302 results about "Imbalance problems" patented technology

The invention discloses a measurement method and a device applied to a layered network. The method of the invention includes the following steps: after receiving the signaling carrying the indication information of the partial bandwidth sent by the base station, the user equipment, according to the indication of the partial bandwidth, The information is measured on the corresponding bandwidth of the used carrier; or, the user equipment determines a part of the bandwidth for measurement by itself, and performs the measurement on the part of the bandwidth of the used carrier. The invention can solve the problem of inaccurate wireless link monitoring caused by unbalanced interference on each sub-band in a layered network.

A reactive power compensation device based on novel modular multilevel topology and a control method thereof relate to a high-voltage and large-power reactive power compensation device. In the prior art, compensation in a high-voltage and large-power field is limited and an inhibiting ability to circulation is insufficient. The invention aims at solving the above problems. A three-phase alternating-current power supply, a resistance-inductance load, a MMC converter, a signal detection circuit, a control circuit and a driving circuit are included. The MMC converter comprises three bridge arms which share a same structure and are connected in parallel. Each bridge arm comprises an upper bridge arm and a lower bridge arm which are symmetrical about a middle point of the bridge arm and are connected in series. Each upper bridge arm comprises a resistor, an inductor, several half-bridge units and an H bridge unit which are connected in series. The inductor of each upper bridge arm and an inductor of each lower bridge arm are connected in series. The middle points of the three bridge arms of the converter are connected in parallel between the three-phase alternating-current power supply and the load through a lead. By using the device and the method in the invention, power-grid reactive power can be compensated; a three-phase imbalance problem of a system is solved; effects of supporting a power grid voltage and inhibiting the circulation are achieved.

The invention discloses a method for early classifying imbalance multi-variable time sequence data. The method comprises the following steps: carrying out the under-sampling for a large category of category data sets according to an imbalance ratio for the inter-category data scale imbalance problem into a plurality of subsets, and combining the subsets with a small category of category data to form a plurality of training subsets; extracting and selecting core features of each training subset, and establishing a sub-classifier based on a rule according to the core features, wherein the feature selection process is realized in a clustering way in order to solve the imbalance problem of the data scale of the intra-category sub-concepts so as to guarantee the diversity of the core features; and solving the weight of a classifying effect of data in the training set by utilizing the sub-classifier on the basis of each sub-classifier, and establishing an integrated classifier. The classifier can solve the multi-variable time sequence classifying problem of the imbalance data set, the accuracy is relatively high, and the earliness degree is good.

The invention relates to the field of bioinformatics and data mining, in particular to a hierarchical multi-label classification method for protein function prediction, and aims at solving the data set imbalance problem, multi-label problem and hierarchical constraint problem when the conventional classification methods are used for predicting protein functions. The method comprises the following steps of: 1, a training stage: training a data set of each node in a class label hierarchical structure by adopting an SVM classifier in the training stage so as to obtain a group of basic classifiers; and 2, a prediction stage: firstly obtaining preliminary results of unknown samples by using the group of basic classifiers obtained in the training stage in the prediction stage, and processing the results by adopting a TPR algorithm with a weight so as to obtain a final result which satisfies a hierarchical constraint condition and realize the prediction of the protein functions. The hierarchical multi-label classification method for protein function prediction is applied to the field of bioinformatics and data mining.

The invention discloses a method and a device for controlling three-phase unbalanced current of a virtual synchronous generator. The method comprises the following steps: establishing a rotor movement equation of the virtual synchronous generator (VSG) and a virtual excitation equation of the VSG; decomposing unbalanced voltage of a power grid into positive sequence voltage, negative sequence voltage, zero sequence voltage and corresponding phase-sequence current, and decomposing active power and reactive power output by the VSG to analyze an encountered imbalance problem; calculating a reference instruction of dq axis current according to the output three-phase voltage signal; and converting the reference instruction of the current into a pulse-width modulation (PWM) voltage modulation signal through a self-designed current inner ring, and controlling the virtual synchronous generator to run through the PWM voltage modulation signal, so as to enable the output three-phase current of the virtual synchronous generator to reach the balance. According to the control method disclosed by the embodiment of the invention, the VSG can still output three-phase balanced current when the three-phase voltage of the power grid is unbalanced, so that the operation safety of the power grid can be relatively well ensured; and the method and the device are safe, reliable, simple and convenient.

The invention relates to a coarse-grained emotion analysis method based on a hierarchical BERT neural network, and belongs to the technical field of Web mining and intelligent information processing.The method comprises the following steps: corpus acquisition: acquiring a corpus of coarse-grained sentiment analysis; corpus preprocessing, wherein character cleaning, subordinate clause segmentationand subordinate clause vector construction are included; constructing sentence vectors: calculating the subordinate clause vectors by utilizing a bidirectional long and short term memory network, a multi-layer perceptron and an attention mechanism to generate sentence vectors; gradient coordination mechanism optimizing: introducing the gradient coordination mechanism to solve the problem of datatype imbalance in coarse-grained sentiment analysis; and carrying out coarse-grained sentiment analysis by adopting a hierarchical BERT neural network. Compared with the prior art, t the sentence vectors containing deep semantic information are constructed for the comment text through the hierarchical BERT neural network, the accuracy of coarse-grained emotion analysis tasks is improved, and the method has a wide application prospect in the fields of information recommendation, public opinion monitoring and the like.

The invention provides a contactless voltage stabilizer to improve voltage quality for users and relates to the electric field, in particular to the field of voltage quality optimizing devices. Existing voltage stabilizers have the disadvantages that a carbon brush is prone to abrasion and needs to be replaced frequently; reaction speed is slow, and using safety of equipment is affected; three-phase integrated adjustment is adopted, and three-phase equilibrium can not be achieved. According to the contactless voltage stabilizer, silicon controlled rectifier gear shift is adopted, and a power electronic device is adopted to replace a mechanical part; carbon brush abrasion and ultrahigh or ultralow voltage at the starting moment are completely avoided structurally; a single-chip microcomputer serves as a main control chip to control a silicon controlled rectifier to be switched on and off, reaction speed is high, efficiency is high, and safety performance is high; a three-phase independent adjustment structure is adopted, and voltage of a certain phase can be adjusted independently, so that output disequilibrium caused by input disequilibrium is avoided.

A binocular display deals with an optical contrast imbalance problem between display screens manifest over a range of interpupillary distances by orienting contrast asymmetries between the display screens in opposite directions.

The invention discloses a VIENNA rectifier neutral point potential alternating-direct-current component balance control method which can reduce design capacity of a capacitor on a direct-current side and a filter on an alternating-current side and save cost of a tri-level VIENNA rectifier. On the basis of common mode component injection modulation, a common mode component direct-current compensation coefficient and an alternating-current compensation coefficient are modified dynamically, so that the problem of neutral point unbalance of direct-current and alternating-current components is solved; more importantly, the method guarantees uniformity between triphase alternating-current side harmonic distortion rate and SVPWM while substantially inhibiting neutral point potential low-frequency oscillation in conventional modulation methods and improving quality of output power and can be popularized and applied in the field of battery testing, charging facilities and driving systems.

The invention relates to a solar and ground source heat pump combined energy supply system and an operation control method thereof. The solar and ground source heat pump combined energy supply system is technically characterized by comprising a solar heat collecting system, a ground source heat pump system and a buried pipe system and being composed of plate heat exchangers, a pump set and switches. The operation control method of the solar and ground source heat pump combined energy supply system includes the steps of calculating equipment parameters of the solar and ground source heat pump combined energy supply system to construct the solar and ground source heat pump combined energy supply system, and carrying out operation control over the heat collecting process, the heat storing process and the heat supplementing process of the solar and ground source heat pump combined energy supply system. Hot water supply requiring time periods of domestic hot water requiring users are analyzed, and by controlling relevant switches and equipment, solar energy and a ground source heat pump are combined to supply energy; and heat in a solar heat collector can be conveyed into soil through the buried pipe heat exchanger system to conduct the heat supplementing process on the soil during time periods when the domestic hot water requiring users do not need hot water supply, and thus the soil thermal imbalance problem caused by long-term operation of an existing ground source heat pump is solved.

A method and system for stabilizing energy consumption in multiple loads, or in single multi-phase loads. The method and system also compensates for unbalance in multi-phase loads. A central controller monitors variable reactances in the loads and identifies situations of power and / or current fluctuation and / or unbalance. It determines appropriate corrective action by the other loads / phases to compensate for the power and / or current change or unbalance due to the problematic load, and it issues control signals instructing variable reactor controllers associated with the other loads to adjust accordingly. The method and system may by applied to electric arc furnace installations. In this context, the method and system provide an electrode position controller coupled to the feed rate controller so as to predictively anticipate the introduction of new source material and lower the electrodes so as to prevent arc extinguishment while the variable reactors maintain predetermined power set-points. The electrode position controller may be used in place of the variable reactance control system to take corrective action to address power and / or current changes or unbalances.

The invention discloses a new method of gas path fault diagnosis of aeroengine based on hybrid particle swarm optimization twin support vector machine (SVM) algorithm. In view of the fact that the gas path fault is the most frequent in the whole aeroengine failure and the strong demand for the intelligent diagnosis method in the field, and TWSVM has the advantages of faster theoretical calculation and better response to the sample imbalance problem, and adopts the TWSVM algorithm to study aeroengine gas path fault diagnosis. In this paper, the hybrid kernel function is introduced to improve the performance of the kernel function, so that the TWSVM algorithm can better balance the generalization ability and good learning ability. In addition, the optimal parameters of TWSVM are optimized by using HPSO, and the fault classification model is obtained. The aeroengine gas path fault diagnosis is realized with high precision.

The invention discloses a self-adaptive oversampling method based on HDBSCAN clustering, and mainly solves the problem of unbalanced data classification by using complete data information in an existing method. The technology comprises the following steps: (1) inputting a training data set; (2) clustering the minority class samples in the training set to obtain different scales of clusters which are not intersected with each other; (3) calculating the number of samples needing to be synthesized in each minority class cluster; (4) adaptively synthesizing new samples according to the number of samples needing to be synthesized by each cluster to obtain a new minority class data set; (5) forming a new balanced data set by the majority class data set and the new minority class data set; and (6) training and testing the classifier by using the new balance data set. According to the technology, noise in an unbalanced data set can be effectively prevented from being generated, meanwhile, theproblem of inter-class and intra-class unbalance is solved, and a brand-new oversampling strategy is provided for unbalanced learning.

The present invention discloses a public bicycle dynamic scheduling method based on an improved ant colony algorithm. The method comprises the following steps: S1, selecting a research range, and collecting basic data; S2, determining a scheduling area and a scheduling amount; S3, establishing a public bicycle dynamic scheduling path optimization model; S4, performing solution through adoption of a method, employing an improved ant colony algorithm to solve a TSPPD model; and S5, formulating a scheduling scheme, calculating the scheduling demand of each rental station in the scheduling area, employing the public bicycle dynamic scheduling path optimization model to perform modeling of the dynamic scheduling problem, performing solution through adoption of the improved ant colony algorithm, and formulating a public bicycle peak hour dynamic scheduling scheme. The unbalance supply and demand problem of peak hour public bicycles can be alleviated.

The invention relates to the technical field of a light emitting diode, and specifically provides a quantum dot light emitting diode and a preparation method thereof. The quantum dot light emitting diode comprises a glass substrate, and a first conductive layer, an electron injection layer, a quantum dot light emitting layer, a hole transport layer, a hole injection layer and a second conductive layer which are stacked on the glass substrate in sequence; and an electron transfer passivation layer is also arranged between the quantum dot light emitting layer and the electron injection layer. Byadding the electron transfer passivation layer between the quantum dot light emitting layer and the electron injection layer, the problem of imbalance current carrier mobility efficiency between theelectron injection layer and the hole transport layer can be solved, the current efficiency of the light emitting device is improved effectively, and the current efficiency of the light emitting device achieves the optimal state.

The invention provides an electromagnetic mixed compensation system for quality of electric energy in electrified railways. The electromagnetic mixed compensation system comprises an RPC, two MSVCs and a V / V traction transformer. The V / V traction transformer is in a V / V type formed by connecting two single-phase transformers. The primary side of the V / V traction transformer is connected to a three-phase public power grid, and traction power supply arms serve as two output ends of the secondary side of the V / V traction transformer. Each MSVC comprises a magnetic controlled reactor, a third single harmonic filter branch circuit and a fifth single harmonic filter branch circuit, and the magnetic controlled reactor, the third single harmonic filter branch circuit and the fifth single harmonic filter branch circuit are connected in parallel. The RPC comprises two H-bridge converters with a shared direct-current side. The MSVCs and the RPC are used in cooperation with each other, and therefore the problem of unbalanced harmonic and negative sequences of the electrified railways is solved. Due to the adoption of MSVC and RPC mixed compensation, compensation capacity of the RPC is effectively reduced, and the construction cost of the system is accordingly reduced.

The invention provides a correction method for a zero intermediate frequency signal and a digital pre-distortion treatment device in digital pre-distortion treatment. The method comprises the following steps: converting the feedback signal of a zero intermediate frequency service signal into the zero intermediate frequency feedback signal synchronous with a transmission signal; and according to the zero intermediate frequency feedback signal, measuring and correcting the zero intermediate frequency error. The method and the device provided by the invention use an arbitrary service signal to measure and correct the problems of imbalance and direct current leakage of ZIF without sending special training sequence, combine ZIF technology with DPD technology and lower the complexity of realizing ZIF correction.

A Chip-On-Lead (COL) type semiconductor package having small chip hidden between leads is revealed. The lower surfaces of the leadframe's leads are attached to a wiring substrate and the leads are horizontally bent to form a die-holding cavity. A smaller chip is disposed on the wiring substrate by passing through the die-holding cavity to be on the same disposing level with the leads. At least a larger chip is disposed on the leads to overlap the smaller chip so that the small chip does not extrude from the leads. The encapsulant encapsulates a plurality of internal parts of the leads, the wiring substrate, and the larger chip. Therefore, the conventional unbalance issue of mold flow above and below the leads leading to cause excessive warpage can be avoided and numbers of stacked larger chips can be increased to have larger memory capacities.

Method and system is described for reducing sensitivity imbalance issues and / or implements target-oriented tomography to enhance tomographic inversion for velocity model building. The method may include performing a preparation stage to construct a measurement vector from seismic data and a kernel matrix from ray-path information; performing a sensitivity optimization stage to generate a data weighting vector; and performing a property optimization stage to reconstruct a subsurface model of one or more geophysical properties.

The invention belongs to the sub-field of the learning theory and application in the technical field of computer application, and relates to a CCA and 2PKNN based automatic image annotation method, in order to solve problems of a semantic gap, a weak annotation, and category imbalance that exist in an automatic image annotation task. The method comprises: firstly, for a semantic gap problem, mapping two features to a CCA sub-space, and solving a distance between the two features in the sub-space; for a weak annotation problem, establishing a semantic space for each annotation; for a category imbalance problem, by combining a KNN algorithm, finding out k nearest neighbors of a test image in the semantic space of each annotation, constituting the k nearest neighbors to an image sub-set, and by using a visual distance between the sub-set and the test image, and by combining a Bayesian formula, assigning a few annotations with the highest score to the test image; and finally, optimizing an image annotation result by using correlation between annotations. The method disclosed by the present invention has a greater degree of improvement for image annotation performance.

The invention relates to an energy efficiency management system based on an intelligent transformer area. The energy efficiency management system comprises a system master station, a transformer area sub meter, an intelligent centralized control device and an intelligent interactive terminal, wherein the intelligent centralized control device is installed in a transformer area; the intelligent interactive terminal is installed on a user side; the transformer area sub meter is an intelligent ammeter provided with a ZIGBEE transmission module; the intelligent centralized control device comprises an ammeter concentrator, an interactive terminal concentrator, transformer area wireless communication equipment, reactive-load compensation equipment, intelligent communication equipment and a distribution transformer monitoring terminal; compensation modes can be adaptively selected by an SVG (Scalable Vector Graphics) digital control system, so that a main circuit can adapt to external connection. According to the energy efficiency management system, an information technique, an electronic technique and an automated technique are applied to establish a stable and reliable communication network and an advanced master station and background integrated management system. By using the energy efficiency management system, the intelligent level of a power supply area is improved, and online monitoring of an operating state of the transformer area is realized; the long-standing problems of harmonic suppression and three-phase imbalance of a rural power network are solved; the loss of the transformer area is reduced; the user initiative is mobilized, and energy conservation and loss reduction are promoted.

The invention discloses an MMC-UPFC-based three-phase imbalance governance system and method of a power transmission line. The system comprises a modular multilevel-unified power flow controller MMC-UPFC, wherein the MMC-UPFC comprises a parallel side converter and a series side converter which are in back-to-back connection and share a same DC bus; the parallel side converter is connected with an AC side of a power grid through a parallel transformer; and the series side converter is connected with a power transmission line side through a series transformer. The system and the method disclosed by the invention have the beneficial effects that control on power flow of the power transmission line can be achieved while a three-phase imbalance problem of the power transmission line is solved, the problems of a light load and an overload of the power transmission line can be solved; the application function is expanded; and safe, economic and stable operation of a power system is ensured.

The invention discloses a medical image segmentation method based on a one-way multi-task convolution neural network. Firstly, the segmentation task is divided into three different but related sub-tasks by using the network. A first sub-task carries out rough segmentation on that image to determine a rough area of the whole organization; a second subtask dilates and fine-segments the area to getthe exact tissue type and exact position of the whole tissue. Subtask 3 uses refined segmentation to scan voxels in the intact tissue and determine the location of the enhanced tissue. Then the threesub-tasks are trained in stages with course learning strategy from easy to difficult to realize data and parameter sharing among sub-tasks. Finally, the post-processing method is used to improve the classification error problem and output the final segmentation results. The invention synthesizes the associated sub-tasks into a single network for simultaneous training, improves the shortcomings ofcascade model training sub-tasks one by one and better solves the category imbalance problem, realizes the segmentation of the image from coarse to fine by one-way operation, and improves the segmentation effect while reducing the complexity of the system.

The invention relates to a method for solving three-phase voltage imbalance by using an SVG (Static Var Generator). The method comprises the steps of sampling, filtering, phase correcting, proportional amplifying and operational processing. A comprehensive control quantity obtained by operating and processing is converted into an IGBT (Insulated Gate Bipolar Translator) triggering time through a function; a controller is used for controlling the IGBT triggering time by taking a PT (Potential Transformer) signal as a synchronous signal; therefore, an all-controlled converter in the SVG generates a voltage signal which is same as the three-phase difference value of a power grid and opposite in direction to offset unbalance voltage in the power grid; and a purpose of balancing the three-phase voltage is achieved. According to the method, by adopting the all-controlled converter, the three-phase voltage imbalance problem can be solved effectively; the response speed is rapid; the tracking precision is high; and the stability and the reliability of the power grid can be greatly improved.

The invention provides a bullet screen text classification method, a bullet screen text classification device, equipment and a storage medium. The method comprises the steps: obtaining an imbalance training data set with a pre-marked category, and dividing the training data set into a sufficient sample and an insufficient sample; training the sufficient samples by adopting a textCNN model; carrying out model training on the insufficient samples by adopting an SVM classifier; inputting a text to be tested into the trained textCNN model, and outputting classification probabilities of various categories in sufficient samples; and if the output classification probability is smaller than a first preset threshold, inputting the to-be-tested text into a trained SVM classifier, and outputting a predicted category. According to the method, the classification models for different text scales are obtained through separate training according to the sizes of the training samples, then the two classification models are combined to be used for classifying the to-be-detected text, the problem of data imbalance of the training samples is solved, compared with single model training, the risk of over-fitting can be reduced, bias is reduced, and the recognition accuracy is higher.

The invention discloses a face attribute recognition method based on a multi-task multi-label learning convolutional neural network, and relates to a computer vision technology. Firstly, multi-task learning is adopted, and two tasks of face key point detection and face attribute recognition are learned at the same time; different learning difficulties and learning convergence rates of different attributes are considered, the attributes are divided into subjective attributes and objective attributes, and the convergence rate of the network is accelerated and the problem of sample imbalance is relieved by adopting a dynamic weight and an adaptive threshold strategy; and finally, according to the trained network model, face attribute recognition results of the subjective attribute and objective attribute sub-networks are taken as final face attribute recognition results. A dynamic weight scheme and self-adaptive threshold adjustment are used, so that the network convergence speed is increased, and meanwhile, the label imbalance problem can be relieved; three different sub-networks are trained by adopting a spatial pyramid pooling method, so that end-to-end training is achieved to perform multi-task multi-face attribute recognition. And the precision of face attribute recognition, especially subjective attributes with high difficulty, is improved.

The present invention discloses a control method for a soft switch circuit in a switch power source, which generates an alternating primary power filter current by controlling first and second primary power switching devices to be closed and opened, and generates an intermittent alternating resonant current in the same direction as the primary power filter current in a resonant branch by controlling forward and backward auxiliary switching devices to be closed and opened to thereby achieve closing of the first and second primary power switching devices at a zero voltage, and which generates a balance current with the same magnitude as and in the opposite direction to the resonant current in the resonant branch in at least a period of time during the resting of the resonant current by further controlling the forward and backward auxiliary switching devices to be closed and opened to thereby achieve an average current of zero across the resonant branch in a switching cycle. Without any additional balance circuit, this control method can address the imbalance problem of output power of the positive and negative direct current input voltage sources in a soft switch circuit of an existing switch power source.

A binocular display deals with an optical contrast imbalance problem between display screens manifest over a range of interpupillary distances by orienting contrast asymmetries between the display screens in opposite directions.

The invention discloses a device with a heating and water-heating device multi-machine parallel system and a use method thereof. The device comprises a heating system, and a water heating circulatory system and a heating circulatory system which are connected with the heating system, wherein the heating system is formed through the parallel connection of a plurality of gas wall-mounted fast heating and water-heating devices; the water heating circulatory system and the heating circulatory system are independent of each other and commonly use the same heating system for heating; and a central controller is used for controlling the heating system, the water heating circulatory system and the heating circulatory system which are connected. The device with the heating and water-heating device multi-machine parallel system, disclosed by the invention, has the characteristics of large load, low noise and low power consumption; and through the design that the water heating circulatory system and the heating circulatory system are independent of each other, the phenomenon that the heat supply is stopped for a long time in the traditional bathing-priority mode is avoided, and the problem of hydraulic imbalance between a heat supply system and the multi-machine parallel system is solved. The invention further discloses the use method of the device.