183results about How to "Self-learning ability" patented technology

The invention relates to the field of elevators. In order to early discover and diagnose A elevator fault, the invention adopts the technical scheme that an elevator fault diagnosis and early-warning method based on data drive is achieved by means of a remote service center, a fault diagnosis and prediction terminal and an elevator controller, and the method comprises the steps as follows: firstly, elevator fault data are mined to obtain characteristic information in an elevator fault data stream, and the mined result is stored in an elevator fault case base of the fault diagnosis and prediction terminal; secondly, an elevator fault knowledge base on the fault diagnosis and prediction terminal is updated by the elevator fault case base; thirdly, the case retrieval is carried out on the characteristic of a new elevator fault problem, and the fault diagnosis is carried out on the elevator system by adopting the fault diagnosis method based on the case-base reasoning; and finally information with the characteristic that is most similar with that of the new elevator fault problem is acquired through retrieval of the knowledge or the case in the elevator fault knowledge base to solve the diagnosis problem. The method is mainly suitable for manufacturing and designing image sensors.

The present invention provides a method of web content filtering, which can be applied to gateway devices at client ends. When an access request for a web page from a browser is sent, by analyzing the web page content returned from the web site, we can decide whether the page is allowed or not, by early blocking and early bypassing algorithms. It allows making the decision as early as we have enough confidence that the web content should belong to some forbidden category or a normal one. Therefore, the web content determination speed can be improved and users' waiting time can be reduced. Moreover, by the present method, the network gateway devices will be allowed to analyze more web content.

The invention discloses a wearable device and method for intelligent monitoring and early warning of chronic diseases. The device includes a wearable equipment terminal, a user mobile terminal and a cloud server; the wearable equipment terminal transmits collected physiological signals, motion data and location information processed through a data signal processing module to the user mobile terminal through the wireless communication module; the user mobile terminal includes a wireless communication module, a data storage module and a data processing module; the data processing module is internally provided with an evaluation and prediction model, data in the data storage module is called for evaluating and predicting chronic diseases of users through the evaluation and prediction model in real time, and the evaluation and prediction result is sent to a user interaction interface; the cloud server includes a cloud database and an evaluation and prediction model update module, the cloud database is used for storing the data evaluation and prediction result sent from the user mobile terminal, and an updated personalized evaluation and prediction model is sent to the user mobile terminal to update the evaluation and prediction model.

Provided are an advertisement filtering system and an advertisement filtering method specific to text characteristics. The advertisement filtering system specific to the text characteristics comprises a content input interface, a characteristic analyzing module, a decision calculation module, a data recording module, an information base, an instruction output interface, a manual operation input interface and a machine learning module. The content input interface is used for receiving user generated contents from an internet interactive product. The characteristic analyzing module is used for analyzing the user generated contents, extracting various characteristics of the user generated contents and calculating characteristic values and generating characteristic vectors according to characteristic history and manual operation records. The information base is used for storing various characteristic data of the user generated contents. The decision calculation module is used for comprehensively judging whether to filter the user generated contents or not according to the characteristic vectors generated by the characteristic analyzing module. The data recording module is used for writing the characteristic data, classified data and the manual operation records into the information base. The instruction output interface is used for sorting the judgement result of the decision calculation module into a display / shield operation instruction and synchronizing the display / shield operation instruction to the internet interactive product. The manual operation input interface is used for receiving and analyzing the operation of the filtering result modified by manual work. The machine learning module utilizes the result obtained from every analysis and the manual operation records to perform learning and updates the decision calculation module according to the learning.

The invention relates to a human body behavioral modeling identification method based on a priori knowledge cluster in a computer system. The method comprises the steps of human body behavioral modeling processing and human body behavioral identification processing. By the adoption of human body behavioral modeling identification method based on the priori knowledge cluster in the computer system, a constructed modeling system has a self-learning capacity and can continuously perfect a classifier along with the classification of observed samples. In cluster distribution obtained in the process of human body behavioral modeling, if outliers exist, the outliers serve as an abnormal behavioral model. When the system is designed, precaution is given as required, the method has great realistic meanings, important influence of priori knowledge on a human body behavioral identification result is considered, an improved cluster method is adopted to model and identify the human body, and foundation can be laid for identifying abnormal behaviors. The method is simple and efficient, can accurately identify different behavioral modes, and has good expandability. Working performance is reliable and stable, and the application range is wide.

The invention provides a method for ensuring navigation safety of a mobile robot in the unknown environments, comprising the following steps: 1) building a local grid map of the current environment and determining occupation probability of each grid; 2) constructing the structure of a fuzzy neural network controller; and 3) enabling the fuzzy neural network controller to output the correct speed command of the mobile robot in accordance with the current partial grid map, the current position information of the robot and the current output movement speed and the turning angle of an autonomous navigation system of the mobile robot. The method can overcome the problem that the single navigating and decision-making system is not high in reliability, avoid the mobile robot from such accidents as collision because of mistaken decision-making of the autonomous navigation system and ensure the safety of the mobile robot in the processes of exploration and movement in the unknown environments.

An advertisement filter system comprises a content input interface, a feature analysis module, a decision calculating module, a data recording module, an information base, a command output interface, a manual operation input interface and a machine learning module, wherein the content input interface is used for receiving user generating content from internet interactive products; the feature analysis module is used for analyzing the user generated content, extracting the multiple features of the user generated content and calculating the feature value according to the feature history condition and a manual operation record so as to generate feature vectors; the information base is used for storing various feature data of the user generated content; the decision calculating module is used for comprehensively judging whether the user generated content is filtered or not according to the feature vectors generated by the feature analysis module; the data recording module is used for writing the feature data, categorical data and the manual operation record into the information base; the command output interface is used for arranging a result judged by the decision calculation module into a display / shielding operation command, and synchronizes the display / shielding operation command to the internet interactive product; and the manual operation input interface is used for receiving and analyzing the operation of a filter result modified manually; and the machine learning module utilizes each analysis result and the manual operation record to learn and upgrades the decision calculating module according to the learning.

The invention discloses a complex environment adaptive active radar system based on a big data analysis technology, which comprises a radar system front-end module, an information integration controlmodule, a depth learner module, an auxiliary parameter extraction module and a data center module, wherein the radar system front-end module receives a waveform instruction sent by the information integration control module, a radar transmitting signal is generated, a radar echo signal is then received, and a radar image is formed; the depth learner module receives the radar image, the radar imageis subjected to target recognition, and target recognition result information is obtained; the auxiliary parameter extraction module obtains target parameters according to the radar image; and the data center module is used for integrating online network information and priori information for storing a typical target environment. While requirements on the prior knowledge of the target and the target environment are reduced, the detection and tracking performance of the radar system towards a weak target in a complicated environment can be effectively improved.

The invention discloses a neural network active-disturbance-rejection controller for an AC radial magnetic bearing, and a construction method thereof. The input of a first tracking differentiator is given radial displacement x*, and the output is a tracking signal x1 and a differential signal x2; the input of a first self-adaptive expanded state observer is controlling quantity u, radial displacement x and three parameters beta01, beta02 and beta03, the other two inputs of a first nonlinear state error feedback control rule is parameters beta1 and beta2, and the output is the controlling quantity u0; the difference of the controlling quantity u0 and an estimation value z3 is the input of a first compensation factor, the output of a second compensation factor is the controlling quantity u,and the controlling quantity u is used as one input of a first self-adaptive active-disturbance-rejection controller. By constructing the self-adaptive extended state observer, the internal disturbance and the externa disturbance of the controlled object are automatically controlled, and the online automatic adjusting of the three parameters beta01, beta02 and beta03 can be realized along the system disturbance change, the estimation and compensation precision on the disturbance by the extended state observer are increased, and the control performance of the active-disturbance-rejection controller is improved.

The invention discloses a high-speed unordered capsule defect detecting system, which comprises a capsule transmitting device, an image collecting system, a defect identifying system and DSP (Digital Signal Processor) system transplantation, wherein the capsule transmitting device adopts a mode that capsules are randomly arranged and transmitted and is provided with a shaking device in front of a transmission belt; the image collecting system adopts the high-speed shutter technology of an industrial camera of 90 frames per second to solve the problem of smearing and adopts an infrared filter and a lens used for filtering infrared rays as double lenses to solve the switching problem of different detecting light sources; and the defect identifying system uses a method for setting a dynamic gray threshold to separate a foreground from a background and adopts a method based on the threshold to extract the edges of undivided unordered capsules. The defect detecting system adopts a machine vision technology to detect the shape and appearance of solid capsules and does not need contact the detected articles, so the system belongs to non-contact detection. The defect detecting system solves the problem of secondary pollution of the detected capsules under the general technology. The defect detecting system is an environment-friendly detecting system, and the detecting efficiency and precision and the degree of automation are high.

The invention provides a detection technology of ocean eddy variation based on historical similarity cases, belonging to the information technical field. The method is mainly applied to quantitative variation detection of ocean eddy, and the implementation technical scheme thereof is as follows: establishing a historical case library combined with an eddy spatial-temporal characteristic relationship extracted by a rough set method on the basis of expression models of the ocean eddy cases, then calculating the similarity between the current cases and the historical cases to obtain the historical case which is most similar to the current cases, and finally detecting variation of the current eddy cases according to the situation of the historical cases. Compared with the eddy variation studied by the existing dynamic analysis method via ocean water masses and dynamics, the method is simpler and more flexible; and the historical case library can be dynamically updated with self-learning capability, thus being capable of quickly adapting to ocean eddy with complicated spatial-temporal characteristics so as to carry out more reasonable and more accurate variation detection.

The invention provides a method for predicting the dirt change trend of a large condenser. The method comprises the steps as follows: 1) a support vector machine prediction structure is established, and two subnets are respectively used for predicting the change trends of soft dirt and hard dirt; 2) the product of the current working condition, the historical dirt degree, the cleaning period and the like of the condenser and proportional divisor is used as the input of a support vector machine, the support vector machine is used for obtaining the prediction values of the soft dirt and the hard dirt, and then the sum of the soft dirt and the hard dirt is multiplied with the proportional divisor to be used as a finial dirt prediction result. The proportional divisor can be adjusted by an online learning algorithm to be fit for different conditions. The invention can overcome the difficulty that the dirt is difficult to be described by an exact mathematic model, and the invention is applicable to prediction under different working conditions, is applicable to the prediction of condenser sets with different capacities, can satisfy the prediction of different time spans and has self learning capacity, thereby realizing the exact prediction of the dirt change trend of the condenser.

The invention discloses a room temperature control algorithm based on a fuzzy neural network. A double-input single-output fuzzy neural network controller is designed, the room temperature is detected in real time and output and a temperature setting value are tracked, and an online learning mechanism is adopted to adjust adjustable parameters in the controller in real time to enable the controller to adapt to the change of room temperature and track the temperature setting value. Learning and computing functions of a neural network are integrated into a fuzzy system, and the human-like IF-Then rule of the fuzzy system is embedded into the neural network, so that the adaptive ability of a fuzzy control system is improved on the premise of maintaining the strong ability of knowledge expression, and the system has the ability of self-learning.

The invention discloses an intelligent measurement and control communication network system, and aims to provide a network system capable of increasing the network operating efficiency and operating capability and satisfying diversified services. The intelligent measurement and control communication network system disclosed by the invention is realized through the technical scheme as follows: an intelligent measurement and control communication network is at least composed of two types of nodes including management nodes and normal nodes; the whole intelligent measurement and control communication network is logically divided into a control plane and a service plane; the control plane and the service plane respectively adopt different networking modes responsible for its responsibility; the control plane selects a shortest-path routing strategy, so that the management nodes on the control plane can be communicated with all normal nodes; the service plane is divided into multiple task sub-networks according to tasks executed by the nodes; the various task sub-networks can select different routing strategies according to respective task requirements; according to application and scene requirements of a task, with the help of externally variable parameters, the control plane performs intelligent learning by utilizing a machine, generates a new mathematical model in real time, andsends a new task instruction to the service plane; a foundation measurement and control station is communicated with the management nodes through a satellite-to-earth link; and measurement and controlcommunication of the normal nodes is indirectly carried out through the management nodes.

The invention provides a springback compensation method for creep age forming. The method comprises the following steps: Step S100, based on MSC.MARC commercial finite element software, constructing creep age forming finite element simulation; Step S200, weighting a springback iterative compensation process based on a molding surface of a finite element simulation result; and Step S300, considering a reverse compensation process of high temperature molding surface thermal expansion effect. The method considers the influence of the strength of a component on surface compensation, the influenceof a component feature structure on surface compensation, and reverse compensation of thermal expansion of a high temperature molding surface. A component formed by the model can still meet the engineering error requirement after being cooled.

The present invention discloses an automatic test paper generation method for network examination. The method comprises: (1) setting a test paper generation constraint condition; (2) generating an initial group P0 according to the constraint condition; (3) calculating the fitness function value of each individual; (4) determining whether the individual fitness function satisfies the test paper generation condition or not, and if individual fitness function satisfies the test paper generation condition, generating the test and ending the automatic test paper generation method; if the individual fitness function does not satisfy the test paper generation condition, determining whether the individual fitness function reaches the maximum iterations or not; if the individual fitness function reaches the maximum iterations, adding 1 to the iterations, and performing the genetic manipulation; and (5) generating a new group P1 through the genetic manipulation, and returning back to the operation (3). The automatic test paper generation method for the network examination employs the genetic algorithm to complete the automatic test paper generation to rapidly obtain a scientific test, solve the problem that the examination system is large in test randomness and unstable in the test paper generation process and improve the examination fairness and rationality.

The invention relates to the technical field of bird-repeller devices in airports and control method, in particular to a wireless-control intelligent bird- repeller system. The intelligent bird-repeller system comprises a central control unit (1) and a bird-repeller work unit (2) and is characterized in that the central control unit (1) comprises a data central processing module (3) which is connected with a control panel (10) by a data wire, as well as a bird monitor module (4), a programmable control module (5), a signal coding module (6) and a wireless transmission module (7) which are respectively connected with the data central processing module (3); and the bird-repeller work unit (2) comprises a wireless receiving module (8), a supply unit (9) and a bird-repeller device (11). The wireless-control bird-repeller system in airports can realize centralized, systematic and intelligent management on various bird-repeller devices in airports, enhance the utilization ratio and use effect of the bird-repeller devices and reduce the management cost on the bird-repeller devices.

The invention discloses an online novel content similarity comparison method which comprises the following steps that a pre-processing step is carried out: a compared online novel is pre-processed, key words are extracted, and the key words are replaced by synonyms to form a standard online novel; a characteristic fingerprint is extracted: the online novel to be compared is divided into a plurality of groups of close order words, a Hash table which is formed by the Hash operation of each group of close order words serves as the characteristic fingerprint; and characteristic fingerprints are compared: the characteristic fingerprint is compared with the characteristic fingerprints of existing online novels which are stored in a characteristic comparison library, and the similarity of the online novel to be compared and the existing online novels is determined according to the quantity or proportion of the same characteristic fingerprints. The method is used for comparing the similarity of the online novels, solves the problems of low efficiency caused by high complicacy of the method by pre-processing and checking omission caused by low purposefulness, and has self-learning ability, and the repetition checking accuracy is improved.

The invention discloses a traffic detection method. The method comprises the steps of classifying the preset number of captured data packets, thereby obtaining a normal traffic data matrix and an abnormal traffic data matrix; training an initial logistic regression model through utilization of the normal traffic data matrix and an abnormal traffic data matrix, thereby obtaining a logistic regression model; and detecting traffic data through utilization of the logistic regression model, thereby classifying traffic data into normal traffic data and abnormal traffic data. According to the method and the system, the problems that an existing traffic detection method is long in delayed time, complex in model and insufficient in robustness are solved; the accuracy of a detection and identification process is high, and the identification efficiency is increased. The invention also discloses a traffic detection system which has the above-mentioned beneficial effects.

The invention provides a multi-feature audio fingerprint fault diagnosis method and system for power equipment based on machine learning. The multi-feature audio fingerprint fault diagnosis method forthe power equipment based on machine learning comprises the steps that training stage is carried out, specifically, training samples are preprocessed, then feature extraction is carried out, 12-dimensional MFCC parameters and 12-dimensional delta MFCC parameters sufficient to represent signal features are extracted as observation sequences; according to the characteristics of continuous mixed Gaussian HMM, the parameters of HMM models are initialized, and initial values of the parameters are stored; the initial values of the parameters are reestimated and trained according to the observationsequences; the parameters of the reestimated HMM model are saved, and the HMM models are established; and diagnosis stage is carried out, specifically, test samples are obtained, the features of the test samples are extracted, and the output probability of observed value sequences under each HMM model is calculated according to the established HMM models. According to the multi-feature audio fingerprint fault diagnosis method and system for the power equipment based on machine learning, whether the equipment is in an abnormal operation state or not can be judged through the changes of audio characteristics such as the timbre of sound, the size of volume and the level of frequency, and even the type and severity of the fault can be judged.

The invention provides an air conditioner control method and an air conditioner control device. The air conditioner control method comprises detecting user body surface moisture according to the preset time interval; adjusting the target adjusting temperature of a current time period to which the current moment belongs of a temperature curve which is operated by an air conditioner according to the body surface moisture; updating the temperature curve according to the adjusted target adjusting temperature and storing the temperature curve. According to the technical scheme, the air conditioner control method enables the air conditioner to have a self-learning capability, reduces user operation and improves user comfort experience.

The invention provides control method and system of the dynamic feedback of the service quality of an end-to-end service. The control system comprises the following modules: a data acquisition module, a service quality analysis module, a service parameter programming module, a service parameter regulating module and a knowledge base module, wherein the data acquisition module is used for acquiring and detecting the network performance parameters of the end-to-end service and the the end-to-end service evaluating information of a user; the service quality analysis module is used for analyzing the network performance parameters and the end-to-end service evaluating information of the user to determine service parameters which need to be regulated; the service parameter programming module isused for programming the service parameters which need to be regulated to form optimized service parameters; the service parameter regulating module is used for setting the optimized service parameters into service implementation equipment; and the knowledge base module is used for providing information support for the data acquisition module, the service quality analysis module, the service parameter programming module and the service parameter regulating module. The control method monitors an end-to-end network service system by utilizing the control system and programs and regulates the optimized service parameters of the service implementation equipment, thereby dynamically enhancing the service quality of the end-to-end service in real time.

The invention provides a method for controlling the frequency of a compressor, a device for controlling the frequency of the compressor and an air conditioner. The method for controlling the frequency of the compressor comprises the steps that after the air conditioner is started, if an initial set temperature is received, historical initial set temperatures in a time interval where the set moment of the initial set temperature is in are obtained, whether the initial set temperature is matched with the historical initial set temperatures or not is judged, and when a judging result is positive, the operation frequency of the compressor is controlled according to a received operation mode and historical temperature records in the time interval. According to the technical scheme, the air conditioner can have the self-learning capability, the frequency of the compressor of the air conditioner can be adjusted in real time according to the use habits of a user, so that the comfort degree experience of the user is improved, operation of the user is reduced, and the energy utilization efficiency is increased.

The invention discloses a method for talent ability evaluation and job matching for college students. The specific steps include the creation of a student-side personal quality model, the creation of an enterprise-side job ability model, a matching and scoring algorithm, an adaptive algorithm, and engineering implementation. The student-side personal quality model adopts the teaching-simulation-practice model, and the job ability model includes two basic ability models and a specific ability model. Starting from the enterprise-side job ability model, the relationship between variables is obtained through linear regression; the overall structure of engineering realization Consists of data acquisition, data preparation and data application. The present invention uses a big data matching algorithm to calculate the matching degree of the personal quality model and the job ability model, and uses this data to obtain each person's competitiveness and application ranking for a specific position, has self-learning ability, improves accuracy, and saves Reduce time cost, improve efficiency, and achieve better results.

The invention discloses a log file processing method, device and system. The method includes: return code in log file is read; when error is determined by virtue of the return code, database is queried according to the return code, wherein the database is initialized with multiple error information and handling flow thereof; and error handling is automatically carried out according to the found handling flow. In the invention, the return code in log file is utilized to be matched with database initialized with various error handling flows, and handling flow corresponding to the error is executed, so as to realize automatic error correction. Compared with visual check and handling of the prior art, the invention is more timely and accurate and time-saving. Besides, the log file can be analyzed, and related information in the log file is recorded into the database every time, thus realizing the aim of retrospecting history log information.

The invention relates to a geological disaster prediction and forecast technology, and aims to provide a rainfall threshold analysis method causing basin landslide risks. The method comprises the following steps: establishing a three-dimensional grid according to DEM topographic data and soil layer thickness data of a target watershed, and generating boundary conditions and initial conditions required by an RIRM hydrological model; calculating space-time changes of hydrological elements in the drainage basin and drainage basin slope stability reference indexes of any three-dimensional positionin the drainage basin; drawing a relation curve that the safety factor SF, the landslide volume and the landslide area respectively change along with time under the conditions of different rainfall durations and different rainfall intensities; analyzing the relationship between landslide deformation damage and rainfall to obtain rainfall thresholds causing basin landslide risks under different conditions. According to the method, natural basin geomorphic hydrological response characteristics are reserved, and prediction errors can be effectively reduced; drainage basin hydrology and slope stability analysis can be carried out simultaneously, the actual occurrence process of drainage basin landslide disasters is met, and the prediction result is more real and reliable.

The invention relates to the technical field of comprehensive emergency management, in particular to an urban comprehensive emergency command scheduling task distributing system and a task distributing method thereof. The urban comprehensive emergency command scheduling task distributing system mainly comprises two parts, namely commanding and dispatching and parameter setting; parameter setting is implemented by setting initial starting conditions for commanding and dispatching; the commanding and dispatching are implemented on the basis of map plotting; after plotting is completed, the map is pushed to an intelligent mobile phone through a rest interface. According to the invention, the problems that current tasks distributed by various comprehensive emergency command scheduling systems are not rich in contents and are inaccurate in position are solved; the task distributing method can be used for the emergency command scheduling task distributing system.

The invention discloses a vehicle-mounted application containerization isolation framework system based on IPFS and an implementation method, belongs to the technical field of vehicle networking safety, information safety and the like, and solves the potential information safety problem existing in operation and deployment of application software of a vehicle intelligent terminal and a vehicle-mounted system in an existing vehicle networking. The system comprises a unique container platform which is used for vehicle-mounted application installation and is deployed in a vehicle machine system of a vehicle, a decentration monitoring management platform which is used for safety monitoring and software review of a vehicle-mounted application program, and an IPFS protocol which is used for realizing communication between the container platform and the decentration monitoring management platform. The method is used for containerization isolation of the vehicle-mounted application.

The invention discloses a service anomaly warning method and device based on time sequence. The method includes: acquiring multiple true values of a preset warning index in an appointed time period according to preset time intervals; determining multiple predicted values of the preset warning index in the time period according to a predicting model which is obtained by fitting according to the time sequence of historical true values of the preset warning index; according the true values in the time period and the predicted values in the time period, determining whether to perform service anomaly warning in the time period or not; after the warning cycle where the time period is located is finished, adding the true values of the preset warning index in the warning cycle into the historicaltrue values of the preset warning index so as to update the predicting model. By the method and device, warning efficiency can be increased, and false warning and warning failure can be reduced.