62 results about "Dynamical modeling" patented technology

A system for developing patient-specific therapies based on dynamic modeling of patient-specific physiology and method thereof are disclosed. The system includes software modules configured to provide access via a computer to one or more data collection protocols defining at least a type of patient-specific data to be collected and a manner in which the patient-specific data is to be collected, and to information from which one or more patient-specific models, configured to simulate one or more aspects of the patient's physiology, is developed. Another software module of the system is configured to provide access via the computer to one or more software tools that apply patient-specific data, collected according to the one or more data collection protocols, to the one or more patient specific models to determine therefrom one or more patient-specific therapies.

The invention relates to the technical field of network maps and space information service, in particular to a method for dynamically constructing an online thematic map. The method comprises the following steps of: constructing a sequenced mapping among three sets, namely a statistical index, a visual variable and a map sign by performing online organization and dynamic modeling on heterogeneous distributed statistic index data by a method for drawing a multi-variable map so as to integrate to form a drawing rule set in which a gathering visual variable is used as a core characteristic; and formalizing description language by using extensible markup language (XML) as a network map sign, dynamically constructing a personal thematic map by using a format of a network thematic map service combination, and forming the online thematic map in a logic layer model organization of a map group, a map picture and an illustration, which is detailed step by step. By the method, a map expression acquired by the user comprises dynamic customization expressed in the forms of a map sign, a color and the like, so the humanized requirement of the map user can be fully met; and aiming at users on different levels, the thematic maps meeting the service requirements can be designed, and the effect is remarkable.

The invention discloses a mobile-end 3D city dynamic modeling method based on geographic semantics. The method comprises the following steps: constructing classification description information and a semantic rule base of a city artificial building facility and a natural geographical entity based on a geometric semantic description method; constructing a geographic semantic library classification mapping analysis module and analyzing a semantic model stored and organized based on the geographic semantic library in a current window range; constructing a model component part 3D geometric outline construction module; and constructing a 3D model dynamic rendering and optimization display module to complete the dynamic construction and the rendering display of a 3D city. A virtual 3D city constructed on the basis of the method has flexible dynamic construction capability. The 3D scene can be constructed and generated dynamically according to a window range and an angle position in order to minimize the utilization rate of hardware resources to the utmost extent and to improve drawing efficiency for satisfying the display and render requirements of a mobile platform.

The invention discloses a dynamic modeling method of a conventional substation based on the IEC standard, which includes the steps as follows: firstly, building an object-oriented model; secondly, obtaining an entity model through mapping and supplementation; thirdly, building a UCIM model of a master station system as per the entity model; and fourthly, updating the UCIM model. The method effectively solves the problems of communication collision and efficiency between communication of an information point oriented dispatching master station and communication of an object-oriented substation, improves the efficiency of technology upgrade and innovation of a great amount of conventional substations, and reduces working pressure of substations and management maintenance staff for dispatching automatization.

The invention relates to a virtual error compensation system of a numerical control machine in the technical field of numerical control machining, comprising an external data acquisition module, a digital modeling module, a data optimization module, a dynamic modeling module, a model generalization module, an error fitting module, an error compensation module and a communication module. In the invention, a multi-module nesting mode is adopted; the external information acquisition module is used for acquiring environment information, numerical control machine information, cutting amount, cutter parameters, and the like; an error model is generated by the acquired information through the dynamic modeling module and the model generalization module; the error fitting module outputs an error fitting curve; the error compensation module outputs NC (Numerical Control) codes; and the communication module realizes the real-time communication with a CNC (Computer Numerical Controller) and data acquisition through a COM (Component Object Model) serial port and an RS232 serial port. The system of the invention can be used for predicting comprehensive errors of the numerical control machine under various cutting conditions and realizing real-time error compensation.

The invention relates to a hybrid depth learning model LSTM-ResNet based metropolitan space-time flow prediction technology. The invention can accurately predicting the change of urban spatio-temporaldata stream so as to provide important reference for urban management, and the key is to extract spatio-temporal dependency features from the data effectively. Currently, convolution neural network,which has been applied to spatio-temporal flow prediction, focuses on the extraction of spatial correlation features, ignoring the temporal dimension dependency and spatio-temporal correlation features. In depth learning model, long and short memory network (LSTM) is suitable for dynamic modeling of time series, and residual convolution network (ResNet) is suitable for large-scale spatial correlation feature extraction. Therefore, we combine LSTM and ResNet to construct a hybrid depth-learning model for spatio-temporal flow prediction: LSTM is used to consider the time dependency before and after, and filter out the invalid time features; the output of LSTM is inputted into ResNet and the spatio-temporal correlation feature is extracted. The model can automatically and accurately capture spatio-temporal correlation features, especially retaining valid temporal features when considering forward and backward dependencies.

The invention relates to an industrial process fault detection method based on an autoregression dynamic hidden variable model. According to the method, by use of data at a normal operation state in an industrial process, a dynamical process model with universality is established, and the model predicts a model structure through a Kalman filter and a smoother and an expectation maximization algorithm, such that dynamical and static coupling relation of industrial data can be effectively extracted; and then, based on dynamic and static noise of the dynamic model, a corresponding monitoring statistical amount is constructed, and a final fault detection result is obtained. Compared with other existing methods, the method provided by the invention can greatly improve the effects of industrial process dynamic modeling and fault detection, reduces the false alarm rate and the missed alarm rate of faults, improves the monitoring performance to a quite large degree, enhances the understanding capability and the operation confidence of process operators for the process and better facilitates automation implementation of the industrial process.

The invention discloses an ANSYS and ADAMS-based large-scale antenna dynamical modeling method, and mainly solves the problem that the existing modeling method cannot be used for accurately describing rigid motion of an antenna and cannot consider the problem of rigid-elastic coupling in a movement process of the antenna. The modeling method comprises the steps: based on a finite element model of an elastic component of a large reflector antenna, according to a modal analysis result, selecting the modality with the maximal contribution to antenna output, replacing node coordinates with a small quantity of modal coordinates to condense the degree of freedom of elastic deformation, and then exporting a corresponding modal neutral file; by establishing the rigid component of the antenna in ADAMS, combining the modal neutral file describing an elastomer, and performing assembly to obtain the dynamical simulation model describing the integral movement of the antenna. The ANSYS and ADAMS-based large-scale antenna dynamical modeling method overcomes the defects of the traditional modeling method, and a development period can be obviously shortened.

The invention discloses a compound identification method of an unmanned helicopter flight dynamical model and belongs to the field of unmanned helicopter dynamical modeling, and the method is characterized in that an unmanned helicopter, a flight control computer, a sensor group, an airborne data broadcasting station, a ground data broadcasting station, a ground station, a remote controlled transmitter and a remote controlled receiver are utilized, wherein a remote controlled command of a ground pilot is in charge of stimulating the unmanned helicopter; an automatic control command of the flight control computer is used for maintaining the unmanned helicopter at the predetermined flight speed, and ensuring the flight safety; and the remote controlled command and the automatic control command are combined by the flight control computer to obtain a steering engine command, so as to manipulate the unmanned helicopter to finish an identification test. In the compound identification methodprovided by the invention, the remote controlled command of the ground pilot and the automatic control command of the flight control computer are simultaneously introduced, so as to ensure the remotecontrolled command and the automatic control command to be mutually matched, thus the flight dynamical model of the unmanned helicopter can be accurately and safely identified.

The invention provides a dynamical modeling least square support vector machine soc estimation method belonging to the SOC estimation method technology field. The dynamical modeling least square support vector machine SOC estimation method comprises the steps that in a discharge initial stage, an LS-SVM is constantly matched with a sample most similar to a current discharge cycle, and is used to constantly build a new model to provide an SOC value corresponding to a current discharge moment, and after discharge is stable, no more new model is built; the current model is used to complete the SOC estimation of the subsequent discharge moment. During SOC estimation of a whole life cycle of a battery, the method provided by the invention is used for different batteries. The charging and the discharging of the battery is a complicated electrochemical reaction, and therefore the SOC is unable to be acquired by direct measurement. During the SOC estimation, the external measurable parameters of the battery are acquired, such as voltage, current, resistance, and temperature. The external parameters are in non-linear correlation with the SOC, and therefore the parameters are mapped to the SOC in a non-linear manner, and an object of measuring estimated SOC from outside is achieved.

The invention discloses a community question and answer website label recommendation method based on a user background, comprehensively considering the to-be-recommended problem text information, theuser background information and the relevance of the text information and the user background information, and modeling a label recommendation problem into a multi-classification prediction problem based on deep learning. The core of the community question and answer website label recommendation method is a deep neural network model PcTagger, and through dynamic modeling of the user background information, the defect that user background static modeling in an existing personalized label recommendation method is difficult to match different recommendation tasks is solved. The model mainly comprises: 1) text feature modeling based on a recurrent neural network and an attention mechanism; 2) user background influence dynamic modeling based on historical question records, and 3) label recommendation by fusing the text features and the user background influence. The experimental results displayed on a real data set show that the community question and answer website label recommendation method based on a user background can significantly improve the prediction precision, compared with an existing same kind label recommendation method.

The invention discloses a dynamic process monitoring method based on structured canonical variable analysis. The dynamic process monitoring method aims to infer a structured canonical variable analysis algorithm and implement dynamic process monitoring based on the algorithm. Specifically, according to the dynamic process monitoring method, by implementing improvement on the optimization target ofthe canonical variable analysis algorithm, the structural thinking is considered, and thus the new algorithm is inferred to excavate autocorrelation characteristics; in the process of extracting thepotential characteristics, the correlation between future data and past data is considered simultaneously, that is, the autocorrelation characteristics on the time series are considered; in addition,during monitoring, the autocorrelation characteristics are monitored by constructing a past vector component and a future vector component; and it can be said that the method infers the brand new dynamic modeling algorithm, namely the structured canonical variable analysis algorithm, and implementation of dynamic process monitoring on this basis should have more excellent fault monitoring performance.

The invention discloses a risk stratification method for myocardial ischemia based on deterministic learning and deep learning. The method includes the steps that conventional 12-lead electrocardiogram signals are collected, based on the deterministic learning theory, neural network modeling and identification are conducted on intrinsic electrocardiodynamic characteristics of the shallow electrocardiogram signals, and the intrinsic dynamic characteristics of ECG signals are obtained; the convolutional neural network under the framework of deep learning is used for achieving the risk stratification of myocardial ischemia. The method combines the deterministic learning dynamic modeling method and the deep learning classification method for the first time, the method is applied to early riskstratification of myocardial ischemia based on the conventional 12-lead electrocardiogram signals, no additional detection equipment is needed, and the method is easy and convenient to use and easy tooperate. Through the deterministic learning method, the dynamic characteristics more sensitive to the ischemic state are extracted, the deep neural network can learn data features independently without further data characterization, and the complexity of the system is reduced.

The disclosed embodiments provide a system that facilitates making updates to a data model, wherein the data model is associated with data items stored in a relational database system. In particular, updates to the data model are automatically propagated to the relational database system so that the structure of the relational database system remains consistent with the data model. This is accomplished by automatically executing queries corresponding to the updates against the relational database system. For example, changes to fields and / or tables in the data model can be automatically propagated to a relational database system by executing “Create,”“Drop,”“Alter,” and / or “Add” statements against corresponding fields and / or tables in the relational database system.

The invention discloses a method for controlling a business system by dynamic modeling of a state machine. The method includes the steps: defining and generating a state machine model file of a business object in an XML (extensive makeup language) mode by a graphing tool; dynamically loading the state machine model file of the corresponding business object by a state machine model driving engine; and controlling the state switching process and dynamic behaviors of the business object according to a state machine model of the business object according to the state machine model driving engine. In the method, the behaviors of the business object are controlled by dynamic modeling of the state machine, a runnable state switching model and a dynamic behavior model of the business object are built by graphing configuration, and only new state node or operation rules need to be defined by the graphing definition tool of the state machine when the dynamic behaviors of the business object are changed, so that the workload of code setting due to change in requirements of the dynamic behaviors of the business object is greatly decreased, and development efficiency is improved.

The invention provides a local structural dynamical modeling method and a device. The method is adopted for solving the problems of big data volume, low calculation efficiency and difficulty in locally dynamical responding and finely analyzing as required of the present dynamical modeling. The method comprises the following steps: dividing a whole structure into a first structure and a second structure; establishing an interface multipoint restriction equation for the first structure under the first grid and the second structure under the second grid according to the finite element stiffness matrix and mass matrix of the first structure and the finite element stiffness matrix and mass matrix of the second structure; confirming a vibration mode of the first structure and a vibration mode of the second structure according to the equation; and confirming a deformation compatibility matrix between the DOF (Degree of Freedom) modal displacement of the first structure and the DOF modal displacement of the second structure according to a Moore-Penrose pseudo-inverse method, thereby confirming a dynamical equivalent support effect matrix of the second structure and a fine dynamical finite element model of the first structure.

The invention discloses a dynamic modeling method of large-scale track lines. In order to perform dynamically three-dimensional track modeling according to different track line data, primary loading is performed to a track model basic unit section established through professional modeling software in the initialization phase, parameters of positions, directions and the like of all tracks on lines are decomposed and computed, and then real-time splicing and displaying are directly performed to tracks and scenes according to a current viewpoint position in the simulation phase. The method is simple and effective, can be used for performing real-time simulation of large-scale track lines on a common personal computer, and is applied to engineering applications, such as design and planning of large-scale track traffic engineering projects, driver training based on vehicle operating scene simulation and safety analysis.

The invention discloses a parallel hybrid electric vehicle energy management method with gearbox gear control, and relates to the field of hybrid electric vehicle energy management. The parallel hybrid electric vehicle energy management method analyzes an evolution law of a state of charge change rate of a battery and engine specific fuel consumption under each gear, combines with a hybrid logic dynamic modeling method, establishes a parallel hybrid electric vehicle energy management system model including gearbox discrete gear information, further introduces a model predictive control idea, defines motor torque as a continuous control variable, and defines gearbox gears as discrete control variables and translates an optimization problem into a mixed integer quadratic programming problemto solve, the strategy optimizes the torque distribution of each power source and decides an optimal gearbox gear. The parallel hybrid electric vehicle energy management method aims at the energy management problem of a parallel hybrid electric vehicle with a gearbox, and has the advantages of high pertinence, high practicability, good optimization effect, high integration and the like.

The invention discloses a wind power ultra-short term probability prediction method and system based on empirical dynamic modeling, and the method comprises the steps of carrying out the standard normalization processing on a time sequence of a to-be-predicted quantity, and carrying out the nonlinear aggregation degree calculation of data after the standard normalization processing, so as to inspect the nonlinear degree of a given dynamic system; calculating an optimal embedding dimension E and delay time tau by adopting a particle swarm optimization algorithm; further, performing phase spacereconstruction on the time sequence of the to-be-predicted quantity; and constructing an empirical dynamic model, and predicting the given dynamic system in the reconstruction phase space by adoptinga simplex projection method to obtain a prediction result of the to-be-predicted quantity. The prediction result shows that the wind power ultra-short term probability prediction method based on empirical dynamic modeling can achieve objective description of the wind power generation dynamic process completely according to data, and the effectiveness of probability prediction is remarkably improved.

The invention discloses a composite fault diagnosis method for a hydraulic AGC system. The method specifically comprises the following steps: S1, mill component dynamic modeling: acquiring the equivalent of each link of a hydraulic AGC system to get a mathematical model thereof; S2, system modeling: simulating a fault, selecting a measuring point, building a multi-signal model of the system, and getting a fault dependence matrix under the multi-signal model; and if the fault is a single fault, getting a fault dependence matrix directly, performing fault diagnosis, and getting a result directly; S3, if the fault is a composite fault, using a blind source separation algorithm to blindly separate measuring point information, getting a fault dependence matrix, and diagnosing the composite fault; and S4, fault diagnosis: after the system fails, getting the information of the measuring point, and locating the fault source by comparing the measuring point information with the fault dependence matrix. Through the diagnosis method of the invention, the purpose of composite fault isolation and location is achieved, and the fault separation rate is improved.

The invention provides a method for designing key geometric parameters of an end mill on the basis of machining vibration. The method includes the steps that modal testing is carried out on a machining system to obtain key modal parameters; dynamical modeling is carried out on the machining system, and a multi-time-lag second-order differential dynamical equation is established; a converted state-space equation is established; stability of the machining system is judged through a generalized Runge-Kutta (GRK) method, and Lobe graphs, namely stability graphs, in a machining parameter space are obtained; different Lobe graphs are obtained by changing design parameters namely values of tooth pitches and helical angles of the end mill; optimized tooth pitches and helical angles of the end mill are obtained by comparing the Lobe graphs under the condition of different design parameters with a goal of obtaining the maximum machining efficiency. Compared with traditional equal-tooth-pitch standard milling cutter machining, the method has the advantages that dynamic characteristics of the machining system are obtained through the GRK method, the optimized key geometric design parameters, namely the tooth pitches and helical angles, of the end mill are obtained, and machining efficiency is greatly improved.

The invention discloses a dynamic modeling method of a boiler combustion system based on an online support vector machine. The method comprises the following steps: firstly, taking the NOx emission and a boiler efficiency value as model output, and then taking the main influence factors of the furnace total coal-supply quantity, the total air, the auxiliary air, the flue gas oxygen content and soon which influence the boiler emission and efficiency as model input; secondly, in the dynamic modeling process, considering the order of input and output variables in order to reflect the dynamic change characteristics of an object; and finally, establishing a dynamic model of the boiler combustion system through an improved online adaptive least square support vector machine algorithm (FVS-ALS). Compared with a traditional steady-state model, the model of the invention has higher prediction accuracy and has an on-line correction function, and can adapt to the change of control characteristics of the combustion system caused by changes in load, coal quality and equipment characteristics, and is of great value to the timely and accurate monitoring of the running state of the boiler combustion system and the operation optimization accordingly.

The hybrid dynamic modeling and optimization control method for the intelligent automobile longitudinal dynamics system comprises the steps of establishing a whole automobile longitudinal dynamics model, and conducting segmented affine approximation on a nonlinear part existing in the model; based on hybrid dynamic model of the longitudinal dynamic system of the whole vehicle and a prediction control method, converting an intelligent vehicle longitudinal motion control problem into a mixed integer quadratic programming problem for solving to achieve the integration of a system optimization control law, and determining the optimal values of the driving torque and the braking torque and the optimal switching sequence between the driving mode and the braking mode in the intelligent vehicle longitudinal motion control process. The method solves the problems that an existing automobile longitudinal motion control method cannot accurately describe hybrid dynamic characteristics existing in asystem and a driving mode and a braking mode are smoothly switched in the automobile speed adjusting process.

The invention discloses a brain-computer adaptive system for brain-computer interface system, The system comprises: Cognitive Status Monitoring Module, a mutual adaptive adjustment module, a cognitivestate control module, and a BCI system control module, wherein the mutual adaptive adjustment module processes the user cognitive state transmitted by the cognitive state monitoring module in real time, and performs real-time closed-loop control on the BCI system control module and the cognitive state control module according to the user cognitive state; The mutual adaptive regulation module comprises three regulation modes of dynamic modeling, dynamic parameter regulation and dynamic audio-visual feedback. The three regulation modes together constitute two closed-loop control paths, including a 'machine' adaptive brain closed-loop regulation path and a 'machine' adaptive brain closed-loop regulation path. The invention dynamically adjusts the stimulation parameters and decoding algorithmof the BCI system according to the user cognitive state closed loop to adapt to the user state, optimizes feedback and regulates the user cognitive state to adapt to the cognitive demand of the BCI.

The application of the method relates to the field of exploitation of the open pit mine and side slope engineering management and the like. With the integration of the research result on single model prediction, the research result on combined model prediction, and the research result on overall deformation prediction of a side slope monitoring line, a deformation predication method for the side slope three-dimensional entity is provided; the deformation predication method skillfully combines a static modeling method with a dynamic method, thus solving the difficulty in the overall prediction of the slope three-dimensional entity; with the deformation prediction method for the slope three-dimensional entity, the deformation condition of the side slope in long time can be predicted; simultaneously, a gray system model and a time sequence ARMA (Autoregressive moving average model) are adopted in predication, a precise predication result is obtained; and the comprehensive research on the long-time deformation predication of the side slope three-dimensional entity refers to the comprehensive application of four methods such as the deformation predication, the long-time predication, the single model predictionand the rolling combination predication of the side slope three-dimensional entity, the predicting outcomes has very high reliability.

The invention discloses a brine pump remote energy saving diagnostic analysis system and method. The system comprises a brine well, a brine pump, an intelligent collector, a server, a client and a flowmeter; a current transformer is electrically connected to the brine pump; an industrial switch is arranged between the brine pump and the intelligent collector through a signal line; the current transformer is electrically connected with the intelligent collector through a signal line; the server comprises a signal processing module, a dynamic modeling analysis module, a work order generating module and a database; a brine pipe is arranged between the brine well and the brine pump; and the flowmeter is arranged on the brine pipe. According to the brine pump remote energy saving diagnostic analysis system and method, the system is simple in structure, and the method is simple and convenient to implement; the intelligent collector is adopted to measure brine data through the flowmeter so that the appropriate brine pump can be selected, the working current is transmitted through the current transformer, the situation that the server remotely controls starting and stopping of the brine pump in a closed loop mode is achieved, resources are utilized reasonably, energy consumption is reduced, and the operating life of the brine pump is prolonged.

The invention discloses an IEC61850-based distributed photovoltaic cluster dynamic modeling method. The method is characterized by comprising the following steps of: 1) establishing a distributed photovoltaic unit model on the basis of an IEC61850 standard modeling method; 2) carrying out cluster dynamic division on all the photovoltaic units in an area according to different regulation targets; and 3) aggregating an IEC61850-based photovoltaic unit layer information model according to the distributed photovoltaic cluster division result, and importing a cluster regulation ability and a topological structure logic node to establish a final photovoltaic cluster information model. The method is capable of achieving the beneficial effects of dynamically selecting distributed photovoltaic stations which participate in clusters according to different regulation functions, carrying out standard modeling on the clusters by combining characteristics of IEC61850, completing the efficient communication between the clusters, realizing the stable operation of photovoltaic power supplies and ensuring the safe control of distributed photovoltaic.