36results about How to "Not universal" patented technology

The invention provides a preparation method of novel doped black titanium dioxide. The method comprises the following steps: (1) a titanium oxide precursor is hydrolyzed, and after full washing, drying is carried out; (2) hydrolysate as well as metal nitrate, sulfide, fluoride, nitride and the like are fully mixed, and the materials are calcinated in an atmosphere furnace or a vacuum furnace at 300-800 DEG C for 2-15 hours. The products are fully washed and dried in order to obtain the doped black titanium dioxide. Doping with Fe, Mn, Sn, Ce and other metal cations as well as N, S, F and other anions are suitable for the method, the application range is wide, the process is simple, the cost is low, and the reaction conditions are simple.

A numerical value simulation method for spot welding connection and failure thereof is characterized in that according to a structural characteristic that two areas: a weld nugget zone and a thermal effect zone are classified according to different changes of material property, solid elements and shell elements are used to respectively establish a solder joint finite element model consisting of the weld nugget zone and the thermal effect zone, thus providing a pertinent execution basis for the arrangement of solder joint failure parameter. The method has the advantages of avoiding the complexity of research on the mathematic models of the solder joint, simplifying the judgment standard for solder joint simulation failure, complying with the principle of one problem with one standard, and widening functions of simulating the failure of the spot welding. The method provides specific establishment rules for forming the model elements of the solder joints, wherein the rules are suitable for automatic realization of program.

The invention discloses a fast and load-balanced service function chain deployment method in a dynamic network environment. Most of the traditional service function chain deployment algorithms are proposed separately for multi-server network or data center network and do not have universality, As that method of the invention can be used in a multi-server network, or in a data center network, and is not limited by the size of the network; compared with traditional deployment algorithms, the invention has a wider application range, the reinforcement learning module is used to learn the network topology links, combined with the load balancing module, the shortest path which is close to the optimal solution can be obtained, Moreover, the invention can accept more SFC deployment requirements, and the total profit of the found deployment scheme is higher than that of the traditional deployment algorithm. Under the condition of effective training, the invention can output the SFC deployment scheme quickly. Compared with other high-accuracy deployment algorithms, the algorithm speed of the invention is much faster.

The invention relates to the technical field of network information security exchange, in particular to a device and method for implementing data transmission between physical isolation networks based on hexadecimal codes. The device comprises a data source computer, a display screen of the data source computer, an image acquisition device and a target computer. The data source computer and the target computer are located in two networks in the physical isolation state respectively. A data source generation module is arranged in the data source computer to convert source data into multiple sets of data embedded with serial numbers, secret keys, effective data marks and screen display rules, and the data are sent to the display screen in a split screen mode and displayed in the form of hexadecimal code characters. The image acquisition device acquires images displayed on the display screen and transmits the acquired images to the target computer, a parsing module in the target computer identifies and parses the images, the parsed information is decoded, combined and uncompressed, and then data consistent with the source data are obtained. By means of the device and method, data transmission between the two networks in the physical isolation state can be conducted in real time, synchronously and automatically, and meanwhile, data transmission is safer and more reliable.

The invention discloses an individual cognitive ability prediction method and system based on dynamic function connection. The method comprises the following steps: acquiring functional magnetic resonance imaging data in a resting state; extracting an average time sequence signal of each brain region through a brain template after preprocessing of the functional magnetic resonance imaging data; calculating and constructing a dynamic function connection matrix by using a sliding time window; with the dynamic function connection matrix as input high-dimensional dynamic brain description originaldata, constructing a model combining convolution and a long-short-term memory network; and substituting a dynamic function matrix and corresponding cognitive behavior data into the model for trainingand testing so as to realize prediction of cognitive competence. According to the invention, dynamic information of brain function connection is fully utilized, the advantages of a deep learning method in big data processing are utilized, calculation is rapid, prediction effect is ideal, various behavior data of a population can be predicted, dependence on the absolute value of the amplitude of imaging signals is avoided, migrations among different MRI machines can be realized, and prediction performance is excellent.

The invention discloses an air filter. The air filter is installed on a motor vehicle and comprises a shell, wherein a filter element is arranged in the shell; a radio frequency identification device(RFID) tag is arranged on the filter element; an RFID reader-writer for reading and writing the RFID tag is arranged on the shell; an air inlet and an air outlet are formed in the shell; a pressure sensor is arranged on the air outlet; and a controller is arranged in the shell and is used for connecting with a vehicle-mounted electronic control unit (ECU) of the motor vehicle. The invention further discloses an anti-fake and pre-alarm method for the filter element the of air filter. The method comprises the steps of checking the authenticity of the filter element, executing conditions, judgingwhether the service time of the filter element is overtime or not, judging the rotate speed, timing the continuous operation of an engine, judging alarm conditions, and updating. According to the airfilter, and the anti-fake and pre-alarm method for the filter element of the air filter provided by the invention, an anti-fake performance is high, so that an anti-fake function of the filter element is difficult to crack; pre-alarm is realized, so that a user using the motor vehicle has enough time to arrange the filter element replacement; and a function on preventing false alarm is realized,and the occurrence of the phenomenon that unstable external conditions cause false alarm is prevented, so that the air filter, and the anti-fake and pre-alarm method for the filter element of the airfilter provided by the invention have favorable popularization and application values.

An antiforge schlieren print technique for bill or package includes such steps as writing the associated data about text, signs and bit map to a database, respectively reading each record from said database by a control module, and processing it by schlieren technique to generate the printable schlieren pattern.

The invention relates to a satellite navigation and positioning method considering non-modeling errors. According to the satellite navigation and positioning method considering non-modeling error, some non-modeling errors which are neglected in a traditional method and cannot be corrected by models and absorbed by parameters are considered, and therefore, the positioning accuracy and reliability of the method of the invention are improved compared with that of the traditional method; since the non-modeling errors have spatial and temporal correlations, a multi-epoch partial parameterization method is used to optimize observation equations enabling non-modeling error significance, and therefore, the accuracy of a model is improved; observation residuals are adopted test the significance ofthe non-modeling errors; since redundant observations are increased in a multi-frequency multi-mode GNSS application environment, the rank defect problem of the residuals can be alleviated, and therefore, the method is especially suitable for multi-frequency multi-mode GNSS observation data; and thus, the satellite navigation and positioning method considering non-modeling errors has the advantages of high precision, high reliability, wide application range and the like.

The invention relates to a method for rejecting wild values of received data in a traffic collision avoidance system (TCAS). By the method, the wild values of the received height data and the received distance data are avoided and rejected quickly; and the wild values exist in the received data of the TCAS and the working speed of the system is high, so the wild values must be rejected quickly. The invention provides a method for avoiding and rejecting the wild values of the height data and the distance data. By the method, the interference problem of the wild values of the data is solved; the effectiveness of the received data and the flight safety are guaranteed; the input of the wild values is reduced by simplifying the height source of the TCAS; and the wild values of the height and the distance are further rejected by a corresponding software processing method.

A snore stop pillow is characterized in that when a snoring patient snores, a sound sensor captures snores and transmits the current to a control chip, the control chip controls a low-speed motor to rotate in low speed according to a program, an eccentric wheel protrudes and ejects rigid rubber, part of a pillow body is protruded, and the patient is stimulated and unconsciously adjusts the sleeping posture until snoring is stopped. The snore stop pillow is low in cost, simple in structure and significant in effect, side effect is avoided, the generality is high, the snore stop effect is better than majority of snore treatment devices on market, disasters, such as sleep apnea, brain oxygen lack, heart disease, caused by snoring are avoided effectively, and the surrounding people are not affected.

The invention discloses a motor vehicle exhaust concentration prediction method and system based on machine learning, and the method comprises the following steps: S1, importing vehicle exhaust data from a corresponding database into an excel or csv file, and carrying out the preprocessing of the data, and obtaining the preprocessed exhaust data; s2, constructing a model framework, combining a neural network method and a support vector regression method in machine learning, adopting a multi-model fusion method, and fusing two models to perform tail gas concentration inversion; s3, utilizing the preprocessed tail gas data to train a tail gas concentration inversion fusion model, wherein a neural network part is used for memorizing deep-level tail gas data, and support vector regression is used for inverting the tail gas concentration; and S4, inputting vehicle tail gas data to predict the tail gas concentration by using the trained tail gas concentration inversion fusion model. The method is high in detection precision and easy to popularize; and multiple feature processing is performed, so that the accuracy is high, and the accuracy can be improved by performing feature processing on the original data twice through the neural network model and the support vector regression model.

The invention discloses a sensitivity analysis and reinforcement learning neural network pruning method, system and device, and the method comprises the steps: setting a sparsity threshold, and selecting a low-sensitivity weight for pruning; obtaining a cutting method and precision, and determining a weight needing random pruning according to the sensitivity weight; carrying out random clipping oneach selected weight, and putting a pruning method and precision of multiple times of random clipping into a buffer area; a training reinforcement learning step: training a reinforcement learning agent by using the data in the buffer area, and putting a cutting method and precision generated after training into the buffer area; and repeating until the network precision reaches a preset value. According to the method, the low-sensitivity weight is selected for pruning, and the sparsity threshold value of each weight is set to ensure that the network descending precision is kept within the preset range after the cut weight is cut by adopting the current sparsity process. Under the condition that the network precision is ensured, the compression ratio of the neural network is improved to themaximum extent.

The invention discloses a vacuum moisture regaining control process for tobacco leaf processing quality homogenization, energy conservation and consumption reduction, belongs to the field of cigaretteraw material processing, and covers threshing and redrying enterprises and tobacco shred producing workshops of cigarette factories or all using units related to vacuum moisture regaining equipment,and the vacuum moisture regaining equipment has software and hardware configuration. The vacuum moisture regaining equipment is mainly composed of a vacuum box body, a switch (box body) door mechanism, a vacuumizing system, a moisture steaming system, a steam distribution system, a compressed air system, a cooling water circulation system, an electric control system and the like. If the vacuum moisture regaining equipment is not provided with a negative pressure monitoring device, a negative pressure sensor for detecting the vacuum degree needs to be installed in a vacuum moisture regaining box body, a corresponding digital quantity input module or a corresponding PLC analog quantity module detection channel needs to be additionally arranged in a PLC system, and input and adjustment can becarried out on an operation page of an industrial control system. A control method and matched parameters are provided, the smoothness of the vacuum moisture regaining technological process is guaranteed, the processing efficiency of technological equipment is improved, and the energy consumption of the technological equipment is reduced while the batch tobacco leaf processing quality is improved.

The invention discloses a mobile application identification method based on adaptive incremental learning. The method is a method for expanding the identification range under the condition that the number of to-be-identified mobile applications is increased based on the incremental learning (Incental Learning). The objective of the invention is to solve the catastrophic forgetting phenomenon of the existing mobile application identification technology, when a new mobile application type identification task is given, an original deep learning classifier with a good effect is kept in an originalstate, and a new mobile application type is added step by step on the basis of the original deep learning classifier. Therefore, the whole mobile application identification system has the self-adaptive incremental learning capability and the capability of expanding the number of the types of the to-be-identified mobile applications.

The invention relates to the technical field of water quality detection, in particular to a detection device capable of quantitatively detecting water at different depths. The detection device comprises a detection frame; a bracket is arranged at the bottom of the detection frame; the detection frame is fixed with the bracket through bolts; a graduated scale is arranged at the top of the detectionframe, and the detection frame is fixed with the graduated scale through bolts; a cylinder is installed in the detection frame; an arc-shaped plate is installed in the cylinder; the arc-shaped plateis slidably connected to the interior of the cylinder; an installation plate is installed in the arc-shaped plate; the installation plate is welded to the cylinder; and a collection cylinder is installed at the top of the installation plate. The detection device is reasonable in design and ingenious in structure; through mutual cooperation of the structures, detection is more convenient, multipletimes of sampling can be achieved, water of different depths can be quantitatively detected, samples are more universal, the detected samples do not have universality, and the method has high market competitiveness and is worthy of recommendation.

The invention relates to a method for automatically discriminating 'four seasons type person' based on the analysis of human face image features. The method comprises the following steps: first, selecting skin, eye, lip and eyebrow four key facial regions based on digital image of human face region; second, according to the established character image database marked with seasonal types, four samples of seasonal types being selected for each of the four key facial regions; third, color similarity algorithm based on color histogram and gray level mean being used to compare the color of each keyarea of the image to be measured with the color similarity of the corresponding area samples of four seasonal types, and the seasonal types of each key area of the character being obtained respectively by comparing the color similarity of each key area of the image; in the fourth step, the seasonal types of the four key regions being analyzed comprehensively, and the final seasonal types of the characters are obtained. Compared with the prior art, the present invention has the advantages of automatic discrimination, objective analysis, accuracy and high efficiency, etc.

The invention discloses an inclined three-dimensional model reconstruction and dynamic scheduling method based on a CAD platform, and relates to the field of computer aided design, in particular to an inclined three-dimensional model architecture adjustment method based on the CAD platform. The invention also discloses a method. The early stage comprises OSGB data hierarchical rendering of the CAD platform, and the later stage comprises architecture adjustment and method improvement; the architecture adjustment and the method improvement comprise two stages of reconstruction and dynamic scheduling of the inclined three-dimensional model. By applying the algorithm model, a foreground and background combined double-buffering type inclined three-dimensional model loading mode of preparing data streams in a background and drawing data images in a foreground is realized, and model reconstruction, image erasing and drawing and memory loading and unloading are efficiently and dynamically performed on inclined three-dimensional model tiles and scenes in real time according to needs.

The invention provides a compound emulsifying agent for preparing an aqueous emulsion, comprising the following components in percentage by mass: 28%-40% of fatty alcohol-polyoxyethylene ether, 19%-40% of alkyl phenol polyoxyethylene ether phosphate or fatty alcohol-polyoxyethylene ether phosphate or kali salt of the alkyl phenol polyoxyethylene ether phosphate and the fatty alcohol-polyoxyethylene ether phosphate, 4.5%-10% of organic base and 19%-45% of emulsifying agent span. The compound emulsifying agent has the advantages of good popularity, easy preparation, stable property and high cost performance and is suitable for the parathion and the production of a plurality of pesticide aqueous emulsions.

A numerical simulation method for spot welded connections and their failures. Its characteristic is that according to the structural characteristics of the weld joints, which are divided into two regions, the weld nugget zone and the heat-affected zone due to the change of material properties. The finite element model of solder joints provides a targeted implementation basis for the setting of solder joint failure parameters. Its advantages are that it avoids the complexity of the research on the mathematical model of the solder joints, simplifies the standard for judging the failure of the solder joint simulation, follows the principle of the same standard for the same problem, and expands the function of the spot welding failure simulation. This method proposes a specific construction rule for the formation of the solder joint model grid, which is suitable for automatic programming.

The embodiment of the invention relates to the technical field of natural language processing, and discloses a statement extraction method and device, electronic equipment and a computer readable storage medium. The statement extraction method comprises the steps that to-be-processed text information is acquired; wherein the to-be-processed text information comprises text information converted and generated according to acquired voice information; inputting the to-be-processed text information into a pre-trained statement partitioning model to obtain a plurality of candidate statements; filtering the plurality of candidate statements according to a preset filtering rule to obtain and output a target statement; wherein the target statement is the statement remaining after filtering in the candidate statements, the statement extraction method provided by the embodiment of the invention can scientifically, reasonably and accurately extract the core statement capable of representing the real intention of the user from the text information, so that high-quality service meeting the actual needs of the user is provided for the user.

The invention relates to the technical field of water quality detection, in particular to a detection device that can quantitatively detect water of different depths, comprising a detection frame, a bracket is installed at the bottom of the detection frame, the detection frame and the bracket are fixed by bolts, and the detection frame A scale is installed on the top of the test frame, the detection frame and the scale are fixed by bolts, a cylinder is installed inside the detection frame, and an arc-shaped plate is installed inside the cylinder, and the arc-shaped plate is slidably connected to the circular Inside the cylinder, a mounting plate is installed inside the arc-shaped plate, the mounting plate is welded with the cylinder, and a collecting cylinder is installed on the top of the mounting plate. The present invention has a reasonable design and an ingenious structure. With the cooperation, the detection is more convenient, it can be sampled multiple times, and it can quantitatively detect water of different depths, making the samples more common, making the samples tested are not universal, and have good market competitiveness, which is worth recommending.

The invention relates to a method for measuring the content of hydrogen peroxide through low-field nuclear magnetic resonance. The method comprises the following steps of: measuring a plurality of standard hydrogen peroxide samples with different concentrations of Ci by using a low-field nuclear magnetic resonance analyzer, respectively acquiring signal values Si of the standard hydrogen peroxide samples, and drawing a working curve of the standard hydrogen peroxide samples by taking Ci and Si as coordinate values; measuring a sample to be measured by using the low-field nuclear magnetic resonance analyzer to obtain a relaxation signal value S of the sample to be measured; and substituting the relaxation signal value S into the working curve of the standard hydrogen peroxide sample, and calculating to obtain the concentration C of the sample to be detected. Compared with other methods for measuring the content of hydrogen peroxide, the method has the advantages that the testing speed is high, the sample does not need to be pretreated in the testing process, the solvent and the sample do not need to be consumed, the method is safe and environmentally friendly, and the risk that other components in the sample cause the inaccurate concentration result of hydrogen peroxide is avoided.

The invention discloses a multi-type precession warhead parameter estimation method based on a double-layer double-channel convolutional neural network. The method comprises the following steps: firstly, emitting single-frequency pulses to one type of bullets with different sizes at one viewing angle to obtain main polarization echo data of the same type of bullets with different sizes at the viewing angle, and then emitting single-frequency pulses from another viewing angle to obtain main polarization echo data of the same type of bullets with different sizes at the viewing angle; and then replacing another type of warhead and repeating the operation until all types of warheads have main polarization echoes under two visual angles; making time-frequency transformation on the obtained echodata to obtain a large number of time-frequency graphs; using four time-frequency graphs of the same warhead under two main polarizations at two visual angles as a group of data, and dividing the obtained data into a training set and a test set; building a double-channel double-layer convolutional neural network regression model, training the obtained training set, testing the test set, and obtaining the relative root-mean-square error of each parameter of each type of warhead. The invention can be used for carrying out parameter estimation on multiple types of bullets.

The invention relates to the technical field of industrial environments, and discloses an authorization method suitable for an industrial environment, a scanning device for confirming an industrial environment authorization method different from a manufacturer, information of a to-be-authorized scanning device is obtained, an asymmetric algorithm is utilized, an authorized device can confirm the identity of an authorization device and obtain an authorization file, the number of authorizable times of a fuse bit on authorization scanning equipment is set, equipment information is encrypted through an encryption algorithm agreed with the authorization equipment on the basis of a pre-acquired first key to obtain an equipment information ciphertext, and the first key and the equipment information ciphertext are sent to the authorization device so as to enable the authorization device to be based on the first key. The authorization method suitable for the industrial environment has the advantages that a user authorizes each system and equipment, is responsible for operation, avoids vulnerabilities, can set the number of burnt authorization files, and limits authorization management of the equipment, further a protection mechanism is improved, and potential safety hazards and the like are avoided.

The invention discloses a lunar soft landing optimal guidance method and system based on a safe landing channel and a storage medium, relates to the technical field of lunar exploration soft landing control. According to the scheme, the problems that an existing lunar soft landing optimal guidance method is generally used for modeling obstacles, however, the modeling precision of the existing lunar soft landing optimal guidance method affects an optimization result and is difficult to adapt to an extremely complex obstacle environment are solved. The method comprises the following steps: 1, based on an obstacle point set of a predetermined landing point, generating an initial piecewise linear path by adopting an error domain merging jump point method; 2, based on the initial path generated in the step 1, generating a safe landing channel constraint for each section of initial path by using a convex decomposition algorithm; and 3, based on the piecewise linear path generated in the step 1 and the landing channel constraint generated in the step 2, converting an original problem into a second-order cone optimization problem by utilizing a deceleration time distribution rule, and solving the second-order cone optimization problem so as to obtain an obstacle avoidance track with optimal fuel. The method is applied to the technical field of lunar exploration soft landing.

The invention discloses an air filter, which is used to be installed on a motor vehicle. The casing is provided with an air inlet and an air outlet, and a pressure sensor is arranged at the air outlet; a controller is arranged in the casing, and the controller is used to connect to the vehicle-mounted ECU of the motor vehicle; the invention also discloses a filter element of the air filter The anti-counterfeit and pre-alarm method includes the steps of checking the authenticity of the filter element, the step of condition execution, the step of judging whether the service time of the filter element is overtime, the step of judging the rotational speed, the step of timing the continuous operation of the engine, the step of judging the alarm condition and the step of updating. The present invention has strong anti-counterfeiting performance, and the anti-counterfeiting function of the filter element is difficult to be deciphered; it realizes the pre-alarm, and leaves enough time for motor vehicle users to arrange replacement of the filter element; it has the function of preventing false alarms, and prevents false alarms caused by unstable external conditions, making The invention has good popularization and application value.

The invention belongs to the technical field of computer file systems, in particular to a method for constructing a secure file system. The present invention realizes the heterogeneity of the file system through multi-dimensional data space transformation, and provides file systems with different structures for users authorized to use the file system. They have the same business processing logic, and there is no difference in logic. Different data structures, organizations, processing, and storage methods are brought about by different data structures, thereby realizing the heterogeneity of file system data organization and access, and enhancing file system security; file systems mainly include file interfaces that provide user access and file system metadata. , file metadata and data space management; a random heterogeneous transformation method is designed for the four parts, so that the file system built each time is random, different from ordinary file systems, the format is fixed, and it is easy to know the organization of the file system through analysis. Ensure file system security.

The invention discloses an explosion-proof intelligent video inspection robot. The explosion-proof intelligent video inspection robot comprises an explosion-proof robot body, the explosion-proof robotbody is provided with a first side box, a second side box, a connecting cylinder, a motor bin, a probe assembly and a power mechanism, wherein the side edge of the first side box is provided with a first side plate, the side edge of the second side box is provided with a second side plate, and the connecting cylinder, the motor bin and the power mechanism are installed between the first side plate and the second side plate. The explosion-proof intelligent video inspection robot is used for solving a series of problems that hydrogen sulfide, ammonia and volatile organic compounds such as methyl mercaptan and dimethyl sulfide generated in the sewage treatment process of underground sewage treatment plants and storage ponds have the possibility of explosion, and the temperature of a batteryis too high due to friction sparks and uninterrupted operation of a running track; and the explosion-proof intelligent video inspection robot is mainly applied to the underground or semi-underground sewage treatment plants, the storage ponds and urban pipe galleries for explosion-proof design and experiments, and safe and stable operation is ensured.

The invention discloses a routing exchange protocol message transformation method for enhancing the safety of internal network equipment, and particularly relates to the technical field of routing protocol safety. The method comprises the following steps: transforming a router exchange protocol stack and developing a protocol message conversion server; the transformation of the router exchange protocol stack comprises the following steps: (1) transforming the length of a field in a message; (2) modifying the position of a field in the message; (3) modifying a protocol stack of an operating system of the routing switch according to the modification work; the development of the protocol message conversion server comprises the following steps: (1) converting a traditional network message received from traditional terminal equipment into a modified network protocol message, and sending the modified network protocol message to the modified network equipment; and (2) converting the reconstructed network message received from the reconstructed network device into a traditional network protocol message, and sending the traditional network protocol message to a traditional terminal device.