46 results about "Process classification" patented technology

The invention belongs to the field of load power use monitoring and energy management, and particularly relates to a method and system for identifying transient process of a non-intrusive load. The method comprises the following steps: by virtue of transient power waveform characteristics of multiple types, measuring similarity of a transient power waveform characteristic parameter sample sequence of an unfixed length and a template sequence by using a dynamic time warping algorithm within a time domain, on the basis, establishing a load transient process classification identification scheme based on a nearest neighbor classification strategy, and confirming power use equipment with the load transient process, thereby achieving artificial working state identification on the non-intrusive type power use equipment. The method and the system have the beneficial effects that accuracy and robustness of power load transient process identification can be improved, the cost of a monitoring system can be effectively controlled, and the practicability of the monitoring system can be effectively improved, therefore, practical popularization of an NILM (Non-Intrusive Load monitoring) technique can be greatly improved.

The invention discloses a method for extracting three-dimensional model view based on panorama and multi-channel CNN. The method comprises the following steps: projecting a 3D model onto a side surface of a cylinder satisfying preset conditions; taking an origin of the 3D model as a center, obtaining an initial panorama by paralleling an axis of the 3D model to one of the principal axes of X, Y and Z; obtaining an initial panorama by projecting a 3D model onto a side surface of a cylinder satisfying preset conditions; 3-D model surface angle phi and y coordinates are sample at a preset rate, acquiring two sets of values of each point in the initial panorama to represent the position characteristics of the 3D model surface and the directional characteristics of the 3D model surface in three-dimensional space; A multi-scale network and a multi-channel convolution neural network are constructed. The 3D model surface location features and 3D model surface orientation features are used as inputs to train the network and measure the similarity between the two different 3D models. The invention retains the local and global information of the structure and vision of the three-dimensional model, automatically calculates the features of the 2D panoramic view, and is used for processing classification and retrieval problems.

The invention discloses a CPU resource allotting method and a terminal, which can determine a process in non-stop state. The method comprises the following steps: according to importanceof the interaction experience of processes among a user and a terminal, the process can be divided into pre-set CPU setting nodes at different levels; states of all CPU cores in the terminal can be monitored and information of the one-line CPU cores can be acquired; and on-line CPU core resources operating the process are distributed to all the CPU setting nodes according to the rank of priorities of all the CPU setting nodes. The CPU resource allotting method and the terminal have the following beneficial effects: according to on-line conditions of all the CPU cores and importance of the process in the non-stop state, CPU core resources operating the process are dynamically distributed so that operation of the important process is ensured, thereby improving overall performance of ANDROID and ensuring smooth experience of a user; furthermore, power consumption of the terminal is reduced to strike a balance between performance and power consumption; when the terminal is used, secondary fine adjustments are made to process classification and a distribution strategy of CPU core resources, which further ensures balance between performance and power consumption of the terminal.

The invention discloses an automatic identification method and system for water floating objects based on artificial intelligence. The method comprises steps of firstly constructing a feature number knowledge base of different ground objects represented by different kinds of floating objects in different bands; using the feature number knowledge base as a basis for classification, and using a classification method of the decision tree to classify floating object monitoring images; processing classification results after classification by a clustering process to obtain results of floating object recognition of the monitoring images; evaluating accuracy of recognition results of floating objects of the monitoring images; and finally, inputting the feature number knowledge base into a supportvector machine for training, and using a trained support vector machine model to automatically identify floating object information in the floating object monitoring images. The method and the systemare adapted to automatic identification of the floating object information in the images, thereby effectively and rationally organizing the floating object cleaning action, and being significant forensuring the navigation of the ship, the safety of the dam, the operation of the reservoir, and the like.

The invention discloses a personal health record system with a process decision support function. By building a personal health record system of characteristic data extraction and characteristic data analytical separation, the characteristic data meeting analysis requirements is expected to be rapidly obtained. A statistics query interface based on characteristic data and time logic is improved, decision support for a user is completed through the steps of characteristic matching, process classification statistics evaluation, process recommendation and the like, and the medical big data can well execute high-level information analysis and decision support functions.

Provided is an underground LED intelligent display management assistance system, which belongs to a display management assistance system. According to the display management assistance system, LED intelligent display terminals are installed in underground personnel-intensive areas and key monitoring areas, and duplex communication is carried out between the LED intelligent display terminals and a ground LED intelligent management service system and other LED intelligent display terminals of all levels in a wired or wireless transmission mode. A sensing information system transmits underground data to the LED intelligent management service system for cluster analysis, a classification result is processed by software, and the processed classification result is transmitted through the Ethernet to the embedded LED intelligent display terminals for displaying all underground information and all ground instructions in real time. The underground LED intelligent display management assistance system has the following advantages: the LED intelligent display terminals are installed underground for displaying monitoring data of various sensors, huge waste of information is prevented, the practicability, visualization, intuitiveness and transparent degree of coal mine monitoring information and management scheduling are improved, underground safety production and efficient management are realized, and the system is simple in operation, high in practicability and convenient in staff scheduling.

The invention provides a processing method and device for predicting aerospace product assembly hours. The processing method comprises the steps of creating a process classification library and a textcorpus; sequentially selecting pre-selection processes in the process card database, and obtaining a pre-selection process text and a pre-selection text corpus of the pre-selection processes; performing text mining on the pre-selected process text according to a pre-selected text corpus to obtain a pre-selected text feature vector; when determining that all the processes in the process card database have been traversed, training a man-hour calculation model corresponding to each process type, and storing the man-hour calculation model in a man-hour calculation model library; and obtaining a man-hour prediction value of the to-be-executed process according to the description text of the to-be-executed process and the man-hour calculation model library. According to the invention, the man-hour calculation model based on the process text features is constructed to carry out man-hour prediction, the problem that the current man-hour prediction is not suitable for aerospace products is solved, the accuracy and effectiveness of aerospace product assembly time prediction are ensured, and the production efficiency is further improved.

The invention provides an assembly time acquisition method and a processor, wherein the method comprises the steps: processing a process text in a target process document in a text mining mode, and determining a target process type of the target process text; obtaining a target neural network model corresponding to the target process type according to the target process type of the target processtext; and inputting the time influence factors in the target process text into the corresponding target neural network model, and obtaining an assembly time output by the target neural network model and corresponding to the target process text. According to the embodiment of the invention, process-level process classification is carried out on a process document in a text mining form, the process-level time prediction model is constructed by acquiring the neural network model corresponding to the process type, and the time corresponding to the process text and the total time corresponding to the whole process document are acquired on the basis of the process-level time prediction model; the accuracy and the efficiency of time prediction are improved, and the estimation accuracy and the management efficiency of an enterprise for the time quota are further improved.

The invention discloses a blast furnace tapping end point online intelligent identification method and system, and the method comprises the steps: collecting a molten iron flow image, extracting molten iron flow image features, collecting molten iron flow data, extracting molten iron flow short time sequence features, fusing the molten iron flow image features and the molten iron flow short time sequence features, and building a tapping process classification prediction model; the blast furnace tapping end point is identified according to the tapping stage of the current moment, the technical problem that the current tapping end point cannot be accurately identified in the prior art is solved, the image, the flow velocity, the slag-iron ratio and other data of the high-temperature molten iron flow are obtained through a non-intrusive device, and the identification accuracy of the blast furnace tapping end point is improved. The features of the multi-source data of the molten iron flow are extracted through the deep neural network for fusion, so that the blast furnace tapping end point at the current moment is recognized, the confidence coefficient of the current recognition result is provided, and the method has the advantages of being high in accuracy, stability and adaptability, suitable for blast furnaces of different scales, low in investment cost and the like.

The invention is suitable for the field of information processing, and provides a method and a system for constructing a service pattern library. The method comprises the following steps of: receiving enterprise commercial pattern information and enterprise value chain information which are input by a user; determining a core service link of an enterprise by a quality house analysis method according to the received enterprise commercial pattern information and enterprise value chain information; determining a core service process in the core service link of the enterprise by using a process classification framework (PCF); extracting service pattern characteristics and application characteristics to construct a service pattern characteristic table of the enterprise based on the received enterprise commercial pattern information and enterprise value chain information, the determined core service link and core service process of the enterprise, and a research result; and storing the service pattern characteristic table of the enterprise, a service process analysis table and contents such as an enterprise overview, an enterprise organizational architecture, basic data management, a service control content and the like in the research result in a database. By the method and the system, a complete and high-value service pattern library can be quickly constructed.

The invention discloses a design method of a multi-class center classification network model, and relates to the technical field of deep learning. According to the method, the problem that the model performance is reduced when a CNN + FC + Softmax framework processes classification data sets with high feature diversity is solved; the thought of a Gaussian mixture model is used for reference; the invention provides a method for setting a plurality of clustering centers for each classification. An optimal value selection mechanism is adopted; different pictures with large feature differences inthe same class can be gathered towards a feature center point most suitable for the pictures; by means of the design method, all the center points in the class can be driven to be separated and prevented from being bonded together, the richness trend of class characteristics is increased, meanwhile, the separability and the judgeability of the model are improved, the training speed is increased, the diversity characteristics of all the classes are reserved, and the problem that the performance of the model is reduced is solved.

Decomposing a process model in an Enterprise Intelligence (‘EI’) framework, including: receiving, by a process model decomposer, a process model for an EI assembly; identifying, by the process model decomposer in dependence upon an EI framework metamodel, component parts of the process model; and generating, in dependence upon the identified component parts, a taxonomy breakdown comprising a definition of a process classification.

The invention discloses a data processing classification storage system, comprising storage device classification, a storage network structure and backup, wherein an SCSI storage device, an SAS storage device, an FC optical storage device and an ISCSI storage device are arranged in the storage device classification, a DAS storage system network structure, an SAN storage system network structure and an NAS storage system network structure are arranged in the storage network structure, a backup cycle, a backup mode, automatic / manual backup and a backup medium are arranged in the backup, the SCSIstorage can conveniently implement the connection with a server and can directly implement the connection with the server through a standard SCSI cable, the overall structure is simple, and the linksare few, so the system has fewer failure points, correspondingly, the system has high security and stability, the complete backup and differential backups can be conveniently and frequently combinedinto a complete backup scheme, the differential backup is performed daily, and the complete backup is performed periodically, in this way, a system administrator only needs two copies of magnetic tapedata during restoration.

The invention discloses a computer data mining and clustering method based on a time sequence. According to the computer data mining and clustering method, denoising and normalization processing is performed on input sample sets X and Y, and extreme point solving is performed on the processed time sequence so that extreme point sequences X' and Y' are obtained; then isometric processing is performed on X' and Y', and classification sequences X" and Y" with equal length are obtained after isometric processing; then class distance calculation is performed on the processed sequences X" and Y", the two classes of the maximum distance are combined and one class is reduced after combination; then class distance calculation is cyclically performed on the processed classification sequences X" and Y" and two classes of the maximum distance are combined until the number of clusters is equal to preset data and then clustering ends; and finally a clustering result is outputted. According to the method, the time sequence data of high data volume and high dimension can be effectively processed, the method is easy and practicable without depending on concrete sequences, data mining and clustering can be effectively performed, and mass data can be effectively compressed and the main characteristics of the data can be maintained.

The present invention provides an input method of symmetry code radical stroke of a computer number keyboard, characterized in that the method uses 1-0 of the number keyboard on the top of the computer main keyboard to process classification and coding according to the rule of symmetry structure and asymmetry structure of the characters, 1-5 number keys are processed with stroke codes. The input method includes following steps: firstly, inputting the character symmetry code classification key of the number keyboard, then determining the radical of the input character by the direction keys or mouse, finally, inputting the strokes of the parts except the character radical by the number keys 1, 2, 3, 4, 5. The invention symmetry classification of the character structure is accordance with the understanding habit and the comprehending habit of the characters. The character input method of the invention simply and quickly inputs any character of any large scale Chinese character library, improving the input speed of rarely used Chinese characters and rarely seen Chinese characters.

The present invention discloses a personal health record system with a process decision support function. By constructing a personal health record system with separated characteristic data extraction and characteristic data analysis, characteristic data that meets analysis requirements is expected to be rapidly obtained. A statistics query interface based on characteristic data and time logic is provided, and decision support for users is completed through steps such as characteristic matching, process classification and statistics evaluation, and process recommendation, so as to make medical big data better execute high-level information analysis and decision support functions.

The invention discloses a power distribution network evaluation, diagnosis and analysis method, device and equipment and a storage medium, and the method comprises the following steps: S1, obtaining atested state parameter of tested power distribution network equipment; and S2, inputting the tested state parameters into a pre-trained decision tree model to obtain an output evaluation analysis result and a processing strategy of the tested power distribution network, and obtaining a strategy classification model based on training of the decision tree model. According to the invention, the fault processing classification model of the power distribution network equipment is trained based on the C4.5 decision tree algorithm, and the problems that unified data analysis is lacked in fault problem sorting of the power distribution network, the quality deviation of technicians is large, and problems are hard to distinguish and cannot be solved in time are solved. And an auxiliary decision-making method is provided for diagnosis analysis and fault processing of the power distribution network.

The present invention provides an input method of symmetry code radical stroke of a computer small keyboard, characterized in that the method uses numbers 0-9 of the small keyboard to process classification and coding according to the rule of symmetry structure and asymmetry structure of the characters, then 1-5 number keys of the number keyboard are processed with stroke codes. The input method includes following steps: firstly, inputting the symmetry code classification key of the number keyboard of the small keyboard, then determining the radical of the input character by the direction keys or mouse, finally, inputting the strokes of the parts except the character radical by the number keys 1, 2, 3, 4, 5. The invention symmetry classification of the character structure is accordance with the understanding habit and the comprehending habit of the characters. The character input method of the invention simply and quickly inputs any character of any large scale Chinese character library, improving the input speed of rarely used Chinese characters and rarely seen Chinese characters, then 1-5 number keys of the number keyboard are processed with stroke codes.

The invention discloses a rapid processing method for batch alarm which comprises the following steps: distinguishing an alarm treatment as emergency operation and non-emergency operation, separating the emergency operation and the non-emergency operation, establishing high priority task TASK_U for the emergency operation, and establishing low priority task TASK_D for the non-emergency operation; and collecting more alarms actively through an alarm packing assembly when generating a plurality of alarms simultaneously, packing emergency operations in the same batch, and delivering the pack to TASK_U for processing. By employing the method in the invention, alarm processing performance of substantial alarms which are generated simultaneously is improved without causing emergency operation overtime of after alarms. The invention also disclose a rapid processing system for batch alarm which comprises an alarm processing classification module, a task creation module, a task execution module and an alarm packing assembly.

The invention relates to a data preprocessing method and device for a business model. The embodiment of the invention provides a novel service model for processing classification services. The servicemodel is realized through a plurality of deep networks, and in the plurality of deep networks, the importance of each service feature for the corresponding classification category can be fully considered in each deep network by introducing feature expression vectors for describing different feature values and each layer label vector corresponding to each classification category. When the businessmodel is used for determining the target category; according to the technical scheme, the possibility that each classification category serves as the target category of the to-be-processed service data can be determined according to the classification category, so that the accuracy of the service model is improved, the service processing result of the service model has traceability due to the fact that the importance coefficient of the corresponding service feature is determined in each deep network, and the use experience is improved.

The invention relates to a malicious encrypted channel detection method based on process behavior analysis. The method comprises the following steps: step 1, acquiring encrypted session flow data and classifying processes; comprising the steps of encrypted traffic data acquisition, preprocessing and process classification; 2, constructing a process file encryption communication behavior feature tree; constructing a three-level feature tree of a meta feature of the encrypted session, namely an IP layer, a TCP segment load length sequence feature, namely a TCP layer, and an SSL message state transition feature, namely an SSL recording layer; step 3, anomaly detection based on the feature tree; comprising the steps of collecting normal encrypted communication behavior data; constructing a normal encryption communication behavior benchmark; constructing a target PEF encrypted communication behavior model; calculating the dissimilarity degree between the feature trees; and performing anomaly detection based on a threshold value. By means of an anomaly detection method, the limitation that malicious encrypted channels are detected purely from the session level is broken through, and effective detection of malicious process files is achieved.

The invention discloses a system and a method for collecting information based on a server and a client framework, comprising the following steps of: (1), collecting data of an suggestion message content of a client; (2) transmitting the data of the suggestion message content to the server through network self-adaption, wherein the transmission realizes data transmission through using a network communication self-adaption module which determines the way to be adopted to realize unordered sending and ordered receiving of the data according to a concrete network environment apparatus and a position where the client is actually located; and (3) carrying out buffer storage on the data of the received suggestion message content by the server and carrying out processing classification on the data of the received suggestion message content by using a data mining module. The scheme provides an indeed effective information collection way on requirements of intending to hear different suggestions of various parities for various industries so as to lighten the work load and really save the human and material resources.

Decomposing a process model in an Enterprise Intelligence (‘EI’) framework, including: receiving, by a process model decomposer, a process model for an EI assembly; identifying, by the process model decomposer in dependence upon an EI framework metamodel, component parts of the process model; and generating, in dependence upon the identified component parts, a taxonomy breakdown comprising a definition of a process classification.

The invention discloses a configurable software dynamic interface loading and data processing method, and relates to software research and development technology. This method divides the business into multiple sub-processes, sets multiple configuration items for each sub-process, adds attributes of configuration items for customers according to requirements, forms a configuration table, and then loads data in the software interface according to the configuration table, realizing the realization of Different sub-processes are individually configured, thereby reducing the adaptation workload, and at the same time, it can support the individual needs of different users for data processing requirements in the case of the same main process, and realize the processing of different management information The effect of process classification configuration, because different sub-processes are displayed on a separate interface, the display information of the software interface is concise and clear.

The invention discloses a wide book classification mechanism. At present, most book returning operations of a book museum are still manual operations, and related facilities capable of continuously and intensively processing classification work are lacked. A support is arranged on one side of a classified storage cabinet, a conveying assembly is arranged on the support, a book carrying plate is arranged above the conveying assembly, a plurality of book grooves are formed in the classified storage cabinet side by side, each book groove is correspondingly provided with an inclined groove body, and the lower end of each inclined groove body is communicated with the corresponding book groove. The high ends of the multiple inclined groove bodies are attached to one side of the book carrying plate, a barrier piece is arranged between each inclined groove body and the book carrying plate, a rack motor is correspondingly arranged below each barrier piece, each barrier piece is driven by the corresponding rack motor to do lifting reciprocating motion in the vertical direction, and a book containing groove is obliquely formed in the other side of the book carrying plate. A book inlet is formed in the high end of the book containing groove, the lower end of the book containing groove is connected with the book carrying plate, and a rotating partition plate is arranged in the book containing groove. The mechanism is used for centralized continuous classification of a large number of books.

The invention discloses a method for diagnosing multiple faults of a power distribution network under imperfect fault information. The method comprises the steps as follows: firstly, obtaining a diagnosis decision table by existing fault information of a power distribution network; then extracting effective information from massive alarm information by complete process classification, and correcting error information in uploaded fault information; finally, writing diagnosis rules into a matrix form, quantifying the probability that a sample to be diagnosed conforms to the diagnosis rules, andobtaining a diagnosis result with the maximum probability. The method can quickly and effectively realize fault diagnosis, and reliability, safety and flexibility of system operation are greatly increased.

The invention discloses a method for displaying custom to-do data. The method comprises the steps of receiving to-be-handled display parameters input by a user from an editing window, wherein the to-be-handled display parameters at least comprise to-be-handled names, to-be-handled classifications, query values, to-be-handled menus and filtering conditions; extracting data corresponding to the filtering condition from a business form defined by the to-be-handled menu information according to the to-be-handled filtering condition to form to-be-handled data; performing statistics on the to-be-handled data according to a condition or a function corresponding to the query value to generate to-be-handled summary information; dividing the display window into a plurality of display areas in the first direction according to the to-be-processed classification; dividing the display area into a plurality of to-be-handled modules in a second direction according to the to-be-handled names; and displaying the summary information on the to-do module in real time. According to the embodiment of the invention, the technical problem that the to-do list is difficult to monitor in the prior art is solved, and the user can conveniently and autonomously monitor the interested process node information.

The invention discloses a monitoring system for working dynamic characteristics of fire-fighting vehicles. The system mainly comprises a working dynamic characteristic information acquisition module for acquiring chassis working state information of the fire-fighting vehicles and loading working condition information of the fire-fighting vehicles, an information processing classification module for processing the acquired information and an analysis module for determining the chassis using state and the maintenance period of the fire-fighting vehicles by combining the processed chassis workingstate information of the fire-fighting vehicles with a weight cases algorithm and a management application model built in the analysis module, and through the analysis module, the processed loading condition information of the fire-fighting vehicles is combined with a command operation model built in the analysis module to determine the vehicle command and dispatching scheme of the fire-fightingoperation. By using the monitoring system, the real-time monitoring of the working dynamic characteristics of the fire-fighting vehicles is realized, the problems of the daily maintenance of the fire-fighting vehicles and the on-site command and dispatching and the problem that reinforcement force have no data guidance are solved, and the digitalization and intelligence of the daily management andcommand and dispatching of the fire-fighting vehicles are improved.