[0055] If no specific conditions are specified in the examples of the present invention, the materials are used according to conventional conditions or conditions recommended by the manufacturer. All raw materials, reagents, etc., which do not specify the manufacturer, are conventional products that can be purchased commercially.

[0056] The dynamic covalent bond (BDB) contained in the cholesteric liquid crystal elastomer film (CLCE) in the examples and comparative examples of the present invention is prepared by the following method: 2.0 g of 1-thioglycerol and 1.5 g of 1,4-phenylenediboronic acid were dissolved in 40 ml of tetrahydrofuran, and then 4.0 g of anhydrous magnesium sulfate and 0.1 g of distilled water were added. The mixture was stirred at room temperature for 24 h in a sealed environment. Subsequently, the magnesium sulfate was removed by vacuum filtration, and the solvent was evaporated in an oven at 80 ° C for 12 h. The obtained solid was washed 3-4 times with n-hexane to remove excess 1-thioglycerol. Finally, the solid was dried at 80 ° C for 4 h to obtain a white solid, which was then ground into powder for later use.

[0057] The CLCE of the initial red reflection was prepared by the following method: 1107.0 mg RM257 and 46.0 mg LC756 were mixed with 400 mg toluene under heating and stirring for 10 min. Then, 81.6 mg of pentaerythritol tetrakis(3-mercaptopropionic acid) (PETMP) and 188.4 mg of 2,2′-(1,2-ethylenedioxy)bis(ethylenethiol) (EDDET), 136.0 mg of dynamic covalent bond (BDB), 10.0 mg of photoinitiator benzoin dimethyl ether (HHMP) and 150 mg of catalyst di-n-propylamine (DPA) (diluted in toluene at a ratio of 1:50) were added to the solution. The solution was stirred at 60 ° C for 10 min and ultrasonicated for 1 min. Then, the precursor solution was poured into a mold (20 mm × 40 mm × 0.5 mm), placed in a fume hood, and reacted in a light-proof environment until a cholesteric phase was formed and the solvent was completely evaporated. Afterwards, the cholesteric LC was exposed to 365 nm, 20 mW / cm² UV light for 10 min to obtain a fully cross-linked CLCE film. Example

[0058] (1) The red cholesteric liquid crystal elastic film was uniaxially stretched to a tensile strain of 60%, and the film was kept stretched and placed in an oven at 70°C. After about 2 hours, when the dynamic bonds were completely exchanged, the film was taken out of the oven and cooled to room temperature. This process caused the initially red CLCE film to be oriented along the stretching direction and turned green. The film was cut into rectangular films of 10 mm × 20 mm for later use.

[0059] (2) Place the green film on a 100°C hot stage. The film will change from green to red within 10 seconds. This process is captured using a smartphone or digital camera to generate an optical image and record the corresponding temperature conditions to form a label file and training data set for subsequent model training.

[0060] (3) A color-based temperature detection model (CTPM) was established based on the convolutional neural network (CNN) architecture. The specific model network architecture is shown in Table 1. Based on the loss curve and output results generated during the training process, the model training parameters and training data sets were continuously adjusted. After multiple training attempts, a better training effect was achieved when the learning rate was 0.0005, the batch size was 16, and the training round (epoch) was 100.

[0061]

[0062] (4) The trained model will be used to identify CLCE and output the temperature value under the corresponding conditions based on its color. The CLCE film is placed on a high-temperature surface, and the temperature range of the surface is preliminarily determined by observing its color change. Then, an image is taken with a smartphone or digital camera and input into the trained model. The model will output the accurate temperature value under the corresponding conditions based on the input image. The above process realizes the temperature detection function.

[0063] (5) After the temperature detection is completed, the CLCE film will automatically restore its original color for next use.

[0064] Figure 1 This is the preliminary temperature detection function achieved by the CLCE film prepared in Example 1. The approximate temperature of the surface of an object can be preliminarily judged based on its color. Red (left) represents a high temperature danger on the surface of the object, while green (right) represents a safe room temperature on the surface of the object.

[0065] Figure 2 It is a flowchart of realizing the temperature detection function based on CLCE and machine vision technology in Example 1 and a network architecture diagram of a color-based temperature detection model (CTPM).

[0066] Figure 3 and Figure 4 This is the training result of the model in Example 1, which shows that the training effect of the model is good and the temperature detection is accurate, with an accuracy rate of more than 95% in the range of 50-70°C.

[0067] (1) The red cholesteric liquid crystal elastic film was uniaxially stretched to a tensile strain of 120%, and the film was kept stretched and placed in an oven at 70°C. After about 2 hours, when the dynamic bonds were completely exchanged, the film was taken out of the oven and cooled to room temperature. This process caused the initially red CLCE film to be oriented along the stretching direction and turned blue. The film was cut into rectangular films of 10 mm × 20 mm for later use.

[0068] (2) Place the blue film on a 100°C hot stage. The film will change from blue to green within 10 seconds and finally back to red. This process is captured using a smartphone or digital camera to generate an optical image and the corresponding temperature conditions are recorded to form a label file and a training data set for subsequent model training.

[0069] (3) A color-based temperature detection model (CTPM) was established based on the convolutional neural network (CNN) architecture. The detailed model network architecture is shown in Table 1. Based on the loss curve and output results generated during the training process, the model training parameters and training data sets were continuously adjusted. After multiple training attempts, a better training effect was achieved when the learning rate was 0.0005, the batch size was 16, and the training epoch was 100.

[0070] (4) The trained model will be used to identify CLCE and output the temperature value under the corresponding conditions based on its color. The CLCE film is placed on a high-temperature surface, and the temperature range of the surface is preliminarily determined by observing its color change. Then, an image is taken with a smartphone or digital camera and input into the trained model. The model will output the accurate temperature value under the corresponding conditions based on the input image. The above process realizes the temperature detection function.

[0071] (5) After the temperature detection is completed, the CLCE film will automatically restore its original color for next use.

[0072] Figure 5 It shows that the CLCE film gradually changes from blue to red as the temperature gradually increases. Figure 6 It indicates that the model training effect is good. Compared with the training results in Example 1, the training results of Comparative Example 1 in the range of 50-70°C are better. The reason is that the color change process of the film is different from the gradual change from green to red in Example 1. In Comparative Example 1, the color of the film will gradually change from blue to green and then to red as the temperature rises, especially in the range of 50-70°C. The image features given to machine vision recognition will be more obvious than in Example 1, so a better training effect can be obtained. Example

[0073] (1) The red cholesteric liquid crystal elastic film was uniaxially stretched to a tensile strain of 60%, and the film was kept stretched and placed in an oven at 70°C. After about 2 hours, when the dynamic bonds were completely exchanged, the film was taken out of the oven and cooled to room temperature. This process caused the initially red CLCE film to be oriented along the stretching direction and turned green. The film was cut into 2 mm × 2 mm "pixels" and the "pixels" were pasted on black oil paper using double-sided tape to form a 10 × 10 CLCE array.

[0074] (2) Place the above array on a glass plate and use an open flame source to heat different positions. The heated area will change color. The above process uses a smartphone or digital camera to capture and generate an optical image, while an infrared camera is used to record the red temperature image under the corresponding conditions. The optical image and the corresponding infrared temperature image form a label file and a training data set for subsequent model training.

[0075] (3) Based on the generative adversarial neural network (GAN) architecture, a color array-based temperature mapping model (CATMM) was established. The model mainly consists of two parts: the generator and the discriminator. The specific network architectures are shown in Tables 2 and 3, respectively. Based on the loss curve and output results generated during the training process, the parameters of the model training and the training data set were continuously adjusted. After multiple training attempts, when the generator learning rate was 0.00005, the discriminator learning rate was 0.000005, the batch size was 16, and the training round (epoch) was 100, a better training effect was achieved.

[0076]

[0077]

[0078] (4) The trained model will be used to identify CLCE and output the temperature distribution map under the corresponding conditions according to its color. Use a glass plate to simulate the surface of a high-temperature object, use an open flame source to heat and simulate the local high temperature of an object. Place the CLCE array on the glass plate, and use a smartphone or digital camera to shoot the color change process. The video takes a frame every 15 seconds and inputs it into the above-trained model. The model will generate the corresponding temperature distribution image based on the input image. The above process realizes the temperature monitoring function.

[0079] (5) After the temperature monitoring is completed, the CLCE array will automatically restore its original color for next use. The ideal effect can also be obtained by changing the heating position of the open flame source in step (2) and monitoring again.

[0080] Figure 7 This is a flow chart of preparing the CLCE array in Example 2.

[0081] Figure 8 Schematic diagram of the network architecture of the color array-based temperature mapping model (CATMM). Fig. 9 It is the loss curve of the generator and discriminator during the training process. The curve results show that the training effect of the model is good.

[0082] Fig.10 It is a flow chart of simulating local high temperature on the surface of an object and using the CLCE array and deep learning model for temperature monitoring. Fig.11 It is the comparison result of capturing an image every 15 seconds during the temperature monitoring process and outputting a temperature distribution image. The "real image" is the infrared temperature image under the corresponding conditions. The results show that the model can output a temperature distribution image that is highly consistent with the infrared temperature image under the corresponding conditions, proving the accuracy of the model recognition.

[0083] Fig.12 The results of temperature monitoring using the CLCE array under different local heating positions show that the model has strong generalization ability.

[0084] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A temperature monitoring system based on a thermally responsive cholesteric liquid crystal elastomer, characterized in that: include: Cholesteric liquid crystal elastomer film, used to characterize the temperature change of the monitored object based on its own color change; An imaging device for capturing the color change of the cholesteric liquid crystal elastomer film and forming an optical image; The processor is used to receive the optical image and input the optical image into a trained color-based temperature detection model for processing to output a temperature value under corresponding conditions.

2. A temperature monitoring method of a temperature monitoring system based on a thermally responsive cholesteric liquid crystal elastomer according to claim 1, characterized in that: The following steps are involved: Step S11, placing a cholesteric liquid crystal elastomer film on the surface of the object to be monitored; Step S12, using an imaging device to capture the color change of the cholesteric liquid crystal elastomer film and form an optical image; Step S13, receiving the optical image through a processor, and inputting the optical image into a trained color-based temperature detection model for processing to output a temperature value under corresponding conditions.

3. The temperature monitoring method of the temperature monitoring system based on the thermally responsive cholesteric liquid crystal elastomer according to claim 2, characterized in that: In step S13, the construction of the color-based temperature detection model includes the following steps: Step S131, after placing the red cholesteric liquid crystal elastic film that has been uniaxially stretched and maintained in a stretched state in an oven for baking until the dynamic covalent bonds are completely exchanged, the cholesteric liquid crystal elastic film is taken out of the oven for cooling to room temperature, and cut into blue sample films; Step S132, placing the blue sample film on a hot stage, capturing the color change of the blue sample film by an imaging device within a preset time to generate an optical image, and preprocessing the optical image to form an RGB image, and simultaneously recording the corresponding temperature conditions to form a first temperature label file; Step S133, using a convolutional neural network architecture to establish a color-based temperature detection model, and inputting the first temperature label file into the color-based temperature detection model for iterative training; Step S134, continuously optimizing the parameters of the color-based temperature detection model through the loss function, evaluating the performance of the color-based temperature detection model through the validation set, and adjusting the hyperparameters of the color-based temperature detection model according to the evaluation results to achieve the optimal color-based temperature detection model.

4. The temperature monitoring method of the temperature monitoring system based on the thermally responsive cholesteric liquid crystal elastomer according to claim 3, characterized in that: In step S134, the loss function is the mean square error loss function, and in the optimal color-based temperature detection model, the learning rate is 0.0005, the batch size is 16, and the training rounds are 100.

5. A temperature monitoring system based on a thermally responsive cholesteric liquid crystal elastomer, characterized in that: include: A plurality of cholesteric liquid crystal elastomer films, wherein the plurality of cholesteric liquid crystal elastomer films are pasted on black oil paper in a pixel array manner, and are used to characterize the temperature change of the monitored object according to the color change of the cholesteric liquid crystal elastomer films; An imaging device for capturing the color change of the cholesteric liquid crystal elastomer film and forming an optical image; Infrared imaging equipment, used to capture the color change of the cholesteric liquid crystal elastomer film and form an infrared temperature image under conditions corresponding to the optical image; The processor is used to receive the optical image and the infrared temperature image, and input the optical image into a trained temperature mapping model based on a color array for processing, so as to output a temperature distribution map under corresponding conditions.

6. A temperature monitoring method of a temperature monitoring system based on a thermally responsive cholesteric liquid crystal elastomer according to claim 5, characterized in that: The following steps are involved: Step S21, after pasting a plurality of cholesteric liquid crystal elastomer films on black oil paper in a pixel array manner to form an array film, the array film is placed on the surface of the object to be monitored; Step S22, using an imaging device to capture the color change of the cholesteric liquid crystal elastomer film and form an optical image; Step S23, using an infrared imaging device to capture the color change of the cholesteric liquid crystal elastomer film and form an infrared temperature image under conditions corresponding to the optical image; Step S24, receiving the optical image and the infrared temperature image through a processor, and inputting the optical image and the infrared temperature image into a trained color array-based temperature mapping model for processing, so as to output a temperature distribution map under corresponding conditions.

7. The temperature monitoring method of the temperature monitoring system based on the thermally responsive cholesteric liquid crystal elastomer according to claim 6, characterized in that: In step S24, the construction of the temperature mapping model based on the color array includes the following steps: Step S241, after placing the red cholesteric liquid crystal elastic film that has been uniaxially stretched and maintained in a stretched state in an oven for baking until the dynamic covalent bonds are completely exchanged, the cholesteric liquid crystal elastic film is taken out of the oven for cooling to room temperature, and the green sample film cut into pieces is pasted on black oil paper in a pixel array manner to form an array film; Step S242, placing the array film on a glass plate, heating different positions with an open flame source, capturing the color change of the heated area of ​​the array film with an imaging device within a preset time to generate an optical image, and preprocessing the optical image to form an RGB image; Step S243, using an infrared imaging device to capture the color change of the array film and form an infrared temperature image under conditions corresponding to the optical image; Step S244, using a generative adversarial neural network architecture to establish a temperature mapping model based on a color array, and inputting a second temperature label file formed by the RGB image and the infrared temperature image into the temperature mapping model based on the color array for iterative training; Step S245, continuously optimizing the parameters of the temperature mapping model based on the color array through the loss function, evaluating the performance of the temperature mapping model based on the color array through the validation set, and adjusting the hyperparameters of the temperature mapping model based on the color array according to the evaluation results to achieve the optimal temperature mapping model based on the color array.

8. The temperature monitoring method of the temperature monitoring system based on the thermally responsive cholesteric liquid crystal elastomer according to claim 7, characterized in that: In step S245, the loss function is a binary cross entropy loss function, and in the optimal color array-based temperature mapping model, the generator learning rate is 0.00005, the discriminator learning rate is 0.000005, the batch size is 16, and the training rounds are 100.

9. The temperature monitoring system based on the thermally responsive cholesteric liquid crystal elastomer according to claim 1 or 5, characterized in that: The method for preparing the cholesteric liquid crystal elastomer film comprises the following steps: Step S31, adding a thermally responsive liquid crystal monomer, a chiral dopant, a cross-linking agent, a photoinitiator, a dynamic covalent bond, and a catalyst into an organic solvent in a preset mass ratio for dissolution to form a precursor solution, wherein the mass ratio of the thermally responsive liquid crystal monomer, the chiral dopant, the cross-linking agent, the photoinitiator, the dynamic covalent bond, and the catalyst is: 1107 mg: 46 mg: 270 mg: 10 mg: 136 mg: 3 mg; Step S32, pouring the precursor solution into a mold and letting it stand, and after the organic solvent evaporates and the cholesteric phase is formed, placing it under ultraviolet light for curing to obtain a completely cross-linked cholesteric liquid crystal elastomer film; Step S33, uniaxially stretching the cholesteric liquid crystal elastomer film, and maintaining the stretched state at a preset temperature for a preset time to complete dynamic covalent bond exchange, so that the initially red cholesteric liquid crystal elastomer film is oriented along the stretching direction and changes to a corresponding color.

10. Use of the temperature monitoring system based on the thermally responsive cholesteric liquid crystal elastomer according to claim 1 or 5 in monitoring the surface temperature of electronic equipment.

Images