27results about How to "Accurate performance" patented technology

The invention discloses a physical sign data monitoring model construction method and device, a medium and equipment, and relates to the technical field of health monitoring. The method comprises the steps of performing feature extraction of physiological sign data by adopting a fractional order reverse scatter entropy method, replacing a Sigmoid activation function in a traditional extreme learning machine with a DiTAC activation function, modifying an L2 norm of an output layer weight coefficient in a loss function in the extreme learning machine into an L1 norm, establishing an optimization model based on the improved extreme learning machine, and performing feature extraction of the physiological sign data based on the optimized model. And solving the optimization model by using an Aribus algorithm and a forty-thief algorithm which introduce discrete fractional order Caputo, so as to obtain a physical sign data monitoring model. The extreme learning machine is constrained by the trainable parameters introduced by the DiTAC activation function and the L1 norm of the output layer weight coefficient in the training process, so that the trained extreme learning machine can be more suitable for physical sign data monitoring, and a more accurate monitoring result is obtained.

The invention discloses a control method, device and equipment for a current loop of a permanent magnet synchronous motor and a storage medium, and relates to the technical field of electric vehicle driving motors, and the control method for the current loop of the permanent magnet synchronous motor comprises the steps: obtaining an input voltage; a current loop is controlled according to a target motor mathematical model and the input voltage, the target motor mathematical model is obtained by adjusting an initial model, and adjustment is based on the required torque, a rectangular axis open-loop transfer function and a zero-pole mutual elimination function. The initial model is constructed according to a target quadrature-axis voltage equation, a dynamic inductance parameter value, a locked-rotor test and a difference strategy, and the dynamic inductance parameter value and the target quadrature-axis voltage equation are at least calculated by three items of an initial quadrature-axis voltage equation, dynamic inductance, a quadrature-axis flux linkage equation and a back electromotive force-flux linkage relation. According to the invention, the control precision, efficiency and response speed of the permanent magnet synchronous motor are improved, and the robustness of the system is enhanced.

The invention relates to a portrait generation and feature similarity recommendation method and system based on a large language model, and the system comprises a data collection module, a feature extraction module, a similarity calculation module and a recommendation module, the data collection module collects multi-modal data from a social media platform, and covers audio, video, text and interactive data information; the feature extraction module performs deep feature analysis and processing based on the collected multi-modal data, and extracts features in multiple dimensions; the similarity calculation module generates a training sample based on the feature matrix generated by the multi-dimensional feature extraction module in combination with a large language model, constructs a similarity matrix of audio, video, text and interactive data features, and calculates similarity scores to determine candidate objects; the recommendation module provides personalized recommendation based on the comprehensive similarity score generated by the similarity determination module. Through combination of a large language model and multi-modal features, the system can accurately analyze behaviors and expressions and generate personalized recommendation.

The invention discloses a case data analysis system based on a large model, and the system comprises a data collection module, a preprocessing module, a large model training and optimization module, an analysis module, a visualization and interaction module, a security and privacy protection module, a data preprocessing module with an innovative algorithm, a large model training module with an adaptive learning rate formula, and a disease risk prediction module with an evaluation formula. System functions and performance indexes are clear, equipment and medium requirements are clear, the steps of the method are complete, and high efficiency, accuracy and safety of case data analysis are guaranteed; the case data processing efficiency and accuracy are improved, and powerful support is provided for medical decision making. A large model is accurately analyzed, visual and convenient to check, and a security module guarantees data privacy. The medical quality can be improved, the treatment scheme can be optimized, and the method has important application value.

The invention discloses a method for testing and evaluating the thermal comfort of an automobile using an air-conditioning dummy in an extreme environment. In the testing method, the air-conditioning dummy and a thermocouple installed on a seat are used for testing; thermocouples are respectively arranged on the left and right sides of the hip position of the seat, and thermocouples are respectively arranged on the left and right sides of the backrest 15cm and 30cm away from the seat; the testing method comprises the steps of refrigeration, heating and extreme sudden change environment testing. The evaluation method is applied to the test method and comprises test effectiveness judgment and scoring rules. If the difference between the subjective evaluation score of the driver and the final score of the air conditioner dummy is less than 30%, the test is considered to be successful, and if the overall scores of the driver and the air conditioner dummy exceed 30%, the test needs to be carried out again; according to the scoring rule, the ratio of the air conditioner dummy to the thermocouple to the subjective evaluation is 5: 2: 3. According to the method for testing and evaluating the intelligent air conditioner, the intelligent air conditioner can be tested and evaluated more efficiently, accurately and safely through the air conditioner dummy in the extreme environment.

The invention provides a method for testing single parameters of a gas compressor. The method comprises the following steps: determining the single parameters of the gas compressor to be tested; multiple adjustment tests of single parameters are carried out, the total pressure ratio of the gas compressor is kept stable through the gas compressor isobaric ratio automatic control method, and the gas compressor isobaric ratio automatic control method comprises the following steps that a target pressure ratio set value is determined; the compressor inlet total pressure and the compressor outlet total pressure are collected in real time, and a pressure ratio feedback value is calculated; calculating the deviation between the pressure ratio feedback value and the target pressure ratio set value; calculating a control output value by taking the deviation as an input quantity by using a PID algorithm; and controlling the opening degree of the exhaust valve according to the target pressure ratio set value, the pressure ratio feedback value, the deviation and the control output value. According to the method, for the single parameter test of the engine, it is guaranteed that the environment variable, namely the pressure ratio, is not changed in the one-way parameter test process, the influence of the variable, namely the pressure ratio, is eliminated, and more accurate test data are obtained.

The invention relates to the technical field of wind speed prediction and artificial intelligence, and discloses a wind speed prediction method and system based on spatial-temporal correlation. Acquiring wind speed data and wind direction data at the plurality of wind speed measuring points, and acquiring geographical location information of the plurality of wind speed measuring points; converting the plurality of wind direction data into angle functions and fusing the angle functions to obtain a plurality of wind direction features; inputting the wind speed data into the self-correlation weight matrix in a time sequence form, and obtaining time characteristics of the wind speed data; the method comprises the following steps: dynamically adjusting weights among nodes in a graph structure through a graph attention mechanism according to geographic position information and wind direction features of a plurality of wind speed measuring points and time features of wind speed data, extracting spatial features, and obtaining a wind speed prediction result based on the wind direction features, the time features of the wind speed data and the spatial features. The problem that an existing wind speed prediction model is limited in performance and cannot effectively predict the wind speed is solved.

A sample observation device 1 includes an irradiation optical system 3 that emits planar light L2A and L2B having different wavelengths so as to pass through cross sections at different positions of the sample S, a scanning unit 4 that scans the sample S with respect to the irradiation surfaces Va and Vb, an imaging optical system 5 that has an observation axis P2 inclined with respect to the irradiation surfaces Va and Vb and images each of light images of observation light L3A and L3B, an image acquiring unit 6 that acquires each of first image data and second image data during scanning of the sample S by the scanning unit 4, and an image generating unit 8 that generates observation image data 24A and 24B of the sample S on the basis of the first image data and the second image data.

The present invention discloses an automated testing method, device and system for motor controllers. The present invention numbers the test torque output instructions of the current batch of motor controllers, and automatically collects and analyzes the temperature and output current of preliminary qualified products under different torque output instructions, thereby reducing human operation errors and improving the consistency and accuracy of the test. Based on the temperature and output current performance of the preliminary qualified products under different torque output instructions, a comprehensive evaluation is performed to obtain the performance evaluation values ​​of the preliminary qualified products under different torque output instructions, thereby accurately evaluating the performance and stability of the preliminary qualified products under different working conditions, solving the problem in the prior art that a single testing process cannot fully cover the performance of the motor controller under various actual working conditions, resulting in the inability to accurately evaluate its performance and stability under different working conditions.

Achieved is a determination method which enables accurate determination as to a bending angle. A bending angle determination method includes: an input reception step of receiving, with respect to a joint at which a first pipe and a second pipe are joined together, inputs of respective parameters of a plurality of predetermined positions set in a circumferential direction of the first pipe; and a bending angle determination step of determining whether a bending angle at the joint is appropriate, on the basis of the parameters that have been inputted in the input reception step.

The invention discloses an electronic parking device comprehensive testing platform, which relates to the technical field of electronic parking device testing, and comprises: a testing platform, a fixture group, and also comprises: a position deviation judgment component and a testing accuracy maintenance component, wherein the position deviation judgment component is located above the testing accuracy maintenance component; through the design of the arc-shaped groove of the 8-shaped groove, when gas is indirectly stored for the testing accuracy maintenance component under the cooperation of the sliding block and the 8-shaped groove, since the stored gas is gradually increasing, its gas pressure is also gradually increasing. In this process, each time the 8-shaped groove and the sliding block involved in gas storage move with the air pushing plate with the connecting rod, the reaction force on the groove wall of the 8-shaped groove will gradually increase, but due to its arc-shaped design, the force can be effectively dispersed, and the point of action of the force can be adjusted in real time at different positions of the groove wall of the 8-shaped groove during the rotation of the rotating disk. The above design can maintain a good performance state during long-term operation.

The present invention relates to the field of wind speed prediction and artificial intelligence technology, and discloses a wind speed prediction method and system based on spatiotemporal correlation. Wind speed data and wind direction data are collected at a number of wind speed measuring points, and geographical location information of a number of wind speed measuring points is obtained; a number of wind direction data are converted into angle functions and the angle functions are merged to obtain a number of wind direction features; the wind speed data is input into an autocorrelation weight matrix in a time series form to obtain the time characteristics of the wind speed data; according to the geographical location information of a number of wind speed measuring points, wind direction characteristics and the time characteristics of the wind speed data, and a graph structure is constructed, the weights between nodes in the graph structure are dynamically adjusted through a graph attention mechanism, spatial features are extracted, and then the wind speed prediction results are obtained based on the wind direction characteristics, the time characteristics of the wind speed data and the spatial characteristics. The problem that the existing wind speed prediction model is limited in performance and cannot effectively predict the wind speed is solved.

The invention relates to the technical field of data processing, in particular to a stability evaluation method based on a big language model, which comprises the following steps: S1, acquiring big language associated data; s2, calculating a stability coefficient according to the big language associated data, and obtaining an evaluation result according to the stability coefficient; s3, adjusting the calculation process of the stability coefficient according to the big language associated data, and correcting the adjustment process of the calculation of the stability coefficient according to the big language associated data; s4, judging the denoising capability according to the evaluation result and the big language associated data, calibrating the correction process of the adjustment of the calculation of the stability coefficient according to the judgment result, and compensating the judgment process of the denoising capability according to the big language associated data; and S5, updating the large language model according to an evaluation result. The method is used for evaluating the stability of the large language model and optimizing the vehicle-mounted large language model system according to the evaluation result.

The present application relates to metal matrix composites, and in particular to a method for designing and predicting the performance of metal matrix composites. The present application uses a finite element method based on the model composite physical properties and structural parameters, fully considers the influence of the interface matrix strengthening microzone on the performance of the composite material during the model establishment process, effectively reflects the influence of the real interface performance and debonding behavior on the service failure of the composite material, truly restores the internal situation of the composite material, realizes the performance-oriented design and performance prediction of the metal matrix composite, and has the characteristics of truly restoring the characteristics of the composite material and accurate performance prediction.

The present invention discloses a cross-network collaborative office management system and method based on big data, and relates to the field of collaborative office technology. The management method comprises the following steps: constructing a collaborative office management system to record each collaborative office process and generate a corresponding collaborative record; conducting a comprehensive performance evaluation on any collaborative record; arbitrarily selecting an office device, analyzing the affinity between the office device and other office devices, and obtaining the actual data processing performance of the office device in the relevant collaborative record; extracting characteristic factors that affect the office device in different network environments, and obtaining the degree of influence of various network environments on the performance of the office device; when generating a real-time collaborative record, based on the affinity between a number of office devices, performing performance prediction on each office device, confirming the network environment of any office device and recommending it to the user.

The present invention discloses a large-scale axle tube stepping mechanism, comprising a fixed frame, the bottom end of the fixed frame is provided with a plurality of evenly arranged supporting feet, the top of the fixed frame is symmetrically provided with two sets of guide plates, the guide plates are obliquely provided with guide grooves, four support rods are symmetrically provided at the four corners of the top of the fixed frame, the inner sides of the support rods are provided with slide rails, a vertical lifting frame is slidably connected between the slide rails, the left and right ends of the vertical lifting frame are provided with roller groups that are compatible with the slide rails, and the bottom end of the vertical lifting frame is symmetrically provided with vertical lifting frame connecting ribs. The internal mechanism of the present invention adopts a hydraulic system as the power, has a strong bearing capacity, stable mechanism performance, can meet the needs of large-scale axle tubes, has precise control, simple mechanism, easy operation, easy maintenance, and greatly improves the application range of axle tubes.

The invention discloses a Mini LED backlight function test method and system based on optical detection. Comprising the following steps of: controlling an industrial camera CMOS (Complementary Metal Oxide Semiconductor Transistor) to shoot a high-resolution gray-scale image of a Mini LED backlight lamp bead in an experimental environment; the method comprises the following steps: marking a shot and captured backlight bead target in an input grayscale image through a connected region marking algorithm, extracting feature information such as a target pixel position, an external rectangle size and an average grayscale, and obtaining average brightness feature information; according to the target set extracted by marking and the target feature information, combining abnormal judgment threshold processing judgment, and extracting a target with abnormal brightness; carrying out positioning output on the abnormal Mini LED lamp bead target by adopting a target self-adaptive positioning algorithm; and adopting a double-label detection correction algorithm based on a pixel space and a target object to detect the LED Open / Sport and carry out positioning output. According to the invention, rapid backlight lamp bead optical quality detection can be realized for Mini LED backlight plates of any large-size backlight lamp bead array, and the detection efficiency and the detection precision are improved.

A pixel according to an embodiment of the present invention includes: a first light source unit, including at least one first light-emitting element connected between a first separation electrode and a second power source; a second light source unit, including at least one second light-emitting element connected between a second separation electrode and the second power source; a driving current generator, including a first transistor, the first transistor being connected between the first power source and the first light source unit and the second light source unit, and generating a driving current corresponding to a first data signal supplied to a first data line; a first switching unit, including a first switching element connected between the driving current generator and the first light source unit; and a second switching unit, including a second switching element connected between the driving current generator and the second light source unit, and controlling the connection between the first transistor and the second light source unit in response to a second data signal supplied to a second data line.

The embodiment of the application belongs to the technical field of detection, and relates to a kit and a method for measuring cytotoxin-related protein A, comprising Ra reagent, Rb reagent, Rc reagent and a luminescent substrate; wherein the Ra reagent comprises streptavidin-modified magnetic microparticles; the Rb reagent comprises a biotin-labeled cytotoxin-related protein A antibody; and the Rc reagent comprises a chemiluminescent substance-labeled cytotoxin-related protein A antibody. The application has high sensitivity and accurate test results.

The invention discloses a fire-fighting equipment fire resistance performance test system and an operation distance calculation system and method, and the system comprises fire-fighting equipment which is used for achieving the circulation of a fire extinguishing medium; the fire simulation device is used for simulating a petrochemical fire; a heat radiation measuring device; the control device is used for adjusting the power of the fire simulation device to enable the fire-fighting equipment to be subjected to heat radiation at different levels and circularly controlling the fire-fighting equipment, and during testing, the critical heat radiation values and time of the fire-fighting equipment under different heat radiation levels are recorded by observing the gasification state of the fire extinguishing medium after circulation; the operation distance calculation device is used for determining the unit area release rate according to the fire scene area of the petrochemical fire, the heat radiation value of the key points around the fire scene and the distance between the fire-fighting equipment and the center of the fire scene so as to analyze the heat radiation distribution state around the fire scene, and based on this, the critical heat radiation value and time of the fire-fighting equipment are combined; and calculating to obtain the operation distance of the fire-fighting equipment.

The invention relates to the technical field of tunnel anti-seismic performance monitoring, in particular to a tunnel anti-seismic performance online monitoring system based on a tunnel supporting component, which comprises a monitoring assembly, a local processor, a data black box and a data center. The monitoring assembly is used for being arranged on a tunnel supporting component, the monitoring assembly is electrically connected with the local processor, and the monitoring assembly comprises at least one of a stress monitoring sensor, a deformation monitoring sensor, a displacement monitoring sensor and a vibration monitoring sensor. The local processor is electrically connected with the data black box, the local processor is used for copying original monitoring data of the monitoring assembly to obtain copied data, and the local processor is further used for storing the original monitoring data to the data black box and sending the copied data to the data center. The data black box only opens the incremental write permission of the local processor. The real-time on-line monitoring of the actual anti-seismic performance of the tunnel can be realized, the integrity and stability of data acquisition are guaranteed, and the data can be recovered at the same time.

The invention relates to the technical field of hydraulic engineering, and discloses a dam safety index prediction method and device, electronic equipment, a medium and a product, and the dam safety index prediction method comprises the steps: carrying out the noise separation of a safety index monitoring data sequence, and obtaining a plurality of modal components; performing correlation analysis on each modal component and the safety index monitoring data sequence, and screening the plurality of modal components according to a correlation analysis result to obtain a plurality of target modal components; reconstructing the plurality of target modal components to obtain a target safety index monitoring data sequence; the safety index of the target dam is predicted according to the environment monitoring data sequence and the target safety index monitoring data sequence, and a safety index predicted value is obtained. Noise separation and screening are carried out on the safety index monitoring data sequence, so that more accurate safety index monitoring data are obtained; safety index prediction is carried out according to the environment data and the safety index monitoring data, and the accuracy of safety index prediction is improved.

A method, system, and medium for evaluating the performance of conductors after a wildfire failure are disclosed. The method includes: inputting images and parameter information of the conductor to be evaluated into a pre-trained appearance evaluation model to perform a preliminary assessment of the damage level; if the preliminary assessment result indicates non-severe damage, performing a secondary assessment of the damage level based on the conductor's diameter and standard requirements; if the secondary assessment result indicates non-severe damage, conducting a tensile strength test to obtain the tensile strength, and performing a tertiary assessment of the damage level based on the tensile strength; if the tertiary assessment result indicates non-severe damage, obtaining the recrystallization status of the conductor's microstructure, and performing a quaternary assessment of the damage level based on the recrystallization status; and determining the damage grade of the conductor based on the assessment results of the preliminary, secondary, tertiary, and quaternary assessments. This application enables rapid, convenient, and accurate evaluation of conductor performance after a wildfire failure, while effectively guiding subsequent operation and maintenance work.

The invention belongs to the technical field of medical equipment, and particularly discloses a use quality evaluation method and system for a digital medical thermometer in a full life cycle, and the method comprises the steps: constructing a parameter database of the digital medical thermometer in the full life cycle, constructing and fusing three monitored object learning models, obtaining a comprehensive evaluation model, and carrying out the evaluation of the use quality of the digital medical thermometer in the full life cycle. Obtaining theoretical monitoring data and actual monitoring data of the digital medical thermometer; establishing a full-life-cycle management model of the digital medical thermometer, performing model training, and then performing correction to obtain a corrected full-life-cycle management model; and adopting the corrected full-life-cycle management model to perform full-life-cycle management on the digital medical thermometers, evaluating the use quality of the whole process, and performing evaluation and grading on the use quality of the thermometers according to the obtained indexes. By adopting the technical scheme, the use quality of the digital medical thermometer in the whole life cycle is comprehensively evaluated, and the performance is accurately evaluated.

The present disclosure provides a communication method and apparatus. The method includes: obtaining N pieces of training data, where for one of the N pieces of training data, the training data corresponds to one piece of mark information, and the mark information indicates an attribute of the training data; and sending indication information to a terminal device, where the indication information indicates information about M artificial intelligence models, the artificial intelligence model is trained based on X pieces of training data in the N pieces of training data, and the X pieces of training data are determined based on the mark information. According to the foregoing method, when the N pieces of training data are obtained, the N pieces of training data may be used for constructing training data corresponding to different mark information, to train the M artificial intelligence models or more artificial intelligence models. When the artificial intelligence model is trained, training data for training one artificial intelligence model meets mark information corresponding to the artificial intelligence model, so that scenario-based artificial intelligence model training can be implemented.

This invention discloses a battery-motor collaborative scheduling method, device, and medium based on real-time data, relating to the field of energy dispatching technology. The method includes: collecting battery health data and motor operation data to obtain a raw battery dataset; preprocessing the raw battery dataset to generate a processed battery dataset; evaluating the battery status of the processed battery dataset and combining it with actual battery performance to generate a battery status evaluation report; combining charging control commands with the battery status evaluation report and adjusting the power flow between the battery and the power grid based on the interaction between the battery and the power grid to generate a power dispatching optimization strategy; detecting the battery charging process based on the power dispatching optimization strategy; adjusting the battery charging and discharging power in real time according to the battery status evaluation report to obtain the real-time charging status; adjusting the energy transmission efficiency between the battery and the power grid; and generating a working status report.

The present invention relates to the technical field of solid-state hard disk performance monitoring, and in particular to a real-time performance detection method and system for solid-state hard disks. The method comprises the following steps: performing free sector identification on a solid-state hard disk, and performing normalized operation implantation processing of a marker block to obtain a plurality of normalized operation implanted marker blocks; performing sequential data access requests on a plurality of normalized operation implanted marker blocks, and performing hard disk head seek tracking analysis to construct a global seek tracking trajectory; performing marker block theoretical access speed calculation on the global seek tracking trajectory, and performing access delay calculation to obtain hard disk normalized access delay data; performing progressive implantation analysis of a check block on a solid-state hard disk, and performing hard disk load evolution to generate hard disk load data for each scheme. The present invention realizes accurate and real-time solid-state hard disk performance detection.