39 results about "Optimal criterion" patented technology

The invention discloses a power allocation algorithm for a UAV (unmanned aerial vehicle) relay system based on an energy efficiency optimal criterion. Firstly, based on a double-hop amplification andrelay transmission model, an optimization model of power allocation is established, and a power allocation problem is converted into an optimization problem of the solution of maximum system energy efficiency. In the process of solving the optimal power allocation, firstly transmitted signal power is fixed to obtain a beamforming optimization scheme. Then, by a large signal-to-noise ratio intervalapproximation, an original non-convex optimization problem is converted into a convex optimization problem. Finally, a KKT condition is used, and a closed form solution of a power allocation scheme is calculated and obtained. A simulation experiment shows that the closed form solution obtained by the algorithm disclosed by the present invention is close to a loop iterative method, and thus the algorithm complexity is reduced.

The invention discloses an OFDM relay network resource allocating method based on the information-and-energy-simultaneous-wireless-transmission. This method constructs an optimization model for the system throughput under the condition that the peak power of a source node is restricted, the peak power of each subcarrier is restricted and the energy causality of the relay node is restricted. In the case that the subcarrier allocation strategy of the source node is given, the optimal criteria for the original problem are sought. Then, through the arrangement of the general energy transmission power of the source node, the original problem is decoupled as two separate sub-problems of the maximization of the throughput, which reduces the complexity of the original problem for solutions considerably. The decoupled sub-problems are then solved according to a water injection manner. Finally, the invention provides four subcarrier allocation schemes, referred to as the energy optimizing scheme, the information optimizing scheme, the equilibrium allocation scheme and the exhaustion scheme with the first three schemes reducing the complexity of the sub-carrier allocation significantly.

A method for computer-supported control and / or regulation of a technical system is provided. In the method a reinforcing learning method and an artificial neuronal network are used. In a preferred embodiment, parallel feed-forward networks are connected together such that the global architecture meets an optimal criterion. The network thus approximates the observed benefits as predictor for the expected benefits. In this manner, actual observations are used in an optimal manner to determine a quality function. The quality function obtained intrinsically from the network provides the optimal action selection rule for the given control problem. The method may be applied to any technical system for regulation or control. A preferred field of application is the regulation or control of turbines, in particular a gas turbine.

A method for combined computed tomography (CT) imaging and positron emission tomography (PET) imaging uses respiration-gated CT imaging in which the optimal criteria for CT scan gating are determined after the PET scan has been performed. After acquisition of first CT scan image data and PET scan image data with strain gauge levels being recorded, optimal gating criteria are calculated based on the strain gauge levels, and a second CT scan is then performed with triggering in accordance with the optimal gating criteria.

The invention discloses a lithium iron phosphate cell lifetime prediction method based on MIV and an SVM model. The method comprises steps of by applying the MIV algorithm, obtaining the influence importance degree of an input variable to an output variable; then, screening out the most important variable to serve as the input variable; preventing unimportant independent variables to be introducedinto training and test processes of the prediction model; after variable optimization, obtaining a new training set and a new test set only including the optimal variables; and by use of the optimaltraining set and the SVM, carrying out training to obtain the prediction model. According to the invention, the structural risk minimization is used as the optimal criterion in the SVM, so the globaloptimal solution can be acquired; and by combining the prediction model which is subjected to the optimization, only includes variables of circulation times, resistance and the like and is obtained through training, prediction efficiency and prediction precision can be effectively improved.

The invention discloses a power demand-side flexible load active power prediction method The prediction method includes the steps of acquiring the historical data of the power demand-side flexible load active power, and counting and calculating the flexible load active power values under the conditions of different weather and time; determining the influence factors of the flexible load active power based on the change trend and features of the flexible load active power; constructing a demand-side flexible load model based on the conventional normal load component, the special event load component, and the random load component of the flexible load; and performing the combined prediction of the demand-side flexible load active power based on the optimal criterion of weighted average of the weight and the fitting degree. According to the method, the prediction accuracy of the demand-side flexible load active power is improved, the technological support for the safe, economic, and stable running of a power grid is provided, and the wide application and efficiency running of the demand-side electric automobile industry and the distributed power and energy storage system are greatly enhanced.

The invention provides a slurry circulating pump optimization method and system based on data driving, and the method comprises the steps: obtaining the historical operation data of related equipmentas the training data of a model, and selecting an index for data preprocessing; constructing a working condition division model by adopting a mean shift algorithm, and obtaining optimal values under different working conditions according to an optimal criterion; acquiring real-time operation data of related equipment as test data, and selecting indexes to perform data preprocessing; and matching the working condition types of the test data based on the working condition division model of the training data to obtain an optimization value, namely an optimization target value of the test data. According to the method, massive historical data of the slurry circulating pump equipment is fully mined through a clustering modeling method, an efficient and practical model is established to performindex optimization and optimization on the slurry circulating pump in real time, and the result relatively more conforms to production operation and is more reasonable and accurate.

The invention relates to a low-sidelobe robust adaptive beamforming method. Firstly, an optimal criterion based on poorest performance is established, and it is ensured by adopting the beamforming method that lower null level can be obtained while low sidelobes are formed when array errors exist after cross coupling correction; then, a new beamforming system is established by combining a robust low-sidelobe theory and can adapt to interference competition and reduction of sidelobe level and achieve wide null. Especially, the system can select reasonable parameters in a self-adaptive mode according to an input signal, so that a beam forming device can obtain unified low sidelobe levels under different receiving signals.

The invention discloses a support vector machine mechanical fault diagnosis method based on a voting mechanism. The method comprises three different feature analysis algorithms: power frequency harmonic surface feature analysis algorithm, a multi-resolution band-pass filter tank spectrum analysis method and Hilbert-huang conversion feature analysis method; the feature extraction is performed on asound signal emitted under the mechanical equipment fault operation state, and then classification identification is performed by utilizing the support vector machine algorithm. The adopted three-feature analysis algorithms have good applicability for the mechanical noise and like unstable nonlinear signal, and the time frequency feature of the signal can be self-adaptively described in multiple aspects; the classification algorithm adopts a support vector machine multi-classifier based on the voting mechanism; the classifier decomposes the multi-classifier into multiple second-class classifiers by adopting a one versus rest method, and the classification method can achieve high identification rate and good algorithm robustness based on the voting mechanism and the confidence optimal criterion judgment method.

The invention discloses a particle reinforced material component robust topological optimization method by considering mixing uncertainty. The method comprises the following steps: considering uncertainty of a component using a particle reinforced material in manufacturing and use, regarding an insufficient external load of a sample as interval uncertainty, and regarding sufficient matrix and particle material attributes of the sample and volume distribution of a particle reinforced phase in the matrix as bounded probability uncertainty; discretizing the volume distribution of the design domain and the particle reinforced phase, setting physical and geometric constraints, and establishing a robust topological optimization design model; and iteratively solving the robust optimization designmodel considering the interval and bounded probability mixed uncertainty by using an optimal criterion method. The component steady topological optimization design model established by adopting the method truly reflects the distribution characteristics of multi-source uncertainty in actual engineering, and the proposed structural performance statistical moment estimation method and the optimal criterion method are used for solving, so that the optimization result can be efficiently obtained, and the method has a very good engineering application value.

The invention provides a robust global threshold segmentation method. The method includes the steps that step1, a threshold value To of a gray-level image is calculated through an Otsu threshold value method; step2, a balance degree factor eta is defined through the variance of a foreground and a background of the gray-level image, and a threshold value Tb with the balance degree factor as the criterion is acquired; step3, a deviation degree lambda and a new optimal criterion zeta are defined according to Tb and To, and a segmentation result is generated. By means of the method, the optimal threshold value can be obtained under the condition that the probability distribution difference between the foreground and the background of the grey-level image is obvious or under general conditions.

The invention discloses a selection method of a cognitive network path. The method comprises the following steps: a secondary user source node, a secondary user relay node and a secondary user destination node respectively obtaining various parameters required for analyzing outageperformance of the cognitive network under mutual interference scene of a primary userand a secondary user, establishing an outage probability closed form expression of any one path in a cognitive relay network under a Rayleigh fading channel in combination withinterference limitation allowed by a primary user receiver end, selecting an optimal path of information transmission from a source node to the destination node in the network according to the optimal criterion of system end-to-end outage performance, and selecting a network path by using minimum standard of the cognitive network outage probability. The technical scheme disclosed by the invention gives full consideration to interference factors of the primary user to the secondary user in the cognitive network path selection and takes corresponding measures so as to improve the reasonability of timely selecting the cognitive network path.

The invention discloses a motion error compensation method for a synthetic aperture radar frequency domain BP algorithm. On the basis of an image sharpness optimal criterion, an equivalent phase center error is estimated to carry out motion compensation. Imaging is carried out by using a constant-speed linear motion trajectory of a radar platform based on a frequency domain BP algorithm, so that an unfocused SAR image obtained by frequency domain BP algorithm imaging is obtained; with SAR image sharpness as an objective function, an antenna phase center position error is estimated by the optimization technique; and then an estimated APC position error is added to the constant-speed linear motion trajectory to obtain an APC absolute position, and final high-precision imaging is performed byusing the frequency domain BP algorithm. According to the invention, imaging processing is performed in the frequency domain and the distance history from a scattering point to an array element doesnot need to be calculated, so that the required computing time is substantially reduced by being compared with that of the time domain self-focusing BP algorithm. Therefore, the method is suitable forlarge-scene long-apertures high-precision SAR imaging.

The invention discloses an underwater heat source detection and location method based on water body hierarchical filtering, and belongs to the crossing field of marine science, thermal physics and pattern recognition, the method is used for hierarchically filtering the water body, filtering depth is determined through an optimal criterion, and a specific location of the underwater heat source is accurately reflected. The method comprises a step for thermal radiation simulation of an underwater heat source in a water body, a step for actually measuring and acquiring an infrared image of an underwater target, a step for hierarchically filtering the water body and a step for detecting and locating the underwater heat source. According to the invention, a thermal field model of a hierarchical ocean water body is established, a thermal field model of a simplified underwater heat source including an underwater vehicle is further established, and a result of the thermal field model of the underwater heat source in the water body can prove that thermal radiation of the water body is hierarchically distributed; according to the acquired infrared image and the hierarchical theory of the water body thermal radiation, the water body hierarchical filtering method can be used for performing detection and location of the underwater heat source, and precision of detection and location is improved.

The invention provides a blind source separation method based on a quantum rhinoceros search mechanism, and designs a hybrid optimization objective function designed based on two different independence criteria, i.e., the hybrid optimization objective function is designed based on two independence criteria of maximization kurtosis and maximization negentropy, and the two criteria are endowed withcorresponding weight coefficients. The optimal criterion of the intelligent calculation method can be judged according to the change condition of the hybrid optimization objective function value alongwith the weight coefficient, so that a more accurate blind source separation result is obtained. Furthermore, a blind source separation method based on a quantum rhinoceros search mechanism and a hybrid optimization objective function is designed. The method designed by the invention can realize blind source separation of aliasing signals, has the advantages of the high convergence speed, high separation precision, stable performance and the like, and has a wide application prospect.

The invention provides a joint optimal criterion based waveform design method of a cognitive SAR. The method comprises the following steps that S1) a system parameter is initialized; S2) a joint optimal criterion objective function is established to maximize the echo SCNR and minimizing a time delay distinguishing constant; S3) a Lagrange multiplier method is used to convert the constraint optimization problem in the step S2) into a constraint-free optimization problem; S4) the constraint-free optimization problem in the step S3) is solved to obtain an expression of the optimal emission waveform based on the joint optimal criterion; and S5) according to an energy constraint condition, a Lagrange multiplier of the optimal emission waveform based on the joint optimal criterion in the step S4) is solved, and the optimal waveform is obtained. The emission waveform of the SAR is designed to use environment and target information, clutters are inhibited in the emission end, the SAR has a cognition capability, and the designed cognitive SAR has a higher distinguishing performance while inhibiting noises.

The present invention relates to a satellite-borne sparse voyage over SAR three-dimensional imaging method under a large orbit distribution range. The method comprises: acquiring an echo signal, and performing pulse compression processing in a range direction to obtain range direction-focused data; performing distance migration correction, azimuth compression and range inverse pulse compression processing on the range direction-azimuth direction two-dimensional data corresponding to each orbital direction sampling unit to obtain azimuth direction-focused data; establishing a range direction-orbital direction two-dimensional linear measurement matrix for range direction-orbital direction two-dimensional data corresponding to each azimuth direction resolution unit; using the compressed sensing theory and the l1 / 2 norm optimal criterion to establish the optimal equation and solving the optimal equation to obtain a backscattering coefficient of the target in the range direction-orbital direction; and performing three-dimensional coordinate transformation to obtain three-dimensional imaging results. According to the method provided by the present invention, the problem of image aliasingcaused by sparse voyage oversampling, and the problem of range direction-orbital direction data coupling under the large orbit distribution range are solved, and the image quality can be effectivelyimproved.

The invention discloses a robust adaptive beamforming method based on spatial smoothing. The invention relates to the field of array antenna self-adaptive beam forming, and aims to solve the problemsthat expected signals are coherent with interference signals, a covariance matrix of array receiving signals can generate a rank loss phenomenon, the expected signals can be cancelled by using a traditional self-adaptive beam forming algorithm, and the performance of a beam forming device is sharply reduced. On the basis of the worst performance optimal criterion, interference and noise power output by the array are minimized on the basis of ensuring that all possible expected signal steering vectors and sampling correlation matrixes pass through a beam former without attenuation. According tothe invention, the phenomenon that the covariance matrix of the array receiving signal has rank loss due to coherence of the expected signal and the interference signal is overcome, and the performance of the beamformer is improved.

The invention discloses a fog penetration identification method for an intelligent traffic monitoring image. The method mainly comprises the following steps: A, acquiring a traffic monitoring image, constructing a video image signal wavelet coefficient, and denoising the image according to the setting of a coefficient threshold; B, constructing a foggy day image degradation model; preliminarily estimating the transmissivity of the image according to a dark channel prior idea, and obtaining a clearer defogged image by adopting a rapid guide filtering method; c, constructing a novel filter basedon phase type distribution and characteristics of a related random model, constructing a consistent approximation method according to the approximation of the filter, and enhancing the response of the filter to image signals; and D, selecting an edge detection optimal criterion to carry out edge detection and segmentation of the image according to consistent approximation of the novel filter, andfinally identifying the image to complete image identification of traffic monitoring. The method has good real-time performance and implantability, the monitoring image can be processed according toimage denoising processing, dark channel defogging processing and edge detection methods, and the fog penetration identification task of the traffic monitoring image is effectively completed.

The invention provides an anti-interference method and device for array antennas. The method comprises following steps: N sets of digital complex signals which are processed with digital quadrature interpolation and are corresponding to N radio frequency channels are received, wherein the N is an integer larger than 1; a cross correlation coefficient matrix corresponding to the N sets of digital complex signals is generated; a constraint steering vector consistent to a preset self-adapted selection condition is calculated on the basis of the cross correlation coefficient matrix; an anti-interference weight value is determined in combination with a preset optimal criterion on the basis of the constraint steering vector. By use of the provided anti-interference method and device for array antennas, the problem in the prior art of weakened anti-interference effects of the array antennas at some angles is solved.

The device (D) has an interfacing module arranged for exchanging request and response message with mobile terminals (T1-T3). A processing module (MT) is arranged for determining geographical position of the required mobile terminal, determining passenger drop off stations (S21-25) based on the determined position and optimal criteria and generating a response message, comprising coordinates of the determined station, to the destination of the required terminal during reception of request for obtaining information on the station of transport units (V21-22).

The invention provides a calculation method of function sizes of part hole series. The method comprises the steps of measuring and obtaining measuring point coordinates on various measured holes, obtaining ideal positions of the holes according to drawing information, converting into a function size of a circle for calculation through coordinate translation, applying a method for solving a nonlinear equation system, changing the diameter of a function surface continuously till a criterion is met, and outputting the function sizes of the hole series and position coordinates of the holes. The method can accurately calculate optimal criterion meeting values of the function sizes of the hole series and the position coordinates of the holes.

The invention belongs to the technical field of event trigger fault detection, and discloses a time-varying random system event trigger fault detection method based on coordinate transformation, and the method comprises the steps that: superposed unknown interference is removed from one subsystem by employing a coordinate transformation method; for each subsystem, an optimal event trigger filter is constructed by utilizing an unbiased and minimum variance optimal criterion; when any system breaks down, a residual error will exceed a preset threshold value, current measurement information willbe immediately sent to a remote filter, and a diagnosis decision triggered by an event does not need to be input. According to the invention, a coordinate transformation method is utilized to realizethe characteristic that the fault detection residual error is only sensitive to system faults and has robustness to additive unknown interference; a considered system can be converted into two subsystems, and interference is removed from one subsystem; and further the optimal gain of each filter is derived, and the optimal gain is applied to an optimization criterion of unbiased and minimum mean square estimation errors.

The invention provides a calculation method of function sizes of part hole series. The method comprises the steps of measuring and obtaining measuring point coordinates on various measured holes, obtaining ideal positions of the holes according to drawing information, converting into a function size of a circle for calculation through coordinate translation, applying a method for solving a nonlinear equation system, changing the diameter of a function surface continuously till a criterion is met, and outputting the function sizes of the hole series and position coordinates of the holes. The method can accurately calculate optimal criterion meeting values of the function sizes of the hole series and the position coordinates of the holes.

The invention discloses a method for improving the overall pose accuracy of components by optimizing and constructing measurement points. According to the coordinates of the measurement points on the components before the posture adjustment and positioning and the coordinates after the posture adjustment and positioning, firstly by constructing effective criteria, Eliminate the points that do not meet the criteria to form valid measurement points, and then, by constructing a transformation matrix, match the components at the initial position and their valid measurement points to the target position, and obtain the construction points of the valid measurement points on the components at the target position and its coordinates. By constructing the optimal criterion, the optimal effective measurement points are obtained from the construction points, and the correction transformation matrix is ​​constructed again for the optimal effective measurement points, and the components at the initial position and the optimal effective measurement points at the initial position are corrected and matched to the target position through the correction transformation matrix, Obtain the corrected construction point and its coordinates of the preferred valid measurement point on the component at the target location.

The invention discloses an online performance evaluation method and system based on a nonlinear generating function and weight optimization. The method comprises the following steps that 1) performance data is filled; 2) according to the filled performance data, performance scores of different evaluated objects corresponding to different evaluation criteria are calculated, and an original score matrix X is obtained; 3) the original score matrix is standardized to obtain a standard score matrix Y of n rows and m columns; 4) a weight optimization model (MIN-MAX model) is established; 5) the weight optimization model is solved, and optimal evaluation criterion weight vectors are calculated; and 6) the standard score matrix Y in step 3) and the optimal criterion weight vectors in the step 5) are used to calculate a comprehensive performance score of all the evaluated objects. An evaluation result of the invention is objective, accurate and effective.

The invention discloses a high-performance material optimal selection method suitable for a space debris protection structure. The method comprises the following steps: a, selecting a protective material, and looking up a table to obtain the Hugoniot and thermodynamic parameters of the material; and b, screening the protective material according to an impact pressure criterion, an internal energyconversion criterion, a minimum thickness criterion and a kinetic energy absorption criterion to obtain a preferred material. In the present invention, the protection mechanism of the protective material under the ultrahigh-speed impact effect is deeply analyzed; the mechanical and thermodynamic characteristics of the protective material are comprehensively considered; the impact pressure, the internal energy conversion, the minimum thickness and the kinetic energy absorption characteristic of the protective material are used as optimal criteria; in combination with a performance coefficient model capable of quantitatively characterizing wave impedance characteristics and thermodynamic characteristics of the protective material, the high-performance material optimal selection method suitable for the space debris protection structure is obtained, optimal selection of various protection materials can be achieved, the test cost for material selection is effectively reduced, and theoretical guidance can be provided for design and preparation of novel high-performance protection materials.

The invention provides a method and a device for acquiring thermal extrusion technological parameters of a nickel base alloy tube stock. The method comprises the following steps: performing numerical simulation analysis for a thermal extrusion process of a tube blank on a thermal extrusion model, thereby acquiring an optimal criterion data of the tube blank under multiple sets of different thermal extrusion technological parameter alternative data; constructing a relation curve between each thermal extrusion technological parameter and the optimal criterion data and confirming a value interval of the thermal extrusion technological parameter corresponding to each optimal criterion datum according to the relation curve and a preset criterion selection condition; performing intersection treatment on the value intervals of the same thermal extrusion technological parameter corresponding to all the optimal criterion data, thereby acquiring optimal values of the thermal extrusion technological parameters. According to the invention, multiple technological parameters of the nickel base alloy tube stock in a thermal extrusion process can be more accurately acquired and the practical production can be more effectively guided.