16results about How to "Improve imbalance" patented technology

The invention provides a soil conditioner for acidified soil remediation and a preparation method thereof, and belongs to the technical field of soil conditioners. The invention discloses a soil conditioner for acidified soil remediation. The soil conditioner is prepared from the following raw materials in parts by weight: 25-30 parts of decomposed organic matters, 15-23 parts of calcined alkaline slag powder, 20-30 parts of a biochar-based composite material, 6-13 parts of a microbial agent and 10-20 parts of a bio-based adhesive. According to the soil conditioner for acidified soil remediation, the unbalance state of acidified soil is improved, and the acidification process is relieved.

The invention provides a tunnel disease monitoring data enhancement method and system based on CTGAN, and relates to the technical field of data enhancement, and the method comprises the steps: carrying out the disease type marking and disease feature preprocessing of a real data table of a tunnel site disease; extracting time sequence features and spatial position features from the preprocessed disease features; taking the time sequence features, the spatial position features and the disease types as conditional information to train a CTGAN, and using the CTGAN to generate synthetic data having similar distribution with the preprocessed real data table to obtain a new sample data table; after the new sample data table is generated, verifying a synthesis effect by adopting a mode of combining a statistical method and a support vector machine; according to the method, the confrontation network is generated by using the condition table, tunnel disease monitoring data is supplemented and optimized based on a real data table of tunnel field diseases, the problems of data scarcity and category imbalance are solved, and the training effect of a machine learning model is improved.

The invention relates to the technical field of electric power systems, in particular to a power distribution network electric energy quality treatment method and device based on a flexible multi-state switch, and the method comprises the steps: enabling the flexible multi-state switch FMS to be installed on a node in an electric power system according to an actual demand; according to the actual operation data of the three-phase unbalanced power system, a mathematical model of a power distribution network accessed to the FMS is established, the mathematical model of the power distribution network accessed to the FMS comprises an objective function, the objective function takes the minimum total network loss including line loss and FMS operation loss as a reference, and a control strategy and an operation mode of the flexible multi-state switch FMS are determined; the method is applied to FMS-based power distribution network optimization control. A simulation experiment is carried out in an IEEE 33-node power distribution network model, and the result shows that the FMS can effectively reduce the system network loss, improve the voltage level and improve the three-phase unbalance degree. The method provides powerful technical support for application of the flexible multi-state switch in an actual power system.

The present invention discloses a method for forecasting the continuous distribution probability of extended-period precipitation by integrating fine terrain features, including: obtaining standardized multi-member multi-factor forecast data of the extended-period numerical model ensemble forecast system and regional historical long-term series precipitation observation data, and extracting the continuous distribution characteristics of historical precipitation; obtaining a descaled key terrain multidimensional feature factor library based on regional high-resolution fine terrain data; constructing an extreme gradient boosting U-Net hybrid deep learning precipitation probability forecast model, and carrying out model training and optimization; using the trained model to generate an improved forecast result of the continuous distribution probability of extended-period precipitation. The present invention integrates fine terrain data, adopts a weighted cross entropy loss function, effectively handles the problem of imbalanced precipitation distribution categories, improves the generalization ability of the model, and improves the forecast performance of the continuous distribution probability forecast of precipitation, which has extremely high application value.

The invention discloses a traditional Chinese medicine tea drink for preventing and treating oral mucosa diseases and a preparation method of the traditional Chinese medicine tea drink. The traditional Chinese medicine tea drink is prepared from 10-15 parts of radix astragali, 8-12 parts of radix codonopsis, 6-10 parts of radix salviae miltiorrhizae, 5-9 parts of radix angelicae sinensis, 3-6 parts of fructus psoraleae, 6-9 parts of rhizoma polygonati, 6-9 parts of radix adenophorae, 6-10 parts of rhizoma atractylodis macrocephalae, 5-9 parts of radix ophiopogonis, 3-5 parts of radix glycyrrhizae, 4-6 parts of pericarpium citri reticulatae and 5-8 parts of herba taraxaci. Aiming at the core pathogenesis of low immunity, long-term deficiency, long-term deficiency and kidney-yang insufficiency of patients suffering from oral mucosa diseases, qi tonifying, blood activating and kidney-qi strengthening are taken as a core treatment method, the formula is scientific and reasonable, all the medicinal materials act synergistically, the overall immune function of the body is adjusted, T lymphocyte subpopulation imbalance and cell factor network disorder are improved, and the curative effect is good. The oral mucosa microcirculation is improved, local inflammation is relieved, the immunity of a patient is improved fundamentally, the recurrence rate of oral mucosa diseases is reduced, and the problems that an existing tea drink is not high in pertinence and limited in curative effect are solved. The tea drink is appropriate in taste, good in compliance, high in safety and suitable for people with various oral mucosa diseases and low immunity.

The present invention discloses a method for mining difficult samples of remote sensing images. The method pre-trains a semantic segmentation module including two semantic segmentation models, inputs the image to be interpreted into the semantic segmentation module, obtains a first semantic segmentation result and a second semantic segmentation result, and then obtains a common semantic segmentation result, a full semantic segmentation result and an intersection-and-union ratio through intersection calculation, union calculation and intersection-and-union ratio calculation. Finally, by comparing the intersection-and-union ratio and a preset IOU threshold, the samples with an intersection-and-union ratio less than the preset IOU threshold are recorded as difficult samples, and the samples with an intersection-and-union ratio greater than the preset IOU threshold are recorded as simple samples. The difficult samples are manually labeled, the simple samples are manually quality-checked, and added to a sample library. The method obtains different semantic segmentation results for the same image through two different semantic segmentation models, and distinguishes simple samples from difficult samples by calculating the intersection-and-union ratio of the segmentation results. Simple samples and manually labeled difficult samples can expand the sample library.

The invention discloses an application of mecobalamin in preparation of a preparation for improving intestinal flora imbalance of a subject or treating and / or preventing related diseases of intestinal flora imbalance of the subject, mecobalamin can regulate intestinal flora in the aspects of Alpha diversity, Beta diversity, gate-level mouse intestinal flora structure, family-level mouse intestinal flora structure and the like, and can be used for treating and / or preventing intestinal flora imbalance of the subject. The pharmaceutical composition has a preparation developed to improve intestinal flora imbalance of a subject, or treat and / or prevent diseases related to the intestinal flora imbalance of the subject.

This application discloses a training method, apparatus, electronic device, and storage medium for a neural network, belonging to the field of data processing technology. The method includes: obtaining a first predicted probability value predicted by a first neural network among two adjacent neural networks in M ​​neural networks when a training sample is used as input; a second neural network in the M neural networks uses the first predicted probability value as the initial value of the input weights to perform weighted processing on the training samples to obtain a weighted result; the second neural network obtains a calculation result based on the weighted result and a preset deviation value, and determines the prediction error value between the calculation result and the actual result of the training samples based on the calculation result and a preset loss function, wherein the loss function has a reward score; updating the reward score based on the prediction error value; and updating the input weights and deviation value based on the prediction error value and the updated reward score until the loss function converges, at which point the training of the second neural network ends.

The present application relates to the technical field of bellows preparation, and particularly relates to a cathode device for preparing an electrodeposited bellows and a preparation method, which comprises a frame body, a core mold assembly, a cathode wire assembly and a power transmission assembly. The frame body is internally provided with the core mold assembly, part of the cathode wire assembly and the power transmission assembly. The inner wall of the cylindrical groove of the frame body is provided with annular sawteeth. The core mold assembly comprises a horizontal connecting rod and a core mold body, the other end of the core mold body extends into the frame body, and each aluminum alloy core mold of the core mold body is detachably connected with a nylon external gear meshing with the annular sawteeth at the end away from the horizontal connecting rod. The present application can effectively control the rotating speed of a single core mold during the electrodeposition process, ensure the stable and uniform rotation of the core mold in the circumferential direction, thereby improving the imbalance of the circumferential current density and ion transmission conditions during the electrodeposition process, and ensuring that the prepared single-layer or multi-layer bellows has a uniform wall thickness in the circumferential direction.

The invention discloses a collaborative optimization control method and system for an energy storage converter and the energy storage converter, and relates to the field of energy storage optimization, and the method comprises the steps: receiving a total output power instruction of an electric energy management system through the energy storage converter; constructing an optimization problem based on the total output power instruction by taking the minimum battery charge state difference value of a battery module in the battery system as a target; and solving the optimization problem, determining the output current of each battery module in the battery system, and performing optimization control on the battery system according to the output current of each battery module. The problem of an unbalanced state in the battery system is improved, and the health level and the available capacity of the battery system are improved.

The application discloses a PCB defect image detection method based on an improved deep learning algorithm. In view of the deficiencies of the YOLOv3 deep convolutional neural network model, such as imbalance of positive and negative samples, feature expression of small defects of the PCB, and extraction of important feature information of the image, the application proposes an improvement method of replacing the MSELoss function in the original YOLOv3 model with the FocalLoss function, so as to dynamically control the weights of positive and negative samples and difficult and easy classification samples. Meanwhile, the application embeds the SPP module in the branch small target detection head and parallelly integrates the CBAM module into the YOLOv3 model, so that the model can pay more attention to the defect information in the image, thereby improving the overall feature extraction capability of the model. The method can detect different types of PCB defect images, and compared with the original YOLOv3 model, the sensitivity and other indicators are improved, has good performance in the PCB defect detection task, can reduce the missed detection rate and the false detection rate, and better realizes the automatic and intelligent detection of the PCB defects.

The present invention relates to the technical field of unmanned aerial vehicle fault diagnosis, and in particular to a method for unmanned aerial vehicle sensor fault diagnosis based on weighted transfer domain adaptation learning. It includes: constructing a multi-scale neural network model, including three multi-scale neural network modules of a feature extractor, a domain discriminator and a classifier; the feature extractor extracts high-dimensional feature information of the data; the domain discriminator performs adversarial training with the feature extractor during the model training process, so that the domain discriminator cannot distinguish whether the data comes from the source domain or the target domain; the classifier identifies and classifies the sample features output by the feature extractor; obtains a source domain sample data set and a target domain sample data set; trains the constructed multi-scale neural network model; and performs fault identification and diagnosis on the target domain data through the trained multi-scale neural network model. The advantage is that in the absence of target domain data set labels, a better learning model is established for unmanned aerial vehicle fault diagnosis, and fault diagnosis of the target domain data set is realized.

The utility model discloses a sewage treatment dosing system which comprises at least one sodium hypochlorite adding line, the sodium hypochlorite adding line comprises a sodium hypochlorite storage tank, a pumping device, a valve, a low-pressure pipe and a check valve which are sequentially connected through a pipeline, the outlet end of the check valve is connected with a dosing pipe, and the dosing pipe extends to a dosing point for dosing. The pipeline comprises a main pipeline connected with the sodium hypochlorite storage tank and a plurality of branch pipelines, the pumping device, the valve, the low-pressure pipe and the check valve are all located on the branch pipelines, and a back pressure valve is arranged on the pipeline between the valve and the low-pressure pipe. The back pressure valves are arranged at the pipeline positions of the branch pipes, and the problem that the pesticide amount of the branch pipes is unbalanced is greatly solved through adjustment of the back pressure valves; in addition, the pressure of the valve element of the back pressure valve is set to be of an automatic adjusting structure, adjustment can be conducted according to needs in the using process so that different opening degrees can be provided, in addition, the back pressure valve can also be used as a control valve, and forward and reverse flushing can be provided when cleaning is needed.

This invention provides the application and method of Streptomyces S16-1 in promoting the growth and salt stress tolerance of Leymus chinensis, belonging to the field of microbial technology. Streptomyces S16-1 was isolated from saline soil, with the preservation number CCTCC M 20251940, and its 16S rDNA sequence is shown in SEQ ID NO:1. The Streptomyces S16-1 has functional activities such as salt and alkali tolerance, phosphorus solubilization, nitrogen fixation, and production of 1-aminocyclopropane-1-carboxylic acid (ACC) deaminase. The microbial fertilizer is obtained by expanding the culture of Streptomyces S16-1. Compared with chemical fertilizers, the microbial fertilizer used in this invention is more environmentally friendly. It can not only improve soil fertility and help plants improve their salt and alkali resistance, but also promote the balance and diversity of soil microbial communities, which helps maintain the health and stability of the soil ecosystem.

This application relates to an ultra-wideband dual-mode mixer circuit with an on-chip balun. The circuit includes: an LO port of a local oscillator differential balun unit connected to the signal output from the local oscillator source in a vector network analysis (VNA) system; an RF port of a mixer unit connected to the test signal preprocessed by the VNA system; an output port of the local oscillator differential balun unit connected to the local oscillator input port of the mixer unit; and an output port of the mixer unit connected to the input port of an intermediate frequency (IF) amplifier unit. The local oscillator signal in the VNA system is processed by the local oscillator differential balun unit to obtain two local oscillator differential signals. The high-frequency test signal preprocessed by the VNA system is input through the RF port of the mixer unit to obtain two radio frequency (RF) differential signals. The two RF differential signals are mixed with the two local oscillator differential signals to obtain the desired IF signal. The IF signal is finally amplified by the IF amplifier unit and output from the output port of the IF amplifier unit, thus solving the problems of low integration, large size, and poor isolation in ultra-wideband mixers.

The present invention provides an application of circular RNA circAASS in preparing a product for diagnosing and / or treating acute kidney injury, belonging to the field of biomedicine technology. The present invention studies and finds that the expression of circAASS is significantly reduced in the renal cortex tissue of mice in three AKI models (IRI-AKI, CP-AKI and SAKI), indicating that the reduced expression of circular RNA circAASS may promote the occurrence and development of AKI; restoring the expression of circAASS effectively reduces the ROS generation of damaged TEC, inhibits TEC apoptosis and pro-inflammatory response; the circular RNA circAASS provided by the present invention is expected to provide a new strategy for preventing and treating AKI.