43 results about "Numerical methodology" patented technology

The invention relates to a wellbore multiphase flow model numerical solution and gas-liquid distribution state inversion method and system, and belongs to the technical field of petroleum engineering, and the method comprises the steps: 1, constructing and training a physical information neural network for drilling wellbore multiphase flow dynamic simulation and overflow gas distribution state inversion; determining input and output of the physical information neural network; determining a loss function of the physical information neural network; training a physical information neural network; 2, designing an adaptive optimization algorithm, optimizing the final solution precision and convergence speed of the physical information neural network, and obtaining an adaptive physical information neural network; designing an adaptive activation function; designing a self-adaptive sampling mechanism based on residual errors; 3, based on the self-adaptive physical information neural network, numerical solution and gas-liquid distribution state inversion of the shaft multiphase flow model are achieved. According to the method, the problem that a traditional numerical method usually needs high-precision grid division and a large number of computing resources is effectively solved.

The invention provides a numerical method for immiscible solution drop swallowing behavior in limited shear flow, and relates to the technical field of immiscible solution drop swallowing behavior, a simulation task based on a phase field model and a Navier-Stokes equation is established, a computational domain is discretized through a finite element method, a physical field is initialized, and in each time step, the physical field of the immiscible solution drop swallowing behavior in limited shear flow is calculated. Updated velocity field and phase field data are obtained through coupling solution, the interface curvature change rate and the flow field vorticity intensity are comprehensively calculated, a global index representing the dynamic intensity of the system is generated, the index is compared with a preset threshold value, and the dynamic intensity of the system is calculated. When the index is higher than a threshold value, a smaller step length and a higher grid resolution are adopted, otherwise, a larger step length and a lower resolution are adopted, finally, solver configuration is automatically reconstructed according to the selected parameters, solving is continued under a new adjusting strategy, a self-adaptive numerical simulation cycle is formed, and a new time step is obtained. And completing simulation and outputting complete field data of the liquid drop swallowing process.

The invention discloses a structural dynamic response method of a physical information radial basis Kolmogorov-Arnold network, and belongs to the technical field of anti-seismic engineering. The method comprises the following steps: acquiring a data supervision range of structural dynamic response under seismic excitation, and acquiring engineering dynamic parameters in the data supervision range in real time according to a timeline; according to the method, the mechanical preciseness of a traditional numerical method is inherited through kinetic equation constraint, and the efficiency and generalization bottlenecks of the traditional method are broken through by means of the self-adaptive learning capability of the RBF network; the analysis limitation of a traditional method under non-stationary seismic excitation is solved, dependence on artificial experience is reduced through an automatic process, and a novel structural dynamic response analysis normal form with precision, reliability and practicability is provided for the field of seismic engineering. And the technology upgrade of the field from empirical driving to data-physical dual driving is promoted.

The invention discloses a stable support analysis method for a tunnel face of a shield tunnel in a composite stratum, which relates to the technical field of support analysis, and comprises the following steps of: dividing the tunnel face and a front area thereof, classifying the areas with similar deformation tendency or instability tendency into the same kind, and setting differentiated support control targets for each area, the method comprises the following steps: constructing a balanced scheduling framework of support and settlement, extracting related risk characteristics before entering a risk working condition, inputting the related risk characteristics into a safety evaluation model, performing preventive correction on a tunnel face pressure regulation amplitude, a propulsion rate change slope and a grouting strategy according to an evaluation result, and in the whole analysis and control process, performing safety evaluation on the tunnel face pressure regulation amplitude and the propulsion rate change slope. And performing cross check by using a numerical method of different dimensions and field measurement, feeding back to the equilibrium scheduling framework when a difference is found, and re-correcting a support strategy. According to the analysis method, a scheduling framework is constructed through real-time data and partition classification, safety evaluation and cross checking are combined, the defects of composite stratum analysis are overcome, and the stability of the tunnel face and the construction safety are improved.

Embodiments perform neural network based behavior determination of physical objects. One such embodiment processes a three-dimensional (3D) numerical-method model representing a physical object to extract (i) 3D geometric data associated with the physical object and (ii) simulation data. The extracted 3D geometric data and simulation data are transformed into a 3D multi-graph. The 3D multi-graph is processed with one or more deep neural network (DNN) and one or more operators to determine behavior of the physical object.

The invention discloses a two-way rocket launching window searching method, system and device and a medium, and the method comprises the steps: rapidly calculating two nearest launching windows of an ascending orbit and a descending orbit of a rocket according to the attributes of a central body, target orbit plane parameters, the geographic coordinates of launching points and the flight duration of the rocket; according to the method, the problem is converted into numerical calculation based on space geometry, the crossing moment is directly solved by using the geometrical relationship between the normal vector of the orbital plane and the inertial position vector of the launching point location and adopting efficient numerical methods such as dichotomy and the like, and complex dynamic iteration is avoided. According to the method, the calculated amount is greatly reduced, rapid searching of the launching window is achieved, the method is suitable for an interstellar detection task scene needing rapid response or multiple times of calculation, and the real-time performance of task planning is remarkably improved.

The invention provides a numerical method and system for high-precision prediction of gas-driven piston motion in a powder dense-phase conveying process, and the method comprises the following steps: 1, taking a powder fuel supply system as a research object, setting an initial field of the research object, obtaining the physical properties and transportation conditions of a powder layer under the initial field of the research object, and calculating the physical properties of the powder layer; the initial static state of the piston is set as a starting point; 2, on the basis of a pre-constructed mesoscale solid phase corresponding force model, the total gas-solid resistance borne by the piston in the static state of the piston is obtained through calculation; 3, based on the gas-solid total resistance borne by the piston, the piston starting critical condition is set; 4, performing numerical simulation on piston motion based on a piston starting critical condition; according to the method, the actual stress environment of the piston is combined, solid phase corresponding force sudden change and the friction effect after powder deformation are comprehensively considered, the whole-process motion characteristic prediction of the piston from static state to dynamic stability has high credibility, and the problem of simulation distortion caused by inaccurate motion characteristics of a traditional model is solved.

The invention discloses a three-dimensional flow field prediction method based on deep learning, and the method constructs a novel deep learning network RKAN-3D through collaborative integration of a nonlinear learnable activation function module and an attention-enhanced residual convolution module in combination with an indirect prediction strategy, and achieves the precise modeling of a complex three-dimensional flow field. Wherein the non-linear learnable activation function module can adaptively capture multi-scale non-linear features in a three-dimensional flow field, and the expression ability of the model to a complex flow phenomenon is enhanced; the attention enhancement residual convolution module effectively improves the perception and modeling precision of the model on local flow characteristics in a three-dimensional space structure by focusing a key flow field area. In order to overcome the high-dimensional challenge of directly predicting the state of the whole flow field, an incremental learning mechanism is introduced, and the capturing capability and generalization performance of the model on complex flow characteristics are effectively improved by gradually evolving the dynamic process of the flow field. Experimental results show that the reasoning speed of the proposed network is improved by two orders of magnitude compared with that of a traditional numerical method, and the network shows excellent prediction precision and robustness in typical three-dimensional flow field prediction tasks such as spheres and cylinders.

The invention provides an airfoil profile streaming flow field rapid prediction method based on a Fourier neural operator, and belongs to the field of intelligent fluid mechanics. According to the method, machine learning and computational fluid mechanics are deeply fused, the future flow field state is rapidly predicted through a series of input and output data, and the calculation efficiency of the method is far higher than that of a traditional numerical method. According to the method, the calculation task of several hours or even several days of traditional computational fluid mechanics simulation can be shortened to the second level, the optimization iteration period of aircraft appearance design is greatly accelerated, and the dependence and cost of research and development on wind tunnel tests and high-performance calculation are remarkably reduced.

The invention discloses a method for analyzing the reliability and failure mechanism of advanced heterogeneous integrated hybrid bonding interconnection. The method comprises the following steps: (1) calculating the contact resistance of a bonding interface based on an advanced heterogeneous integrated hybrid bonding process condition; (2) taking the bonding interface contact resistance as the input of electromigration simulation, and simulating a vacancy evolution multi-physical process induced by electromigration in interconnection by using a numerical method; and (3) analyzing an interconnection failure position based on a simulation result. The provided method for analyzing the reliability and failure mechanism of the advanced heterogeneous integrated hybrid bonding interconnection can provide theoretical guidance for interconnection design optimization and future research on multi-physical processes in complex circuits and systems.

The present application relates to the field of intelligent control and optimization, and discloses a kind of energy consumption dynamic management method, comprising the following steps: S1, the mathematical model of energy system is constructed, and the state space and control space of energy system are defined;S2, and the stability of energy system state is analyzed by Lyapunov function;S3, the optimal control law of system is obtained by solving HJB equation;S4, design Kalman type state estimator;S5, model parameters are converted between micro and macro levels, for the consistency of energy and constraint conditions between scales;S6, the HJB equation is discretized and solved using numerical method;S7, the performance of the system is verified by simulation and experiment.The effect of automatic avoidance of high electricity price operation interval, budget overrun warning and adaptive cost optimization is achieved by constructing stochastic dynamics energy consumption model, HJB optimal control solution and cross-scale energy consumption mapping, dynamic prediction and optimal scheduling control of energy consumption behavior under different loads, different electricity prices and different budget upper limits.

The invention relates to a pipe network division method and system for urban infrastructure simulation, and the method comprises the steps: S1, building a pipe network structure model for a target region, and enabling the pipe network structure model to comprise the topological structure information of pipe network segments and the attribute information of each pipe network segment; wherein the pipe network section between any two connection nodes is regarded as a to-be-dispersed pipe network section, and S2, performing grid division on each to-be-dispersed pipe network section in the pipe network structure model until a division cut-off condition is met. According to the method, on the basis of ensuring the solving precision and convergence, the human experience dependence and unnecessary modeling debugging time are remarkably reduced, and a scientific and rapid spatial discrete scheme is provided for engineering tasks such as urban pipe network digital twinning, transmission and distribution system simulation optimization and operation condition prediction; the method is suitable for fluid transmission and distribution process simulation based on numerical methods such as a finite volume method and a finite difference method.

The application discloses a method for obtaining C / C composite ablation morphology based on physical information neural network, and relates to the field of aerospace, comprising the following steps: taking the coordinate data of a C / C composite three-dimensional space model under a Cartesian coordinate system, an ablation time, a dimensionless parameter Sh and a dimensionless parameter A as inputs of a pre-trained multilayer perceptron network, outputting oxygen molar concentration and solid volume fraction through the pre-trained multilayer perceptron network; solving a gas phase diffusion equation and a solid phase evolution equation through the oxygen molar concentration and the solid volume fraction output by the multilayer perceptron network to obtain the C / C composite ablation morphology. The method can efficiently and accurately solve the gas-solid coupling equation, overcome the limitations of traditional numerical methods, accurately obtain the ablation morphology of the C / C composite under the gas-solid coupling effect at a low calculation cost, and provide strong theoretical support and technical support for the application of the C / C composite in the field of aerospace and the like.

The application discloses a transient multiphase flow numerical simulation method of a high-temperature structure water-entry boiling mechanism, which designs a motion-flow-heat transfer-phase change synchronous calculation method for the high-temperature structure, realizes real-time solving of a flow field and a cavity shape in a high-temperature phase change process through joint solving of a motion equation, a multiphase flow equation and an energy equation, designs a boiling state determination method based on wall surface superheat, simultaneously calculates the flow, the high-temperature phase change, the heat transfer and the rigid body motion by using the same set of numerical methods, makes the cavity shape, the drag coefficient and the heat flux consistent in the physical mechanism and the numerical results, and can accurately reflect the influence of the steam film on the cavity shape, the water-entry motion and the drag characteristics by analyzing the quantitative correlation between the steam film and the stress and motion response of the structure, thereby improving the reliability of water-entry cavity evolution, water-entry trajectory and drag reduction prediction results, and providing a theoretical basis for the research on the material and motion control of the high-temperature structure.

The invention discloses a numerical method for predicting aerodynamic force of a large-span bridge through a vortex structure around a flow field, and belongs to the field of bridge wind engineering. The method comprises the following steps of: simulating and generating a velocity field and vortex structure information around a flow field by utilizing computational fluid mechanics; determining geometric boundary conditions based on a vortex force diagram method, constructing and solving a hypothetical potential equation, and calculating a vortex force vector; finally, the additional mass force, the vortex pressure, the viscosity pressure and the friction force are calculated through integration, and the total aerodynamic force is synthesized. The method combines flow field information and aerodynamic force, can analyze contribution of different vortex structures, is suitable for a single box girder bridge and a multi-box girder bridge, has the advantages of being accurate in result, high in efficiency and small in resource occupation, can be used for flow separation aerodynamic force mechanism analysis, and solves the problem that a traditional pressure integral method cannot analyze the influence of a flow field.

This invention relates to the field of space debris environment modeling technology, specifically to a short-term evolution method for continuum disintegration debris based on numerical methods. The method includes: Step 1, establishing an initial distribution model of disintegration debris, including a disintegration debris velocity increment model and a numerical model of the initial distribution of disintegration debris; Step 2, sampling feature points of the continuum using a gradient-based sampling method, selecting the top N samples with the highest L2 norm scores of the gradient vectors as feature points of the continuum to be evolved; Step 3, processing the continuum equations using numerical solutions, considering the six orbital roots and four perturbation factors to perform short-term evolution of feature points; Step 4, reconstructing the final state distribution of disintegration debris using a Student's t-mix model. This invention fills the gap in previous research in this field, which only focused on ultra-short-term and long-term evolution, and has the beneficial technical effects of requiring fewer samples, having lower computational costs, shorter computation time, higher result accuracy, and stronger robustness.

The application provides a wharf rock-embedded pile critical pile length calculation method and system, relates to the critical pile length calculation technical field, through the deep integration and data calibration of numerical simulation and physical test, a high-fidelity comprehensive parameter database is constructed, and a continuous mapping relationship of the critical length-diameter ratio about the base rock strength and the inclination is established. First, numerical analysis is carried out under multiple working conditions, then physical model test is carried out under representative working conditions, the test data are used to construct a correction model to calibrate all simulation results, so that the accuracy of the end resistance ratio prediction is significantly improved. The application effectively solves the dual limitations that the pure numerical method lacks entity verification and the pure theoretical method is difficult to simulate the complex pile-rock contact behavior, through systematically diffusing the discrete test information into the comprehensive simulation data, the stress mechanism of the inclined rock-embedded pile is accurately described, and it is ensured that the mapping relationship established subsequently can truly reflect the physical law under the multi-factor coupling.

The present application relates to the technical field of numerical simulation of hydraulic engineering, and particularly relates to a dam-break simulation method and system based on Pinn, which comprises the following steps: constructing a data set and training a first prediction model by numerically simulating different river channels, constructing an error matrix based on the model prediction results and fitting the correlation between time and space and error, using the relationship to construct an error-weighted quadratic training loss function to optimize and train the model, and finally screening the best dam-break simulation model to achieve accurate prediction of dam-break water flow and water depth in the target scene. The present application combines physical information neural network with numerical simulation data, physical constraints and error-weighted quadratic training, which not only gets rid of the grid dependence of traditional numerical methods, improves the calculation efficiency and scene adaptability, but also solves the problem of physical deviation caused by data scarcity, significantly improving the prediction accuracy and reliability under extreme dam-break working conditions.

The invention discloses a subsurface erosion simulation numerical method based on a semi-implicit two-phase substance point method, and relates to the technical field of geotechnical engineering numerical simulation, and the method comprises the following steps: generalizing unsaturated easily subsurface-eroded accumulated soil into a two-phase multi-component medium based on a mixture theory; establishing a solid-liquid double-phase double-layer grid system; establishing a core control equation of seepage subsurface erosion in the soil body; transferring the material point information of the solid phase and the liquid phase to corresponding background grid nodes in the double-layer grid system; applying a boundary condition on a background grid of the double-layer grid system, and carrying out semi-implicit solution on the core control equation; mapping the obtained node speed to a substance point; judging whether the relative velocity of the solid phase and the liquid phase exceeds a seepage velocity threshold value or not; state variable information carried by the two-phase substance points is updated; correcting permeability and saturation; and judging and calculating whether the specified time is up or whether the soil body reaches the yield limit. The method provided by the invention has important theoretical and engineering practical values for solving the difficulty in simulating the subsurface erosion disaster of the accumulated soil.

The present invention provides a method for analyzing the horizontal displacement of free-end pile foundations in a dual-parameter foundation model. The boundary conditions of pile foundations such as friction piles and isolation piles can be simplified to free-end pile foundations. Combined with the Fourier series method, a method for analyzing the horizontal displacement of free-end pile foundations in a dual-parameter foundation model is proposed. The method is used to calculate the horizontal displacement and internal force of pile foundations caused by shield construction adjacent to pile foundations. Compared with the numerical simulation method, the physical quantities required for calculation can be obtained by substituting the corresponding calculation parameters. The calculation process is simple. Compared with the theoretical method using the numerical method, the method only needs to solve the horizontal displacement of the pile end. w O 、 w L The related two-variable linear equation can be used to further obtain other unknown coefficients, which has high calculation efficiency and provides an effective theoretical method for the safety risk assessment of existing structures in shield construction adjacent to pile foundations.

This invention discloses a numerical method for predicting aerodynamic forces of long-span bridges based on the vortex structure of the flow field, belonging to the field of bridge wind engineering. The method first uses computational fluid dynamics to simulate and generate the velocity field and vortex structure information of the flow field; then, based on the vortex diagram method, it determines the geometric boundary conditions, constructs and solves the assumed potential equation, and calculates the vortex vector; finally, it calculates the additional mass force, vortex pressure, viscous pressure, and frictional force through integration, synthesizing the total aerodynamic force. Combining flow field information and aerodynamic forces, it can analyze the contribution of different vortex structures, is applicable to both single-box girder and multi-box girder bridges, and features accurate results, high efficiency, and low resource consumption. It can be used for the analysis of flow separation aerodynamic mechanisms, solving the problem that traditional pressure integration methods cannot analyze the influence of the flow field.

The invention provides a numerical method for transient thermal simulation of a core particle integration system, and the method comprises the steps: decomposing the whole system into a plurality of modules according to the function and structural characteristics of the core particle integration system; converting a time domain heat conduction control equation in each module into a Laguerre domain; extracting heat transfer characteristics of each module based on a Laguerre domain heat conduction equation, and establishing a corresponding Laguerre macro model; taking the Laguerre macro model as an equivalent boundary condition, and solving the transient thermal response of each module in an order stepping mode by adopting a finite volume method. According to the method, on the premise that the calculation precision is guaranteed, the calculation complexity of the space domain and the time domain is reduced at the same time through the Laguerre macro model, the simulation efficiency is remarkably improved, and calculation resource consumption is reduced.

This invention belongs to the field of rotating machinery technology, specifically relating to a method for solving the dynamic characteristics of a seal-rotor and seal-rotor bearing system. The method includes: 1. Obtaining four rotor dynamic coefficients at different speeds and whirl frequencies using steady-state or transient computational fluid dynamics methods; 2. Establishing a database of seal rotor dynamic coefficients based on two-dimensional cubic polynomial interpolation; 3. Setting the initial speed for the solution and specifying the initial operating condition; 4. Calculating the instantaneous whirl frequency of the rotor, finding the current seal rotor dynamic coefficients, and calculating the seal excitation force on the rotor; 5. Establishing the dimensionless dynamic model to be solved; 6. Substituting the seal excitation force obtained in step S4 into the dynamic equation established in step S5, performing numerical calculations, and determining whether the numerical solution converges. If not, returning to step S4; 7. Retaining the result when the calculation converges and determining whether the current speed is the upper limit of the desired speed.

The application discloses a kind of plate pile embedded stability evaluation method of foundation pit considering unsaturated effect.Model test and numerical simulation results show that the failure sliding surface of foundation pit is the combination of curve and straight line, and the combination is not yet applied to the research of unsaturated foundation pit stability at present.Based on the upper limit principle of limit analysis, a kind of horizontal piece half analytical analysis method is proposed, the external power done by unsaturated soil gravity is calculated using this method, and based on the maximum energy consumption principle, the explicit semi-analytical solution of the moment of the lowest way of internal support point by active and passive earth pressure is obtained, and then the stability safety factor of plate pile embedded foundation pit is obtained.The semi-analytical analysis method constructed by the application has the advantages of analytical method and numerical method, effectively combines the unsaturated soil strength theory with limit analysis method, can reasonably explain the strengthening mechanism of suction effect, and has certain practical guiding significance for guiding the design and excavation of foundation pit under complex conditions.

A computer and machine-implemented method based on integrated numerical / machine learning for general-purpose simulation of real-world systems. The purpose is to allow machines to carry out simulations and analyze the surrounding environment for advanced monitoring and control tasks. The processors available in the devices are limited in their capabilities and they cannot perform complicated calculations. In addition, the boundary conditions of the real-world systems are in a continuous change, which makes the simulation task very difficult for machines. Moreover, validating the simulation is a human task. The method is set to overcome the challenges by compromising a numerical method with two- round machine learning. The first round helps to simplify the numerical model, whereas the second round allows the device to adjust the numerical algorithm in responding to any change in the boundary conditions. Additionally, the method can be used as computer software to reduce the computational and validation effort.

This invention belongs to the field of stress corrosion cracking prediction and simulation technology. It provides an efficient method for predicting the entire stress corrosion cracking process based on an evolutionary mass transport model and a stress distribution model derived from Irwin's small-scale yield theory. The method includes the following steps: after completing the theoretical derivation and modeling in the first two parts, a MATLAB program based on the mathematical model is designed. The parameters are then verified using a dimensionless method while referencing experimentally obtained data points. Finally, the goal of predicting the entire stress corrosion cracking process is achieved. The method provided by this invention fully leverages the efficiency of numerical methods, effectively solving the problems of excessive computational resources consumed by detailed finite element modeling and analysis, and inaccurate predictions caused by pure data fitting neglecting the actual physical process. It provides an important method for estimating crack initiation time in practical engineering. This method is highly versatile, easy to master, and easy to implement.

The invention discloses a flow field prediction method based on a Lagrange operator network. The method comprises the following steps: constructing a Lagrangian operator network, obtaining and preprocessing flow field data at continuous moments, and training the network by taking the previous flow field data as input and the latter flow field data as output to obtain a flow field prediction neural network; and performing dimensionless processing on the flow field data at the current moment in the to-be-predicted area, inputting the processed flow field data into the flow field prediction neural network, and outputting predicted flow field data at the next moment after processing to realize flow field prediction. According to the method, physical knowledge is embedded into a neural network architecture, the strict dependence of a traditional numerical method on a small time step is broken, the precision equivalent to that of direct numerical simulation is achieved under a coarse-resolution space-time grid and a large time step, the efficient processing capacity of the space-time dimension and low data dependence are achieved, and the method is suitable for large-scale popularization and application. The method can maintain high precision and stability in a non-seen scene, has a strong generalization capability, and finally can predict the flow field more accurately and efficiently.

A high-performance simulation method for heat conduction and thermal stress of embankment in non-homogeneous half-space. Based on the principle of geometric similarity transformation, the method constructs a scaling boundary modeling strategy, introduces a scaling line coordinate system, realizes the spatial mapping of Cartesian coordinates to scaling coordinates, and uses the Jacobian matrix to complete the coordinate transformation of differential operators, thereby accurately representing the non-uniform characteristics of the medium material. The Galerkin method is used to establish the control equation, and the semi-analytical solution technique is used to obtain the temperature field distribution; then the fourth-order Runge-Kutta algorithm is used for reverse iteration to further calculate the thermal stress field caused by the temperature field. The invention can effectively solve the problems of high calculation cost, complex boundary modeling and poor interface continuity of existing numerical methods in non-homogeneous media. It has high computational efficiency, excellent numerical stability and high precision characteristics, and is suitable for thermal-mechanical coupling analysis of complex engineering structures such as multi-layer foundation, embankment-water-foundation coupling system, etc.

The invention provides a satellite material surface charging potential calculation method based on a Bayesian PINN, and relates to the technical field of aircraft application, and the calculation method comprises the steps: constructing a physical information neural network, training a Bayesian physical information neural network model through the fusion of electrostatic field equation constraint and a Bayesian inference framework, and calculating the charging potential of a satellite material surface. Obtaining a trained Bayesian physical information neural network model; acquiring space coordinate data of the surface of the target satellite material; and inputting the space coordinate data into the trained Bayesian physical information neural network model to obtain a potential distribution prediction result of the surface of the target satellite material and a corresponding confidence interval. According to the method, potential distribution and confidence intervals can be rapidly obtained only by inputting space coordinates for new material surface geometrical shapes or boundary conditions, and the tedious process that a traditional numerical method needs to conduct modeling again and large-scale calculation for each new problem is avoided.

The application discloses a kind of composite stratum shield tunnel face stability support analysis methods, it is related to supporting analysis technical field, and the regional division is carried out to face and its front area, the region with similar deformation trend or instability tendency is classified into the same kind, and the different supporting control target is set for each region, the balanced scheduling framework of support and settlement is constructed, before entering risk condition, relevant risk characteristics are extracted, and input safety evaluation model, according to the evaluation result, the preventive correction of face pressure regulation amplitude, advancing rate change slope and grouting strategy is carried out, in the whole analysis and control process, cross-checking is carried out using different dimensional numerical method and field measurement, when difference is found, feedback to balanced scheduling framework, and support strategy is revised again.The analysis method constructs scheduling framework by real-time data and partition classification, combined with safety evaluation and cross-checking, overcomes the deficiency of composite stratum analysis, improves the stability of face and construction safety.