43 results about "Linear problem" patented technology

The invention discloses a circulating fluidized bed temperature intelligent prediction method based on a physical information neural network, and mainly relates to the technical field of industrial process intelligent control and energy power engineering crossing. Comprising the following steps: S1, acquiring operation data of the circulating fluidized bed, and preprocessing the operation data to obtain a data set; s2, constructing a physical information residual neural network; s3, constructing a loss function of adaptive weight adjustment; s4, training the physical information residual neural network by using the data set, and performing iterative optimization in combination with the constructed physical information residual neural network and the loss function to obtain a bed temperature prediction model; s5, inputting operation data of the circulating fluidized bed into the bed temperature prediction model to obtain a bed temperature prediction value; the method can solve the problem of complexity and nonlinearity of dynamic prediction of the bed temperature of the circulating fluidized bed, can optimize the operation efficiency of the boiler, improves the combustion stability, and reduces the emission of pollutants.

According to the plateau area load prediction method and system based on the signal decomposition and the neural network, an original signal is decomposed into a plurality of intrinsic mode components with physical significance by adopting completely adaptive noise set empirical mode decomposition, mode aliasing is suppressed, and multi-scale features are extracted to effectively suppress load demand oscillation; the problem of non-stationarity of the load of the service area is solved; then, K-means clustering grouping is carried out on the components, similar modes are combined to reduce calculation redundancy, and the problems of heterogeneity and local feature redundancy in complex time series data are effectively solved; performing nonlinear dimension reduction processing on the grouped reconstructed signals by adopting kernel principal component analysis, eliminating noise interference and extracting key features, thereby effectively solving the multiple nonlinear problem; and finally, inputting the dimensionality-reduced features into a long-short-term memory network modeling time sequence dependency relationship, and realizing high-precision prediction through adaptive weighted integration of component prediction results.

The invention provides a structure reliability analysis method based on an adaptive variable fidelity model, and the method comprises the steps: generating an initial sample point set which is uniformly distributed and has representativeness through an improved random sampling method KMODMC, enabling the initial sample point set to comprise a low-fidelity sample set and a high-fidelity sample set, training a BP neural network through employing the low-fidelity sample set, and carrying out the training of the BP neural network through employing the high-fidelity sample set; a low-fidelity BP neural network model is obtained; meanwhile, based on an error training Kriging model of a high-fidelity sample set and a low-fidelity model predicted value, an error correction Kriging model is constructed, the low-fidelity BP neural network model and the error correction Kriging model are combined to form a multi-fidelity mixed agent model, and adaptive iterative optimization is performed through a double-model alternate point adding sampling strategy. And finally obtaining a high-precision multi-fidelity hybrid agent model for structural reliability evaluation. According to the method, the problems of high cost of high-fidelity simulation calculation and insufficient precision of a low-fidelity model are solved, and the adaptive capacity and prediction precision of the model in a complex nonlinear problem are effectively improved.

The invention provides a topological optimization structure design method based on space control points and a space point control technology of structure topological optimization. The SPCT constructs a field function by utilizing spatial arrangement of control points, so that the number of design variables of topological optimization is effectively reduced, the optimization calculation efficiency is improved, and a chessboard effect and medium-density elements in a traditional topological optimization method are avoided. The control point placement mode is visual, the initial configuration or sample data can be conveniently controlled, and the optimization efficiency is further improved. Besides, sensitivity information based on an SPCT topological optimization model is deduced, a gradient-based optimization framework is established, and a non-gradient algorithm solution scheme can be used to solve the complex nonlinear problem without the help of gradient information. The technology has obvious advantages in the aspects of reducing design variables and improving calculation efficiency, and particularly has wide application prospects in the field of structure optimization design.

According to the source-load-storage coupled multi-energy micro-grid double-layer optimization method and system for the oil and gas well field, an upper-layer model, namely a load optimization layer model with the purpose of maximizing the green power consumption rate, is constructed respectively, the working system and production parameters of production equipment are optimized, and the wind and light consumption level is enhanced while the production requirement is guaranteed; a lower-layer model, namely a power generation dispatching layer model taking the lowest daily operation cost of the multi-energy micro-grid as a target, is constructed, the energy storage peak regulation capability is fully exerted, and the economic benefit of system operation is improved; meanwhile, due to the fact that the coupling relation exists between the upper layer and the lower layer of the double-layer model, the double-layer nonlinear problem is converted into single-layer mixed integer linear programming through the KKT condition and Big-M linearization, the solving time is greatly shortened, and the method is suitable for a real-time scheduling scene. And the green electricity consumption rate and the well group daily fluid production capacity are respectively improved by 9.34% and 9.61%. And the double-target balance of economy and environmental protection is realized.

The invention discloses a semi-analytic solution-based energy flow calculation method and system for an electricity-gas integrated energy system, and the method comprises the steps: firstly building an ordinary differential-algebraic model of an electricity-gas network based on a numerical boundary extrapolation and semi-discrete method; based on a semi-analytical theory of differential transformation, establishing a linearized electricity-gas network model; carrying out block calculation on a matrix and self-adaptive adjustment on a time window; and finally solving the model to obtain an energy flow calculation result. According to the method, the original nonlinear partial differential equation of the gas network is converted into the linear algebraic equation, the complex nonlinear problem in traditional gas network modeling is simplified, errors caused by local linearization and time difference are avoided, the calculation scale is reduced, and the calculation robustness is improved.

The invention discloses a damaged road network recovery sequence optimization method with maximum toughness as a target based on user balance flow distribution, and aims to solve the problems of low recovery efficiency of a post-disaster traffic road network and lack of scientific basis for decision making. The method comprises the following implementation steps: firstly, constructing a damaged road network toughness evaluation framework, and defining a toughness index as a reciprocal of a product of recovery time and total transit time in combination with a traffic network topological structure and a flow demand; secondly, establishing damaged road network toughness optimization scene parameters, abstracting a road network into a connected graph, and defining a recovery scheme set; then, a mixed integer linear programming model with the maximum toughness as the target is constructed, a nonlinear problem is converted into a linear problem through secant approximation, and constraint conditions such as flow balance and travel requirements are set; and finally, performing global optimization solution on the model by using a high-performance solver to generate an optimal recovery scheme. According to the method, the road network toughness can be scientifically quantified, the recovery sequence is optimized, blindness of experience decision making is avoided, the passing efficiency of a post-disaster traffic system is rapidly improved, efficient and scientific decision support is provided for post-disaster traffic recovery, and the method has important practical application value.

The invention relates to a low-carbon economic dispatching method and system for an integrated energy system, and the method comprises the steps: accurately describing the variable efficiency characteristics of energy conversion equipment through constructing a dynamic energy concentrator model based on a BP neural network, and processing a nonlinear problem through employing a linearization method; a typical source load prediction scene is generated in combination with a scene generation and reduction technology, and an energy flow-carbon flow double-layer optimization model is established. Wherein the energy flow layer takes the minimum operation cost as a target and considers equipment operation, multi-energy flow balance and transaction constraints; a stepped carbon transaction mechanism and demand response guided by carbon potential are introduced into a carbon flow layer, and the purpose is to minimize the cost of carbon transaction and demand response. And through iterative solution of the double-layer model, collaborative optimization of a system energy purchase and sale plan, equipment output and load change is realized until an optimal scheduling scheme is converged. According to the method, collaborative optimization of the energy flow and the carbon flow is realized, and double improvement of the scheduling of the comprehensive energy system in economy and low carbon is promoted.

This invention discloses a real-time attitude estimation system for non-cooperative industrial targets based on strong tracking filtering, relating to the fields of industrial machine vision and automation control technology. This system decouples the complete six-degree-of-freedom attitude estimation problem into a linear translational sub-state space and a nonlinear rotational sub-state space. This decomposes the nonlinear calculations that would normally be performed in six dimensions into a three-dimensional linear problem and a three-dimensional nonlinear problem. The translational sub-state only performs a simplified Kalman gain iteration with closed-form solutions, and matrix operations are simplified to low-order linear recursion. Although the rotational sub-state uses strong tracking filtering, it only needs to perform nonlinear updates in the reduced low-dimensional space. The Jacobian matrix sparsification pre-computation and incremental update unit pre-stores the constant part of the observation equation's derivative calculation offline and only incrementally updates the sparsely changing part online. This allows the system to be fully deployed on ARM architecture edge devices and stably meet millisecond-level processing requirements.

The application discloses a kind of space target ultra-short arc initial orbit intelligent determination method, belong to aerospace technology field.The application is based on artificial intelligence technology to realize space target ultra-short arc initial orbit determination, there is no strict requirement to complex nonlinear problem, and can develop more complex perturbation condition and initial orbit determination under the operation law of multiple body, coverage is wider;The application generates multiple groups of target orbits randomly and calculates the observation and state quantity of each visible arc segment, constructs the sample data set with ultra-large capacity, which is helpful for neural network to learn the nonlinear relationship between observation and state quantity using rich samples;The application builds and trains long short-term memory network model, determines the optimal network model weight parameter by designing and adjusting hyperparameter, can obtain the fast prediction solution of unknown space target ultra-short arc initial orbit according to observation;The application can meet more space target ultra-short arc initial orbit determination tasks by considering different inputs and outputs.

The present invention provides an active suspension control method, which includes the following steps: constructing a state space equation of the model according to the established dynamic model of the multi-variable non-linear active suspension to obtain an active suspension system model; linearizing the multi-variable non-linear links in the active suspension system model to obtain a corresponding quasi-linear system model of the active suspension; optimizing the quasi-linear system model of the active suspension by the Lagrange multiplier method, and solving by the bisection method to obtain the feedback gain K of the multi-variable non-linear active suspension and the feedback control law of the multi-variable non-linear active suspension. By adopting the above method, not only the non-linear problem of the inherent components is solved, but also the network non-linear problem is considered, improving the control accuracy and the stability of the system. At the same time, because the method is simple to reproduce, it is easy to implement in actual work.

The invention discloses an electrocatalyst activity attenuation prediction method based on microdynamics, and relates to the field of electrochemical catalyst.According to the electrocatalyst activity attenuation prediction method based on microdynamics, an electrocatalytic reaction and an inactivation reaction are dynamically coupled in a microdynamics model for the first time, the limitation of traditional thermodynamic stability analysis is broken through, and the activity attenuation process under the working condition can be truly reflected. By introducing a residual active site coverage variable and adopting quasi-steady-state approximation, the modeling and solving problems of a dynamic process of continuous loss of active sites are successfully solved, and a complex rigid differential equation set is simplified into an analyzable linear problem. The method not only can predict a stability attenuation curve, but also can quantitatively diagnose key steps influencing the service life through half-life period control degree analysis, realizes crossing from'phenomenon prediction 'to'mechanism interpretation' and'reverse design ', and provides a powerful theoretical tool for rationally designing a high-stability electrocatalyst.

The invention provides an original real-time electricity price prediction method based on a quantum complex neural network and Hilbert-Huang Transform (HHT), and the real-time electricity price prediction method based on the quantum complex neural network and the Hilbert-Huang Transform (HHT). Aiming at the non-stationarity of electricity price data on the historical level, firstly, the time internal correlation of each feature channel is extracted through an HHT time sequence analysis method, and complex multi-dimensional time sequence prediction is simplified into a simple regression task, so that a tedious time sequence modeling process is avoided; for the non-linear problem of electricity price data, a quantum neural network is used for capturing the coupling relation between different factors, and the calculation speed and the model efficiency are further improved by means of the parallelism of quantum calculation. Based on the two advantages, the method can realize accurate real-time electricity price prediction.

The application provides a topology optimization structure design method based on a space control point, and a space point control technology for structure topology optimization. The SPCT constructs a field function by using the space arrangement of the control points, effectively reduces the number of design variables of the topology optimization, improves the optimization calculation efficiency, and avoids the appearance of the chessboard effect and the medium density elements in the traditional topology optimization method. The control point placement method is intuitive and convenient for controlling the initial configuration or sample data, and further improves the optimization efficiency. In addition, the application derives the sensitivity information of the SPCT topology optimization model, establishes a gradient-based optimization framework, and can use a non-gradient algorithm to solve the scheme without the aid of gradient information to solve complex nonlinear problems. The technology has obvious advantages in reducing design variables and improving calculation efficiency, and has broad application prospects in the field of structure optimization design.

A method and apparatus for simulating the operation of a cascade reversible hydropower station based on piecewise McCormick relaxation is disclosed. The method includes: constructing a simulation model of the reversible hydropower station operation based on the operating conditions of the station and aiming to maximize the total output of the station; linearizing the nonlinear part of the simulation model using piecewise McCormick relaxation and piecewise linearization to simplify it into a mixed-integer linear programming model; and iteratively relaxing the simulation model based on upstream runoff data and operating parameters to obtain the simulation results. This invention uses piecewise McCormick relaxation to transform the nonlinear problem into a linear problem, while employing an iterative relaxation algorithm to improve the solution accuracy. It does not require narrowing the feasible boundary of McCormick relaxation, has fewer iterations, and a faster convergence speed, thus improving the speed and stability of the solution while ensuring simulation accuracy.

The application provides a structure reliability analysis method based on an adaptive variable fidelity model, and the method comprises the following steps: generating an initial sample point set with uniform distribution and representativeness through an improved random sampling method KMODMC, the initial sample point set comprising a low-fidelity sample set and a high-fidelity sample set; subsequently, training a BP neural network by using the low-fidelity sample set to obtain a low-fidelity BP neural network model; at the same time, training a Kriging model based on the error of the high-fidelity sample set and the predicted value of the low-fidelity model to construct an error correction Kriging model; combining the low-fidelity BP neural network model and the error correction Kriging model to form a multi-fidelity hybrid proxy model; and performing adaptive iterative optimization through a double-model alternating point sampling strategy to finally obtain a high-precision multi-fidelity hybrid proxy model for structure reliability evaluation. The application solves the problems of high cost of high-fidelity simulation calculation and insufficient precision of a low-fidelity model, and effectively improves the adaptability and prediction precision of the model in complex nonlinear problems.

The invention discloses a structure nonlinear dynamic response incremental modal overlay analysis method and system, and belongs to the field of aerospace structure dynamics calculation. The method comprises the steps of constructing a finite element model and obtaining an initial matrix; dispersing the time-varying temperature field into a plurality of time increment intervals, updating material parameters in each interval, and calculating an instantaneous matrix; carrying out eigenvalue analysis and converting the kinetic equation into a modal coordinate system; solving a modal response increment by adopting a numerical integration method and converting the modal response increment to a physical coordinate system for superposition; and circularly executing until a full-time-history response is output. According to the method, the problem of rigidity nonlinearity caused by the temperature is solved through incremental linearization and modal order reduction, high calculation precision and high efficiency are achieved, and the method is suitable for dynamic design and analysis of aerospace structures in the high-temperature variable-temperature environment.

The invention belongs to the technical field of electric vehicle load optimization, and particularly discloses an electric vehicle load multi-target double-layer optimization method considering peak load shifting, comprising the following steps: establishing an electric vehicle load multi-target double-layer optimization model; converting a double-layer optimization problem of the multi-target double-layer optimization model into a single-layer multi-target linear problem through a KKT condition and a maximum M method; further converting a single-layer multi-target linear problem into a single-layer single-target problem; solving a single-layer single-target problem to obtain a Parteo leading edge; and determining an ideal solution at a Parteo leading edge, and taking the ideal solution as a final result of an original double-layer multi-objective optimization problem to realize electric vehicle load multi-objective double-layer optimization. According to the method, the problem that the upper-layer model in the double-layer model is single-target optimization due to the fact that comprehensive indexes such as power distribution network side peak-valley difference and load fluctuation are not considered in an existing double-layer optimization operation model is solved, and the peak-valley difference and the load fluctuation can be reduced while the income of an electric vehicle user is guaranteed.

The present invention discloses a distribution network reconstruction method and device based on operation uncertainty. An interval uncertainty set is constructed according to the uncertainty of distribution network operation. A linearized power flow constraint is established for the distribution network, and a maximum slack variable is used to replace the nonlinear power flow over-limit of the distribution network, so that the original nonlinear problem can be converted into a linear problem. Minimization of the over-limit frequency and degree of distribution network state variables is used as an objective function, and the objective function is solved in combination with the interval uncertainty set and the linearized power flow constraint to reconstruct the distribution network with the minimum risk. Therefore, the influence of the uncertainty of distribution network operation on the network reconstruction model can be considered, which is conducive to the rapid network reconstruction of the distribution network and is suitable for application in the scenario of distribution network network reconstruction.

This invention discloses a method, apparatus, and medium for solving variable quantum linear problems in subspaces. The method includes: determining the system of linear equations to be solved, constructing a variable quantum circuit and a Krylov subspace, and using the generalized minimum residual method and the variable quantum circuit to calculate an approximate solution of the system of linear equations to be solved in the subspace. This addresses the shortcomings of existing technologies and can reduce the time complexity and computational load of solving linear problems.

The application discloses a kind of based on user balance flow distribution with maximum toughness as target damaged road network recovery order optimization method, to solve the problem of low efficiency, decision lacks scientific basis after disaster traffic network recovery.The method comprises the following implementation steps: first, construct damaged road network toughness evaluation framework, in combination with traffic network topological structure and flow demand, define toughness index as the reciprocal of recovery time and total travel time product;Second, establish damaged road network toughness optimization scene parameter, abstract road network into connected graph and define recovery scheme set;Then, construct the mixed integer linear programming model with maximum toughness as target, convert nonlinear problem into linear problem by cutting line approximation, and set flow balance, travel demand and other constraint conditions;Finally, the model is globally optimized and solved using high-performance solver, to generate the best recovery scheme.The method of the application can scientifically quantify road network toughness and optimize recovery order, avoid the blindness of experience decision, quickly improve the traffic efficiency of post-disaster traffic system, provide efficient and scientific decision support for post-disaster traffic recovery, and have important practical application value.

The invention discloses a displacement estimation method and system for transcranial ultrasonic shear wave viscoelastic fluid imaging, and belongs to the technical field of ultrasonic medical image processing, and the method comprises the steps: dividing ultrasonic radio frequency data after beam forming into a plurality of independently processed data blocks; the ultrasonic radio frequency data comprises two frames of ultrasonic radio frequency signals collected before and during shear wave propagation; constructing a cost function for each data block, wherein the cost function comprises a data similarity item and a mixed-order displacement continuity constraint item; performing linearization processing on the cost function, and converting a nonlinear optimization problem into a linear problem; establishing and solving a linear equation set to obtain local displacement of each data block; and splicing the local displacement of each data block to form a global displacement field. According to the method, high-precision estimation of the transcranial shear wave displacement and reliable evaluation of the viscoelastic fluid property of the brain tissue can be realized, relatively efficient and accurate displacement estimation is kept in a transcranial high-noise environment, and the performance of transcranial ultrasonic shear wave elastography is improved.

The invention discloses a battery DCR prediction method and related equipment, and the method comprises the steps: dividing the early-stage decrease stage and the later-stage increase stage of battery aging DCR through inflection points, constructing a growth rate model with a temperature index relation and a time power function coupled, solving parameters through dual linear fitting, and combining with a comprehensive working condition temperature to establish a prediction model. And substituting the duration and the initial DCR to obtain a target DCR value. According to the method, the model is enabled to accurately adapt to the DCR change rules of different aging stages through stage division, an actual mechanism is fitted, and prediction rationality is greatly improved. The coupling model can quantify the cooperative influence of temperature and time, effectively improves the prediction precision of the whole temperature interval, and is especially suitable for extreme temperature conditions. Nonlinear parameter solving is converted into a linear problem through double linear fitting, and the parameter matching degree and the solving precision are improved while calculation is simplified. The prediction model based on the comprehensive working condition can adapt to actual temperature fluctuation to output an accurate result, and a reliable basis is provided for battery aging evaluation.

This invention discloses an improved canonical perturbation high-order expansion prediction method and system for the critical flutter wind speed of long, flexible structures. The method includes: Step 1: Acquiring aerodynamic and structural parameters based on the Scanlan self-excited force model; Step 2: Constructing the bending-torsional coupling control matrix equation and introducing small parameters for perturbation expansion; Step 3: Using the improved canonical perturbation method, distinguishing between equal and unequal frequency cases, eliminating long-term terms by constructing eigenvalues, and solving for eigenvalues ​​and modal vectors step by step; Step 4: Deriving explicit analytical solutions for modal frequencies, damping ratios, amplitude ratios, and phase differences; Step 5: Determining the modal frequencies and directly solving for the damping ratio through a single iteration to determine the critical flutter wind speed. This invention decomposes complex nonlinear problems into linear superpositions of components, obtaining explicit closed-form solutions for the parameters. The calculation can be completed in a single iteration, significantly simplifying the analysis process and clearly revealing the influence mechanism of various structural and aerodynamic parameters on flutter instability.

The application discloses a fast complementary filter method for attitude estimation by using a low-cost MARG sensor, and comprises the following steps: 1) estimating an attitude quaternion by using data output by an accelerometer; 2) using a gravity filter for compensation of a gyroscope; and 3) fusing the gravity and the magnetometer. The application discloses the fast complementary filter method for attitude estimation by using the low-cost MARG sensor, and converts the attitude estimation problem based on the accelerometer into a linear problem in view of a nonlinear problem in the attitude quaternion calculation process. In order to improve the accuracy of MARG attitude estimation, a new complementary filter structure is designed, and the new complementary filter structure is characterized in that the gravity is estimated by fusing the measurements of the accelerometer and the gyroscope. Meanwhile, the magnetic distortion detection is introduced into the algorithm, and the output of the magnetometer is fused with the estimated gravity based on the Markley algorithm.

The invention discloses a health data classification method based on an Ising machine training TSK fuzzy system, and the method comprises the steps: collecting target health data to construct fuzzy sets, calculating leading parameters of different fuzzy sets through a Gaussian membership function, solving the activation degree of each sample for different rules, carrying out the weighted average of the output of the rules according to the activation degree, and carrying out the classification of the health data. The output of the TSK fuzzy system is expressed as a linear regression form; linear regression output is converted into a QUBO problem through equivalent mathematical changes; and solving the optimal unwinding solution by using a quantum simulated annealing algorithm, and realizing classification of the target health data by analyzing the unwinding solution. According to the method, efficient approximate calculation is performed locally through a quantum simulated annealing algorithm, or the method is deployed on a quantum computer platform to obtain a higher-speed global optimal solution. While the calculation efficiency is ensured, the quantum optimization process can effectively avoid a common local optimal trap in a traditional method, so that the overall performance of the model in a complex nonlinear problem is improved.

The present invention relates to a multi-sensor fusion positioning method and system under complex scenarios of driverless vehicles. The multi-sensor fusion positioning method uses a factor graph optimization method, that is, after converting the factor graph model constructed by positioning information into a non-linear problem by using maximum a posteriori probability estimation, the non-linear problem is solved to obtain the positioning result of the driverless vehicle, so as to fuse the observation data of the sensor positioning model, which can enhance the robustness of multi-sensor fusion positioning under complex scenarios of driverless vehicles while effectively improving the positioning accuracy and speed estimation accuracy of driverless vehicles.

The present invention relates to the technical field of adjustable nozzle design, and discloses a method and system for equivalent analysis of aerodynamic loads of adjustable nozzle flaps. By constructing a three-dimensional structure model of the adjustable nozzle, dividing the inlet section and the throat section of the adjustable nozzle, and analyzing the engine outlet and the throat section of the adjustable nozzle, the proportional amplification coefficient of the flap is obtained. Then, the proportional amplification coefficient is used to equivalent the initial aerodynamic load of the flap, and the equivalent value of the wall aerodynamic load on the flap is obtained. The equivalent value of the aerodynamic load includes the load transmitted by the seal flap, and there is no need to consider the load of the seal flap during the aerodynamic analysis of the adjustable nozzle, which solves the non-linear problem of line-surface contact between the fish scales (seal flap and flap) during dynamic analysis, and greatly improves the convergence of the calculation model.

The invention provides a multi-element energy storage collaborative fluctuation stabilizing method, which comprises the following steps of: 1, performing empirical mode decomposition on an original new energy grid-connected power signal to obtain a plurality of IMF components; 2, reconstructing the IMF component according to the grid-connected power and the amplitude-frequency response capability of each energy storage device, and dividing a low-frequency component, an intermediate-frequency component and a high-frequency component corresponding to different time scales respectively; step 3, establishing a charge-discharge efficiency linearization model of each energy storage device, performing linearization modeling based on an MLD theory, and processing a nonlinear problem through a piecewise linearization method; and 4, by taking the minimum grid-connected fluctuation and the highest energy utilization rate as targets, constructing a multi-element energy storage collaborative fluctuation stabilizing stochastic optimization model, and optimizing the output strategy of each energy storage device. Through EMD decomposition and reconstruction, separation of power fluctuation at different time scales is realized, so that different types of energy storage devices give full play to the frequency response advantage, and the cooperation efficiency and the energy utilization rate are improved.

The application discloses a kind of based on BEO-XGBoost algorithm's side slope stability evaluation method, and the invention includes the following steps: S1, the factor that influences side slope stability is analyzed, determines the parameter for being used for analyzing side slope stability, constructs side slope dataset;S2, the data distribution situation is analyzed, and data set is handled;S3, establishes XGBoost model, and data set is divided into training set and test set;S4, constructs BEO model and optimizes XGBoost;S5, BEO-XGBoost model is trained and tested.The BEO algorithm used in the application is a powerful heuristic algorithm developed based on the biological behavior of black eagles, which simulates the hunting, migration and breeding behavior of black eagles.Through simulating these behaviors of black eagles, BEO can gradually find the optimal solution from a complex search space, which makes BEO suitable for optimizing complex nonlinear problems such as machine learning models.XGBoost is an ensemble algorithm based on decision trees, which uses gradient descent to optimize the residual error of each tree, thereby continuously improving the accuracy of the model.Using the BEO-XGBoost model to predict the stability of the slope provides a new and highly adaptable hyperparameter optimization strategy that combines the global and local search mechanisms of BEO and significantly improves the accuracy of slope stability prediction through dynamic adaptive search.