75results about How to "Avoid simplification" patented technology

The invention discloses an image domain conversion network based on generative adversarial networks (GAN) and a conversion method. The image domain conversion network includes a U-shaped generative network, an authenticity authentication network and a pairing authentication network. An image domain conversion process mainly includes the following steps: 1) training the U-shaped generative network,and establishing a network model of the U-shaped generative network; and 2) normalizing a to-be-converted image, and then input the image into the network model established through the step 1) to complete image domain conversion of the to-be-converted image. According to the image domain conversion network, an image domain conversion task of a local region in the image can be achieved, image local-domain conversion quality is high, judgment ability of the network is high, stability of image conversion is high, and authenticity of a generated image is greatly improved.

The invention discloses a near spacecraft fault diagnosis and fault-tolerant control method. Firstly, a dynamical model of an attitude control system of a reentry phase of a near spacecraft is established and the dynamical model of the attitude control system is transferred to a nonlinear equation in a general type through coordinate transformation; secondly, a fault model with actuator invalidation and actuator dynamic damage existing simultaneously is established; thirdly, real-time online estimation of self-adaption estimation observer to actuator dynamic damage factors and actuator invalidation factors is respectively established; and finally a sliding module fault-tolerant controller is designed through the estimated actuator dynamic damage factors and the estimated actuator invalidation factors and parameters of the controller is updated in real time. According to the near spacecraft fault diagnosis and fault-tolerant control method, in the condition that when the actuator dynamic damage and the actuator invalidation occur simultaneously in a flight control system, fault information can be estimated rapidly and accurately, strong fault-tolerant capacity is achieved and an expected requirement of the flight control system is reached.

The invention discloses a method for dynamically grading major hazard sources of liquid chemicals in a storage tank region of a petrochemical wharf. The method comprises the following steps: creating an accident consequence model of the major hazard sources of the liquid chemicals in the storage tank region of the petrochemical wharf and calculating a major hazard source injury index; taking the major hazard source injury index as initial data for grading the major hazard sources and performing standardization calculation; initially grading the major hazard sources of the liquid chemicals; continuously adjusting the grades of all the hazard sources on the basis of initial grading by a cluster analysis principle and outputting a grading result until reasonable grading. The method has the effects of accurately reflecting a relative risk relationship among various hazard sources when the storage capacity is changed, effectively solving the current situations of simple and fixed grading standards of the major hazard sources and high subjectivity and generality of the grading result, guaranteeing the real-time variability of the grades of the major hazard sources, and setting protection grades corresponding to the conventional control measures and safety technical means of the production enterprises of the petrochemical wharf.

The invention discloses a rapid and lasting re-greening method for high and steep rocky slope vegetations. The method comprises the following steps: (1) shaping and checking a slope surface, and removing dangerous rocks; (2) paving a protective net along the whole slope surface from top to bottom, fixing the protective net on the slope surface by adopting grouting anchor rods, and arranging transverse and longitudinal ecological belts between the protective net and the slope surface; (3) casting V-shaped planting grooves in situ in the slope surface; (4) spraying a slope matrix layer onto theprotective net, fixing a coconut fiber net on the matrix layer by using U-shaped nails, and spraying and sowing an organic material mixed with grass seeds on the coconut fiber net; and (5) backfillingsoil in the V-shaped planting grooves, and planting vine plants and shrubs. Through the synergistic effect of the ecological belts, the planting grooves and the hanging net spraying and mixing technology, the defects that in a conventional greening technology, a base material is easily scoured, the greening effect is slow, greening is not uniform, and the re-greening effect is difficult to last are overcome, by arranging the planting grooves and the ecological belts, the stability of the side slope is improved, lasting nutrient supply is achieved, and rapid and lasting re-greening of high andsteep rocky slopes can be realized.

The embodiment of the invention discloses a radiation source individual identification method and a device based on a convolutional neural network, and a computer storage medium. The method can comprise the following steps: acquiring emission signals of a plurality of radiation sources which have the same model and are in the same working mode to obtain acquired emission signals; establishing a data set for training a convolutional neural network according to the acquired emission signals; constructing a convolutional neural network model; training the convolutional neural network model by using the established data set to obtain a trained convolutional neural network model; acquiring a target emission signal of any target radiation source in the plurality of radiation sources, and inputting the target emission signal to the trained convolutional neural network model by using a forward propagation algorithm, obtaining the output of the trained convolutional neural network model, and identifying a target radiation source corresponding to the target emission signal based on the output.

The invention relates to a complex network key node cluster mining method, and specifically discloses a complex network key node cluster mining method based on combination optimization, which is usedfor solving problems of simplified technical indexes, high cost needed for node mining, limited propagation range, insufficient amount of propagation information and no consideration of a strengthening function of reachable neighbors to node centrality in the existing method. The method provided by the invention has the following steps: 1, building an input network model; 2, selecting indexes forpreprocessing network nodes; 3, calculating an initial key node cluster S<2> and a to-be-optimized node cluster S<3>; 4, selecting an objective function of key node cluster combination optimization; 5, performing combination optimization; 6, outputting a key node cluster S. The method provided by the invention has the advantages of preprocessing network node groups, selecting limited nodes for performing combination optimization, being low-complexity in calculation, considering various indexes, preprocessing the network nodes and performing combination optimization for the obtained nodes, andmining the key node cluster which needs low cost for mining, has wide propagation range and large amount of propagation information, and considers the strengthening function of the reachable neighborsto node attributes.

The invention discloses a mould processing and grinding device, comprising a workbench, a lifting device is arranged on the left side of the workbench, a work box is arranged on the right side of the lifting device and above the workbench, and the work box is A motor and an electric telescopic device are respectively arranged inside, the electric telescopic device is located on the right side of the motor, the output shaft of the motor is sleeved with a first gear, the first gear is meshed with a second gear, and the first gear is engaged with the first gear. The inner ring of the second gear is sleeved with a rotating rod. The rotating rod penetrates the working box and extends to the outside of the working box. Two grinding blocks are arranged inside the protective shell. The grinding blocks penetrate through the protective shell and are fixed with the grinding plate arranged outside the protective shell. connect. The mold processing and grinding device of the invention has simple structure and convenient use, can grind the mold holes or grinding tool grooves during processing, and has better grinding effect on the burrs in the holes or grooves, so that the utilization rate of the mold is higher when it is put into use. .

The embodiment of the invention provides a network quality detection method and equipment, relates to the technical field of communication, and aims to solve the problem that the network quality detection is not comprehensive or accurate as the amount of information obtained by a detection method in the prior art is relatively small. The network quality detection method can include: a first router receives a detection task, sends a preset length of UDP (User Datagram Protocol) message to each node on a transmission route, records timestamps and the number of the sent UDP message, and receives ICMP (Internet Control Message Protocol) messages fed back by the different nodes on the transmission route, wherein the ICMP messages include MTU (Maximum Transmission Unit) values; based on the timestamps and the number of the sent UDP message, the timestamps and the number of the received ICMP messages, and the MTU values in the received ICMP messages, the network quality between the first route and the different nodes is detected.

A method for culturing the seeds of photosensitive and thermo-sensitive male sterility line of high-temp sterile or long-day high-temp sterile rice includes such steps as choosing male parent and female parent (photosensitive and thermo-sensitive sterility line), sowing, irrigating with cold water when the ambient temp is higher than the critical temp of fertility transform, hybridizing, growing and harvesting.

The invention relates to a timing reminding method of a mobile terminal and the mobile terminal. The method includes: setting timer information, wherein the setting timer information includes setting the reminding time, the reminding mode and at least one function related to a timer and storing the identification and link address of the function application to a storage area of the mobile terminal; starting the timer to monitor whether the current time reaches the preset reminding time or not; and if the monitoring result is yes, reminding a user in the preset reminding mode and meanwhile automatically starting the function application or starting a shortcut list, wherein the shortcut list provides a shortcut access of the function application related to the timer. Due to the fact that the correlation between the timing reminding function and other function applications is arranged, the current timing reminding function is expanded, the function that business can be handled more conveniently and rapidly is provided, the operation of the user is simplified, the experience of the user is improved, and the practical effect of the timing reminding of the mobile terminal is promoted.

The invention discloses a text recommendation method and device, equipment and a medium, and relates to the technical field of financial science and technology, and the method comprises the steps: monitoring an operation behavior of a target user, so as to determine a keyword associated with the target user; searching more than one updating text containing at least one keyword from a preset text database set to serve as a first candidate text; calling a preset cause map, selecting an updated text of which the total association degree with the first candidate text is not less than a preset association threshold value from a preset text database set according to the map to serve as a second candidate text, the cause map including association relationships between texts, and each associationrelationship having a corresponding association degree; and according to the operation behavior, screening out a selected text from the first candidate text and the second candidate text and recommending the selected text to the target user. The technical problems that in the existing recommendation process, recommendation is too simple, and the recommendation accuracy is low are solved.

An intelligent control method for efficient extraction of coalbed methane in goafs includes judging whether extracting the coalbed methane in each goaf or not by utilizing a mobile mining pump station or constructing a fixed downhole extraction pipe network and according to whether concentration of CH4 in the goafs, a working face and a return gasway exceeds the standard or not. In addition, the control method includes real-time monitoring of other gas compositions and gas temperature restricting extraction of the coalbed methane, and the gas compositions include ethylene, carbon monoxide and oxygen. The invention particularly provides the intelligent control method for efficient extraction of the coalbed methane in the goafs. During application and practice, the method basically solves the problems of independent operation and single extraction control parameter in each goaf, no safety control guarantee during extraction process and low extraction efficiency. Prospective objective of system optimization is achieved, and utilization rate and extraction efficiency can be brought into full play under the condition that the extraction capacity is consistent.

The invention relates to an extensible distributed power transaction method in a multi-microgrid environment. The method comprises the following steps of S1, constructing a block chain network; S2, examining the permission allowing a user to enter a transaction market; S3, allocating the public and private keys and the wallet addresses; S4, before each transaction gap, enabling a consumer to submit the quotation information, and a producer and a salesman to submit the bidding information; encrypting all quotation information and bidding information by the public and private keys; S5, completing the bidding matching according to a bidirectional auction mechanism; S6, judging whether a transaction rule is met or not, if yes, executing S7, and if not, returning to S4; S7, carrying out power transmission, and enabling an intelligent electric meter to monitor the electric energy transmission of a node in real time; S8, completing the market clearing; and S9, enabling each microgrid operatorto compete for an accounting right according to a workload proof mechanism, and enabling a winner to be an accountant to make a block and embed the block into a block chain network. According to theinvention, the distributed energy transaction system can realize the higher energy transaction efficiency and the stronger stability.

The invention discloses complex liquid coated medical fertilizer for organically preventing and curing root diseases and the manufacture process thereof. Complex liquid of organic bactericide pesticide and a swelling agent are coated on the surfaces of compound fertilizer particles, and a coating agent is coated on the outermost layer. The compound fertilizer particles are nitrogen-phosphorus-potassium mixed fertilizer particles. The bactericide pesticide is one of DT bactericide, antibiotic 120 bactericide or validamycin antibiotics, the swelling agent is double baking powder, and the two preparations are mixed to be complex liquid according to maintenance dose. The coating agent is seaweed glue, and the raw material weight percentage is as follows: 96 to 98 percent of compound fertilizer particles, 0.3 to 1 percent of bactericide pesticide, 1 to 2 percent of swelling agent and 0.5 to 1.7 percent of coating agent. In the invention, organic bactericide pesticide such as the DT bactericide and the like is mixed with the swelling agent to form complex liquid, a drug micron sprayer is used for spraying on the outer surfaces of the compound fertilizer particles, and then a certain amount of seaweed glue coating agent is added for coating the outermost layer. Therefore, the complex liquid coated medical fertilizer is prepared. The nutrition slowly releasing and disease-resisting and sterilizing functions of the compound fertilizer can be added, and the lasting effect and the utilization efficiency of fertilizer and pesticide can be improved.

The invention discloses a multifunctional optical needle cutter which mainly comprises a needle seat, a needle tube and optical fibers. The end of the needle tube penetrates into the needle seat by passing through a pore arranged at the front end of the needle seat, and the front end of the needle tube extends towards the frontage of the needle seat so as to form a needle tip provided with a cutting edge; the optical fiber output end penetrates through an inner cavity of the needle tube and continuously extends to the needle tip part of the needle tube; the optical fiber input end is fixed ina conductive inserting core, and the conductive inserting core is arranged behind the needle tube and is connected with the end of the needle tube through conductive fibers or directly; an optical fiber joint which can be connected with a laser output port on a host is formed at the back part of the conductive inserting core; and when the joint is connected to the laser output port, the optical fiber input end is in coupling with a laser device which is arranged on the host and can output curing lasers, and the conductive inserting core is connected with a conductive device which is arranged on the host and can output currents. The multifunctional optical needle cutter can simultaneously and comprehensively enable energy including the lasers, the currents and ozone water to be acted on a human body so as to achieve more comprehensive and better therapeutic efficacy.

The invention relates to a HLOD pretreatment method for real-time rendering a large-scale static scene. The method comprises pre-calculating single-model HLOD (Hierarchical Levels of Detail); storing the pre-calculation result, downloading single-model HLOD information according to the scene, and pre-calculating the scene HLOD; and selecting and rendering the HLOD based on viewpoint according to an error specified by a user. The invention has the advantages that the method can flexibly divide the model and the scene according to different hierarchies and granularities such that good detail and less calculation amount can be kept during the rendering, and the method is suitable for pretreatment of model data of large-scale static scene. A HLOD file format is designed to solve the problem of long pre-calculation time, thereby improving the reusability thereof.

The present invention discloses a grid equivalent simplification method considering system grid strength and generator effective transmission capacity, belonging to the technical field of power system analysis. According to the method, the grid strength of a retention system and an equivalent system are characterized through the equivalent branch impedances of the retention system and the equivalent system, the grid strength in the equivalent system is characterized through the step-up transformer impedance from an equivalent generator in the equivalent system to equivalent buses, and the electromagnetic ring network link between different master stations is characterized through the equivalent electromagnetic loop impedance between the equivalent buses. At the same time, the generator effective transmission capacity in the equivalent system is characterized through effective transmission power of the generator via transfer impedance. The method can be used in the regional grid equivalence connected through a line and the low voltage grid equivalence which is connected to a main grid through a transformer branch. Compared with a disclosed grid equivalent simplification method, the topology relationship of an original grid can be more accurately reflected by the network after equivalence, and the physical meaning is clear.

The invention discloses an encryption method and device and a terminal. The method includes the following steps that identity information of equipment is acquired through an encryption card; an encryption key is generated according to the identity information; the equipment is encrypted through the encryption key. By the implementation of the encryption method and device and the terminal, the identity information of the equipment is acquired through the encryption card, the encryption key is generated according to the identity information, and the equipment is encrypted through the encryption key. The equipment is encrypted in the mode that the external encryption card is inserted, the equipment is encrypted through the encryption key independently generated by the encryption card, the simplification of an encryption password is avoided, and the encryption safety of the equipment is improved.

The invention discloses a strip mine high-step large-aperture buffer blasting method. The method can improve the effective utilization rate of the depth of blast holes and reduce the influence of blasting vibration on the stability of a side slope on a high step close to a mining boundary through a blasting technology of distributing buffer holes in the last row and filling explosives in the holesat intervals, so that the slope boundary efficiency is improved, the model of stope drilling machine equipment is simplified, and finally efficient, safe and low-cost boundary operation is achieved.

According to the flexible DC power transmission system centralized energy consumption device topology provided by the invention, the second controllable semiconductor device T2 is additionally arranged on the switch sub-module, so a problem that the controllable semiconductor device in the semiconductor switch is simplified is effectively solved, and a problem that the semiconductor switch cannotbe used when a single controllable semiconductor switch refuses to operate is avoided; the second controllable semiconductor device T2 is added to effectively replace the rejected first controllable semiconductor device T1, sustainability of the semiconductor switch is improved, effective bypass of the switch sub-module is ensured, and after the controllable semiconductor device T1 breaks down, the MOV cannot be exploded. When the offshore wind power flexible direct-current power transmission system has an alternating-current fault on a land station, the energy release device can consume surplus energy, and the fault is prevented from being diffused to an offshore system; direct current is monitored, and surplus energy is discharged; DC overvoltage is reliably suppressed by presetting a target DC voltage; emergency power control becomes possible, and more control is provided for operation and maintenance personnel.

The invention relates to a smart simulation method for an indoor fire simulation training scene. The method includes steps of establishing a flue-gas temperature model before indoor fire flashover; obtaining a curve of flue-gas temperature varying along with time according to the flue-gas temperature model and predicating flashover time; establishing combustion combinations with different fire extinguishing difficulties; applying the flashover time obtained through prediction and the combustion combinations with different fire extinguishing difficulties to a fire scene environment. According to the invention, the flue-gas temperature model can be used for predicating flashover time of a plurality of indoor scenes such as offices, hospitals, theaters and the like and flashover time in two conditions including different positions in a room and different combustibles of an initial fire source can be predicated at the same time, so that multiple fire scene environments can be provided forfire control simulation training and training diversity is achieved.

The invention relates to an amalgamating address quotation system and method and auto quotation device. Commodity sales send quotation code and supervising device address to switch-in amalgamating address device by quotation device; switch-in amalgamating address device compounds quotation code and quotation device address to generate new quotation code and compares it with quotation code of quotation server, then outputs the result to quotation device and supervising device to assure commodity property; Commodity sales send commodity supervising code to switch-in amalgamating address device by supervising device; switch-in amalgamating address device compounds supervising code and supervising device address to generate new supervising code and compares it with supervising code of supervising server to assure commodity property; commodity producer receives buyer's supervising device address by quotation device, and writes sale information to engage in commodity sales.

The invention discloses a flat-top beam forming method based on a neural network. The method comprises the following steps: step S1, collecting sample data; step S2, constructing a convolutional neural network model; step S3, training the convolutional neural network model by using the sample data; and step S4, using the trained convolutional neural network mode to perform flat-top beam forming. According to the flat-top beam forming method based on the neural network provided by the invention, a closed-loop control is introduced in the flat-top beam forming, so that a system is more robust, and meanwhile, training is carried out by using the neural network, only a large amount of data are needed for training, and complex formula derivation and simplification are avoided. After the training is completed, an antenna is adjusted by using the trained neural network, a transmitting antenna and the amplitude of the transmitting antenna needing to be adjusted can be quickly determined according to the input change, and the real-time performance is strong.

The invention discloses a behavior data collection and analysis system comprising a data collection module which is used for collecting the behavior data of students through the unified data access interface; a data processing module which is used for performing data processing on the scholastic achievement inputted by the teachers and the record content submitted by the students and the teachers,acquiring the comprehensive evaluation data for performing comprehensive evaluation of the students and performing original image storage and compressed image storage on the images in the submitted record; a data display module which is used for displaying the record and the corresponding images submitted by the students and the teachers; and a data analysis module which is used for performing statistical analysis of the comprehensive evaluation data on all the behavior data submitted within a period of time, acquiring the comprehensive evaluation score of each student and generating the comprehensive evaluation report of each student. The beneficial effects of the behavior data collection and analysis system are that all the influence factors involved in the evaluation are comprehensively considered so that the score result is enabled to be more objective and reliable.

The invention relates to an auto quotation device used to do amalgamating address quotation. It receives commodity ID quotation code of commodity tag by the auto receiving device and receives quotation device address by network interface card and network device; it sends the quotation code and supervising device address to switch-in amalgamating address device; switch-in amalgamating address device compounds quotation code and quotation device address to generate new quotation code and compares it with quotation code of quotation server, then outputs the result to auto quotation device and buyer's supervising device to assure commodity property and receives commodity information and sale information of quotation commodity which is sent from switch-in amalgamating address device.

The invention discloses an image fine-grained classification method, system and equipment, and the method comprises the steps: carrying out the preprocessing and normalization processing of an image, inputting the image into a convolutional neural network, and carrying out the feature extraction, thus the conventional image spatial features are considered and the features of an image channel are considered, the fine-grained image classification effect can be improved; then obtaining a proportional relation between different channels through an SENet module, and identifying more fine discriminative regions of the image; adopting a clustering method of metric learning to find a plurality of attention areas in one image, so that the distance of the local features of the same category in the image is further shortened, and the distance of the local features of different categories in the image is further increased; and finally, training the image classification model continuously until the loss function converges to obtain an optimal image classification model, so that the images are classified through the model, and the technical problem of poor accuracy of fine-grained classification of the images in the prior art is solved.

The invention provides a beaker washing device. The beaker washing device comprises a iron frame platform, a cross-shaped clamp, an adjustable-speed motor, a connector, pins, a hair brush, a to-be-washed beaker, magnet limiting blocks and a speeder; the cross-shaped clamp is fixedly connected with the iron frame platform and is arranged on the upper part of the iron frame platform; the adjustable-speed motor is fixedly connected with the cross-shaped clamp; the connector is placed under the adjustable-speed motor; the pins are capable of adjusting positions and are placed under the connector;the hair brush is fixedly connected with the pins; the magnet limiting blocks are fixedly connected with the iron frame platform; the to-be-washed beaker is placed on the upper parts of the magnet limiting blocks; and the speeder is connected with the adjustable-speed motor. According to the beaker washing device, in the washing process of beakers, the singleness is avoided, and the washing problem of the beakers with various specifications can be effectively solved.

The invention relates to an equipment monitoring method and device, equipment and a storage medium. The method comprises the following steps: collecting first operation data of the equipment in a marketing business system of a low-voltage transformer area and second operation data of the equipment in a production business system, wherein the first operation data comprises meter data and distribution transformer monitoring data, and the second operation data comprises distribution transformer monitoring data and environment security data; performing fusion and edge calculation on the meter data and the distribution transformer monitoring data to obtain first result data; performing fusion and edge calculation on the distribution transformer monitoring data and the environment security data to obtain second result data; and sending the first result data to a metering master station system, and sending the second result data to a cloud server. According to the technical scheme provided by the embodiment of the invention, the accuracy of monitoring the operation state of the low-voltage transformer area equipment can be improved.