40 results about "Simple sample" patented technology

The invention discloses a fluorescent probe for detecting spermine and spermidine in a water sample and urine as well as a preparation method and application of the fluorescent probe, and relates to the technical field of supramolecular fluorescent probes, and the fluorescent probe is formed by reacting 14-membered cucurbit urils (tQ [14]) and a compound (E)-1-(4-boron benzyl)-4-(4-hydroxyl styryl) quinoline-1-ammonium bromide according to a molar ratio of 1: 1. The novel fluorescent probe provided by the invention can specifically detect spermine and spermidine in a water sample and urine, and the preparation method of the fluorescent probe is simple and easy to implement; the method for detecting spermine and spermidine in a water sample and urine by using the fluorescent probe has the advantages of high sensitivity, high selectivity, low cost, simple sample treatment, convenient operation, rapid determination and the like.

The application discloses a seminal plasma metabolic biomarker combination for male infertility spermatogenesis obstacle typing diagnosis, and relates to the field of biological science and technology.The application records the metabolic fingerprints of 475 seminal plasma samples by means of nanoparticle-enhanced laser desorption / ionization mass spectrometry, and constructs a seminal plasma metabolic fingerprint database related to spermatogenesis obstacle (SPGF). Then, through machine learning and statistical analysis, six metabolic biomarkers for SPGF typing diagnosis are screened and identified in the seminal plasma, and the six metabolic biomarkers can realize accurate typing of SPGF (AUC=0.814). The application has the advantages of simple sample pretreatment, fast analysis speed and low cost by means of recording the metabolic fingerprints of seminal plasma by means of nanoparticle-enhanced laser desorption / ionization mass spectrometry. Through the synthesized iron oxide nanoparticles as a matrix, small molecule metabolites in the seminal plasma can be efficiently captured and ionized, and complex pretreatment steps such as deproteinization and chromatographic separation are not needed, so that super-fast analysis, super-high sensitivity, extremely low sample consumption and excellent repeatability detection are realized.

The invention discloses a method for detecting sialic acid in a dairy product. The method comprises the following steps: pretreating raw materials: weighing a uniformly mixed liquid dairy product, adding water for diluting, adding a precipitant consisting of potassium ferrocyanide and zinc acetate, uniformly shaking, adding water for fixing the volume, uniformly shaking, and filtering with a filter membrane of 0.5 mu m or less to obtain a filtrate; the concentration of the liquid dairy product is 1-10g / 100mL, the concentration of the potassium ferrocyanide is 0.1-0.26 g / 100mL, and the concentration of the zinc acetate is 0.2-0.55 g / 100mL after the water is added to a constant volume; a chromatographic column is a hydrogen type cation chromatographic column, a 0.005-0.01 mol / L sulfuric acid aqueous solution is used as an eluent for isocratic elution, and after elution, a PAD detector is adopted for detection under the detection wavelength of 210-220 nm. The method is suitable for detecting sialic acid in milk, and has the advantages of simple sample treatment, low treatment cost and accurate detection.

The application discloses a quality control method of Shen Shuai Kang capsules, and applies isoxanthohumol to the quality control of the Shen Shuai Kang capsules, carries out thin-layer identification and fingerprint spectrum detection on the Shen Shuai Kang capsules, and determines the content of isoxanthohumol in the Shen Shuai Kang capsules, so that the quality control standard of the Shen Shuai Kang capsules is improved. The method can not only more accurately, comprehensively and conveniently evaluate the quality of the Shen Shuai Kang capsules, but also indirectly guarantee the effectiveness and safety of the product. The quality control method has the advantages of simple sample preparation method, reliable and stable detection result and the like.

The present application provides a science and education resource relationship prediction method and system based on large model zero sample prediction, comprising: obtaining unlabeled science and education resource data, dividing the unlabeled science and education resource data into simple samples and difficult samples based on the zero sample prediction and self-check of the large model, and constituting a few sample training set; labeling the few sample training set by artificial means; selecting a corresponding large model instruction fine-tuning algorithm according to different application scenarios, training the model on the labeled few sample training set, and obtaining a trained model; loading the trained model to perform automatic relationship extraction and complete relationship prediction. By obtaining unlabeled science and education resource data and using the zero sample prediction capability of the large model to preliminarily process these data, the dependence on a large amount of labeled data in the traditional method is avoided, so that effective relationship prediction can be performed in the science and education field where labeled data is scarce.

The invention discloses a model training method and device, and relates to the technical field of machine learning, and the method comprises the steps: carrying out the training of a target model through a training set in a data set, and obtaining an evaluation index of each sample; dividing each to-be-divided sample into a simple sample or a difficult sample according to the evaluation index; and discarding the simple samples, and returning to execute the next round of training based on the adjusted data set until the preset round of training is completed. According to the method, all key operations are completed in the training process, so that the sample division does not need to start any calculation process independent of model training, and independent links such as preprocessing, post-processing or special index calculation are avoided; and the scale of a data set of subsequent training is dynamically reduced by discarding the sample subsets which are divided into simple samples after each training round, so that the model training pays more attention to difficult samples, the data volume which needs to be processed in subsequent iteration of the model is directly reduced, and the calculation resources are saved and the training efficiency is improved.

The invention provides an SERS (Surface Enhanced Raman Scattering) substrate and a detection method for the blood concentration of epirubicin, and belongs to the technical field of detection.The SERS substrate is prepared by the following steps: firstly, respectively preparing 40nm and 20nm gold nanoparticles through a sodium citrate reduction method, and constructing the SERS substrate on a silicon dioxide sheet through ABA sandwich dispensing; an epirubicin blood sample is pretreated through an acetonitrile precipitation method and then dropwise added to the substrate to be dried away from light, and the 1236 cm <-1 > characteristic peak intensity is detected through a 785 nm Raman spectrum. By establishing a quantitative curve of epirubicin concentration and peak intensity in serum, rapid quantitative detection of clinical samples is realized. The method is simple in sample treatment, high in detection sensitivity, good in repeatability, consistent with LC-MS / MS results, and suitable for bedside treatment drug monitoring.

The invention belongs to the technical field of in-vitro nucleic acid detection, and discloses a pyrosequencing primer and a kit for detecting CYP2D6-5 genetic typing. And the base sequence of the pyrosequencing primer is as shown in SEQ NO.3. The kit comprises dNTPs, a buffer solution, a recombinant DNA polymerase (rTaq), double distilled water, a specific amplification primer 1, a specific amplification primer 2, a sequencing primer, a blank reference substance and a positive reference substance, the base sequence of the specific amplification primer 1 is as shown in SEQ NO.1; the base sequence of the specific amplification primer 2 is as shown in SEQ NO. 2; the base sequence of the sequencing primer is as shown in SEQ NO.3. When the kit is used for detecting the CYP2D6 * 5 genotyping, the kit has the advantages of accurate qualitative detection, high sensitivity and strong specificity; in addition, the method has the advantages that sample treatment is simple, sequencing steps are simple, the sequencing speed is high, one on-machine reaction is completed in half an hour, detection site frequency analysis is directly given, and the result is visual.

The utility model relates to the technical field of sample preparation equipment, in particular to a simple sample preparation machine for a thick-wall PVC pipe sample. Comprising an upper die device and a lower positioning device, the lower positioning device comprises a bottom plate, a movable plate, a base plate and a jack, and the upper die device comprises an upper fixing plate, a connecting plate and a material pressing die; the jack is fixedly connected with the bottom plate, the movable plate can move up and down under the action of the jack, the upper fixed plate is elastically connected with the movable plate, and the base plate is fixedly connected with the movable plate; the pressing die for sample preparation is arranged below the connecting plate; the upper mold device is further provided with a sample plate demolding pin and a reset spring which are used for enabling the upper mold device to reset automatically after sample preparation. According to the thick-wall PVC pipe tensile sample manufacturing machine tool, large-diameter thick-wall pipe samples can be manufactured according to the shapes of the test pieces, the sizes of the manufactured samples meet the standard requirement and have accuracy, the sample manufacturing machine is simple in structure, small in size and convenient to operate, and the tensile detection requirement of the samples is met.

The invention discloses a kit and a method for rapidly detecting and diagnosing feline calicivirus. The kit comprises composite magnetic beads, a reverse transcription system and a fluorescent quantitative PCR (Polymerase Chain Reaction) detection system, the composite magnetic bead is Fe3O4-coated UiO-66, the reverse transcription system comprises a reverse transcription primer and a reverse transcription reagent, and the fluorescent quantitative PCR detection system comprises an amplification primer and a reaction reagent. Composite magnetic beads are synthesized from an MOFs material and magnetic nanoparticles through a solvothermal method, then reverse transcription primers are pre-anchored on the surfaces of the composite magnetic beads, and a one-step mRNA detection platform is derived; after the feline calicivirus is hybridized with the surface anchor primer, reverse transcription is immediately carried out, the generated cDNA can be directly amplified through quantitative polymerase chain reaction, and quantitative analysis can be realized without pre-RNA extraction. Compared with the traditional kit, the method has the advantages of simple sample pretreatment, less time consumption, low cost, no need of using toxic organic reagents such as chloroform in the traditional nucleic acid extraction method, no need of tedious steps such as centrifugation and elution, safety and environmental protection.

The application relates to an industrial equipment residual life prediction method and medium based on classification and error correction knowledge distillation, and relates to the technical field of industrial equipment life prediction. The similarity between the teacher output prediction value and the corresponding true value is used to divide the difficult samples and simple samples of each time sequence, so that the distillation loss function can be constructed based on the obtained difficult samples and simple samples, thereby obtaining the total loss function, so that the student network model can perform knowledge self-reflection in the training process combined with the loss function, instead of directly replacing the true value, emphasizing self-correction, and the adaptability of the student model to the real data is enhanced, so that the prediction accuracy can be improved while reducing the reasoning memory, thereby achieving a better balance between accuracy and reasoning memory, so as to facilitate easier deployment on mobile devices.

The invention discloses a method for detecting folic acid and derivatives thereof. The method comprises the following steps: 1) detecting by adopting fluorescent liquid chromatography to obtain peak appearance time and a peak area-folic acid concentration linear equation corresponding to each folic acid monomer and folic acid precursor; and 2) preparing a to-be-detected sample into a folic acid extracting solution, performing fluorescent liquid chromatography detection, and obtaining the types and contents of folic acid monomers and folic acid precursors in the to-be-detected sample according to the appearance time and the peak area-folic acid concentration linear equation in the step 1). The method for detecting folic acid and derivatives thereof has the advantages of high degree of separation and sensitivity, good repeatability and stability, simple sample pretreatment, low detection cost, high detection efficiency and the like, and is suitable for accurate quantitative analysis of folic acid content in food such as mushrooms, corn, soybeans, rice, potatoes, spinach and the like.

This application discloses a microfluidic chip, a nucleic acid detection cartridge, and a nucleic acid extraction cartridge, comprising a chip body and thin films respectively disposed on two surfaces of the chip body. The chip body is provided with n first microfluidic channels and m second microfluidic channels. The first microfluidic channels include m sequentially connected first reaction chambers, and the second microfluidic channels include n sequentially connected second reaction chambers. The first and second reaction chambers are connected in a one-to-one correspondence, and a phase change material isolator is disposed at the connection point. m and n are both natural numbers not less than 2. One of the first and second microfluidic channels is used to inject a template-free nucleic acid amplification system, and the other is used to inject the sample to be tested. This application can achieve simultaneous detection of multiple nucleic acid targets in multiple samples to be tested through simple sample addition, greatly simplifying the operation process for technicians and reducing consumable costs.

The invention belongs to the technical field of atomic absorption spectrometry analysis, and particularly relates to a method for rapidly determining trace Pb in a complex sample under high acidity based on atomic absorption spectrometry, in particular to an analysis method for accurately and rapidly determining trace Pb in cosmetics. According to the method for rapidly determining the trace Pb in the complex sample under the high acidity based on the atomic absorption spectrometry, the HCl concentration of a test system can be greatly improved to 10% by adopting a designed chemical atomizer and optimizing the conditions of reduction reaction; the method breaks through the bottleneck that trace Pb in a few simple samples can only be measured under low acidity by a hydride generation method, so that the chemical reaction of Pb and NaBH4 can be stably carried out under high acidity, the problem of testing trace Pb in complex samples is solved, and the method is high in sensitivity, simple and rapid to operate and reliable in test result.

The invention discloses a fluorescent probe, preparation thereof and application of the fluorescent probe in detection of glyphosate in water. The fluorescent probe provided by the invention is constructed by self-assembling symmetric tetramethyl cucurbit [6] uril and guest molecules E as raw materials. The fluorescent probe is a novel self-assembly body, is simple in preparation method, can be used for qualitatively or quantitatively detecting glyphosate in water, and has the characteristics of high sensitivity, high selectivity, low cost, simplicity in sample treatment, convenience in operation and rapidness in determination.

The application discloses a method for identifying the authenticity or year of brewed wine based on dynamic light scattering, comprising the following steps: (1) taking a brewed wine standard product, adding different exogenous substances into the brewed wine standard product respectively, dissolving and mixing uniformly, and then performing membrane treatment to obtain a standard product sample; (2) determining the hydrodynamic average radius of each exogenous substance in the standard product sample by dynamic light scattering, and obtaining the hydrodynamic average radius basic data of the corresponding exogenous substance in the standard product; (3) taking a to-be-detected brewed wine, and performing the same treatment as steps (1) and (2) on the to-be-detected brewed wine to obtain the hydrodynamic average radius data of the corresponding exogenous substance in the to-be-detected brewed wine; and (4) comparing the hydrodynamic average radius data of the corresponding exogenous substance in the to-be-detected brewed wine with the hydrodynamic average radius basic data of the corresponding exogenous substance in the standard product, and identifying the authenticity or year of the to-be-detected brewed wine. The application has the advantages of small sampling amount, simple sample pretreatment and analysis process, and less interference in the detection process.

The invention belongs to the technical field of pharmaceutical analysis, and relates to a method for detecting the residual quantity of glacial acetic acid in a crisaborole raw material by high performance liquid chromatography, which adopts a chromatographic column with high-purity silica gel particle chemically bonded octadecylsilane as a filler, a 0.1% phosphoric acid aqueous solution as a mobile phase A and acetonitrile as a mobile phase B for gradient elution. The detection method is simple in sample treatment, strong in specificity, relatively good in durability and high in detection sensitivity.

The invention discloses a zebra fish-based spatial resolution lipidomics method, and belongs to the technical field of biological medicines. The zebrafish-based spatial resolution lipidomics method provided by the invention comprises the following steps: separating a zebrafish trunk from yolk; sample pretreatment and chromatography-mass spectrometry detection. According to the method, yolk is sucked by a pipettor, trunks are cleaned in batches by a vortex instrument, after the trunks and the yolk are pretreated, a sample is subjected to lipidomics detection by an ultra-high performance liquid chromatography tandem quadrupole electrostatic field orbitrap high-resolution mass spectrometer, and further data analysis is performed. According to the method, the trunk and yolk samples which can be used for high-quality and high-coverage lipidomics analysis are obtained while high sample flux is maintained for the first time, and regionalized analysis of zebrafish embryo / larva lipid metabolism is realized. The method has the advantages of simple sample mechanism, low lipid signal interference, high sensitivity and the like, and can detect inrecognizable subtle changes and key lipid molecules in the whole sample.

The invention discloses a kit for chemiluminiscence detection of interleukin 13 and application of the kit. The kit comprises a reagent R1, a reagent R2 and a reagent R3, the reagent R1 comprises a magnetic bead, and the surface of the magnetic bead is coated with streptavidin; the reagent R2 and the reagent R3 respectively comprise an interleukin 13 monoclonal antibody 1 and an interleukin 13 monoclonal antibody 2, and the interleukin 13 monoclonal antibody 1 and the interleukin 13 monoclonal antibody 2 are respectively coupled with biotin and alkaline phosphatase. The kit has the advantages of being high in sensitivity, wide in linear range, simple in sample treatment and high in automation degree when being applied to detection of the interleukin 13, and has the application prospect of being applied to clinical determination of the concentration of the interleukin 13 in a serum or plasma sample.

The invention relates to the technical field of biological detection and medical diagnosis, and discloses a method capable of rapidly detecting mycobacterium tuberculosis, which comprises the following steps: S1, providing a detection test strip which sequentially comprises a sample pad, a combination pad, a chromatography membrane and a water absorption pad; wherein the combination pad is coated with a labeled mycobacterium tuberculosis specific nucleic acid probe, the chromatography membrane is provided with a detection line and a quality control line, capture molecules are fixed on the detection line and the quality control line, the method integrates a nucleic acid probe chromatography technology and simple sample treatment, expensive instruments are not needed, and a result can be obtained by visual inspection within 15 minutes. The kit remarkably reduces the detection threshold and cost, is simple and convenient to operate and high in sensitivity, can synchronously identify drug resistance, perfectly meets the urgent demand of primary medical scenes on rapid and accurate diagnosis of tuberculosis, and assists in early intervention and epidemic prevention and control.

The application discloses a method for quantitatively analyzing 21 arginine synthesis metabolites in mouse liver, kidney and serum based on liquid chromatography-mass spectrometry technology and application of the method in cantharidin hepatorenal toxicity research. The method comprises the following steps: preparing 21 metabolites and two internal standard solutions; optimizing mass spectrometry multiple reaction monitoring parameters; performing sample analysis after protein precipitation by acetonitrile, nitrogen blowing dry and 50% methanol water reconstitution; and establishing eight concentration level standard curves and adopting a weighted least square method to fit a linear equation. Methodology verification shows that the linear relationship of each metabolite is good (r 2 >0.99), and the precision, accuracy, matrix effect, recovery rate and stability all meet the requirements of biological sample analysis. The method has simple sample pretreatment, high sensitivity and strong specificity, is successfully applied to determination of the content of 21 arginine metabolites in a mouse hepatorenal toxicity model caused by cantharidin, and provides a reliable analysis tool for arginine metabolism research and cantharidin toxicity mechanism exploration.

The application belongs to the technical field of pharmaceutical analysis, and relates to a method for detecting residual acetic acid in a clilborol raw material by high performance liquid chromatography. An octadecylsilane bonded silica gel is used as a filling agent of a chromatographic column, 0.1% phosphoric acid aqueous solution is used as a mobile phase A, and acetonitrile is used as a mobile phase B to perform gradient elution. The detection method has the advantages of simple sample processing, strong specificity, good durability, and high detection sensitivity.

The present application relates to the technical field of food detection, and more particularly to a method for simultaneously detecting multiple antibiotic residues in prepared dishes. The method comprises the following steps: (1) mixing prepared dishes, a mixed internal standard solution and an extraction solution, ultrasonicating, centrifuging, taking supernatant, and obtaining a mixed solution; (2) purifying the mixed solution, drying, and obtaining a dried product; (3) mixing the dried product with a redissolution solution, centrifuging, taking supernatant, and obtaining a to-be-tested sample after membrane filtration; and (4) detecting the content of veterinary drug residues in the to-be-tested sample by high performance liquid chromatography-mass spectrometry. The method has simple sample processing steps, is convenient to operate, does not require complex derivatization treatment, has low detection cost, can realize rapid detection of batch samples, is suitable for both routine detection in laboratories and rapid screening on site by food supervision departments, and has important practical value for guaranteeing the safety of prepared dishes and standardizing the development of the industry.

The utility model relates to the technical field of liquid nitrogen storage, in particular to an automatic liquid nitrogen storage tank. Comprising a tank body, an annular cryopreservation frame, a driving part and an in-tank automation mechanism, the opening of the tank body is in butt joint with an external sample conveying mechanism; the annular cryopreservation frame is mounted in the tank body and is used for storing samples; the tank body is provided with an operation space for forming avoidance with the annular cryopreservation frame; the driving part is used for driving the annular cryopreservation frame to rotate in the tank body and is in butt joint with the operation space in sequence, and the in-tank automatic mechanism is installed in the tank body and is used for storing and taking samples on the annular cryopreservation frame in the operation space according to a sample storing and taking instruction. The liquid nitrogen storage tank is simple in structure, large in sample storage amount, high in automation degree and stable in operation, and the sample storage and taking efficiency is improved.

The application discloses a polystyrene micro-nano plastic immunochromatography time-resolved fluorescence kit and application thereof. The kit comprises a fluorescence test strip and a sample reaction bottle containing quantum dot fluorescence microsphere labeled anti-polystyrene monoclonal antibody freeze-dried product. The fluorescence test strip comprises a bottom plate, the adhesive surface of the bottom plate is sequentially pasted with a water absorption pad, a detection pad and a sample pad from top to bottom, adjacent pads are connected through overlapping connection at a connecting position, the detection pad takes a nitrocellulose membrane as a base pad, a transverse quality control line and a detection line are arranged on the nitrocellulose membrane from top to bottom, the quality control line is coated with anti-mouse polyclonal antibody, and a polystyrene-hemocyanin conjugate is coated on the detection line. The immunochromatography time-resolved fluorescence kit provided by the application can realize quantitative detection of polystyrene, has the advantages of high specificity, high sensitivity, simple sample pretreatment, small environmental pollution hazard and suitability for on-site batch detection.

The embodiment of the invention relates to the technical field of target tracking, in particular to a multi-target association method and system, computer equipment and a storage medium. The method mainly comprises the steps of obtaining a plurality of candidate detection frames corresponding to a plurality of to-be-associated targets in current frame data, taking each to-be-associated target and the plurality of candidate detection frames corresponding to the to-be-associated targets as a sample, determining the sample type of each sample, extracting appearance features of the candidate detection frames in difficult samples and the to-be-associated targets, and determining the target type of the to-be-associated targets according to the appearance features of the candidate detection frames in the difficult samples and the appearance features of the to-be-associated targets. And based on a preset classification model, obtaining association probability data according to the basic detection frame characteristics of the simple samples, the basic detection frame characteristics of the difficult samples and the appearance characteristics, and determining a target detection frame having an association relationship with each to-be-associated target from the plurality of candidate detection frames according to the association probability data and a preset matching algorithm. By adopting the method, the accuracy of target object recognition can be improved.

The utility model discloses a kind of pollutant rapid detection test cards, including long strip shell, detachably equipped with inner shell in shell body, quick release structure is equipped between inner shell and shell body;Back surface is equipped with friction groove in shell body;Liquid preparation groove is equipped in friction groove;Inner shell front and close to lower end are equipped with containing groove, and several different test paper strips are equipped in containing groove;The front of inner shell is movably equipped with the stop lever that can limit test paper strip.The utility model is equipped with two parts of inner shell and shell body, and inner shell is used to store different test paper strips, and different pollutant detection is coped with;Shell body can carry out quick rubbing breakage to food sample, and simple sample can be quickly prepared after adding reagent, and detection is carried out directly through test paper strip, improve detection efficiency.

The invention discloses a quantitative detection model for multi-element heavy metals in a paper packaging material and application of the quantitative detection model. According to the model, to-be-detected heavy metal elements in the paper food packaging material are calibrated by making a standard sample to obtain a standard curve, and through multiple times of test fitting, the influence of background noise on result accuracy is eliminated, and the detection limit is effectively reduced, so that the test sensitivity is improved; the sample is subjected to simple tabletting treatment, so that the X-ray excitation intensity can be enhanced, and the detection limit is effectively reduced; on the basis of a common energy spectrum X-ray fluorescence method, a bent crystal monochromatic X-ray fluorescence method is introduced, so that the detection limit of trace heavy metals can be effectively reduced. The method has the advantages of being simple in sample pretreatment, high in detection speed, high in accuracy and the like, the heavy metal detection capacity and efficiency in the paper packaging material can be remarkably improved, and the quality and safety of the paper packaging material are effectively guaranteed.

The invention discloses a construction method and application of an HPLC (High Performance Liquid Chromatography) specific chromatogram of nephritis-relieving granules, and belongs to the technical field of quality control and evaluation of Chinese patent medicines. 16 components such as chlorogenic acid are selected as reference substances, a variety of bioactive components are covered, a specific chromatogram is constructed through high performance liquid chromatography and high resolution mass spectrometry, 35 common peaks are contained, the chemical components and relative contents of different batches of particles are visually displayed, the component difference is accurately presented, and the defects of a traditional method are overcome. 16 characteristic peaks are identified and cover 6 medicinal materials, and an efficient method is provided for quality control. Meanwhile, one or more of the 16 components are used as quality markers, the particle quality is comprehensively evaluated, the limitation of a traditional method is avoided, the 16 components can be detected according to clear requirements, and the result integrity is guaranteed. According to the method, sample pretreatment is simple and convenient, methodology investigation verifies that reproducibility and stability are good, quality standard formulation and industrialization requirements are met, and support is provided for quality control of the nephritis-relieving granules.

This invention discloses a training method and identification method for a power system instability state identification model, belonging to the field of power system stability analysis technology. The training method includes noise identification of the original sample set D0 carrying instability state labels before training the instability state identification model. Noise identification includes: training model M0 with sample set D0, and dividing sample set D0 into simple sample sets D based on the magnitude of the loss. s and complex sample set D H / N , sample set D s The unstable state labels of some samples in the dataset are changed to form an artificially labeled noisy sample set D. a ; with sample set D a Train model M1, and adjust the sample set D according to the loss value. a Divided into simple sample sets D a，s and complex sample set D a，H / N According to sample set D a，H / N Training model M a Through model M a Construct a mapping relationship between the loss value sequence and the label type; then based on model M a Identify sample set D H / N The above process allows for the identification of noise in the original sample set, enabling more effective training of the power system instability identification model.