20results about How to "Reduce dependence on experience" patented technology

The invention provides a method for determining a mixing ratio, a storage medium and an electronic device, and the method comprises the steps: determining particle attribute parameters of a target material, and obtaining the particle attribute parameters of the target material under the condition of obtaining simulation data corresponding to mixing equipment; determining a first contact parameter of the target material and the mixing equipment based on an equipment intrinsic parameter and a particle attribute parameter in the simulation data, and determining a target mechanical basic model of the target material matched with a motion attribute under the condition that the motion attribute of the mixing equipment input by the target object is received, inputting the particle attribute parameter, the first contact parameter and the target mechanical basic model into a preset engineering application, and setting a particle generation strategy corresponding to the mixing equipment based on the mixing demand of the target object in the preset engineering application to obtain a to-be-operated simulation engineering application file, and operating the to-be-operated simulation engineering application file. And generating the target ratio parameter of each sub-strategy in the particle generation strategy. The problem that the coating is not uniform due to inappropriate ratio in the related technology is solved.

The invention relates to the technical field of power dispatching optimization, in particular to a multi-source power real-time dispatching optimization method and system based on reinforcement learning. The method comprises the following steps: acquiring required load information data of power generation equipment in different time periods; constructing an optimal scheduling strategy model of hybrid wind-fire distributed power generation; adopting a Monte Carlo method and a double-Q learning algorithm to optimize the optimal scheduling strategy model; and performing multi-source power real-time scheduling optimization by using the optimal scheduling strategy model. The multi-source electric power dispatching method is optimized in real time through the reinforcement learning method, the optimal electric power dispatching strategy is searched, compared with a traditional method, the electric power cost is reduced by 15% or above, meanwhile, automatic dispatching can be achieved, and manual intervention and experience dependence are reduced.

The invention discloses an intelligent assessment method and system for regional geological disaster risk, and relates to the technical field of intelligent assessment. During operation of the system, various real-time data related to debris flow are acquired by integrating real-time sensors and satellite remote sensing data, original data are preprocessed and stored in a data storage system; based on the collected data, a physical model and a machine learning model are constructed, a debris flow disaster risk index MRI, a rainfall intensity coefficient RIC, a geological vulnerability coefficient GVC and a hydrodynamic pressure coefficient HPC are obtained after calculation, the probability of debris flow occurrence is predicted, and a result is output according to the model. The debris flow disaster risk index MRI is compared with a preset first safety threshold value M and a preset second safety threshold value N, the risk level of debris flow occurrence is evaluated, different areas are managed in a classified mode, debris flow early warning is given out according to real-time monitoring data and risk evaluation results, and automatic emergency response measures are supported.

The invention discloses an NAS hyperspectral image crop classification method combined with noise inspiration and attention mechanism improvement, and relates to the field of remote sensing image classification. The method sequentially comprises the following steps of: a, acquiring a hyperspectral image data set, and artificially injecting random noise into image data; step b, constructing a neural network architecture search model for hyperspectral image crop classification, integrating a channel attention mechanism and a position scale perception mechanism so as to improve the calculation efficiency and the adaptability of the model, and enhancing the processing and anti-noise performance of difficult samples by using a focus loss function; c, searching a framework and performing framework evaluation; step d, performing model training by using the finally output architecture; and e, testing by using the complete test set, and outputting a classification result. According to the invention, by introducing the attention mechanism, the position and scale perception mechanism and the focus loss function based on the channel, the processing capability of the hyperspectral crop image mixed with noise is obviously enhanced, so that the classification precision and the robustness of the model are obviously improved.

The invention relates to an impression for repairing a complete denture, the complete denture and a preparation method. The impression terminator capable of being used for controlling the thickness of the impression material is designed by adopting computer modeling software, and the thickness of the impression material is limited by designing a bulge structure which is matched with the shape of an intraoral mucosa and has a certain thickness on an individual tray tissue surface, so that the intraoral impression of a patient has uniform thickness; the preparation precision of the edentulous jaw final impression is ensured; meanwhile, the individual tray and the diagnosis denture are designed into an integrated structure, so that a doctor can be assisted in acquiring the occlusion relation of a patient while the edentulous jaw final impression is acquired, the two steps of preparing the final impression and recording the jaw position relation in the traditional process are combined into one step, the clinical process is simplified, and the working efficiency is improved. And the experience dependence and the technical sensitivity of the operation process are reduced, so that the accuracy of the two key steps in the edentulous jaw repair process is improved.

The invention belongs to the technical field of data processing, and particularly relates to an energy storage transaction data processing method and model based on an electric power spot market environment, and the method comprises the steps: analyzing the historical energy storage state of an electric power spot market platform to obtain a historical energy storage state feature value, and analyzing the historical transaction state of the electric power spot market platform to obtain a historical energy storage state feature value; obtaining a historical transaction state characteristic value, combining a real-time meteorological state data set, performing comparison to obtain a predicted daily charging frequency of an energy storage battery, a predicted power transmission line power factor and a predicted power grid load capacity, and performing comparison based on a scheduling optimization characteristic value to obtain an electric power spot market platform scheduling optimization scheme; the system comprises an energy storage state characteristic value analysis module, a transaction state characteristic value analysis module, a meteorological state data set acquisition module, a prediction data analysis module and a scheduling optimization scheme comparison module. The method can improve the capability of judging the future power demand and supply condition, helps to balance the power grid load, and enhances the flexibility and stability of the power system.

The invention belongs to the technical field of coal mine safety monitoring, and discloses a coal mine tunnel roof disaster real-time early warning system and method. Comprising the following steps: performing time sequence reasoning on monitoring data acquired in real time, respectively judging whether the key physical quantity of the roof triggers primary early warning or secondary early warning by acquiring an expected value prediction curve and an uncertainty boundary curve of the key physical quantity of the roof in a future time step, and acquiring corresponding early warning information when the early warning is triggered; constructing a dynamic causal graph, inputting the early warning information and the currently acquired key physical quantity into the dynamic causal graph for reasoning, calling candidate intervention measures based on the acquired current state node to perform anti-fact reasoning in the dynamic causal graph, and acquiring a risk evolution trajectory and a risk suppression efficiency index of the candidate intervention measures; and the risk suppression efficiency indexes are sorted and preferentially selected, and a causal decision early warning report containing the optimal intervention scheme is output, so that the scientificity and effectiveness of risk suppression are improved.

The invention provides a turbine blade cooling structure optimization method and device based on reinforcement learning, and the method comprises the steps: responding to an input preset design condition, obtaining an initial cooling structure of a turbine blade based on the design condition, responding to the initialization of a preset intelligent agent, and carrying out the optimization of the cooling structure of the turbine blade under a preset reinforcement learning environment framework, the intelligent agent is trained based on a preset reinforcement learning algorithm to obtain a trained intelligent agent, the cooling parameters of the cooling structure are adjusted and optimized based on the trained intelligent agent to obtain an adjusted and optimized cooling parameter, and the adjusted and optimized cooling structure is determined based on the adjusted and optimized cooling parameter. According to the method and the device, a more complex or complete cooling structure can be applied, the cooling parameters of the cooling structure can be automatically adjusted, optimized and determined, and the experience dependence on designers and the time cost in the cooling parameter design process are reduced.

The application relates to the technical field of fault diagnosis, in particular to an information system fault tracing method based on log event clustering and time sequence relationship, which comprises the following steps: extracting a log information set; classifying event description information to obtain a plurality of event category clusters; counting the occurrence frequencies of log events corresponding to each event category cluster before a fault event occurs, and screening abnormal event category clusters; determining adjacent cluster pairs and calculating time sequence correlation strength; constructing a directed correlation graph, locating a target node, traversing reversely to a starting node with an in-degree of zero, and determining a fault root cause. The application has the effect that event semantic correlation patterns can be automatically extracted from massive heterogeneous logs, a tracing model reflecting a real fault propagation path is constructed in combination with time sequence characteristics, a fault root cause is accurately located, fault diagnosis efficiency and accuracy are improved, and manual intervention and experience dependence are reduced.

The present application belongs to the technical field of probe processing, and particularly relates to a novel welding method for a MEMS probe. The novel welding method for the MEMS probe constructs and initializes a digital twin model of core components of the MEMS probe welding, combines a real-time collected synchronous three-dimensional data set and a lightweight agent model, and accurately outputs thermal stress distribution data, so as to early understand the thermal stress change in the welding process, effectively control potential risks, and when the risk condition is determined, the potential root cause can be diagnosed in-depth and the final adjustment suggestion can be generated. The accurate dynamic regulation mechanism enables the welding process to be flexibly optimized according to the real-time working condition, ensures the stable and reliable welding quality, avoids the welding defects caused by the out-of-control thermal stress, greatly improves the yield of the MEMS probe welding, reduces the production cost, and improves the production efficiency.

The invention discloses a plasma polishing machine which comprises a vacuum cavity, a bracket, an angle adjusting mechanism, a multi-axis movement mechanism, a data processing unit and a numerical control system. The bracket is positioned in the vacuum cavity; the angle adjusting mechanism is used for adjusting the inclination angle of the to-be-polished workpiece. The multi-axis movement mechanism is arranged on the side, opposite to the bracket, in the vacuum cavity and carries an ion source and a measuring probe, and the measuring probe moves to multiple positions of the surface of the to-be-polished workpiece through the multi-axis movement mechanism so as to obtain coordinate values of the surface of the to-be-polished workpiece at the multiple different positions in the thickness direction of the to-be-polished workpiece; the data processing unit analyzes the coordinate values and calculates the inclination angle of the workpiece to be polished; the numerical control system controls the angle adjusting mechanism according to the inclination angle provided by the data processing unit so as to adjust the inclination angle of the workpiece to be polished, dependence on professional skills and experience of operators is reduced, the intensity of manual operation and the occurrence rate of human errors are reduced, and the machining efficiency and precision are effectively improved.

The invention discloses a portable pipeline blockage point rapid positioning instrument and method based on pressure transient analysis, the positioning instrument comprises a positioning instrument host, a butt joint of a three-way structure, a high-speed electromagnetic valve and a high-frequency pressure sensor, the high-speed electromagnetic valve and the high-frequency pressure sensor are integrated in the butt joint; and signal and power supply connection with the host is realized through the quick connector. The positioning method comprises the following steps: connecting equipment and recovering a signal; the host controls the high-speed solenoid valve to instantly open and close to generate pressure pulse; the high-frequency pressure sensor collects incident and blocking point reflection pulse signals; the host preprocesses the data through wavelet transform, cross-correlation analysis and other algorithms and calculates a pulse time difference delta t; and calculating and displaying the distance of the blocking point according to a formula S = (wave velocity a * delta t) / 2, wherein the wave velocity can be calibrated through a self-adaptive algorithm. The positioning precision reaches the meter level, equipment is portable, operation is easy and convenient, point distribution and excavation in advance are not needed, the method is suitable for multi-material pipelines, and the detection cost and the technical threshold are reduced.

This invention relates to the field of metal cutting technology and discloses a steel coil slitting device integrating blade deep processing. The device includes a feeding machine tool with stabilizing side plates fixed at both ends. A drive assembly is mounted on the feeding machine tool, comprising two support frames fixed to the stabilizing side plates at both ends of the machine tool. A first drive mechanism and a second drive mechanism are mounted on the support frames. The first drive mechanism includes a rotating rod rotatably connected between the two support frames. By integrating blade maintenance functions onto the slitting machine tool and directly utilizing the cutting kinetic energy of the CNC blade rotation to drive the grinding and dust removal mechanisms, constant-pressure fine grinding and surface cleaning of the CNC blade sides can be simultaneously completed while the metal cutting and slitting operation continues. This achieves true online blade deep processing, avoiding the drawbacks of traditional methods that require stopping the machine to disassemble the blades, and greatly improving the overall operating efficiency and productivity of the equipment.

The invention relates to the field of image processing, and provides an image control point processing method and electronic equipment, and the method comprises the steps: obtaining at least one to-be-detected control point in a to-be-detected image, setting a plurality of epipolar lines corresponding to the to-be-detected control point according to the relative relation between a plurality of reference images and the to-be-detected image, the to-be-inspected control points are automatically inspected as the stable control points or the unstable control points by means of the multiple epipolar lines, manual intervention is not needed, experience dependence and subjective deviation of workers are remarkably reduced, and the efficiency, accuracy and reliability of processing the image control points can be considered easily.

The utility model relates to the technical field of medical instruments, in particular to a wound infected tissue sampling device which comprises an inner cylinder, a screw cap is fixedly arranged at the top of the inner cylinder, rotatable outer cylinders are arranged on the outer side of the inner cylinder at intervals, a through hole is formed in the screw cap in a penetrating mode, and a supporting connecting rod capable of moving up and down is arranged in the through hole in a penetrating mode. The lower end of the supporting connecting rod is fixedly connected with a cutting tool bit of a conical structure. According to the utility model, the cutting tool bit can be driven to move up and down through the arranged knob and the supporting connecting rod, so that the sampling depth can be accurately controlled, the dependence on the proficiency and experience of an operator is reduced, the sampling accuracy is improved, unnecessary wounds are reduced, point-to-surface cutting sampling is realized through the cutting tool bit, and the sampling efficiency is improved. The device is simple in structure and convenient to use, can cut into the hardened skin surface, reduces the pressure of an operator, avoids aggravating the pain of a patient, is more suitable for different parts and different types of skin by adopting a girdling mode, and improves the sampling convenience.

This invention proposes a reconfigurable metasurface beamforming design method based on CGAN and Gumbel-Sinkhorn. This method constructs a conditional generative adversarial network (GAN) as the inverse design network. The generator takes the target beam pattern as a conditional input to generate the corresponding encoding matrix. The Gumbel-Sinkhorn operator is introduced to discretize this continuous matrix, transforming it into an approximately doubly random permutation matrix. The encoding is sorted and selected by multiplying this permutation matrix with a predefined structure vector (SV), outputting a discrete 0 / 1 metasurface encoding array. A noisy forward prediction network is trained to predict the beam pattern corresponding to a given encoding array. In the training of the inverse design network, the discretized encoding matrix output by the generator is input into the noisy-trained forward prediction network. The mean square error between the predicted beam pattern and the target beam pattern is calculated as the target loss term to optimize the network parameters and accurately learn the complex mapping relationship from the beam pattern to the encoding array.

The invention relates to an intelligent alignment compensation method for assembly of a liquid crystal panel and a backlight module, and aims to solve the problems of dynamic pose deviation and weight mismatch caused by material batch difference, nonlinear deformation and environmental factors in a high-precision fitting process. According to the core technology, an industrial camera array and a multi-source sensor are used for collecting multi-dimensional micro texture and environment data, a fingerprint spectrum of intrinsic deformation characteristics of a material is established in a fusion mode, spectrum matching and weight self-adaptive calibration are carried out in combination with real-time working conditions and visual feedback, and six-degree-of-freedom micro dynamic correction is achieved through FPGA acceleration driving control. According to the scheme, the sub-pixel-level precision in the attaching process of the liquid crystal panel and the backlight module is effectively improved, the nonlinear deformation trend of materials is restrained, the control instruction response speed is optimized, continuous and accurate calibration of self-adaptive compensation weights among batches is guaranteed, and the process consistency and the automation level are improved.

The invention discloses a failure analysis method of a storage structure, a failure analysis device of the storage structure, a computer readable storage medium and a terminal. The method comprises the following steps: acquiring a test structure data graph; comparing the plurality of test structure data graphs with the yield data graph of the wafer to be analyzed one by one, and judging whether the test structures corresponding to the test structure data graphs are abnormal or not; when the test structure corresponding to the test structure data graph is judged to be abnormal, obtaining wafer test data of a hierarchical structure in the test structure; obtaining a hierarchical structure data graph; comparing the hierarchical structure data graph with the yield data graph one by one; and when the hierarchical structure corresponding to the hierarchical structure data graph is abnormal, acquiring defect scanning data related to the abnormal hierarchical structure so as to analyze the failure reason of the failure position in the yield data graph. According to the method, the degree of dependence on experience of analysts is reduced, the burdensome labor of data processing is reduced, and the failure position of the storage structure is positioned more quickly.

The present invention discloses a plasma polishing machine, comprising a vacuum chamber, a bracket, an angle adjustment mechanism, a multi-axis motion mechanism, a data processing unit and a numerical control system. The bracket is located in the vacuum chamber; the angle adjustment mechanism is used to adjust the inclination angle of the workpiece to be polished; the multi-axis motion mechanism is arranged on a side opposite to the bracket in the vacuum chamber, and is equipped with an ion source and a measuring probe, and the measuring probe is moved to multiple positions on the surface of the workpiece to be polished through the multi-axis motion mechanism to obtain the coordinate values ​​of the surface of the workpiece to be polished at multiple different positions in the thickness direction; the data processing unit analyzes the coordinate values ​​to calculate the inclination angle of the workpiece to be polished; the numerical control system controls the angle adjustment mechanism according to the inclination angle provided by the data processing unit to adjust the inclination angle of the workpiece to be polished, thereby reducing the dependence on the professional skills and experience of the operator, reducing the intensity of manual operation and the incidence of human errors, and effectively improving the processing efficiency and precision.

The application discloses a method for extracting and storing along-line building space data based on traffic digital twinning, which comprises the following steps: obtaining along-line engineering CAD data and traffic digital twinning base map data, and generating a joint point set input structure; calling a coherent point drift algorithm to establish a continuous deformation mapping relationship, and generating an initial deformation field; introducing a Bayesian parameter estimation mechanism to generate a Bayesian deformation field; performing segmented deformation inference and cross-segment continuous consistency processing on the joint point set input structure to generate an along-line consistent deformation field and a segmented uncertainty distribution; performing discretization inversion processing and substantial re-projection processing to generate corrected building geometric data; generating building attribute data based on the along-line consistent deformation field; and completing data storage based on the corrected building geometric data and the building attribute data. The application realizes high-precision alignment, stable correction and automatic storage of along-line building space data.