Personnel and goods elevator top plate bearing capacity evaluation method with self-adaptive early warning function

Abstract

The invention provides a passenger and goods elevator roof bearing capacity evaluation method with a self-adaptive early warning function, which comprises the following steps of: acquiring data of a multi-source sensor, and performing data acquisition and preprocessing by adopting a dynamic synchronous sampling optimization algorithm; extracting time-frequency characteristics, separating load modes by adopting a multi-dimensional time-frequency analysis algorithm, and constructing a dynamic load spectrum; establishing multi-scale state mapping, calculating system response by adopting a generalized dynamic response analysis method, and evaluating structural risk; and constructing a self-adaptive early warning system, dynamically optimizing an early warning threshold value, and generating an optimal decision scheme. According to the method, the sampling strategy can be adaptively adjusted, the load characteristics can be accurately recognized, multi-scale state evaluation is achieved, early warning decisions are dynamically optimized, and the accuracy and reliability of people and goods elevator top plate bearing capacity evaluation are remarkably improved.

Description

Technical Field

[0001] The invention relates to bearing capacity assessment, in particular to a method for assessing the bearing capacity of a passenger and freight elevator top plate with an adaptive early warning function. Background Art

[0002] With the rapid development of urban construction, the demand for passenger and freight elevators in construction is increasing. As a key structure for carrying construction personnel and materials, the load-bearing capacity of the passenger and freight elevator roof is directly related to construction safety and efficiency. Especially in the construction process of super-high-rise buildings, the passenger and freight elevator roof needs to withstand complex and changeable load conditions, including the combined influence of dynamic impact loads, asymmetric load distribution, temperature stress and other factors. Therefore, the development of a roof load-bearing capacity assessment method with adaptive early warning function is of great practical significance for preventing safety accidents, protecting the lives and property of construction personnel, and improving construction efficiency.

[0003] At present, the bearing capacity of passenger and freight elevator roofs is mainly assessed by traditional static analysis methods, such as linear finite element analysis and empirical formula calculation. Some researchers have begun to introduce dynamic response analysis, using acceleration sensors to monitor the vibration characteristics of the roof and evaluating the structural state through spectrum analysis. There are also studies that obtain stress distribution data by arranging strain gauges and identify structural damage by combining modal analysis methods. In terms of early warning, single indicator monitoring based on fixed thresholds is mainly used, such as displacement over-limit alarm and stress over-limit alarm. Some researchers have tried to use artificial intelligence methods such as neural networks to establish early warning models, but they are mainly based on static models trained offline.

[0004] However, the existing technical solutions still have the following problems: First, the sensor data collection lacks an adaptive optimization mechanism, the sampling frequency is fixed, and it is difficult to dynamically adjust according to the changes in working conditions, which may lead to the loss of important data features at critical moments; second, the time-varying characteristics are not fully considered in the process of load feature extraction, especially under non-stationary working conditions, the traditional time-frequency analysis method is difficult to accurately capture the dynamic changes of the load mode; third, the state assessment model is mostly based on a single scale analysis, ignoring the coupling effect between the macro-structural response, local stress distribution and material performance evolution, and it is difficult to fully reflect the structural state; fourth, the early warning decision lacks adaptive ability, the early warning threshold setting is too rigid, and it fails to dynamically adjust according to the real-time risk assessment results. At the same time, the decision optimization process does not fully consider the multi-objective trade-off, and it is difficult to achieve the optimal balance between benefits, timeliness and cost; finally, the system parameter optimization lacks an effective feedback learning mechanism, and it is impossible to automatically adjust and optimize the system performance according to the historical decision effect. These technical problems seriously restrict the actual application effect of the passenger and freight elevator top plate bearing capacity assessment system. Summary of the invention

[0005] The purpose of the invention is to provide a method for evaluating the bearing capacity of the top plate of a passenger and freight elevator with an adaptive early warning function, in the hope of solving the above-mentioned problems existing in the prior art.

[0006] A technical solution, a method for evaluating the bearing capacity of a passenger and freight elevator top plate with an adaptive early warning function, comprises the following steps:

[0007] Step S1, collecting data streams of multiple sensors, including load sensor data, vibration sensor data, displacement sensor data and temperature sensor data, and preprocessing them to obtain a final processed data stream;

[0008] Step S2, receiving the data stream after final processing, constructing a time-frequency window function according to the data time series, performing convolution operation on the window function and the data stream, and calculating the time-frequency energy distribution matrix according to the extracted time-frequency features; separating different load modes from the time-frequency energy distribution matrix, calculating the correlation coefficients between the modes, and constructing the modal weight matrix; evaluating the energy contribution rate of each load mode, calculating its coefficient of variation and stability index, generating a feature importance score, performing feature screening and fusion based on the score, and obtaining a fused feature vector; calculating the load probability density distribution according to the fused feature vector, constructing a dynamic load spectrum and establishing a reliability index;

[0009] Step S3, receiving the dynamic load spectrum and reliability index, inputting these data into the macro-structural response analysis module, the local stress distribution analysis module and the material performance analysis module respectively, and calculating the state vectors of three levels; establishing the state transfer matrix according to the distance relationship between the state vectors of the three levels; inputting the state transfer matrix into the dynamic equation solver, and calculating the acceleration response spectrum, velocity response spectrum and displacement response spectrum of the system; calculating the static risk value, dynamic risk value and fatigue risk value respectively based on the response spectrum data, and determining the weight coefficient according to the information entropy of each risk index, and generating a comprehensive risk index and a risk evolution trend; predicting the future state distribution according to the risk evolution trend, calculating the probability of the system entering the failure state, and obtaining the reliability prediction value and the failure probability prediction value;

[0010] Step S4, receiving the reliability prediction value, the failure probability prediction value and the comprehensive risk index, calculating the initial warning threshold value according to the historical data, combining the warning threshold value with the risk change rate to construct a dynamic threshold function; calculating the warning discrimination value based on the dynamic threshold function and the dynamic response characteristics, combining the warning discrimination value with the prior probability distribution to obtain the warning level; constructing the decision space according to the warning level and the risk evolution trend, calculating the benefit value, timeliness index and cost index of each optional decision plan, and obtaining the optimal decision plan through multi-objective optimization; comparing the execution effect of the optimal decision plan with the expected goal, calculating the evaluation error, and updating the knowledge base parameters and system parameters according to the error.

[0011] Beneficial effects include being able to adaptively adjust sampling strategies, accurately identify load characteristics, realize multi-scale state assessment, dynamically optimize early warning decisions, and significantly improve the accuracy and reliability of passenger and freight elevator top plate bearing capacity assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a flow chart of the present invention.

[0013] Figure 2 It is a flow chart of step S1 of the present invention.

[0014] Figure 3 It is a flow chart of step S2 of the present invention.

[0015] Figure 4 It is a flow chart of step S3 of the present invention.

[0016] Figure 5 It is a flow chart of step S4 of the present invention. DETAILED DESCRIPTION

[0017] Embodiment 1, as Figure 1 As shown, the method for evaluating the top plate bearing capacity of a passenger and freight elevator with an adaptive early warning function includes the following steps:

[0018] Step S1, collecting data streams of multiple sensors, including load sensor data, vibration sensor data, displacement sensor data and temperature sensor data, calculating the importance score of each sensor according to the signal strength, noise level and signal change rate of each sensor, adjusting the sampling frequency of each sensor based on the importance score, and obtaining a synchronously optimized data stream; decomposing the synchronously optimized data stream into multi-dimensional harmonic components and residual terms, calculating the local variance of each frequency component, constructing a filter function, filtering the data, and obtaining a denoised data stream; mapping the denoised data stream to a multi-dimensional manifold space, calculating the gradient vector in the manifold space, identifying abnormal data based on the local average eigenvector and the eigenvector standard deviation, and obtaining a data stream with abnormal markings; constructing a local reconstruction function according to the data stream with abnormal markings, calculating the reconstruction error, updating the compensation coefficient, and obtaining the final processed data stream.

[0019] Step S2, receiving the final processed data stream output by step S1, constructing a time-frequency window function according to the data time series, performing convolution operation on the window function and the data stream, and calculating the time-frequency energy distribution matrix according to the extracted time-frequency features; separating different load modes from the time-frequency energy distribution matrix, calculating the correlation coefficients between the modes, and constructing the modal weight matrix; evaluating the energy contribution rate of each load mode, calculating its coefficient of variation and stability index, generating a feature importance score, performing feature screening and fusion based on the score, and obtaining a fused feature vector; calculating the load probability density distribution according to the fused feature vector, constructing a dynamic load spectrum, and establishing a reliability index.

[0020] Step S3, receiving the dynamic load spectrum and reliability index output by step S2, inputting these data into the macro-structural response analysis module, the local stress distribution analysis module and the material performance analysis module respectively, and calculating the state vectors of three levels; establishing the state transfer matrix according to the distance relationship between the state vectors of the three levels; inputting the state transfer matrix into the dynamic equation solver, and calculating the acceleration response spectrum, velocity response spectrum and displacement response spectrum of the system; calculating the static risk value, dynamic risk value and fatigue risk value respectively based on the response spectrum data, and determining the weight coefficient according to the information entropy of each risk index, and generating a comprehensive risk index and a risk evolution trend; predicting the future state distribution according to the risk evolution trend, and calculating the probability of the system entering a failure state.

[0021] Step S4, receiving the reliability prediction value, failure probability prediction value and comprehensive risk index output by step S3, calculating the initial warning threshold value according to historical data, combining the warning threshold value with the risk change rate to construct a dynamic threshold function; calculating the warning discrimination value based on the dynamic threshold function and the dynamic response characteristics, combining the warning discrimination value with the prior probability distribution to obtain the warning level; constructing a decision space according to the warning level and the risk evolution trend, calculating the benefit value, timeliness index and cost index of each optional decision plan, and obtaining the optimal decision plan through multi-objective optimization; comparing the execution effect of the optimal decision plan with the expected goal, calculating the evaluation error, and updating the knowledge base parameters and system parameters according to the error.

[0022] This technical solution realizes the intelligent and adaptive evaluation of the elevator top plate bearing capacity by constructing a complete technical chain of "data collection-feature extraction-state assessment-early warning decision". The innovation of the solution is mainly reflected in: ① The adaptive collection and intelligent preprocessing of multi-source data ensure the quality of the original data; ② The application of time-frequency joint analysis and modal decomposition technology improves the accuracy of feature extraction; ③ The construction of a multi-scale state assessment model realizes comprehensive state monitoring; ④ The design of an adaptive early warning decision mechanism ensures the timeliness of the early warning and the optimality of the decision. These technical innovations cooperate with each other to form a closed-loop intelligent evaluation system, which significantly improves the accuracy, reliability and practicality of the elevator top plate bearing capacity evaluation. In practical applications, this solution can effectively prevent safety accidents, protect the lives and property of construction workers, and improve construction efficiency, and has important engineering application value.

[0023] According to one aspect of the present application, step S1 specifically comprises:

[0024] Step S11: According to the preset reference sampling frequency, load sensor data, vibration sensor data, displacement sensor data and temperature sensor data are respectively obtained, and the signal strength, noise level and signal change rate of each sensor are calculated. These parameters are multiplied by the preset weight coefficient and summed to obtain the sensor importance score. The optimal sampling frequency is calculated based on the sensor importance score, and the original data stream is resampled using the optimal sampling frequency to obtain a synchronously optimized data stream.

[0025] Among them, the adaptive sampling frequency optimization method f(i,t) = f_base·exp(η(t)·W(i,t))·(1 + δ(t)·ΔW(i,t)); where: f(i,t) is the optimal sampling frequency of sensor i at time t; f_base is the base sampling frequency; η(t) = η o ·(1-exp(-λ·||dW / dt||)) is the adaptive gain coefficient; W(i,t) is the sensor importance score; ΔW(i,t) = W(i,t) - W̄(t) is the relative importance deviation, W̄(t) is the average importance score; δ(t) = δ o ·exp(-μ·var(W(:,t))) is the compensation coefficient; η o ,λ,δ o , μ are control parameters; ||·|| is the Euclidean norm; var(·) is the variance operator.

[0026] Step S12, receiving the synchronized optimized data stream outputted from step S11, dividing the data stream into a plurality of time windows according to the time series, applying the generalized harmonic basis function to the data in each time window to decompose, and obtaining a series of harmonic components and corresponding residual terms; calculating the variance value of each harmonic component in its local time window, and constructing an adaptive filtering function according to the variance value; applying the adaptive filtering function to each harmonic component respectively, performing threshold filtering on the residual terms, and recombining the processed harmonic components and residual terms to obtain a denoised data stream.

[0027] Step S13, importing the denoised data stream obtained in step S12, establishing a multidimensional coordinate system, mapping the data points into the coordinate system to form a manifold space; calculating the Euclidean distance between adjacent data points, and constructing a local connection relationship matrix; calculating the local gradient vector of each point based on the connection relationship matrix; calculating the average eigenvector and the standard deviation of the eigenvector in the local area based on historical data; comparing the current gradient vector with the average eigenvector, and determining the abnormal point according to a preset standard deviation threshold; adding marking information to the identified abnormal point, and outputting the data stream with the abnormal mark.

[0028] Step S14, read the data stream with abnormal marks output by step S13, divide the data sequence into multiple overlapping local intervals in chronological order; in each local interval, select a suitable combination of basis functions according to the distribution characteristics of non-abnormal points; calculate the weight coefficients of the basis functions by the least squares method, and establish a local reconstruction function; calculate the mean square error between the reconstruction function and the actual data; based on the gradient information of the mean square error, iteratively update the weight coefficient until the preset convergence condition is met; use the reconstruction function to correct and compensate the data of the abnormal points; smoothly connect the reconstruction results of all local intervals, and finally output the processed data stream.

[0029] By constructing a sensor importance scoring function and an adaptive sampling frequency optimization mechanism, intelligent collection and synchronous optimization of multi-source sensor data are achieved. This step uses wavelet transform to reduce data noise, combines manifold learning methods for anomaly detection, and finally corrects and compensates for abnormal data through local reconstruction functions. This multi-level data preprocessing method significantly improves the quality and reliability of raw data. Specifically, in the elevator top plate bearing capacity assessment scenario, it can effectively handle noise interference and abnormal data of load sensors, vibration sensors, displacement sensors, and temperature sensors to ensure the accuracy of subsequent analysis. By dynamically adjusting the sampling frequency, the system can increase the sampling density at critical moments (such as load mutations) and reduce the sampling frequency in a stable state, which not only ensures the integrity of the data, but also optimizes the efficiency of system resource utilization.

[0030] According to one aspect of the present application, step S2 is specifically:

[0031] Step S21, importing the final processed data stream output by step S1, segmenting the data stream, and weighting each data segment using a preset time window function; constructing an adaptive kernel function, which includes a time domain window function and a phase compensation function; performing a convolution operation on the weighted data segment and the adaptive kernel function to calculate a time-frequency distribution matrix; respectively extracting amplitude information and phase information from the time-frequency distribution matrix; calculating the time-frequency energy distribution based on the amplitude information and phase information to form a time-frequency feature matrix.

[0032] The specific process of time-frequency feature extraction is: TF(t,ω) = |∫x(τ)·g((t-τ) / α(t))·exp(-jωτ+ jφ(t,ω))dτ| 2 ; Where: TF(t,ω) is the time-frequency energy distribution; x(τ) is the signal time series; g(·) is the adaptive window function, g(u) = exp(-u 2 / 2)·(1 + β(t)·u 4 ), α(t) is the time-varying window width parameter, φ(t,ω) = θ(t)·ω 2 + γ(t)·ω³ is the phase compensation function; β(t), θ(t), γ(t) are dynamic optimization parameters; t is the time variable; ω is the frequency variable; j is the imaginary unit.

[0033] Step S22, read the time-frequency characteristic matrix output by step S21, construct a time-varying amplitude function and a frequency phase function, and combine these functions to form a load modal basis function; calculate the projection coefficient of the time-frequency characteristic matrix on each basis function; sort the load modes according to the size of the projection coefficient, and retain the main load modes; calculate the inner product between the retained load modes to obtain the modal correlation coefficient matrix; determine the weight coefficient of each mode according to the correlation coefficient matrix and the modal energy value; combine the weight coefficient with the corresponding load mode, and output the load mode set.

[0034] Among them, the process of load modal decomposition is: B_k(t,ω) = A_k(t)·exp(jΦ_k(ω))·exp(-ζ_k(t)·|ω-ω_k(t)| 2 );where: B_k(t,ω) is the time-frequency basis function of the kth load mode; A_k(t) is the time-varying amplitude function, which is obtained by EMD decomposition; Φ_k(ω) is the frequency phase function, which is calculated by phase unwrapping algorithm; ζ_k(t) is the frequency localization parameter; ω_k(t) is the center frequency function; k is the mode number; t is the time variable; ω is the frequency variable; |·| is the absolute value operator.

[0035] Step S23, receiving the load mode set output by step S22, calculating the energy value, variance and time series stability index of each mode; multiplying the energy value, variance and stability index with the preset weight coefficient to obtain the importance score of each mode; calculating the redundancy penalty factor according to the correlation coefficient between the modes; combining the importance score with the redundancy penalty factor to determine the final feature selection criterion; screening the modes based on the selection criterion, weightedly combining the retained modes according to the importance score, and generating a fused feature vector.

[0036] Step S24, import the fused feature vector output by step S23, and segment the feature vector within a preset time window; calculate the statistical features of each segment of data, including the mean vector and the covariance matrix; construct a multidimensional Gaussian mixture model based on the mean vector and the covariance matrix, and calculate the probability density distribution of the load; derive the probability density distribution function to obtain the load change rate; construct a dynamic load spectrum according to the load change rate; calculate the system reliability index based on the dynamic load spectrum, and output the dynamic load spectrum and reliability index.

[0037] By introducing time-frequency joint analysis and modal decomposition technology, multi-dimensional extraction and fusion of elevator roof load characteristics are achieved. This step first uses the time-frequency transformation method to obtain the time-frequency feature matrix of the signal, then separates different load modes through the modal decomposition algorithm, and performs feature screening and fusion based on energy contribution rate and stability index. This method can not only capture the static characteristics of the load, but also reflect its dynamic change process. In practical applications, the system can accurately identify the load characteristics of the elevator roof under different working conditions, such as static load, dynamic impact, fatigue load, etc., and by constructing a dynamic load spectrum, it provides a reliable feature basis for subsequent state evaluation. Especially when dealing with non-stationary working conditions, this method can effectively capture the dynamic changes of the load mode and improve the accuracy and adaptability of feature extraction.

[0038] According to one aspect of the present application, step S3 is specifically:

[0039] Step S31, import the dynamic load spectrum and reliability index output from step S2, calculate the overall structural response vector through the finite element model, calculate the local stress distribution vector using the stress concentration factor, and generate a microscopic performance vector in combination with the material performance curve; calculate the Euclidean distance between adjacent state vectors, and construct a state transfer function with a time decay term; dynamically adjust the bandwidth parameter of the state transfer function according to the rate of change of the reliability index; apply the adjusted state transfer function to each state vector to generate a state transfer probability matrix; apply time weight to the state transfer probability matrix, and output a state mapping matrix.

[0040] The specific multi-scale state mapping is: Q(t) = exp(-||S(t)-S_ref||2 / θ(t))·exp(λ(t)·R(t))·M(t); where: Q(t) is the state mapping matrix; S(t) is the current state vector; S_ref is the reference state vector; θ(t) = θ o ·exp(-η·|dR / dt|) is the adaptive bandwidth function; R(t) is the reliability index; λ(t) is the state weight function; M(t) is the quality matrix; ||·|| is the Mahalanobis distance; θ o , η are control parameters; t is the time variable.

[0041] Step S32, read the state mapping matrix output by step S31, construct a time-varying mass matrix according to the mass distribution of the system; construct a damping matrix based on the nonlinear relationship between displacement and velocity; calculate the stiffness matrix by combining material properties and geometric features; substitute these matrices into the dynamic equation; construct a set of orthogonal dynamic basis functions, and calculate the combination coefficients of the basis functions using the variational principle; solve the dynamic equation by an iterative optimization method to obtain time domain response data; perform Fourier transform on the time domain response data to obtain acceleration response spectrum, velocity response spectrum and displacement response spectrum respectively.

[0042] The solution to the dynamic equation is: x** + C(x,x*)·x* + K(x)·x = F(t); where: C(x,x*) = C o + α·|x*| + β·x 2 is the nonlinear damping matrix; K(x) = K o + γ·x + δ·x³ is the nonlinear stiffness matrix; F(t) is the external load vector; C o is the linear damping coefficient; K o is the linear stiffness coefficient; α, β, γ, δ are nonlinear parameters; x is the displacement vector; x* is the velocity vector; x** is the acceleration vector; t is the time variable.

[0043] Step S33, receiving the three response spectrum data outputted from step S32, calculating the static over-limit probability based on the extreme value distribution theory to obtain the static risk index; calculating the dynamic impact effect using the time integral of the response spectrum to obtain the dynamic risk index; calculating the fatigue damage accumulation value according to the number of cycles and amplitude distribution of the response spectrum to obtain the fatigue risk index; calculating the time series entropy value of the three risk indicators; determining the weight coefficient of each risk indicator according to the entropy value; substituting the weighted risk indicator into the risk evolution equation; solving the risk evolution equation by the numerical integration method to obtain the comprehensive risk indicator and the risk evolution trend.

[0044] The risk assessment process is specifically as follows: R(t) = w1(t)·R_s(t) + w2(t)·R_d(t) + w3(t)·R_f(t); where: R(t) is a comprehensive risk index; R_s(t) = P(max|x(t)| > x_cr) is a static over-limit risk, calculated based on the extreme value distribution theory; R_d(t) = ∫|x**(t)|dt / T_cr is a dynamic impact risk; R_f(t) = Σ(n_i / N_i)^m is a fatigue damage risk; w_k(t) = exp(H_k(t)) / Σexp(H_j(t)) is an entropy weight coefficient; H_k(t) is information entropy; x_cr is a critical displacement; T_cr is a characteristic time; n_i is the number of cycles; N_i is the number of lifespans; and m is a material constant.

[0045] Step S34, import the comprehensive risk index and risk evolution trend output from step S33, establish a prediction model that describes the system state transition; construct a risk impact function based on the risk index; introduce random disturbance terms into the state prediction model to simulate the random fluctuation characteristics of the system; define the safety domain boundary conditions of the system; use the Monte Carlo method to generate a large number of state samples; count the proportion of samples that are within the safety domain to obtain a reliability prediction value; calculate the probability of crossing the safety domain boundary for the first time in the sample to obtain a failure probability prediction value; output the reliability prediction value and the failure probability prediction value.

[0046] The reliability prediction process is as follows: P_f(t,τ) = ∫∫p(z,ω)·I(g(z,ω) < 0)dzdω·exp(-λ(t)·R(t)); where: P_f(t,τ) is the predicted failure probability; p(z,ω) is the state-load joint probability density; g(z,ω) is the limit state function; I(·) is the characteristic function; z is the state vector; ω is the load vector; λ(t) is the time-varying attenuation coefficient; R(t) is the reliability index; t is the current moment; τ is the prediction time step.

[0047] By constructing a multi-scale state assessment model and a nonlinear dynamic response analysis method, a comprehensive assessment and risk prediction of the load-bearing state of the elevator roof is achieved. This step innovatively introduces the coupling analysis of three levels: macroscopic structural response, local stress distribution, and material properties, and describes the interaction between different levels through a state transfer matrix. At the same time, the use of a dynamic equation solver combined with the entropy weight method for risk assessment can accurately predict the reliability and failure probability of the system. This multi-scale analysis method is particularly suitable for key structures such as elevator roofs that bear complex loads. It can simultaneously consider multiple aspects such as overall deformation, local stress concentration, and material performance degradation, providing more comprehensive and accurate state assessment results.

[0048] According to one aspect of the present application, step S4 is specifically:

[0049] Step S41, import the reliability prediction value, failure probability prediction value and comprehensive risk index output by step S3, and read the initial threshold of each level from the warning level database; calculate the time derivative of the comprehensive risk index to obtain the risk change rate; substitute the risk change rate into the exponential decay function to calculate the threshold adjustment factor; multiply the initial threshold by the adjustment factor to generate a dynamic warning threshold; calculate the mean square error between the current warning threshold and the historical optimal threshold to obtain the threshold credibility; construct a constraint equation according to the threshold credibility, and obtain the target threshold by numerical solution; use the gradient descent method to iteratively optimize the warning threshold until the convergence condition is met, and output the dynamic warning threshold set and threshold credibility.

[0050] The specific process of dynamic threshold optimization is: Th(k,t) = Th o (k)·exp(-γ(t)·|dR / dt|)·(1+ η(t)·ΔR(t)); where: Th(k,t) is the k-level warning threshold; Th o (k) is the initial threshold; γ(t) = γ o ·(1-exp(-α·||dR / dt||)) is the adjustment factor; η(t) is the compensation coefficient; ΔR(t) is the risk change; R(t) is the comprehensive risk index; k is the warning level; t is the time variable; γ o , α are control parameters.

[0051] Step S42, read the dynamic warning threshold set output by step S41 and the dynamic response characteristics output by step S3, obtain the historical statistical distribution of each response characteristic from the characteristic database; calculate the degree of deviation between the current response characteristic and the warning threshold; calculate the alarm weight coefficient of each characteristic according to the degree of deviation; substitute the weighted characteristic deviation into the discriminant function to obtain the warning discrimination value; read the prior probability distribution from the warning level database; calculate the probability density of the warning discrimination value under the conditions of each level; calculate the posterior probability distribution according to the Bayesian formula; select the level with the largest posterior probability as the current warning level; calculate the entropy value of the posterior probability distribution, obtain the level stability index, and output the warning level and level stability.

[0052] The specific process of warning level assessment is: L(t) = argmax_k{π_k·exp(-||W(t)-W_k|| 2 / σ 2 _k(t))}; where: L(t) is the warning level; π_k is the prior probability; W(t) is the current warning feature vector; W_k is the k-level standard feature vector; σ 2_k(t) is the dynamic variance parameter; k is the level number; t is the time variable; ||·|| is the Mahalanobis distance.

[0053] Step S43, receiving the warning level and level stability output by step S42, and the risk evolution trend output by step S3, and reading the optional alarm strategies, maintenance suggestions and response measures from the decision database; constructing a decision benefit evaluation function, substituting the warning level, level stability and risk evolution trend into the function to calculate the benefit value; calculating the timeliness index according to the availability and scheduling constraints of maintenance resources; estimating the cost index based on manpower, material resources and time consumption; substituting the benefit value, timeliness index and cost index into the objective function; constructing decision constraints, including resource limitations and operating procedure requirements; using the Lagrange multiplier method to solve the constrained optimization problem, and obtaining the optimal decision plan and optimization index.

[0054] The specific decision optimization process is as follows: J(d,t) = λ1·E(d,t) + λ2·T(d,t) + λ3·C(d,t); where: J(d,t) is the decision objective function; E(d,t) = β1·R(t)·S(t) + β2·dR / dt is the benefit function; T(d,t) = exp(-α·t_d) is the time function; C(d,t) = c o + c1·||d|| + c2·||d|| 2 is the cost function; d is the decision vector; R(t) is the risk index; S(t) is the level stability; t_d is the decision delay; λ_i, β_i, α, c_i are weight coefficients.

[0055] Step S44, import the optimal decision plan and optimization index output from step S43, obtain real-time feedback data of decision execution from the monitoring system; calculate the deviation between the actual execution effect and the optimal decision plan; construct a decision evaluation index, substitute the deviation into the evaluation function to calculate the evaluation score; read system parameters and evaluation history data from the knowledge base; calculate the knowledge update gradient according to the evaluation score; substitute the gradient information into the parameter update equation; obtain new system parameters through iterative calculation; store the updated system parameters and execution effect data into the knowledge base; output the updated knowledge base and optimized system parameters.

[0056] The parameter self-learning process is as follows: θ(t+1) = θ(t) - η(t)·grad J(θ(t))·exp(-μ·E(t)); where: θ(t) is the system parameter vector; η(t) is the adaptive learning rate; J(θ) is the performance objective function; E(t) =||d*(t) - d_real(t)|| 2is the execution error; grad is the gradient operator; d*(t) is the optimal decision; d_real(t) is the actual execution effect; μ is the attenuation coefficient; t is the time variable.

[0057] By establishing a dynamic warning threshold and a multi-objective decision-making optimization mechanism, the adaptability of the elevator roof bearing capacity warning and the optimization of decision-making are achieved. This step pioneered a dynamic threshold adjustment method based on the risk change rate, and combined historical data and real-time monitoring results to build a warning level assessment model. In the decision-making optimization link, the intelligent and precise warning decision is achieved by considering factors such as benefits, timeliness and costs through multi-objective trade-offs. The system also has self-learning capabilities, and can continuously optimize system parameters and knowledge bases according to the decision execution effect, thereby improving the adaptability and reliability of the warning system. This adaptive warning mechanism is particularly suitable for scenes such as elevator roofs with high safety requirements and complex working conditions. It can detect potential risks in a timely manner and provide the best disposal plan.

[0058] According to one aspect of the present application, step S11 is specifically:

[0059] Step S111: According to the preset sensor sampling initial frequency, collect raw data from the load sensor, vibration sensor, displacement sensor and temperature sensor respectively, and perform timestamp calibration on the data stream of each sensor; reorder the data stream according to the timing relationship of the timestamp; read the calibration curve of each sensor from the sensor parameter library, and convert the raw data into physical quantities; perform dimensional normalization processing on each data stream, and output the normalized multi-source sensor data stream.

[0060] Step S112, import the normalized data stream output by step S111, divide the data into data segments according to the preset time window; calculate the signal variance for each data segment to obtain the signal strength coefficient; use the multi-scale permutation entropy algorithm to calculate the noise level coefficient of the data segment; calculate the signal change rate based on the LZ complexity algorithm; multiply these three coefficients with the preset weight coefficients read from the weight database; perform feature space mapping on the weighted coefficients, and output the importance score value of each sensor.

[0061] Step S113, receiving the sensor importance score value outputted from step S112, and reading the reference sampling frequency from the sampling parameter library; constructing a frequency adjustment equation based on the Sigmoid function; substituting the importance score value into the frequency adjustment equation; introducing a dynamic compensation term, which is adaptively adjusted according to the score difference of adjacent sensors; obtaining the optimal sampling frequency of each sensor through iterative calculation; quantizing the sampling frequency to ensure that the hardware sampling capability constraint is met, and outputting the optimized sampling frequency of each sensor.

[0062] Step S114, read the optimized sampling frequency output by step S113, and re-collect the data of each sensor; apply anti-aliasing filtering based on wavelet transform to the newly collected data; interpolate and align the filtered data according to the minimum time interval; construct a time synchronization evaluation function to calculate the phase difference between data streams; use a dynamic time warping algorithm to eliminate the phase difference; perform linear interpolation on the aligned data to ensure that all data streams have a uniform time step; output the synchronized optimized data stream.

[0063] By building a multi-level sensor data acquisition and optimization mechanism, the intelligent and efficient data acquisition is achieved. This step first builds a sensor importance scoring function based on signal strength, noise level and signal change rate, and dynamically adjusts the sampling frequency to ensure that sufficiently dense data is obtained at critical moments. Then, quantization processing and anti-aliasing filtering technology are used to ensure the synchronization and integrity of the data. In the elevator top plate bearing capacity assessment scenario, this step is particularly important because it can adaptively adjust the data acquisition strategy according to the characteristics of different sensors and the importance of data. For example, when a sudden load change or abnormal vibration is detected, the system automatically increases the sampling frequency of the relevant sensors to ensure that key dynamic features are captured. The phase difference is eliminated through the dynamic time warping algorithm, and the precise alignment of multi-source data is achieved, providing a high-quality data foundation for subsequent feature extraction and state evaluation. This intelligent data acquisition mechanism significantly improves the system's response speed and data quality, while optimizing the efficiency of system resource utilization.

[0064] According to one aspect of the present application, step S14 is specifically:

[0065] Step S141, receiving a data stream with an abnormal mark, and reading a smoothing window length parameter in a signal analysis parameter library; dividing the data stream into multiple overlapping local time intervals based on the rate of change of the data; calculating the Fourier spectrum for the data in each interval; determining the main frequency component of each interval based on the spectrum analysis results; selecting a basis function combination corresponding to the main frequency component from a basis function library; orthogonalizing each basis function to form a complete local basis function set; and outputting the local interval division result and the corresponding basis function set.

[0066] Step S142, import the local interval division result and basis function set output by step S141, and read the error evaluation parameters from the data processing parameter library; calculate the projection coefficient of the non-abnormal data point in each local interval on the basis function set; construct a local reconstruction function based on the projection coefficient; substitute the reconstruction function into the original data to calculate the residual; use the Huber loss function to evaluate the residual; construct an optimization objective function based on the gradient information of the residual; iteratively optimize the projection coefficient through the L-BFGS algorithm; output the optimized reconstruction function parameters.

[0067] Step S143, read the reconstruction function parameters output by step S142, extract the time position information of the abnormal point from the abnormal database; calculate the theoretical value of the position of the abnormal point based on the reconstruction function; construct an adaptive weighting function, which takes into account the time distance and amplitude difference between the abnormal point and the adjacent normal point; apply the weighting function between the theoretical value and the actual value; use the Kalman filtering algorithm to correct the abnormal point; calculate the local consistency index of the corrected data; output the corrected and compensated data sequence.

[0068] Step S144, receiving the corrected and compensated data sequence outputted from step S143, calculating the overlapping parts of adjacent local intervals; constructing a transition function based on improved Hermite interpolation; applying the transition function to the overlapping intervals for data fusion; calculating the gradient continuity index of the fusion area; adaptively adjusting the transition function parameters according to the continuity index; performing final smoothing on the entire data sequence; verifying the overall smoothness and local fidelity of the processed data; and outputting the final processed data stream.

[0069] By constructing a data correction and compensation mechanism based on local reconstruction, the intelligent processing of abnormal data and the improvement of signal quality are achieved. This step innovatively introduces a local interval partitioning method based on Fourier spectrum analysis, and combines the Huber loss function and L-BFGS optimization algorithm to achieve accurate reconstruction of abnormal data. In the scenario of elevator top plate bearing capacity assessment, this method can effectively handle abnormal values ​​in various sensor data, such as data anomalies caused by load mutations, sensor failures, etc. By using the Kalman filter algorithm to correct abnormal points and combining the improved Hermite interpolation transition function for data fusion, the continuity and smoothness of the reconstructed data are guaranteed. The technical innovation of this step not only improves the reliability of the data, but also maintains the intrinsic characteristics of the signal, providing high-quality data support for subsequent feature extraction and state evaluation. Especially when processing data under high dynamic conditions, this method shows excellent robustness and adaptability.

[0070] According to one aspect of the present application, step S22 is specifically:

[0071] Step S221, import the time-frequency feature matrix, determine the number of modal basis functions according to the singular value distribution of the matrix; construct a generalized Gabor basis function that takes time-varying characteristics into consideration; use the Schmidt orthogonalization method to orthogonalize the basis function; calculate the time-frequency localization characteristic index of the basis function; optimize the basis function parameters based on the characteristic index; obtain the optimal basis function group through iterative calculation; and output the optimized modal basis function group.

[0072] Step S222, receiving the modal basis function group outputted from step S221, constructing a multi-resolution analysis framework based on wavelet packets; calculating the projection coefficients of the signal and the basis function at each resolution level; performing sparse decomposition using a matching pursuit algorithm; performing a time-frequency consistency test on the decomposition results; screening the main projection components according to the consistency test results; sorting the screened projection coefficients according to energy size; and outputting an ordered set of modal projection coefficients.

[0073] Step S223, read the modal projection coefficient set output by step S222, construct an instantaneous correlation analysis method based on Hilbert transform; calculate the instantaneous phase difference between each mode; use the mutual information algorithm to evaluate the nonlinear correlation between modes; construct a modal correlation matrix; calculate the eigenvalue decomposition of the correlation matrix; determine the number of independent modes according to the size of the eigenvalue; output the modal correlation evaluation result.

[0074] Step S224, import the modal correlation evaluation result output by step S223, and adopt a weight initialization method based on information entropy; construct a weight optimization objective function considering modal independence; apply the alternating direction multiplier method to solve the weight optimization problem; calculate the stability index of the optimized weight; fine-tune the weight according to the stability index; combine the adjusted weight with the corresponding mode; and output the load modal set.

[0075] By constructing a modal decomposition framework based on generalized Gabor basis functions and multi-resolution analysis, accurate extraction of load features and modal identification are achieved. This step first determines the number of modal basis functions based on singular value decomposition, and uses the Schmidt orthogonalization method to optimize the basis function characteristics. Then, sparse decomposition is performed using the matching pursuit algorithm, and instantaneous correlation analysis is performed in combination with the Hilbert transform, ultimately achieving accurate separation and feature extraction of load modes. In the scenario of elevator top plate bearing capacity assessment, this method can effectively identify and separate different types of load modes, such as static loads, dynamic impacts, and periodic vibrations. The reliability and stability of the modal decomposition results are ensured by a weight optimization strategy based on information entropy. The technical innovation of this step significantly improves the accuracy and resolution of load feature extraction, providing a reliable feature basis for subsequent state assessment.

[0076] According to one aspect of the present application, step S32 is specifically:

[0077] Step S321, receiving the state mapping matrix, reading the mass distribution data from the system parameter library; constructing the time-varying mass matrix using the Newmark integral format; constructing the nonlinear damping matrix based on the coupling relationship between velocity and displacement; reading the constitutive relationship data in the material parameter library; constructing the geometric stiffness matrix using the piecewise linearization method; introducing the influencing factors of ambient temperature and humidity into the stiffness calculation; correcting the singular values ​​of the stiffness matrix; and outputting the mass matrix, damping matrix and stiffness matrix of the system.

[0078] Step S322, read the system matrix output in step S321, construct the dynamic basis function based on the Krylov subspace method; use SVD decomposition to determine the main components of the basis function; construct a generalized basis function that takes into account high-order modes; calculate the orthogonality index of the basis function; use the Gram-Schmidt method to orthogonalize the basis function; evaluate the computational complexity of the basis function; optimize the number of basis functions according to the complexity and accuracy requirements; and output the optimized dynamic basis function group.

[0079] Step S323, import the dynamic basis function group output by step S322, construct a displacement interpolation function based on the variational principle; discretize the dynamic equations using the Galerkin method; construct an iterative solution format that takes nonlinear terms into account; design an adaptive time step control strategy; solve the nonlinear equation group using the prediction and correction method; calculate the residual vector for each time step; dynamically adjust the iteration parameters according to the size of the residual; and output time domain response data.

[0080] Step S324, receiving the time domain response data outputted from step S323, and using the Welch method to perform power spectrum estimation; constructing a multi-window time-frequency analysis framework; calculating the frequency response function of the system; extracting transient response characteristics using wavelet transform; resampling the response data in the frequency domain; respectively calculating the acceleration spectrum, velocity spectrum, and displacement spectrum; smoothing the response spectrum; and outputting three types of response spectra.

[0081] By constructing a framework for solving nonlinear dynamic equations, high-precision analysis of the dynamic response of the elevator roof is achieved. This step innovatively introduces a hybrid algorithm based on Newmark integral and Krylov subspace, and combines the Galerkin method to discretize the equation. By designing an adaptive time step control strategy and a prediction correction method, the solution accuracy and computational efficiency are significantly improved. In the elevator roof bearing capacity assessment scenario, this method can accurately capture the dynamic response characteristics of the structure under complex loads, including acceleration, velocity, and displacement responses. Especially when dealing with high-frequency dynamic loads and nonlinear responses, this method exhibits superior performance and stability. Through multi-window time-frequency analysis and wavelet transform, the transient response characteristics are accurately extracted, providing reliable dynamic characteristics for risk assessment.

[0082] According to one aspect of the present application, step S43 is specifically:

[0083] Step S431, receive warning level and level stability data, read the historical decision plan set from the decision knowledge base; use association rule mining algorithm to analyze the characteristics of historical plans; construct a multidimensional decision variable space; define the value range of decision variables based on fuzzy set theory; use Latin hypercube sampling method to generate candidate decision plans; evaluate the feasibility index of each plan; filter the decision space according to the feasibility index; output a set of feasible decision plans.

[0084] Step S432, import the feasible decision-making scheme set outputted in step S431, read the index calculation parameters from the evaluation index library; construct a benefit evaluation model based on fuzzy hierarchical analysis; use the DEA method to calculate the timeliness index of the scheme; estimate the cost of the scheme based on the life cycle theory; incorporate the risk evolution trend into the benefit calculation; use the entropy weight method to determine the index weight; calculate the comprehensive evaluation score of each scheme; and output the multi-objective evaluation index value.

[0085] Step S433, read the multi-objective evaluation index value outputted in step S432, and import resource restriction parameters from the constraint condition library; construct a resource occupancy model taking into account time series; use a constraint satisfaction algorithm to verify the feasibility of the scheme; establish a constraint processing model based on fuzzy optimization; calculate the degree of constraint violation; design a penalty function for constraint violation; screen the schemes according to the penalty value; and output a set of decision schemes that satisfy the constraints.

[0086] Step S434, receiving the set of decision solutions that meet the constraints output by step S433, constructing a solution ranking model based on the PROMETHEE method; using grey correlation analysis to calculate the pros and cons relationship between solutions; establishing a comprehensive optimization model that considers multiple objectives; designing a solution algorithm based on simulated annealing; calculating the Pareto optimality of each solution; selecting the final solution based on the optimality ranking; evaluating the robustness index of the solution; and outputting the optimal decision solution and optimization index.

[0087] By building a multi-objective decision optimization and execution evaluation mechanism, the intelligentization and optimization of early warning decisions are achieved. This step pioneered the introduction of a comprehensive evaluation model based on fuzzy hierarchical analysis and DEA methods, and combined it with a constraint satisfaction algorithm for decision optimization. In the elevator top plate bearing capacity assessment scenario, this method can comprehensively consider multiple objectives such as benefits, timeliness and cost to generate the optimal early warning decision plan. Through the PROMETHEE method and gray correlation analysis, the scientific sorting and optimization of decision plans are achieved. The technical innovation of this step not only improves the scientificity and reliability of decision-making, but also realizes the continuous optimization and performance improvement of the early warning system. By establishing a perfect feedback mechanism, the system can continuously optimize the decision model according to the actual execution effect, and improve the accuracy and adaptability of early warning decisions.

[0088] In the above embodiment, the specific implementation process of the relevant algorithm is as follows:

[0089] The LZ complexity algorithm is: LZ(x) = (c·log_b(n)) / n + λ(t)·H(x); where: LZ(x) is the LZ complexity value; c is the pattern count; n is the sequence length; b is the logarithm base; H(x) = -Σp_i·log(p_i) is the information entropy term; λ(t) is the adaptive weight; p_i is the probability of pattern i; x is the input sequence; t is the time variable.

[0090] The dynamic time warping algorithm is: DTW(X,Y) = min{Σw_k·d(i_k,j_k)·exp(-α·|i_k-j_k|)}·exp(-β·S(X,Y)); where: DTW(X,Y) is the time warping distance; X, Y are the sequences to be aligned; d(i,j) is the local distance; w_k is the path weight; i_k,j_k are the corresponding point indexes; S(X,Y) is the sequence similarity; α, β are control parameters.

[0091] The Huber loss function is: L(δ,r) = {0.5·r 2 / σ 2 (t), |r|≤δ(t); δ(t)·|r| / σ 2 (t)-0.5·δ 2 (t), |r|>δ(t)}·exp(-λ·E(t)); where: L(δ,r) is the loss function value; r is the residual; δ(t) is the adaptive threshold; σ 2 (t) is the dynamic variance; E(t) is the system energy; λ is the attenuation coefficient; t is the time variable.

[0092] The Kalman filtering algorithm is: x-(t+1) = F(t)·x-(t) + K(t)·[y(t)-H(t)·x-(t)]·exp(-α·tr(P(t))); where: x-(t) is the state estimate; F(t) is the state transfer matrix; K(t) is the gain matrix; y(t) is the observation vector; H(t) is the observation matrix; P(t) is the error covariance; tr(·) is the matrix trace; α is the adaptive coefficient.

[0093] Schmidt orthogonalization method is: q_k = v_k - Σ(β_j·<v_k,q_j> / ||q_j|| 2)·q_j·exp(-λ·κ(Q)); where: q_k is the orthogonalized vector; v_k is the original vector; β_j is the dynamic weight; <·,·> is the inner product; ||·|| is the norm; Q is the orthogonal basis matrix; κ(Q) is the condition number; λ is the stability parameter.

[0094] The matching pursuit algorithm is: MP(x,D) = Σα_k·φ_k·exp(-λ_k·||r_k|| 2 / E_k); where: MP(x,D) is the signal reconstruction result; x is the input signal; D is the overcomplete dictionary; α_k is the adaptive coefficient; φ_k is the selected atom; r_k is the residual of the kth iteration; E_k is the residual energy; λ_k is the dynamic weight; ||·|| is the norm.

[0095] The mutual information algorithm is: MI(X,Y) = Σp(x,y)·log(p(x,y) / (p(x)·p(y)))·exp(-α·T(X,Y)); where: MI(X,Y) is the mutual information value; p(x,y) is the joint probability; p(x), p(y) are the marginal probabilities; T(X,Y) is the time delay function; α is the attenuation coefficient; X,Y are the input sequences.

[0096] The alternating direction multiplication method is: L(x,y,λ) = f(x) + g(y) + <λ,Ax+By-c> + (ρ / 2)||Ax+By-c|| 2 ·exp(-μ·E(t)); where: L(x, y, λ) is the augmented Lagrangian function; f(x), g(y) are the objective functions; λ is the Lagrangian multiplier; ρ is the penalty parameter; A, B are the constraint matrices; c is the constant vector; E(t) is the system energy; μ is the control parameter.

[0097] The Newmark integral format is: u(t+Δt) = u(t) + Δt·v(t) + (Δt 2 / 2)·[(1-2β)·a(t) + 2β·a(t+Δt)]·exp(-γ·|Δa|); where: u is displacement; v is velocity; a is acceleration; β is integration parameter; Δt is time step; γ is stability parameter; Δa is acceleration increment.

[0098] The Krylov subspace method is: K_m(A,b) = span{b,Ab,A 2 b,...,A^(m-1)b}·exp(-λ·κ(A_m)); where: K_m is the m-dimensional Krylov subspace; A is the system matrix; b is the initial vector; κ(A_m) is the subspace condition number; λ is the adaptive parameter; m is the subspace dimension.

[0099] The Gram-Schmidt method is: q*_k = v_k - Σ(w_j·<v_k,q_j> )·q_j·exp(-α·||R||_F); where: q*_k is an orthogonal vector; v_k is an original vector; w_j is a dynamic weight; <·,·> is an inner product; q_j is an orthogonalized vector; R is an upper triangular matrix; ||·||_F is the Frobenius norm; α is a stability parameter.

[0100] The Galerkin method is: (Au,v) + α(t)·(∇u,∇v) + β(t)·(u,v) = (f,v)·exp(-λ·E(u)); where: A is the operator; u, v are trial functions; α(t), β(t) are adaptive coefficients; ∇ is the gradient operator; (·,·) is the inner product; f is the source term; E(u) is the energy functional; λ is the control parameter.

[0101] The Welch method is: P(ω) = (1 / K)·Σ|X_k(ω)| 2 ·W(ω)·exp(-α·V(ω)); where: P(ω) is the power spectrum estimate; X_k(ω) is the Fourier transform of the kth data segment; W(ω) is the adaptive window function; K is the number of data segments; V(ω) is the frequency variance; α is the smoothing parameter; ω is the frequency variable.

[0102] The association rule mining algorithm is: Conf(A→B) = sup(A∪B) / sup(A)·exp(λ·MI(A,B)); where: Conf(A→B) is the confidence; sup(·) is the support function; MI(A,B) is the mutual information measure; λ is the weight coefficient; A, B are item sets.

[0103] The Latin hypercube sampling method is: X(i,j) = (π(i,j) - U(i,j)) / n·exp(-β·C(X)); where: X(i,j) is the sample matrix; π(i,j) is the permutation matrix; U(i,j) is a uniform random number; n is the number of samples; C(X) is the correlation measure; β is the control parameter; i,j are indexes.

[0104] The DEA method is: E(k) = (Σu_i·y_i) / (Σv_j·x_j)·exp(-λ·D(k)); where: E(k) is the efficiency value of the kth decision unit; u_i is the output weight; y_i is the output index; v_j is the input weight; x_j is the input index; D(k) is the crowding function; λ is the penalty coefficient.

[0105] The constraint satisfaction algorithm is: S(x) = min{Σw_i·g_i(x) 2}·exp(-α·V(x)); where: S(x) is the satisfaction function; w_i is the constraint weight; g_i(x) is the constraint function; V(x) is the violation measure; α is the adaptive parameter; x is the decision vector.

[0106] Grey relational analysis is: γ(x0,x_i) = (1 / n)·Σ[(min_i min_k|x0(k)-x_i(k)| +ρ·max_i max_k|x0(k)-x_i(k)|) / (|x0(k)-x_i(k)| + ρ·max_i max_k|x0(k)-x_i(k)|)]·exp(-λ·D(x0,x_i)); where: γ(x0,x_i) is the degree of association; x0 is the reference sequence; x_i is the comparison sequence; ρ is the resolution coefficient; D(x0,x_i) is the distance function; λ is the attenuation parameter; n is the sequence length; k is the time index.

[0107] The permutation entropy algorithm is PE(x) = -Σp(π)·ln(p(π))·exp(-λ·C(x)); where: PE(x) is the permutation entropy; p(π) is the probability distribution of the permutation pattern π; C(x) is the sequence complexity; λ is the adaptive parameter; x is the input sequence.

[0108] The phase unwrapping algorithm is: Φ(ω) = ϕ(ω) + 2π·N(ω)·exp(-α·D(ω)); where: Φ(ω) is the unwrapping phase; ϕ(ω) is the wrapping phase; N(ω) is an integer function; D(ω) is a discontinuous measure; α is a smoothing coefficient; and ω is a frequency variable.

[0109] SVD decomposition is: A = U·Σ·V^T·exp(-μ·κ(A)); where: A is the original matrix; U, V are orthogonal matrices; Σ is the singular value matrix; κ(A) is the condition number; μ is the stability parameter.

[0110] The prediction and correction method is: x(t+Δt) = x_p + α(t)·(x_c - x_p)·exp(-β·||r||); where: x_p is the predicted value; x_c is the correction value; α(t) is the adaptive weight; r is the residual vector; β is the control parameter; and t is the time variable.

[0111] EMD decomposition is: x(t) = Σc_i(t)·exp(-λ_i·E_i(t)) + r_n(t); where: c_i(t) is the intrinsic mode function; E_i(t) is the modal energy; λ_i is the attenuation coefficient; r_n(t) is the residual; and t is the time variable.

[0112] Embodiment 2,

[0113] S1. Multi-source data collection and intelligent preprocessing

[0114] S11. Adaptive multi-source data acquisition

[0115] Get the multi-sensor raw data stream R(t), including: load sensor data: L(t); vibration sensor data: V(t); displacement sensor data: D(t); temperature sensor data: T(t);

[0116] 1) Construct sensor importance scoring function:

[0117] W(i,t) = α·S(i,t) + β·N(i,t) + γ·C(i,t)

[0118] Where: i is the sensor number; S(i,t) is the signal strength coefficient; N(i,t) is the noise level coefficient; C(i,t) is the signal change rate; α, β, γ are dynamic weight coefficients.

[0119] 2) Calculate the optimal sampling frequency: f(i,t) = f_base · exp(W(i,t)); where f_base is the base sampling frequency.

[0120] The data stream R'(t) after synchronization optimization is obtained.

[0121] S12. Nonlinear noise suppression

[0122] Read the synchronously optimized data stream R'(t), then:

[0123] 1) Decompose the signal into multi-dimensional components:

[0124] R'(t) = Σ[H_k(t)] + ε(t)

[0125] Where: H_k(t) is the k-dimensional harmonic component; ε(t) is the residual term.

[0126] 2) Construct an adaptive filter function:

[0127] F(ω,t) = 1 exp(-λ·|H(ω,t)| 2 / σ 2 (t))

[0128] Where: ω is the frequency component; H(ω,t) is the harmonic amplitude; σ 2 (t) is the local variance; λ is the adaptive coefficient;

[0129] The denoised data stream R''(t) is obtained.

[0130] S13. Intelligent anomaly detection

[0131] Read the denoised data stream R''(t), then,

[0132] 1) Constructing data manifold space:

[0133] M(t) = {x(t) ∈ R^n | φ(x(t)) = 0}

[0134] Where: φ(x) is the manifold constraint function; n is the data dimension.

[0135] 2) Calculate the topological eigenvector:

[0136] v(t) = grad(M(t)) / ||grad(M(t))||;

[0137] 3) Anomaly detection criteria:

[0138] A(t) = ||v(t) v̄(t)|| / σ_v(t)

[0139] Where: v̄(t) is the local average eigenvector; σ_v(t) is the standard deviation of the eigenvector;

[0140] Get the abnormal marked data stream R'''(t).

[0141] S14. Data reconstruction and compensation

[0142] Read the abnormal mark data stream R'''(t), then

[0143] 1) Construct a local reconstruction function:

[0144] Φ(t,x) = Σ[w_i(t)·B_i(x)];

[0145] Where: w_i(t) is the time-varying weight coefficient; B_i(x) is the basis function;

[0146] 2) Optimizing reconstruction error:

[0147] E(t) = ||R'''(t) Φ(t,x)|| 2 ;

[0148] 3) Update the compensation coefficient: w_i(t+1) = w_i(t) η·∂E / ∂w_i;

[0149] Get the final processed data stream R_final(t);

[0150] S2. Multidimensional feature extraction and load analysis

[0151] S21. Time-frequency feature extraction

[0152] Read the final data stream R_final(t) output by S1, then:

[0153] 1) Construct the generalized time-frequency transform operator:

[0154] Ψ(t,ω) = ∫R_final(τ)·K(t-τ,ω)dτ;

[0155] Where: K(t,ω) is the adaptive kernel function; K(t,ω) = g(t)·exp(jωt + h(t,ω)); g(t) is the time domain window function; h(t,ω) is the phase compensation function;

[0156] 2) Calculate the time-frequency feature matrix: TF(t,ω) = |Ψ(t,ω)| 2 exp(jφ(t,ω)); where φ(t,ω) is the phase information;

[0157] Get the time-frequency feature matrix TF_matrix;

[0158] S22. Load pattern decomposition

[0159] Read the time-frequency feature matrix TF_matrix, then:

[0160] 1) Construct load modal basis function:

[0161] B_k(t,ω) = α_k(t)·exp(jβ_k(ω));

[0162] Where: α_k(t) is the time-varying amplitude function; β_k(ω) is the frequency phase function; k is the mode number;

[0163] 2) Optimize the modal decomposition equation:

[0164] TF_matrix = Σ[λ_k·B_k(t,ω)] + ε(t,ω);

[0165] Where: λ_k is the modal weight coefficient; ε(t,ω) is the residual term;

[0166] 3) Calculate modal correlation:

[0167] C(i,j) =<B_i,B_j> / ||B_i||·||B_j||; get the load mode set LM = {(λ_k, B_k)};

[0168] S23. Load feature fusion

[0169] Read the load mode set LM;

[0170] 1) Construct a feature importance evaluation function:

[0171] I(k) = μ·E(k) + ν·V(k) + ξ·S(k);

[0172] Where: E(k) is the energy contribution rate; V(k) is the coefficient of variation; S(k) is the stability index; μ, ν, ξ are weight coefficients;

[0173] 2) Feature selection criteria:

[0174] F(k) = I(k)·exp(-ρ·C_max(k));

[0175] Where: C_max(k) is the maximum correlation coefficient; ρ is the penalty factor;

[0176] 3) Fusion feature vector:

[0177] X(t) = Σ[F(k)·λ_k·B_k(t)]; get the fused feature vector X(t).

[0178] S24. Dynamic load spectrum construction

[0179] Read the fused feature vector X(t).

[0180] 1) Construct load probability density function:

[0181] p(x,t) = Σ[w_i(t)·G(x|μ_i(t),Σ_i(t))];

[0182] Where: G(·) is the generalized Gaussian kernel function; μ_i(t) is the time-varying mean; Σ_i(t) is the covariance matrix; w_i(t) is the mixing weight;

[0183] 2) Calculate the dynamic load spectrum:

[0184] L(x,t) = -∂ln(p(x,t)) / ∂x;

[0185] 3) Construct reliability indicators:

[0186] R(t) = exp(-∫L(x,t)dx);

[0187] Obtained: dynamic load spectrum L(x,t); reliability index R(t);

[0188] S3. Multidimensional status assessment and risk analysis

[0189] S31. Multi-scale state mapping

[0190] Read dynamic load spectrum L(x,t); reliability index R(t);

[0191] 1) Constructing a multi-scale state space:

[0192] S(t) = {M(t), L(t), D(t)};

[0193] Where: M(t) is the macroscopic state vector (overall structural response); L(t) is the local state vector (stress distribution); D(t) is the microscopic state vector (material properties);

[0194] 2) Define the state transfer operator:

[0195] Γ(s1,s2) = exp(-||s1-s2|| 2 / θ(t));

[0196] Where: s1, s2 are state vectors; θ(t) is the adaptive bandwidth function; θ(t) = θ o ·exp(-η·|dR(t) / dt|).

[0197] 3) Calculate the state mapping matrix:

[0198] Q(t) = {q_ij(t)}; q_ij(t) = Γ(s_i,s_j)·ω(t); where ω(t) is a time-varying weight function;

[0199] Get the state mapping matrix Q(t);

[0200] S32. Nonlinear Dynamic Response Analysis

[0201] Read the state mapping matrix Q(t);

[0202] 1) Constructing nonlinear dynamic equations:

[0203] M(t)·x** + C(x,x*,t)·x* + K(x,t)·x = F(t); where: M(t) is the time-varying mass matrix; C(x,x*,t) is the nonlinear damping matrix; K(x,t) is the stiffness matrix; F(t) is the external load vector;

[0204] 2) Propose a new solver:

[0205] x(t+Δt) = x(t) + Σ[α_k·Φ_k(x,t)];

[0206] Where: Φ_k is the dynamic basis function; α_k is the optimization coefficient; based on the variational principle, solve: min J = ||M·x**+ C·x* + K·x F|| 2 ;

[0207] 3) Calculate the response characteristics:

[0208] ψ(t) = {A(t), V(t), D(t)}; where: A(t) is the acceleration response spectrum; V(t) is the velocity response spectrum; D(t) is the displacement response spectrum; the dynamic response characteristic ψ(t) is obtained;

[0209] S33. Risk Assessment

[0210] Read the dynamic response characteristic ψ(t);

[0211] 1) Constructing a risk assessment indicator system:

[0212] R(t) = {R_s(t), R_d(t), R_f(t)};

[0213] Among them: R_s(t) is the static risk index; R_d(t) is the dynamic risk index; R_f(t) is the fatigue risk index;

[0214] 2) Define the comprehensive risk assessment function:

[0215] Ω(t) = ξ1·R_s(t) + ξ2·R_d(t) + ξ3·R_f(t);

[0216] Where: ξᵢ = exp(H(R_i)) / Σ[exp(H(R_j))]; H(·) is the information entropy function;

[0217] 3) Constructing the risk evolution equation:

[0218] dΩ / dt = λ(t)·Ω + μ(t)·grad 2 Ω + f(ψ,t);

[0219] Where: λ(t) is the attenuation coefficient; μ(t) is the diffusion coefficient; f(ψ,t) is the source function;

[0220] Get the comprehensive risk index Ω(t); risk evolution trend dΩ / dt;

[0221] S34. Reliability prediction

[0222] Read the comprehensive risk index Ω(t); risk evolution trend dΩ / dt;

[0223] 1) Build a state prediction model:

[0224] Z(t+τ) = Λ(Z(t)) + Φ(Ω(t)) + ε(t);

[0225] Where: Z(t) is the system state vector; Λ(·) is the state transition function; Φ(·) is the risk impact function; ε(t) is the random disturbance term;

[0226] 2) Define the reliability evaluation function:

[0227] P(t,τ) = P{Z(t+τ) ∈ S_safe|Z(t),Ω(t)}; where S_safe is the safety domain;

[0228] 3) Calculate the failure probability:

[0229] P_f(t,τ) = 1 ∫∫p(z,ω)·I(z,ω)dzdω;

[0230] Where: p(z,ω) is the joint probability density; I(z,ω) is the indicator function;

[0231] Obtain the reliability prediction value P(t,τ); failure probability P_f(t,τ);

[0232] S4. Adaptive warning decision system

[0233] S41. Dynamic warning threshold construction

[0234] Read the reliability prediction value P(t,τ); failure probability P_f(t,τ); comprehensive risk index Ω(t);

[0235] 1) Construct a multi-level warning threshold function:

[0236] Th(k,t) = Th o (k)·exp(-γ(t)·Δr(t));

[0237] Where: k is the warning level (k=1,2,3,4); Th o (k) is the initial threshold vector; γ(t) is the time-varying adjustment factor; Δr(t) is the risk change rate; γ(t) = γ o (1-exp(-η·|dΩ / dt|));

[0238] 2) Define the threshold adaptive equation:

[0239] dTh / dt = α(t)·[Th_target Th(k,t)] + β(t)·d 2 Th / dt 2 ;

[0240] Where: α(t) is the asymptotic coefficient; β(t) is the inertia coefficient; Th_target is the target threshold;

[0241] 3) Calculate the threshold credibility:

[0242] C(k,t) = exp(-||Th(k,t) Th*(k,t)|| 2 / σ 2 (t)); where Th*(k,t) is the historical optimal threshold;

[0243] Get the dynamic warning threshold set Th(k,t); threshold credibility C(k,t);

[0244] S42. Warning level assessment

[0245] Read the dynamic warning threshold set Th(k,t); the dynamic response characteristic ψ(t) output by S3;

[0246] 1) Constructing early warning discriminant function:

[0247] W(t) = Σ[ω_i(t)·φ_i(ψ_i(t),Th_i(k,t))];

[0248] Among them: ω_i(t) is the dynamic weight coefficient; φ_i is the discriminant function; ψ_i is the response characteristic component;

[0249] 2) Define the level conversion rules:

[0250] L(t) = arg max_k{P(k|W(t))};

[0251] Where: P(k|W(t)) = π_k·f_k(W(t)) / Σ[π_j·f_j(W(t))]; π_k is the prior probability; f_k is the conditional density function;

[0252] 3) Calculation of grade stability index:

[0253] S(t) = 1 H(P(k|W(t))) / log(K); where H(·) is the information entropy function

[0254] Get the warning level L(t); level stability S(t);

[0255] S43. Decision making and optimization

[0256] Read the warning level L(t); level stability S(t); risk evolution trend dΩ / dt output by S3;

[0257] 1) Constructing the decision space:

[0258] D(t) = {d_i(t) | i=1,2,...,N};

[0259] Where d_i(t) is an optional decision vector: d_i(t) = [a_i(t), m_i(t), r_i(t)]; a_i(t) is the alarm strategy; m_i(t) is the maintenance suggestion; r_i(t) is the response measure;

[0260] 2) Define the decision optimization goal:

[0261] J(d,t) = λ1·E(d,t) + λ2·T(d,t) + λ3·C(d,t);

[0262] Where: E(d,t) is the benefit function; T(d,t) is the time function; C(d,t) is the cost function; λᵢ is the weight coefficient;

[0263] 3) Find the optimal decision:

[0264] d*(t) = arg min_d{J(d,t)};st G(d,t) ≤ 0; H(d,t) = 0;

[0265] Where: G(d,t) is an inequality constraint; H(d,t) is an equality constraint;

[0266] Get the optimal decision solution d*(t); decision optimization index J*(t);

[0267] S44. Feedback and Self-Learning

[0268] Read the optimal decision solution d*(t); decision optimization index J*(t); historical decision database D_hist;

[0269] 1) Construct a decision evaluation function:

[0270] E(t) = ||d*(t) d_real(t)|| 2 ; Where d_real(t) is the actual execution effect;

[0271] 2) Define knowledge update rules:

[0272] K(t+1) = K(t) + μ(t)·grad E(t);

[0273] Where: K(t) is the knowledge base state; μ(t) is the learning rate; grad E(t) is the gradient term;

[0274] 3) Optimize system parameters:

[0275] θ(t+1) = θ(t) η(t)·∂J* / ∂θ;

[0276] Where: θ(t) is the system parameter vector; η(t) is the optimization step size; J* is the optimal decision indicator;

[0277] Get the updated knowledge base K(t+1); the optimized system parameters θ(t+1).

[0278] Example 3, Comparative Example

[0279] 1. According to the SC200 / 200 construction elevator manual (100m installation height), the foundation bearing capacity is as follows:

[0280] (1) Total deadweight G = cage weight + outer cage weight + guide rail frame total weight + counterweight weight + load weight

[0281] =2×2000kg+1480kg+66×150kg+2×1000kg +2×2000kg

[0282] =4000kg+1480kg+9900kg+2000 kg +4000kg

[0283] =21380kg

[0284] =21380kg×0.0098 =209.6 KN

[0285] (2) Calculation of foundation bearing P:

[0286] P=2×G (considering the influence of dynamic load, deadweight error and wind load on the foundation, the coefficient n=2)

[0287] P = 2 × 209.6 = 419.2 KN

[0288] The foundation bearing capacity is 419.2 KN

[0289] The weight of the foundation concrete is: 5.9*3.8*0.3*25=168.15 KN

[0290] P total = 419.2 + 168.15 = 587.35 KN

[0291] The uniformly distributed load during construction is: 587.35 / (5.9*3.8)=26.2KN / ㎡;

[0292] 2. According to the SC200 / 200 construction elevator manual (100m installation height), the foundation bearing capacity is as follows:

[0293] (1) Total deadweight G = cage weight + outer cage weight + guide rail frame total weight + counterweight weight + load weight

[0294] =2×2000kg+1480kg+132×150kg+2×1000kg +2×2000kg

[0295] =4000kg+1480kg+19800kg+2000 kg +4000kg

[0296] =31280kg

[0297] =31280kg×0.0098 =306.6 KN

[0298] (2) Calculation of foundation bearing P:

[0299] P=2×G (considering the influence of dynamic load, deadweight error and wind load on the foundation, the coefficient n=2)

[0300] P = 2 × 306.6 = 613.2 KN

[0301] The foundation bearing capacity is 613.2 KN

[0302] The weight of the foundation concrete is: 5.9*3.8*0.3*25=168.15 KN

[0303] P total = 613.2 + 168.15 = 781.35 KN

[0304] The uniformly distributed load during construction is: 587.35 / (5.9*3.8)=34.9KN / ㎡;

[0305] The basement grid is designed as 8m×8m, the floor height is 5m, and the slab thickness is 250mm.

[0306] The preferred embodiments of the present invention are described in detail above; however, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all belong to the protection scope of the present invention.

Claims

1. A method for evaluating the bearing capacity of a passenger and freight elevator roof with an adaptive early warning function, characterized in that: The steps include: Step S1, collecting data streams of multiple sensors, including load sensor data, vibration sensor data, displacement sensor data and temperature sensor data, and preprocessing them to obtain a final processed data stream; Step S2, receiving the finally processed data stream, constructing a time-frequency window function according to the data time series, performing a convolution operation on the window function and the data stream, and calculating a time-frequency energy distribution matrix according to the extracted time-frequency features; Different load modes are separated from the time-frequency energy distribution matrix, the correlation coefficients between the modes are calculated, and the modal weight matrix is ​​constructed; the energy contribution rate of each load mode is evaluated, its coefficient of variation and stability index are calculated, and the feature importance score is generated. Based on the score, feature screening and fusion are performed to obtain the fused feature vector; according to The characteristic vectors are integrated to calculate the load probability density distribution, construct the dynamic load spectrum and establish the reliability index; Step S3, receiving the dynamic load spectrum and reliability index, inputting these data into the macro structure response analysis module, the local stress distribution analysis module and the material performance analysis module respectively, and calculating the state vectors of three levels; establishing a state transfer matrix according to the distance relationship between the state vectors of the three levels; inputting the state transfer matrix into the dynamic equation solver, and calculating the acceleration response spectrum, velocity response spectrum and displacement response spectrum of the system; Based on the response spectrum data, the static risk value, dynamic risk value and fatigue risk value are calculated respectively, and the weight coefficient is determined according to the information entropy of each risk indicator to generate a comprehensive risk indicator and risk evolution trend; Predict the future state distribution according to the risk evolution trend, calculate the probability of the system entering the failure state, and obtain the reliability prediction value and failure probability prediction value; Step S4, receiving the reliability prediction value, the failure probability prediction value and the comprehensive risk index, calculating the initial warning threshold according to the historical data, and combining the warning threshold with the risk change rate to construct a dynamic threshold function; The early warning discrimination value is calculated based on the dynamic threshold function and the dynamic response characteristics, and the early warning discrimination value is combined with the prior probability distribution to obtain the early warning level; the decision space is constructed according to the early warning level and the risk evolution trend, and the benefit value, timeliness index and cost index of each optional decision plan are calculated, and the optimal decision plan is obtained through multi-objective optimization; the execution effect of the optimal decision plan is compared with the expected goal, the evaluation error is calculated, and the knowledge base parameters and system parameters are updated according to the error.

2. According to the method for evaluating the load-bearing capacity of the top plate of a passenger and freight elevator with an adaptive early warning function as described in claim 1, It is characterized in that The step S1 is specifically as follows: Step S11: according to the preset reference sampling frequency, respectively obtain the load sensor data, vibration sensor data, displacement sensor data and temperature sensor data, calculate the signal strength, noise level and signal change rate of each sensor, multiply these parameters with the preset weight coefficient and sum them to obtain the sensor importance score, calculate the optimal sampling frequency based on the sensor importance score, resample the original data stream using the optimal sampling frequency, and obtain the synchronized optimized data stream; Step S12, receiving the synchronized optimized data stream outputted from step S11, dividing the data stream into a plurality of time windows according to the time series, applying the generalized harmonic basis function to the data in each time window to decompose, and obtaining a series of harmonic components and corresponding residual terms; calculating the variance value of each harmonic component in its local time window, and constructing an adaptive filtering function according to the variance value; applying the adaptive filtering function to each harmonic component respectively, performing threshold filtering on the residual term, and recombining the processed harmonic component and residual term to obtain a denoised data stream; Step S13, importing the denoised data stream obtained in step S12, establishing a multidimensional coordinate system, mapping the data points into the coordinate system to form a manifold space; calculating the Euclidean distance between adjacent data points, and constructing a local connection relationship matrix; calculating the local gradient vector of each point based on the connection relationship matrix; calculating the average eigenvector and the standard deviation of the eigenvector in the local area based on historical data; comparing the current gradient vector with the average eigenvector, and determining the abnormal point according to a preset standard deviation threshold; adding marking information to the identified abnormal point, and outputting the data stream with the abnormal mark; Step S14, read the data stream with abnormal marks output by step S13, divide the data sequence into multiple overlapping local intervals in chronological order; in each local interval, select a suitable combination of basis functions according to the distribution characteristics of non-abnormal points; calculate the weight coefficients of the basis functions by the least squares method, and establish a local reconstruction function; calculate the mean square error between the reconstruction function and the actual data; based on the gradient information of the mean square error, iteratively update the weight coefficient until the preset convergence condition is met; use the reconstruction function to correct and compensate the data of the abnormal points; smoothly connect the reconstruction results of all local intervals, and finally output the processed data stream.

3. According to the method for evaluating the load-bearing capacity of the passenger and freight elevator top plate with adaptive early warning function as described in claim 1, It is characterized in that The step S2 is specifically as follows: Step S21, importing the final processed data stream output from step S1, segmenting the data stream, and weighting each data segment using a preset time window function; constructing an adaptive kernel function, which includes a time domain window function and a phase compensation function; performing a convolution operation on the weighted data segment and the adaptive kernel function to calculate a time-frequency distribution matrix; respectively extracting amplitude information and phase information from the time-frequency distribution matrix; calculating the time-frequency energy distribution according to the amplitude information and the phase information to form a time-frequency feature matrix; Step S22, reading the time-frequency characteristic matrix outputted in step S21, constructing a time-varying amplitude function and a frequency-phase function, and combining these functions to form a load modal basis function; Calculate the projection coefficient of the time-frequency characteristic matrix on each basis function; sort the load modes according to the size of the projection coefficient and retain the main load modes; Calculate the inner product between the retained load modes to obtain the modal correlation coefficient matrix; determine the weight coefficient of each mode according to the correlation coefficient matrix and the modal energy value; combine the weight coefficient with the corresponding load mode and output the load mode set; Step S23, receiving the load mode set outputted in step S22, calculating the energy value, variance and time series stability index of each mode; multiplying the energy value, variance and stability index with a preset weight coefficient to obtain the importance score of each mode; Calculate the redundancy penalty factor based on the correlation coefficient between each mode; Combine the importance score with the redundancy penalty factor to determine the final feature selection criterion; The modalities are screened based on the selection criteria, and the retained modalities are weighted and combined according to the importance scores to generate a fusion feature vector; Step S24, importing the fused feature vector outputted from step S23, segmenting the feature vector within a preset time window; calculating statistical features for each segment of data, including a mean vector and a covariance matrix; constructing a multidimensional Gaussian mixture model based on the mean vector and the covariance matrix, and calculating the probability density distribution of the load; The probability density distribution function is differentiated to obtain the load change rate; a dynamic load spectrum is constructed according to the load change rate; the system reliability index is calculated based on the dynamic load spectrum, and the dynamic load spectrum and reliability index are output.

4. The method for evaluating the top plate bearing capacity of a passenger and freight elevator with an adaptive early warning function according to claim 1, It is characterized in that The step S3 is specifically as follows: Step S31, import the dynamic load spectrum and reliability index output from step S2, calculate the overall structural response vector through the finite element model, calculate the local stress distribution vector using the stress concentration factor, and generate the microscopic performance vector in combination with the material performance curve; calculate the Euclidean distance between adjacent state vectors, and construct a state transfer function with a time decay term; dynamically adjust the bandwidth parameter of the state transfer function according to the rate of change of the reliability index; apply the adjusted state transfer function to each state vector to generate a state transfer probability matrix; apply time weight to the state transfer probability matrix, and output a state mapping matrix; Step S32, reading the state mapping matrix outputted in step S31, and constructing a time-varying mass matrix according to the system mass distribution; Construct the damping matrix based on the nonlinear relationship between displacement and velocity; calculate the stiffness matrix by combining material properties and geometric features; Substitute these matrices into the dynamic equation; construct a set of orthogonal dynamic basis functions, and use the variational principle to calculate the combination coefficients of the basis functions; solve the dynamic equation through iterative optimization method to obtain time domain response data; perform Fourier transform on the time domain response data to obtain acceleration response spectrum, velocity response spectrum and displacement response spectrum respectively; Step S33, receiving the three response spectrum data outputted from step S32, calculating the static over-limit probability based on the extreme value distribution theory to obtain the static risk index; calculating the dynamic impact effect using the time integral of the response spectrum to obtain the dynamic risk index; calculating the fatigue damage accumulation value according to the number of cycles and amplitude distribution of the response spectrum to obtain the fatigue risk index; calculating the time series entropy value of the three risk indicators; determining the weight coefficient of each risk indicator according to the entropy value; substituting the weighted risk indicator into the risk evolution equation; By solving the risk evolution equation through numerical integration method, we can obtain the comprehensive risk index and risk evolution trend; Step S34, import the comprehensive risk index and risk evolution trend output from step S33, establish a prediction model describing the system state transition; construct a risk impact function based on the risk index; introduce random disturbance terms into the state prediction model to simulate the random fluctuation characteristics of the system; define the safety domain boundary conditions of the system; use the Monte Carlo method to generate a large number of state samples; count the proportion of samples in the safety domain to obtain the reliability prediction value; Calculate the probability of crossing the safety domain boundary for the first time in the sample and obtain the failure probability prediction value; output the reliability prediction value and failure probability prediction value.

5. According to the method for evaluating the top plate bearing capacity of a passenger and freight elevator with an adaptive early warning function as described in claim 1, It is characterized in that The step S4 is specifically as follows: Step S41, importing the reliability prediction value, failure probability prediction value and comprehensive risk index outputted in step S3, and reading the initial threshold value of each level from the warning level database; Calculate the time derivative of the comprehensive risk index to obtain the risk change rate; substitute the risk change rate into the exponential decay function to calculate the threshold adjustment factor; multiply the initial threshold by the adjustment factor to generate a dynamic warning threshold; Calculate the mean square error between the current warning threshold and the historical optimal threshold to obtain the threshold credibility; construct the constraint equation based on the threshold credibility and obtain the target threshold through numerical solution; The warning threshold is iteratively optimized using the gradient descent method until the convergence condition is met, and the dynamic warning threshold set and threshold credibility are output; Step S42, read the dynamic warning threshold set outputted by step S41 and the dynamic response characteristics outputted by step S3, obtain the historical statistical distribution of each response characteristic from the characteristic database; calculate the degree of deviation between the current response characteristic and the warning threshold; calculate the alarm weight coefficient of each characteristic according to the degree of deviation; substitute the weighted characteristic deviation into the discriminant function to obtain the warning discriminant value; read the prior probability distribution from the warning level database; calculate the probability density of the warning discriminant value under the conditions of each level; Calculate the posterior probability distribution according to the Bayesian formula; select the level with the largest posterior probability as the current warning level; Calculate the entropy value of the posterior probability distribution, obtain the level stability index, and output the warning level and level stability; Step S43, receiving the warning level and level stability outputted from step S42, and the risk evolution trend outputted from step S3, and reading optional alarm strategies, maintenance suggestions, and response measures from the decision database; Construct a decision-making benefit evaluation function, substitute the warning level, level stability and risk evolution trend into the function to calculate the benefit value; calculate the timeliness index based on the availability of maintenance resources and scheduling constraints; estimate the cost index based on manpower, material resources and time consumption; substitute the benefit value, timeliness index and cost index into the objective function; Construct decision constraints, including resource restrictions and operating procedure requirements; use the Lagrange multiplier method to solve the constrained optimization problem and obtain the optimal decision solution and optimization index; Step S44: import the optimal decision-making plan and optimization index outputted in step S43, and obtain real-time feedback data of decision execution from the monitoring system; Calculate the deviation between the actual execution effect and the optimal decision-making solution; construct a decision evaluation index, substitute the deviation into the evaluation function to calculate the evaluation score; read the system parameters and evaluation history data from the knowledge base; calculate the knowledge update gradient according to the evaluation score; substitute the gradient information into the parameter update equation; obtain new system parameters through iterative calculation; store the updated system parameters and execution effect data into the knowledge base; output the updated knowledge base and optimized system parameters.

6. The method for evaluating the load-bearing capacity of the passenger and freight elevator roof with adaptive early warning function according to claim 1, It is characterized in that The step S11 is specifically as follows: Step S111: according to the preset sensor sampling initial frequency, collect raw data from the load sensor, vibration sensor, displacement sensor and temperature sensor respectively, perform time stamp calibration on the data stream of each sensor; reorder the data stream according to the time sequence relationship of the timestamp; Read the calibration curve of each sensor from the sensor parameter library and convert the raw data into physical quantities; perform dimension normalization processing on each data stream and output the normalized multi-source sensor data stream; Step S112, import the normalized data stream output by step S111, divide the data into data segments according to the preset time window; calculate the signal variance for each data segment to obtain the signal strength coefficient; use the multi-scale permutation entropy algorithm to calculate the noise level coefficient of the data segment; calculate the signal change rate based on the LZ complexity algorithm; multiply the three coefficients by the preset weight coefficients read from the weight database; perform feature space mapping on the weighted coefficients, and output the importance score value of each sensor; Step S113, receiving the sensor importance score value outputted in step S112, and reading the reference sampling frequency from the sampling parameter library; constructing a frequency adjustment equation based on the Sigmoid function; substituting the importance score value into the frequency adjustment equation; introducing a dynamic compensation term, which is adaptively adjusted according to the score difference of adjacent sensors; The optimal sampling frequency of each sensor is obtained through iterative calculation; the sampling frequency is quantified to ensure that the hardware sampling capacity constraints are met, and the optimized sampling frequency of each sensor is output; Step S114, read the optimized sampling frequency output by step S113, and re-collect the data of each sensor; apply anti-aliasing filtering based on wavelet transform to the newly collected data; interpolate and align the filtered data according to the minimum time interval; construct a time synchronization evaluation function to calculate the phase difference between data streams; use a dynamic time warping algorithm to eliminate the phase difference; perform linear interpolation on the aligned data to ensure that all data streams have a uniform time step; output the synchronized optimized data stream.

7. The method for evaluating the load-bearing capacity of the passenger and freight elevator roof with adaptive early warning function according to claim 1, It is characterized in that The step S14 is specifically as follows: Step S141, receiving a data stream with an abnormal mark, reading a smoothing window length parameter in a signal analysis parameter library; dividing the data stream into a plurality of overlapping local time intervals based on the rate of change of the data; calculating the Fourier spectrum for the data in each interval; determining the main frequency component of each interval according to the spectrum analysis result; selecting a basis function combination corresponding to the main frequency component from a basis function library; orthogonalizing each basis function to form a complete local basis function set; outputting the local interval division result and the corresponding basis function set; Step S142, import the local interval partition result and basis function set outputted in step S141, and read the error evaluation parameters from the data processing parameter library; calculate the projection coefficients of the non-abnormal data points in each local interval on the basis function set; construct a local reconstruction function based on the projection coefficients; substitute the reconstruction function into the original data to calculate the residual; use the Huber loss function to evaluate the residual; construct an optimization objective function based on the gradient information of the residual; iteratively optimize the projection coefficients by the L-BFGS algorithm; and output the optimized reconstruction function parameters; Step S143, read the reconstruction function parameters outputted in step S142, extract the time position information of the abnormal point from the abnormal database; calculate the theoretical value of the position of the abnormal point based on the reconstruction function; construct an adaptive weighting function, which takes into account the time distance and amplitude difference between the abnormal point and the adjacent normal point; apply the weighting function between the theoretical value and the actual value; use the Kalman filtering algorithm to correct the abnormal point; calculate the local consistency index of the corrected data; and output the corrected and compensated data sequence; Step S144, receiving the corrected and compensated data sequence outputted from step S143, calculating the overlapping parts of adjacent local intervals; constructing a transition function based on improved Hermite interpolation; applying the transition function to the overlapping intervals for data fusion; calculating the gradient continuity index of the fusion area; adaptively adjusting the transition function parameters according to the continuity index; performing final smoothing on the entire data sequence; verifying the overall smoothness and local fidelity of the processed data; and outputting the final processed data stream.

8. The method for evaluating the top plate bearing capacity of a passenger and freight elevator with an adaptive early warning function according to claim 1, It is characterized in that The step S22 is specifically as follows: Step S221, importing the time-frequency characteristic matrix, determining the number of modal basis functions according to the singular value distribution of the matrix; constructing a generalized Gabor basis function considering the time-varying characteristics; orthogonalizing the basis function using the Schmidt orthogonalization method; calculating the time-frequency localization characteristic index of the basis function; optimizing the basis function parameters based on the characteristic index; obtaining the optimal basis function group through iterative calculation; and outputting the optimized modal basis function group; Step S222, receiving the modal basis function group outputted in step S221, constructing a multi-resolution analysis framework based on wavelet packets; calculating the projection coefficients of the signal and the basis function at each resolution level; performing sparse decomposition using a matching pursuit algorithm; and performing a time-frequency consistency test on the decomposition result; Filter the main projection components according to the consistency test results; sort the filtered projection coefficients according to energy size; output an ordered set of modal projection coefficients; Step S223, read the modal projection coefficient set outputted in step S222, construct an instantaneous correlation analysis method based on Hilbert transform; calculate the instantaneous phase difference between each mode; use the mutual information algorithm to evaluate the nonlinear correlation between modes; construct a modal correlation matrix; Calculate the eigenvalue decomposition of the correlation matrix; determine the number of independent modes based on the eigenvalue size; output the modal correlation evaluation results; Step S224, importing the modal correlation evaluation result outputted from step S223, adopting a weight initialization method based on information entropy; constructing a weight optimization objective function considering modal independence; applying the alternating direction multiplier method to solve the weight optimization problem; calculating the stability index of the optimized weight; and fine-tuning the weight according to the stability index; Combine the adjusted weights with the corresponding modes; output the load mode set.

9. The method for evaluating the top plate bearing capacity of a passenger and freight elevator with an adaptive early warning function according to claim 1, It is characterized in that The step S32 is specifically as follows: Step S321, receiving the state mapping matrix, reading the mass distribution data from the system parameter library; constructing the time-varying mass matrix using the Newmark integral format; constructing the nonlinear damping matrix based on the coupling relationship between velocity and displacement; reading the constitutive relationship data in the material parameter library; The geometric stiffness matrix is ​​constructed using piecewise linearization method; Introduce the influence factors of ambient temperature and humidity into the stiffness calculation; correct the singular values ​​of the stiffness matrix; output the mass matrix, damping matrix and stiffness matrix of the system; Step S322, read the system matrix output in step S321, construct a dynamic basis function based on the Krylov subspace method; use SVD decomposition to determine the main components of the basis function; construct a generalized basis function considering high-order modes; calculate the orthogonality index of the basis function; use the Gram-Schmidt method to orthogonalize the basis function; evaluate the computational complexity of the basis function; optimize the number of basis functions according to the complexity and accuracy requirements; output the optimized dynamic basis function group; Step S323, import the dynamic basis function group output from step S322, construct a displacement interpolation function based on the variational principle; discretize the dynamic equation using the Galerkin method; construct an iterative solution format that takes nonlinear terms into account; design an adaptive time step control strategy; and solve the nonlinear equation group using a predictive correction method; Calculate the residual vector for each time step; dynamically adjust the iteration parameters according to the residual size; Output time domain response data; Step S324, receiving the time domain response data outputted from step S323, and using the Welch method to perform power spectrum estimation; constructing a multi-window time-frequency analysis framework; calculating the frequency response function of the system; extracting transient response characteristics using wavelet transform; resampling the response data in the frequency domain; respectively calculating the acceleration spectrum, velocity spectrum, and displacement spectrum; smoothing the response spectrum; and outputting three types of response spectra.

10. The method for evaluating the top plate bearing capacity of a passenger and freight elevator with an adaptive early warning function according to claim 1, It is characterized in that The step S43 is specifically as follows: Step S431, receiving warning level and level stability data, reading a set of historical decision solutions from a decision knowledge base; and using an association rule mining algorithm to analyze the characteristics of historical solutions; Construct a multidimensional decision variable space; define the value range of decision variables based on fuzzy set theory; use Latin hypercube sampling method to generate candidate decision plans; evaluate the feasibility index of each plan; screen the decision space according to the feasibility index; Output a set of feasible decision solutions; Step S432: import the feasible decision solution set outputted in step S431, read the index calculation parameters from the evaluation index library; and construct a benefit evaluation model based on fuzzy hierarchical analysis; Use DEA method to calculate the timeliness index of the scheme; estimate the cost of the scheme based on the life cycle theory; incorporate the risk evolution trend into the benefit calculation; use the entropy weight method to determine the index weight; calculate the comprehensive evaluation score of each scheme; output the multi-objective evaluation index value; Step S433, read the multi-objective evaluation index value outputted in step S432, import resource restriction parameters from the constraint condition library; construct a resource occupancy model taking into account time series; use a constraint satisfaction algorithm to test the feasibility of the solution; establish a constraint processing model based on fuzzy optimization; calculate the degree of constraint violation; design a penalty function for constraint violation; screen the solutions according to the penalty value; and output a set of decision solutions that satisfy the constraints; Step S434, receiving the decision solution set satisfying the constraints outputted from step S433, constructing a solution ranking model based on the PROMETHEE method; using grey correlation analysis to calculate the pros and cons relationship between solutions; establishing a comprehensive optimization model considering multiple objectives; designing a solution algorithm based on simulated annealing; calculating the Pareto optimality of each solution; selecting the final solution according to the optimality ranking; and evaluating the robustness index of the solution; Output the best decision-making plan and optimization indicators.

Images