Solid waste heat storage body data generation method based on material intrinsic signal decoupling

Abstract

The invention discloses a solid waste heat storage body data generation method based on material intrinsic signal decoupling. The solid waste heat storage body data generation method comprises the steps of S1, thermally induced microstructure acoustic emission capturing; s2, carrying out electromagnetic eddy current attenuation inversion; s3, a phase change grating memory effect is achieved; s4, generating thermodynamic prior knowledge injection based on confrontation data of physical constraints; s5, generating a framework through double-channel confrontation; s6, carrying out microscopic-macroscopic correlation modeling; and S7, reinforcement learning of the failure critical point.

Description

Technical Field

[0001] This invention relates to the generation of high-temperature performance data for industrial materials, and more specifically, to a method for generating data on solid waste thermal storage bodies based on the decoupling of intrinsic material signals. Background Technology

[0002] In the field of generating high-temperature performance data for solid waste thermal storage bodies, traditional data acquisition methods are limited by the high-temperature failure of external equipment and the challenge of lacking thermodynamic credibility when generating virtual data.

[0003] Although Chinese invention patent applications CN114067915A and CN120296690A have proposed innovative data processing methods in their respective fields, such as the application of deep learning in single-cell RNA sequencing data mining and the short-term prediction method of marine environment based on conditional adversarial networks, they have not solved the problem of generating material performance data under high temperature conditions.

[0004] In particular, these methods fail to adequately consider the utilization of intrinsic material signals and the constraints of physical rules when generating data, resulting in deficiencies in the real-time performance, accuracy, and cost-effectiveness of data under high-temperature conditions. Summary of the Invention

[0005] This invention provides a method for generating solid waste thermal storage body data based on decoupling of intrinsic material signals, in order to solve the technical problems existing in the prior art.

[0006] To achieve the above objectives, the present invention provides a method for generating solid waste thermal storage body data based on decoupling of intrinsic material signals, comprising: Step S1. Thermally induced acoustic emission capture of microstructures: A non-metallic resonant cavity is pre-embedded inside the material. The cavity structure expands with temperature to generate characteristic audio frequency signals. The sound waves in the 20-100kHz frequency band are captured by an external piezoelectric ceramic receiver. A cavity resonant frequency-temperature mapping model is established to achieve single-point temperature measurement. Step S2. Electromagnetic eddy current attenuation inversion: By applying a 10-100MHz alternating electromagnetic field to the surface of the material, the eddy current loss generated by the metal oxides in the solid waste is inverted through the attenuation coefficient θ=exp(-α·T), thus realizing passive temperature measurement of the entire area. Here, α is the linear thermal expansion coefficient of the metal oxide, and T is the temperature. Step S3. Phase change grating memory effect: Thermosensitive color-changing microcapsules are embedded in the material forming stage. Rare earth fluorescent substances are released at temperatures above 800°C. After cooling, the fluorescence distribution is captured by laser scanning to reconstruct the high-temperature thermal history. Step S4. Injecting thermodynamic prior knowledge into adversarial data generation based on physical constraints: Construct piecewise continuous constraint equations for the material's thermal expansion coefficient α(T) and specific heat capacity C(T), and force the generated data to satisfy the physical rule dα / dT≤0; Step S5. Dual-channel adversarial generative architecture: Generator G: Inputs measured low-temperature data and outputs predicted high-temperature parameters; Discriminator D: Simultaneously receives real high-temperature data and generated data, and uses the second law of thermodynamics as the judgment criterion to train the generator G through game theory to generate thermodynamically feasible and undistinguishable data. Step S6. Micro-macro correlation modeling: A convolutional network was trained using scanning electron microscope images to establish a pore structure-thermal conductivity correlation model. High-temperature thermal conductivity data was generated by taking room-temperature microscopic images as input. Step S7. Reinforcement learning for failure critical points: A material thermal fatigue life prediction model is constructed. When the generated data approaches the phase transition temperature, data augmentation is automatically triggered to generate critical failure data in the temperature-stress coordinate space.

[0007] In an optional embodiment of the present invention, in step S1, the metal resonant cavity is made of AlO / SiC composite material and is resistant to temperature of 1600°C.

[0008] In an optional embodiment of the present invention, optionally, in step S1, the temperature resolution in the frequency-temperature mapping model is ±5°C.

[0009] In an optional embodiment of the present invention, optionally, in step S1 and step S2, the metal oxides in the solid waste are FeO and / or CuO.

[0010] In an optional embodiment of the present invention, optionally, in step S1 and step S3, the particle size of the thermochromic microcapsules is between 50-200 μm.

[0011] In an optional embodiment of the present invention, optionally, in step S1 and step S5, the low-temperature measured data is obtained at a temperature below 500°C.

[0012] In an optional embodiment of the present invention, optionally, in step S1 and step S6, the scanning electron microscope is a SEM (scanning electron microscope).

[0013] In an optional embodiment of the present invention, optionally, in step S1, the acoustic emission sensor receives the ultrasonic waves emitted by the microstructure through a hydrophone coupling array.

[0014] In an optional embodiment of the present invention, in step S1, step 2 is determined by an eddy current attenuation experiment.

[0015] In an optional embodiment of the present invention, optionally, in step S1, the material is silicon carbide zirconium solid waste.

[0016] The solid waste thermal storage body data generation method based on material intrinsic signal decoupling provided by this invention has the following beneficial technical effects: First, the present invention provides a method for generating high-temperature performance data of solid waste thermal storage bodies based on the decoupling and adversarial generation of intrinsic material signals. By building a pore-thermal conductivity correlation model through a convolutional neural network, it realizes the generation of high-precision thermal conductivity data from high-definition SEM images with low energy consumption. Compared with traditional methods, it can significantly improve the reliability and accuracy of predicting the high-temperature performance of solid waste materials.

[0017] Secondly, this invention uses the original "material itself as a sensor" paradigm to invent a physical constraint-resistance generation method for thermal expansion coefficient and specific heat capacity, which significantly reduces costs and simplifies the process, enabling the acquisition of solid waste thermal storage body performance data at the minute level, all-weather, and with high precision.

[0018] Third, this invention, by leveraging deep learning technology, has for the first time achieved the reverse engineering of high-temperature performance data of solid waste, and has developed high-throughput computing capabilities that can adapt to different materials and conditions, resulting in excellent economic and social benefits. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the system and sensor principles implemented in this invention.

[0021] Figure labeling: 1-Non-metallic resonant cavity (Al2O3 / SiC composite material, temperature resistance 1600°C); 2-Thermosensitive color-changing microcapsules (particle size 50-200μm); 3-Piezoelectric ceramic receiver (acoustic emission capture); 4-Electromagnetic field generator (eddy current attenuation inversion); 5-Laser scanner (phase change grating memory effect); 6-AI data processing and generation unit. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below 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.

[0023] This invention achieves full-area passive temperature measurement and high-temperature thermal history reconstruction by pre-embedding a non-metallic resonant cavity inside the material, using acoustic emission signals to capture changes in the material's internal microstructure, and combining electromagnetic eddy current attenuation inversion and phase change grating memory effect.

[0024] Meanwhile, by constructing an adversarial data generation model with physical constraints, the generated data is ensured to meet thermodynamic rules, thus solving the problem of thermodynamic credibility of virtual data.

[0025] Furthermore, this invention also uses failure threshold reinforcement learning to generate critical failure data in a targeted manner, which further improves the data acquisition frequency and the fidelity of high-temperature data, while significantly reducing the data acquisition cost.

[0026] Example 1: This invention provides a method for generating solid waste thermal storage body data based on decoupling of intrinsic material signals, comprising: Step S1. Thermally induced acoustic emission capture of microstructures: A non-metallic resonant cavity is pre-embedded inside the material. The cavity structure expands with temperature to generate characteristic audio frequency signals. The sound waves in the 20-100kHz frequency band are captured by an external piezoelectric ceramic receiver. A cavity resonant frequency-temperature mapping model is established to achieve single-point temperature measurement. Step S2. Electromagnetic eddy current attenuation inversion: By applying a 10-100MHz alternating electromagnetic field to the surface of the material, the eddy current loss generated by the metal oxides in the solid waste is inverted through the attenuation coefficient θ=exp(-α·T), thus realizing passive temperature measurement of the entire area. Here, α is the linear thermal expansion coefficient of the metal oxide, and T is the temperature. Step S3. Phase change grating memory effect: Thermosensitive color-changing microcapsules are embedded in the material forming stage. Rare earth fluorescent substances are released at temperatures above 800°C. After cooling, the fluorescence distribution is captured by laser scanning to reconstruct the high-temperature thermal history. Step S4. Injecting thermodynamic prior knowledge into adversarial data generation based on physical constraints: Construct piecewise continuous constraint equations for the material's thermal expansion coefficient α(T) and specific heat capacity C(T), and force the generated data to satisfy the physical rule dα / dT≤0; Step S5. Dual-channel adversarial generative architecture: Generator G: Inputs measured low-temperature data and outputs predicted high-temperature parameters; Discriminator D: Simultaneously receives real high-temperature data and generated data, and uses the second law of thermodynamics as the judgment criterion to train the generator G through game theory to generate thermodynamically feasible and undistinguishable data. Step S6. Micro-macro correlation modeling: A convolutional network was trained using scanning electron microscope images to establish a pore structure-thermal conductivity correlation model. High-temperature thermal conductivity data was generated by taking room-temperature microscopic images as input. Step S7. Reinforcement learning for failure critical points: A material thermal fatigue life prediction model is constructed. When the generated data approaches the phase transition temperature, data augmentation is automatically triggered to generate critical failure data in the temperature-stress coordinate space.

[0027] In an optional embodiment of the present invention, in step S1, the metal resonant cavity is made of AlO / SiC composite material and is resistant to temperature of 1600°C.

[0028] In an optional embodiment of the present invention, optionally, in step S1, the temperature resolution in the frequency-temperature mapping model is ±5°C.

[0029] In an optional embodiment of the present invention, optionally, in step S1 and step S2, the metal oxides in the solid waste are FeO and / or CuO.

[0030] In an optional embodiment of the present invention, optionally, in step S1 and step S3, the particle size of the thermochromic microcapsules is between 50-200 μm.

[0031] In an optional embodiment of the present invention, optionally, in step S1 and step S5, the low-temperature measured data is obtained at a temperature below 500°C.

[0032] In an optional embodiment of the present invention, optionally, in step S1 and step S6, the scanning electron microscope is a SEM (scanning electron microscope).

[0033] In an optional embodiment of the present invention, optionally, in step S1, the acoustic emission sensor receives the ultrasonic waves emitted by the microstructure through a hydrophone coupling array.

[0034] In an optional embodiment of the present invention, in step S1, step 2 is determined by an eddy current attenuation experiment.

[0035] In an optional embodiment of the present invention, optionally, in step S1, the material is silicon carbide zirconium solid waste.

[0036] Example 2: The specific implementation process for generating data on the thermal storage performance of solid waste in the silicon carbide zirconium system is as follows: S1. A non-metallic resonant cavity is pre-embedded inside the silicon carbide zirconium solid waste. As the temperature rises, the cavity expands and the internal pressure increases. When the internal pressure is greater than the bonding strength of the external glass slide, the acoustic signal is captured by the piezoelectric ceramic chip. The resonance frequency f-temperature F mapping relationship is obtained by performing Fourier transform processing on the acoustic spectrum. S2. The internal temperature of solid waste is inverted by electromagnetic eddy current attenuation method. The passive temperature measurement result obtained by fitting the dielectric constant ε-temperature T is used as the true value label. Adversarial data is generated and adversarial training is carried out using real data and generated data. S3. During the solid waste molding stage, rare earth photoresist is filled to prepare thermosensitive color-changing microcapsules. The capsules are placed at a temperature higher than the phase transition point for reaction and then cooled. The distribution of fluorescent substances is captured by laser scanning to reconstruct the high-temperature thermal history of the material. S4. Physically Constrained Adversarial Data Generation: Construct piecewise continuous constraint equations for the specific heat capacity C(T) and thermal expansion coefficient α(T) of the material, and force the generated data to satisfy the boundary dα / dT≤0; S5. Enhance the diversity of solid waste surface thermal conductivity data by maliciously attacking it. Conduct adversarial training based on the measured sparse high-temperature data and the generated data to make the generated data more diverse while ensuring accuracy. S6. Using powder images as input, a porosity-thermal conductivity correlation model is constructed through a deep neural network to generate thermal conductivity data. S7. Based on our unique cloud platform multidisciplinary coupling theory (from top to bottom: field layer, feature layer, law layer, essence layer), the thermal fatigue life of materials is predicted by reinforcement learning algorithm under the constraint of the second law of thermodynamics. When the generated data gets closer and closer to the phase transition point, the data augmentation mechanism is triggered to generate more samples located near the phase transition point.

[0037] Figure 1 This is a schematic diagram of the system and sensor principle implemented in this invention, wherein 1 is a non-metallic resonant cavity (Al2O3 / SiC composite material, temperature resistance 1600°C), 2 is a thermosensitive color-changing microcapsule (particle size 50-200μm), 3 is a piezoelectric ceramic receiver (acoustic emission capture), 4 is an electromagnetic field generator (eddy current attenuation inversion), 5 is a laser scanner (phase change grating memory effect), and 6 is an AI data processing and generation unit.

[0038] The solid waste thermal storage body data generation method based on material intrinsic signal decoupling provided by this invention has the following beneficial technical effects: First, the present invention provides a method for generating high-temperature performance data of solid waste thermal storage bodies based on the decoupling and adversarial generation of intrinsic material signals. By building a pore-thermal conductivity correlation model through a convolutional neural network, it realizes the generation of high-precision thermal conductivity data from high-definition SEM images with low energy consumption. Compared with traditional methods, it can significantly improve the reliability and accuracy of predicting the high-temperature performance of solid waste materials.

[0039] Secondly, this invention uses the original "material itself as a sensor" paradigm to invent a physical constraint-resistance generation method for thermal expansion coefficient and specific heat capacity, which significantly reduces costs and simplifies the process, enabling the acquisition of solid waste thermal storage body performance data at the minute level, all-weather, and with high precision.

[0040] Third, this invention, by leveraging deep learning technology, has for the first time achieved the reverse engineering of high-temperature performance data of solid waste, and has developed high-throughput computing capabilities that can adapt to different materials and conditions, resulting in excellent economic and social benefits.

[0041] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of one embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention.

[0042] Those skilled in the art will understand that the modules in the apparatus of the embodiments can be distributed in the apparatus of the embodiments as described in the embodiments, or they can be located in one or more devices different from this embodiment with corresponding changes. The modules of the above embodiments can be combined into one module, or they can be further divided into multiple sub-modules.

[0043] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for generating solid waste thermal storage body data based on decoupling of intrinsic material signals, characterized in that, include: Step S1. Thermally induced acoustic emission capture of microstructures: A non-metallic resonant cavity is pre-embedded inside the material. The cavity structure expands with temperature to generate characteristic audio frequency signals. The sound waves in the 20-100kHz frequency band are captured by an external piezoelectric ceramic receiver. A cavity resonant frequency-temperature mapping model is established to achieve single-point temperature measurement. Step S2. Electromagnetic eddy current attenuation inversion: By applying a 10-100MHz alternating electromagnetic field to the surface of the material, the eddy current loss generated by the metal oxides in the solid waste is inverted through the attenuation coefficient θ=exp(-α·T), thus realizing passive temperature measurement of the entire area. Here, α is the linear thermal expansion coefficient of the metal oxide, and T is the temperature. Step S3. Phase change grating memory effect: Thermosensitive color-changing microcapsules are embedded in the material forming stage. Rare earth fluorescent substances are released at temperatures above 800°C. After cooling, the fluorescence distribution is captured by laser scanning to reconstruct the high-temperature thermal history. Step S4. Injecting thermodynamic prior knowledge into adversarial data generation based on physical constraints: Construct piecewise continuous constraint equations for the material's thermal expansion coefficient α(T) and specific heat capacity C(T), and force the generated data to satisfy the physical rule dα / dT≤0; Step S5. Dual-channel adversarial generative architecture: Generator G: Inputs measured low-temperature data and outputs predicted high-temperature parameters; Discriminator D: Simultaneously receives real high-temperature data and generated data, and uses the second law of thermodynamics as the judgment criterion to train the generator G through game theory to generate thermodynamically feasible and undistinguishable data. Step S6. Micro-macro correlation modeling: A convolutional network was trained using scanning electron microscope images to establish a pore structure-thermal conductivity correlation model. High-temperature thermal conductivity data was generated by taking room-temperature microscopic images as input. Step S7. Reinforcement learning for failure critical points: A material thermal fatigue life prediction model is constructed. When the generated data approaches the phase transition temperature, data augmentation is automatically triggered to generate critical failure data in the temperature-stress coordinate space.

2. The method for generating solid waste thermal storage body data based on decoupling of intrinsic material signals according to claim 1, characterized in that, In step S1, the metal resonant cavity is made of AlO / SiC composite material and is resistant to temperature up to 1600°C.

3. The method for generating solid waste thermal storage body data based on decoupling of intrinsic material signals according to claim 1, characterized in that, In step S1, the temperature resolution in the frequency-temperature mapping model is ±5°C.

4. The method for generating solid waste thermal storage body data based on decoupling of intrinsic material signals according to claim 1, characterized in that, In step S2, the metal oxides in the solid waste are FeO and / or CuO.

5. The method for generating solid waste thermal storage body data based on decoupling of intrinsic material signals according to claim 1, characterized in that, In step S3, the particle size of the thermochromic microcapsules is between 50-200 μm.

6. The method for generating solid waste thermal storage body data based on decoupling of intrinsic material signals according to claim 1, characterized in that, In step S5, the low-temperature measured data were obtained at temperatures below 500°C.

7. The method for generating solid waste thermal storage body data based on decoupling of intrinsic material signals according to claim 1, characterized in that, In step S6, the scanning electron microscope is a scanning electron microscope (SEM).

8. The method for generating solid waste thermal storage body data based on decoupling of intrinsic material signals according to claim 1, characterized in that, In step S1, the acoustic emission sensor receives the ultrasonic waves emitted by the microstructure through a hydrophone coupling array.

9. The method for generating solid waste thermal storage body data based on decoupling of intrinsic material signals according to claim 1, characterized in that, In step 2, the eddy current attenuation experiment was used for determination.

10. The method for generating solid waste thermal storage body data based on decoupling of intrinsic material signals according to claim 1, characterized in that, In step S1, the material is silicon carbide zirconium solid waste.

Images