160 results about "Functional equation" patented technology

The invention provides a method for calculating chemical mechanical polishing material removal rate. The method comprises the following steps: determining a calculation formula MRR=kPV of the material removal rate; analyzing a factor which affects the material removal rate, and obtaining a function equation of the factor which affects the material removal rate; and substituting the function equation of the factor into the MRR=kPV, obtaining a specific calculation formula of the material removal rate, and calculating the material removal rate MRR according to the specific calculation formula. Besides, the invention further provides a device for calculating the chemical mechanical polishing material removal rate. According to the schemes provided by the invention, the action rule among multiple bodies is deeply revealed by the analyzing of the factor which affects the material removal rate and the comprehensive considering of an interaction relationship among four bodies including wafers, particles, polishing pads and polishing fluids, the actual process of chemical mechanical polishing can be relatively objectively and truly described, and the method for calculating the chemical mechanical polishing material removal rate has active guiding functions on the mechanism analysis and the layout design of the CMP (Chemical Mechanical Polishing), as well as the research and the development of a large production line process.

The invention relates to a minimum base attitude disturbance track planning method for a redundant space manipulator, and belongs to the technical field of aerospace. The method comprises the steps of (1), designing a three-dimensional model of the redundant space manipulator; (2), establishing a kinematic model of the redundant space manipulator; (3) establishing a kinematic equation of the redundant space manipulator by employing a generalized Jacobian matrix; (4), parameterizing joints by employing a 5-order sine polynomial function; (5), establishing a cost function equation under a minimum constraint condition of base attitude disturbance; (6), providing an improved hybrid particle swarm optimization; and (7), carrying out optimization solution to the cost function by employing the hybrid particle swarm optimization, thereby obtaining planning tracks under the constraint condition. According to the method, the track planning problem of the redundant space manipulator under the minimum constraint condition of the base attitude disturbance is solved, and the planning tracks solved by the improved hybrid particle swarm optimization is high in precision and good in effect and is steady.

The invention discloses a noise parameter measurement uncertainty evaluation method based on a multi-chip module (MCM) and belongs to the field of measurement metering. The method includes utilizing an equivalent noise parameter equation as a measurement model for noise parameter measurement uncertainty evaluation, utilizing physical quantity measured from a measurement platform to calculate the function equation and obtain data reflecting equivalent noise parameter distribution condition, leading out noise parameter and noise parameter distribution from the equivalent noise parameter and finally obtaining uncertainty for noise parameter measurement. The method effectively combines digital random simulation, physical measurement boundary criterion and least square method optimization criterion, and enables noise parameter uncertainty evaluation to be effective, real and reliable. The method satisfies the national standard, and is flexible in simulation mode, multiple in applicable measurement system variety and strong in universality.

The invention discloses a hybrid electric vehicle energy conservation predictive control method based on traffic signal lamp information. Real-time road grades and the traffic signal lamp information are acquired from the global positioning system and an intelligent transportation system to serve as system input, and a hybrid electric vehicle mathematic model is established to serve as a reference for predicting further vehicle states; a hybrid electric vehicle energy management optimal control problem is defined, and a functional equation for solving the optimum control quantity is provided; optimum control is fed back in real time to solve the optimum control quantity; under the condition of meeting hybrid electric vehicle storage battery electric quantity balance and rotation speed and torque physical limit restraints, energy flow of a hybrid electric vehicle is adjusted and optimized on line according to practical vehicle driving working conditions, the system optimal performance of the hybrid electric vehicle is obtained, a planetary gear mechanism is used to serve as an electric infinitely variable transmission, an engine is made to constantly work at the best operating point, road traffic information is used for predicting the charge state of a target storage battery, the energy flow of the hybrid electric vehicle is adjusted on line, and therefore the purposes of energy conservation and emission reduction are achieved.

A method, apparatus and computer program product are provided for facilitating combinatorial logic modeling at an asynchronous clock domain crossing. The modeling technique employs a simulation value of X in combinatorial logic at the asynchronous clock domain crossing of a circuit being modeled to facilitate modeling of a potential combinatorial logic glitch at the crossing during metastability periods thereof. Employing the simulation value of X includes: generating one or more equivalent functional equations for one or more combinatorial paths through the combinatorial logic at the crossing; propagating the simulation value of X through the combinatorial logic using the at least one equivalent functional equation; and then converting the simulation value of X at an output of the combinatorial logic of the asynchronous clock domain crossing to a random logic value for further propagation within the circuit being modeled.

The invention discloses a cost based method for optimizing an external PHEV (Plug-in Hybrid Electric Vehicle) power assembly and application thereof. The method is characterized by comprising the steps of firstly confirming variables to be optimized in a PHEV power assembly design according to an external PHEV power assembly arranging structure and establishing a PHEV cost objective functional equation and a constraint condition; then figuring out the variable to be optimized of the PHEV power assembly, which meets the constraint condition and has the lowest cost, by utilizing the cost objective functional equation and the constraint condition and by adopting a quadratic programming or matrix partition optimization algorithm; and then obtaining an optimum design scheme of the PHEV power assembly, which meets the constraint condition and has the lowest cost. By means of the method provided by the invention, reasonable quantitative indicators can be provided for the manufacturing and the use cost of the external PHEV and a fast and efficient design for a PHEV power assembly system is realized.

The invention discloses a hybrid electric vehicle energy-saving predication control method based on decentralized control. The method comprises the following steps that real-time vehicle traffic information, real-time front vehicle traffic information and real-time rear vehicle traffic information are obtained from a global positioning system, an inter-vehicle communication system, a vehicle condition communication system and an intelligent traffic system to serve as system input; a hybrid electric vehicle queue driving decentralized control mathematical model is built to serve as a basis for predicating the state of a vehicle in the future; the hybrid electric vehicle decentralized control queue driving optimal control is defined, and a functional equation for solving the optimal control quantity is provided; the optimal control is fed back in real time, the optimal control quantity is solved, a planet gear mechanism is used as an electronic infinitely-variable transmission, and an engine works at the optimal working point; and road traffic information is applied to predicate the driving state of the front vehicle and the driving state of the rear vehicle, and energy flowing of the hybrid electric vehicle is adjusted online, so that the targets of energy saving and emission reduction are achieved, the calculation time is greatly shortened, and real-time control characteristics of the vehicle are improved.

The invention relates to a pre-stack three-parameter inversion implementation reservoir stratum and fluid prediction method based on mixed norm regularization. Based on the Bayesian parameter estimation theory, the AVO basic principle, the comprehensive sparse theory, remaining residual distribution, geological significance and other factors, a mixed penalty functional equation is built, and regularization constraint is performed on the pre-stack three-parameter inversion process. A model and actual material application prove that result stability and geological significance can be improved through mixed norm regularization, and particularly, the density inversion result is more accurate under the situation of the same offset distance and the same signal-to-noise ratio. The method has the advantages of being capable of being directly applied to actual materials, a reasonable inversion result is obtained, and the reservoir stratum and fluid prediction accuracy is improved.

The invention discloses a weld seam fatigue life calculation method based on total strain energy density. The method mainly comprises the steps that 1, the weld-seam total strain energy density, a life function equation and a fatigue strength coefficient, a fatigue strength index, a cyclic strengthening coefficient and a cyclic strain hardening index which are needed by half-cycle life stress and a strain function equation are acquired through a welded joint fatigue test; 2, the weld seam structure is simulated through a shell unit module, the stress-strain response of the weld seam structure under the cyclic load action is calculated through a multi-load step method, and a stress-strain hysteretic curve is output; 3, the total strain energy density is calculated according to the stress-strain response, and the hot point fatigue life and a damage value of a weld toe of the weld seam are calculated by combining the energy density with the life function equation. According to the method, contribution of elastic-plastic stress and strain to the fatigue damage is comprehensively taken into account, scalar quantities are taken as damage parameters, the position and direction problems related to vectors are effectively avoided, the calculation precision is improved, and the time is saved.

The invention provides a method of determining a stage-discharge relation in a convergent area of main and branch waterways. The method comprises steps of estimating the jacking strength of convergent of main and branch waterways on a target section stage, determining a stage-discharge function relation under influences of jacking effects, solving a stage-discharge function equation under the influences of jacking effects, and drawing multivalued stage-discharge relation curve groups. Compared with the existing experience method, the stage-discharge relation in the convergent area of main and branch waterways deduced according to the river hydraulics principle has an explicit physical meaning, stage-discharge dimensional relations between main and branch waterways can be accurately reflected, the calculation precision is higher, stage-discharge relation curve groups in different main and branch waterway combinations can be given in a detailed and thorough mode, and a convenient way is provided for stage estimation for the convergent area of main and branch waterways.

The invention discloses an energy-saving control method for plug-in hybrid electric vehicles based on road traffic information. The energy-saving control method comprises the following steps: obtaining the real-time road grade, the traffic information of front vehicles, and the information of a next charging station from a global positioning system and an intelligent traffic system; establishing a three-dimensional mathematical model of a plug-in hybrid electric vehicle for predicting the vehicle state in the future; defining the power consumption of the vehicle and the optimum control problem of energy management at the power sustaining stage, and providing a functional equation for solving optimum control amount; performing real-time feedback optimum control, solving the optimum control amount so as to realize the optimum allocation for charging and discharging of a storage battery, adjusting and optimizing the energy flow of the plug-in hybrid electric vehicle on line according to the traveling conditions of the vehicle, obtaining the optimum performance of the system of the plug-in hybrid electric vehicle, and using a planetary gear mechanism as an continuously-variable electronic transmission, so that an engine works at an efficient region; predicting the charging state of the target storage battery according to the road traffic information, and adjusting the energy flow of the plug-in hybrid electric vehicle on line so as to achieve the purposes of saving energy and reducing exhaust.

The invention belongs to the technical field of measurement of roughness or irregularity of surfaces, and discloses a surface roughness measurement method based on auto-correlation value curvature characteristics of an interference image. The method comprises the following steps: removing the noise of the interference image by adopting Gaussian filter pretreatment; calculating second-order statistical characteristics of the image by applying an image auto-correlation analysis method, and extracting out a statistical characteristic curve of 10 pixel offsets in an i direction; concluding the curve change of the statistical characteristic curve under different roughness, and performing polynomial fitting to solve the maximum curvature of the curve; and fitting the functional relation betweenthe maximum curvature of the curve and the change of the roughness by applying a least square method so as to obtain a functional equation of which the curvature changes with the roughness. The methoddisclosed by the invention can measure high-reflection smooth surfaces which can not generate speckles and can measure rough surfaces which can not generate interference fringes, thereby expanding the measurement range; the method can effectively measure high-reflection surfaces with smaller roughness and rough surfaces with larger roughness, thereby expanding the measurement range; and the maximum value of the relative error of a measurement result is 8.6%.

The invention relates to a method and device for judging the stability of grid-connected sub-synchronous oscillation of a double-fed induction generator. The method comprises the steps of setting positive sequence harmonic voltage and positive sequence harmonic current of a stator of the double-fed induction generator according to stator phase-a voltage, stator phase-a current and rotor phase-a current of the double-fed induction generator; determining an output voltage expression of a current regulation link according to the positive sequence harmonic voltage and the positive sequence harmonic current; establishing a stator-side impedance analytical expression of the double-fed induction generator according to the positive sequence harmonic voltage, the positive sequence harmonic currentand the output voltage expression so as to obtain double-fed induction generator-side impedance characteristics; building a grid-side impedance model so as to obtain grid-side impedance characteristics; establishing a characteristic function equation of the double-fed induction generator-side impedance characteristics and the grid-side impedance characteristics; and solving the characteristic function equation so as to quantitatively judge the oscillation frequency and damping level of the double-fed induction generator. The oscillation frequency and damping level of can be quantitatively judged according to the method and device.

The invention discloses a room temperature shim coil of a superconducting NMR spectrometer. The room temperature shim coil is formed by a winding region, a support frame, and a base from top to bottom in sequence. The inner layer of the winding region is axial coils, and the outer layer is radial coils. On the manufacturing method for a coil structure, the multiple order axial coils are compressed on a same cylinder to improve space utilization rate, and a winding method of self-bonding enamelled wires and PET films (also named as high-temperature polyester film) is used, so as to solve technical problems of wire positioning and fixing in a winding process of the room temperature shim coil. An optimization function is also designed and established aimed at an algorithm of the shim coil, and a problem of design of the room temperature shim coil is converted to a problem of solving the minimum value of the optimization function, and then through a particle swarm optimization, the functional equation is solved, and finally position distribution and number of turns of the coils can be calculated. According to the room temperature shim coil, reliability and adaptability of the shim coil can be obviously improved.

The invention provides a technical scheme of a power distribution system equivalent ground distributed capacitor measuring method, comprising the following steps: 1) injecting signal current with constant frequency into a power distribution system; 2) measuring neutral current of primary side of an arc suppression coil and busbar voltage of a busbar in the power distribution system, filtering the measured neutral current and busbar voltage, and filtering out power frequency quality to obtain current injected into the power distribution system and pressure drop generated by the current on an equivalent ground distributed capacitor; 3) modeling the power distribution system based on a lumped parameter model, and calculating a complex functional equation of the injected signal current, voltage and power distribution system equivalent ground distributed capacitance according to Ohm's law; and 4) calculating equivalent ground distributed capacitance value of the power distribution system and ground insulation resistance value of the power distribution system according to the equation in the step 3). The method provided by the invention is clear in principle, is easy to implement, and has a wide application range and high accuracy in computation.

The invention belongs to the technical field of the power electronic technology and in particular relates to a parallel outer ring flow equalizing control method for DC / DC convertor for the DC converter parallel outer ring flow equalizing controller. The method includes: converting the parallel system into small signal transfer function model type; obtaining the inner ring voltage controller according to the small signal open-loop transfer function from the duty ratio to the output voltage; obtaining the voltage closed loop output impedance of the single module; obtaining the transfer functional equation of the flow equalizing controller; using the loop in the Hoo robust control for forming the theory, and setting the frequency domain weighting function different in function for forming the flow equalizing loop gain with ideal expectation characteristic. The design of the flow equalizing controller is converted to the robust loop forming problem for setting the frequency domain weighting function for processing, the ideal flow equalizing loop characteristic can be conveniently obtained, especially, better robust performance can be obtained at the condition that the parallel DC converter model difference and model parameter perturbation information are lacked.

The invention, which belongs to the technical field of automatic mining of the fully mechanized caving face, discloses an intelligent coal drawing method based on real-time monitoring of the top coalthickness variation. Top coal thickness information is monitored in real time by using an ultra-wideband radar and a laser three-dimensional scanning technology; and the opening or closing of the coaldrawing hole is adjusted automatically according to the top coal thickness variation. The method comprises the following steps: step one, calculating a top coal thickness variation by an ultra-wideband radar transmitting device based on a radar echo signal; step two, calculating a coal drawing amount by a laser three-dimensional scanning device according to scanned geometrical characteristics ofthe coal flow, the coal flow speed, and coal drawing time of the coal drawing hole; step three, establishing a functional equation between the top coal thickness variation and the coal drawing time and calculating optimal opening / closing time of the coal drawing hole at different coal drawing stages; and step four, sending out opening and closing instructions to the coal drawing hole by a coal drawing top-coal controller to realize intelligent coal drawing of the fully mechanized caving face. The method has the following advantages: the whole coal drawing process is controlled accurately; theadaptability is high and the manual intervention is avoided; the coal drawing efficiency is improved effectively; and the labor intensity is reduced.

The invention relates to a finite element simulation method capable of removing a microwave tube high-frequency circuit in a pseudo-DC mode, comprising the flowing steps: (A) using a zero electric displacement vector as an electric field constraint equation and acquiring an integrated form of the electric field constraint equation according to an boundary value problem of an electromagnetic field in a microwave tube high-frequency circuit and acquiring a functional equation of the boundary value problem of the electromagnetic field through a standard variation principle of a finite element method; and (B) necessarily ensuring the match between grids on a main surface and grids on a secondary surface on a peripheral boundary when solving a solution domain and considering a pseudo-periodic boundary condition by adopting a dissection solution domain of tetrahedral grids. The invention has the advantages that the finite element simulation method can greatly improve the accuracy, the efficiency and the robustness of microwave tube high-frequency system simulation.

The invention discloses a DEM-auxiliary-intensity-related glacier surface motion three-dimensional vector inversion method, and relates to the technical field of glacier motion monitoring. Lifting rail SAR image pairs are registered at high precision based on a high-precision DEM, and then displacement observation results in four different directions of a glacier area are obtained by utilizing satellite-borne SAR amplitude information and applying an offset tracking technology; and shadowing and layover areas with the threshold value lower than the offset tracking relativity threshold value are masked, according to the SAR image forming geometrical relationship, the functional equation system between the deformation observation results of the four directions and three-dimensional motion components of three glaciers in the east-west, south-north and vertical directions is built, the three-dimensional motion components are calculated, and then the actual flowing speed and direction of the surfaces of the glaciers are obtained. Actual three-dimensional motion on the surfaces of the glaciers is inversed only through satellite-borne SAR, when the terrain is steep and an SAR image spacebase line is large, the offset error is lowered, and the precision of the offset tracking result is improved.

The invention discloses a method for realizing the acceleration of anisotropic diffusion filtration of an overlarge synthetic aperture radar (SAR) image by a graphic processing unit (GPU), which solves the problem of low speed caused in the process of processing the overlarge SAR image by adopting anisotropic diffusion filtration. An anisotropic diffusion and filtration process adopts a compute unified device architecture (CUDA) and executes the following steps in the GPU: (1) copying image data I from a host memory of a computer to a memory area A of the GPU; (2) computing diffusion dimension data c(q) of the image data I by using an anisotropic diffusion dimension function; (3) computing the graphical data of an anisotropic diffusion filtration result according to an anisotropic diffusion dimension functional equation; and (4) circularly repeating step (2) and step (3) for T times to obtain the final anisotropic diffusion filtration result graphic IT, and copying the data IT in the memory area C to the host memory of the computer after iteration ends. In the invention, the acceleration is accomplished by the GPU parallel computation under the CUDA, compared with the central processing unit (CPU) serial computation, the processing speed is obviously improved, and the method can be applied to sites with high real-time processing requirements.

The invention provides a power grid annual power purchase plan formulation method and system based on the game theory. The method comprises the following steps of: collecting the annual total load electric quantity of a power grid, the generator set number of a total power grid, the grid purchase price of each generator set, the coal consumption of each generator set, the upper limit and the lower limit of the annual power generation amount of each generator set, the annual average load rate of each generator set and the upper limit of the difference of the annual average load rate of each generator set; setting the parameter of the annual power generation amount of each generator set; determining a target to be optimized in the power grid annual power purchase plan; building a sub-target functional equation and a constraint equation; converting the power purchase plan into a multi-target optimization problem; on the basis of the game theory, comprehensively optimizing each target to be optimized to obtain the weight value of each target functional equation; converting a multi-target optimization problem into a single-target optimization problem; and determining to obtain the power generation plan value of the annual power generation amount of each generator set. According to the power grid annual power purchase plan formulation method disclosed by the invention, a plurality of targets can be both considered while the power purchase plan is formulated, and meanwhile, a plurality of targets are optimized.

The invention relates to an intrinsic-analysis method for assigned frequency of a periodic structure, which includes steps: A, setting a functional equation which can consider boundary value of an electromagnetic field in conductors and medium loss periodic structure and can obtain the boundary value of the electromagnetic field by means of standard variation principle of the finite element method; B, solving solution domain by gridding subdivision of a tetrahedron and guaranteeing gridding matching of a primary side and a secondary side of the periodic boundary; C, selecting primary function, expanding electric-field intensity vector in all grids by the primary function and obtaining a matrix equation of expansion coefficient by Ritz method; D, writing the matrix equation of the expansion coefficient into a linear generalized intrinsic equation about propagation constant; E, solving the linear generalized intrinsic equation by setting a frequency, obtaining attenuation constant and phase constant corresponding to the set frequency according to the propagation constant, and obtaining interaction impedance through after treatment; and F, repeating the step of E to obtain attenuation constants, phase constants and interaction impedance corresponding to different frequency.

The invention discloses a general inverse kinematic solving method of a six-freedom-degree serial robot and belongs to the field of inverse kinematics of robots. The closed solving method simple in computational process and easy to achieve is proposed based on an exponential product model. The method mainly utilizes the basic property of a spinor theory, a Rodrigues rotation formula and a special geometric construction to convert a complicated inverse solution problem into a simple trigonometric functional equation for solution, so that six joint angles can be represented only by needing two expressions, and a form is simple and convenient to memorize. The method is wide in practical range, can be applied to any robot conforming to a Pieper principle and having intersected or parallel relation between two adjacent shafts in front three joints and promotes popularization and application of the robots, and the application process is simplified.

The invention discloses an eliminating method for Gibbs ring pseudoshadow when reconstructed MRI, and comprises steps as follows: (1) acquiring some data from K space; (2) transforming acquired vector array of original frequency domain; (3) filtrating and getting vector array data 2 of frequency domain from transmitted vector array data 1 of frequency domain; (4) taking edge detection using vector array data 2 of frequency domain; (5) rewriting the image function equation to imaging equation selected second type Chebyshev polynomial as basic functions; (6) by edge information from step (4), substituting vector array data 1 of frequency domain to imaging equation from step (5) in every continuous domain, obtaining needed image domain data array, then transmitting it to vector, resulting in corrected image. The invention can eliminate efficient Gibbs ring pseudoshadow, meantime ensures high resolution for reconstructed image.

The embodiment of the invention provides a chemical-mechanical grinding end-point detecting method based on a shallow trench isolation technology. The chemical-mechanical grinding end-point detecting method comprises the following steps of: establishing a function equation between the relative sliding speed of a wafer surface element and a grinding pad and the friction force born by the element; solving the functional equation, substituting the friction born by the element into the torque formula of the grinding pad, and establishing a functional relation among a wafer, the adjustable torque of the grinding pad and the mechanical parameter of a grinding driving device; and according to the difference of the friction coefficients caused by material differences, analyzing the influence of the surface characteristic of the grinding pad on the appearance change of the wafer, and recognizing the grinding end point by virtue of the functional relation between the adjustable torque of the grinding pad and the mechanical parameter of the grinding driving device, thus realizing the detection of the end point. According to the scheme provided by the invention, the torque change is analyzed by surveying the friction factor between the wafer and the grinding pad, the mutual action relation between the wafer and a random-fluctuation rough pad in the manufacture procedure of the grinding process is deeply analyzed, and the difference of the grinding removing rate caused by the material difference can be recognized directly by the torque of the grinding pad, so that the detection of the end point can be realized flexibly.

The invention discloses a multi-physical coupling method in transient calculation of a pressurized water reactor. The method comprises the following steps of 1, for a multi-group space-time neutron kinetics equation, obtaining neutron-flux density and delayed-neutron precursor density of a prediction step by using a fully-implicit backward difference method; 2, obtaining a shape function of the prediction step through a normalization condition, linearly interpolating the shape function in a medium time step, and by solving an amplitude function equation on a small time step, finally obtaining neutron-flux density on the medium time step, thereby obtaining power and delayed-neutron precursor density; and 3, performing fuel temperature calculation and coolant temperature calculation through the power in the medium time step, realizing feedback through a nuclear reaction cross section, and realizing coupling with neutronics calculation. When the neutronics calculation is performed by using an improved quasi-static method of prediction and correction, heat engineering calculation and delayed-neutron precursor density calculation are performed on the proper time step to realize efficient coupling.

The invention provides an automatic power adjusting method of a microprogramming technique 1327 (MPT1327) mobile hand-held station with electricity saving, longer standby time of terminal and excellent performance. The transmitting power of the MPT1327 mobile hand-held station is automatically adjusted according to a received signal strength indication (RSSI) value transmitted by a base station. Two RSSI threshold values are set for the MPT1327 mobile hand-held station, the transmitting power values corresponding to the two threshold values are respectively calculated, modeling is carried out according to the threshold values and the corresponding power values and by referencing an electric ware space transmission theory so as to establish a mathematical model. The MPT1327 mobile hand-held station detects the strength of signals transmitted by a base station at any time, and the transmitting power of the hand-held station is dynamically calculated according to a model function, thereby the automatic adjustment of the transmission power of the MPT1327 mobile hand-held station is realized. The invention can reduce electricity consumption, prolong the standby time of the terminal, has more excellent performance, improve the user experience of users and fully use the inherent technological advantages of the MPT1327 mobile hand-held station.

The invention discloses a time-varying convex quadratic programming solver design method. The method includes steps: establishing a time-varying quadratic programming standard model of a practical physical system through mathematical modeling; according to the Lagrange multiplier method, subjecting the quadratic programming standard model to optimal value solving to acquire optimal solution and Lagrange's multiplier related partial derivative information; converting the partial derivative information into a standard time-varying matrix form; designing an error functional equation according tothe standard time-varying matrix; according to the error functional equation and a power type varying-parameter recurrent neural dynamic method, designing a time-varying quadratic programming problempower type solver in a real number field through a monotone increasing odd activation function, so that a network state obtained through the time-varying quadratic programming problem power type solver is the optimal solution to the practical physical system or numerical solution time-varying quadratic programming problems. The time-varying convex quadratic programming solver design method has global convergence characteristics, errors can be converged to zero at a super-exponential speed, and accordingly calculation speed is greatly increased.

The invention provides a design method of a blade suction surface groove for controlling flow separation, which comprises the steps of: firstly, determining the initial position of the initial point of a groove, the initial projection length of the groove on the molded line of a blade suction surface and the initial function equation for describing the curved line of the groove; and secondly, optimizing the groove by adopting genetic algorithm, wherein the initial position of the groove, the projection length of the groove on the molded line of the blade suction surface and the initial function equation for describing the curved line of the groove are taken as optimization variables, and the static pressure rise, the total pressure recovery coefficient, the total pressure coefficient and the airflow turning angle of a blade grid passage are taken as variables of the fitness function in the genetic algorithm, so that the optimization variable combination which can be maximize the fitness function is taken as the modeling parameters of the groove. According to the design method, the flow in the blade grid passage can be improved, a burbling zone in the passage can be decreased or even eliminated, the efficiency and the total pressure recovery coefficient of the passage are improved, the loss coefficient of the passage is reduced, the high attack angle of the elementary blade type is improved, and the available attack angle range of the elementary blade type is enlarged.

The present invention discloses an action vocable layered automatization test system, which includes virtual programming interface module, command transmission module, at least one command module and outside programming interface module. Said virtual programming interface module locates bottom layer providing uniform interface for top level module calling automatization testing system functional equation; said command transmission module calling virtual interface module provided functional equation to proceed command transmitting and receiving; said command module calling specific function command packaging according to different version, transmitting command to tested equipment and obtaining equipment responding; said outside programming interface module located upper most layer for providing uniform interface for testing logic call instruction modular function. Said invention realizes separation of testing logic with AW and AW with automatization testing system environment through AW inner layering process, to reduce tested version and process variation influence to testing logic, reducing AW maintenance quantity, raising AW maintainability.