40results about How to "Effective exploration" patented technology

The invention discloses a humanoid robot gait control method based on model correlated reinforcement learning. The method comprises steps of 1) defining a reinforcement learning framework for a stable control task in forward and backward movements of a humanoid robot; 2) carrying out gait control of the humanoid robot with a model correlated reinforcement learning method based on the sparse online Gaussian process; and 3) improving a motion selection method of a reinforcement learning humanoid robot controller by a PID controller, and taking the improved operation as an optimizing initial point for the PID controller obtaining the motion selection operation of the reinforcement learning controller. The invention utilizes reinforcement learning to control gaits of the humanoid robot in movement, and thus the movement control of the humanoid robot can be automatically adjusted via interaction with the environment, a better control effect is achieved, and the humanoid robot is enabled to be stable in forward and backward directions.

The invention discloses a method and a system for social relationship recommendation. The method includes that musical behavior data of users are acquired by the social relationship recommendation system, values of preference of the users on each music tag are calculated according to the musical behavior data, inter-user musical similarity is computed according to the values of preference of the users on the music tags, and user recommendation is performed on a current use about to accept recommendation according to the inter-user musical similarity. Besides, the social relationship recommendation system can obtain inter-user intimacy according to user relationship link data and / or user interaction data statistics, and user recommendation is performed according to the inter-user musical similarity and intimacy. By the method and the system, the users can use accurate and effective user recommendation information for exploring social relationships in the music field, social relationship building efficiency in internet social music services is improved, and time cost for the users to build social relationships is reduced.

The invention discloses a method and system of extracting information of unexpected events based on literary forms and word lists. First, analyzing the characteristics of Web texts and news styles; extending a list of words built by experts with Google word 2 Vec; developing different extraction methods for different attributes of unexpected events: using the word lists for event classification; using stylistic features to extract time and event abstracts; using literary style and word lists to extract locations, casualties and economic loss attributes. Experiments show that the average accuracy rate is 87.89% and 91.29%, respectively, and the average recall rate is 81.76% and 87.91% respectively when using the method based on literary style and word lists to extract the attributes of unexpected events in crawled Web news corpora and open corpora, which can meet emergency management needs.

The invention relates to the field of seabed natural gas exploration and provides a three-dimensional time-shifting seismic observation method and system. The three-dimensional time-shifting seismic observation method and system are used for solving the exploration problem of seabed natural gas. The three-dimensional time-shifting seismic observation system comprises a data collecting device, a data transmission device and a data analysis device, the data collecting device is arranged at the seabed, the data transmission device is connected with the data collecting device, and the data analysis device is connected with the data transmission device. The collecting device is arranged at the seabed, thus the seismic data can be effectively collected, and sufficient support is provided for reservoir analysis.

A method is provided for transforming data from a high-dimensional to low-dimensional design space, and for inspecting the transformed data in such a manner as to permit effective navigation and exploration of the high-dimensional design space. Conveniently, an optimum / conditional value for a prescribed functional representation of the transformed data can be derived by visual inspection of a 2-D image map representation of the transformed data. Significantly, the invention has utility for various aircraft design applications and although this technology has been developed with reference to aircraft aerodynamic design in particular, the design space visualisation and curve-fitting technology developed is general. It should therefore be equally applicable to other disciplines such as cost analysis, structures and computational electromagnetics, in which expensive analysis tools are used to find optima for complicated design problems. It is expected to be particularly useful in multi-disciplinary design and situations where there are multiple optima in the design space.

The invention discloses a three-dimensional electromagnetic imaging device applied to road disaster and collapse detection and a use method thereof. The device is formed by a transmitting part, a receiving part, an acquisition device and a processing system. The GPS timing synchronization is carried out between the transmitting part and the receiving part. The processing system is installed in electronic equipment. The acquisition device is a rectangular body and is provided with a center plate. The center plate is provided with sensors and is provided with four extending brackets outward. The ends of the extending brackets are provided with auxiliary brackets which are wound by multiple turns of transmitting coils. The electronic equipment can be used in a field to carry out real-time imaging, simple real-time processing and feedback of construction quality information, the ability of penetrating and distinguishing low resistance coverage is low, a high resistance shielding phenomenon does not exist, a hole or soft stratum in a road can be effectively explored, the interference of atmospherics and an artificial electromagnetic field from the outside is reduced, the electromagnetic coupling effect between the transmitting coils and receiving coils can be effectively eliminated, the equipment is small, the weight is small, and the device is suitable for being used in cities and high speed roads.

The invention relates to a method for detecting the seepage performance of rock with filled connected fracture. The method includes steps of 1, taking samples from construction sites, preparing a plurality of standard test specimens, designing seepage orthogonal test schemes for the filled fracture and preparing study test specimens; 2, implementing the orthogonal test schemes, processing orthogonal test data by the aid of seepage coefficients which are used as detection indexes and finding factors with significant influence on seepage of the filled fracture; 3, designing multi-factor and multi-level comprehensive tests for the significant factors and measuring seepage coefficients in each test; 4, acquiring relations between each single factor and the seepage coefficients and relational functions of multi-factor coupling effects and the seepage coefficients by means of regression according to comprehensive test data. The method has the advantages that the method is scientific and rigorous, the various factors with influence on the seepage performance of the rock with the filled connected fracture can be comprehensively studied, and the method has a guiding significance on managing seepage during underground engineering and designing and implementing supports in seepage environments.

The invention discloses a multifunctional portable micro unmanned aerial vehicle system which comprises a vehicle body and a flight control mainboard. The vehicle body and the flight control mainboardare integrated on a PCB. The multifunctional portable micro unmanned aerial vehicle system has the advantages of being simple in structure, low in mass, small in size, low in cost, good in invisibility, convenient to use, wide in application range, convenient to carry and unfold, capable of flying under complex terrain indoors or outdoors, capable of executing detective tasks invisibly and conveniently, and meanwhile not high in requirement for the using environment.

The invention discloses a variant aircraft aerodynamic optimization method for improving the expected improvement degree of a position vector, and belongs to the technical field of aircraft overall optimization. The method comprises the following steps: selecting an initial reference airfoil profile and a related shape coefficient, and determining a design working condition; establishing an optimization model considering structural consistency constraint to solve the shape coefficient of the deformed airfoil profile, and establishing a high-precision aerodynamic analysis model to solve the lift coefficient and the resistance coefficient of the airfoil profile; establishing a Kriging agent model of the high-precision pneumatic analysis model, optimizing the Kriging agent model to obtain a pseudo-non-dominated solution, and judging whether a prediction variance of the Kriging agent model is greater than a threshold value or not; and if the prediction variance is not less than the threshold, obtaining newly added sample points for updating the Kriging proxy model by adopting a position vector expectation improvement criterion, otherwise, screening the newly added sample points from the pseudo-non-dominated solution to update the Kriging proxy model. By screening newly added sample points, the precision of the Kriging agent model at the Pareto frontier is improved, the optimality and the distributivity of an optimized non-dominated solution are improved, meanwhile, the optimization efficiency is improved, and the cost is saved.

The invention provides a stratospheric large-scale MIMO user grouping and precoding method and system, and the method comprises the steps: an optimization problem design step: designing an optimization problem according to an outer-layer precoding matrix and an inner-layer precoding matrix on the basis that it is assumed that multiple users of a stratospheric large-scale MIMO communication systemare grouped; Optimization problem simplification step: carrying out optimization problem simplification; According to the method, a matrix norm inequality and a zero space criterion are adopted, an obtained optimization problem is simplified, the power of a signal is explored to be mainly concentrated on a statistical intrinsic mode StacticalEgigades of a channel, and the statistical intrinsic mode StacalEgigades are obtained. In the invention, the CSI of the effective channel for designing the inner-layer precoding matrix is also effectively reduced, so that the grouping scheme and the precoding scheme can effectively explore the key technology of the stratospheric large-scale MIMO communication system.

A method (20) for generating / updating a database of current-voltage characteristic curves is disclosed. This method comprises simulating (25), for at least one combination of a topology of a photovoltaic cell group, an internal cell temperature(s) and a cell irradiation(s), a model of the photovoltaic cell group to provide a current-voltage characteristic curve representative of that combination,and clustering (24) the current-voltage characteristic curves to identify at least one plurality of similar current-voltage characteristic curves. The method also comprises generating (26) a many-to-one mapping in the database to map query requests that correspond to each of the at least one plurality of similar current-voltage characteristic curves onto a single representative current-voltage characteristic curve for that plurality of similar current-voltage characteristic curves, each query request identifying a topology of a photovoltaic cell group, at least one internal temperature for the photovoltaic cells in the photovoltaic cell group and at least one cell irradiation for the photovoltaic cells in the photovoltaic cell group.

The invention relates to a comprehensive evaluation method for selection of favorable zones of medium-order coal bed gas reservoirs in small and medium-sized subsidence basins. The comprehensive evaluation method comprises the following steps: 1, evaluating the gas production capacity of a target coal seam; 2, evaluating the reservoir capacity of the target coal seam; 3, evaluating the storage condition of the target coal seam; 4, evaluating the permeability of the target coal seam; 5, defining coal bed methane accumulation main control factors of the target zone, predicting distribution characteristics of the main control factors by utilizing logging, earthquake and experiment analysis data of the research zone, superposing favorable zones of the main control factors, and comprehensivelyanalyzing and predicting favorable high-permeability enrichment zones. The method aims at the actual situation that the exploration degrees of a plurality of coal-bearing basins are different; according to the method, the parameters are reasonably classified, through comprehensive evaluation of the evaluation parameters, the coal bed methane reservoir area can be accurately and effectively judgedto the maximum extent, and the problems that according to an existing evaluation method, due to limitation of the evaluation parameters, evaluation is inaccurate, the selection target of coal bed methane is not clear, and manpower and material resources are wasted are solved.

A transmit power control (TPC) method is provided. The method includes: obtaining a first channel quality of a pilot channel for a user equipment (UE) (S310); performing an interference cancellation on the pilot channel to obtain a second channel quality of the channel (S320); and instructing the UE to adjust the transmit power based on at least the first channel quality (S330).

The present invention discloses a determination method of permeability rule of fracture surrounding rock bolting and grouting in underground engineering. The method is based on a large number of experiments and comprehensive and reliable data. A rock crack density as a variable, and significant factors affecting the permeability rule of filling cut-through crack surrounding rock bolting and grouting as a set of variables, the degree of rock crack is controlled unchanged, and a multi-factor and multi-level comprehensive experiments are conducted on the significant factors; and based on the minimum permeability coefficient principle, the optimal levels of the significant factors corresponding to different crack densities of the rock are determined. After comprehensive tests, the method obtains a relationship of multiple factors to surrounding rock crack density through regression analysis; the principle of optimal experimental results, experience form of the surrounding rock crack density corresponding to each significant factor is organized, so as to guide seepage governance in underground engineering and supporting design and implementation of in seepage environment.

The invention discloses a bionic-based flying mechanical neck eye system and a control method thereof. The system involves an aircraft system, a mechanical neck eye system, an airborne computing device and a flight control unit; the mechanical neck eye system involves a mechanical neck arranged on an aircraft body, a binocular vision inertia system arranged at the tail end of the mechanical neck,and a mechanical neck main control for controlling the movement of the mechanical neck; and the binocular vision inertial system involves a binocular camera and an inertial measurement unit, the binocular vision inertia system is used for carrying out active observation and completing pose estimation on the tail end of the mechanical neck, and the mechanical neck is a six-degree-of-freedom seriesmechanical neck. According to the bionic-based flying mechanical neck eye system and the control method thereof, the binocular vision inertial system consisting of the double fisheye cameras and the inertial measurement unit is placed at the tail end of the mechanical neck to form a sensing system, so that the observation range of the system is larger; and the cameras is variable in pose, and hasthe capability of actively sensing the environment, so that the problem that the view field of an aircraft in the indoor environment is limited can be solved to a certain extent, and meanwhile, the shaking of pictures can be compensated.

The invention relates to equipment capable of detecting a uranium ore in real time in a coring drilling process. A while-drilling gamma uranium measuring instrument based on rock core drilling is characterized by comprising an outer pipe, an inner pipe, a battery component, a fastening pressure spring and a sensor, wherein the outer pipe is a non-magnetic drill rod; the inner pipe is positioned in the outer pipe, and comprises a power supply switch fixing piece, a battery bin compression-resisting cylinder, a first compression-resisting cylinder connecting head, a circuit board bin compression-resisting cylinder, a second compression-resisting cylinder connecting head, a sensor bin compression-resisting cylinder and a plug screw; the battery component supplies power to a signal processing circuit and the sensor; the sensor is positioned in the sensor bin compression-resisting cylinder, and is connected with a sensor signal input end of the signal processing circuit; the signal processing circuit comprises a single-chip microcomputer; and the single-chip microcomputer is connected with a storer through an interface circuit. The instrument can perform the uranium ore exploration for a stratum difficult to obtain a rock core, through measuring natural gamma strength of the stratum, and measures the natural gamma strength while drilling in the coring drilling process to judge if the drilled stratum is the uranium ore.

Provided is a genetic cluster method based on multiple partial searches. Evolution in the genetic clustering process is supported through the method with the three partial searches including cluster combining, cluster splitting and k average value operation, and the searches have different characteristics according to search tasks. The first two partial searches are used for partially increasing the number of clusters in a solution, and how the first two partial searches are used is decided in the clustering process in a self-adaptive mode. The third partial search is used for improving the clustering center of the solution. The three partial searches are used for mutually cooperating and competing from the aspect of complementation to partially improve the solution in genetic evolution so that the common optimization target can be completed. The decision space can be rapidly and effectively found, and therefore the proximity global optimal clustering result is given. The genetic cluster method based on the multiple partial searches is applied to image cutting, and the precise cutting effect is achieved.

The invention discloses a convolutional neural network model for static image behavior recognition, and relates to the technical field of image processing methods. The model comprises a VGG16 convolution layer module used for processing an input image; a multi-branch convolution layer module used for respectively learning features output by the last layer of the VGG16 convolution layer module through different branches, wherein the convolution layer weight of the module is randomly initial in the training process; and a softmax classifier layer module used for classifying the features outputby the three-branch convolution layer module. The model can explore information in a VGG16 convolutional layer module channel more effectively, and can play a role in finely adjusting the highest layer weight in the neural network; therefore, the action information in the single static human behavior image can be identified more effectively.

The invention discloses a PID parameter setting optimization method based on an improved Bayesian model. The method comprises the following steps: constructing an optimization model of a PID control system, and selecting an objective function which is an error absolute value time integral performance index; establishing an agent model of a PID parameter space and an objective function value spaceto realize efficient prediction of an unknown parameter objective function value; establishing a selection function of PID parameters, and realizing automatic iterative exploration of PID optimal parameters; performing an optimal parameter space reduction strategy; and determining a termination condition of iterative search, and outputting an optimal parameter and a performance index thereof. According to the method, on the basis of an improved Bayesian model, an agent model from a PID parameter space to an objective function value space is constructed, and efficient prediction of an unknown parameter objective function value is realized. By establishing a selection function for PID parameters, effective exploration of a potential optimal solution in an unknown parameter space is realized,so that the optimization efficiency and effect of the algorithm are improved.

The invention discloses a method for analyzing a transformer winding and a sleeve direct-current resistance change amount threshold value by utilizing mathematical statistics. In order to accurately analyze the change quantity threshold value of the direct-current resistance of the transformer, the direct-current resistance data of the transformer is firstly sorted, and the variation of the direct-current resistance of the adjacent years is calculated; then, by combining a subjective representation method (distribution diagram and a Q-Q graph) with an objective hypothesis testing method (W checking, K-S checking and A-D checking), whether the direct-current resistance change quantity meets normal distribution or not is verified; and finally, if the direct-current resistance change quantitymeets normal distribution, the sample data mean value and the standard deviation are calculated, a normal distribution confidence interval is utilized and a certain margin condition is considered, and a direct current resistance variation threshold value under different voltage grades is determined. By analyzing the mathematical statistics result, the threshold value of the direct-current resistance change of the transformer under the difference condition can be judged, and reference is provided for operation and maintenance personnel.

An unmanned aerial vehicle group flight path optimization method for multi-radiation source tracking comprises an establishment module, an estimation module, a matching module, a positioning module and a tracking module. The establishment module is used for establishing an unmanned aerial vehicle group trajectory optimization problem under a multi-constraint condition; the estimation module obtains a mapping relation between the received signal strength and the distance by adopting a deep neural network; the matching module adopts an interactive matrix generation method to obtain an unmanned aerial vehicle and radiation source matching scheme; the positioning module obtains a reference position of the radiation source by adopting a multi-ball intersection positioning method; and the tracking module adopts a deep reinforcement learning method to design a flight path optimization algorithm of the unmanned aerial vehicle group. Compared with a traditional method, the method has obvious advantages in the index aspects of average tracking time, task completion rate, convergence speed and the like.

The invention discloses a multi-modal sentiment classification method based on an attention guidance two-way capsule network. Due to the fact that the capsule network has the trainable viewpoint invariant transformation characteristic, the effectiveness of the capsule network is proved in the aspect of analyzing heterogeneity of multi-modal learning at present. In the preprocessing stage, a multi-modal dynamic interaction enhancement module is provided, cross-modal homogeneity is explicitly enhanced on the feature level, and the model can effectively execute the multi-modal decoupling process in a more compact local public space. On the basis, a bidirectional capsule network (ABCN) based on attention guidance is provided, and a global multi-mode public message is explored through a new bidirectional dynamic routing mechanism. Secondly, guiding a multi-modal dynamic routing process by utilizing a global multi-modal context, and researching a global optimal common clue of each modal; this greatly improves learning efficiency and provides superior ability to erect a bridge between all modes.

The invention provides a stratospheric large-scale MIMO user grouping and precoding method and system, and the method comprises the steps: an optimization problem design step: designing an optimization problem according to an outer-layer precoding matrix and an inner-layer precoding matrix on the basis that it is assumed that multiple users of a stratospheric large-scale MIMO communication systemare grouped; Optimization problem simplification step: carrying out optimization problem simplification; According to the method, a matrix norm inequality and a zero space criterion are adopted, an obtained optimization problem is simplified, the power of a signal is explored to be mainly concentrated on a statistical intrinsic mode StacticalEgigades of a channel, and the statistical intrinsic mode StacalEgigades are obtained. In the invention, the CSI of the effective channel for designing the inner-layer precoding matrix is also effectively reduced, so that the grouping scheme and the precoding scheme can effectively explore the key technology of the stratospheric large-scale MIMO communication system.

The aerodynamic optimization method of a variant aircraft for improving the expected improvement degree of the position vector disclosed by the invention belongs to the technical field of overall optimization of the aircraft. Select the initial reference airfoil and related shape coefficients to determine the design conditions; establish an optimization model considering structural consistency constraints to solve the shape coefficient of the deformed airfoil, and establish a high-precision aerodynamic analysis model to solve the lift coefficient and drag coefficient of the airfoil; establish The Kriging surrogate model of the high-precision aerodynamic analysis model, optimize the Kriging surrogate model to obtain a pseudo-non-dominated solution, and determine whether the prediction variance of the Kriging surrogate model is greater than the threshold; if the prediction variance is not less than the threshold, use the position vector expectation improvement criterion to obtain new samples The point is used to update the Kriging surrogate model, otherwise the new sample points are filtered from the pseudo-non-dominated solution to update the Kriging surrogate model. By filtering new sample points, the accuracy of the Kriging surrogate model at the Pareto frontier is improved, the optimality and distribution of the optimized non-dominated solution are improved, and the optimization efficiency is improved and the cost is saved.

The present invention addresses the problem of providing a method capable of detecting unknown virus sequences and that can efficiently perform virus detection and searches. The method includes a stepfor randomly fragmenting target double-stranded (ds) RNA to obtain dsRNA fragments; a step for providing the obtained dsRNA fragments to a reverse transcription reaction, followed by a polymerase chain reaction (PCR), to obtain corresponding DNA fragments; and a step for providing the obtained DNA fragments to a sequence analysis operation to determine sequences. The reverse transcription reaction is preferably initiated from the 3' terminal of the dsRNA fragments.