20results about How to "Ensure availability" patented technology

The invention discloses a manufacturing industry production data real-time management system and method based on Internet of Things edge computing. Comprising a data acquisition and preprocessing module, a lightweight agent module, an edge computing module, a network transmission module, a platform management module, an application service module, a security protection module, a self-adaptive resource configuration module and a monitoring and diagnosis module. According to the invention, nearby acquisition, preliminary cleaning and real-time analysis of multi-source heterogeneous data from a sensor, a controller and industrial equipment are realized, the cloud burden is effectively reduced, and the overall processing efficiency is improved. A real-time anomaly detection mechanism based on a sliding window and machine learning is constructed, parameter deviation, equipment faults or process anomalies in the production process can be recognized within millisecond-level time, and early warning or control feedback is triggered.

This disclosure proposes a method for synthesizing multimodal data for decision-making attribution explanation in the mining industry, comprising the following steps: First, extracting causal logic from industry texts to construct a structured rule base; second, collecting on-site multimodal data and filtering out data pairs strongly correlated with rules through scenario-based weight modeling and cross-modal association mechanisms; third, using a multimodal generative adversarial network with rule constraint mechanisms and combined with rule mutation technology to synthesize multimodal data with complete logical chains and covering various complex working conditions; finally, ensuring data quality and achieving self-evolution of the knowledge base through a three-level verification and rule feedback mechanism, and storing qualified data in a hybrid storage database. This invention solves the problems of complex causal relationships and scarce high-quality training samples in the mining field, and can automatically and in batches generate logically rigorous and scenario-diverse multimodal data, improving the training efficiency and generalization ability of attribution decision models.

The invention discloses a skill server cluster management method and device, a storage medium and an electronic device, and relates to the technical field of smart home, the skill server cluster management method is applied to a management server, and comprises the following steps: sending an online query request to each skill server in a skill server cluster at a set period; wherein the skill node information of the skill server is stored in the management server; if the online feedback of the skill server is still not received under the preset feedback receiving cut-off condition, determining that the skill server is abnormal; and setting the state identifier of the skill server to be abnormal so as to move the skill server out of the skill server cluster. According to the technical scheme provided by the embodiment of the invention, the abnormal skill server nodes can be eliminated in time in the skill service execution process, so that the reliability and availability of the skill service are ensured.

The application discloses a high-resolution satellite image urban area shadow-removed ARD production method and device, and belongs to the technical field of remote sensing image processing. The method comprises the following steps: firstly, screening out images covering urban areas; performing radiation calibration on high-resolution satellite images covering urban areas to obtain apparent radiance images; based on the apparent radiance images, performing preliminary shadow detection by using a pre-trained deep learning model; using a near-infrared band of an atmospheric-corrected ground reflectivity image, performing false detection elimination on the preliminary detection result by adaptive threshold segmentation to obtain a final shadow mask; performing histogram matching processing on the shadow area based on the shadow mask on the atmospheric-corrected ground reflectivity image to realize shadow removal; and finally, outputting the shadow-removed ARD product. The application significantly improves the shadow detection precision and the quantitative reliability of the ARD product, and is suitable for large-scale ARD production of high-resolution satellite images in urban areas.

The invention relates to the technical field of biological medicines, and particularly discloses a drug-resistant cell line MC38-IR for immunotherapy of mouse colon cancer as well as a preparation method and application of the drug-resistant cell line MC38-IR. The cell line is preserved in the China Center for Type Culture Collection, and the preservation number is CCTCC NO: C2025358. Experimental results show that the MC38-IR cell line has a remarkable PD-1 antibody drug resistance characteristic while maintaining the in-vitro proliferation and migration capabilities of parent cells. An in-vivo experiment shows that the tumor volume of a mouse inoculated with the cell line is not obviously inhibited after PD-1 antibody treatment, the lifetime has no statistical difference compared with that of a control group, and the parent MC38 cell shows obvious treatment response. The invention solves the problem of lack of colon cancer immunotherapy drug resistance models in the prior art, provides an important tool for researching drug resistance mechanisms, screening reverse drug resistance drugs and developing combined treatment strategies, and has higher scientific research and clinical application values.

The invention relates to the technical field of computer networks, and discloses a communication connection adaptive adjustment method, device, electronic equipment and system, and the method comprises the steps: attempting to obtain server certificate information based on a certificate dynamic obtaining strategy when a server address received by an application program meets a communication protocol switching condition, and obtaining an information obtaining result; verifying the server certificate information based on a multi-level certificate verification system to obtain a certificate verification result; determining a first target communication protocol which can be used for the server address according to the information acquisition result and / or the certificate verification result; the first target communication protocol comprises an HTTP protocol or an HTTPS protocol; and establishing a dynamic communication connection between the application program and the server based on the first target communication protocol. Therefore, the communication connection strategy can be adjusted in a self-adaptive mode, the selection flexibility and reliability of the communication connection between the client side and the server are improved, and therefore the availability and continuity of communication services are ensured.

The invention provides an adaptive SSMVEP-MI fusion brain-computer interface method and system based on eye movement tracking. The method comprises the following steps: determining a current effective fixation area of a user in real time through eye movement tracking; based on the region, only activating a corresponding local SSMVEP visual stimulation target, and generating associated specific motor imagery prompt information; electroencephalogram signals are synchronously collected, and SSMVEP features and MI features are extracted; self-adaptively selecting a decoding path according to whether the accumulated number of the MI effective samples with the labels reaches a threshold value or not: if not, independently decoding by using the SSMVEP features; if yes, fusion decoding based on signal quality evaluation is carried out on the bimodal features; and finally outputting a control signal to drive external equipment or update an interactive interface. According to the method, the visual interference is remarkably reduced, the recognition accuracy and the training efficiency at the initial stage of motor imagery are improved, and the robustness and the availability of the system are ensured through a dual adaptive mechanism of the sample number and the signal quality.

The application relates to the technical field of battery safety detection, and discloses a power battery safety detection and evaluation system based on multi-modal visualization, which comprises a detection device and an evaluation device, wherein the detection device is connected with the evaluation device; the detection device comprises a sealed test cavity, a safety trigger and loading module, a multi-modal parameter acquisition module and a data processing and visualization module; an analysis module constructs a coupling characteristic vector based on temperature characteristics, smoke characteristics and deformation characteristics, and determines a safety evaluation value of a battery to be detected according to the coupling characteristic vector; a judgment optimization module judges whether the safety evaluation value needs to be optimized according to power data; if yes, an optimization coefficient of the safety evaluation value is determined based on the air pressure in the bin and the power data, and an optimized safety evaluation value is obtained; and a judgment module determines the safety grade of the battery to be detected according to the optimized safety evaluation value. The application can comprehensively, accurately and intuitively detect and evaluate the safety condition of the power battery.

The application relates to a fusion context-aware hierarchical state space speech recognition method, comprising the following steps: acquiring an original audio signal to be recognized, performing acoustic feature preprocessing on the original audio signal to obtain a first acoustic feature sequence; constructing a hierarchical state space speech recognition model, inputting the first acoustic feature sequence into the constructed hierarchical state space speech recognition model; sequentially performing high-resolution state space modeling, low-resolution semantic bottleneck reasoning and high-resolution time grid alignment through the hierarchical state space speech recognition model to obtain a target acoustic representation; and inputting the target acoustic representation into a decoder to generate a recognized text corresponding to the original audio signal. The application can realize efficient and accurate speech recognition.

The application discloses a dynamic redundancy robust digital blind watermarking method and system, and relates to the technical field of image data processing. The method embeds watermark information into pixel data (robust channel) and metadata (fast channel) of a carrier image simultaneously through a double embedding strategy. When the watermark is extracted from the carrier image subjected to geometric attacks, the system firstly attempts to extract from the metadata; if the fast extraction fails, the system automatically switches to the pixel domain for robust extraction. The method comprises the following steps: optimizing and packaging the data to be embedded; embedding the packaged data stream into the pixel and writing the compressed watermark data into the metadata; and adopting the aforementioned double-channel extraction strategy for extraction. The design effectively avoids the defects of simply relying on spatial synchronization, and significantly improves the extraction performance and reliability of the watermark system under geometric attacks.

The application relates to the technical field of knowledge base generation, and particularly discloses an education knowledge base content generation method and system based on AI big data. The method comprises the following steps: comparing user answer content with answer information to obtain a matching degree, calibrating positive samples and negative samples to train an evaluation model, and determining a reference standard score according to the quantity relationship between the positive samples and the negative samples; based on the reference standard score and a tolerance threshold, a plurality of candidate knowledge contents are screened to calibrate them as stable knowledge contents or to-be-corrected knowledge contents, and are output. By constructing an evaluation system based on user real answer data and an evaluation standard reflecting the actual application effect of the knowledge content, and automatically calibrating the evaluation standard according to the data distribution, objective screening and continuous iteration of the knowledge content are realized, and the accuracy and reliability of the knowledge base content are improved.

This invention discloses a text correction method, apparatus, electronic device, and storage medium. The method includes: acquiring first text, which is a text result generated by speech recognition of target audio data; detecting the first text using a target text detection model to obtain a first detection result; wherein the target text detection model is a pre-trained neural network model used to detect whether the text input to the model is a reference type statement; in response to the first text not being a reference type statement, performing text correction on the first text using a target text correction model to obtain a text correction result; the target text correction model is a pre-trained neural network model used to perform character recombination on the text input to the model and filter reference noise characters for text correction. This invention enables automatic detection and correction of text results from speech recognition in multi-speaker mixed scenarios.

The invention relates to the technical field of traffic control, in particular to a second-hand car market mixed flow traffic control method and system based on multi-target recognition. The method comprises the following steps: in response to an activation signal of a barrier gate triggering device, collecting multi-dimensional identification original data of vehicles passing through a barrier gate of a second-hand vehicle market through a multi-source identification device; performing cross-device data calibration processing on the multi-dimensional recognition original data, performing feature matching, and generating a vehicle identity recognition result and an identity confidence coefficient parameter; according to the vehicle identity recognition result and the identity confidence coefficient parameter, a dynamic passing strategy engine is called to generate a vehicle passing permission evaluation result and a strategy adaptation coefficient; and generating vehicle passing state record data according to the passing authority evaluation result and the strategy adaptation coefficient, performing classification analysis, generating an intervention processing work order, and feeding back the intervention processing work order to the system optimization module for iteratively updating the identification model and the passing strategy. The barrier gate passing efficiency of the second-hand car market can be greatly improved.

This invention relates to the field of automatic control of nitrogen valves, and particularly to an automatic control method for nitrogen valves based on intelligent sensing. First, based on the measured values ​​of the acquired original sensing variables, preprocessing is performed to obtain preprocessed measured values ​​of the sensing variables. Based on the preprocessed measured values ​​of the sensing variables, a time-varying energy consistency fusion algorithm is used to obtain the state-driven energy value of the sensing variables. Based on the state-driven energy value of the sensing variables, the contribution of the sensing variables is determined, and a fusion situation indicator is constructed. Then, based on the fusion situation indicator, a control potential energy function is constructed using a continuous potential energy-driven valve position decision algorithm. Based on the control potential energy function, the valve adjustment speed is obtained. This method solves technical problems such as insufficient utilization of multi-source sensing information, inaccurate state assessment, delayed adjustment response, and lack of effective intervention for abnormal risks.

The invention relates to the technical field of artificial intelligence, and particularly provides a dialogue tool calling method and device, a storage medium and electronic equipment, and the method can comprise the steps: analyzing a user dialogue message, and determining a current dialogue stage; wherein the current dialogue stage comprises a user initial access stage, an intention identification stage, a task execution stage or a verification stage; obtaining an initial tool list matched with the current dialogue content; wherein dialogue tools in the initial tool list are used for outputting corresponding reply content according to the current dialogue content; updating and sorting the initial tool list based on the current dialogue stage, and obtaining a target tool list; calling a current dialogue tool from the target tool list; wherein the current dialogue tool is used for generating the reply content. According to the embodiment of the invention, the dialogue tool can be automatically and accurately called, and the dialogue reply precision is improved.

The application discloses a Bluetooth transmission method and system, the method comprises the following steps: sending a plurality of Bluetooth data packets while detecting Bluetooth connection; after sending the Bluetooth data packet each time, judging whether the ACK response is correctly received; if yes, increasing the initial maximum transmission window of Bluetooth in response to the correct reception of the ACK response; if not, using dichotomy to find the optimal window size of the current transmission between the initial maximum transmission window and the transmission window threshold, the transmission window threshold being smaller than the initial maximum transmission window; repeatedly calculating the optimal window size of several transmissions, and determining the best Bluetooth transmission window based on the optimal window size of several transmissions for data transmission; the application can improve the efficiency of Bluetooth transmission data.

The application discloses a kind of virtual power plant distributed transaction method and platform based on alliance block chain, create account in application state;Become available state account by admission mechanism application;Virtual power plant carries out power and fund pre-allocation to available state account according to each equipment day before reported energy supply / consumption energy;User carries out electric energy transaction according to virtual power plant real-time equipment power and account pre-allocation power and fund, with other node user in platform and virtual power plant interior;Check the sell sign less than or equal to the buy sign price in the same period, select the strongest seller of transaction stability to knock down transaction;Record in each node account book of block chain by alliance block chain;Before next pre-allocation, calculate the remaining electric energy of all user accounts last time pre-allocation, obtain the latest stability value of each account, and carry out m times electricity fee punishment to the user with deviation from pre-allocation.Guarantee that user can complete block chain transaction in front-end interface.

This invention discloses a method, apparatus, medium, and device for generating prompt words based on a large language model. Belonging to the field of prompt word generation, this application acquires multi-source runtime data from a development scenario processed according to a model context protocol, performs semantic understanding on this data using a large language model, and selects the optimal prompt word template suitable for the current development intent from a pre-set prompt word template library based on template metadata specifications. Further, according to the semantic mapping relationship established by the model context protocol, parameter values ​​matching the optimal template are selected from the multi-source runtime data, and the accuracy and consistency of the parameter values ​​are ensured through multiple verifications. Finally, the verified target parameter values ​​are filled into the optimal prompt word template to generate a complete prompt word for large model inference. This application effectively solves the problem that existing technologies cannot accurately and efficiently generate prompt words for large language models.

This invention discloses a method and system for recovering encrypted files based on key context. The method includes: acquiring the underlying binary image of the target storage medium; parsing the file system structure to locate the inode table; filtering encrypted file inodes whose status is deleted, and extracting key information including data block pointers, master key identifiers, and encryption policy descriptors; obtaining the master key from the associated system kernel based on the master key identifier, and calculating and generating a file content encryption key in combination with the encryption policy descriptor; decrypting the extracted encrypted data blocks using the key, and reassembling them in sequence to complete file recovery. This invention avoids the huge computational consumption of traditional brute-force attacks by directly utilizing residual metadata of the file system and the master key cached by the kernel for key reconstruction, and, based on the general FBE mechanism, does not rely on specific vulnerabilities, significantly improving the recovery efficiency and applicability for Android FBE encrypted files.

The invention relates to the field of big data, in particular to a big data change trend real-time prediction method, which comprises the following steps of: determining a data stream and a public opinion perception data stream by lagging, analyzing a time-delay coupling relationship between a historical data change position and a sales peak, calculating a dynamic delay window and constructing a parameterized mapping model; a predictive complementary sequence is generated based on current public opinion sensing data, a panoramic data sequence is constructed in combination with confirmed data, then real-time abnormal fluctuation verification is performed on public opinion sensing flow, a delay window is dynamically adjusted according to matching reliability to reconstruct a mapping relation, and the complementary sequence passing the verification is output as a prediction result. The technical effects of multi-source heterogeneous data fusion, time-delay relation adaptive evolution and prediction process real-time calibration are achieved, the timeliness and accuracy of trend prediction are improved, and scientific data support is provided for drug sales and drug management.