103 results about "Back analysis" patented technology

A method and a system are presented for determining the observability of faults in an electronic circuit. In the method, for each element the time periods are determined in which an occurrent fault could cause a deviation in analysis output signals.

The invention discloses a system for temperature control and crack prevention intelligent monitoring of a concrete dam. The system for the temperature control and crack prevention intelligent monitoring of the concrete dam is capable of quickening a construction process of a concrete dam and satisfying complex construction and environment conditions. The system for the temperature control and crack prevention intelligent monitoring of the concrete dam comprises a concrete temperature control information collecting unit, a concrete temperature control information control and temperature control implementation evaluating unit, a concrete temperature stress simulated analysis and back analysis unit, a concrete cracking risk real-time early-warning unit and a concrete temperature control and crack prevention feedback and automatic interfering unit, wherein the concrete temperature control information collecting unit, the concrete temperature control information control and temperature control implementation evaluating unit, the concrete temperature stress simulated analysis and back analysis unit, the concrete cracking risk real-time early-warning unit and the concrete temperature control and crack prevention feedback and automatic interfering unit are connected in sequence. Temperature control monitoring data information is collected in real time through the concrete temperature control information collecting unit, the temperature control monitoring data information is stored, controlled, shared and evaluated by the concrete temperature control information control and temperature control implementation evaluating unit, calculation is carried out by the concrete temperature stress simulated analysis and back analysis unit, concrete cracking risk real-time early-warning information is combined, and on-spot temperature control measures are interfered, revised and adjusted through the concrete temperature control and crack prevention feedback and automatic interfering unit. The invention further provides a method using the system.

The invention relates to a method for water inrush prediction and seepage control for underwater-tunnel broken surrounding rocks. The method includes the steps of S1), exploring by adopting a geophysical exploration and advanced-level geological drilling method and performing tests; S2), establishing a saltation prediction model of analytic hierarchy grey correlation of water inrush of the surrounding rocks by adopting an analytic hierarchy grey correlation method; S3), establishing a three-dimensional porous continuous medium fluid-structure coupled finite element model of the underwater-tunnel broken surrounding rocks by adopting an orthogonal back-analysis method; S4), performing dynamic prediction and seepage control on water inflow of the broken surrounding rocks and performing intelligent fuzzy logic control and instability early-warning forecast on high-pressure water inrush of the broken surrounding rocks; S5), adopting comprehensive prevention and control measures. Compared with the prior art, the method has the advantages that instability of water inrush of the underwater-tunnel broken surrounding rocks under high water pressure can be predicted and subjected to economical, reasonable, safe and reliable comprehensive seepage control.

The invention belongs to the technical field of geology prediction, and relates to a method for detecting influence of underground construction on a structure of a ground building. The method comprises the following steps: firstly detecting a ground subsidence value and the vibration of a concrete construction during blast, determining surface movement parameters by using a random medium theory back analysis method, adopting a population-level-based adaptive genetic algorithm, and using a back analysis calculation program prepared by Matlab to perform back analysis of the surface movement parameters; calculating the ground subsidence value by a random medium theory according to the obtained surface movement parameters; and building a concrete construction finite element model after detecting the concrete construction and component sizes on site, then calculating concrete construction stress of the ground building under the action of subsidence and blast, and lastly extracting the stress and performing quantitative damage prediction on the ground building structure. The process principle of the design is scientific and reliable, the data acquisition is simple and accurate, the calculation analysis is good in reliability and high in precision, the prediction result is accurate, the safety is good, and environment protection is realized.

The invention provides a multi-means, dynamic and whole-process landslide prewarning method. The method includes: performing macroscopic observation, engineering exploration data monitoring and displacement monitoring on landslide; performing macro damage phenomenon analysis and influence factor analysis according to a result of macro observation; building a numerical value model and performing parameter indicator determination according to a result of engineering exploration data monitoring; performing displacement back analysis and displacement tendency analysis according to a result of displacement monitoring; building a correcting numerical value model according to influence factor analysis and the built numerical value model; determining calculating parameters through parameter indicator determination and displacement back analysis; performing numerical value analog calculation analysis according to the correcting numerical value calculating model and the calculating parameters; calculating a displacement-time curve under a series of different stability safe coefficients according to numerical value analog calculation analysis; drawing a monitoring displacement-time curve according to displacement tendency analysis; comparing the above two curves to determine stability state of the landslide. By the method, landslide disaster recognition, prewarning and control level can be improved effectively.

The invention discloses a near-G-series high-speed train ground settlement prediction method based on a three-dimensional geologic model and back analysis. The method comprises the following steps of S1, obtaining geological survey data, the underground water level and ground settlement monitoring data in a study area; S2, constructing a regional three-dimensional geological model; S3, calculating the regional underground water level change; S4, constructing a regional ground settlement model; S5, conducting back analysis calculation on stratum mechanical parameters; S6, conducting prediction calculation on the regional ground settlement amount. According to the along-G-series high-speed train ground settlement prediction method, the geological survey data, the underground water level and the ground settlement monitoring data of G-series high-speed train engineering are utilized, and with the combination of back analysis, a three-dimensional ground settlement prediction model which can consider the uneven spatial distribution of the soil layer thickness and the soil body mechanical parameters is constructed. The along-G-series high-speed train ground settlement prediction method has high prediction precision, and can be widely applied to prediction of the ground settlement amount caused by the change of the water level along the G-series high-speed trains.

The invention discloses a method for time alignment between interference SAR echo data and platform motion and gesture data. The method comprises the steps of selecting the interference SAR echo data corresponding to a flat ground scene, acquiring the platform motion and gesture data corresponding to the interference SAR echo data and measured by a platform inertial navigation / GPS device, compensating the interference SAR echo data by means of the platform motion and gesture data when a roll angle is 0 degree to further obtain a scene digital elevation model through back analysis, acquiring an elevation azimuth wave function from the scene digital elevation model, acquiring the changing function of the roll angle from the platform motion and gesture data, solving a time value corresponding to the most value of a cross-correlation function of the elevation azimuth wave function and the changing function of the roll angle and taking the time value as time displacement quantity, and carrying out time displacement on the platform motion and gesture data according to the time displacement quantity and achieving the time alignment between the interference SAR echo data and the platform motion and gesture data.

The invention discloses a comprehensive brittleness evaluation method of supracrustal rocks of gas storage, and belongs to the technical field of rock brittleness evaluation. Firstly, the brittleness of a collected rock sample is evaluated on the basis of brittleness evaluation methods respectively based on a stress-strain curve, hardness and mineral composition; then the weighted value of the three brittleness evaluation methods is calculated by means of a weight back analysis method based on optimization theory; finally, comprehensive evaluation on the brittleness of the supracrustal rocks of gas storage is carried out. The method effectively solves a problem of limitation of single brittleness evaluation method in the prior art and has excellent practicability and high evaluation reliability. The collected rock sample has multiple applications, so that test cost is low.

A highly convenient design support method and design support system for a heat convection field or a mass diffusion field which significantly reduce the number of times of numerical simulation required to examine the designing parameters for achieving the design purpose. The design support method includes a forward analysis step of analyzing the heat convection field or the mass diffusion field by solving an equation of the heat convection field or the mass diffusion field based on an initially set value of a designing parameter, an inverse analysis step of analyzing a sensitivity defined by a change ratio of the design purpose to a designing parameter change by solving an adjoint equation corresponding to the design purpose based on the set design purpose, and a sensitivity display step of displaying information on the sensitivity analyzed by inverse analysis step as a graphic image on the display device.

The invention discloses a method for constructing a light gauge welding residual stress field on the basis of small-hole process tested data. The method comprises the following steps of: firstly, testing a residual stress of a weld assembly by adopting a small hole process, carrying out interpolation fitting on the tested stress values, then sequentially substituting each first-order polynomial in a finite element model by using a high-order Chebyshev polynomial as a primary function of representing unknown intrinsic strain causing the welding residual stress, carrying out calculation of elasticity to obtain calculated stress values of coordinate points with same tested stresses, selecting a coefficient enabling the quadratic sum of differences of the calculated stress values and the fitted tested stress values to be minimum as an intrinsic strain coefficient, multiplying the intrinsic strain coefficient by corresponding all order polynomials and then linearly overlapping to be used as an intrinsic strain, and finally obtaining a welding residual stress field through a finite element process. On the basis of the small hole process tested data, the intrinsic strain causing the welding residual stress is constructed by using a thermo-elasticity finite element method and a reverse analysis method, therefore, the distribution of the light gauge welding residual stress field is obtained; and the method is convenient to operate and high in efficiency.

The invention discloses a rock mass mechanical parameter back analysis method, and relates to a parameter back analysis method. The method comprises the following steps: S1, determining rock mechanical parameters to be inversed, and constructing a calculation scheme of a training sample on the basis of a uniform test design method; S2, performing numerical calculation on each constructed scheme toobtain a valley amplitude deformation value corresponding to each scheme, and forming an input and output value of an SDCS-LSSVM algorithm by the calculation scheme and the corresponding valley amplitude deformation calculation value; S3, based on the input and output samples obtained in the step S2, establishing a nonlinear mapping relationship between rock mechanics parameters and valley amplitude deformation values by learning input and output sample data; and S4, solving an optimal solution of the target function, and determining an optimal mechanical parameter combination. According to the method, a rock-soil body material parameter back analysis regression model is constructed, an improved cuckoo algorithm is adopted to search inversion parameters most matched with actual measurement displacement from the global space, and the prediction effect of the model is improved.

The invention relates to an indicator system generation method based on an entropy weight method. Based on the safety management indicator system of the chemical enterprise, the method researches reasonable improvement and updating on the basis of the original indicator system to construct a new indicator system. The basis of improvement is that the historical chemical accident cases act as the analysis data according to the irrationality of the data reverse analysis indicator weight. The entropy weight method is not the algorithm of a large number of sample data assignment, the entropy weightmethod is adopted to evaluate the sample quality to form the sample weight, the weight of the safety indicator system is reversely analyzed and calculated on the basis of the data and the indicator weight is appropriately adjusted and updated. Reasonable and necessary modification is made according to the actual situation so that continuous improvement of enterprise safety supervision and safetyevaluation can be promoted and the risk avoidance can be better performed.

The invention relates to a spring-back analysis and full-profile compensation method for a cold-stamped part of an automobile fender and belongs to the field of manufacturing of stamping molds for automobile covering parts. The method comprises specific steps as follows: step 1: a spring-back result of the part of the fender is predicted as follows: analog simulation is performed on the forming process of the part of the fender by the aid of finite element analysis software AutoForm, and the spring-back result is predicted; step 2: a spring-back compensation scheme for the part of the fender is determined; step 3: full-profile spring-back compensation for the part of the fender is performed, and multiple rounds of iterative compensation is performed by the aid of the finite element analysis software AutoForm until the size requirements are met; step 4: product data of the part of the fender are reconstructed. The method has the positive effects that the defects of low fender size precision and high mold transformation workload in the past are overcome by the aid of fender spring-back analysis and spring-back compensation based on the finite element analysis technology, the qualityof a fender product is substantially improved, the mold debugging cost is reduced, and the mold debugging period is shortened.

The invention relates to remote three-dimensional digital alarm method and system of the deformation stability of a primary tunnel supporting body. In the invention, a three-dimensional strain numerical model of a tunnel excavation supporting body in a construction area is established on the basis of tunnel construction area geological survey data and a FLAC (Free Lossless Audio Codec) 3D technology; an intelligent displacement back analysis model is established on the basis of a differential evolution method, and a three-dimensional visualized safe alarm platform is established on the basis of a VTK (Visualization Toolkit) business software system; the real-time deformation monitoring data of the tunnel supporting body is acquired by a full-section laser scanner arranged between two lining areas and transmitted to a remote user computer; obtained actual measurement displacement data is subjected to the displacement forward and back analysis of the differential evolution method by the three-dimensional strain numerical model and the intelligent displacement back analysis model of the tunnel excavation supporting body and displayed through the three-dimensional visualized safe alarm platform; the grading alarm of the deformation stability of a tunnel supporting structure is carried out by analyzing an actual measurement displacement value and the deformation threshold value of the primary tunnel supporting body, which is obtained through the displacement forward and back analysis, and alarm information is sent to a mobile phone of a related person in a mobile phone text message way so that the alarm of the deformation stability of the primary tunnel supporting body is completed, and therefore, the construction safety is enhanced.

The invention discloses a back analysis identification method for parameters of a dynamic re-crystallizing model. Based on a genetic neural network cellular automatic machine, the method is characterized in that the method comprises the following steps of: 1) confirming a value range of parameters and designing and establishing a test scheme; 2) performing analog computation and establishing a learning sample library; 3) establishing a BP (Back Propagation) neural network; 4) training the BP neural network and establishing a nonlinear mapping relation; 5) randomly selecting at least one set of parameters and forecasting a flow stress; 6) comparing a precast value with a really measured value till an error meets a design demand; and 7) storing an optimum parameter value with the error meeting the design demand. Haphazardry of a dynamic re-crystallizing parameter value is overcome; the dynamic re-crystallizing parameter value is calculated by utilizing an orthogonal test, a genetic algorithm, a neural network and numerical simulation optimization; the problems of instability, strong nonlinearity, large calculating quantity and uneven learning sample during a back analysis optimizingcalculation are solved; and the accuracy and the rapidity of re-crystallizing parameter value are obviously promoted.

The invention establishes a construction area tunnel excavation three-dimensional strain numerical model based on tunnel construction areal geology reconnaissance documents and an FLAC3D (Fast Lagrangian Analysis of Continua in 3-D) technology, establishes an intelligent displacement back analysis model based on a difference evolution method and establishes a three-dimensional visualization safety early warming platform based on a VTK (Visualization Tool Kit) commercial software system. The method comprises the following steps: acquiring real-time monitoring data of tunnel initial convergence displacement; transmitting the real-time monitoring data to a user computer; analyzing the obtained real-time monitoring displacement data by using the tunnel excavation three-dimensional strain numerical model and the intelligent displacement back analysis model; carry out displacement normal and back analysis by using the difference evolution method; displaying through the visualization safety early warming platform; analyzing and comparing real-time monitoring displacement values with a surrounding rock convergence deformation threshold obtained from the normal and back analysis; and subsequently carrying out graded early warming about the deformation stability of a tunnel supporting structure, wherein the early warming information is transmitted to cell phones of relevant on-site construction and ground management personnel through a short message module connected with a computer through a cell phone short message, so as to accomplish the remote three-dimensional digital safety early warming for tunnel construction landslides.

The invention discloses a temperature and stress analysis and back analysis method used for crack control of a concrete dam. The temperature and stress analysis and back analysis method includes that a design of a three-layered B/S structure is adopted, wherein a first layer is a presentation layer, namely a data acquisition layer, a second layer is an application layer comprising a WEB server and an application server as well as implementation of the middleware technology, a third layer is a data layer used for storing and managing various information by adopting an SQLserver database; the structure comprises a data acquisition and analysis module, a simulation calculation module, a back analysis module and a post-processing analysis warning module. Through the three-layered structure and the four modules, temperature and stress changes of concrete during construction can be tracked and analyzed in real time, so that the current temperature and stress can be timely known and future temperature and stress can be predicted.

The invention discloses a slope back analysis method based on anchoring load monitoring data, which starts with the anchoring load monitoring data to establish the relations among monitoring data and equivalent mechanical parameters of a rock-soil body as well as the variation trend of slop stability coefficients according to an objective functional expression. The slope back analysis has the technical effect that the mapping relation between the equivalent mechanical parameters of the rock-soil body and the anchoring load monitoring data is established through an objective function to calculate the equivalent mechanical parameters of the rock-soil body where an anchoring body is positioned, and further deduce the variation trend of the slope stability coefficients.