45results about How to "Reduced characteristics" patented technology

The invention relates to the field of blood products, in particular to a preparation method of human activated C1 esterase. The preparation method comprises the following steps: carrying out contact adsorption on cryoprecipitate-removed human plasma and anion exchange resin, washing and eluting to obtain an eluent containing C1 proesterase, carrying out membrane filtration, and carrying out ultrafiltration desalination treatment; adding an anionic substance into the eluent, incubating in vitro to effectively activate the C1 esterase, and filtering by a membrane to obtain a feed liquid containing the activated C1 esterase; and carrying out purification treatment on the feed liquid to obtain the human activated C1 esterase, and carrying out cryopreservation. According to the method, an anion substance is introduced into an eluent system obtained through anion exchange, mild and effective activation of the C1 proesterase is achieved under the condition of not depending on foreign proteolytic enzyme, the risk of non-specific hydrolysis and self-cutting inactivation is remarkably reduced, good synergy is formed with a follow-up purification process, and the method is suitable for industrial production. The human activated C1 esterase product with relatively high activity and good stability is obtained.

The application discloses an article recommendation method and related device, and relates to the field of article recommendation. According to the article ID contained in the article recommendation request, a positive query result is obtained. If the positive query meets the first missing level, and the incremental benefit evaluated based on the article recommendation scene data is not less than the first benefit threshold, a semantic vector is generated according to the article key field to recall a first group of candidate articles similar to the current article from a vector library, a second group of candidate articles corresponding to the current article is recalled from other multiple different databases, multi-dimensional feature fusion evaluation and sorting are performed on each candidate article, and the top N articles are selected as recommended articles. The application reduces invalid filling of low-value recommendation requests through information missing level and incremental benefit evaluation. When the positive feature is missing, candidate articles are recalled from the vector library and multiple different databases outside the vector library, multi-dimensional feature fusion evaluation and sorting are performed on each candidate article, and article recommendation is realized.

The invention relates to the technical field of computer vision and coal mine intelligent mining, and discloses a lightweight working face three-dimensional reconstruction method and device fusing geometric prior constraints, and the method comprises the steps: firstly screening a dynamic effective region based on a projection relation and texture features, and removing static redundant scenes; performing sparse visual reasoning and depth regression only for the effective area to generate an initial depth map; carrying out physical structure correction on the depth map by utilizing geometric priori such as vertical upright posts and parallel top plates of the hydraulic support; and finally, constructing an optimization objective function containing deviation constraints of the vertical column perpendicularity and the top plate parallelism, and completing high-precision fusion of the incremental point cloud. According to the method, inference delay is reduced by reducing invalid calculation, structural distortion and registration drift caused by an underground severe environment are effectively inhibited by using a geometric law, and real-time and high-fidelity reconstruction of a working face scene is realized.

The application discloses a feature placement method and system for multi-GPU sampling graph neural network training, uniformly models a sampling graph neural network training process, and obtains hardware topology information; in a pre-sampling process, the number of times of accessing node features is counted as node heat, the node features are sorted according to the node heat, the node features are divided into a plurality of feature blocks according to a single-block capacity, and the feature blocks are divided into hot blocks or cold blocks according to block heat; based on the feature block set obtained by the division and the obtained hardware topology information, a mixed integer programming model is constructed, with the minimum bottleneck link time in a feature extraction stage and the memory load imbalance degree as targets, the mixed integer programming model is solved, and the placement position, access path and link flow of the globally jointly optimized hot block are obtained. The application can obtain a high-quality feature placement scheme under a given hardware constraint, minimize the extraction time, and improve the resource utilization rate of a multi-GPU system.

This utility model relates to a PE foam car cargo box mat, comprising: a base plate, an insert groove inside the base plate, multiple sets of mounting holes on both sides of the bottom of the insert groove, six sets of positioning slots on both sides of the top of the base plate, a set of snap-fit ​​components on both sides of the inner wall of the positioning slots, a set of limiting slots on both sides of the inner wall of the insert groove between adjacent sets of positioning slots, multiple mat body parts inserted into the insert groove at positions corresponding to the limiting slots, pressure plate grooves on both sides of the top of the mat body and at the center of the mat body, multiple sets of positioning insertion holes at the bottom of the pressure plate grooves on both sides, and a pressure plate bracket inserted into the top of the base plate above the mat body. The beneficial effects of this utility model are: the design facilitates the disassembly and replacement of damaged areas of the mat, enhancing overall practicality.

This invention relates to the field of image restoration technology, specifically disclosing an image restoration method and device based on discrete wavelet transform. An image restoration network, called Multi-Wavelet Attention Memory Neural Network (MWA-MNN), is designed, consisting of three sub-networks. These sub-networks include Discrete Wavelet Attention Blocks (DWABs) and Coordinate Channel Attention Blocks (CCABs) to extract spatial and channel feature information, effectively separating degraded image layers such as rain streaks from the image to obtain a clean image. The invention leverages the expressive power of wavelet transform for high-frequency information, thereby improving image restoration performance. The sub-networks also include Supervised Selective Kernel Blocks (SSKBs) for filtering effectively extracted features in each sub-network. This invention also proposes a memristor-based cross-array MWA-MNN circuit implementation scheme. Memristors are a novel type of non-volatile memory with faster read speeds and lower power consumption. Experimental results show that this method has good performance and versatility in various image restoration tasks.

The invention discloses a photoelectric sensor system conversion efficiency calibration system and method based on a reverberation chamber, and belongs to the technical field of photoelectric sensor conversion efficiency calibration, the system comprises an electric wave reverberation chamber, a transmitting antenna, a reference antenna and a photoelectric field sensor system; the photoelectric field sensor system is composed of a photoelectric sensor, a laser source, a photoelectric detector, a polarization maintaining optical fiber and a single-mode optical fiber. In the calibration process, the reference antenna and the photoelectric sensor system are used as receiving ends respectively, receiving power at different stirring positions in the reverberation chamber is sampled to obtain two groups of transfer functions, and the total efficiency of the photoelectric sensor system is determined by comparing the two groups of transfer functions under the condition that the radiation efficiency of the reference antenna is known. The conversion efficiency of the photoelectric sensor system is obtained by combining the relation among the power, the field intensity and the total efficiency in the free space; and the effective averaging of the incident direction and the polarization state is realized by using the uniform and isotropic electromagnetic field characteristics of the reverberation chamber.

The application relates to the field of artificial intelligence and power equipment operation and maintenance data processing technology, and particularly discloses a power energy storage operation and maintenance health evaluation method based on deep learning, which comprises the following steps: a pre-processing unit first performs caliber unification according to a historical caliber change set and extracts single data in real time, eliminates equipment differences, makes the algorithm identify only one set of caliber, and reduces special cases and error sources; an edge gateway performs clock consistency test, aligns the time axis, and avoids wrong time misjudgment; if the test fails, the data is improved and data reliability is output, and the caliber change set is rewritten and corrected; then, the data and the reliability are sent to a model to perform health evaluation and physical consistency verification, and the correlation is improved to be explainable physical reliability; finally, the evaluation result is mapped to a standardized operation and maintenance disposal suggestion, and the operation and maintenance plan is synchronized to a platform; the standardized action and plan are synchronized to realize a closed loop of diagnosis, disposal and recovery, and the whole process is auditable and reproducible.

The invention relates to a hydrogen-electricity energy turbine hybrid propulsion engine system, which relates to the technical field of engines, and is characterized in that a motor driven by a hydrogen fuel cell and a hydrogen fuel gas turbine engine are connected in parallel by using a gear transmission unit, and the gear transmission unit has three switchable transmission modes; the propulsion structure establishes a transmission relationship with the motor and a power turbine of the hydrogen fuel gas turbine engine at the same time through a first transmission mode; the propelling structure only establishes a transmission relation with the motor through the second transmission mode; the propulsion structure establishes a transmission relationship only with the power turbine of the hydrogen fuel gas turbine engine through a third transmission mode. At least one of the hydrogen fuel gas turbine engine and the hydrogen fuel cell can be selectively used, so that the response speed of the engine system is increased, and infrared characteristics and noise of the engine system are reduced.

The invention discloses a concrete mixing quality prediction method based on mixing condition self-adaption and an AI model, and belongs to the technical field of concrete mixing quality prediction, and the method comprises the following steps: collecting batch working condition information related to a concrete mixing technology and a multi-source process signal in a mixing process, judging the working condition type of the current batch, and determining the working condition type of the current batch; determining feature normalization parameters of the current batch, performing feature extraction on the collected multi-source process signals according to process stages to obtain normalized feature expression, constructing the normalized feature expression and batch working condition information as feature input of the current batch, and inputting the feature input into an artificial intelligence model to obtain an artificial intelligence model; and obtaining a prediction result including the concrete quality index set. According to the method, the problems that the cross-working-condition generalization ability is insufficient, the dependence on calibration data is relatively strong, and the output stability and the engineering suitability need to be improved in an engineering scene with complex working condition fluctuation, process event disturbance, equipment difference and long-term drift are solved.

This invention provides a fault detection method and system based on low-rank decomposition and common trend decoupling. The method first acquires historical process data under normal operating conditions of an industrial process as training data. Then, it decomposes the training data into clean data representing the main structural information and sparse noise data representing noise interference using low-rank decomposition. Next, it constructs a common trend learning model based on the clean data, reconstructs the non-stationary common trend, and uses the residual between the clean data and the non-stationary common trend as a stationary feature. Decoupling is achieved by applying stationarity constraints. Further, a fault detection model is obtained through joint optimization. In the detection stage, the process data to be detected is input into the model, and monitoring statistics based on the reconstruction error of the stationary feature and the common trend are constructed respectively. The result of the control limit comparison determines whether a fault has occurred, thereby improving the robustness of fault detection in non-stationary industrial processes.

The utility model provides a kind of for special-shaped vacuum bag of special-shaped mould, the special-shaped mould includes the mould cavity of trapezoidal cross section;The special-shaped vacuum bag is set on the mould cavity and seals the mould cavity;The cross-sectional shape of the special-shaped vacuum bag includes the main body structure of convex letter shape, and the main body structure includes sequentially arranged from top to bottom center protruding portion, lateral recess portion and base portion.The special-shaped vacuum bag provided by the utility model can meet the requirement of air tightness and vacuum keeping when applied to special-shaped mould, ensure the forming quality of composite material, avoid bubble, void and other problems.

This invention relates to the field of cable defect identification technology, solving the technical problems of low accuracy in cable defect identification caused by gradient diffusion and feature degradation in existing deep networks, which lead to loss of edge details and difficulty in effectively extracting true weak thermal deformation features. Specifically, it relates to a method for identifying surface thermal deformation defects in extruded insulated low-voltage fire-resistant cables. The method acquires a visual image of the cable surface in the extrusion cooling state and a one-dimensional temporal physical signal, preprocesses the surface visual image to obtain a standard image, and uses a weighted gradient edge detection algorithm to construct templates in multiple directions to perform edge detection on the standard image. This invention reduces the probability of network parameter combinations getting trapped in local optima during model training, lowers false positive and false negative rates, improves the classification accuracy of cable surface thermal deformation defects, and meets the high-precision and high-robust defect detection requirements of extruded insulated low-voltage fire-resistant cables.

The invention discloses a method and a system for protecting portrait feature data of multilayer space mapping, which are used for scenes such as user portraits / article portraits and the like. According to the method, a three-layer mapping architecture of a scalar input layer (S)-a middle random layer (R)-a target output layer (C) is adopted, and multi-stage irreversible transformation processing is carried out on original scalar data. The method comprises the following steps: firstly, mapping a scalar into an intermediate vector by utilizing a one-way function of cryptography security and combining a secret key and an offset; secondly, distance features of the intermediate vector are extracted, direction features are discarded, and the distance features are converted into two-dimensional space coordinates through a deterministic mapping algorithm; according to the method, the irreversibility of reversely deducing the scalar quantity from the coordinates is ensured by erasing the angle information of the middle layer; the space isolation in different business scenes is realized by introducing the offset; according to the method, a deterministic rejection sampling algorithm based on HMAC is preferably adopted, the problem of center aggregation existing in a traditional mapping method is solved, uniform distribution of coordinates in a set area is achieved, sensitive feature value scalar quantities of all dimensions in portrait feature vectors are mapped into two-dimensional coordinate points, and the two-dimensional coordinate points are aggregated to form an object portrait feature matrix; the method is used for cross-system analysis, clustering, similar retrieval and visual display, and meanwhile plaintext feature value leakage can be avoided.

This invention relates to the application of AMPA receptor antagonists in the preparation of drugs for treating peritoneal metastases of gastric cancer, belonging to the field of biomedical technology. This invention discloses the application of GRIA2 as a drug target in the preparation of drugs for the prevention or treatment of peritoneal metastases of gastric cancer. It discloses the use of AMPA receptor antagonists NBQX and / or Selurampanel in the preparation of drugs for the prevention or treatment of peritoneal metastases of gastric cancer. NBQX and Selurampanel can significantly inhibit the migration, invasion, spheroidization ability, and adhesion to mesothelial cells of gastric cancer cells; in vivo experiments have confirmed that AMPA receptor antagonists can significantly reduce tumor burden and the number of metastatic nodules in a mouse model of peritoneal metastases of gastric cancer. Patient-derived organoid xenograft models further demonstrate that AMPA receptor antagonists have significant efficacy against peritoneal metastases of gastric cancer with high GRIA2 expression. This invention provides a new treatment strategy for peritoneal metastases of gastric cancer.

The present specification relates to a compound of Chemical Formula 1 and an organic light-emitting device comprising the same. The compound described in the specification can be used as a material for an organic layer of an organic light-emitting device. The compound according to at least one embodiment of the present specification can achieve an improvement in efficiency, a lower driving voltage, and / or an improvement in lifespan characteristics in an organic light-emitting device. In particular, the compound described in the specification can be used as a material for a light-emitting layer. In addition, compared with an existing organic light-emitting device, the organic light-emitting device has the effects of lower driving voltage, high efficiency and / or long service life.

The utility model discloses a booster-level flame-retardant low-intermodulation feeder line, and relates to the technical field of cables. The booster-level flame-retardant low-intermodulation feeder line comprises an inner conductor, an insulating layer, an outer conductor and a sheath which are coaxially arranged from inside to outside, the outer conductor is a spiral corrugated metal tube, the insulating layer is made of low-density polytetrafluoroethylene with a fibrous microporous structure, and the sheath is made of booster-level flame-retardant fluorinated ethylene propylene; the booster-level flame-retardant low-intermodulation feeder line provided by the utility model has the characteristics of high temperature resistance, high flame retardance, high shielding, low loss, low passive intermodulation and super-flexible bending, is suitable for booster-level cables or cable assemblies of a UL system, and is particularly suitable for interlayers of ceilings and high-flame-retardant safe application occasions of super-flexible wiring connection.

The present disclosure provides a fluorinated organic silicon modified waterborne polyurethane coating and a preparation method thereof, and belongs to the technical field of port power protection coating. The preparation method comprises the following steps: mixing diisocyanate, oligomeric diol and hydrophilic monomer, adding a catalyst after heating treatment, and stirring to obtain a polyurethane prepolymer; adding a neutralizing agent to the polyurethane prepolymer, stirring uniformly, and obtaining a neutralized prepolymer; adding deionized water to the neutralized prepolymer, shearing and dispersing, and standing to obtain a uniformly dispersed prepolymer; adding fluorinated organic silicon to the uniformly dispersed prepolymer, stirring and reacting, adding a chain extender, and stirring and reacting to obtain a fluorinated organic silicon modified waterborne polyurethane emulsion; coating the fluorinated organic silicon modified waterborne polyurethane emulsion into a mold, drying, and obtaining a fluorinated organic silicon modified waterborne polyurethane coating film, so as to endow the waterborne polyurethane coating with weather resistance and aging resistance in a high-humidity and high-temperature, high-ultraviolet environment of a coastal port.

The invention discloses a network intrusion detection feature selection method based on a multi-strategy improved neural network algorithm, and the method comprises the steps: firstly carrying out the cleaning, normalization and feature coding processing of original network flow data, constructing an initial feature space, and representing a feature selection scheme in a binary vector form; then Tent chaotic mapping is introduced to initialize a population, and the diversity of an initial solution is enhanced by using the ergodicity and uniformity of a chaotic sequence; in the feature optimization process, the Levy flight strategy is adopted to update the position of a leader, the global search ability is improved through a search mechanism combining long and short steps, local optimum is avoided, meanwhile, a follower update strategy based on the adaptive inertia weight is designed, global exploration and local development ability is dynamically balanced, and the algorithm convergence speed is increased; and further constructing a weighted fitness function taking a classification error rate and a feature subset scale as double targets, evaluating candidate feature subsets, and determining a global optimal feature subset through an elitist retention and iteration termination mechanism.

This invention belongs to the field of free-space laser communication technology, specifically disclosing a second-level anti-disturbance link establishment method and ground simulation system for inter-satellite laser communication. The method includes: using an inverse dynamics feedforward algorithm to compensate in real-time for the inertial coupling torque of the coarse tracking mechanism caused by satellite platform disturbances, achieving inertial spatial stability of the line of sight; employing a spectrum grading and saturation unloading strategy to further suppress line-of-sight jitter caused by satellite platform disturbances through the high-bandwidth operation of the fine tracking mechanism, and unloading low-frequency components to the coarse tracking mechanism to prevent saturation of the fine tracking mechanism; and initiating a high-frequency probabilistic spiral scan at the initial stage of link establishment to achieve rapid target acquisition. This invention solves the problem of lacking efficient anti-disturbance strategies and optimization strategies for second-level rapid response during mode switching from large-scale "blind search" to high-precision "lock-in" in inter-satellite laser link establishment.

The invention provides a composition, a kit and a method for detecting cell species and interspecific contamination. Specifically, primers and Taqman probes are designed for people, mice and rats respectively, and cell species and interspecies contamination are detected in a single reaction system through multiple real-time fluorescent quantitative PCR. The method is high in specificity and sensitivity, strong in specificity and short in consumed time.

The application provides an operation circuit and a microprocessor architecture, which are applied to the technical field of computers, and the circuit comprises a gating circuit and a multiplier. The operation circuit is responsive to a switching control signal. In a first mode, the operation circuit controls the gating circuit to send a real operand pair (a, b) into the multiplier for operation. Alternatively, in a second mode, the operation circuit is responsive to the switching control signal to send a non-zero mask data pair into the multiplier for operation. In this way, the multiplier has data input and performs operation in any clock cycle, that is, the multiplier is continuously in a running state. The power consumption generated by the running of the multiplier is maintained stable, and differential power consumption characteristics are not generated. The side channel attack of an attacker to obtain sensitive information by analyzing the power consumption characteristics is prevented, the security of the password operation is improved, and security vulnerabilities are eliminated.

This invention discloses a multi-model fusion method and system for permanent magnet motor fault diagnosis, belonging to the field of permanent magnet motor fault diagnosis technology. It aims to improve the flexibility and adaptability of models in multi-task learning and hierarchical classification by introducing multiple Softmax-layer CNNs and fusing them with multiple models, while effectively solving the problems of weak noise resistance and insufficient generalization ability of single models. Specifically, the technical solution of this invention first collects the vibration and current signals of the faulty permanent magnet motor, then uses Markov transfer fields to adaptively enhance the time-domain signals into images. Using a multi-Softmax-layer CNN and XGboost as base learners and an IWOA-SVM model as the meta-learner, a high-precision fault diagnosis result is finally obtained. This invention combines the advantages of multiple Softmax layers and multi-model fusion, effectively solving the multi-task fault diagnosis problem under complex working conditions, and improving the accuracy, reliability, and adaptability of fault diagnosis.

The invention discloses a phasor synthesis type flexible interconnection transformer topological structure and a control method. The phasor synthesis type flexible interconnection transformer topological structure comprises a first parallel branch and a second parallel branch which are connected into an alternating current input end in parallel. Each branch is composed of two rotating phase-shifting transformer units, rotor windings of the two rotating phase-shifting transformer units are connected in parallel, and stator windings of the two rotating phase-shifting transformer units are connected in series to form an output port. The internal rotors of the two output ports are respectively driven by independent servo motors to rotate synchronously, and the control module generates corresponding rotating speed instructions according to power control targets of the ports. The structure is based on the phasor synthesis principle, and continuous and accurate adjustment of voltage and power flow is achieved through mechanical rotation conversion. The defects that traditional power electronic equipment is high in harmonic pollution and a mechanical transformer is slow in response are effectively overcome, and the controllability and economical efficiency of a system are remarkably improved while the advantages of high reliability and low loss of a full-electromagnetic structure are kept; and an efficient and compatible flexible interconnection solution is provided for a complex power distribution network containing high-proportion renewable energy sources.

The invention belongs to the technical field of molecular diagnosis / detection, and particularly discloses a detection method of an aminoglycoside drug resistance gene armA. Comprising the following steps: extracting genome DNA from a sample to be detected; the method comprises the following steps: in the presence of Bst DNA polymerase, an MCDA primer and a reaction buffer solution, carrying out MCDA isothermal amplification on DNA to be detected to obtain an amplification product containing a PAM site; the obtained amplification product is mixed with Cas12a protein, gRNA molecules, a double-labeled DNA probe and a reaction buffer solution, and CRISPR / Cas12a reaction is carried out in the same reaction system; and detecting a fluorescence signal of the reaction system in the step 3, and judging that the armA gene exists in the sample when an effective fluorescence signal is generated. The invention provides a method (MCDA-CRISPR / Cas12a) capable of rapidly, sensitively and specifically detecting armA by combining an MCDA isothermal amplification technology and a CRISPR / Cas12a fluorescence detection technology.

This invention discloses a dual-polarization angle-selective surface structure for reducing radar sidelobes, comprising an upper structure and a lower structure, with an air gap between them. Both the upper and lower structures include a plurality of frequency-selective surface units arranged in an array. Each frequency-selective surface unit includes a dielectric substrate. A first metal resonant unit structure is disposed on the upper surface of the dielectric substrate, and a second metal resonant unit structure is disposed on the lower surface of the dielectric substrate. The first and second metal resonant unit structures are connected through metal vias on the dielectric substrate. This structure can solve the problem that radar sidelobe blanking technology cannot simultaneously reduce radar sidelobes and avoid introducing redundant additional interference sources.

The invention belongs to a personalized intelligent medicine feeding regulation and control system based on machine learning, and the system comprises the steps: collecting a skin multi-source bioelectricity signal, carrying out the case matching and feature extraction through employing an ant colony-mantis shrimp visual fusion algorithm, and generating a bioelectricity feature data set; the optimization direction of the bioelectricity characteristic data set is shared through decentralized lightweight federal learning, and a medicine feeding equipment interaction model with communication and energy supply capabilities is constructed; establishing a minimum block chain network, and recording equipment operation behaviors to generate a physical hash value; decrypting the encrypted data set and extracting key features, constructing a database according to user health information, and establishing a correlation model of medication parameters and health states; and predicting an optimal medication scheme of the user through the LSTM neural network, converting the scheme into a control instruction, and issuing the control instruction to the medicine feeding equipment interaction model to form a personalized medicine feeding regulation and control closed loop. According to the invention, multiple advantages are formed in physiological data medication regulation individuation and data safety protection.

The application discloses a sodium dihydrogen phosphate reinforced high-temperature supermolecular fracturing fluid and a preparation method thereof, and belongs to the technical field of fracturing fluids for oil and gas fields. The fracturing fluid takes a supermolecular associated thickening agent mother liquor as a thickening main body, and second-stage sodium dihydrogen phosphate, a gel breaking regulator, a complex stabilizer, nano silicon dioxide, a gel breaker and water are added; the thickening agent is prepared from acrylamide, 2-acrylamido-2-methylpropane sulfonic acid, octadecyl methacrylate, acryloyl morpholine, hydroxypropyl-beta-cyclodextrin, sodium dodecyl sulfate, AEO-7, ammonium persulfate, first-stage sodium dihydrogen phosphate and water. The application introduces sodium dihydrogen phosphate in stages and constructs a supermolecular association network, improves the high-temperature shear stability of the fracturing fluid, reduces gel breaking residues, and is suitable for fracturing reconstruction of deep and ultra-deep reservoirs.

This application provides a method, system, device, and storage medium for identifying abnormal states of traditional Chinese medicine (TCM), relating to the field of anomaly identification. The method includes: acquiring product information of the TCM to be detected, and determining data acquisition standards for the TCM based on the product information and preset detection rules; inputting image data into a preset image anomaly detection model based on the data acquisition standards to obtain preliminary image anomaly regions; fusing visual feature data and spectral feature data to obtain anomaly feature vectors; inputting the anomaly feature vectors into a preset anomaly state discrimination model to output anomaly state data, and determining the anomaly state based on the anomaly state data and a preset anomaly state judgment threshold to obtain the anomaly state detection result of the TCM to be detected. This method enables timely and accurate identification of abnormal states in TCM.

This invention provides a cooling rate control method based on thermal stress prediction during the sintering and cooling stage of grinding wheels. It generates standardized multimodal data through spatiotemporal registration and noise filtering, identifies key nodes in thermal stress evolution using finite element simulation, constructs a thermal stress response fingerprint spectrum specific to the material structure, and encodes it as a structured feature vector. This enables real-time characterization of the stress state without relying on meshes or material parameter libraries. A historical knowledge base is used to quickly map the feature vector to stress trends, enabling prediction and dynamic calibration of abnormal disturbances. Closed-loop error compensation generates high-confidence stress trend fragment flow, driving low-latency automatic adjustment of the cooling system. This invention improves the accuracy of thermal stress prediction and the efficiency of cooling control response, ensuring the performance and safety of the grinding wheel.