286 results about "Deep mining" patented technology

A big data real-time storage, processing and querying system includes a distributed support module, a data acquisition module, a message-oriented middleware module, a data cleaning module, a data processing module, a data deep mining module, a data management module, a data querying module, and a system scheduling module; all the modules run under the coordination of the system scheduling module, and scheduling and circulation of data flows among all the modules can be achieved Processing flows among all the modules can be coordinated, and the date real-time performance is improved; effective improvement is performed for high throughput, mass data, fast response messages and data high availability; and then the comprehensive ability of a big data real-time data service system can be improved.

The invention discloses a recommendation system click rate prediction method based on a deep neural network, and the method comprises the steps: collecting a user click behavior as a sample, extracting numerical features of the sample with a numerical value relation, and inputting the numerical features into a GBDT tree model for training, and obtaining a GBDT leaf node matrix E1; inputting a behavior sequence formed by clicking the articles by all the users in the sample into an Attention network to obtain an interest intensity matrix E2 of all the users in the sample for the articles; summing and averaging the article feature vectors of the click interaction of the user to obtain a click interaction matrix E3 corresponding to the user, splicing E1, E2 and E3, and inputting the E1, E2 andE3 into a deep neural network model with three hidden layers and one output layer to output a prediction result. According to the method, user clicking behaviors are decomposed into attribute characteristics, a GBDT tree model, an Attention network and a deep neural network model are subjected to nonlinear fitting, a recommendation system clicking rate prediction model is constructed, a prediction result is obtained through model training, and the method has the advantages of deep mining of recent interests of users, high generalization degree and high expansibility.

The invention discloses an underground pressure three dimensional physical simulation test platform, which is characterized in that a frame type model platform used for bearing a model, a swing structure used for causing the frame type model platform to incline in a whole in a range of set maximum inclination angle, and a loading device used for pressing on the model in the direction of the opening of the frame type model platform are included, wherein, the frame type model platform comprises an upper opening, side wall plates arranged at the front, at the back, at the left and at the right, as well as a base plate; the base plate is provided with at least a mining hole, and the mining hole is provided with a demountable sealing lath. The invention can simulate a big inclination angle coal seam mining experiment and simulate a coal seam deep mining experiment, and meanwhile, can also simulate a bottom plate coal mining experiment, simulate a middle coal mining experiment, and the like.

The invention discloses a power grid comprehensive visualization method based on multi-service system information fusion. On the basis of comprehensively organizing service flows and data models of links of an intelligent power grid, an information exchange bus crossing a production control region and a management information region is utilized to obtain relative operational and management information of systems such as a production management system (PMS), a distribution management system (DMS), an energy management system (EMS), an electric transmission and transformation equipment state monitoring system, a distribution network production quick-repairing commanding platform, a centralized video monitoring system, a marketing management system, an electricity consumption information acquisition system, a geographical information system (GIS), an intelligent district energy consumption management system, a residential area intelligent electricity consumption interaction and energy efficiency management system, an electric vehicle charging facility comprehensive monitoring and management system and a weather system, and furthermore, based on the relative operational and management information, the data is subjected to deep mining, analysis and comprehensive displaying. The power grid comprehensive visualization method based on multi-service system information fusion helps the production operational and management personnel to master the real-time running state of the power grid more comprehensively, visually and effectively.

The invention relates to a coal mine gas disaster monitoring and pre-warning system and a pre-warning method. According to the system and the method, automatic monitoring, transmission, real-time acquisition, storage and dynamic sharing of coal mine gas disaster related safety information are firstly achieved through a gas disaster safety information acquisition platform; then, deep mining and analysis are conducted on the related safety information by utilizing the gas disaster safety information acquisition platform according to gas disaster monitoring and pre-warning indexes and rules established on the four aspects of objective dangerousness, precautionary measure defects, management measure defects and catastrophe control, gas disaster danger level and range are automatically judged, and pre-warning results of different levels are issued in sound, light, short message and network forms and the like; finally, emergency linkage control of a coal mine power monitoring system, an underground personnel location system, an underground broadcasting system, an online ventilation monitoring system and other coal mine monitoring systems is achieved by utilizing a pre-warning linkage control software system according to the issued pre-warning results, and finally the purposes of online monitoring, real-time analysis and intelligent pre-warning and linkage control of coal mine gas disasters are achieved.

The invention discloses a continuous monitoring system of overburden rock displacement in an ore bed mining process, which comprises an inclination measuring drilling hole, an inclination measuring pipe assembly, a monitoring device and a monitoring station. When in use, a plurality of fixed inclinometers are arranged at different positions as required so as to realize the continuous whole-coursemonitoring on mined overburden rock displacement; meanwhile, the monitoring device for a flexible chain structure can be used for preventing the damage of overburden rock movement on instruments so as to ensure monitoring is carried out smoothly; and the monitoring system is used for evaluating the space-time influence degree of the mining to overburden rock so as to provide a basis for the judgment of the influenced degree and the safety of deep mining operation execution, facilities in an overburden rock stratum, ground facilities, and the like. In addition, the invention also discloses a method for structuring the monitoring system, wherein methods for embedding an inclination measurement pipe, carrying out anti-sliding safety protection during lowering the inclination measurement pipe, lowering the inclination measurement pipe when water or no water exists in the drilling hole, fixing the inclination measurement pipe by deep hole grouting, and the like not only effectively solve the problem of huge difficulty on installing deep-hole inclination measurement pipes, but also effectively improve the construction efficiency and the safety.

The invention belongs to the technical field of generator set equipment fault diagnosis, in particularly to a fault diagnosis method based on fault data deep mining and learning, which comprises the steps of collecting and preprocessing the historical data of a generator set, then learning and training through a deep long-short time memory network algorithm, obtaining a fault data screening model, and then traversing a large number of historical databases, and screening and forming a fault data sample set; estimating the number of the fault types of the fault data sample set by adopting a Medoids surrounding classification method, and adopting K-Means clustering algorithm for clustering analysis to form a multi-class typical fault sample set; training and learning the multi-class typical fault sample set by adopting an LSTM neural network algorithm, and establishing a fault diagnosis model; and monitoring the real-time operation data of the system based on the fault diagnosis model, judging the operation state of the system, recording a newly generated fault sample, and updating the fault diagnosis model by using the updated multi-class typical fault sample set.

The invention provides a drainage internet-of-things system comprising a sensing layer, a network layer and an application layer. The sensing layer is located at the bottommost layer of the drainage internet-of-things system and composed of various transmission units; the network layer is located in the intermediate layer of the drainage internet-of-things system, and is a link for linking a monitoring center and each field site. The network layer realizes bidirectional information transmission between the application layer and the sensing layer through an existing wired network, a wireless network and the like; the application layer is located in the topmost layer of the drainage internet-of-things system, acquires the sensing layer mainly based on data analysis methods such as correlation analysis and cluster analysis or a hydrodynamic model, and performs summary analysis and deep mining through data transmitted by the network layer so as to guide management and operation maintenance of a drainage system. By the application of the drainage internet-of-things system, a traditional drainage management mode is changed, running supervision, scientific warning and unified scheduling of the drainage system are realized, and intelligent and digital management level of the drainage system is increased effectively.

The invention discloses a demonstration-based photovoltaic power station testing diagnosis and forecasting database establishment method. In the method, deep mining and probability reliability assessment are performed on the statistical data of different components and balancing parts in a demonstration system based on long-term running performance testing data by using a reliability analysis principle to form reliability and economy-coordinated comprehensive evaluation mechanism and establish a photovoltaic power station field fault diagnosis and maintenance forecasting database. The method has the advantage of comprehensively evaluating the reliability, running performance and cost of photovoltaic components, the parts and the system.

The invention provides a main control and roadway retaining method for anchor injection and top-cutting in a deep mining roadway, which comprises the following steps: step 1, determining the technicalparameters required for realizing the top-cut and roadway retaining in the deep mining roadway; step 2, carrying out the grouting anchor cable support construction during the tunnelling period of thedeep mining roadway; step 3, carrying out the advanced overcut and perforating construction at the roadway side during the mining of the working face; step 4, carrying out grouting anchor cable support construction during the mining of the working face; step 5, carrying out monomer strut support construction at the roadway side during the mining of the working surface; step 6, carrying out auxiliary support construction. The method has the advantages that the supporting requirements of the deep gob-side entry can be satisfied when carrying out non-pillar mining construction in the deep mine.

The invention discloses a multi-factor short-time traffic flow prediction method. The method comprises the steps that 1, distances between a specific detector and meteorological stations are calculated, the nearest meteorological station is screened out, and weather data of the meteorological station serves as weather data of the detector; 2, historical traffic flow data and historical weather data of the detector are preprocessed respectively and merged according to the time; 3, a plurality of features are constructed, and feature screening is carried out on the basis of LightGBM; 4, modelingis carried out on time sequence features and periodic features of the traffic flow data through LSTM; 5, the time sequence features and the periodic features of the traffic flow are fused with the features screened out in step 3 through a fully-connected network in a neural network; 6, the model is trained, and the short-time traffic flow is predicted. The method overcomes the defect that existing methods cannot fully use known data, can be used for deeper mining and analysis of the traffic flow data, and is high in prediction accuracy and good in robustness.

The invention relates to a text classification method based on a graph kernel and a convolutional neural network, and belongs to the technical field of data mining and information retrieval. The coreconcept is that: firstly, a text is preprocessed into an expression mode of a graph structure, wherein nodes in a graph correspond to words in the text; then weights of the nodes are calculated on thebasis of the graph structure, then the graph structure is decomposed into a plurality of sub-graphs by using a community discovery algorithm, the graph is mapped to a high-dimensional space by usinga graph kernel technology so as to obtain a tensor expression of the graph; finally, the tensor expression is input into the convolutional neural network, deep mining is carried out on graph features,and a category of the text is output. Compared with the prior art, according to the text classification method, an internal structure and context semantics of the text can be sufficiently utilized, so that text contents are sufficiently expressed; node information is more reasonable; the problem of complex and fussy processing process in the text classification is effectively solved.

The invention discloses an event modeling-based event personalized classification method and system. The method comprises the steps of 1, performing data crawling on daily news, corresponding microblogs and reviews on various forums, and establishing a news library; 2, performing deep mining on data in the news library, building corresponding event models, generating a plurality of tags for each event, and storing event information in the event library; 3, guiding different users to input required different classification conditions and keywords of all categories through a client; and 4, based on the event tags and other attributes, performing different category division on events in the event library according to category matching and category similarity, and visually displaying the events on a user platform. The system can perform personalized classification on news event data according to different requirements by fully utilizing the fast-developing internet and a large amount of emerged news data at present, can be fully connected with various news platforms conveniently, and can provide personalized information services for various platforms.

The invention discloses a road traffic accident risk factor analysis method based on a random forest. According to the road traffic accident risk factor analysis method, influence factors related to four aspects of people, vehicles, roads and environments and corresponding accident severity data are extracted when traffic accidents occur; a road traffic accident severity prediction model is established according to a random forest prediction model algorithm; model training is carried out, and model parameters are adjusted; and then variable importance measurement is carried out on the established prediction model by using a Gini exponential method, and importance sorting is carried out on related risk factors, and factors which have a remarkable influence on the severity of the accident are identified. According to the road traffic accident risk factor analysis method, influence factors of road traffic accidents, deep mining risk factors and accident severity can be considered in many aspects, and road safety management is facilitated.

The invention provides an advertisement management system operating on a CMMB (China Mobile Multimedia Broadcasting) super hotspot service platform. The advertisement management system comprises a management server, a management terminal and a user terminal, wherein the management server is communicated with the CMMB super hotpot service platform, an advertiser accesses to the management server through the user terminal, the management terminal accesses to the management server through a WEB application program, the management server comprises a storage module and an advertisement order management module, an advertisement space management module, a material management module, an advertisement management module, an advertisement delivery management module and an advertisement statistics module which are communicated with the storage module. The advertisement management system provided by the invention is used for delivering different types of advertisements, and can perform relevant scheduling on the advertisements at the same time, thereby realizing advertisement resource management and full-service precise advertisement delivery management based on deep mining and intelligent analysis for user behavior data.

The invention discloses a project data management platform for a power supply company. The project data management platform comprises a mobile client, a WEB client and a server, The mobile client is used for logging in the server to perform project management work and realize project initialization information acquisition, information acquisition in project construction and project display; The WEB client is used for logging in the server and carrying out management of platform user permission, data import, data export and summarization; And the server is used for providing storage and management services of project data of a power supply company for the mobile client and the WEB client. According to the platform, unified supervision of a decentralized development system can be realized, aplatform for effective utilization of data is provided, and multi-dimensional analysis of data integration is realized, so that advanced practicability of deep mining of the data and flattening monitoring of the data in an abnormal state are achieved.

The invention discloses a high-voltage circuit breaker mechanical fault diagnosis method. Opening and closing coil current obtained by a high-voltage circuit breaker fault simulation experiment is used as a target data sample for fault diagnosis; data obtained through simulation by establishing an opening and closing coil mathematical model serve as an auxiliary data sample, deep mining and self-adaptive extraction of sample data features are achieved through a deep belief network (DBN), and information matching of the auxiliary data and target data is achieved in combination with a transfer learning method. According to the method, transfer learning and a deep belief network are combined; the deep belief network is utilized to carry out deep mining and self-adaptive extraction on the current time domain signal of the opening and closing coil of the circuit breaker, the transfer learning method is combined to solve the problem that the actual fault training sample data size is small, and the generalization capability of the fault diagnosis model is improved.

The invention discloses an analytical method of deep mining floor rock stratum damage and fracturation evolution and failure law. The analytical method comprises the following steps: Step 1, stope floor stress state data is acquired; Step 2, stope floor rock stratum damage and fracturation data is acquired; Step 3, deep well floor rock stratum damage and fracturation data is acquired; Step 4, deep well floor mining-induced damage and fracturation data is acquired; Step 5, the above stope floor stress state data, the stope floor rock stratum damage and fracturation data, the deep well floor rock stratum damage and fracturation data and the deep well floor mining-induced damage and fracturation data are analyzed so as to obtain deep mining floor rock stratum damage and fracturation evolution and failure law. By obtaining the deep mining floor rock stratum damage and fracturation evolution and failure law, data foundation is provided for specific deep mining, so as to further provide technical support for raising deep mining efficiency and mining safety.

The invention provides an intelligent sensing and remote and unified regulating system for deep-mining mine ventilation. The intelligent sensing and remote and unified regulating system is mainly composed of intelligent sensing equipment, a PLC control cabinet (3), a middle-layer optical fiber ring network (18), a ring network exchanger (17) and an industrial personal computer (19). The PLC control cabinet (3) is provided with an analog input terminal (4), a frequency converter (12), an analog input module (13), a communication module (14) and a CPU module (15), wherein the analog input terminal (4) is connected with the intelligent sensing equipment through a cable, the frequency converter (12) is connected with a ventilation motor (2) and can automatically acquire operation parameters of the motor, and then the frequency converter is communicated with the industrial personal computer (19) of an upper-layer monitoring terminal through the CPU module, the communication module, the analog input module, the middle-layer optical fiber ring network and the ring network exchanger. The intelligent sensing and remote and unified regulating system is strong in expansibility and improves the unified regulating efficiency and the real-time controlling and optimizing capacity to multiple fans in deep-mining mines.

The invention discloses an individual brain function mapping method based on electrocorticogram high-frequency Gamma nerve oscillation and belongs to the field of nerve engineering. The individual brain function mapping method based on electrocorticogram high-frequency Gamma nerve oscillation comprises the steps that collected ECoG data are processed, the correlation synchronization intensity of time-frequency events is calculated, statistical significance testing is conducted and a brain function index is extracted, and individual brain function mapping is conducted. According to the individual brain function mapping method based on electrocorticogram high-frequency Gamma nerve oscillation, through the deep mining of the ECoG data, relevant electroencephalogram data of brain function areas can be analyzed rapidly, accurately, comprehensively and systematically, the brain function index is obtained, individually targeted brain function mapping is achieved, people can better understand brain mechanisms for processing complicated cognition tasks, and great assistance is provided for study on cognitive neuroscience brain functions and fundamental study on clinic neurosciences.

The invention discloses a target track optimization method based on information entropy weight and nearest neighbor data association. The method comprises the following steps: obtaining measurement vectors of a measurement point at the first two moments after track initiation, and calculating an initial state vector of a target; carrying out initialization on a Kalman filter; calculating a state estimation vector, an estimation error covariance matrix, a prediction vector, an innovation covariance matrix and Kalman gain of the target at the moment k through an iteration method; obtaining candidate measurement points falling into the center of a tracking gate at the moment k; and when the number of the measurement points at the moment k is larger than 1, optimizing and updating the target track at the moment k through an entropy weight method. Through deep mining of known measurement information, and on the basis of not changing the priori information, the method makes full use of the information carried by the target measurement point, so that the method has lower transformation cost and better applicability, and meanwhile, improves accuracy of radar track tracking.

The invention discloses a testing system for an evolution process of a deep mining stress field, and belongs to the field of testing systems for the stress evolution of working faces. The system can perform non-uniform dynamic loading on large coal according to mechanical conditions of the coal and really simulate the mechanical conditions of the coal in elastic and plastic states, and comprises a test loading component, a servo component, a control component and an acoustic wave monitoring component, wherein the test loading component comprises a coal sample, a main loading frame, two side pressure loading units and an axial pressure loading unit; the axial pressure loading unit is provided with five independent loading points; the two side pressure loading units are arranged on the two sides of left and right upright posts; the side pressure loading unit on each side is provided with three independent loading points, and different loads can be loaded by adjusting the loading points above a top plate and the loading points on the left and right sides to realize the non-uniform loading of the coal.

The invention provides a data mining system used for an enterprise innovation system. The data mining system comprises a science and technology innovation system decision assisting and supporting subsystem, a science and technology enterprise innovation knowledge mining subsystem, a science and technology innovation service subsystem and an innovation resource catalogue service subsystem. The science and technology innovation system decision assisting and supporting subsystem comprises an information collection module, an analytical prediction module, a decision making module and a decision execution module, wherein the information collection module is used for finding mass data on a knowledge base or a network and comprehensively and accurately obtaining all types of basic data related to items where decision making is to be conducted. The data mining system has the advantages of being comprehensive in function and high in intelligent degree, the effect of conducting deep mining on enterprises is achieved, and enterprises are scientifically guided to conduct technical innovation.

The invention discloses a simulation test method for heat and humidity exchange of a heading tunnel in deep mining of a metal mine. The simulation test method comprises the steps of arranging a special simulation test platform and implementing operations, wherein the arranged simulation test platform is formed by combining a simulation tunnel, a tunnel heat source simulation device, a tunnel internal humidity environment simulation device, a tunnel internal fresh air quality simulation device, a tunnel cooling and refrigerating simulation device and airflow parameter monitoring and controlling components; the operations implementation step is as follows: determining the working condition of a refrigerating unit when an internal heat source of the heading tunnel and an air blowing parameter of blown fresh air are known; verifying theoretically derived parameters of the refrigerating unit. According to the simulation test method for the heat and humidity exchange of the heading tunnel in the deep mining of the metal mine, which is disclosed by the invention, a field real heat and humidity exchange environment is provided for the study on the heat and humidity exchange of internal air of a blind drift, the simulation test method and theoretical calculation form a mutually complemented system, the reliability of various theoretically calculated parameters can be greatly verified, the reliability of theoretical calculation is improved, and excellent practical significance and popularization and application values are provided to solve a heat damage problem in the face of the deep mining at present.

The invention discloses a thin-layer limestone ground directional drilling horizontal multi-stage branch grouting drill hole. The thin-layer limestone ground directional drilling horizontal multi-stage branch grouting drill hole comprises a ground main hole. The ground main hole vertically extends to the underground, penetrates through a vertically-drilled straight hole section of a base rock layer from the ground to the underground and downwards penetrates through an ash consequently-drilled bedding section along a coal bed. A kickoff point is arranged at the end, away from the ground and close to the coal bed, of the straight hole section. Inclined drilling is conducted between the kickoff point and the middle of lime-cement-flyash below the coal bed, so that a directional inclination guide section is formed. The two ends of the directional inclination guide section communicate with the straight hole section and the bedding section correspondingly. The main hole bedding section is divided into a plurality of first and second stage branch holes distributed in a fan shape in the approximately horizontal direction of the bedding section. By adopting the technical scheme, compared with the former conventional water control technology, the safe reliability is greatly improved, deep mining bottom plate limestone water disaster governance is achieved, bottom plate water disaster accidents are avoided, and safety production is ensured.

The invention discloses a power grid GIS-based multi-business system information fused power grid visualization method. The method comprises the following steps of: obtaining information of six business links such as power transmission, power transformation, power distribution, power utilization, dispatching and communication of an intelligent power grid by utilizing an information interaction bus; and establishing a visualization platform which comprises power grid GIS panorama simulation, power grid running state monitoring, distribution network running state monitoring, power utilization energy efficiency service monitoring, clean energy application and power-off management application in a display and application manner. Through carding business flows and data models of the links of the intelligent power grid, and carrying out deep mining, analysis and comprehensive display on data on above basis, the running and management personnel are helped to grasp the real-time running state of the power grid more comprehensively, visually and effectively in a more intelligent and richer power grid visualization monitoring, alarm and analysis manner, so that the company running management decisions are intelligently assisted and supported, and the method has favorable application prospect.

The invention relates to a medical data joint learning system and method based on trusted computing and privacy protection. A joint learning center control layer receives non-sensitive meta-information uploaded by a data contributor through a data contributor management layer of a data node where the joint learning center control layer is located for filing, and original data is locally registered, stored and subjected to isolation calculation; and the joint learning center control layer processes a joint learning request initiated by the data miner through the data miner interaction layer, summarizes insensitive intermediate results obtained by performing local isolation calculation on the data nodes based on the original data in a safe calculation area, and returns a final joint learningresult to the data miner interaction layer. The invention provides a whole set of service system based on medical big data security sharing, trusted computing, deep mining, authority authentication and multi-platform joint learning, and solves the problems of scattered, single and incomplete medical data privacy protection and data mining at the present stage.

The invention relates to the field of mining. A gob expanding material prestressed filling method is characterized by comprising the following steps of (1) preparing a prestressed filling material, i.e. stirring and blending the following raw materials in percentage by mass, 20 to 80 percent of burnt lime powder, 20 to 50 percent of water, 0 to 60 percent of filling aggregate and 0 to 5 percent of additive, to obtain the prestressed filling material; (2) according to the actual needs of a mine, on the basis that all the prestressed filling material or a part of filling material which is originally used by the mine is filled into a gob, filling the prestressed filling material into the gob through a filling pipe until the prestressed filling material reaches the top of the gob. After the prestressed filling material enters the gob for a certain time, the volume of the prestressed filling material is expanded; after the whole gob is fully filled, the prestressed filling material applies certain pre-stress to the surrounding rock mass. The problem that the mine is difficultly filled to the top is solved by the method and the filling material, the deformation and displacement of the surrounding rock mass are fully restrained, the stress distribution of the surrounding rock is improved, and a good condition for deep mining and the mining in a complex ground stress environment is also created.

The invention discloses a source code file multi-service label automatic classification method. The method comprises four parts of data preparation, data preprocessing, word vector representation andmulti-label classification. According to the method, semantics in codes can be mined more deeply through multi-label classification learning based on deep learning; the method comprises steps of decomposing the program file into an annotation part and a code part by utilizing the text characteristics of the program file, and respectively encoding by adopting different preprocessing and vectorization modes. According to the method, a strategy of fusing multiple models is adopted, and the method has higher accuracy than a single model; based on a Fast Text model, a classification result can be obtained more quickly on the basis of the layered softmax; based on the TextCNN model, deep mining of text features in a fixed window in the code is carried out; code context semantics of the dynamic length are captured by the TextRNN. The method can help programmers to screen effective codes more quickly and help programmers understand the meaning of deep-layer services of the codes.

The invention discloses a method for determining a rigid change of a bridge structure based on deformation data of bridge health monitoring. The method comprises six steps: analyzing a bridge structure; installing a dynamic weighing system at a position where a vehicle gets on the bridge to monitor a vehicle load; getting bridge load data in a normal state; collecting deflection deformation data of the bridge within one month or longer time and establishing a standard model graph; after establishment of the standard model graph, continuously collecting deflection deformation and vehicle load data for a period of time to guarantee accuracy of the model; and realizing change determination of the bridge rigidity. According to the invention, on the basis of a constructed relation model of bridge member rigid changes and deformation normal distribution curve graphs, a rigid change of the bridge is determined according to changing rule of bridge deformation value. Compared with the traditional static test, the provided method has the following advantages: wasting of lots of manpower and materials can be prevented; bridge traffic blocking is not required; and utilization of the method plays a positive role in deep mining and application of bridge health monitoring data.