188results about How to "Get rid of dependency" patented technology

This invention relates to permanent simultaneous motor control system based on NN with five free degree and with no bearing, which is characterized by the following: one two-free degree with no bearing simultaneous motor, one three-free degree alternating radium and axis direction mixed magnetic bearing form one sample body; three Clark anti-conversion, three current tracing anti-converter and one linear power amplifier, sample machine and its load work as one integration to form compound controlled subject; using static NN summer through study formula to form NN reverse for non-linear dynamic decoupling control on five free degrees bearing.

The invention provides an intelligent personal medical treatment archival system and belongs to the technical field of medical treatment software development. Medical treatment information interaction between patients and hospitals, medical information sharing of the hospitals and reasonable allocation and effective utilization of medical treatment resources are achieved, and the problem of medical treatment resource waste is solved. Examination results and patient personal information are stored, then are analyzed and integrated along with experiment value examination and image data obtained during the non-stay in hospital through mobile phone software, the recent detection information is compared and used for drawing a curve, and various indexes are monitored and stored in a mobile phone or a network cloud. According to abnormal indexes, a treatment hospital is automatically recommended to a patient, and the patient makes an appointment according to prompts of the registration program or selects an appropriate hospital for registration by himself. When severely abnormal data are monitored, signals are sent to a diagnosis and treat center to ask for help. A doctor obtains the personnel medical treatment archives of the patient by using the network cloud or directly using a mobile phone software terminal with the permission of the patient. The patient can know the body changes of himself immediately, and is guided to seek medical advices.

The invention discloses a mechanical equipment fault intelligent diagnosis method based on a generation model under small sample data. The mechanical equipment fault intelligent diagnosis method comprises: carrying out zero-mean-value standardization preprocessing on a small number of obtained mechanical signals; Establishing a composite network for mechanical signal generation; Training a generative adversarial network model in an adversarial manner in combination with a Wasserstein distance and a gradient punishment method; Establishing a deep convolutional neural network model for classifying the operation states of the mechanical equipment by using the mechanical signals; And by combining the generative adversarial composite neural network model and the deep convolutional neural network model, training two networks by using a small amount of real mechanical signals, and finally realizing intelligent fault diagnosis of the mechanical equipment under small sample data. The method hasthe advantages of being good in mechanical signal feature extraction effect, high in state classification accuracy and good in mechanical signal data expansion performance.

The invention provides a diabetic retinopathy grade classification method based on deep learning. The diabetic retinopathy grade classification method comprises the steps of: constructing a sample library; removing backgrounds and noise of ophthalmoscope photographs in the sample library; normalizing the images of different brightness and different intensity to the same range by adopting a local mean value subtracting method; adopting random stretching and rotating methods for different samples for data augmentation, and constructing a training set and a test set; training an initial deep learning network model by establishing an input portion architecture, a multi-branch feature transformation portion architecture and an output portion architecture separately; and inputting samples to betested into the trained initial deep learning network model for diabetic retinopathy grade classification. Compared with the traditional processing method, the diabetic retinopathy grade classification method gets rid of the dependence on prior knowledge, and has good generalization ability; and by adopting the designed multiple grades, a small-sized convolution kernel can be used for extracting very tiny lesion features, thereby making the classification results more reliable.

The invention provides an alarm processing analysis system and method of a broadcasting and TV access network. The system comprises a collecting module used for collecting various warning messages sent by network equipment, an analysis module used for analyzing the relevance of various received equipment warning messages according to a correlation model established through data mining in advance between different devices and different warning types, and a judging model used for further judging the relevance types of the warning devices according to an expert knowledge base which is established in advance to obtain the specific relevance type between different devices and different warning types and to find out a root alarm. By means of the alarm processing analysis system and method of the broadcasting and TV access network, the dependency of original warning relevance analysis on TOPO is removed, meanwhile, the expert knowledge base is used for overcoming the defects of data mining analysis warning relevance, the warning real-time performance processing demand can be met, and the analysis accuracy requirement can be met as well.

The invention discloses a method for upgrading a terminal through a wireless local area network (WIFI), and the method comprises the following steps: connecting a WIFI terminal with an upgrading server through the WIFI; and downloading contents which need to be upgraded into a cache, carrying out software upgrading after completing the download. The invention further discloses a system for upgrading the terminal through the WIFI, and the adoption of the method and the system can get rid of the dependence on a data download line of a universal serial bus (USB) during the process of upgrading the WIFI terminal, realize automatic upgrading and simultaneously carry out rapid batch upgrading.

The invention discloses a facial expression identification method and relates to a method for identifying images. The facial expression identification method is a facial expression identification method which extracts expression textural characteristics using a center symmetrical ternary patterns (CSTP) algorithm. The method comprises steps of preprocessing a facial expression image; extracting facial expression textural characteristics on each sub-block of the facial expression image; and determining final facial expression textural characteristics of the facial expression image. Therefore, the facial expression identification is completed. The method overcomes problems that texture description is not fine due to a complex identification background and the identification rate is not high in the prior art.

The invention discloses a lithium-ion battery electrolyte taking LiBF2SO4 as basic lithium salt. The electrolyte comprises the following components: electrolyte lithium salt and an aprotic solvent, wherein the electrolyte lithium salt can be unitary LiBF2SO4 and also can be a mixture of LiBF2SO4 and other common lithium salts; the aprotic solvent is one or a mixture of some of common carbonic ester, sulphite, sulfone compounds and the like. Through the lithium-ion battery electrolyte, disclosed by the invention, the compatibility between the electrolyte and the electrode materials can be improved; meanwhile, the rate performance, the high-temperature performance and the cycle performance of the lithium-ion battery can be effectively improved.

The invention discloses a flight path prediction method. The method comprises the steps: considering the meteorological environment of a flight, extracting meteorological features through a convolutional neural network, modeling flight path features through a recurrent neural network, and finally outputting a series of future flight paths, specifically including the steps of flight path data preprocessing, meteorological feature extraction, time sequence model definition, model compiling and training, and flight path prediction. Through the technical scheme of the invention, the long-term flight trajectory can be accurately predicted in an actual environment.

The invention relates to a fuel cell vehicle energy management method based on a deep reinforcement learning algorithm, and belongs to the field of new energy vehicles. The method comprises the following steps: S1, acquiring state information of a fuel cell vehicle; s2, building a fuel cell vehicle energy management system model; and S3, constructing a fuel cell vehicle energy management strategyby using a deep reinforcement learning algorithm, and solving a multi-objective optimization problem including fuel economy and fuel cell efficiency, thereby obtaining an optimal energy distribution result. According to the invention, the deep reinforcement learning algorithm is applied to the energy management system of the fuel cell vehicle, so that the optimization and real-time performance aregood; meanwhile, the working efficiency of the fuel cell is considered in the reward function, and a new thought is provided for energy management.

The invention provides a spatial pose calibration method for multi-line laser sensor combined measurement, and belongs to the technical field of measurement. Firstly, a linear laser sensor is fixed ona multi-degree-of-freedom adjustment displacement table, and then a stepped calibration block and a calibration block are placed on a motion shaft of a measurement platform; measuring different positions of the calibration block by using a linear laser displacement sensor to obtain a deflection angle of a laser plane; adjusting the positions of the plurality of line laser sensors in the measurement sensitive direction through a standard cone; and measuring a plurality of feature points on the calibration block, establishing a matrix equation of coordinate system space transformation accordingto the measurement base coordinates of the feature points and the sensor coordinates, and solving an optimal solution based on a least square method to complete calibration. The sensors can completeangle deflection, position alignment and multi-sensor calibration of a laser plane under one-time clamping, and the dependence on auxiliary calibration of a numerical control machine tool and an industrial robot is eliminated; the calibration process is quick, efficient, high in practical value, convenient to operate and high in calibration accuracy.

The invention discloses a power distribution network high fault-tolerance on-line fault locating method based on fault auxiliary factors, comprising the steps of: utilizing a main station to collect the current spillage information of each feeder line switch of a power distribution network; setting a switch function set; building a power distribution network fault locating non-linear complementation optimization fault locating model; building a non-linear programming fault locating model meeting KKT extremum conditions; and utilizing a lagrangian multiplier and a continuous space non-linear optimization extremum theorem to obtain fault auxiliary fault factors, and realizing a fault locating and FTU defect diagnose fault locating non-linear equation set model. The method realizes single or multiple fault high fault-tolerance location on a power distribution network feeder line fault section, realizes latent defect FTU apparatus position accurate identification and large scale complex power distribution network on-line fault positioning, provides theoretical guidance for FTU apparatus state maintenance, and has the advantages of convenient implementation, high reliability, strong fault tolerance capability, high fault locating efficiency, great adaptability to multiple faults, etc.

The invention discloses an intelligent fault diagnosis method under a small sample based on an attention mechanism element learning model. According to the intelligent fault diagnosis method, an attention mechanism and a meta-learning method are used for establishing an association network model; short-time Fourier transform is carried out on mechanical signals to obtain a time-frequency spectrogram of the mechanical signals; feature extraction and operation state recognition are further carried out on the time-frequency spectrogram; and rich fault information contained in the mechanical signals can be effectively mined. According to the intelligent fault diagnosis method, a pseudo distance can be trained adaptively to evaluate the similarity between related data; clear mathematical formula definition is not needed; and high mechanical equipment fault diagnosis accuracy can be obtained. Therefore, the dependence of a feature extraction process on artificial experience and the dependence of an existing intelligent fault diagnosis algorithm on a large amount of training data in a traditional diagnosis method are eliminated, and the problem of mechanical equipment fault diagnosis under the condition of small sample data is practically solved.

The invention discloses a sewage treatment process for realizing sludge reduction by taking elemental sulfur as the medium. The sewage treatment process comprises the following steps: allowing the sewage to be treated to flow into a sulfur reduction reactor for anaerobic biological treatment, removing the organic matters in the sewage during the sulfur reduction process to produce a minus bivalence sulfide, introducing the minus bivalence sulfide into a sulfur oxidation reactor inoculated with sulfur oxidizing bacteria for reoxidizing the minus bivalence sulfide into elemental sulfur, and then allowing the elemental sulfur to reversely flow to the sulfur reduction reactor. The sulfur reducing bacteria are first applied to the field of water treatment to form various sewage treatment processes with less sludge production, thereby realizing the innovation of the sewage treatment method; in the traditional anaerobic sulfate sewage treatment process, the sulfate is used as the electron acceptor during the COD (chemical oxygen demand) mineralization, however, the process utilizes the elemental sulfur for replacing the sulfate and sulfite; and in the whole process, the equipment is simple, the energy consumption is low, and the cost is saved.

The invention relates to a manufacture method of an environment-friendly lignin composite material. A thermal pressing forming treatment technology is adopted to prepare the environment-friendly lignin composite material by using forest remains and rejected wood materials as main raw materials; after the and pretreating, adding stuffing, such as a larch extract, lignin in papermaking waste liquid, and the like; adding biological degradable resin-polylactic acid; and assembling and combining blanks. The invention is free from reliance on a formaldehyde-contained cementing compound, not only improves the plate-forming performance, but also is biologically degradable after being used by adding the degradable resin, thereby not causing environmental pollution and partially substituting the prior petroleum source products. The preparation method of the invention is simple and suitable for industrialized production. The products are provided with multifunction lignin forming products with high strength ratio, environment-friendly types, natural degradation, and the like, can be applied to the fields, such as building industry, interior decoration engineering, electric appliances, manufacturing industry, and the like and can substitute qualitative material products with similar performance.

The invention discloses a cross-scene target automatic identification and tracking method and the application thereof. The method comprises the following steps: a tracking target is tracked in a monitoring scene; when the tracking target is lost, the extraction model detects the tracked actor; the Euclidean distance between the actor sequence feature and the tracking target feature is calculated, and doer re-identification is carried out; the actor detection and actor re-identification is performed on multiple paths of scenes, and cross-scene actor identification and tracking is performed; the FACENET convolutional network face recognition technology is used for recognizing the actor, and identity recognition of the tracking target is carried out; the crowd density is estimated by adopting CSRnet, computing resources are allocated according to a crowd density threshold value, and early warning is performed on the monitoring area exceeding the set threshold value; key targets are searched and matched in all video monitoring scenes according to the retrieval conditions, and all targets meeting the conditions are selected. According to the invention, the generalization ability of the actor re-identification model in different scenes is improved.

The invention discloses a method for observing a left inverse state of a neural network of a permanent magnet synchronous motor, and belongs to the technical field of motor drive control. An observer of the left inverse state of the neural network is constructed through the method for observing the left inverse state of the neural network to achieve the state observation of a multivariable and strong-coupled time-varying nonlinear system of a permanent magnet synchronous motor system without a sensor. According to the method for observing the left inverse state of the neural network of the permanent magnet synchronous motor, the observation problem of a complex system mutually coupled by the current, the voltage and the speed of a stator is translated into a simple observation problem of a unit mapping system, therefore, high-precision observation of the rotate speed of the permanent magnet synchronous motor and the angular position of a rotor is achieved, and the interference of system noise is overcome. An observation system is made to have excellent dynamic performance, static performance and anti-interference and high-precision tracking performance.

The invention relates to a triple point recurrence device with a refrigerating machine serving as a cooling source. The triple point recurrence device with the refrigerating machine serving as the cooling source comprises a measuring room and the refrigerating machine. The measuring room is used for offering measuring spaces of a thermometer to be measured, a standard thermometer and a triple point bottle. The refrigerating machine offers cooling capacity for the triple point recurrence device. In the triple point recurrence device, heat balance among a cooling head, a temperature control copper block and a heat evening block, so that materials in the at least one triple point bottle are in a triple point state and calibration of temperature is conducted. According to the triple point recurrence device with the refrigerating machine serving as the cooling source, operation of the refrigerating machine only needs maintenance of water and electricity, so that operation cost is low; the refrigerating machine serves as the cooling source so that dependence on liquid helium and supply limitation of the liquid helium can be eliminated. The triple point recurrence device with the refrigerating machine serving as the cooling source is compact in structure, easy to move and capable of conducting low temperature calibration tests at any time anywhere, and has notable advantages.

The invention discloses a construction method for a neural network generalized inverse decoupling controller of a bearing-free synchronous reluctance motor, which comprises the steps: taking two Park inverse converting type inverters, two Clark inverse converting type inverters and two direct current tracking type inverters as a wholly-formed composite controlled object after the two Park inverse converting type inverters, the two Clark inverse converting type inverters and the two direct current tracking type inverters are respectively and sequentially connected with one another in series and before the two Park inverse converting type inverters, the two Clark inverse converting type inverters and the two direct current tracking type inverters are connected with the bearing-free synchronous reluctance motor; forming a generalized imitative linear system before a constructed neural network generalized inverse is connected with the composite controlled object in series, and forming a linear closed loop controller by two position controllers and a speed controller; and jointly forming the neural network generalized inverse decoupling controller by the means that the linear closed loop controller, the neural network generalized inverse, the two Park inverse converting type inverters, the two Clark inverse converting type inverters and the two direct current tracking type inverters are respectively and sequentially connected with one another in series. The independent decoupling control between the electromagnetic torque and the radial levitation force and the independent decoupling control of the radial levitation force between two components on the vertical direction are realized according to the closed ring control and the PID (proportion integration differentiation) parameter adjustment, and the control performance of the bearing-free synchronous reluctance motor is obviously improved.

The invention discloses a constructing method for a decoupling controller of a support vector machine inverter of a bearingless asynchronous motor. The support vector machine inverter is constructed by a support vector machine with 10 input nodes and 4 output nodes and six integrators s-1 and is arranged in front of a complex controlled object to form a pseudo-linear system, and two corresponding displacement regulators, one speed regulator and one flux linkage regulator are respectively designed for integral type pseudo-linear subsystems; a linear closed-loop controller is constructed by four regulators; and the decoupling controller is constructed by the linear closed-loop controller, the support vector machine inverter, two Clark inverse transformers and two current tracking inverters. According to the constructing method, control on a multi-variable, nonlinear and strong-coupling time-changing system is converted into control on two displacement second-order integral linear subsystems, one speed first-order integral linear subsystem and one flux linkage first-order integral linear subsystem, and dynamic decoupling of the radial force, the speed and the flux linkage can be realized.