1308 results about "Benchmark data" patented technology

The invention discloses a fine-grained multi-modal prompt learning method based on a visual language pre-training model, and relates to the technical field of visual language pre-training models.The method comprises the steps that a multi-modal prompt learning vector used for learning visual features and text features is constructed, and a fine-grained multi-modal prompt learning model is constructed based on the fine-grained multi-modal prompt learning model; global differences, subtle distinguishing visual features and category difference text features among different visual categories are generated, and a multi-modal prompt learning vector is optimized based on a cosine similarity loss function. According to the invention, through combination of dual-granularity visual prompt and text prompt, the problems that in an existing prompt learning tuning method, cooperative reasoning cannot be carried out in a dual-mode prompt representation space, and fine distinguishing features are difficult to capture are effectively solved; and superior migration generalization ability and few-sample learning ability are verified on a plurality of main image identification reference data sets, and the performance is remarkably improved in downstream tasks.

Disclosed is a three-dimensional deformation field modeling method fusing global navigation satellite system (GNSS) / synthetic aperture radar interferometry (InSAR) observation data. The three-dimensional deformation field modeling method includes: five parts of solving an InSAR line-of-sight velocity field, unifying a spatial reference datum of the GNSS data and the InSAR data, verifying precision of a result of the InSAR data, solving a three-dimensional deformation field having high precision based on a least square method, and analyzing and verifying a result. According to the method, the InSAR data is processed according to a high-precision time series method, a long-wavelength error of the InSAR data is corrected on the basis of a difference of two data spatial reference datums, and the accuracy of the result is ensured through verification of internal coincidence precision.

The invention discloses an intelligent financial risk early warning method and system based on a management decision, and relates to the technical field of data intelligence, and the method comprises a multi-source heterogeneous data collection module which obtains enterprise financial data, supply chain data, market public opinion data and industry reference data in real time through an API interface, risk keywords are extracted from news, social media and policy documents through a natural language processing technology according to the market public opinion data; the streaming data processing engine is constructed based on an Apache Flink framework, performs windowing processing on the real-time data stream, calculates the dynamic fluctuation ratio of financial indexes by sliding a time window, and compares the dynamic fluctuation ratio with a preset industry risk threshold value; and a risk decision fusion model, a dynamic threshold adaptive module and a man-machine collaborative early warning terminal. According to the method, millisecond-level financial index fluctuation monitoring is realized through a streaming computing framework, a knowledge graph and a natural language processing technology are fused, a risk entity and a causal chain are extracted from unstructured data, and a multi-dimensional risk portrait is constructed.

The invention relates to a cable online operation fault positioning, monitoring and early warning method and system based on a neural network algorithm, and the method comprises the steps: collecting the operation environment and state data of a cable through a multi-source sensor, and generating an original monitoring data set through noise reduction and time sequence alignment; aiming at the problem of periodic distortion caused by seasonal fluctuation of environment temperature and humidity, performing dynamic reference compensation to generate a reference data set without seasonal influence; aiming at a coupling effect of a dynamic load and temperature drift, calculating insulation performance evaluation data of temperature compensation through load-insulation correlation mapping; aiming at the problem of disconnection of waterproof sealing monitoring and fault positioning, a sealing degradation grade is generated based on regional high-humidity detection; aiming at the defect of a fixed threshold strategy, generating a fault coordinate and a risk probability in combination with spatial positioning and probability analysis; and finally, self-adaptive updating of the parameter library is realized through credibility verification.

The invention discloses a fault diagnosis method and system based on a multi-scale adaptive hypergraph convolutional neural network, and belongs to the technical field of fault diagnosis. The method comprises the following steps: acquiring original vibration data of mechanical equipment, and carrying out data preprocessing; constructing a fault diagnosis model based on the multi-scale adaptive hypergraph convolutional neural network; and inputting the vibration data after data preprocessing into the constructed fault diagnosis model, and outputting a fault diagnosis result. According to the method, a vibration signal is decomposed into different frequency bands, features under various scales are extracted respectively, prior knowledge is fused, local topological graphs are constructed through various weighting measures, and a plurality of local topological graphs jointly form a spatial hypergraph to serve as input of a model; a dual-channel graph convolutional network is introduced to extract features, weights of different channels are adaptively distributed through an attention mechanism, and deep fusion of the features is realized. Experiments on two reference data sets show that the method can effectively improve the fault diagnosis accuracy, and has good stability and generalization.

The invention provides a self-adaptive adjustment Internet operation and maintenance strategy generation method and system, and relates to the technical field of Internet, and the method comprises the steps: 1, dynamically collecting the operation state data of a target operation and maintenance environment through a distributed sensor and an edge node, and generating a multi-dimensional time series data set; 2, performing space-time correlation analysis on the multi-dimensional time sequence data set, determining a reference data node, constructing a two-dimensional correlation structure, and generating a dynamic judgment interval; and step 3, respectively selecting monitoring sample sets in the inner domain and the outer domain of the dynamic judgment interval, generating a trajectory feature sequence according to the time evolution relationship of the sample sets, and calculating a dynamic correction coefficient based on the trajectory feature sequence. According to the method, the self-adaptive circulation control is formed by dynamically adjusting the threshold baseline, the judgment interval and the strategy generation rule, the accuracy and effectiveness of the internet operation and maintenance strategy are improved, and the stability of internet operation is enhanced.

The invention relates to the technical field of agricultural product storage management, in particular to an agricultural product storage intelligent management system based on big data, which comprises a multi-source data acquisition layer, an edge processing layer, a cloud processing layer and an execution control layer, compared with the prior art which only depends on manual recording of the picking date or simple visual sampling inspection for judging the freshness, the method has the advantages that the freshness key indexes such as the skin color degree and the texture structure of the agricultural product are synchronously and quantitatively extracted through the deep learning model in the warehousing link of the agricultural product; performing dual cross validation with the registration time to eliminate abnormal samples; the verified freshness reference data and parameters such as temperature, humidity, gas concentration and the like monitored in real time in a corresponding storage microenvironment are fused and input, and a self-adaptive shelf life prediction model based on a dynamic environmental stress factor is constructed; therefore, continuous and accurate quantification and early warning of decay processes of different types of agricultural products under various storage conditions are realized.

The invention relates to the technical field of process parameter analysis, and discloses a process parameter analysis method and system for a foaming material forming process and a storage medium, and the method comprises the following steps: carrying out parameter collection and data statistics on an MPPO material under a 160 DEG C standard process condition to obtain process parameter reference data; carrying out a load pressure parameter orthogonal test on the material to obtain a load pressure process parameter database; forming temperature gradient analysis is carried out, sampling testing is carried out at the interval of 5 DEG C from 160 DEG C to 190 DEG C, and a temperature control curve is obtained; performing coupling analysis and pressure dynamic compensation to obtain a forming process parameter combination; carrying out wet weight, dry weight and size detection on a product produced under the forming process parameter combination, and carrying out multi-dimensional parameter calculation to obtain a quality control index; and performing association verification and parameter correction on the quality control index and the forming process parameter combination, and outputting target process specification data. The method improves the control precision of the process parameters.

The invention relates to a fire prevention and control smart network method and system based on Beidou satellite navigation. The method comprises the following steps: firstly, acquiring real-time operation data of fire-fighting equipment in a building, and scanning a building structure through a three-dimensional imaging device to generate a three-dimensional digital model; secondly, fusing the real-time operation data, the three-dimensional digital model and preset environment parameters to form a fire prevention and control reference data set; thirdly, carrying out multi-dimensional data comparison on the reference data set based on a preset fire-fighting standard program, and generating an alarm signal with spatial positioning information; then, determining a fire-fighting facility failure area according to the alarm signal, and calculating a fire risk diffusion parameter; and finally, generating a grading early warning instruction according to the risk diffusion parameters, and constructing an electronic combat map containing the optimal rescue path. The method provides global and accurate basic data support for fire prevention and control, improves the timeliness and accuracy of early warning, optimizes the rescue path, and improves the fire disposal efficiency.

The invention discloses an ACM data analysis method and a data monitoring system, and belongs to the technical field of aircraft maintenance, and the method comprises the steps: constructing an ACM data set, setting a feature engineering processing method, carrying out coupling feature extraction through setting a dynamic window and calculating a dynamic correlation coefficient, dividing working conditions, eliminating fluctuation interference, and generating a structured time series data set; the method comprises the steps of obtaining ACM reference data of aircrafts of the same model, setting a reference fault judgment method, judging latest data in a structured time sequence data set, screening abnormal parameters, judging whether the ACM breaks down or not by combining an actual temperature adjusting effect, reducing single-parameter misjudgment, setting a coupling fault positioning method, diagnosing the faulty ACM, and distinguishing true faults and false anomalies. Meanwhile, a prediction scheduling method is set, fault prediction, health state prediction and service life evaluation are carried out on the fault-free ACM, the probability of sudden faults is reduced, and non-planned flight stopping is reduced.

The invention relates to the technical field of building simulation, and discloses a BIM-based special-shaped hyperboloid building site model generation method and system, and the method comprises the steps: collecting the multi-source heterogeneous point cloud data of a special-shaped hyperboloid building site to construct a three-dimensional feature tensor model, and recognizing the key curvature mutation point of the tensor model to construct a space reference topology network; based on the digital skeleton model, geometric contour features are analyzed, hyperboloid fitting is carried out, and an initial BIM model is obtained; performing multi-physics field coupling simulation on the initial model to obtain a target BIM model; extracting curvature distribution and structural stress parameters of the target BIM model to obtain basic model parameters, and constructing a building performance prediction model in combination with historical performance reference data; and inquiring a bearing capacity standard and a deformation threshold value, and analyzing the optimal curved surface parameter combination and a prestress applying strategy to obtain optimized simulation parameters, so as to simulate the special-shaped hyperboloid building by utilizing target model simulation. According to the method, the simulation precision of the special-shaped hyperboloid building can be improved.

The invention provides a distributed energy access-oriented intelligent electric meter multi-device collaborative optimization method and device, equipment and a medium, and the method comprises the steps: carrying out the interpolation of a unified electric meter, an inverter and energy storage data granularity, building a joint time error model, obtaining calibration reference data, and completing the multi-device time synchronization calibration; based on the reference data, acquiring operation data of three devices, and using a joint energy consumption model to distinguish electric meter hardware faults and multi-device collaborative anomalies to complete fault diagnosis; calculating local line loss by combining an inverter-to-ammeter line and electric quantity data, verifying residual electricity metering deviation, and tracing and correcting settlement data to complete line loss-bidirectional metering linkage verification; associating the fault result with the corresponding relation between the long-term statistical fault type of the metering data and the deviation rate and the line loss to establish a model to predict deviation trend early warning; and determining a multi-device cooperative abnormal source and transformer area load interference relation, calculating an interference coefficient, feeding back a scheduling system, and marking a line reconstruction priority. By adopting the method, the distributed energy consumption efficiency and the operation reliability of the power distribution network are improved.

The invention relates to the technical field of radio frequency identification, in particular to an online test method and system for an electronic tag. The method comprises the following steps: arranging a radio frequency signal transmitting device and a radio frequency signal receiving device on a production line, and placing a to-be-tested electronic tag at a test position; according to production line speed data and environmental parameter data; testing time sequence parameter data are generated according to the staying time of the electronic tag to be tested in the testing area according to the production line speed data; generating a test threshold according to the environmental parameter data to obtain test reference data; according to the test time sequence parameter data, radio frequency excitation signals of different frequency bands are sequentially sent to the electronic tag through the controller, response signals of the electronic tag under the corresponding frequency bands are collected, and response signal data of the electronic tag are obtained. According to the invention, a test time sequence and reference data are dynamically generated according to the production line speed and environmental parameters, diagnosis and correction optimization of multi-band response signals are carried out in real time, and a more accurate test result is provided.

The invention discloses a medical image segmentation method and system based on basic model assisted semi-supervised learning. In a medical image segmentation task, by introducing an innovative method of combining a visual basic model (SAM) and a U-Net model, the labeling precision and the segmentation effect are remarkably improved. Compared with an existing semi-supervised learning method, the method has the advantages that the problem of low-quality false labels is solved by introducing a reliability scoring mechanism and a foreground conflict processing strategy, and unreliable images are effectively utilized through a region exchange strategy. Finally, the invention provides a new solution for solving the problems of low false label quality and high data annotation cost in medical image segmentation, greatly improves the performance of medical image segmentation, and shows an excellent segmentation effect on a plurality of reference data sets.

The invention relates to the technical field of food processing, in particular to a system for predicting flavor formation and optimization decision in food processing based on machine learning, which comprises a sensing unit, a multi-source data acquisition and fusion module, a dynamic modeling module, an optimization decision module and an execution module which are in signal connection with one another, and the optimization decision module is used for receiving the flavor perception probability distribution data and the updated scoring reference data, solving a Pareto optimal solution set through a multi-target particle swarm optimization algorithm in combination with equipment physical constraint conditions, generating a candidate processing parameter scheme, inverting equipment control parameters for the candidate processing parameter scheme through a physical constraint neural network, and obtaining the flavor perception probability distribution data and the updated scoring reference data. And a final machining parameter adjusting instruction is generated and transmitted to the execution module. According to the method, through dynamic threshold modeling, a semantic-chemical attention mechanism and a time-space preference map dynamic correction technology, multi-source data and a multi-target optimization algorithm are fused, so that closed-loop accurate regulation and control of processing parameters are realized, and the flavor quality is improved.

The invention discloses a TCBL wind field high-resolution LES simulation method facing gray area parameterization, a medium and a program product. According to the method, a TCBL idealized initial gradient wind field is constructed based on a modified Rankine vortex model, controllability and stability of initial conditions are ensured by specifying key kinetic parameters, and thermodynamic profiles under different wind speed conditions are obtained based on synthetic analysis of tropical cyclone sounding data. And correcting the bottom of the boundary layer by adopting a quadratic and logarithmic function fitting method. Through high-resolution grid configuration and dynamic boundary conditions, secondary grid turbulence is processed in combination with an SGS-TKE closed model, and adaptability of the model to strong rotation and high wind shear environments is optimized by adopting a dynamic correction mechanism. Besides, the stability of thermodynamic variables is maintained through large-scale tendency correction, and high-resolution three-dimensional wind field and turbulence statistical reference data are output and used for verifying and improving a turbulence parameterization scheme of the TCBL under gray region resolution.

The invention discloses a metal grinding machine tool grinding wheel wear compensation type intelligent control system, and relates to the technical field of mechanical manufacturing automation, and the system comprises a contact type reference detection module which is used for obtaining the initial contour and reference wear loss of a grinding wheel through discrete point measurement in a non-machining state; the multi-parameter non-contact real-time detection module is used for collecting acoustic emission, vibration and spindle current signals in the grinding process through three types of sensors; the grinding environment sensing module collects the temperature of a grinding area and the cutting fluid concentration; the intelligent control unit receives reference data and an environment signal, performs deep learning on a wear loss prediction model after multi-source data fusion, performs periodic reference calibration, analyzes the dynamic wear loss, and synchronously transmits the data to the data storage and tracing module; the executing mechanism adjusts grinding wheel feed amount compensation according to the dynamic abrasion loss; and the data storage and tracing module stores full-process data in real time. According to the system, high-precision and real-time wear compensation is achieved, the machining precision and consistency are improved, and high-end manufacturing requirements are met.

The invention discloses a pipe liner stability analysis and health monitoring system for an urban water supply pipe network, and particularly relates to the field of water supply pipe network health monitoring, which comprises the following steps: a data acquisition module acquires three-axis vibration data and adds a timestamp and a topological label; the signal separation module extracts independent components of turbulence and structural vibration; the modal offset labeling module calculates an energy coupling ratio and identifies an offset modal; the stability attenuation evaluation module compares the reference data through the transfer function matrix to quantify the rigidity loss and converts the rigidity loss into a pipe liner stability attenuation rate; the synchronous association module analyzes the time sequence characteristics and marks synchronous association nodes; the pipe lining degradation traceability module constructs a traceability path tree and converts the traceability path tree into a decision result with geographic coordinates; the system can effectively distinguish fluid disturbance and structural abnormity, performs pipe lining degradation monitoring and traceability, realizes visual early warning, and ensures long-term stable operation of a water supply pipe network.

The invention discloses a face depth forgery detection method based on an enhanced double-branch fusion model. According to the method, a novel enhanced double-branch fusion model is adopted, so that the robustness and generalization ability of a CNN-based face depth forgery detector are improved. Firstly, RGB high-frequency noise in a face image is enhanced to extract abnormal features of the face image, and then reservation fusion is carried out. Specifically, a depth separable convolution module is used to improve model performance when image features are extracted. A selective kernel module is also used to adaptively extract more representative noise features by dynamically adjusting a convolution kernel. Besides, a multi-scale channel space attention fusion module is designed, and feature information of each part is effectively fused, so that overfitting of the model is reduced, and robustness and generalization ability of the model are enhanced. And finally, comprehensive evaluation on a plurality of reference data sets proves that the method is remarkably improved in the aspects of robustness and generalization.

The invention relates to the technical field of construction safety management, and discloses a construction engineering site construction monitoring management method and system, and the method comprises the steps: obtaining an original data set, carrying out the data preprocessing according to the original data set, obtaining a preprocessed feature data set, carrying out the sequential optimization of the preprocessed feature data set, and obtaining a feature data set; the method comprises the steps of optimizing a data uploading sequence to obtain an optimized data uploading sequence, performing time synchronization on the optimized data uploading sequence to obtain a unified time reference data set, performing spatial calibration on the unified time reference data set to obtain a spatially calibrated data set, and performing data clustering analysis on the spatially calibrated data set to obtain a data clustering result. And obtaining a region-level data deviation analysis result, and performing abnormal region identification according to the region-level data deviation analysis result to obtain an abnormal region report. According to the method, the abnormal region can be dynamically identified.

The invention provides an open vocabulary industrial defect detection method based on multi-modal prior prompt, which comprises the following steps: S1, collecting and sorting defect images of industrial products to be detected, and constructing a large industrial reference data set; s2, building a visual priori prompt pool, and injecting refined and fine-tuned priori into multi-scale industrial features; s3, designing a significance Gaussian distribution modeling mechanism, and capturing a complex spatial mode of position prior, so that fine-grained prior has better expression ability, and the generalization of prior is improved; s4, establishing a decoupling LoRA text encoder, and extracting an industrial semantic basis in the industrial text template through a hierarchical prompt template and a hierarchical decoupling LoRA mechanism; and S5, constructing a visual text fusion unit, and performing multi-modal fusion on text prompt embedding and visual features. According to the method, high-precision and strong-generalization defect detection is realized by fusing visual and text prior prompts, and the robustness and adaptability in a complex open environment are remarkably improved.

The embodiment of the invention discloses a double-truck-crane cooperative hoisting monitoring method and device, electronic equipment and a storage medium. The method comprises the steps that component parameters and operation environment parameters of hoisting operation are collected, a reference data model is constructed, and an action expected value and a safety control threshold value are set; state acquisition devices are arranged on the two truck cranes to acquire multi-source data in real time and transmit the multi-source data to a central control system so as to realize synchronous sensing of the two trucks; processing multi-source data based on a data fusion algorithm, and constructing a dynamic attitude model and a collaborative operation state model; a state prediction correction value in the model is compared with the reference data model in real time, deviation is judged, and safety early warning is triggered; and generating a feedback control instruction, dynamically adjusting the hoisting operation, correcting the coordination deviation, and realizing the coordination control of the hoisting posture. In this way, through multi-source data fusion and real-time feedback control, the safety and coordination of cooperative lifting operation of the double truck cranes are remarkably improved.

The invention relates to the technical field of image recognition and storage, and discloses a driving video recording method and system based on four-way monitoring. The method comprises the following steps: synchronously acquiring and processing the surrounding environment of a vehicle and the head posture of a driver through a video stream to obtain driving attention reference data; compressing and storing the sight blind area data according to the head posture features to obtain hierarchical video data; performing enhanced mapping on the graded video data to obtain an attention compensation image; performing threat identification and sorting on the compensation image to obtain threat level data; establishing a threat level index structure; and performing hierarchical storage on the data according to the attention coverage degree to obtain a storage data set. According to the method and the device, the technical problem that key scene data may be lost due to lack of consideration of driver attention distribution in a multi-channel video real-time recording process is solved.

The invention discloses a UKEY-based network submission method and system and a storage medium, and relates to the technical field of information security, and the method comprises the steps: receiving a login authentication request sent by a mechanism end; after verification, generating authentication information containing an organization number and a current timestamp; generating a session identifier, and integrating the session identifier, the organization number and the current timestamp to generate secret key negotiation reference data; encrypting the key negotiation reference data to obtain first encrypted data; receiving second encrypted data, and generating a symmetric encryption key according to the first encrypted data and the second encrypted data; receiving the encrypted data packet encrypted by the mechanism end by using the symmetric encryption key, and decrypting the encrypted data packet by using the symmetric encryption key to obtain submitted data; and after the validity of the submitted data is confirmed according to the session identifier, storing the submitted data into a preset database, and returning storage confirmation information to the mechanism end. By implementing the application, the security of network submission can be improved.

The invention discloses a construction method of a self-adaptive gated spectrum-space-graph collaborative fusion network. The construction method comprises the following steps: step 1, constructing a network overall architecture; step 2, spatial branching-hierarchical spatial feature modeling is carried out; step 3, spectrum branching-adaptive spectrum feature optimization; step 4, designing an adaptive gating fusion module AGFM; the invention provides a spectrum-space-graph collaborative fusion network, which is a double-flow architecture, and solves the problems of complex space-spectrum interactive modeling, spectrum redundancy and low calculation efficiency in hyperspectral image classification. According to the network, through integration of hierarchical spatial feature learning, adaptive spectrum optimization and a dynamic cross-modal fusion mechanism, complementary advantages of a graph attention mechanism, intelligent agent self-attention and data-driven spectrum modeling are effectively combined; experimental results on three reference data sets verify the advanced performance of the SGCFN, and ablation studies prove that each module has an irreplaceable effect on enhancing classification robustness.

The invention provides an exercise risk dynamic identification system for a cardiac patient, and relates to the technical field of data processing. A user side; the server comprises a reference data matching module which is used for receiving the pre-exercise type and the pre-exercise duration, and analyzing the index by combining the body detection information and the environment temperature to obtain a reference index interval; the reference data verification module sends the reference index interval to the medical node to obtain a verified index interval; the alarm end comprises a dynamic comparison module used for carrying out risk analysis on the detection value based on the approved index interval to obtain risk levels; and an early warning device. By means of the exercise risk monitoring method and device, the technical problems that due to the fact that traditional exercise risk monitoring equipment usually conducts monitoring and early warning based on a fixed threshold value, the phenomena of erroneous judgment and missed judgment frequently occur, and the accuracy and reliability of monitoring and early warning are poor can be solved, the target of accurate monitoring and early warning can be achieved, and the effect of effectively guaranteeing the exercise safety of a patient is achieved.

The invention provides a legal provision retrieval method and retrieval effect evaluation based on legal data embedding optimization, and is characterized in that the method comprises the following steps: obtaining a user query request, extracting related keywords by utilizing a semantic information extension prompt-based keyword method, constructing a legal query-legal provision fine adjustment data set training word embedding model, and establishing a legal query-legal provision fine adjustment data set training word embedding model; the model is used for constructing a vector database, the weight of retrieval results is dynamically adjusted through retrieval legal provisions combining sparse retrieval and dense retrieval so as to merge the results, and a reference data set is constructed to evaluate the retrieval effect. Compared with the prior art, the legal service system has rigorous words and a fixed structure, is suitable for being systematically stored in a knowledge base, effectively supports various application scenes such as legal consultation, case analysis and legal education, and further improves the efficiency and accuracy of legal services. The invention further provides a reference data set construction method for evaluating the retrieval effect, and the problem that effective evaluation standards are lacked in the field of legal provision retrieval is solved.

The invention discloses a PCB laser drilling deviation detection method and system, and the method comprises the following steps: obtaining original image data corresponding to a PCB laser drilling region, standard hole site coordinate data in a drilling design file, and real-time operation parameter data corresponding to laser drilling equipment; performing preprocessing to generate preprocessed image data, a standard coordinate data set and an equipment operation reference data set; extracting edge contour feature data of a drilling area based on the preprocessed image data, and generating an initial deviation vector of an actual drilling contour and a design contour in combination with the standard coordinate data set; constructing a drilling deviation influence factor matrix based on the equipment operation reference data set in combination with the initial deviation vector, and generating a weight coefficient of each influence factor for drilling deviation; and correcting the initial deviation vector based on the weight coefficient, judging whether deviation exists in the drill hole or not, and generating a drill hole deviation detection result. According to the invention, the drilling deviation detection precision of the PCB can be improved.

The invention provides an optimization method based on an industrial energy consumption big data model. The optimization method comprises the following steps: S1, acquiring a reference data set; s2, a target function with the purpose of minimizing the total energy consumption within the unit industrial production time is determined, and construction of an energy consumption prediction model embedding layer is completed; s3, enriching data background and influence factors through a real-time energy consumption data set Oj collected from a monitoring station j set in the current industrial production cycle, and realizing real-time calibration and display of the fine-tuned model by establishing a contribution matrix; and S4, forming an energy consumption point set. According to the method, the bag-of-words model is constructed by formulating the constraint conditions, so that the construction and output of the bag-of-words model can meet the constraint requirements of the optimization scheduling model, the obtained model input data is text data with constraint, and the subsequent energy consumption model training and prediction process not only considers the statistical characteristics of energy consumption data, but also considers the energy consumption data. And the computing power constraint in the prior art is reduced.

The invention discloses a financial knowledge graph reasoning method and system based on multilayer path semantic modeling. The method comprises the steps that a financial knowledge graph to be reasoned is collected; constructing a reasoning model based on multilayer path semantics; and completing the inference of the financial knowledge graph by using the inference model. According to the method, the confidence coefficient of the triple is comprehensively evaluated from two levels of path context semantics; in addition, the understanding ability of the pre-training language model for path semantics is remarkably enhanced. Multiple tests are carried out on a reference data set, and it is shown that the reasoning ability of the model can be effectively improved.