66results about How to "Solve the problem of ambiguity" patented technology

The invention discloses an urban water disaster risk prediction method based on an RBF (radial basis function) neural network-cloud model. The method includes (1) determining evaluation factors, levels and the indicator range under corresponding levels; (2) determining an expectation Ex and an entropy En of the cloud model; (3) determining the weight of each evaluation factor according to measured values of the evaluation factors and the indicator range of each level; (4) training the RBF neural network, finishing model establishment for the RBF neural network, inputting the measured values of the evaluation factors of the cloud model to the trained RBF neural network to perform simulated prediction, and obtaining a prediction value of each evaluation factor; and (5) substituting the prediction value of each evaluation factor to the integrated cloud model to allow the integrated cloud model to calculate corresponding certainty degree of the prediction value of each evaluation factor belonging to each risk level and multiply the corresponding weight to obtain integrated risk level distribution. The urban water disaster risk prediction method is visualized and reliable and strong in operability, and accuracy of prediction is improved.

The invention provides a large-scale complex equipment maintenance guarantee capability assessment method comprising the specific steps: firstly building a large-scale complex equipment maintenance guarantee capability assessment index system; then, building a large-scale complex equipment maintenance guarantee capability assessment model; finally, gathering large-scale complex equipment maintenance system feature parameters, importing the parameters into the assessment index system and the assessment model, thus obtaining a large-scale complex equipment maintenance guarantee capability assessment conclusion. The maintenance guarantee capability assessment index system comprises three-level index parameters and index weight calculation: the first level index of the three-level index parameters refers to the manpower resource guarantee capability, the maintenance facility guarantee capability, the maintenance equipment guarantee capability, the maintenance apparatus guarantee capability, the information resource guarantee capability and the maintenance guarantee management level; the index weight calculation method is determined by a group decision and an AHP method.

The invention discloses a parallelization method of convolutional neural networks in a fuzzy region under a big-data environment. The parallelization method comprises the following steps: firstly, constructing the convolutional neural networks in the fuzzy region, putting a given target assumption region and object identification into the same network, carrying out convolutional calculation, and updating the weight of the whole network in a training process; and secondly, dividing an input log data set into a plurality of small data sets, introducing multiple workflows to pass through the convolutional neural networks in the fuzzy region in parallel for convolution and pooling, and independently training each small data set by virtue of gradient descent. By virtue of the parallelization method, a network structure and parameters are optimized, and relatively good analysis performance and precision are realized; furthermore, the number of FR-CNN obfuscation layers is adjusted aiming at different log data sets, so that the extracted features can well reflect the characters of oil-gas reservoirs, and the fuzzification problem of the log data can be solved; and the parallel training and execution of FR-CNN are carried out by virtue of multiple GPUs, so that the efficiency of the FR-CNN is improved.

The invention discloses a natural image denoising method based on regionalism and dictionary learning. The natural image denoising method based on the regionalism and the dictionary learning mainly solves the problems that in an image denoising method based on kernel singular value decomposition (KSVD), blurring occurs in a weak texture region and fake texture occurs in a smooth region. The realization scheme includes that: removing high-frequency information of a noise-contained image through alternation of a stationary wavelet, and extracting structural information through a primal sketch algorithm, dividing the noise-contained image into three regions including a structural region, a texture region and a smooth region; obtaining a dictionary of the structural region and the texture region through a KSVD method; denoising the three regions respectively, merging denoising results, and obtaining a denoising image. An idea of combination of the regionalism and the dictionary learning is utilized, a dictionary which is obtained by the dictionary learning is enabled to conduct sparse presentation on corresponding signal composition of the image , information of edges and texture of the image is kept effectively, a denoising effect is improved, and the natural image denoising method can be used for obtaining high-quality images from noise-contained low-quality images.

The invention discloses a method for weighting multiple example studying features based on a master space classifying criterion. The realization scheme of the method comprises three steps of initializing a positive package representative example and a negative package representative example, building a problem to be optimized, and updating three kinds of unknown variables of the problem to be optimized. A representative example which can right express the category mark of a package in a positive package is found by adopting a heuristic search method, so that the problem of the fuzzification of the category mark of the example in the positive package is solved; repeated iteration is performed by adopting a coordinate rising method, so that the problem to be optimized can be converged into a local optimum solution; a relative weight is given according to the size of the contribution of each feature to recognition, and compared with the method of using original data for recognition, when the data which are weighted by features are used for recognition, higher recognition precision can be obtained.

The invention discloses a deep excavation risk evaluation method based on network reasoning, in particular to a deep excavation engineering risk evaluation method based on a risk tree network for probabilistic reasoning, which comprises the following steps: establishing a deep excavation engineering risk evaluation and computation model according to risk identification and analysis; taking preliminary statistic deep excavation engineering quality safety accident data as basis for probabilistic computation and loss assessment, and adopting the risk tree network to conduct the linear reasoning of risk event occurrence probability and the dynamic update of the probability; adopting expert investigation and historical experience to conduct the assessment of losses caused by risks; and combining the established risk evaluation model to evaluate deep excavation risks to obtain the overall risk level of the deep excavation engineering and the average risk level of each risk event. The invention has the advantages that the method can be used for conducting deep excavation engineering construction risk evaluation and dynamic risk evaluation, the risk evaluation is enabled to more accurate and objective, the dynamic risks in the whole construction process can be reflected, the relevant prevention, control and reduction of the construction risks are facilitated and the safety is guaranteed.

The invention relates to a cutting force model building and experimental testing method, in particular to a trapezoidal external thread turning instant cutting force model building and experimental testing method.The problem that according to an existing instant main cutting force study method, the influence mechanism of tool nose cutting motion track changes and tool cutting edge inclination changes on instant cutting force in the coarse-pitch thread turning process cannot be revealed is solved.The trapezoidal external thread turning instant cutting force model building and experimental testing method specifically comprises a tool nose cutting motion track under the vibration action, the instant cutting postures of a left cutting edge and a right cutting edge of a tool under the vibration action, instant cutting layer parameters of the left cutting edge and the right cutting edge of the tool, the instant cutting force of the left cutting edge and the right cutting edge of the tool, a coarse-pitch trapezoidal external thread turning experimental method and the instant cutting force of the left cutting edge and the right cutting edge of the tool during turning of a trapezoidal external thread with the thread pitch of 16 mm.The influence mechanism of the tool nose cutting motion track changes and the tool cutting edge inclination changes on instant cutting force in the coarse-pitch thread turning process is revealed.

The invention discloses an evaluation method for a hole collapse risk of a foundation of a cast-in-situ bored pile and application. On the basis of the basic principle of a fuzzy mathematic theory, the method is a fuzzy synthetic evaluation method for carrying out analysis and evaluation on the stability of a hole wall of the cast-in-situ bored pile and the hole collapse risk through multiple-factor analysis and evaluation. The fuzzy evaluation method not only evaluates and sorts evaluation objects according to the comprehensive scores, but evaluates the order of evaluation of the evaluation objects according to values on a fuzzy evaluation set and a maximum membership principle. The method overcomes the defect of oneness of a result of the conventional mathematical method, and the result contains abundant information quantity. The method is simple and feasible, and as for problems which cannot be subjected to number analysis from some traditional views, the problems of fuzziness and uncertainty of stability evaluation of many projects can be well solved.

The invention provides a reliability detecting and evaluating method of a power distribution system on the basis of a cloud model. The reliability detecting and evaluating method includes steps of searching line fault rate of a power distribution system to be detected and evaluated and historic statistics data of the fault rate of transformers; processing in standardization; calculating the cloud model digital characteristics of the line fault rate and the fault rate of the transformers by means of a backward cloud generator; calculating cloud droplets of the line fault rate and the fault rate of the transformers by means of a forward cloud generator; processing in inverse standardization; calculating reliability of the power distribution system by the feeder line partitioning algorithm to obtain values SAIFI, SAIDI and ASAI of the reliability of the power distribution system; drawing the values into diagrams and analyzing the reliability of the power distribution system to be detected and evaluated according to the diagrams. By the reliability detecting and evaluating method, qualitative laws of the parameters can be obtained, and the reliability of the power distribution system can be quantitatively detected and evaluated. In addition, the reliability detecting and evaluating method is good in universality and applicable to reliability detection and evaluation of the complicated power distribution system.

The invention discloses a comprehensive evaluation method and system for a mine construction project design scheme. A specific detailed evaluation index system is built according to the content of the mine construction project design scheme; and comprehensive evaluation of the projection design scheme is carried out by an AHP-fuzzy valuation method by combining the advantages of an analytic hierarchy process (AHP) and a fuzzy comprehensive evaluation method (FCE). The AHP can quantify a qualitative problem, provides a quantitative basis for decision by comparing the importance of various associated factors layer by layer and has relatively good scientificity and reasonability as a method for determining various levels of index weights; and the FCE can quantify experience judgment of a decision maker and can relatively well solve the fuzzy problem in comprehensive evaluation. A scientific and reasonable evaluation can be made for the overall effect of the projection design scheme; timely discovery of problems of the design scheme in some aspects is facilitated; completion and optimization of the design scheme are facilitated; and the comprehensive evaluation method and system have relatively good application value and practicability.

The invention discloses a formalized modeling based software security requirement acquisition method. The method includes functional requirement acquisition, security environment analysis, security objective analysis, security requirement elicitation and final generation of a system security requirement file, wherein the file at least includes information of system asset, behavioral sequence, defects, threats, security strategies, security assumptions, security objectives, security requirement levels, security guarantee levels, security functional modules and security guarantee modules. Compared with the prior art, the formalized modeling based software security requirement acquisition method has the advantages that automation in acquisition of software security requirements is realized while precision in acquisition of the security requirements is greatly improved; the problems of failure in realization of computer processing, ambiguity and fuzziness of security knowledge bases in natural language description are solved to lay the foundation for automatic elicitation of the software security requirements; the method which is a typical software security requirement acquisition method is applicable to different types of software systems and high in universality.

The invention belongs to the field of power distribution network reliability evaluation, and particularly relates to a comprehensive evaluation method for the reliability and economy of a radiation type power distribution network. A multi-objective optimization model based on the mixed integer nonlinear programming is constructed, so that the problem of complexity between the reliability investment and the reliability level improvement amplitude is solved, and the distribution network technical constraints and the reliability index constraints are fully considered during the multi-objective optimization process. A fuzzy membership function of the failure rate and the failure duration is constructed according to the historical power failure data by applying the fuzzy theory to estimate the power failure parameter failure rate and the failure duration, so that the problems of limited historical power failure data and power distribution system fuzziness are solved. The method provided by the invention can help the power supply company to analyze the investment cost required by a reliability transformation scheme, can select the optimal reliability transformation scheme on the premise that the distribution network meets a certain reliability level, and has a certain practical value.

The embodiment of the present application relates to a method for activating and deactivating a secondary cell and a terminal device. The method includes that the terminal device receives first indication information at a first start time that is sent by a network device to instruct the terminal device to activate or deactivate the secondary cell of the terminal device; the terminal device receives, at a second start time later than the first start time, second indication information that is sent by the network device to instruct the terminal device to activate or deactivate the secondary cell, wherein the terminal device does not activate or deactivate the secondary cell according to the first indication information at the second start time; the terminal device determines, according to apreset rule, that the secondary cell is activated or deactivated according to the first indication information or the second indication information after the second start time. The method for activating and deactivating a secondary cell and the terminal device in the embodiment of the present application can determine to perform activation or deactivation of the secondary cell in case of indication information conflict.

The invention discloses a method and a system for acquiring answers of tax business questions based on semantic graphs, and belongs to the technical field of intelligent questions and answers. The method comprises the steps of defining a question Q and an answer of the question Q according to a state transition graph G, and mapping a semantic interpretation process of the question Q into a sub-graph sub-G query process of the state transition graph G; generating a logic expression; action transfer and conditions in the logic expression are repeated; and obtaining a tax business problem, inputting the tax business problem into the model, checking whether the tax business problem completes the logic expression X, and outputting a tax business problem answer when the tax business problem completes the logic expression X. The semantic parsing process is mapped into the retrieval process of the query graph, so that the retrieval space of problem parsing is reduced, the operation between entities is judged by using the deep convolutional neural network, and the predicate synonym ambiguity problem mentioned in the preamble is solved.

A method performed by a network node (111, 112), namely eNB, for handling communications in a radio communications network (100) is provided. The network node (111, 112) is configured to be in the radio communications network (100). The network node (111, 112) receives a report of a radio link failure (RLF) report of user equipment (UE), which RLF report comprises an identity of a re-establishment cell, configures an RLF indication comprising the received RLF report to further comprise a first indication of the cell identity of the current cell the UE is connected and / or a second indication of a result of a comparison between the cell identity of the cell the UE was connected to upon sending the RLF report and the identity of the re-establishment cell, and sends the RLF indication to another network node (110), namely eNB.

The invention provides a gravitational dam risk assessment calculation method based on analytic hierarchy process and fuzzy comprehensive evaluation. The method comprises the following steps: designing a gravitational dam, including drawing up the profile of the gravitational dam, calculating the load of the gravitational dam, analyzing the stability of the gravitational dam, and analyzing the stress of the gravitational dam; and performing risk assessment on the gravity dam by performing risk estimation on the gravity dam, establishing fuzzy comprehensive evaluation on risk factors, performing risk evaluation on gravity dam engineering, and analyzing a risk evaluation result of the gravity dam engineering. The step of performing risk estimation on the gravity dam further comprises the following steps of performing fuzzy estimation on risk probability and establishing a risk evaluation standard.

The invention discloses a method for expanding Call-Info header carried service information in SIP information, comprising the following steps: service-param is expanded in the info elements of the Call-Info header in the SIP information and definition of the type of service-param is added; and an application server can add, delete or modify the info elements of the Call-Info header according to the service logic, wherein the info elements include a uniform resource identifier (URI) absolute, info-param and service-param. The method solves the problems of uncertainty and ambiguity existing when the existing Call-Info header is used, supports connect processing of different service logics when a plurality of services simultaneously use the header and simultaneously supports implementation of service interaction when a plurality of services use the header.

The invention discloses a visual sense-based fatigue driving identification method fusing heart rate and facial features. The method comprises the following steps: S1, extracting three fatigue characteristics, namely heart rate, eye opening degree and mouth opening degree, by using a depth camera; s2, applying a recurrent neural network (RNN) layer to obtain time information of the eye opening degree and the mouth opening degree; s3, combining fuzzy reasoning with RNN, and extracting time information of the heart rate; s4, extracting relations among the three features by using a relation layercontaining two stages of RNNs; and S5, outputting the fatigue degree of the driver. According to the method, the time sequence information related to the fatigue driving characteristics and the connection information among the characteristics are extracted, so that the fatigue detection performance of the driver is improved. Fuzzy reasoning and RNN are combined to solve the problems of fuzzinessand noise, temporal information related to the heart rate is extracted, the opening degree of eyes and the opening degree of the mouth can be determined more accurately through the depth image, and the driving fatigue recognition precision is effectively improved.

The invention discloses a heterogeneous distributed detection information target identification optimization method based on threat assessment, which is applied to an attack and defense countermeasuresystem simulation system. The method is characterized in that a heterogeneous distributed sensor network is formed based on signal-level semi-physical systems such as multiple satellites and multipleradars, the property of an attacking target is evaluated to form a threat sequence, and a ground command control system schedules the satellites and the radars to preferentially track and identify ballistic missile targets with high threat values according to the threat values in the threat sequence, and finally an interception or striking sequence is determined according to a target identification result. According to the method, redundant and complementary information fusion is carried out on collected detection information, target and environment information is collected and processed to agreater extent, accuracy and reliability of battlefield target identification are improved, and therefore the method is of great significance in subsequent situation evaluation, threat estimation andinterception strategy formulation, and the winning probability in confrontation simulation is greatly improved.

The invention relates to a comparison-ratio increasing method and device. The comparison-ratio increasing method includes the steps that when a mobile device works in a preset display mode, the gray level distribution information of a to-be-displayed frame is determined, and the preset display mode is a night-light display mode for showing that the screen luminance is lower than a preset luminance threshold value; according to the gray level distribution information, a comparison-ratio adjustment curve of the to-be-displayed frame is determined; according to the comparison-ratio adjustment curve, the comparison ratio of the to-be-displayed frame is increased. By means of the comparison-ratio increasing method and device in the technical scheme, the comparison-ratio adjustment curve of the to-be-displayed frame can be automatically determined, the comparison ratio of the to-be-displayed frame is further increased, the readability of the display content is improved, and the problem that as the comparison ratio of the display content cannot be increased in the related technology, the display content is fuzzy is solved.

The invention discloses a visual emotion label distribution prediction method based on automatic estimation, and belongs to the technical field of computer vision. According to the method, the problemof fuzziness existing in visual emotion is solved through label distribution learning, and meanwhile a depth framework capable of conducting emotion label classification and label distribution prediction at the same time is provided. In addition, as most of the visual emotion data sets only provide a single category label, in order to improve the practicability of the framework, weak priori knowledge, namely similarity information among the labels, is utilized, and corresponding emotion label distribution is generated according to emotion categories. In the process of framework learning, forclassification tasks, a Softmax function is used for constraint, and for distributed tasks, Kullback-Leibler (KL) loss is used for constraint. And the two loss weighting values are added to obtain a final loss function, and end-to-end label distribution prediction of the frame is realized.

The invention provides a logging phase recognition method based on a fuzzy theory and a neural network, which comprises the following steps: firstly, constructing a fuzzy region convolutional neural network, putting a given target assumed region and target recognition into the same network, sharing convolution calculation, and updating the weight of the whole network in a training process; next, performing convolution and pooling operation on the logging data through a fuzzy region convolutional neural network; interacting the convolution layer and the pooling layer; carrying out fuzzy operation on the convolution layer and the pooling layer; starting from a first layer of a fuzzy region convolutional neural network so that the number of fuzzified layers is gradually increased; adjusting the number of fuzzy layers according to different data sets, the last layer of the fuzzy region convolutional neural network obtains a feature vector, the feature vector maps features into a low-dimensional vector through a sliding window, then the features are input into two full-connection layers, one full-connection layer is used for positioning, and the other full-connection layer is used for classifying.

The invention discloses a traffic intersection risk level evaluation method and device, electronic equipment and a storage medium and belongs to the technical field of traffic safety evaluation. According to the method, an index weight determination method based on local difference is proposed to determine the weight of each risk index passing through the intersection among different objects fromthe nonlinear characteristic of the traffic intersection evaluation standard, and risk level evaluation is performed in combination with a cloud model, so a problem of fuzziness of each evaluation index is effectively solved, and risk level evaluation accuracy is improved. A conflict evaluation index attribute discrete standard is established, natural conversion between qualitative concepts of conflict risk levels among different objects and conflict evaluation index quantitative concepts is achieved, and evaluation accuracy and reliability are improved; problems that an existing traffic intersection conflict evaluation index is single in selection and unreasonable in weight distribution are solved.

The invention discloses a VLC dynamic positioning method and system based on mean shift and unscented Kalman filtering. The method comprises the following steps: firstly, controlling an LED lamp to beon and off at high frequency by an LED driving circuit, finding an area where an LED exists through LED-ID identification, identifying the ID of the LED, and obtaining the initial position of the positioning terminal; secondly, dynamically tracking LEDs in the image sequence by using a mean shift algorithm and unscented Kalman filtering, and calculating the relative positions of the LED pixel coordinates of the current frame and the LED pixel coordinates of the initial frame; and then, obtaining the position of the positioning terminal in the real world in combination with the relative position relationship between the initial position of the positioning terminal and the positioning terminal in a subsequent frame, so that real-time positioning is realized. The method has the capability oftracking the high-speed target, improves the positioning precision when the LED is shielded, and can maintain the precision even if half of the LED is shielded. In addition, the method has good robustness and real-time performance, and has a wide application prospect in the field of indoor positioning.

The invention discloses a pollution source area detection method and method for delineating LNAPLs in aeration zone based on ERT and IP. According to the technical scheme, the method comprises the steps of determining an ERT and IP detection wiring mode; collecting ERT data and IP data of the same measuring line according to the wiring condition; carrying out data preprocessing and data inversionto obtain resistivity and polarizability profile maps; and delineating a suspected pollution area according to an inversion result. According to the method, the problem of fuzziness of the ERT in delineating the range of the LNAPLs pollution source area in the aeration zone is solved, and a reliable basis is provided for accurate positioning of the LNAPLs pollution source area.

The invention relates to a milling cutter designing method and a milling cutter, in particular relates to a method for designing a low-entropy and safe high-speed milling cutter and the high-speed milling cutter, and aims to solve the problems that the entropy cannot be controlled due to disordering of particle swarms of the milling cutter and the safety of the high-speed milling cutter is reduced. According to the characteristic that the disordering of the particle swarms of a high-speed milling cutter assembly is associated with the reducing of the safety, a safety reducing process is described by virtue of an entropy determination method; by virtue of an entropy model, a mesoscopic motion state is judged, and the safety reducing process is determined; by virtue of an entropy control method, a control variable for controlling the safety reducing process is revealed, and the disordering of the particle swarms can be effectively controlled; macroscopic and mesoscopic structure parameters are collaboratively designed by virtue of a low-entropy high-speed milling cutter safety designing method; the safety and stability of the low-entropy milling cutter are verified. The diameter of the milling cutter is 63mm, the number of tool teeth is 4, the tool teeth are unequally distributed, included angles between the teeth are 88 degrees, 89 degrees, 90 degrees and 93 degrees, the tooth root of a tool body is of a chamfer and transition circular arc structure, and a front blade installation angle is 2 degrees. The method is applied to milling cuter designing and high-speed milling.

The invention provides a single end-point characteristic description based line segment matching method. According to the single end-point characteristic description based line segment matching method, one end-point is used as a main characteristic point and the other end-point is used as an auxiliary characteristic point to construct a characteristic description unit; a specific line segment direction is used as a reference direction of the characteristic unit and a characteristic descriptor is constructed and main characteristic point matching is carried out; and by a matching relationship of any end-points of two line segments, the matching relationship of two line segments is determined, and finally, verification of line segment matching correctness is completed by geometric verification of the matched characteristic points. According to the invention, the calculation complexity is greatly simplified and the matching speed is obviously improved. As long as any end-points or breakpoints on the corresponding line segments of two images can be matched, the two line segments or sub line segments of the line segments can be matched. The line segment direction is used as the reference direction of the descriptor, so that the ambiguity problem of calculating the reference direction according to the neighborhood gradient is solved, the wrong match of two line segments, of which the end-points are close, but the directions are different, is eliminated and the robustness of the line segment matching is improved.

The invention provides a method for judging whether a radiation source radar belongs to a target platform. The method comprises the following steps: calculating the membership degree of each characteristic parameter of each radar in a database to which each intercepted radiation source belongs; carrying out normalization processing; calculating the entropy weight of each intercepted radiation source belonging to each radar in the database; selecting the maximum entropy weight from the entropy weights of the radars belonging to the database of each interception radiation source as the credibility of the corresponding radar belonging to the database of the corresponding interception radiation source, and constructing a credibility matrix of the interception radiation sources and the radars; obtaining a target platform associated with all radars in the database in which the interception radiation source belongs and corresponding confidence from the database; constructing a confidence coefficient matrix of the radar and the target platform; multiplying the credibility matrix by the confidence coefficient matrix; and selecting the platform corresponding to the maximum multiplication result value as a target platform to which the corresponding radar belongs in the database to which each interception radiation source belongs. According to the invention, the discrimination accuracy and discrimination efficiency of the radiation source platform target are greatly improved.

The invention discloses an intelligent charging station site selection optimization method based on a fuzzy TOPSIS method. The intelligent charging station site selection optimization method comprisesthe following steps: 1) establishing an evaluation index system related to unattended intelligent charging station site selection; 2) evaluating the selected indexes, using a triangular fuzzy numberto represent the evaluation indexes, and further obtaining the evaluation index performance and the corresponding weight of each site selection scheme; and 3) sorting all the alternative schemes by using a TOPSIS method, and selecting an optimal site selection scheme. According to the invention, the problem of a fuzzy phenomenon generated in the evaluation process of the evaluation indexes can bewell solved, the calculation process is simple and clear, the influence of subjective factors on the evaluation process is small, and the method has high objectivity and high operability in the wholeevaluation process.