164 results about "Approximation algorithm" patented technology

The invention discloses a multi-modal fusion lithium iron phosphate battery thermal runaway early warning method and system, and the method comprises the steps: collecting multi-source heterogeneous data, such as temperature, voltage, gas concentration and shell strain pressure, in real time through deploying a heterogeneous sensor network, and carrying out the noise reduction and time sequence feature extraction through employing a sub-linear time low-rank approximation algorithm of a Hankel matrix; constructing a cross-modal feature association network by applying a secondary time algorithm of a maximum weight sparse subgraph problem, inputting a fused feature vector into a Bayesian network health degree evaluation model for probabilistic reasoning calculation to obtain a battery health degree score and a thermal runaway risk level, generating a graded early warning signal through multi-level early warning threshold comparison, and performing early warning on the battery health degree score and the thermal runaway risk level. And corresponding prevention and control suggestions are matched. The method solves the technical problems that single physical quantity monitoring is difficult to comprehensively reflect the complex change in the battery and the response delay of a centralized processing architecture causes the early warning lag, and achieves the timely capture and accurate early warning of the early weak characteristics of thermal runaway.

The invention provides a reliability-aware dependent task unloading method based on topology reconstruction in industrial internet edge computing, which comprises the following steps of: constructing a system model covering an edge cloud network platform, an industrial cloud platform and internet of things equipment, and establishing a transmission delay model and a reliability model; constructing a task unloading mathematical model with the maximum reliability level; the method comprises the following steps: modeling a micro-service dependency relationship into a weighted directed acyclic graph based on a network flow theory, and carrying out topology reconstruction on micro-services applied to the Internet of Things through a Ford-Fulkerson approximation algorithm to obtain a micro-service grouping structure formed by minimum cut division; the micro-service grouping structure serves as priori knowledge to be input into the deep Q network, the deep Q network is used for solving a task unloading mathematical model, a task unloading strategy is dynamically adjusted, resource allocation is optimized, and an optimal calculation unloading scheme in the industrial internet edge calculation environment is obtained. The micro-service deployment is optimized, the communication overhead is reduced, and the system reliability and the resource utilization rate are improved through real-time network state dynamic decision making.

The invention relates to the field of power system planning, in particular to a micro-grid boundary quantitative evaluation method and system for multi-dimensional analysis and dynamic verification. The method comprises the steps that power distribution network and micro-grid scheme data are acquired, and scene recognition and decoupling modeling are carried out after standardization processing; the method comprises the following steps: extracting end supply-preserving scene data, analyzing cost elements, and generating a critical cost threshold table through normalization processing and a threshold approximation algorithm; performing multi-dimensional parameter correlation analysis and integrated learning training based on the table, and constructing a multi-dimensional boundary index model; performing benefit matching calculation and green value accounting according to the result to generate an economic benefit decomposition structure; real-time streaming data processing and stability verification are combined to generate an economical efficiency boundary index set; and finally, generating a standardized evaluation file through matrix mapping and weight dynamic adjustment. According to the method, dynamic quantitative evaluation of the economy boundary of the micro-grid is realized, the capacity substitution benefit and the green value are effectively integrated, and an accurate basis is provided for planning decision.

The invention relates to a beam forming design method of an active RIS-assisted ISAC system, and belongs to the field of communication perception integration. A multi-user large-scale multiple-input single-output system model assisted by communication and perception integration and a system model based on an ISAC system assisted by an active RIS are constructed, and a beam forming vector and a reflection coefficient of the active RIS are respectively optimized by adopting an alternating optimization algorithm. According to the method, the propagation blockage problem is solved, the system performance is further improved, the multiplicative fading effect problem of the passive RIS is solved by using the active RIS, the weighted minimum mean square error algorithm, the continuous convex approximation algorithm and the positive semidefinite relaxation algorithm are combined, the system effectiveness is improved to the maximum extent, and the system performance is improved to the maximum extent. The system utility is the sum of the weighted sum rate of the system and the target detection power, the target detection capability is remarkably enhanced while the weighted sum rate of the user is improved, and a new solution is provided for performance optimization of the ISAC system.

The invention provides an SQL statement optimization suggestion generation method and device, a medium and electronic equipment, and is applied to the technical field of artificial intelligence. The method comprises the following steps: acquiring SQL log monitoring data, including an execution timestamp and a time consumption value of an SQL statement, dividing the data according to a preset time window, and constructing a to-be-processed data set; a median approximation algorithm is used for calculating the median of SQL execution time consumption, and then abnormal SQL statements in the running log are screened out. And for each abnormal SQL, reconstructing an execution context thereof, obtaining an abstract syntax tree and an execution plan, and substituting the abstract syntax tree and the execution plan into a preset cue word template to generate cue words. And finally, calling a large model to analyze the cue word so as to obtain an optimization suggestion for the abnormal SQL. By collecting and analyzing the SQL logs and combining the median algorithm and the large model technology, efficient SQL statement optimization is achieved.

The present invention provides a method for optimization of active and passive beamforming and signal reception configuration of a dual IRSs-assisted ISAC system. By jointly optimizing active beamforming at a base station, the reception of a sensing signal by the base station, and passive beamforming at an IRS, the achievable rate of communication users is maximized while ensuring that the signal-to-noise ratio of the sensing signal meets the minimum requirement. In the present invention, to solve the complex nonconvex problem generated, first, fractional programming is used to decouple an optimization problem, then, a successive convex approximation algorithm and an alternating direction method of multipliers are used to transform an intractable nonconvex problem into multiple tractable subproblems, and finally, an alternative optimization method is used to efficiently solve for a high-quality sub-optimal solution. Simulation results show that the provided solution has good convergence and effectiveness, and the solution can effectively improve the performance of IRS-assisted ISAC systems.

The invention discloses an unmanned ship dynamic obstacle avoidance control system and method based on stochastic nonlinear model predictive control, and adopts a control structure of a planning layer and a control layer to solve the problem of dynamic obstacle avoidance of an unmanned ship in a complex obstacle environment. The planning layer is based on a random nonlinear model predictive control technology, and firstly, random modeling is carried out on the unmanned ship and dynamic obstacles encountered by the unmanned ship in the sailing process; designing obstacle avoidance switching conditions according to the whole actual obstacle avoidance process; in an obstacle avoidance random nonlinear model prediction planning control algorithm, uncertainty propagation of a dynamic obstacle and an unmanned ship in a whole prediction time domain is evaluated through unscented transformation, then obstacle avoidance constraint is described in a chance constraint mode, a chance constraint expectation operator transformation algorithm and a unit jump function approximation algorithm are designed, and a chance constraint expectation operator transformation algorithm and a unit jump function approximation algorithm are designed. Opportunity constraints are converted into deterministic constraints, and then a nonlinear optimization problem is constructed for planning. And the control layer adopts a PI D algorithm to track a planning value so as to realize dynamic obstacle avoidance of the unmanned ship.

A method for the joint active and passive beamforming and received signal optimization in an ISAC system assisted by dual IRSs is provided. The method jointly optimizes active beamforming at Base Station (BS), reception of sensing signals at the Base Station (BS), and passive beamforming at IRSs, so as to maximize communication sum-rate of users while ensuring that SNR of sensing signals meets a minimum requirement. To address the complex non-convex optimization problem, the method first applies fractional programming to decouple problem, then adopts successive convex approximation algorithm and alternating direction method of multipliers to transform intractable non-convex problem into multiple tractable subproblems, and finally employs an alternating optimization method to efficiently acquire the high-quality suboptimal solutions. The simulation results demonstrate that disclosed scheme exhibits satisfactory convergence and effectiveness, and can significantly improve the performance of IRS-assisted ISAC systems.

The invention discloses an auxiliary decision-making method and system for regional coal market purchase, and belongs to the technical field of coal market purchase decision-making, and the method comprises the steps: building model input parameters through collecting the data of a fire coal demand plan of a power plant, the maximum supply of a supplier, a coal quality index, a purchase price and a transportation cost; constructing a mixed integer linear programming model taking the minimization of the total purchase cost as a target function, and setting a demand satisfaction constraint, a supply capability constraint and a coal quality standard reaching constraint; solving the mixed integer linear programming model by adopting a two-stage approximation algorithm; and converting the obtained approximate optimal solution into a visual purchasing scheme chart and report, and supporting a user to dynamically adjust the purchasing scheme through an interactive interface to generate a final purchasing decision. According to the invention, on the basis of comprehensively considering the coal quality, the purchase cost, the transportation mode and the supplier capability, the coal purchase plan is optimized, the cost minimization is realized, and the overall quality of coal purchase is improved.

The invention discloses a multi-mode online detection method and system for transformer oil, and relates to the technical field of online monitoring of transformer oil, and the method comprises the steps: deducing a dielectric constant based on a Maxwell equation and an electric displacement invariant principle in combination with a real-time voltage; segmenting, marking and sharpening a real-time visual image by combining a curvature prior smoothing approximation algorithm with a segmentation mark sharpening method; performing denoising enhancement on the real-time thermodynamic image by adopting a filtering mapping enhancement method to generate a real-time enhanced thermodynamic image, and deducing the temperature based on region screening and colorimetric mapping; deriving an acid value based on the gray value of the second target visual image in combination with a fluorescence standard curve, obtaining the actual height of an oil column in a third target visual image through an improved edge detection and scale scaling method, and calculating the change height of the oil column to reflect the viscosity; an aging index is generated through analysis of a logic regression device, and whether to adjust a monitoring interval or send an early warning prompt is dynamically decided based on an early warning decision mechanism, so that online self-adaptive detection and early warning of the transformer oil are realized.

The invention relates to the technical field of train operation optimization, and discloses an air braking efficiency uncertainty-oriented freight train operation optimization method, which comprises the following steps of: generating a state value function approximate value and a working condition selection set by adopting a constraint-based state value function table learning algorithm according to an initial state and a disturbance sampling value of a freight train; according to the state value function approximate value and the working condition selection set, a state value function table approximate algorithm based on operation time estimation is adopted to generate an operation time approximate value and an operation time estimation value; and according to the operation time approximate value and the operation time estimated value, a freight train operation sequence meeting the air braking constraint, the speed limit constraint and the operation time limit is generated by adopting an operation optimization algorithm meeting the time constraint. According to the method, the freight train operation sequence meeting various constraints can be calculated in a short time, meanwhile, the influence caused by uncertain disturbance is reduced, and the overall optimization effect of the calculation result is improved.

The invention belongs to the technical field of wireless optical communication, and particularly relates to a mobile edge computing task unloading and resource allocation method in wireless optical communication assisted by an unmanned aerial vehicle. According to the method, the data model, the queue model and the communication model are constructed, a task unloading strategy is optimized, and efficient and low-power-consumption edge computing resource allocation is achieved; an optimization technology is adopted to decompose a long-term optimization problem into sub-problems of each time slot, and the sub-problems are solved through a continuous convex approximation algorithm and a BFGS algorithm; finally, joint optimization of task unloading, computing resources, transmitting power, task return transmitting power, time slot scheduling and UAV trajectory is realized, and an optimization framework with more fine granularity in multi-dimensional allocation is formed. The characteristics of flexibility of the unmanned aerial vehicle and high bandwidth of wireless optical communication are utilized, the long-term throughput of the system is improved, the long-term power consumption of Internet of Things equipment is reduced, the high real-time performance of task processing is ensured, and the method is suitable for outdoor mobile Internet of Things application scenes with strict requirements for low power consumption and high real-time performance.

The invention discloses a digital twin high-voltage motor-based strong current operation state monitoring and maintenance system, which comprises the following steps: collecting current, temperature rise and vibration signals in motor operation, carrying out high-precision fitting processing on the signals by using a Chebyshev polynomial approximation algorithm, carrying out trend offset detection on a fitting result based on a CUSUM algorithm, and carrying out high-precision fitting processing on the fitting result based on a CUSUM algorithm; therefore, tiny abnormal changes are identified. And meanwhile, a prediction signal is generated in combination with a digital twinborn model, residual calculation is performed on the prediction signal and the fitting signal, fault features are further analyzed, and early warning information is output. Real-time sensing, trend judgment, fault prediction and intelligent early warning of the running state of the high-voltage motor are achieved, and the system is suitable for motor monitoring and maintenance in an industrial scene.

The invention belongs to the technical field of optimization design of concrete materials, and particularly relates to a multi-objective optimization method of a strain hardening alkali-activated concrete material, which comprises the following steps: selecting an experimental material, setting a plurality of alkali-activated mortar mixing ratios, and predicting the performance of alkali-activated mortar based on a response surface method; performing parameter analysis based on the response curved surface model, and constructing a response index prediction model; based on the response index prediction model, defining basic parameters; the response indexes are considered, alkali-activated mortar multi-objective optimization is carried out through an NSGA-II algorithm, and a Pareto leading edge is obtained; determining an optimal compromise solution set based on a Pareto leading edge; based on the optimal compromise solution set, determining the objective weight of each response index by adopting an entropy weight method; based on the determined objective weight of each response index, performing multi-target evaluation on the alkali-activated mortar mix proportion in the optimal compromise solution set by adopting an ideal approximation algorithm to obtain an optimized alkali-activated mortar mix proportion; and the performance optimization of the strain hardening alkali-activated concrete material is indirectly realized.

The invention discloses a two-stage Raman hyperspectral imaging method, which relates to the technical field of spectral signal analysis and comprises the following steps: S0, preprocessing Raman hyperspectral data to remove abnormal values; the method comprises the following steps: S1, taking a wave number where a target characteristic peak is located as a target wave number, and based on a neighborhood of the target wave number in Raman hyperspectral data, performing spectrum denoising on the Raman hyperspectral data by using an adaptive low-rank matrix approximation algorithm to obtain the Raman hyperspectral data after spectrum denoising; and S2, using an improved BM3D method based on a rotating block to carry out image denoising on a Raman image after Raman imaging is carried out on a target wave number in the Raman hyperspectral data after spectrum denoising. According to the method, the neighborhood of the target wave number is used as the target area, the adaptive low-rank matrix approximation algorithm is used for spectrum denoising, then the improved BM3D method based on the rotating block is used for image denoising, Raman hyperspectral fast and efficient imaging is achieved, and the method has the advantages of being good in denoising effect, high in speed and light in weight.

The invention discloses a federated learning method (FedDKA) based on difference cognition personalized knowledge absorption, and belongs to the technical field of federated learning. Aiming at the problems that a global model adaptive to each client is difficult to construct under Non-IID data in traditional federated learning, and an existing personalized federated learning method depends on the global model and neglects client difference knowledge and privacy protection and the like, the FedDKA formalizes an optimization target of a personalized model; and a near-regularization term is introduced to deal with the customer single drift problem caused by data heterogeneity, and the over-fitting risk is reduced. In addition, in order to reduce the trial and error cost during personalized exploration of local clients, the FedDKA innovatively designs a distance approximation algorithm, and the difference cognition between the clients is improved. In order to accurately realize personalized knowledge absorption, a personalized feature fusion algorithm is designed, and a client is promoted to realize highly personalized updating of a local model on the basis of differential cognition. The method aims to enhance the personalized performance of the local model in the personalized federated learning system.

The invention discloses a combined control method for an MMC grid-connected inverter, and the method comprises the steps: employing passive sliding mode control as a main structure of a current inner loop, introducing slope control on the basis, constructing slope passive sliding mode control, effectively suppressing the buffeting problem, and improving the system output precision; meanwhile, the model prediction control strategy is used in the nearest level approximation algorithm of the modulator, and the grid-connected electric energy quality is further improved. An MMC grid-connected inverter electromagnetic transient model is built by using a Simulink simulation platform, and simulation verification is performed on different working conditions such as normal operation and load abrupt change. According to the method, passive sliding mode control is used as a current inner loop control structure, slope control is introduced on the basis, and buffeting is suppressed through a simple and efficient method; moreover, model predictive control is introduced in a modulation link, and a composite control structure with high robustness and high precision is formed.

The invention discloses an earthquake risk evaluation method and system, and relates to the technical field of earthquake numerical simulation, and the method comprises the steps: receiving a three-dimensional geometric model of a grid fault, and setting an initial condition; constructing a simulation operation equation to solve a shear stress loading rate, a normal stress loading rate and a state variable, and constructing an ordinary differential equation system; based on the three-dimensional geometric model and the initial conditions, accelerated simulation operation of the shear stress loading rate, the normal stress loading rate and the state variables is achieved through the ordinary differential equation system, and the accelerated simulation operation method comprises the steps that accelerated simulation operation is achieved through the method of combining an H-matrix and MPI, comprising the steps that a layered matrix is obtained based on a Green function under the half-space condition in combination with an adaptive cross-approximation algorithm, the layered matrix comprises a normal stiffness H-matrix and a shear stiffness H-matrix, and MPI parallel operation matrix-vector multiplication is adopted. According to the invention, the hierarchical matrix is introduced and the MPI parallel computing framework is fused, so that the simulation period is greatly shortened.

The invention discloses a symbiotic radio network hybrid beamforming method and system based on assistance of a reconfigurable holographic surface (RHS), and the method comprises the steps: constructing a communication architecture with the RHS as a transmitting antenna, and eliminating dual path loss; a dual-mode signal demodulation strategy is designed, the strategy I adopts MMSE-SIC combined with MRC technology to realize demodulation of an environment signal, deletion of a received signal from an SN through SIC and demodulation of a backscattering signal one by one, and the strategy II adopts design of the environment signal and a transmission frame structure thereof, combination of zero-forcing beamforming and demodulation of the backscattering signal to avoid transmission of an environment signal demodulation error; the transmitting power is optimized through a gradient descent method, the digital beam forming non-convex constraint is processed through fractional programming, the RHS amplitude control matrix is optimized through a sequence parameter convex approximation algorithm, and efficient alternating optimization of the three is achieved. The method has the advantages that energy efficiency of a multi-user scene is improved, system power consumption is reduced, dynamic network topology is supported, high reliability is kept, and limitation of an existing single-user scheme is broken through.

The present invention provides a single-stage high-frequency isolated DC / AC inverter, comprising a primary-side bridge inverter circuit, a high-frequency transformer, a series resonant cavity and a secondary-side AC circuit which are connected in sequence. The inverter is modulated by a control circuit, and an inverter modulation method is set in the control circuit and comprises the steps of: determining an inner phase shift angle between primary-side bridge arms, an outer phase shift angle between each primary-side bridge arm and a secondary-side bridge arm, and switching frequencies of a primary-side switching transistor and a secondary-side switching transistor on the basis of a fundamental harmonic approximation algorithm, so as to modulate control variables of the inverter, wherein the control variables include a power transmission characteristic and a soft switching characteristic; and calculating precise current values at switching instants of the primary-side switching transistor to compensate the control variables on the basis of the precise current values.

The invention provides an Internet of Vehicles DNN collaborative reasoning method based on a rate splitting multiple access technology. The method comprises the following steps: S1, establishing an edge computing system model; s2, establishing a communication model; s3, based on a rate splitting multiple access strategy, a road side unit RSU divides the information sent to each group of vehicles into a public flow and a private flow, and distributes a precoding matrix for each group; s4, in combination with a given DNN partition strategy and an RSU computing resource allocation method, calculating the communication rate received by each vehicle by using a continuous convex approximation algorithm; s5, on the basis of the communication rate and RSU computing resources, constructing a directed acyclic graph model of a DNN reasoning task, expanding the model into a network flow graph, and performing task partitioning on the model by adopting a genetic algorithm to minimize reasoning time delay to obtain a corresponding edge-vehicle partition node set; s6, converting an optimization problem into a KKT condition, and solving an optimal allocation scheme of RSU computing resources; and S7, repeating the steps from S4 to S6, and carrying out iterative updating to obtain an optimal solution of the system.

The invention discloses a construction method and evaluation method based on a vehicle network interaction multi-dimensional evaluation model, and the method comprises the steps: automatically recognizing all influence factors in a region through a machine learning algorithm based on intelligent power grid big data and geographic information data, and determining region-level evaluation parameters; fusing the evaluation parameters by a multi-dimensional data fusion method based on deep learning to obtain a comprehensive evaluation index; then, the number of input variables is reduced through feature selection and dimension reduction technologies, meanwhile, an approximation algorithm is adopted to replace complex mathematical operation, and a lightweight assessment model is constructed; and constructing a vehicle network interaction multi-dimensional evaluation model based on the lightweight evaluation model in combination with resource value information and multi-scene requirements. According to the method, the functions of differentiated scheduling, peak clipping and valley filling, frequency adjustment and emergency response are realized, the problems of resource mismatching, scheduling lag and poor regional adaptability caused by traditional static assessment are solved, the power grid stability, the scheduling efficiency and the user participation enthusiasm are improved, and the application of the V2G technology is promoted.

The invention discloses a dynamic beam hopping and resource allocation method for NGSO satellite security communication, and belongs to the field of sixth-generation mobile communication security communication and wireless resource allocation. According to the method, significant non-uniformity of distribution of ground flow requirements in a time domain and a space domain is considered, and a satellite-ground security communication network dynamic beam hopping and resource allocation problem model is constructed by considering ground user flow requirements and composition elements and channel characteristics of an NGSO multi-beam satellite network. Through joint optimization of satellite hopping beam scheduling, power resource allocation and auxiliary interference unmanned aerial vehicle deployment strategies, service requirements of different users and ground eavesdropping environments are dynamically adapted, and the safety throughput and queue delay fairness of the system are improved. According to the method, mixed integer linear programming modeling is adopted, and a low-complexity approximation algorithm is combined, so that the method not only has optimality guarantee, but also can be deployed and operated in an actual satellite communication system, and thus unification of secure communication and efficient resource scheduling is realized.

This invention discloses a floating-point-fractional hybrid anchoring method and computational system for suppressing iteration errors, belonging to the field of high-precision numerical computation technology. It solves the technical problems of pseudo-chaos caused by the accumulation of iteration errors in pure floating-point calculations, the computational power explosion in pure high-precision fractional calculations, and the inability to simultaneously achieve both computational power and precision. This invention completes high-speed iterative computations of nonlinear systems using a general floating-point format. After each floating-point calculation, an optimal rational number approximation algorithm is used to convert the calculation result into a simplified rational number with a denominator not exceeding a preset threshold, eliminating the original errors introduced by floating-point truncation and rounding. The purified rational number is then converted back to floating-point format to participate in the next iteration, relying on a bounded rational number anchoring mechanism to progressively block the error amplification chain. The denominator threshold is set to 10¹²~10¹ based on the effective number of bits of the floating-point number. 5 This invention can adaptively adjust to various scenarios such as embedded terminals, supercomputing, and industrial control, and optimizes the continued fraction algorithm to achieve the globally optimal rational approximation under a given upper limit of the denominator. It requires no dedicated large number hardware and combines the advantages of high-speed, low-load floating-point operations with zero approximation error in rational numbers. It can be stably used in nonlinear computing scenarios that are sensitive to initial values ​​and require long-term stable iterations, such as chaotic simulation, vehicle trajectory prediction, industrial PID closed-loop control, 5G / 6G communication channel simulation, microscale meteorological simulation, and large-scale number theory zero-point verification. It effectively eliminates numerical trajectory drift, reduces equipment computing power consumption, and minimizes the workload of manual calibration and data screening, possessing extremely high engineering and academic application value.

A convolution operation apparatus according to a first aspect of the present invention comprises: an input unit that receives input data as an input; a constant generation circuit that generates a plurality of constants and includes one or more nodes corresponding to each of the plurality of constants; a memory capable of storing computer-executable instructions; and a processor that performs a convolution operation on the basis of the input data that has been input and the plurality of constants, by executing the instructions. Here, the plurality of constants are determined by approximating a plurality of predetermined reference constants to a similar value between each reference constant and each reference constant less than or equal to a predetermined threshold, and the plurality of reference constants may be values corresponding to a plurality of weights that are predetermined through training of a neural network.

This invention discloses a multi-clinical-staging disease auxiliary classification method and system based on federated learning, belonging to the fields of big data and medical technology. To improve the accuracy of the classification model and the security of privacy data, this invention collects case data to construct a case dataset. This dataset is input into a client for analysis and processing to obtain an optimized case dataset. The client trains an XGGridBoost model on the optimized dataset. The trained model parameters are then encrypted and compressed, and sent to a central server for decompression. The decompressed model parameters are then securely aggregated, and an approximate algorithm is used to determine the optimal split point. The central server sends the optimal split point to multiple clients. Upon receiving the split point, each client determines whether its local decision tree has reached its maximum depth and performs iterative training or outputs the trained model. This invention solves the problem of data silos and improves the security of privacy data.

The present invention discloses a method for managing partially shared inventory of spare parts based on horizontal transfer, comprising the following steps: real-time monitoring of the degradation status of equipment, triggering preventive or post-event maintenance according to the equipment degradation status; after maintenance is triggered, if the local warehouse has sufficient spare parts inventory, directly performing the maintenance; if the inventory is insufficient, initiating a horizontal transfer request to other warehouses; using a Markov model to describe a multi-warehouse system; and using a decomposition approximation algorithm to solve the Markov model to derive key performance indicators of the multi-warehouse system. Compared with the prior art, the present invention optimizes critical thresholds based on the derived key indicators to achieve the lowest-cost inventory sharing strategy, thereby providing an efficient and scalable solution for maintenance management of modern industrial equipment.

The application discloses a seismic risk evaluation method and system, relates to the technical field of seismic numerical simulation, and receives a three-dimensional geometric model of a gridded fault and sets initial conditions; a simulation operation equation is constructed to solve a shear stress loading rate, a normal stress loading rate and a state variable, and a common differential equation system is constructed; based on the three-dimensional geometric model and the initial conditions, the common differential equation system is used to realize accelerated simulation operation of the shear stress loading rate, the normal stress loading rate and the state variable, wherein the method of the accelerated simulation operation comprises: using a method combining an H-matrix and MPI to accelerate simulation operation, comprising obtaining a layered matrix based on a Green function under a half-space condition in combination with an adaptive cross approximation algorithm, comprising a normal stiffness H-matrix and a shear stiffness H-matrix, and using MPI parallel operation matrix-vector multiplication. The application introduces a hierarchical matrix and fuses an MPI parallel computing framework, so that the simulation period is greatly shortened.

This invention discloses an infrasound localization method for minor pipeline leaks, comprising: acquiring the original infrasound signal of the pipeline leak; performing variational mode decomposition on the original infrasound signal to obtain IMF components; based on the IMF components, using an improved generalized cross-correlation function to obtain the time difference between the infrasound reflection to the upstream and downstream ends of the pipeline; calculating the wave velocity of the infrasound according to the pipeline characteristics; and determining the location of the pipeline leak point based on the time difference and wave velocity. The improved generalized cross-correlation function method involves: constructing an analytical signal based on the IMF components; using an improved linear adaptive weighted local approximation algorithm to obtain the local optimal solution of the IMF components; and performing inverse attenuation dispersion compensation analysis based on the generalized cross-correlation function to obtain the time difference between the infrasound reflection to the upstream and downstream ends of the pipeline. This invention can accurately locate the pipeline leak point.

The application discloses an automatic planning method and device for a walking route of a land controllable seismic source, and belongs to the field of physical exploration. The application introduces a TSP idea, and finally obtains a seismic source vehicle driving route which is short in path, less in barrier crossing, less in driving restriction, and in line with the driving operation habit of a driver through steps of constructing a map, setting relevant parameters and weight ratios, and solving the walking route of the land controllable seismic source by using a TNH combination algorithm. The application combines the advantages of a two-approximation algorithm, a Heuristic algorithm and an NSGA_II algorithm to propose the TNH combination algorithm for solving the walking route of the land controllable seismic source, so that the automatic planning of the walking route of the controllable seismic source is realized, the operation efficiency is accelerated, the accuracy of operation results is improved, the calculated route is more in line with the demand of field construction, and the exploration operation is further developed intelligently. The application is applied to the field of physical exploration to plan the walking route of the controllable seismic source.