114 results about "Vector distribution" patented technology

The invention provides a curtain wall deformation condition safety detection method and device based on laser point clouds, and relates to the technical field of constructional engineering, and the method comprises the steps: 1, carrying out the coordinate system unification processing of a three-dimensional point cloud data set covering the whole surface of a curtain wall, generating a curtain wall surface digital twin model, and carrying out the coordinate system unification processing of a three-dimensional point cloud data set covering the whole surface of the curtain wall in the digital twin model; defining a first detection target point and a second detection target point according to geometrical characteristics and material joint characteristics of the curtain wall sheet; 2, according to the digital twinborn model, surface normal vector distribution characteristics are extracted, actually-measured wind pressure load distribution data and temperature field parameters are fused, a dynamic flexural deformation coupling calculation model is constructed, and global deformation parameters and local deformation parameters are output. According to the invention, the digital twinborn model is constructed through point cloud data processing, deformation and displacement parameters are extracted, the safety coefficient is evaluated, and high-precision, omnibearing and dynamic detection and safety risk early warning of the flatness of the building outer wall are realized.

The invention provides a double-pulse high-frequency switching power supply control method and system for precise electroplating, and relates to the field of electroplating, and the method comprises the steps: collecting area surface data through a three-dimensional scanner, and obtaining a three-dimensional model; extracting a high-curvature area grid according to the three-dimensional model, and calculating to obtain a surface concave-convex degree index; performing current field simulation, and calculating current density vector distribution; when the deviation of the current density vector distribution exceeds a preset deviation threshold value, pulse parameters are optimized through a genetic algorithm; selecting a pulse parameter which is most matched with the surface concave-convex degree index, and carrying out iterative calculation to obtain updated current density vector distribution; based on the difference between the updated current density vector distribution and the initial distribution, predicting a coating thickness value, and determining a preliminary evaluation index; and optimizing pulse parameters to obtain a final control scheme of uniform current distribution. According to the method, pulse parameters can be dynamically optimized according to the geometrical shape of the workpiece, and uniform current distribution and stable plating quality are realized.

The invention provides a hydro-junction water level adjusting method and system. Wherein the sediment deposition space-time atlas containing the sedimentary gradient is constructed and periodically corrected through scanning of an underwater robot and synchronous acquisition of satellite remote sensing data. A radar array is deployed at a bend section of a flood discharge channel, and vector distribution of a three-dimensional sediment flow velocity field is analyzed by using multi-frequency pulse signals. Through correlation analysis of the dynamic coupling sedimentary gradient and the flow velocity field, a correlation area of flow velocity sudden change and sediment deposition is accurately identified, and a constraint condition set including coordinate positioning and a flow-sediment matching rule is established. And in combination with real-time water level monitoring data, calculating and generating a dynamic adjustment parameter sequence of the opening degree of the flood discharge gate by using a constraint condition set, thereby realizing accurate regulation and control of the water level. According to the technical scheme, the flood discharge parameters are intelligently optimized, the water level of the hydro-junction can be accurately regulated and controlled, the sediment deposition problem of a high-sediment-content water area is effectively relieved, and safe operation and flood control efficiency of the junction are guaranteed.

The invention relates to a Chinese language and literature computer networking query reading system. According to the system, a query request processing module carries out mixed feature analysis processing according to a query request of a user, and a mixed query vector fusing radicals and semantics is generated; a distributed retrieval module performs distributed retrieval in a preset space-time fragment index structure based on the mixed query vector to obtain a candidate literature data set; the preliminary retrieval optimization module is used for carrying out self-adaptive conversion processing on abnormal characters and traditional and simple characters in the candidate literature data set according to a Chinese character evolution rule and a context to obtain standardized text data; and the visual query output module performs knowledge graph enhancement processing on the standardized text data to generate a visual query result. According to the system, efficient query and visual display of Chinese language and literature works are achieved through multi-modal processing, multi-dimensional indexing and knowledge graph technologies, and the problems that a traditional query mode is single in input, limited in retrieval dimension, insufficient in text processing and the like are effectively solved.

The invention relates to the technical field of multi-sensor identification, in particular to a coal gangue detection and classification method and system based on multi-sensor fusion discrimination, and the method comprises the following steps: obtaining visible light and infrared images, calculating brightness and intensity judgment feature conditions, executing edge detection to extract gray segments, and fusing textures and a thermal field to generate a vector set. And performing clustering analysis to finish classification judgment, and outputting a coal gangue detection classification result. According to the method, a precise trigger mechanism is established through brightness and thermal radiation double-feature screening, boundary recognition sensitivity is enhanced through gray abrupt change point division, salient region extraction capacity is enhanced through weighted fusion of texture energy and gray gradient, and a cross-modal consistency feature group is constructed through combination of two-dimensional vector construction and similarity screening. The recognition expression integrity is improved, static threshold classification is replaced by vector distribution clustering, accurate mapping and classification decision making of material attributes in a complex scene are achieved, and the stability and the recognition rate of a coal gangue detection result are guaranteed.

The invention provides an unmanned aerial vehicle route planning method for hydropower engineering, and relates to the technical field of hydropower engineering safety monitoring. The method comprises the steps of generating a three-dimensional terrain grid model corresponding to a target operation area, performing voxelization processing to obtain an initial voxel map, and determining a first candidate waypoint by combining the initial voxel map; based on the average normal vector distribution of the initial voxel map, segmenting and refining the voxel units to generate a fine voxel map, and determining second candidate waypoints in combination with the fine voxel map; performing collision detection on all candidate waypoints to obtain a target waypoint set; and generating an unmanned aerial vehicle operation route according to the target waypoint set. According to the scheme, the multi-level voxel map is constructed and optimized, candidate waypoint screening and space collision detection are combined, and unmanned aerial vehicle patrol monitoring route planning which adapts to complex terrains and is complete in coverage and high in feasibility is achieved at low calculation cost.

The invention relates to the technical field of power electronic device diagnosis, and discloses a reverse conducting IGBT intelligent power module fault automatic diagnosis method and system. The method comprises the following steps: acquiring multi-source monitoring data including a grid voltage waveform, a collector current waveform and a shell temperature change curve when the power module operates; then establishing a dynamic feature extraction model, performing time domain and frequency domain conjoint analysis on the multi-source monitoring data, and generating a feature parameter set; then constructing a fault feature space, and mapping the feature parameter set to a high-dimensional space to form a feature vector distribution diagram; carrying out regional division on the feature vector distribution map by adopting a self-adaptive clustering algorithm, and identifying an abnormal feature aggregation region; and finally, comparing the abnormal feature gathering area with a preset fault feature library through a mode matching engine, and outputting a fault type identification result. According to the method, multi-source data can be integrated to realize dynamic feature extraction and adaptive fault identification, and the real-time performance and accuracy of fault diagnosis are improved.

The invention provides an intelligent semiconductor chip and a thermal management method thereof, and the method comprises the steps: extracting the gradient features of the interlayer temperature of each semiconductor chip from the temperature information of each barrier layer of a three-dimensional stacked chip in a high-power-consumption module under a current working condition; constructing a temperature coupling field of the high-power-consumption module in the current working condition through each gradient feature and the heat dissipation feature of the packaging layer; a plurality of heat flux concentration areas of the high-power-consumption module in the current working condition are positioned based on the temperature coupling field and the specified sampling frequency, and thermal runaway risk points in the high-power-consumption module are screened out from all the heat flux concentration areas through peak characteristics in junction temperature information of all the heat flux concentration areas; and determining the vector distribution characteristics of the heat flux density in each thermal runaway risk point, and performing layered heat dissipation on the three-dimensional stacked chip in the high-power-consumption module through the vector distribution characteristics of the heat flux density in each thermal runaway risk point. Based on the scheme, thermal field reconstruction of dynamic thermal management in the three-dimensional stacked chip can be realized.

The invention relates to a three-dimensional reconstruction method and system for strong ambient light filtering and depolarization error correction, and the method comprises the steps: obtaining the reflection components of ambient polarized light and polarization structure light, respectively calculating the Kirth vector distribution of the reflection components of the ambient polarized light and the polarization structure light, and obtaining a polarization image of the ambient light and a reflection image of the polarization structure light, determining a polarization degree threshold according to the polarization image of the ambient light, and extracting the polarization degree of the projection area according to the structured light reflection polarization image; establishing an ambient light separation model based on a reflection depolarization principle of light, and generating an abnormal pixel mask according to a polarization degree threshold and the ambient light separation model; establishing an error traceability model based on the ambient light separation model, and performing component decomposition and error correction on the polarization structure light phase data of the abnormal pixel mask area; and reconstructing the three-dimensional point cloud by fusing the corrected phase data with the image captured by the polarization camera. According to the method disclosed by the invention, precise three-dimensional measurement work of the polarized structured light under strong ambient light interference can be realized.

A method for enhancing detection of morphed biometric modality data is provided that includes receiving, by an electronic device, biometric modality data of a person, extracting feature vectors from the biometric modality data, normalizing the feature vectors, encoding the normalized feature vectors into qubits, and expanding, using at least one quantum algorithm, the normalized feature vectors into a high-dimensional space. Moreover, the method includes generating a distribution from the high-dimensional space based on the qubits, calculating a deviation between the generated distribution and a corresponding record high-dimensionality feature vector distribution of the person, and comparing the calculated deviation against a threshold deviation value. In response to determining the deviation satisfies the threshold deviation value, the method determines the received biometric modality data was morphed.

The invention provides a hysteresis loss calculation method suitable for motor electromagnetic field finite element calculation, and relates to the technical field of motor simulation calculation. The method comprises the following steps: calculating flux density vector distribution of a motor core grid through a transient electromagnetic field finite element; calculating a fundamental wave period according to the vector distribution; selecting a flux density vector in a fundamental wave period, performing flux density ellipse fitting, and rotating an ellipse along an original point to enable a long axis of the ellipse to be parallel to a Y axis; translating the long axis of the ellipse along the X axis, and enabling the long axis to coincide with the Y axis; establishing a hysteresis loss calculation model; checking parameters through a hysteresis loss curve, and calculating to obtain Kx; and calculating the magnetic hysteresis loss of all grids, and summing to obtain the magnetic hysteresis loss of the whole motor iron core. By analyzing the flux density of each grid of the iron core and combining the pulsating magnetic hysteresis loss curve of the silicon steel sheet, a motor magnetic hysteresis loss algorithm model considering rotation and pulsating loss is obtained, and the precision of finite element calculation of the motor magnetic hysteresis loss is improved.

The invention relates to the technical field of image processing, and discloses a house quality detection method and system based on a live-action three-dimensional model. The method comprises the following steps: converting an OSGB format live-action three-dimensional building model into a body coordinate system and carrying out quality evaluation, carrying out semantic segmentation on a standardized model to obtain a component geographic entity set, establishing an independent local coordinate system of each component to form a single three-dimensional model set, generating a multi-view projection feature image set based on a normal vector distribution planning view angle, and carrying out multi-view projection on the single three-dimensional model set. And detecting defects by adopting a 3D-YOLO-BuildingDefect algorithm, and carrying out inverse transformation to obtain three-dimensional space defect distribution and a quality grade evaluation result. The technical problems that a traditional two-dimensional detection method is insufficient in spatial positioning precision, incomplete in detection coverage rate and insufficient in geometric information utilization are solved, and the spatial positioning accuracy of house quality detection and the comprehensiveness of defect recognition are improved.

The invention provides a multi-source data fusion analysis system for electronic commerce digitization insight, and relates to the technical field of data processing, and the system comprises a reference analysis unit used for selecting a reference analysis unit in a spatial relation structure, calculating fractal dimension features of feature vectors, carrying out multi-level recursive division on the analysis unit based on fractal dimension values, and carrying out multi-level data fusion analysis on the fractal dimension values; each level of division keeps a self-similarity characteristic, and feature partitions are obtained; the partition space distribution module is used for mapping the feature vector distribution to corresponding feature partitions and calculating partition space distribution weights based on vector density distribution and space position relations of the feature partitions; the value evaluation module is used for performing parameter adjustment on the deep learning network by using the partition space distribution weight to obtain a value evaluation result; and the value evaluation module is used for processing a value evaluation result through a multi-modal fusion algorithm, generating a personalized recommendation strategy based on a reinforcement learning framework, and calculating resources and an intelligent routing algorithm so as to complete a digital insight analysis process. The e-commerce data utilization efficiency and the decision-making effect can be improved.

The invention relates to the technical field of data quality evaluation, in particular to an automatic quality evaluation method for a high-quality data set in the cultural field, which comprises the following steps: extracting semantic fragments and vector information, performing clustering analysis on a coverage structure, screening redundant samples, labeling image-text differences, and outputting a consistency evaluation result. According to the method, high-precision recognition of semantic coverage and structure density can be realized through structured extraction of semantic information and vector distribution measurement, and the discriminability of redundancy detection is enhanced through linkage screening of vector coincidence frequency and similarity threshold. The systematicness of image-text consistency marking is improved through an image-text position difference and semantic span combined marking mechanism, and it is ensured that redundant and deviated intersection samples can be accurately recognized through a synchronous contrast mechanism between text screening content and image-text difference marking results. The intensive clustering of the cultural semantic content in the vector space and the quantitative analysis capability of the structure deviation point are synchronously improved, and the comprehensive dimension and logic closure of the evaluation result are optimized.

The invention discloses a power cable outer sheath fault positioning method, and belongs to the technical field of cable fault positioning. Aiming at the problems of poor adaptability, weak anti-interference capability and low positioning precision of the existing cable outer sheath fault positioning technology, the method comprises the following steps of: acquiring horizontal and vertical components of an electromagnetic signal by deploying an array sensor, and obtaining pure vector distribution through Fourier transform and wavelet denoising processing; judging coordinates of an abnormal point according to an amplitude threshold value and a phase ratio, and determining a fault cluster center by adopting K-means clustering; and fusing the environmental data, refining the position through iterative filtering, comparing and confirming a final fault point in combination with triangulation and historical data, and generating a vector distribution map and a diagnosis report. According to the invention, high-precision and anti-interference positioning and visual diagnosis of cable faults in a complex environment are realized.

The invention discloses a fossil pattern recognition and classification method and system based on image recognition, and relates to the technical field of hydraulic engineering, and the method comprises the steps: obtaining three-dimensional point cloud data of a fossil sample, projecting the three-dimensional point cloud data into a multi-view two-dimensional texture mapping graph, and obtaining a surface element distribution graph of the fossil sample; utilizing an improved GLCM gray level co-occurrence matrix algorithm to calculate the contrast ratio, the correlation and the entropy value of fossil textures, and calculating the curvature and the normal vector distribution of fossils in the three-dimensional point cloud data and geometric parameters of biological parts; analyzing the surface element distribution diagram through a PCA (principal component analysis) method, and extracting a significant element combination related to species; and mapping the feature data to a unified feature space based on a cross-modal attention mechanism, distributing contribution degrees of all modals through learnable weights, and outputting a fossil pattern classification result. Therefore, a large number of dinosaur fossil samples can be quickly and accurately classified.

The invention provides a data security sharing method under an AI platform, and belongs to the technical field of AI platforms, and the method comprises the steps: building a distributed architecture of a confusion center node and a plurality of confusion edge nodes in an AI platform management server, carrying out the knowledge vector extraction of to-be-shared data, and converting the to-be-shared data into a high-dimensional vector for representation; generating a plurality of ellipsoidal confusion bubbles based on vector distribution characteristics and constructing a multi-dimensional authorization strategy, adopting a game boundary model comprising an upper-layer game model and a lower-layer game model to perform separation processing on a crossed and overlapped region, establishing a data training reasoning mapping relation according to a game solving result, and calculating confusion parameters through a confusion intensity optimization function; and finally, real-time confusion protection in the data transmission process is executed, and data flow is continuously monitored. The technical problem of data leakage risk caused by mismatching of a confusion protection strategy and an actual security demand in the data sharing process under the AI platform is solved.

The invention discloses a laser etching adjusting method and system, and the method comprises the steps: obtaining model data of a machined part, and constructing a normal vector distribution diagram of the surface of the machined part, so as to divide a machining region of the machined part; in each processing area, the laser incident angle is adjusted according to the normal vector data of the processing area; according to the laser incident angle, the material attribute of the machined part and the machining depth requirement, the power parameter adjusting value of each machining area is calculated, and a preliminary power distribution scheme is obtained; and aiming at the laser incident angle and the power parameter of each processing area, generating a control instruction sequence and transmitting the control instruction sequence to the laser processing system. According to the method, the laser incident angle and power parameters can be dynamically adjusted.

A volume holographic grating, a method for formulating a grating vector distribution formulation thereof, and a method and apparatus for manufacturing the volume holographic grating. The grating vector distribution is estimated by sampling a position of an eye movement range effective region of a plurality of sub-fields of view and a grating vector, matching with Bragg diffraction conditions, of a ray thereof, without considering the diffraction of ineffective rays that ultimately enter beyond the eye movement range, such that most of the effective rays with different incident angles have higher diffraction efficiency on the volume holographic grating, which ensures the overall brightness uniformity of the output beams and extend the field of view.

The invention relates to a vector retrieval method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: receiving a query vector; searching according to the query vector and the global vector index of the vector index based on the search parameter to obtain an initial cluster; wherein the global vector index is determined based on a single-layer hierarchical navigable small-world graph; the search parameters are used for controlling the size of the initial cluster; under the condition that the vector distribution in the initial cluster set does not meet a preset condition, executing expansion search in the global vector index by taking the initial cluster set as a starting point to obtain each adjacent cluster; determining a target cluster set based on the initial cluster set and each adjacent cluster; the target cluster set is searched in a local vector index of the vector index, a target vector is obtained, and the local vector index is determined based on the inverted file plane vector index. By adopting the method, the vector recall precision can be improved.

The application relates to the technical field of power electronic device diagnosis, and discloses a fault automatic diagnosis method and system for an inverse-conducting IGBT intelligent power module. The method comprises the following steps: acquiring multi-source monitoring data of the power module during operation, wherein the multi-source monitoring data comprises a gate voltage waveform, a collector current waveform and a shell temperature change curve; then, a dynamic feature extraction model is established, time domain and frequency domain joint analysis is performed on the multi-source monitoring data, and a feature parameter set is generated; subsequently, a fault feature space is constructed, the feature parameter set is mapped to a high-dimensional space to form a feature vector distribution graph; then, a self-adaptive clustering algorithm is used to divide the feature vector distribution graph into regions, and an abnormal feature aggregation area is identified; finally, a pattern matching engine is used to compare the abnormal feature aggregation area with a preset fault feature library, and a fault type identification result is output. The method can integrate multi-source data to realize dynamic feature extraction and self-adaptive fault identification, and improves the real-time performance and accuracy of fault diagnosis.

The application relates to the field of engineering surveying and mapping technology, in particular to an engineering surveying and mapping marking method and system based on an AI robot, which comprises the following steps: controlling a surveying and mapping robot to move to a to-be-marked engineering area to collect environment point clouds and environment images, constructing a three-dimensional environment map based on the environment point clouds and the environment images, identifying the geometric features and spatial distribution of key surveying and mapping control points based on the three-dimensional environment map, planning a candidate position of a marking point, and generating a preliminary coordinate sequence of the marking point. The application introduces normal vector distribution analysis and a depth compensation mechanism, before surveying and mapping marking spraying, the robot calculates the local flatness and roughness of a base surface through line laser scanning, automatically fits an inclined plane and calculates a compensation depth, simultaneously adjusts the pitch angle of a mechanical arm, so that an inkjet head is always perpendicular to the base surface and is within a depth of field range, which ensures that no matter how the ground fluctuates, the marking point can keep a standard circle and clear boundaries, and solves the problem of difficult marking on uneven ground.

The invention discloses a risk management and control zoning method and system based on multi-medium interactive pollution data, and relates to pollution management and control, and the method comprises the steps: delimiting an initial risk pollution region through random pollution point detection, and determining main pollutants of each medium and the content of the main pollutants; analyzing the content change gradient of the pollution point, and constructing a content reduction vector line; setting a multidirectional edge region, analyzing vector line difference, and if the criterion does not reach the standard, expanding the edge until the criterion is met or crossed; accurately measuring an actual pollution range, extracting geographic features, and establishing a correlation group of pollutants, vector distribution and the geographic features; according to the method, vector distribution, geographic features and pollutants are taken as elements, a multi-medium interactive pollution comparison model is constructed, and real-time data are analyzed, so that the accuracy, dynamic adaptability and real-time response efficiency of multi-medium pollution risk management and control are improved, and efficient partitioning and early warning in a complex terrain scene are supported.

The invention discloses an intelligent food detection method and system based on image recognition, a terminal and a medium, particularly relates to the technical field of food surface detection, and is used for solving the problem that in the prior art, due to the frequency domain aliasing effect, food surface microtextures and real defects are difficult to distinguish. The method comprises the following steps: acquiring a food surface image and generating a frequency domain phase distribution diagram, and extracting a phase consistency feature and a spatial domain texture directivity feature; reconstructing a frequency domain response function based on the phase-texture dynamic correlation weight, and generating an optimized frequency domain graph; detecting a mutually exclusive conflict region in combination with the frequency domain singular vector distribution entropy and the phase coherence attenuation, and obtaining a de-aliasing image through spatial domain inverse transformation; defect identification and judgment are realized based on self-adaptive morphological parameters and multi-feature classification; through cross-domain feature coupling analysis and dynamic parameter adjustment, frequency domain aliasing interference under a complex texture background is suppressed, and the sensitivity and misjudgment resistance of defect detection are remarkably improved.

The invention discloses an NPC type converter common-mode voltage suppression method with active midpoint potential control, which adopts an improved virtual space vector pulse width modulation method to redefine a small vector and a middle vector, so that the maximum value of common-mode voltage generated by an obtained virtual voltage vector is Vdc / 6, and the maximum value of the common-mode voltage generated by the virtual voltage vector is Vdc / 6. Compared with a traditional virtual vector pulse width modulation method, the common-mode voltage value of Vdc / 3 is reduced by half, and therefore the suppression effect on the common-mode voltage is achieved. On the basis, the value of a virtual vector distribution coefficient k is reasonably selected, the problem of neutral-point potential imbalance caused by accumulation of neutral-point potential offset due to the initial voltage difference of upper and lower capacitors of the direct-current bus and the active current component effect is solved, the fluctuation of the neutral-point voltage is actively inhibited, and the neutral-point voltage balancing effect is improved. And the correctness and the effectiveness of the method are verified through an MATLAB / Simulink simulation test.

The invention provides a motor control method and device, a vehicle and a storage medium, and the method is applied to the field of vehicles, and comprises the steps: obtaining first to third phase current values of a motor when the motor is in a locked-rotor state, and determining a zero vector adjustment value of the motor according to the first to third phase current values based on a pre-constructed zero vector adjustment distribution table, and adding the zero vector adjustment value to the current three-phase duty ratio to obtain a final three-phase duty ratio of the motor, and then determining a control signal of each phase of bridge arm of the motor based on the final three-phase duty ratio so as to control the motor according to the control signal of each phase of bridge arm. The method can dynamically change the current distribution of each phase of bridge arm of the motor, optimize the loss distribution of a power device, improve the torque output of the system and further improve the locked-rotor capability of the motor by dynamically adjusting the zero vector distribution strategy.

The invention relates to the technical field of motor driving, and discloses a motor driving direction adjusting method and system based on an attitude sensor, and the method comprises the steps: carrying out the motion state analysis of a target vehicle, and obtaining a target operation mode; calculating a zero vector distribution proportion and a vector switching sequence according to the target operation mode; generating a first motor torque instruction and a second motor torque instruction according to the attitude angle data of the target vehicle; and when it is detected that the target operation mode is switched, smooth transition processing is carried out on the zero vector distribution proportion, the vector switching sequence, the first motor torque instruction and the second motor torque instruction, and a PWM control signal is output. The control intention of a rider can be accurately predicted, mode misjudgment and frequent switching are effectively avoided, reliability and continuity of operation mode recognition are ensured, parameter sudden change impact during mode switching is effectively eliminated, and vehicle operation stability and riding comfort are improved.

This invention relates to the field of large-model latent space mapping technology and discloses a cross-scene hazard migration identification method based on large-model latent space mapping. The method collects historical hazard data and normal operation data from multiple scenarios, and obtains a standardized scene dataset after format unification and noise filtering. Then, a large-model latent space mapping module is constructed to map hazard features and operation features from the standardized data to the latent space of a pre-trained large model, generating scene feature latent vectors. Based on these latent vectors, a cross-scene feature alignment model is constructed, obtaining a unified feature representation across scenarios by aligning the differences in latent vector distributions across different scenarios. Next, the hazard migration identification model is trained, using the unified cross-scene feature representation and real-time data from the target scenario to predict the probability of hazard presence. Finally, the scenario-level hazard identification threshold is calculated by combining the probability distribution of historical hazard data from the target scenario, and the feature alignment model parameters are updated periodically.

The application provides a water level regulation method and system for a water conservancy hub. The method includes the following steps: collecting data by underwater robots and satellite remote sensing data simultaneously, constructing a sediment deposition space-time atlas containing a deposition gradient, and periodically correcting the space-time atlas. A radar array is deployed in the curved section of the spillway, and a multi-frequency pulse signal is used to analyze the three-dimensional sediment flow velocity field vector distribution. Through the correlation analysis of the dynamic coupling of the deposition gradient and the flow velocity field, the correlation area of the flow velocity mutation and the sediment deposition is accurately identified, and a constraint condition set containing coordinate positioning and flow-silt matching rules is established. Combined with real-time water level monitoring data, the constraint condition set is used to calculate and generate a dynamic adjustment parameter sequence of the spillway gate opening, and the water level is accurately regulated. The technical scheme provided by the application can accurately regulate the water level of the water conservancy hub by intelligently optimizing the spillway parameters, effectively alleviate the sediment deposition problem in the high-sediment water area, and ensure the safe operation and flood control efficiency of the hub.

The application relates to a speech synthesis model training method, a speech synthesis method, equipment and products. The speech synthesis model training method comprises the following steps: obtaining training speech spectrum information corresponding to a training speech sample of a training speech object; inputting the training speech spectrum information into a first coding module in a speech synthesis model to be trained, obtaining a first coding vector corresponding to the training speech spectrum information through the first coding module, and determining training vector distribution parameters corresponding to the first coding vector; obtaining a training object timbre feature corresponding to the training speech object based on the training vector distribution parameters, obtaining first synthesized speech according to the training object timbre feature and training text information corresponding to the training speech sample; obtaining a first model loss value according to the difference between the first synthesized speech and the training speech sample; and adjusting model parameters according to the first model loss value to obtain a pre-trained speech synthesis model, which can effectively improve the similarity between the timbre of the synthesized speech and the timbre of the speaker.