21results about How to "Has engineering application value" patented technology

The invention discloses a multi-task electromagnetic model based on a hybrid expert network, and belongs to a wireless communication technology. The model comprises a preprocessing module, a feature extraction module, a task output module and a pre-training-fine tuning learning strategy. The preprocessing module carries out standardization processing on the multi-source electromagnetic signals; the feature extraction module is based on a Transform structure, introduces a hybrid expert network to replace part of a traditional feedforward neural network, and dynamically selects an expert sub-network through a task specific routing mechanism; the task output module configures a special structure according to different task targets; in the pre-training stage, a mask auto-encoder is used for pre-training large-scale label-free data, and a downstream task is subjected to full-amount fine adjustment through small-scale label data. According to the method, multi-task collaborative learning and differential expression are realized, and the recognition performance, robustness and processing efficiency of the model in a complex electromagnetic environment are improved.

The invention belongs to the field of data processing and water resources, and particularly relates to a running water ecological landscape water demand calculation method based on a mountainous area. The method comprises the following steps: identifying characteristics such as karst landforms and ancient terraces through remote sensing data; constructing a nonlinear regulation and storage model of the underground water storage reservoir to simulate a base flow hysteresis release effect; calculating a landscape fragmentation degree and a connectivity index, and establishing a mapping relation between a space structure and a hydrological process; constructing an ecological water demand quota correction matrix based on nonlinear regression; and outputting a final water demand calculation result in combination with the base flow supply sequence and the dynamic landscape parameters. According to the method, by introducing the nonlinear regulation and storage theory and the landscape pattern index, deep coupling and dynamic correction of the ecological water demand and the space structure are achieved, and the precision and scientificity of water demand calculation in the complex environment are improved.

This invention belongs to the field of thermal power unit automation control technology, and discloses a configuration batch replacement method and related device based on hash caching technology. The configuration batch replacement method includes: reading a modified measurement point mapping table to obtain all source KKS and target KKS; updating the original hash table cached in memory based on the obtained source KKS and target KKS to obtain an updated hash table; traversing the updated hash table to perform configuration batch replacement; wherein, during configuration replacement, location information is obtained through the measurement point type, and then the source KKS in the indexed variable is replaced with the target KKS through the location information index variable. This invention can achieve fast and accurate replacement of KKS within the configuration, helping engineers to perform efficient configuration design; this invention solves the technical problems of slow configuration replacement speed and high system resource consumption in existing technologies, and has high engineering application value.

This invention discloses a directional coupler with reconfigurable filtering characteristics that reuses resonant stubs, belonging to the field of microwave and millimeter-wave circuit technology. The coupler includes a main body consisting of a main transmission line, a coupling transmission line, and four ports. A serpentine resonant stub unit is connected in series on both the main and coupling transmission lines. Each resonant stub leads out an equivalent inductance stub and is connected to a reflective single-pole double-throw switch. The other end of each switch is connected to a variable capacitor. This invention allows the variable capacitor to be selectively connected to either the main or coupling resonant stub via switching, achieving time-division multiplexing of the resonant stubs and forming an adjustable parallel LC resonant circuit, resulting in two operating modes: when the capacitor is connected to the main path, both the main and coupling channels simultaneously possess low-pass filtering characteristics with continuously adjustable cutoff frequencies; when the capacitor is connected to the coupling channel, only the coupling channel is filtered, while the main channel achieves low insertion loss pass-through across the entire frequency band. This invention has the advantages of compact size, low insertion loss, and a wide filtering tuning range.

The invention aims to provide a rolling bearing weak fault feature enhancement extraction method, and belongs to the field of rolling bearing fault diagnosis. Firstly, a 1 / 3 binary tree filter bank is adopted to divide an original signal into a plurality of frequency bands, then spectrum amplitude modulation (SAM) is carried out on each filtered signal, a discrete normalized envelope spectrum peak factor (DNESP) index is established, the most MO range is selected in a self-adaptive mode, and then a normalized square envelope spectrum is reconstructed. And finally, constructing a normalized envelope spectrum peak factor (NESP) as an evaluation index, quantizing and reconstructing the richness of fault information in the normalized square envelope spectrum, and adaptively selecting a demodulation band with significant fault information according to the richness, thereby realizing enhanced extraction of fault features. The method provided by the invention can effectively extract weak fault features of the rolling bearing, and has good robustness for various interferences and noises.

The invention belongs to the crossing field of artificial intelligence and geophysics, and particularly discloses a transient electromagnetic forward modeling method and system considering an induced polarization effect based on deep learning, and the method comprises the steps: inputting Cole-Cole key parameters representing the induced polarization effect into a CNN-LSTM mixed deep learning network model for forward modeling, and obtaining transient electromagnetic response data; wherein Cole-Cole key parameters representing the induced polarization effect serve as multi-dimensional input features of the CNN-LSTM mixed deep learning network model, transient electromagnetic response data of a corresponding time channel serve as output, and the CNN-LSTM mixed deep learning network model is trained by adopting a segmented weighted mean square error loss function. According to the method, the segmented weighted mean square error loss function is adopted in the training stage, higher weights are given to the symbol inversion points and the prediction errors of the adjacent time windows, and the reliability of all-time response prediction is guaranteed.

The present application belongs to the field of radio wave propagation theory and electromagnetic field numerical calculation technology, and discloses a low-dispersion very low frequency radio wave propagation characteristic prediction method.The present application method aims at the defects of large numerical dispersion, low calculation efficiency and high resource occupation of the existing FDTD method in long-distance, complex layered medium scene, and realizes high-precision and high-efficiency prediction of the very low frequency radio wave propagation characteristic in seawater-air-ionosphere medium by combining HIE-FDTD method and M(2,4)FDTD method through three-layer processing of the whole calculation area.The present application method solves the contradiction between reducing numerical dispersion error and reducing computer consumption of the traditional FDTD method, can greatly reduce computer memory resource occupation and improve calculation speed while ensuring prediction accuracy, is suitable for electromagnetic propagation simulation of large-size, complex medium and terrain, and has obvious advantages in phase delay prediction of very low frequency ground wave propagation.

The invention discloses an intelligent optimization energy storage efficient charging and discharging method. The method comprises the following steps: S1) acquiring parameters of a power system; s2) calculating an average value of 24-hour electricity consumption according to a daily load demand curve of the power system; s3) identifying a whole interval graph formed by enveloping the average value and the daily load demand curve, defining an area above the average value as a positive area, and defining an area below the average value as a negative area; s4) performing charging operation in the negative area region and performing discharging operation in the positive area region according to the pumping and storage parameters and the energy storage parameters; the method comprises the steps of (S1) carrying out pumping and storage treatment on a pumping and storage system, (S2) carrying out pumping and storage treatment on the pumping and storage system, (S5) repeating charging and discharging operations until the total charging and discharging area requirement of the day is met or the maximum operation frequency limit is reached, (S6) forming a new demand curve after all pumping and storage treatment is finished within one day, and (S7) carrying out energy storage treatment on the new demand curve to form an optimized demand curve, optimizing the charging and discharging strategy of the pumping and storage and energy storage system, and reducing the calculated amount.

The application discloses a preparation method of ZnCl2-PEI composite modified ceramsite and application of the ZnCl2-PEI composite modified ceramsite in removal of PAC, and belongs to the field of water treatment environmental protection materials and wastewater treatment technologies. The modified ceramsite is prepared by taking ordinary ceramsite as a base body, performing water washing pretreatment, ZnCl2 calcination activation pore forming and PEI grafting, is rich in micropore structure and amino functional groups on the surface, and has double characteristics of physical adsorption and chemical chelation. Meanwhile, an ecological filter adsorption continuous treatment process suitable for the modified ceramsite is designed, and continuous, efficient and stable treatment of PAC-containing wastewater is realized. The modified ceramsite has the advantages of large adsorption capacity, high mechanical strength, water washing resistance, and suitability for different PAC concentration ranges and pH ranges, and is simple to operate and controllable in running parameters, and the removal rate of PAC in water can be up to 90%.

The invention discloses a nuclear reactor few-group diffusion coefficient solving method based on anisotropic neutron transport, which comprises the following steps: firstly, generating a material area multi-group macroscopic cross section by a multi-group cross section library according to the geometric dimension and material information of a nuclear reactor fuel assembly; secondly, on the basis of the multi-group macroscopic cross section of the material area, adopting anisotropic P1 neutron transport calculation to obtain the neutron flux density of the flat source area, and normalizing according to volume weight to obtain a component neutron energy spectrum; then, establishing a multi-group diffusion coefficient solving linear equation set, and solving to obtain a multi-group diffusion coefficient; further, carrying out transport correction by utilizing a multi-group diffusion coefficient to obtain a transport cross section and a corrected zero-order self-scattering cross section; finally, isotropic neutron transport calculation of transport correction is carried out, leakage correction and component homogenization are carried out after the neutron flux density of a flat source region is obtained, and a few-group diffusion coefficient is generated. According to the method, the error caused by approximate processing is reduced while the requirement for component calculation efficiency is met, and the method has engineering application value.

The chemical component of the carbide-based eutectic medium-entropy alloy is CrFeNi (NbC) x, x is larger than or equal to 0.1 and smaller than or equal to 0.3, the carbide-based eutectic medium-entropy alloy is prepared through a vacuum arc melting method, and after vacuum arc melting, the solidification phase of the carbide-based eutectic medium-entropy alloy is FCC, MC carbide and M7C3 carbide; and when x is equal to 0.1, the solidification structure of the alloy is primary FCC + reticular eutectic FCC / MC + superfine eutectic FCC / M7C3. When x is equal to 0.2-0.3, the solidification structure of the alloy is primary MC + primary FCC + reticular eutectic FCC / MC + superfine eutectic FCC / M7C3; the Vickers hardness of the alloy ranges from 235.6 HV to 287.0 HV, the yield strength of the alloy ranges from 439 MPa to 534 MPa, and the compressive strength (Rpc30) of the alloy under the specified strain (30%) ranges from 1482 MPa to 1607 MPa.

This invention provides a method for measuring the volume of hydraulic perforation in underground coal mines. The method includes closing two valves and a throttle valve; opening the air pump and valves to extract high-pressure nitrogen from a storage tank into the measuring instrument; and closing the air pump and valves when the pressure in the measuring instrument reaches a preset value, and recording the nitrogen pressure value in the measuring instrument. P 1. Volume V 1 and temperature T 1. Open the throttle valve to allow nitrogen gas from the measuring instrument to enter the measuring bladder and expand it, and observe the second pressure gauge; when the pressure value reaches the preset value of the measuring instrument, close the throttle valve and record the pressure of nitrogen gas in the measuring instrument. P 2. Volume V 2 and temperature T 2; Read the reading of the second pressure gauge. P 3. Input the measuring instrument; operate the measuring instrument to calculate the volume of the punched hole. This invention can effectively overcome errors caused by factors such as differences in coal body parameters and coal powder loss, and control the volume measurement error within 2%.

This invention discloses a multi-resource collaborative scheduling method for distribution networks under earthquake disasters, designed to address the challenges posed by the uncertainty of line faults and the interruption of power supply to critical loads caused by earthquakes to the operation of distribution systems. The method constructs a line fault probability model using a seismic attenuation model and vulnerability curves from the US Earthquake Disaster Loss Assessment System, and introduces line reinforcement decisions to form a decision-dependent uncertainty model, accurately reflecting the coupled impact of earthquake intensity and planning decisions on line faults. Simultaneously, a multi-dimensional flexible resource scheduling framework integrating mobile hydrogen energy resources, building thermal inertia, and demand response is adopted to enhance energy supply capacity and system resilience during the disaster phase. A 1-norm-based multi-resource collaborative scheduling optimization model is introduced to characterize the probability shift of earthquake scenarios, and a column constraint generation algorithm is designed to achieve efficient collaborative solution of planning and scheduling. This invention can significantly improve the operational reliability of distribution networks under earthquake disaster scenarios and has certain engineering application value.

The application discloses a sea evaporation duct path loss adaptive prediction method based on small sample transfer learning, and belongs to the technical field of sea wireless communication. In view of the problem that the prediction accuracy of a pure simulation data driven model is reduced due to the difference between a simulation scene and a measured scene in actual sea area application, a transfer learning framework of'source domain pre-training-target domain fine-tuning' is constructed. First, a Transformer model optimized by a tree structure Parzen estimator is pre-trained by using large-scale simulation data, so as to extract general physical features of evaporation duct propagation; second, for sparse single-point measured data, the encoder parameters are frozen and the decoder is fine-tuned, and a point constraint loss function is introduced, so as to calibrate the full path prediction curve by using the measured path loss value of a link specific distance point. The application can reduce the demand for measured data, improve the path loss prediction accuracy and time sequence stability, and is suitable for sea over-the-horizon communication link modeling and performance evaluation.

The application discloses a soft-hard interbedded rock mechanical parameter prediction method and system based on a residual attention network, relates to the technical field of rock mechanical parameter prediction, and has the advantages that the traditional neural network is prone to gradient disappearance when processing a deep network, which influences the training effect; the existing method lacks an attention mechanism and cannot effectively identify and strengthen key features; and the modeling capability for interlayer interaction is insufficient; the application provides a soft-hard interbedded rock mechanical parameter prediction method based on a residual attention network, which comprises the following steps: obtaining structure parameters and target mechanical parameters of a soft-hard interbedded rock sample; converting the structure parameters into an enhanced feature vector; constructing a residual attention network model; inputting the enhanced feature vector into the residual attention network model; and outputting a mechanical parameter prediction result and reliability evaluation information.

The invention discloses a structural stability topological optimization method and system considering a dead load and live load separation effect. The method comprises the following steps: establishing a finite element model of a to-be-optimized structure; taking the material pseudo density of each finite element as a design variable, taking maximization of a buckling load factor corresponding to a live load as an optimization objective function, and taking a constraint function that the total volume and the total flexibility of the structure do not exceed a set threshold value to obtain an optimization model; performing eigenvalue buckling analysis considering a dead load and live load separation effect on the finite element model to obtain a buckling load factor corresponding to the live load; calculating the sensitivity of the optimization objective function and the constraint function to the design variable; on the basis of the sensitivity information, an optimization algorithm is adopted to update design variables; and outputting a final structure topology until an optimization result meets a preset convergence condition. According to the method, the stable bearing capacity of the structure can be accurately optimized, and topological design with better mechanical properties and higher material utilization rate is obtained.

A kind of hydraulic bulging process for thick-walled cylinder instability control belongs to hydraulic bulging technical field, overcome the defect that instability easily occurs in existing thick-walled cylinder hydraulic bulging process, comprising the following steps: S1, blank pretreatment;S2, determine the judging basis of blank hydraulic bulging instability;S3, finite element simulation modeling and parameter setting;S4, mold assembly and blank installation;S5, hydraulic bulging, in the process of hydraulic bulging, the shape size data of blank is monitored in real time by visual displacement sensor, if it is found that blank occurs bulging, concave waist or horn shape instability phenomenon, immediately stop bulging process, unload and accurately measure the shape size of blank, according to the shape shape and size measurement result of blank, adjust the loading rate of upper punch, then blank is reloaded on the force-reducing hydraulic bulging die, continue to bulge processing to blank.The present application effectively suppresses bulging, concave waist, horn mouth and other instability phenomena, realizes the uniform deformation of thick-walled cylinder bulging process.

The utility model discloses a railway power supply equipment detection device, which solves the problem that the existing railway power supply equipment detection device is difficult to install and poor in detection effect caused by infirm installation. The detection device comprises a host, a rigid platform, and a Beidou mobile station module, an attitude acquisition module, a laser radar acquisition module, a mileage coding module, an image acquisition module, a power supply module and a Beidou base station module which are electrically connected with the host, the upper end and the lower end of the transition connecting profile are each fixedly provided with a pipe hoop through a connecting corner connector, and the rigid platform is installed at the tail or the front section of a rail operation vehicle through the pipe hoops. A pair of adapter plates is fixedly installed at the four tail ends of the cross-shaped support. The device adopts an anti-vibration design, can be firmly mounted, meets a complex test environment of the rail operation vehicle, and ensures long-term stable operation of the system.

The invention belongs to the field of experimental aerodynamics, and discloses a data analysis method for a digital valve flow control unit. The data analysis method for the digital valve flow control unit comprises the steps that a digital valve system is installed; installing a sensor; carrying out a flow measurement test; calculating the outlet flow of the digital valve system. According to the data analysis method for the digital valve flow control unit, the data processing and analysis method for the digital valve flow control unit is established through strict theoretical derivation and flow measurement test verification on the basis of a hydromechanics basic equation, test and theoretical support is provided for accurate control of high-pressure air flow, and the method has engineering application value.

The invention discloses a field-circuit coupling numerical simulation method for calculating spatial thermal field distribution of a high-voltage tunnel cable, and belongs to the technical field of energy research, and the simulation method specifically comprises the following steps: S1, simulating important geometric parameters collected and recorded by a simulation system according to high-voltage tunnel cable equipment which is actually put into operation; the method can comprehensively analyze the temperature field distribution of the cable system under the conditions of different loads, tunnel structures, laying modes and the like, improves the precision and reliability of temperature monitoring, and compared with the traditional technology, reduces the computing resources, accurately simulates the operation condition of the KM-level long-distance tunnel cable, improves the analyzability of the simulation result, and reduces the construction cost. According to the method, the engineering reference precision of practical application is deepened, the temperature field distribution condition of the high-voltage tunnel cable in a long-time operation state can be analyzed, and the method has important engineering application value and particularly has great significance in operation monitoring, fault early warning and optimization design of the high-voltage tunnel cable.

The present application relates to the technical field of foundation pit engineering monitoring, and discloses a deep and large foundation pit deformation monitoring method and system based on close-range photogrammetry, wherein the deep and large foundation pit deformation monitoring method based on close-range photogrammetry comprises the following steps: receiving risk area data to generate a standardized image set; combining domain knowledge to perform adaptive three-dimensional reconstruction to output a three-dimensional grid model with semantic annotation; extracting a model deformation mode and performing semantic annotation to form a structured set; constructing a bidirectional information channel between a geometric domain and a knowledge domain based on this to create a knowledge-geometry collaborative measurement graph; using a multi-scale space-time memory structure to realize self-evolution learning and resource optimization allocation, outputting a resource allocation strategy and updating a monitoring knowledge base; performing risk assessment according to the deformation mode, semantic interpretation and updated knowledge base to produce intervention suggestions, and outputting hierarchical early warning information and multi-level decision support schemes; and the present application realizes automatic processing and optimization from data to decision.