25results about How to "Optimization parameters" patented technology

The application relates to the technical field of sewage treatment, in particular to a pressure dosing method and system for sewage treatment. The application first establishes a stable gas pressure environment in a dosing device, and uses a high-precision pressure sensor to collect pressure data in real time; based on Bernoulli equation, a total pressure-dosing flow relationship model and a dosing pipe static pressure-dosing flow relationship model of the dosing device are constructed, and the applicable flow intervals of the two models are determined; after a target flow value is input, the corresponding model is selected according to the interval to calculate the target total pressure or static pressure value; the system adopts a hierarchical PID control strategy, a primary PID is responsible for total pressure adjustment, and a secondary PID is responsible for static pressure fine adjustment, so that the rapidity and accuracy of pressure control are cooperatively ensured; meanwhile, the actual dosing amount is continuously monitored through a flow measurement unit, and when the flow deviation exceeds the set range, a calibration program is automatically started, and the pressure-flow model parameters are optimized in real time; finally, through closed-loop control of pressure and flow, high-precision adjustment of the dosing flow is realized.

This invention provides a method and system for constructing a multi-agent cooperative interception decision model in the field of agent control. The method includes determining the actions that multiple agents independently execute at the current decision moment; updating the relative motion state based on the actions to obtain the relative motion state at the next decision moment and determining the individual immediate reward; determining the total reward of the agent after executing the action based on the individual immediate reward of the agent and the individual immediate rewards of neighboring agents; generating an experience tuple for the agent at the current decision moment by combining the state, action, total reward, and state at the next moment; calculating a loss function based on the experience tuple and optimizing the parameters of the agent's deep Q-network to obtain an optimized deep Q-network; and determining the multi-agent cooperative interception decision model based on the optimized deep Q-network of each agent. This method improves the efficiency of constructing the cooperative interception decision model.

This invention discloses a fuel cell energy management optimization method, comprising the following steps: 1) The fuel cell energy management system maximizes the utilization of fuel cell energy by adjusting and optimizing system parameters; 2) Modeling the fuel cell energy management system, including the stack, battery management system, and DC / DC converter; 3) Simulating and analyzing the fuel cell energy management system, considering the system's operating state, load requirements, and environmental conditions; 4) Optimizing and analyzing the simulation results to obtain the optimal parameter configuration and control strategy; 4-1) Optimization analysis of the fuel cell energy management system; 4-2) Adjustment analysis of the fuel cell energy management strategy; 4-3) Adjustment analysis of the fuel cell energy management system.

The invention relates to the field of intelligent operation and maintenance of power equipment, in particular to a transformer maintenance-free respirator control method integrating temperature, humidity and pressure monitoring and electric heating regeneration and a storage medium. The method comprises the following steps: firstly, deploying a sensor in the respirator, collecting humidity and internal temperature and pressure data of a drying agent, and obtaining accurate data after moving average filtering and temperature compensation; and the control module is combined with the data, a humidity threshold value dynamically adjusted along with the load rate of the transformer and a regeneration period to trigger the regeneration system to start. During regeneration, the electric heating piece heats the drying agent, the fan adjusts the rotating speed according to the real-time temperature and humidity to discharge steam, meanwhile, the moisture absorption capacity recovery rate is corrected by combining aging of the drying agent, and starting and stopping or temperature adjusting delay are conducted according to the standard condition. The process data is uploaded to the operation and maintenance cloud platform through the LoRa module encrypted by AES-128, and the platform analyzes the trend through weighted linear regression and gives an alarm; and operation and maintenance personnel optimize parameters according to the parameters and regurgitate a judgment link. Related programs are stored in the storage medium and can be executed by the processor to achieve the method, and support is provided for maintenance-free of the respirator.

This invention discloses a self-optimizing control method and system for a fabric spreading machine based on multi-source scoring feedback, belonging to the field of textile machinery automation technology. The method includes: real-time acquisition of multi-source data from the fabric spreading machine; calculation of parameter jitter score Q1, operational continuity score Q2, and fabric smoothness score Q3 respectively, and weighted fusion to obtain a comprehensive quality score; when the comprehensive quality score is below a threshold, root cause analysis is performed to generate parameter tuning suggestions; parameter regression prediction is performed using historical high-scoring samples, optimized parameter combinations are output and executed, and the score is updated based on newly acquired data, forming a self-optimizing closed loop. This invention, through multi-dimensional quantitative evaluation and data-driven closed-loop optimization, solves problems such as reliance on experience in manual parameter tuning, subjective quality assessment, and difficulty in fault attribution, significantly improving the consistency of fabric spreading quality and the level of equipment intelligence.

The invention discloses an automatic mining method for a marketing document popular formula based on big data analysis, and belongs to the field of digital marketing intelligent generation. Emotional expression vectors and structural design modes are obtained by analyzing historical document data, emotion generality categories are determined after clustering, and user response intensity is evaluated; and the association rule of the high-response emotion type and the structure mode is mined. A platform exclusive creation rule is determined according to a platform difference refining rule subset, copywriting templates suitable for various crowds are generated by fusing crowd data, and finally the copywriting is output and optimized by matching promotion information and adjusting parameters, so that the problems that traditional copywriting creation depends on artificial experience and is difficult to quantify and optimize are solved, and the copywriting efficiency is improved. And data-driven accurate copywriting generation is realized.

Embodiments of the present application provide a vehicle-cloud collaborative perception defense method, electronic equipment, storage medium and program product. It relates to the technical field of vehicle-cloud collaboration. The method is applied to a vehicle client, a local model is deployed on the vehicle client, the local model is used for attack detection and active defense response, and the vehicle-cloud collaborative perception defense method comprises the following steps: feature extraction and reasoning are performed on multi-source sensor data by the local model to obtain a predicted logic value and a local prediction probability distribution corresponding to the predicted logic value; a variational free energy is calculated based on a global knowledge prior distribution received from a cloud server and the local prediction probability distribution; the variational free energy and the predicted logic value are sent to the cloud server, and a global logic value and a global prediction probability distribution returned by the cloud server are received; and parameter optimization is performed on the local model with the aim of minimizing a composite loss. The present application is used to improve the detection performance of the global model and enhance the effect of vehicle-cloud collaborative perception defense capability.

This invention belongs to the field of smart city information technology, specifically involving a closed-loop management method for the lifecycle risk of smart city building safety using a digital twin. Specifically, it includes constructing a dynamic mapping model of the building's full lifecycle digital twin, designing a core system encompassing full lifecycle data fusion algorithms, virtual-real linkage risk evolution, and closed-loop management iterative optimization. Through digital twin modeling, full-cycle fusion, virtual-real linkage, and closed-loop iteration, it achieves precise management of building safety risks throughout its entire lifecycle, from design, construction, operation and maintenance to decommissioning.

The application provides a kind of cold bending type steel quick online adjustment manufacturing system, including square rectangular tube production line;Steel coil unwinding unit, shearing butt welding unit, loop storage unit, fixed-length sawing unit, automatic collection unit and automatic packing unit are integrated with corresponding control system;It also includes host computer and PLC, the PLC includes multiple, with rough forming unit, high-frequency welding unit, sizing and finishing unit and square rectangular tube production line host one-to-one corresponding setting;The host computer is used to receive external instructions and send it to the corresponding PLC, the PLC is used to control rough forming unit, high-frequency welding unit, sizing and finishing unit and square rectangular tube production line host drive rotating speed according to external instructions respectively;The host computer is also used to control the corresponding control system of steel coil unwinding unit, shearing butt welding unit, loop storage unit, fixed-length sawing unit, automatic collection unit and automatic packing unit.The application can realize the full-line linkage control of square rectangular tube production line.

The application relates to a steel production cost calculation and sales order decision method and system, wherein the method comprises the following steps: synchronously collecting various data of a steel enterprise to form a multi-source data set; converting the multi-source data set into decision rules for steel production, constructing an optimal process path of steel production and constraint conditions of each process of the optimal process path of steel production based on the decision rules; simultaneously realizing the collaborative optimization of the whole-process cost of steel production and the cost of each process based on the multi-source data set, the optimal process path and the constraint conditions of each process to obtain a cost calculation result of steel production; and outputting a steel production suggestion and a cost report based on the cost calculation result of steel production to provide guidance for sales order. The application can provide effective steel cost calculation and guidance for sales order.

The invention provides an E2400 electrically-driven precision seeding operation monitoring terminal, and relates to the field of medical apparatuses and instruments. The terminal adopts a five-layer architecture of'perception-control-execution-interaction-cloud platform ', the core innovation lies in that traditional mechanical transmission and land wheel speed measurement are abandoned, a Beidou satellite positioning speed measurement unit and a millimeter wave speed measurement radar are integrated to form double speed measurement sources, and non-contact accurate speed measurement is realized; 24 rows of seeding and fertilizing execution mechanisms are driven in an electric-direct-drive mode, and sensors such as a seed induction grating and a fertilizing probe are matched to construct a whole-process real-time monitoring and intelligent alarm system. The speed measurement precision reaches the centimeter level, the miss-seeding / re-seeding recognition precision is larger than or equal to 99.5%, the plant spacing deviation is smaller than or equal to + / -3%, the seed and fertilizer saving rate is 10%-15%, the maintenance cost is reduced by 60% or above, the operation efficiency is improved by 30%, the method is suitable for scenes such as large-scale planting bases and agricultural synergy, and the precision, intelligence and universality of seeding operation are remarkably improved.

The invention discloses a simulation test method for a dexterous hand. The simulation test method comprises the following steps: constructing a finite element simulation model based on the configuration of the dexterous hand; performing multi-physical-field multi-working-condition coupling simulation on the finite element simulation model to obtain simulation data; verifying the simulation data according to a pre-constructed bidirectional calibration mechanism and experimental data so as to correct the finite element simulation model; and when the corrected finite element simulation model meets a preset condition, outputting a dexterous hand performance evaluation result. By constructing a finite element simulation model, multi-physical field and multi-working-condition coupling simulation is carried out, so that the simulation model can output multi-dimensional performance data, and evaluation limitation is avoided. And meanwhile, aiming at a complex coupling environment in real operation, dynamic feedback optimization of the model is realized through a pre-constructed bidirectional calibration mechanism and experimental data verification, so that simulation data can be directly used for subsequent model iteration, and a stable and reliable dexterous hand simulation test process is realized.

The present application relates to a kind of non-cooperative communication signal modulation identification method based on multiscale feature fusion extraction, belong to cognitive radio communication technical field, including generating the time domain signal of communication signal under different modulation style and being preprocessed to obtain multiple baseband IQ signal as sample data, with corresponding modulation style as sample label composition communication signal dataset;Initialization multiscale feature fusion extraction network model, utilize the modulation signal dataset to train the multiscale feature fusion extraction network model, obtain trained multiscale feature fusion extraction network model;Real-time acquisition non-cooperative communication signal time domain signal and pre-processing, input the trained multiscale feature fusion extraction network model, obtain the identification result of corresponding time domain signal modulation style.Accurate and efficient identification of non-cooperative communication signal modulation style is realized.

This invention discloses a nanoscale α-alumina and its preparation method. Polyethylene glycol, citric acid, and ammonium aluminum sulfate are sequentially added to a urea solution and mixed to obtain a mixed solution. Ammonia water is then added dropwise to the mixed solution, and the mixture is allowed to stand and age to obtain a precursor suspension. The precursor suspension is then separated into solid and liquid phases, and the resulting solid phase is the alumina precursor. The alumina precursor is then calcined to obtain α-alumina. The α-alumina provided by this invention is a pure phase, free of impurities such as γ-Al₂O₃, with a complete crystal structure, a particle size ≤150 nm, a particle size distribution CV value <15%, excellent dispersibility, no obvious hard agglomerates, and a higher specific surface area, making it suitable for applications in high-end electronic ceramics, precision polishing, and other fields.

This invention relates to the field of offshore wind power operation and maintenance technology, and discloses an intelligent collaborative replacement system and method for large components of offshore wind turbines. The system includes a bottom-mounted work platform, a hoisting control device, a multimodal collision avoidance and attitude stabilization device, and modular tooling assemblies. The hoisting control device plans and commands the main and auxiliary cranes to coordinate their movements through a collaborative controller. The multimodal collision avoidance and attitude stabilization device, through an attitude sensing unit, a data processing center, and an actuator, integrates and processes real-time data to generate trajectory correction commands that suppress swaying and prevent collisions; the actuator is an active guy rope array. The modular tooling assembly includes contour-adjustable elephant-leg tooling, a multi-point adaptive balancing beam, and a buffer protective gear assembly. The replacement method includes steps such as intelligent platform positioning, intelligent collaborative disassembly and reassembly of the rotor assembly, and intelligent replacement of the main unit, achieving precise control of large component replacement operations under wind disturbance conditions and improving operational safety and reliability.

This invention discloses a method and system for controlling slag entrapment defects on the surface of cold-rolled steel coils. The method includes: controlling the temperature and composition of molten iron in the pretreatment process based on the steel grade and smelting process; controlling the oxygen content and temperature at the converter endpoint in the converter steelmaking process according to the steel grade; wherein the oxygen content at the converter endpoint ranges from 350ppm to 500ppm; performing slag modification on the top slag after tapping from the converter; controlling the vacuum degree in the refining process, with oxygen controlled below 350ppm before deoxidation, and the pure circulation time controlled above 6 minutes, followed by secondary slag modification after refining; and in the continuous casting process, using a pre-structured tundish and determining the submerged entry nozzle, nozzle angle, and argon flow rate based on the steel grade to control liquid level fluctuations and reduce slag entrapment.

The application discloses a kind of digital twin motion simulation method and system of tobacco in loose leaf moistening cylinder, the method comprises: establishing 1:1 of loose leaf moistening cylinder three-dimensional geometric simulation model, loose leaf moistening cylinder three-dimensional geometric simulation model includes: tobacco inlet, copy board, cylinder and tobacco outlet, cylinder is inclined downward from the tobacco inlet to tobacco outlet, the axis of cylinder and horizontal plane is preset inclination angle;Establish the three-dimensional geometric simulation particle model of tobacco;The motion of the three-dimensional geometric simulation particle model of initial holding amount of tobacco in loose leaf moistening cylinder three-dimensional geometric simulation model is simulated and calculated.The digital twin motion simulation method of tobacco in loose leaf moistening cylinder provided by the application obtains a large number of experimental data sets through computer simulation test, and compares and analyzes the data with actual production, so as to accurately obtain the parameter design rule and basis of loose leaf moistening cylinder, and avoid the waste of debugging before tobacco production.

This invention relates to the field of textile technology, specifically to a textile process evaluation and optimization system based on big data analysis. The system includes a data acquisition layer, a data transmission and storage layer, a data processing and analysis layer, and an optimization decision layer. The data acquisition layer collects textile process data in real time through multi-dimensional sensors and detection modules. The data transmission and storage layer can transmit data to the cloud in real time and in batches, storing structured and unstructured data long-term. The data processing and analysis layer performs data cleaning, feature extraction, statistical analysis, and machine learning modeling. The optimization decision layer is used to establish process parameter optimization models and generate adjustable solutions. Through big data analysis, it achieves precise control from "experience-based trial and error" to "data-driven" approaches.

This invention discloses a damage localization method for elastomeric layered carbon fiber composite materials based on elliptic probability fusion, comprising: acquiring Lamb wave signals from multiple sensing paths using different sensors as excitation sources; performing continuous wavelet transform using complex Morlet wavelets to extract the measured flight time; introducing a layered wave velocity correction model to optimize the elliptic trajectory method and calculating the theoretical flight time and damage probability; introducing a dynamic minor axis optimization model to optimize the probability weighting method and calculating the damage probability; generating a prior distribution using the results calculated by the elliptic trajectory method and the probability weighting method; constructing a likelihood function based on the difference between the measured flight time and the theoretical flight time; obtaining the posterior distribution of the damage location by combining the prior distribution and the likelihood function using Bayes' theorem; sampling the posterior distribution using an adaptive MCMC algorithm and generating a probability cloud map of the damage location; and outputting the final imaging result of the damage location for damage localization.

The invention relates to the technical field of locomotive lighting, in particular to an intelligent lighting controller for a locomotive headlamp and a control method. The shell is composed of an upper shell body, a lower shell body, a left end cover and a right end cover, the control panel assembly is fixed in the shell body, and an indicating lamp, an SD card slot, a lamp panel temperature / voltage and illuminance data interface, a noise and TVOC data interface, a power supply and light state conversion data interface and an ETH Ethernet communication interface are integrated on the control panel assembly. And a power supply circuit, a main control chip, a clock chip, a data acquisition circuit, a light state switching circuit, a storage circuit, an Ethernet communication circuit and a display circuit are integrated in the control panel assembly. According to the application, full-light and semi-light automatic switching of the headlamp under the scenes of entering and exiting a tunnel and the like is realized through an external illuminance sensor; by collecting the temperature, voltage and internal illuminance data of the lamp panel and combining with the accumulated lighting time, the pre-estimation and early warning of the service life of the headlamp are realized.

The application relates to the technical field of laser micro-nano processing, and particularly provides a femtosecond laser processing system and method, wherein the femtosecond laser processing system comprises a laser emission module and a beam shaping module; the beam shaping module comprises a pulse separator, a spatial light modulator and a beam expander; the laser emission module, the pulse separator, the beam expander and the spatial light modulator are sequentially arranged; the laser emission module is used for emitting laser pulses; the laser pulses sequentially pass through the pulse separator and the beam expander and then reach the spatial light modulator; the pulse separator is used for splitting the laser pulses and introducing a pulse delay time between the split laser pulses; the beam expander is used for expanding the laser beam; and the spatial light modulator is used for shaping a Gaussian beam into a flat-top beam. The application has the effect of improving the processing quality of nanochannels on a grating surface.

The application is a power grid false data injection identification method based on improved CNN-LSTM. It includes the power FDIA feature extraction method based on the stacked autoencoder to solve the problem of large model calculation caused by the feature redundancy of power system operation data. In order to further improve the performance of feature extraction, the attention mechanism module is introduced to improve the model, and the attention mechanism layer is introduced at the input layer of each sub-autoencoder, so as to provide greater weight to the highly relevant features of the attack. The false data injection attack identification model based on CNN-LSTM is constructed, and the sparrow search algorithm is used to optimize the parameters of the CNN-LSTM model to solve the problem that the parameter selection has a great influence on the identification accuracy. The method is scientific and reasonable, has high accuracy, can be applied to the identification of false data injection attacks in power grids, and has certain practical significance for maintaining the safety of power grids.

The invention provides a main body flow determination method, electronic equipment, a storage medium and a program product. The subject flow determination method comprises: according to historical travel data of a road network, determining a plurality of associated paths including a target road section, the historical travel data recording a plurality of road section sets actually driven by each subject in a historical travel process, and the paths being any one road section set; according to the multi-source traffic data of the associated path, determining a main body shunt volume of the associated path in a target time period, the main body shunt volume representing the number of main bodies from a starting point area of the associated path to an end point area of the associated path along the associated path in the target time period; and taking a result obtained by adding the main branch flows of the associated paths as the main flow of the target road section in the target time period. According to the technical scheme, dependence on large-scale sample data training is avoided, and the subject flow of any road section in the road network can be accurately and efficiently estimated without a neural network.

This invention provides a design method for a casing treatment with stabilization and noise reduction functions. The method includes: constructing the wall impedance equation of the casing treatment using the equivalent distributed source method and the characteristic impedance model of porous materials; matching the perturbation continuity equation and momentum equation of the casing treatment section with the bladeless region of an axial compressor using the transfer element method; constructing the perturbation frequency characteristic equation of the axial compressor with casing treatment based on small perturbation theory; solving the perturbation frequency characteristic equation to obtain the perturbation frequency; determining the instability point of the axial compressor based on the imaginary part of the perturbation frequency; calculating the instability point of the axial compressor under different casing treatment parameters; and selecting the casing treatment parameters corresponding to the lowest instability point as the design parameters of the casing treatment. Implementing the technical solution of this invention can optimize the stabilization of the casing treatment. The casing treatment designed using the technical solution of this invention can achieve noise reduction due to the use of porous materials.

The invention discloses a magnetic suspension bearing suspension force and zero sequence current decoupling control system, particularly relates to the technical field of magnetic suspension, and comprises a four-bridge-arm inverter module, a double closed-loop control unit, a multi-source sensing acquisition module, a power driving module and a magnetic suspension bearing body. According to the decoupling control system for the suspension force and the zero-sequence current of the magnetic suspension bearing, the fourth bridge arm connected with the neutral point of the three-phase winding is additionally arranged, an independent zero-sequence current loop is constructed, and zero-sequence current closed-loop control based on the alpha-beta O coordinate system is matched, so that the suspension force and the zero-sequence current of the magnetic suspension bearing are decoupled. Zero-sequence current generated by three-phase imbalance, parameter asymmetry and magnetic field distortion in a traditional three-phase inverter architecture can be pertinently dredged and suppressed; and the fourth bridge arm adopts a SiCMOSFET power switch device and fly-wheel diode combined design, so that a compensation instruction can be quickly responded, winding current distortion caused by superposition of zero-sequence current to working current is avoided, and a linear control relationship between suspension force and current is guaranteed.