30 results about "Bayesian formulation" patented technology

The invention relates to a CVT error evaluation method and device, error evaluation equipment and a storage medium, and belongs to the technical field of mutual inductor error recognized.The CVT error evaluation method comprises the steps that a first data set is restored to a primary voltage level based on voltage proportionality coefficients of multiple CVTs to obtain a second data set, and a difference matrix is constructed based on the second data set; establishing a likelihood function based on the difference matrix, and converting the likelihood function into a posterior probability density function based on a Bayesian formula; extracting data corresponding to a preset number of CVTs from the second data set to construct a third data set, and determining a prior probability density based on the third data set; and based on a multi-chain MCMC algorithm fused with plant rhizome growth and a posterior probability density function, determining error values corresponding to a plurality of CVTs when the posterior probability density is maximum. According to the method, the accuracy of CVT error evaluation is effectively ensured, and meanwhile, the error evaluation efficiency is also improved.

The invention discloses a data analysis system and method for tracing an information management platform, and relates to the technical field of data analysis, and the method comprises the following steps: dividing full-process transportation nodes, and deploying a sensor collection environment; according to the difference of node types, different missing value filling strategies are adopted, and environmental parameters after interpolation are subjected to standardization processing; screening qualified batch data according to quality scores, constructing environmental parameter safety thresholds of each node to form a threshold pool, and iteratively updating the thresholds according to the qualified batch data; determining an effective monitoring time period by combining node circulation time, calculating a standard environmental parameter over-threshold proportion, and judging whether the node is abnormal according to a proportion threshold value; and constructing a node causal map, reversely positioning an abnormal source through a Bayesian formula, calculating a node influence weight, and screening key nodes for feedback optimization. The condition that in the prior art, effective traceability is difficult when refrigerated goods transportation is abnormal can be effectively improved.

The invention discloses an ambiguity deviation vector classification and probability decoupling collaborative protection level construction method and device, and belongs to the technical field of integrity monitoring. The method comprises the following steps: firstly, constructing a model of a positioning error under a fixed solution, then constructing an initial protection level model based on a Bayesian formula, then performing protection level parameter decoupling by utilizing an ambiguity deviation vector probability, classifying deviation vectors according to a modulus length, calculating a modulus length boundary value, and finally obtaining a protection level parameter; and finally, calculating a final protection level based on decoupling and classification results. According to the method, the influence of ambiguity error fixation on the protection level can be quantified, and the requirements of safety critical applications on reliability and interpretability are met.

The invention provides a data processing method and device based on clinical test data, in the application, a prior probability and a historical likelihood of a symptom corresponding to target clinical test data can be determined through a historical clinical database, the prior probability reflects a basic epidemic rate of an indication in a population, and the historical likelihood of the symptom corresponding to the target clinical test data can be determined through the historical clinical database. According to the historical likelihood, association rules between symptoms and indications are mined from historical cases, the association rules and the indications are dynamically updated through a Bayesian formula, an inference chain conforming to clinical logic is formed, then, a complex multi-feature joint probability estimation problem is converted into product calculation of single-feature statistics through conditional independent assumption, and a probability estimation result is obtained. The problem of calculation feasibility under high-dimensional data is solved, probabilistic output provides a quantitative basis for auxiliary determination of indications, a data-driven statistical rule is converted into a clinically understandable auxiliary support tool, and objectivity, consistency and scientificity of diagnosis decisions are effectively improved.

The present invention discloses a method for estimating formation pore pressure while drilling based on Bayesian theory, comprising the following steps: collecting target well logging data, while drilling logging data, and data from adjacent wells in the same block as the target well; predicting the target well's confidence-containing formation pressure interval profile based on the logging data and while drilling logging data, and determining the prior probability of formation pressure at any well depth; calculating the confidence-containing formation pressure interval profile of the adjacent well based on the adjacent well data, and determining the likelihood function of formation pressure at any well depth; selecting the prior probability and likelihood function of formation pressure at the same layer and substituting them into the Bayesian formula to obtain an updated posterior probability of formation pressure. The present invention realizes the while drilling update and correction of the confidence-containing formation pressure interval profile, thereby more accurately obtaining formation pressure parameters; providing more accurate formation pressure information for drilling risk assessment, reducing drilling risks caused by unclear understanding of formation pressure, and effectively improving drilling efficiency.

The invention belongs to the technical field of power system operation and planning, and discloses a power grid net load fluctuation scene generation method, system and device based on an ARIMA and Copula combined model and a medium, so as to solve the problem of poor scene generation accuracy. The method comprises the following steps: decomposing and reconstructing an original time sequence of the net load of the power grid by using discrete wavelet transform; taking permutation entropy minimization as an optimization target, adopting a variable chromosome length hybridization genetic algorithm to divide time segments for the low-frequency linear subsequences, and respectively establishing ARIMA models to generate linear trend scenes; establishing a joint probability distribution model of the high-frequency fluctuation subsequences at adjacent moments based on a Copula function, and deducing conditional probability distribution in combination with a Bayesian formula to generate a fluctuation scene; and the linear trend scene and the fluctuation scene are superposed to form an initial net load scene set, and a k-means clustering algorithm is adopted to reduce the initial net load scene set to obtain a representative scene set.

Disclosed is a natural language interaction-based method for extracting a parameter of an energy storage management system, belonging to the technical field of energy storage of new energy power systems. The method comprises parameter type division, natural language interaction corpus collection, parameter labeling, construction of a word segmentation set, natural language interaction corpus encoding, computation of a prior probability and a conditional probability, persistence of the prior probability and the conditional probability, new corpus input, loading of the prior probability and the conditional probability, computation of a posterior probability on the basis of a naive Bayes formula, and extraction of a parameter corresponding to the maximum value of the posterior probability.

This invention discloses an adaptive reclosing method, system, and electronic equipment for power transmission lines in winter based on evidence theory fusion. The method includes calculating the posterior probabilities of different subdivided fault causes under corresponding weather characteristics during winter power transmission line faults using a Bayesian formula model. These probabilities are then used as weights and incorporated into the fault cause identification and classification results output by a pre-constructed waveform image recognition model. Major fault categories are further subdivided to obtain the identification probabilities of different subdivided fault causes. The posterior probabilities and identification probabilities are fused using a D-S evidence theory fusion model to obtain the fused probabilities of different fault causes. Based on these probabilities, the reclosing operation mode, reclosing time, and number of reclosing operations are dynamically adjusted. Compared to existing technologies, this invention significantly improves the accuracy of fault cause identification and enables the reclosing system to dynamically adjust according to weather changes and fault analysis, providing effective decision support for adaptive reclosing of power transmission lines in winter.

The application discloses a kind of water supply system control methods based on flow self-distribution, specifically related to water supply system technical field, including by using historical record to construct the probability density function of monitoring feature, and by sensor determines the actual situation of current monitoring feature, based on bayes formula determines water supply probability information, and by cross entropy adjusts the weight of monitoring feature, by monitoring the steam flow of each position in boiler pipeline, based on historical data determines the influence of each position steam flow variation on water supply system water supply decision, by logistic regression model, determines water supply deviation information, by the comprehensive analysis of water supply probability information and water supply deviation information, quantifies the influence degree of each time point on water supply system decision, by comprehensively considering steam flow, water supply flow and drum water level, construct neural network model, further optimize the regulation and control of water supply system, the application is helpful to accurately adjust water supply rate, improve boiler operating efficiency and safety.

The invention discloses a new energy power generation prediction error analysis method and system considering meteorological conditions, and belongs to the technical field of computer data processing and prediction.The new energy power generation prediction error analysis method includes the steps that new energy power generation historical data, prediction error data and meteorological data are obtained and preprocessed, an initial data set is generated, and the statistical magnitude of prediction errors is calculated; combining the meteorological data in the initial data set, using a kernel density estimation method to estimate the joint probability density of the meteorological data and the prediction error data, generating joint probability density distribution, using a Bayesian formula to calculate the conditional probability distribution of the prediction error under a preset meteorological condition, and generating a conditional probability model; and performing multi-dimensional conditional probability modeling on the conditional probability model based on the climate and the position to generate a multi-dimensional conditional probability model. According to the method, kernel density estimation and the Bayesian theory are combined, and multi-dimensional space-time factors are fused to carry out refined modeling, so that the uncertainty of new energy power generation prediction can be accurately quantified, and prospective risk early warning can be realized.

The present application relates to a kind of CVT error evaluation method, device, error evaluation equipment and storage medium, belong to mutual inductor error identification technical field, wherein, the CVT error evaluation method includes: based on the voltage proportionality coefficient of multiple CVTs, first data set is restored to primary voltage level and obtains second data set, and difference matrix is constructed based on second data set;Based on difference matrix, likelihood function is constructed, and likelihood function is converted into posterior probability density function based on Bayes formula;From second data set, the data corresponding to the data of the extraction of the preset number of CVT is constructed third data set, and prior probability density is determined based on third data set;Based on the multi-chain MCMC algorithm of fusion plant rhizome growth and posterior probability density function, when the posterior probability density maximum, the error value corresponding to multiple CVTs is determined.The present application effectively guarantees the accuracy of CVT error evaluation, also improves the efficiency of error evaluation.

The invention discloses a single-base MIMO radar parameter joint estimation information amount calculation method, and the method comprises the steps: constructing a multi-dimensional joint probability density function of a uniform linear array through the property of an array receiving signal covariance matrix, and simplifying the joint estimation probability density function according to the vector form of a receiving signal; deducing a joint posterior probability density function of distance angle estimation under the condition of given received signals according to a Bayesian formula in combination with the joint probability density function, and obtaining distance direction information of multiple targets according to a distance-direction information formula; and deducing the distance direction information by using Taylor expansion under the condition of a high signal-to-noise ratio to obtain an upper bound of the distance direction information, and obtaining an entropy error, an entropy error lower bound and a Cramer-Rao bound according to the information amount and the information amount upper bound. According to the method, a posterior probability density function used for distance and angle joint estimation of multiple targets of the uniform linear array is deduced based on the information theory, and an information structure in radar observation data can be more completely described through joint distance-angle parameter modeling and mutual information theory analysis. Through introduction of mutual information indexes in an information theory, the perception capability of a radar system to target parameters can be quantified from a statistical perspective, and a theoretical basis is provided for system design.

The invention relates to a method and equipment for tracing and repairing water conservancy data quality abnormity, and a medium, and belongs to the technical field of intelligent water conservancy. Collecting multi-source water conservancy data, extracting data features and generating a standardized feature vector set; constructing a three-layer Bayesian network topological structure of circulation link-business influence factor-anomaly type, training to obtain a dynamic weight Bayesian network model adaptive to the water conservancy scene, inputting the standardized feature vector into the dynamic weight Bayesian network model obtained by training, outputting quality anomaly root cause probability distribution, and obtaining the quality anomaly root cause probability distribution. The posterior probability of each abnormal root cause is calculated through a Bayesian formula, and a root cause output traceability result is judged; performing graded repair and verification according to different traceability results; and the repaired qualified data is synchronized to a data asset module of the water conservancy credible data space for service output and generation of a visual data quality report. According to the method, the traceability precision is remarkably improved, and the service adaptability is higher.

The application discloses a hydrogen energy storage system electrolytic cell selection method for complex multiple working conditions, comprising the following steps: according to existing large number platform data statistics results, determining different renewable energy hydrogen production scene probability distributions of various regions, multiple type electrolytic cell use probabilities and multiple type electrolytic cell use probabilities under different scenes, taking the data as input (prior probability matrix) of an electrolytic cell selection decision system, and then calculating electrolytic cell use probability matrix (posterior probability matrix) under different application service working conditions through a Bayesian estimation method according to the known prior probability. According to the dynamic correction of the Bayesian formula, the use probability of different types of electrolytic cells under different hydrogen production scenes is corrected, the multi-subject matching characteristics of the renewable energy hydrogen production and energy storage system are effectively ensured, the distribution problem of maximum resource utilization and maximum economic benefit in the renewable energy hydrogen production system is solved, and the economic construction of the hydrogen energy storage system applied to multiple types of scenes in the future can be effectively guaranteed.

The application discloses a data cleaning method based on a Bayesian formula, a terminal and a storage medium, and the method comprises the following steps: acquiring original data and pre-defined prior knowledge; constructing a Bayesian network and an association relationship according to the prior knowledge, performing causal reasoning on the original data according to the Bayesian network, and obtaining a probability score of the Bayesian network; determining an association relationship score of the original data according to the association relationship, and cleaning the original data according to the sum of the probability score and the association relationship score, to obtain a cleaned data version. The application combines the user knowledge prior information which is easy to master, the modeling capability of the Bayesian network on dirty data and the association relationship of mutual information existing in the data, performs scanning and cleaning on the original data, reduces the difficulty of data cleaning, and improves the accuracy and recall rate of data cleaning.

The application discloses a kind of methods for measuring the sound reflection coefficient of material in acoustic tube, which comprises the following steps: firstly, the sample to be tested is placed in the test environment to emit and collect signals, and discrete Fourier transform is carried out to establish the sparse matrix form of sound receiving signal; then, the Bayesian formula containing hyperparameters is determined, and the hyperparameters are recursively solved to obtain the estimation result of multipath time delay; then, single-frequency PCW signal is emitted and collected, and the least square method is used to solve the amplitude parameter to obtain the estimation result of optimal multipath amplitude; finally, according to the estimation result of multipath time delay and the estimation result of multipath amplitude, the incident direct wave and the first reflected wave are extracted and separated, and the reflection coefficient of the sample to be tested is calculated. By using the sparse Bayesian learning method and the least square method, the application realizes the time delay estimation and amplitude estimation of multipath signal in acoustic tube, and achieves the purpose of separating multipath signal and calculating the reflection coefficient of acoustic material.

The invention discloses a rotary table multi-source data reliability prediction method based on the Bayesian theory, and belongs to the technical field of numerical control machine tools. The method comprises the following steps: constructing a mechanism reliability model based on geometric errors; according to the constructed mechanism reliability model, obtaining prior distribution and a joint prior probability density function matched with the overall Weibull reliability model under least square method line fitting; a turntable precision degradation test is carried out, and a likelihood function of life data is established according to an overall Weibull reliability model; substituting the likelihood function and the joint prior distribution probability density function into a Bayesian formula to obtain a posteriori distribution probability density function; and estimating parameters by adopting an MCMC sampling method, and carrying out reliability prediction analysis on the worm gear turntable under the condition of fusing different samples. According to the method, the prior information of the mechanism reliability model can be fully utilized under the condition that the service life sample size is small, and the reliability prediction precision in the operation process of the rotary table is effectively improved.

The invention provides a power transmission line fault reason identification method, system and device and a medium, and relates to the field of power systems, and the method comprises the steps: obtaining fault data and a to-be-identified fault, and processing the fault data and the to-be-identified fault to obtain discrete cause information; constructing posterior probability distribution of various fault causes under a given cause condition based on historical discrete cause information; fusing the posterior probability distribution into a preset Bayesian model to obtain a fault identification model; and different impedance angle information is acquired, and the fault identification model predicts and obtains a prediction category based on the impedance angle information. The method is used for identifying power transmission line faults caused by six reasons of lightning stroke, foreign matters, windage yaw, icing, pollution flashover and forest fire. Based on the Bayesian formula, the fault condition probability is calculated by using the fault cause information of the existing fault case and the impedance angle of the fault waveform, and the real fault cause is obtained.

The application discloses a local knowledge base RAG method based on Bayesian inference and belongs to the technical field of information retrieval. The method comprises the following steps: acquiring all knowledge data of a local knowledge base, performing paragraph classification and coding on the knowledge data to obtain a paragraph set, and calculating semantic probability of each paragraph in the paragraph set; acquiring a target professional vocabulary set of a target question, calculating semantic probability of each vocabulary in the target professional vocabulary set in each paragraph and semantic probability of all vocabularies in the target professional vocabulary set in each paragraph in the paragraph set; calculating conditional probability of each vocabulary in the target professional vocabulary set in each paragraph based on the frequency of occurrence; calculating conditional probability of the target professional vocabulary set in each paragraph by using a Bayesian formula and sorting to obtain a conditional probability sorting result; selecting a paragraph with a total number of words not exceeding a preset threshold as a target paragraph set; and generating target retrieval information based on the target paragraph set and the target question. The method improves the accuracy and reliability of question retrieval.

The invention discloses a single-base MIMO radar multi-target DOA estimation method, and belongs to the field of signal processing. According to the method, a Shannon information theory method is combined, and a joint conditional probability density function of a received signal of the MIMO radar, a target direction angle and a target reflection coefficient is deduced by utilizing the property of noise in the received signal; according to a Bayesian formula, firstly, a posterior probability density function of a single-base MIMO radar multi-target reflection coefficient based on a uniform linear array is derived, and a posterior probability density function of a multi-target direction angle when a signal is received is further obtained by combining a derivation result. The azimuth angles of a plurality of targets can be obtained at the same time by performing spectral peak search on the posterior probability density function of the target direction angle. The invention provides a DOA (direction of arrival) detection method based on a posterior probability density function and aiming at the condition that a plurality of targets are densely distributed. According to the method, the posterior probability density function used for uniform linear array monostatic MIMO radar multi-target DOA estimation is deduced based on the information theory, all available information is organically combined through the Bayesian theory, and therefore more comprehensive angle estimation is provided.

The present application relates to the technical field of data verification, especially relates to a kind of distributed photovoltaic irradiation data verification and correction method.The method first obtains grid-connected power station operating data, and the actual irradiance of each point is deduced based on photovoltaic module physical model;Subsequently, the reference point is screened from multiple dimensions, and the total deviation of measured and original simulated irradiance is decoupled into regional system deviation factor, local shading deviation factor and installation equipment deviation factor.On this basis, combined with historical meteorological statistics experience, the prior distribution of candidate point correction factor is established, the posterior mean and variance of each point correction factor are calculated by using Bayes formula to fuse measured likelihood information.Finally, the original data is calibrated using the expected value of correction factor, and the confidence interval under the preset confidence level is output.

This invention relates to the field of coating process monitoring and analysis technology, specifically to a coating process anomaly prediction system and method. The method includes: acquiring the core processes of coating production; establishing a data flow model between the core processes based on a multiple linear regression model; optimizing the regression coefficients of the model using Gaussian process regression combined with particle swarm optimization to determine the optimal regression coefficients; performing anomaly propagation path inference based on a Bayesian network path probability inference model; determining the edge weights of directed edges between nodes based on the calculation results of the improved Bayesian formula, and determining the influence coefficients between processes; processing the influence coefficients between processes, determining the predicted abnormal process for each process when an anomaly occurs, and sending notification information. This invention achieves accurate anomaly prediction and efficient notification, enhancing the accuracy and targeting of anomaly prediction, and is more suitable for the anomaly prevention and control needs of complex coating production processes.

The invention relates to a weather forecasting method, in particular to a rainfall level integrated forecasting method based on a Bayesian formula and an evidence theory. The invention aims to provide a rainfall level integrated forecasting method based on a Bayesian formula and an evidence theory, and provides a firm and reliable technical support for further improving the rainfall forecasting precision and promoting the application of the rainfall forecasting method in flood and drought disaster early warning and water resource scheduling. According to the technical scheme, the rainfall level integrated forecasting method based on the Bayesian formula and the evidence theory comprises the following steps of 1, rainfall forecasting and actually measured data collection; step 2, rainfall level determination; 3, determining and calculating Bayesian formula distribution; step 4, evidence fusion; and 5, integrating forecast evaluation.

The application discloses a multi-modal remote sensing image semantic segmentation method based on a discrete diffusion model, and relates to the field of remote sensing image semantic segmentation. The method first extracts isomorphic features of each mode, uses a parameter non-shared full connection layer to extract attention weights of the isomorphic features of each mode and performs weighting, adds the weighted mode features to obtain multi-modal features after adaptive fusion, then adds random noise on a real probability distribution matrix by using a state transition matrix to obtain a probability matrix in which probabilities of all categories are equal, derives inverse diffusion real distribution of the discrete diffusion model through a Bayesian formula, and finally takes the fused multi-modal features as a condition, takes the diffusion real probability matrix as input, and predicts the real probability matrix without diffusion. The method can effectively deal with the heterogeneity and information conflict problems among different modes, significantly improves the understanding ability of the model to complex environment, and enhances the generalization ability and robustness of the model in various scenes.

The small current grounding system single-phase grounding fault line selection method solves the problem of how to improve the line selection accuracy, and belongs to the field of small current fault line selection. The two power frequency period data before the fault are used as the reference, and the data after the fault are compared by difference, so that the inherent interference such as load imbalance, harmonic background and line parameter asymmetry during normal operation of the system is eliminated, and the fault characteristics are more prominent. Secondly, four criteria (differential steady-state phase, wavelet packet differential energy spectrum, differential instantaneous phase consistency and envelope weighted cumulative polarity) are used to describe the fault from different physical aspects. Each criterion independently outputs a 0~1 confidence level, forming a multi-dimensional complementary verification mechanism. Even if a certain criterion fails under certain conditions, other criteria can still provide reliable evidence. Finally, the prior probability and the four criterion likelihood values corrected by dynamic weight are fused through the Bayes formula to obtain the posterior probability of each branch and bus grounding, and the high and low threshold values are used for hierarchical decision-making.

The invention provides a method for calculating the contribution rate of maintenance regulations of a rail transit vehicle. The method comprises the following steps: S1, calling historical maintenance data; s2, extracting historical maintenance features; s3, obtaining a plurality of groups of maintenance characteristics; s4, obtaining fused maintenance features; s5, calculating a posterior probability; and S6, calculating the contribution rate. In the step S2, feature extraction is carried out on the historical maintenance data based on a BERT model. And S3, classifying the historical overhaul features by using a neural network. And S4, fusing the plurality of groups of maintenance features based on a gating mechanism in the neural network. And S5, calculating a posterior probability based on a Bayesian formula. And S6, calculating the contribution rate based on the regularized mahalanobis distance matrix. The invention further provides a system for calculating the contribution rate of the maintenance regulations of the rail transit vehicle. According to the method, deep mining is carried out on historical maintenance data, applicable scenes of different maintenance regulations can be systematically sorted, and the association strength of the maintenance regulations and fault types can be quantitatively evaluated.

The invention relates to an LDPC low-complexity scheduling decoding method based on network coding, and the method comprises the following steps: S1, obtaining a superposed signal of two users in a channel at a relay, calculating initial channel information according to a Bayesian formula, and initializing related parameters; s2, calculating the stability of a variable node through two adjacent decoding differences, and calculating the reliability of a check node and the variable node through a connection relation of a check matrix; s3, only allowing unreliable or unstable variable nodes and unreliable check nodes to perform information updating; and S4, calculating the posterior probability of the variable node, and carrying out decoding judgment according to PNC mapping to judge whether an iteration termination condition is met or not. The theory and simulation both prove that the complexity can be reduced by reasonably utilizing the current node state and FFT operation, the problem that short rings exist in LDPC short codes is analyzed, only subsets of part of nodes are updated, then a trap set is destroyed, and the decoding performance is improved.

The embodiment of the invention discloses a coal mine rock burst occurrence probability dynamic speculation method, and belongs to the technical field of coal mine danger prediction, and the method comprises the steps: carrying out the preprocessing of historical monitoring data of a target working face; using a comprehensive index method to calculate and quantify a danger index as the prior probability of Bayesian; dividing the target area into n sub-areas according to a preset step pitch; constructing a prediction system, wherein the prediction system comprises a prior layer and a likelihood layer; setting each monitoring layer of the likelihood layer as a 4-level discrete state, and determining a judgment threshold value of the 4-level discrete state; respectively judging the current discrete state of each monitoring layer of the likelihood layer according to the judgment threshold, and integrating the current states of all the monitoring layers; setting a probability transfer rule of an initial default dynamic state of each monitoring layer state of the likelihood layer; and calculating the posterior probability of rock burst occurrence in each monitoring state based on a Bayesian formula. According to the method, the engineering suitability, the dynamic flexibility and the field pertinence can be improved.

The small current grounding system single-phase grounding fault line selection method solves the problem of how to improve the line selection accuracy, and belongs to the field of small current fault line selection. The two power frequency period data before the fault are used as the reference, and the data after the fault are compared by difference, so that the inherent interference such as load imbalance, harmonic background and line parameter asymmetry during normal operation of the system is eliminated, and the fault characteristics are more prominent. Secondly, four criteria (differential steady-state phase, wavelet packet differential energy spectrum, differential instantaneous phase consistency and envelope weighted cumulative polarity) are used to describe the fault from different physical aspects. Each criterion independently outputs a 0~1 confidence level, forming a multi-dimensional complementary verification mechanism. Even if a certain criterion fails under certain conditions, other criteria can still provide reliable evidence. Finally, the prior probability and the four criterion likelihood values corrected by dynamic weight are fused through the Bayes formula to obtain the posterior probability of each branch and bus grounding, and the high and low threshold values are used for hierarchical decision-making.

The invention provides a multilevel resource management method and system for communication equipment, and relates to the technical field of power optical communication networks. A resource relation graph is constructed by obtaining multi-layer resource configuration data, nodes represent resource entities, and edges represent connection relations; fusing LLDP, equipment configuration, SNMP and manually recorded multi-source data, and calculating the existence probability of each edge in the resource relation graph through a Bayesian formula to represent the connection uncertainty; receiving a path generation request, converting the existence probability into an edge weight, and searching an optimal path as a candidate service path by adopting a Dijkstra algorithm integrated with engineering constraint pruning; and acquiring an actual network path through routing tracking, calculating a consistency index of the actual network path and the candidate path for verification, and dynamically updating the existence probability of the edge and the weight of the data source according to a verification result to form closed-loop optimization. According to the invention, the accuracy of service path planning and the adaptive capability of the system in a complex network environment are effectively improved.