11results about How to "Reduce data" patented technology

The invention discloses an electronic nose gas classification model construction method and system, and relates to the technical field of gas sensing detection and artificial intelligence crossing. The method comprises the steps of performing fine adjustment on a pre-trained Transform encoder based on drift-free source domain data, constructing a source domain reference model and initializing an online updating component; in the real-time classification stage, after preprocessing and feature textualization conversion are carried out on target domain data, the drift mode type of the target domain data is judged by calculating feature distribution similarity, and representative samples are screened from current data; aiming at the new drift mode, training a new LoRA adapter; for a known drift mode, loading an existing LoRA adapter and carrying out lightweight fine tuning; and finally, obtaining a model for gas classification. According to the method, self-adaptive and lightweight online learning of sensor drift is realized, the calculation overhead and the storage requirement are reduced while the classification precision is ensured, and the long-term stability of an electronic nose system in a dynamic environment is improved.

The invention discloses a real-time storage optimization method based on block data, which relates to the technical field of big data processing and distributed storage, introduces a block data concept, realizes block-by-block writing, in-block ordering, block-level ordering and an intelligent flush trigger mechanism, and realizes real-time writing with high throughput, low delay and low memory occupation. The method specifically comprises the steps that block data is introduced into a data memory model to serve as a basic processing unit, multiple rows of data are aggregated into a block according to an aggregation key CKValue, row data in the block are sequenced according to a primary key PK, and write-in, namely aggregation is achieved; writing data, aggregating according to blocks, and uniformly sorting in the blocks; one block, namely multiple rows of data, generates an FRC file, and small files are reduced; exception processing: supporting block-level recovery to skip a failed block; the core problems of too high memory occupation, large CPU consumption, high writing delay and the like in a high-concurrency and high-throughput real-time writing scene can be solved, and the method is widely applied to scenes with extremely high requirements on data real-time performance and system stability.

The present invention relates to an image decoding method using residual information in an image encoding system and an apparatus thereof. According to the present document, an image decoding method performed by a decoding apparatus includes the steps of receiving a bitstream including residual information of a current block; deriving a particular number of context-coded bins for a context syntax element of a current sub-block of the current block; decoding the context syntax element of the current sub-block included in the residual information based on the particular number; deriving transform coefficients of the current sub-block based on the decoded context syntax element; deriving residual samples of the current block based on the transform coefficients; and generating a reconstructed picture based on the residual samples.

The invention relates to the field of DC micro-grid stability analysis, and provides a mechanism and data driving fused DC micro-grid low-frequency dynamic parameter identification and evaluation method, which comprises the steps of determining equivalent output impedance of a DC voltage control unit when the DC voltage control unit works in a droop mode; defining an impedance equivalent transformation ratio considering the impedance influence of the outlet line of the direct-current voltage control unit; establishing a constant power load equivalent circuit model; carrying out constant-power load equivalent modeling to obtain a reduced order form of the total parallel impedance of the direct-current micro-grid; measuring DC voltage and DC current output by a port of the DC voltage control unit before and after disturbance, extracting steady-state voltage and current values, calculating and obtaining steady-state deviation values of the voltage and current before and after disturbance, and obtaining a droop coefficient estimation value of the DC voltage control unit; an objective function for solving the optimization problem is determined; acquiring equivalent load resistance; and the low-frequency dynamic characteristics of the system are analyzed by using the damped oscillation frequency, the damping coefficient, the peak time and the overshoot.

The invention relates to the technical field of general electric digital data processing, in particular to a thermal process control method and system for a quartz diffusion tube, and the system comprises a simulation unit which is used for carrying out the modeling processing according to the physical structure of the quartz diffusion tube, obtaining a digital twinborn body representing the quartz diffusion tube; a related process group related to the thermal process is obtained, the process parameter range of the thermal process is determined according to the process parameters of the thermal process in the related process group, and the related process group is a process combination which jointly influences the quality of the quartz diffusion tube; performing thermal simulation processing of transient temperature distribution on the digital twin in the process parameter range to obtain a simulation result; and based on the simulation result, determining a target process parameter from the process parameter range, and sending the process parameter to equipment related to the thermal process, so that the equipment executes the target process parameter.

The present disclosure provides a lighting device and a method and system for controlling the lighting device, wherein the lighting device control method is a method for controlling the lighting device by an effect application executed by at least one processor on a director terminal, comprising the steps of: generating a library setting a light emitting mode of a plurality of lighting devices; generating a scenario setting a light emitting mode of the plurality of lighting devices within a specified time; generating a performance preparation data including at least one of the library and the scenario; providing the generated performance preparation data to the plurality of lighting devices; transmitting a lighting device control signal to the plurality of lighting devices, the lighting device control signal commanding execution, interruption or termination of at least one of the library and the scenario within the provided performance preparation data; and implementing a comprehensive lighting control performance effect according to the transmitted lighting device control signal.

A biometric identification device based on a surface light source is characterized by comprising: a housing including a light-emitting surface; a sensor located inside the housing for receiving reflected signals from a palm or face; multiple lateral light sources located inside the housing, emitting active light towards a center point; the center point being the center of a shape formed by the multiple lateral light sources; a light guide plate located on the light path of the lateral light sources illuminating the center point, for causing the active light to exit from the side; and a light-diffusing plate located above the light guide plate to ensure uniform emission of the active light. This invention utilizes lateral light sources illuminating along the light guide plate and a light-diffusing plate to ensure uniform emitted light, resulting in a more uniform light path, lower cost, fewer control components, and is beneficial for large-scale applications.

The present invention relates to an image decoding method using residual information in an image encoding system and an apparatus thereof. According to the present document, an image decoding method performed by a decoding apparatus includes the steps of receiving a bitstream including residual information of a current block; deriving a particular number of context-coded bins for a context syntax element of a current sub-block of the current block; decoding the context syntax element of the current sub-block included in the residual information based on the particular number; deriving transform coefficients of the current sub-block based on the decoded context syntax element; deriving residual samples of the current block based on the transform coefficients; and generating a reconstructed picture based on the residual samples.

The application discloses a lithium battery state of health estimation method and system based on physical information and Mamba, which comprises the following steps: firstly, the battery voltage and current data of each cycle of the lithium ion battery in the charging stage are acquired, and pretreatment is performed to form a pretreatment data set; secondly, two features of constant voltage charging time and constant voltage charging capacity are extracted based on the pretreatment data set, two features of the maximum incremental capacity and the incremental capacity variance are extracted based on the ICA curve, and a feature vector is constructed; then, the feature vector is input into a PINN-Mamba neural network to obtain a SOH prediction value; finally, the PINN-Mamba neural network is optimized and trained by using a loss function, and is tested and verified. The application solves the problem of uneven distribution of features of the battery in different states, and effectively improves the prediction accuracy of the SOH.

This invention relates to the field of electricity theft detection technology, and in particular to a method for determining electricity theft using multi-source data. The method includes: determining initial electricity theft probabilities based on several evidence sources of a user; determining corresponding lower-level and upper-level credibility based on the evidence sources to synchronously adjust the initial electricity theft probabilities and generate a base electricity theft probability; determining a global conflict coefficient representing the differences between different evidence sources based on the base electricity theft probability; and constructing a formula for calculating the initial user electricity theft probability, which includes a confidence factor mapped from the global conflict coefficient. This invention effectively solves the problem of multi-source evidence conflict fusion by dynamically adjusting the weights of the corresponding initial electricity theft probabilities based on the upper and lower-level credibility of the evidence sources and adaptively updating them. The confidence factor preserves the weighted average information of conflicting evidence to avoid fusion paradoxes.

The present invention relates to an image decoding method using residual information in an image encoding system and an apparatus thereof. According to the present document, an image decoding method performed by a decoding apparatus includes the steps of receiving a bitstream including residual information of a current block; deriving a particular number of context-coded bins for a context syntax element of a current sub-block of the current block; decoding the context syntax element of the current sub-block included in the residual information based on the particular number; deriving transform coefficients of the current sub-block based on the decoded context syntax element; deriving residual samples of the current block based on the transform coefficients; and generating a reconstructed picture based on the residual samples.