62 results about "Domain testing" patented technology

The invention relates to the technical field of artificial intelligence and databases, in particular to an NL2SQL method and system based on reinforcement learning, and the method comprises the following steps: natural language input preprocessing, semantic analysis and abstract representation construction, mode linking and candidate set determination, SQL generation, query execution and feedback optimization, and multi-round interaction processing. The method has the beneficial effects that complex scenes such as nested sub-query, multi-table JOIN and aggregation function combination are supported, and the SQL generation accuracy is improved by more than 30% compared with that of a traditional method (test data is from a spider data set); a mode link mechanism based on reinforcement learning can quickly adapt to a new database mode, and in a cross-domain test (such as switching from an e-commerce database to a medical database), the accuracy rate decrease amplitude is less than 15%.

The invention discloses a cross-working-condition bearing fault diagnosis method based on a multi-module combination model, and the method comprises the following steps: 1, collecting bearing vibration time domain signals when different bearing faults are generated under different working conditions, dividing a source domain training set with fault type labels, and obtaining a source domain training set; the target domain training set and the target domain test set are not provided with fault type labels; step 2, constructing a diagnosis model, wherein the diagnosis model comprises a feature extraction module, a fault classifier module, a global domain confrontation module, a condition domain confrontation module and a feature weighted alignment module; 3, inputting the source domain training set, the target domain training set and the target domain test set into a diagnosis model for model training; and 4, carrying out bearing fault identification by adopting the trained diagnosis model. The generalization ability of the bearing fault diagnosis model is improved.

The invention relates to the technical field of electronic testing, and discloses a USB module multi-station synchronous testing and data association method and system, and the method comprises the steps: constructing a global schedulable resource map of a testing station, so as to create a resource load balancing table and a global testing database of the testing station; configuring a synchronous test controller of the test station by using the test task flow generated by the test station and the global schedulable resource atlas; reading the unique identity label of the USB module so as to establish an association mapping network of the USB module in the global test database; and executing multi-station synchronous testing and data association processing of the USB module based on the global schedulable resource atlas, the resource load balancing table, the synchronous testing controller, the collaborative optimization strategy set and the association mapping network. According to the invention, accurate synchronous control of multi-station test and automatic correlation analysis of test data can be realized.

The invention relates to the technical field of industrial fault diagnosis, in particular to a multi-sensor and cross-working-condition industrial fault diagnosis method. The method comprises the following steps: S1, acquiring equipment operation data of a plurality of sensors under different working conditions through a data acquisition system, and dividing the equipment operation data into a training set, a verification set and a test set; s2, constructing the original data into a 3D space-time collaborative tensor as model input; s3, CBT, LMSCB, ESRM and KAN-TD are embedded into a network architecture, LightM-ConvKNet intelligent fault diagnosis modeling is completed, then pre-training of the model is completed by using a source domain sample, and fine tuning is performed on the model based on a target domain sample; and S4, inputting the target domain test set into the fine-tuned model to generate a fault diagnosis result. By adopting the method, the limitation of the CNN convolution kernel size is broken through, the local correlation among sensor data can be fully acquired, the parameter quantity is reduced, the calculation efficiency is improved, and the calculation complexity is reduced.

The invention discloses an aerospace field test case generation method based on a knowledge graph. The method comprises the following steps: establishing a test field knowledge graph model comprising four layers of entities, namely a test item, a test case, a test process and a problem list; identifying test entities from a historical test document, and establishing inclusion, execution, generation and association relationships among the entities; filling the identified entities and relationships into the knowledge graph model to construct a knowledge graph, and optimizing the constructed knowledge graph; based on an input test item of a to-be-tested item, recommending a most relevant test case entity and a relevant test process and question list information thereof in the constructed and optimized knowledge graph; and inputting the recommended test case, test process and question list information into the large model to generate a structured test case text conforming to a project language style and a test environment. The invention aims to solve the problems of insufficient structuralization, weak reasoning ability and poor dynamic adaptability of the existing test case generation.

This invention discloses a multi-source information fusion-based equipment health diagnosis and management system, relating to the field of equipment management technology. It includes: an information acquisition module, a feature perception module, a processing and fusion module, an equipment modeling module, an evolution prediction module, a root cause diagnosis module, a cross-domain testing module, a learning and sharing module, a status assessment module, a collaborative decision-making module, a chain ledger module, and a visual decision-making module. This invention significantly improves the foresight and sensitivity of anomaly detection, reduces noise interference and sampling bias, makes the fused data more stable and physically consistent, supports differentiated diagnosis and individual-level prediction, improves the interpretability of diagnosis and the credibility of decision-making, avoids false alarms and missed alarms, enhances the reliability of early warnings, and achieves scientific, transparent, and traceable health management.

The invention discloses a depth estimation method and system based on domain memory perception guidance, and the method comprises the steps: obtaining a source domain data set and a target domain data set which respectively comprise a segmentation label true value and a depth label true value; dividing the target domain data set into a target domain fine tuning set and a target domain test set; inputting the target domain fine tuning set into a pre-trained source domain model for cross-domain adjustment to obtain a cross-domain fine tuning model; wherein the pre-trained source domain model is trained by a source domain data set, a high-order polynomial memory operator HPM, a hierarchical memory interaction mechanism and a Gaussian guidance factor, and the cross-domain adjustment is the same as the training process of the source domain model; and inputting a target domain test set into the cross-domain fine tuning model to obtain a target domain depth estimation prediction result. According to the method, the dependency problem of a large-scale depth data set is effectively avoided, and the application bottleneck of small sample depth estimation is broken through.

This invention discloses a software testing requirements analysis method based on knowledge graphs, comprising the following steps: Step 1: Collecting requirements-side and verification-side data to form a test requirements analysis corpus; Step 2: Extracting semantic objects of requirements, behaviors, and assertions; Step 3: Constructing a three-domain test knowledge graph; Step 4: Generating a test assertion tension field; Step 5: Inputting the test assertion tension field into an improved Hypergraph Transformer model to obtain a test requirement tension propagation feature map; Step 6: Generating a test requirement gap map; Step 7: Obtaining the test requirement decomposition results; Step 8: Updating the three-domain test knowledge graph and the test assertion tension field based on test execution feedback. This invention employs an assertion gap tension resonance mechanism and an improved Hypergraph Transformer model to improve the accuracy of software testing requirement gap identification and decomposition.

The application relates to a visual language model test adaptation method based on logarithmic calibration and consistency cache, and relates to the technical fields of computer vision, pattern recognition, machine learning and artificial intelligence. The method aims at the problems of class prediction bias, insufficient cache sample coverage and low utilization rate of boundary samples of a visual language model in a target domain test stage, and constructs an online adaptation framework composed of an image encoder, a text encoder, a dynamic logarithmic calibration module, a consistency guide exploration cache module and a cross-modal joint optimization module. The method improves the identification opportunities of difficult classes and low-frequency classes through a dynamic logarithmic calibration mechanism, and improves the overall class balance and classification stability. Through the consistency guide exploration cache mechanism, the coverage range of the cache to the real distribution of the target domain is expanded, and the perception ability of the model to the decision boundary region is enhanced, so that the adaptation effect in a complex distribution offset scene is improved.

The application discloses a test device, a test cabinet and a test method, and relates to the technical field of testing. The test device comprises a supporting piece, a containing groove for containing a tested piece and limiting the tested piece; a base in sliding connection with the supporting piece; a test terminal fixed to the base; a moving assembly for driving the supporting piece to slide on the base so that a physical interface of the tested piece is plugged with or separated from the test terminal; a first supporting assembly with a first end rotationally arranged on the supporting piece; and a linkage assembly connecting the moving assembly and the first supporting assembly. The application drives the supporting piece to slide by the moving assembly so that the physical interface is plugged with or separated from the test terminal, and the power of the moving assembly is transmitted to the first supporting assembly by the linkage assembly. When the moving assembly drives the supporting piece to slide away from the test terminal, the first supporting assembly is rotated to the second end supported on the bearing surface of the bearing base, so that the physical interface and the test terminal are stably separated and are prevented from being damaged.

The invention discloses a test case generation method and device, a medium and a product, and relates to the technical field of automatic testing, and the test case generation method comprises the steps: obtaining a Schema file of a financial industry specification standard for generating a test case; under the condition that the Schema file is a first version Schema file, analyzing the first version Schema file to obtain first target field information; under the condition that the Schema file is a second version Schema file, comparing an old version Schema file with the second version Schema file, determining difference contents, and analyzing the difference contents to obtain first target field information; and generating a test case according to the first target field information. According to the invention, automatic generation of the corresponding test case after the financial industry specification standard is changed is realized.

The invention belongs to the technical field of vibration data identification, and discloses a real-time adaptive learning bearing fault classification method and system based on prototype alignment and a parameter-by-parameter adaptive learning rate, and the method comprises the steps: inputting a source domain sample into a source domain pre-training module, obtaining a basic classification model through supervised training, and carrying out the training of a source domain pre-training module; extracting a feature mean value of each category to construct a source domain feature prototype set; in a target domain test stage, generating a relatively stable pseudo tag for a target domain sample by using an EMA Teamer module; then, an online self-adaptive updating process is executed on a target domain sample by using a pseudo tag, and feature alignment is performed in the process by combining symmetric cross entropy loss and prototype-based comparison loss; meanwhile, model parameters are stably updated by adopting a parameter-by-parameter adaptive learning rate strategy; and finally, outputting an accurate prediction label for a target domain sample by the model after adaptive training.

The present invention provides a weighted multi-source domain transfer learning method for bearing remaining life prediction, comprising: obtaining vibration signals over the entire bearing life cycle to determine a bearing degradation dataset; dividing the bearing degradation dataset into a source domain and a target domain, and dividing the target domain into a target domain training portion and a target domain test portion, thereby constructing a multi-source domain transfer learning task dataset; training a source prediction model based on the source domain; fine-tuning the source prediction model based on the target domain training portion to obtain a bearing target prediction model; and obtaining a final bearing remaining life prediction, final prediction uncertainty, and root mean square error based on the target domain test portion and the bearing target prediction model. Taking into account the data scarcity of the target domain, the present invention incorporates a calibration term into the loss function to calibrate the prediction uncertainty during model training. This calibration term can be used to identify negative transfer and quantify the contribution of different prediction models to the target domain prediction task, thereby achieving better remaining life prediction.

The application discloses a test case generation method and system, and belongs to the technical field of functional safety testing. The test case generation method automatically extracts structured data by performing semantic analysis on a functional safety requirement document, the structured data includes an ASIL level, a functional parameter threshold boundary and a safety target, and automatically determines a test method set based on the ASIL level and a preset rule, and automatically generates a test data set based on the ASIL level and the functional parameter threshold boundary, and then automatically generates a test case set in combination with the test method set, the test data set and the safety target, so that manual checking of standards or manual writing of cases are not needed in the whole process, and therefore, the generation time of functional safety test cases can be shortened, and the efficiency of test case generation can be improved.

The invention discloses an aerospace equipment data-real fusion test cross-domain scene-task generation and evaluation method, and belongs to the technical field of equipment digital testing. The method comprises the following steps: constructing a cross-domain test definition model of a layered architecture; analyzing a natural language demand by using a large language model in combination with a domain knowledge base, and generating a cross-domain task flow and a variable scene instance; generating a scene image conforming to a physical rule through a fine tuning generative model and a visual language model closed-loop verification mechanism, and reconstructing the scene image into a dynamic simulation scene with physical attributes; and finally, driving a digital-real fusion system to operate, and completing automatic evaluation through a digital-real synchronous verification mechanism and a multi-dimensional performance evaluation model. According to the method, the problems that scene construction depends on manpower and the number-real consistency is poor in aerospace equipment testing are effectively solved, and high-credibility and automatic number-real fusion testing is achieved.

The invention discloses a test case generation method and device, a vehicle and a storage medium, and the test case generation method comprises the steps: obtaining a target demand document; determining target demand information according to the target demand document; determining an initial test path according to the target demand information; performing generalization operation on the initial test path to obtain a plurality of target test paths; and generating a target test case according to the plurality of target test paths. The technical problem of low precision of the vehicle body domain test case generated by a large model in the prior art is solved.

The application provides a building group energy consumption prediction method and system based on space-time graph convolution and adversarial domain adaptation, comprising: obtaining source domain related data and target domain related data, preprocessing to obtain initial input data, etc.; constructing a space-time graph convolution network model and pre-training to obtain a pre-trained space-time graph convolution network model, splitting the model into a space-time feature extractor and a predictor; performing weight initialization and adversarial training on the target domain space-time feature extractor and the target domain predictor to obtain a trained target domain space-time feature extractor and a trained target domain predictor; inputting a target domain test set into the trained target domain space-time feature extractor and the trained target domain predictor to obtain a target domain regional building group energy consumption prediction result; and the application captures the spatial dependence relationship between buildings based on graph convolution and graph attention mechanism, adopts a transfer learning strategy to transfer source domain prediction related knowledge to a target domain, and accurately predicts energy consumption by using a small amount of data.

The invention relates to the technical field of electrical equipment diagnosis, and particularly discloses a sample scarcity-oriented transformer fault identification and generalization improvement method and system, and the method comprises the steps: constructing a frequency domain wave equation and an impedance boundary condition of a transformer; embedding a frequency domain wave equation and an impedance boundary condition into a neural network node and a loss function of the initial physical information network model to obtain a target physical information network model, and constructing a training sample set and a target domain test set of multiple meta-learning tasks based on historical voiceprint data and fault sample data, training the target physical information network model by adopting model-independent meta-learning, and performing generalization ability test on the trained model based on the target domain test set to obtain a target meta-model; the target meta-model is embedded into an online voiceprint recognition system for feature extraction and fault type recognition, transformer fault tags are output, and the method improves the generalization ability under the conditions of fault sample scarcity, cross-equipment and cross-working conditions.

The application discloses a three-directional vibration high-end machine tool joint surface contact characteristic testing device and method, and belongs to the field of machine tool dynamic characteristic analysis. The testing device and process mainly comprise the following steps: before testing, a first support and a second support are fixed on a base through bolts, a tested piece is screwed on a boss of the base through a pre-tightening bolt, and a vibration exciter system is hung on the support through a rope; a vibration mode database in a computer is called, the number of sensors and a layout position scheme are determined according to the vibration mode of a tested piece with different structures; the sensors are pasted on the tested piece according to the obtained scheme, the output ends of the sensors are connected to the computer through a signal acquisition system, and the contact characteristic of the machine tool joint surface is obtained by processing the computer. The device has the advantages of simple structure, convenient loading and unloading, easy repeated experiment, sensor layout scheme according to the shape of the tested piece, and fast and accurate measurement of the dynamic characteristic of the machine tool joint surface.

The application relates to the technical field of wind power generation equipment predictive maintenance, in particular to a wind power generation key component multi-working condition residual life prediction method S and a target domain sample set D T ; a feature extractor module is constructed; a multi-feature adaptive module is constructed; a migratable attention mechanism module is constructed; a residual service life predictor module is constructed; a domain adaptation module is constructed; a residual service life prediction model is trained; residual service life of a wind power generation key component under a target domain working condition is predicted, target domain test sample sets are input into the established network model, and residual service life prediction results of the key component at each time are output. The wind power generation key component multi-working condition residual life prediction method proposes a migratable attention mechanism module, the module can dynamically activate degradation features with high migration in model training, so that the generalization capability of the model is improved.

A method and system for identifying small-sample bearing fault modes based on multi-source data integration, relating to the field of intelligent operation and maintenance and health management of mechanical equipment. It solves the problems of difficulty in identification and inaccurate identification due to the limited number of samples in existing bearing fault modes. The method includes: constructing multi-source data samples based on a public dataset; extracting features from the multi-source data samples to construct source domain training sets and target domain training sets; combining the target domain training sets and source domain training sets to obtain a combined source domain training set; training a base classifier based on the target domain test set and the combined multi-source domain sample set; calculating inter-domain distribution metrics and sample similarity; constructing a weight matrix; classifying the target domain test set using the base classifier to obtain class probabilities; and weighting and integrating the class probabilities according to the weight matrix to obtain the classification results of the target domain test set, thus completing fault mode identification. This invention is applied in the field of fault identification.

The application provides a SPN test method and platform, and belongs to the field of communication technology. The SPN test method is applied to a SPN test platform, and comprises the following steps: determining configuration parameters of a test case to be executed by a SPN device according to information of the SPN device; sending a configuration message to the SPN device, wherein the configuration message comprises the configuration parameters of the test case; and receiving operation results returned by the SPN device after the test case is executed based on the configuration message. The technical scheme of the application can improve test efficiency and save test time.

The present invention proposes a semi-supervised medical image segmentation method based on adaptation, which is implemented by the following steps: obtaining source domain and target domain 3D medical images and preprocessing them; obtaining a source domain training sample set; obtaining a labeled target domain training sample set, an unlabeled target domain test sample set, and a target domain training sample set; constructing a semi-supervised medical image segmentation model O and defining its loss function Loss; iteratively training the semi-supervised medical image segmentation model O; and obtaining the segmentation results of the semi-supervised medical image. The source domain training sample set of the present invention contains domain adaptive images corresponding to all source domain image slices. During the training of the semi-supervised medical image segmentation model, the uncertainty of the domain adaptive teacher model and the target domain teacher model is used to guide the student model to learn more reliable target domain prediction probability maps and domain adaptive prediction probability maps, thereby fully utilizing the source domain data and effectively improving the segmentation accuracy of the medical image.

The invention provides a large language model cross-domain illusion detection method and system, and the method comprises the steps: screening an optimal template from a candidate template set according to the authenticity direction variance ratio of each candidate template; splicing the unknown field test text with the optimal template to obtain a spliced test text; performing feature extraction on the spliced test text through a large language model to obtain a to-be-detected feature vector; and detecting the to-be-detected feature vector through the trained detector to obtain reliability data of the cross-domain test. Compared with the prior art that a cross-domain consistency mode is difficult to capture by features directly extracted by LLM and generalization ability of a detector is limited, the method has the advantages that the optimal template is determined according to the authenticity direction variance ratio of the candidate template, and the plug-and-play optimal template is used as a prompt template to optimize the feature extraction process, so that the accuracy of feature extraction is improved. The significance and cross-domain consistency of the authenticity related structure are enhanced, and the generalization performance of the detector in the unknown field is improved.

The invention discloses a batch verification maximization fault diagnosis system and method for life stage migration. The dependence on target domain labeling is remarkably reduced while cross-stage high-precision fault recognition is achieved. The system comprises a life stage pseudo-domain division label generation module which inputs source domain and target domain test data containing timestamps or residual service life information and outputs data subsets corresponding to all stages; the stage perception sharing-specific feature coding module is used for inputting original signal data of each stage and a corresponding stage label, and outputting a shared feature and a stage specific feature; the channel-stage fusion selection module is used for receiving the shared features and the stage specific features from the stage perception sharing-specific feature coding module and outputting fused features; and the multi-stage batch nuclear norm maximization migration module is used for inputting fusion features and stage label information, respectively calculating a Softmax output matrix for each pseudo-domain, and maximizing the batch nuclear norm of the Softmax output matrix so as to enhance cross-stage discrimination and feature consistency.

The embodiment of the invention provides a test case generation method and device, a test method and device, electronic equipment and a computer readable storage medium, and relates to the field of natural language processing. The test case generation method comprises the following steps: obtaining test information; repeatedly executing the first operation until the trap text output by the first large language model meets a preset condition, and taking the trap text meeting the preset condition as a test case of the RAG system; the first operation comprises obtaining a trap text based on the test information and the first prompt information through a first large language model, and the trap text comprises semantic information of each trap fragment in the fragment set; the preset condition comprises that the output trap text comprises a correct answer, and the first relevancy between the output trap text and the question text is not lower than a relevancy threshold value. According to the embodiment of the invention, a basis is provided for analyzing and generating semantic understanding of the model from more levels.

The invention provides a testing system and a testing method, and belongs to the technical field of testing. The test system comprises an upper computer, an IO board card and a real-time simulation device. The upper computer and the controller to be tested are connected through an IO board card, and the two ends of the real-time simulation device are connected with the upper computer and the controller to be tested respectively. The upper computer can read and traverse the test table line by line and obtain the first actual observed quantity and the second actual observed quantity of the tested signal, and if the first actual observed quantity and the second actual observed quantity are within the tolerance range allowed by the expected value of the first observed quantity and the expected value of the second observed quantity in the test table, the transmission of the tested signal is successful. The test system can carry out signal verification of two observed quantities on the same signal, and the robustness and safety of the controller software to be tested are enhanced. And meanwhile, automation of testing multiple points of each tested signal is realized, and the coverage degree and the testing efficiency of software testing are improved.

The application discloses an IGBT module state prediction method based on a domain adversarial long short-term memory network, and comprises the following steps: collecting aging data of different IGBT modules respectively, cutting different sample sequences from sample data of known IGBT modules respectively, and labeling after normalization processing; an adversarial transfer state prediction network for unknown IGBT modules is constructed; in the training stage, source domain and target domain data are mapped into a high-dimensional feature space to obtain data feature distribution; domain invariant features of the two domains are learned by using a domain adversarial module to perform feature distribution matching; a weighting discrimination mechanism is further provided to evaluate the similarity degree of target domain sample data and source domain data, discriminate the transferability of data, and thus effectively improve the classification performance of the state category; target domain test data is input into the trained prediction model for testing, the current running state of the equipment is discriminated by calculating the obtained weight value, and the final classification prediction result is output.

The application discloses a total nitrogen and total phosphorus inversion method based on sample migration and lasso regression, which comprises the following steps: step 1, collecting water quality data of multiple sites and sorting the water quality data of each site according to time respectively; step 2, pre-processing; step 3, for each site, dividing a target domain training set and a target domain test set respectively; and determining a source domain auxiliary training set; step 4, constructing a TrAdaboost transfer learning model based on a weak learner lasso algorithm, training a model corresponding to the site by using a combination of the source domain auxiliary training set and the target domain training set, obtaining a trained sample transfer learning model corresponding to each site, and inputting the target domain test set corresponding to the site into the trained sample transfer learning model to obtain a model output of total nitrogen and total phosphorus results corresponding to the site. The application can effectively improve the detection efficiency and the detection precision of measurement results.