691 results about "Resource consumption" patented technology

Systems for explainable large language model routing with immutable audit trails are disclosed. The system receives a query and determines its characteristics including complexity, domain, regulatory constraints, and performance requirements. It retrieves profiles for multiple LLMs from a model matrix containing performance attributes, resource consumption, and compliance parameters. The system selects a particular LLM by balancing resource consumption with performance requirements, evaluating regulatory compliance, ranking LLMs based on these factors, and prioritizing models with successful processing history. The system generates a human-readable explanation of the selection including decision factors, rationale, and alternatives considered. Finally, it records the selection and explanation in a tamper-evident, immutable audit trail data structure.

The invention provides a vehicle fault root cause diagnosis method, equipment, equipment and a medium, and the method comprises the steps: obtaining a vehicle target fault phenomenon, determining a to-be-detected fault event set based on a preset fault logic relation, constructing an initial scoring matrix, enabling a row vector to correspond to a fault event, enabling a column vector to correspond to the state feature parameters of a plurality of evaluation dimensions, and carrying out the detection of the fault event set; executing the detection task of the highest comprehensive score fault event, obtaining feedback data, updating the initial score matrix parameters based on the feedback data to obtain an updated score matrix, and calculating the comprehensive score of each event based on the updated score matrix again. Iteratively executing the detection task corresponding to the updated highest score event to identify whether the detection event is a fault root cause or not until the fault root cause of the target fault phenomenon is determined; according to the method, through a dynamic priority scheduling mechanism, real-time feedback data is fused in multi-dimensional evaluation, a high-value diagnosis task is executed preferentially, the resource consumption of traditional traversal diagnosis is remarkably reduced, and the fault positioning efficiency is effectively improved.

Systems and methods disclosed herein automatically detect, analyze, and mitigate anomalous resource distribution among artificial intelligence (AI)-based agents within a distributed computational network. The system receives a resource allocation request specifying computational resources, agent parameters, and performance objectives for a network of agents. A first AI model set monitors agent activity by tracking resource consumption and behavioral deviations from baseline profiles. The system compares resource usage of agents with historical norms and / or predetermined thresholds to generate an anomaly score for each agent. A second AI model set aggregates scores to construct a multi-dimensional data structure that indicates the comparison and anomaly score. The system ranks agents by anomaly severity to isolate agents with high scores (e.g., misaligned agents), and reallocates resources and updates access privileges for the misaligned agents.

The invention relates to the technical field of wind power, in particular to a dynamic response test method for a wind power variable pitch system. Comprising the steps of building a high-fidelity digital twinborn model, performing virtual testing and case optimization, performing physical self-adaptive testing execution, performing real-time analysis and decision-making driven by machine learning, dynamically adjusting testing stress, performing data feedback and model calibration and generating a comprehensive health report. Compared with the prior art which mainly depends on a fixed physical test process and artificial experience judgment and has the defects of low test efficiency, large resource consumption and difficulty in comprehensively covering extreme working conditions, the method innovatively introduces a high-fidelity digital twinning technology and a virtual test optimization process; according to the method, massive tests are executed in the virtual environment in a risk-free manner, and the enhanced test cases are intelligently generated, so that test forward movement and accurate optimization are realized, the pertinence and efficiency of physical tests are remarkably improved, and the test cost and period are greatly reduced.

The invention relates to the technical field of data analysis, and particularly provides a park enterprise space efficiency evaluation method and system based on a machine learning AI model, and the method comprises the steps: collecting space efficiency correlation parameters including a space utilization parameter, an economic energy efficiency parameter, an innovative output parameter, a shared resource consumption parameter and a collaborative correlation parameter; constructing a space efficiency evaluation model, extracting basic indexes and value-added indexes, training the model by taking industry characteristics as constraints, and outputting an evaluation result; generating spatial configuration schemes such as spatial layout optimization and resource allocation adaptation; and monitoring dynamic parameters in real time, and dynamically optimizing an evaluation result and a configuration scheme through a hierarchical correction strategy. According to the method, the problems of single index, static solidification, insufficient industry adaptation, closed-loop deficiency and the like in the prior art are solved, full-dimension accurate evaluation and dynamic optimization of space efficiency are realized, the feasibility of a configuration scheme and the refinement level of park operation are improved, and support is provided for intelligent park operation.

The invention relates to the technical field of data processing, and discloses a college budget management system fused with a new quality productivity theory, comprising: a state characterization interface collects process characterization data and maps the process characterization data into a discrete innovation state bit vector; the differential efficiency mapping unit responds to the state bit jump to calculate an efficiency marginal increment and generate an injection instruction; the physical bypass monitoring channel collects physical consumption time sequence data, and the entropy analysis closed-loop unit calculates a metabolic entropy representing the resource consumption order degree; the budget dynamic response control unit generates a state locking instruction when the state bit is constant and the metabolic entropy meets the low-entropy convergence condition, the physical working condition entropy is used for correcting the logic state, and the project resource state is maintained by recognizing the low-entropy resource consumption characteristics when no dominant output exists; the problem that discrete evaluation indexes cannot represent blind areas of continuous scientific research activities is solved.

The invention discloses a low-altitude economic intelligent scheduling method based on multi-dimensional data fusion and dynamic topological optimization, and belongs to the technical field of aircraft air traffic control, and the method comprises the following steps: S1, collecting and fusing multi-dimensional data, including meteorological data, equipment data, airspace data, task data and historical data, S2, constructing a dynamic airspace topological network based on the fused data, the topology network comprises a vertex set and an edge set, vertexes represent unmanned aerial vehicles, and edges represent cooperative paths between the unmanned aerial vehicles, S3, generating a global optimal scheduling strategy through a multi-objective optimization algorithm according to the dynamic airspace topology network, S4, executing the global optimal scheduling strategy, updating historical data and model parameters after a task is completed, and S5, executing the global optimal scheduling strategy according to the updated historical data and model parameters. And closed-loop feedback is formed. Multi-target dynamic balance of unmanned aerial vehicle dispatching safety, efficiency and energy consumption is achieved through multi-dimensional data fusion and dynamic topological optimization, the system supports full-process closed-loop automation, and the accident rate and resource consumption are remarkably reduced.

The invention relates to the field of low-altitude economy and geographic information, in particular to a low-altitude scene-based twinborn data rapid updating method, which comprises the following steps of: acquiring multi-source sensing data of a target low-altitude area in real time; comparing with a historical twinborn model, and identifying a change area and evaluating a priority by using an artificial intelligence algorithm; dynamically generating an optimal unmanned aerial vehicle data acquisition route by combining the airspace information and priority of the change area; then the unmanned aerial vehicle is controlled to collect updated data along the route; performing lightweight real-time three-dimensional reconstruction on the updated data to form a local updated model; and finally, fusing the model with a historical twinborn model to generate a final target twinborn model. Through intelligent change identification, dynamic route planning, lightweight reconstruction and incremental updating, the updating efficiency is significantly improved, the resource consumption is reduced, the real-time accuracy of the model is ensured, and the method is suitable for multiple fields such as low-altitude economy and urban governance.

The invention discloses an AI dynamic game virtualization resource scheduling system and method for a 5G low-delay service, and aims to solve the problems of insufficient QoS guarantee and low resource utilization rate caused by insufficient static rule adaptability, lack of service differentiation scheduling and low fusion degree of a game model and AI. A service feature extraction module is used for distinguishing a low-delay service (T = 0) and a common service (T = 1), constructing an AI dynamic game model taking the two types of services as participants, defining a payment function comprising a QoS satisfaction function and a resource consumption cost function, solving Nash equilibrium by adopting a deep reinforcement learning algorithm, outputting an optimal resource demand strategy, and realizing the optimal resource demand strategy. And the bandwidth and CPU allocation are dynamically adjusted through the virtualized resource management module. According to the invention, the delay compliance rate of the low-delay service can be improved by 20%-30%, the resource utilization rate of the common service can be improved by 15%-25%, the strategy iteration period is shortened to millisecond level, and differential fine scheduling and dynamic optimization in a 5G mixed service scene are realized.

The invention relates to the technical field of engineering simulation, and discloses a GPU heterogeneous resource scheduling method based on mixed execution of multiple simulation jobs, which considers the calculation characteristics of different simulation tasks, also introduces a calculation time prediction model and a video memory usage amount prediction model, and prejudges the resource consumption condition during job execution in advance. And accurate depiction of resource demands in a complex multi-physics field simulation scene is realized. When potential conflicts are found, a Monte Carlo tree search technology is utilized to explore a plurality of scheduling strategies, and a scheduling scheme which achieves optimal balance among calculation delay, throughput and energy consumption is screened out from the scheduling strategies. According to the method, historical data and real-time job requirements of a simulation platform are comprehensively analyzed, accurate prediction and dynamic allocation of GPU heterogeneous resources are achieved, the utilization rate and the overall performance of the GPU heterogeneous resources in a multi-simulation job mixed execution environment are effectively improved, and support can be provided for efficient resource scheduling in a complex simulation scene.

The invention provides a project design and optimization method and system and a storage medium, and the method comprises the steps: obtaining initial project information, and carrying out the cleaning, normalization and correlation processing of the initial project information, so as to obtain processed project information, which comprises processed historical project information and processed current project information; processing the processed historical project information based on a natural semantic algorithm and a fusion model to construct a knowledge graph; constructing a project performance prediction model by adopting a machine learning method based on the processed historical project information and the knowledge graph; obtaining a project demand, determining a constraint condition based on the project demand, and obtaining a candidate project based on the constraint condition, the project performance prediction model and the processed project information; a final item is determined based on the candidate items. The design efficiency can be improved, and the resource consumption can be reduced.

The invention discloses a subject corpus labeling method and system based on classifier multi-round iteration, and belongs to the field of natural language processing and machine learning. The method comprises the steps that a multi-level subject classification system is constructed, multiple rounds of iterative reasoning are conducted on a target subject based on a Fasttext classifier and a seed set, positive and negative samples are dynamically generated in each round of iteration, and classification accuracy is gradually improved; performing multi-dimensional scoring on the positive samples by using the large model to screen high-quality positive samples, and extracting keywords to optimize classification boundaries; a subject problem is generated through a WebQA method, and a low-recall-rate subject corpus is retrieved and supplemented; training an information filter to identify, filter and reject low-quality contents such as advertisements and garbage; and finally, efficient and accurate subject labeling is realized through a classifier and filter series connection process. According to the method, the problems of low efficiency, large resource consumption and subject understanding deviation of traditional labeling are solved, and the method is suitable for efficient and accurate subject labeling scenes of large-scale text data.

The invention relates to an electrolytic hydrogen production system operation strategy adjusting method and device, equipment and a storage medium. The method comprises the steps of predicting a node electricity price of a power system based on obtained load prediction data, renewable energy output prediction data, unit operation parameters, power grid topology data and a constructed electricity price prediction model, and obtaining an electricity price prediction result; based on the electricity price prediction result, the obtained hydrogen demand data and operation data of the electrolytic hydrogen production system, the constructed electrolytic hydrogen production system model and a preset electric energy balance constraint condition, a preset optimization objective function is solved with the goal of minimizing the hydrogen production cost, and the optimal hydrogen production cost is obtained. Obtaining a target start-stop plan and a target operation power plan of the electrolytic hydrogen production system, wherein the optimization target function is constructed based on the operation cost of the power system and the operation cost of the electrolytic hydrogen production system; and adjusting the operation strategy of the electrolytic hydrogen production system based on the target start-stop plan and the target operation power plan. The method is beneficial to reducing the consumption of hydrogen production resources.

The invention provides a distributed cooperative processing method and system based on event driving, and relates to the technical field of data processing, and the method comprises the steps: analyzing event stream data, screening the event stream data into time sequence features and semantic features, and fusing codes to construct event multi-dimensional feature vectors; constructing a load situation prediction model on the basis, and generating a comprehensive load descriptor; grouping the processing nodes according to the descriptors and reconstructing a resource topology; mapping to an optimization space to generate a scaling decision scheme; and constructing a risk assessment function to realize co-evolution of the prediction model and resource configuration. The system processing efficiency can be effectively improved, the resource consumption is reduced, and the load fluctuation adaptability is enhanced.

The embodiment of the invention provides a model training method, an information recommendation method, equipment, a storage medium and a program product. According to the embodiment of the invention, a multi-encoder-multi-sub-decoder-total decoder hybrid model architecture is provided, sample data of different information modes correspond to different encoders-sub-decoders, and a mode of processing all training sample data by a single model is converted into a divide-and-conquer mode. The internal complexity of each encoder-sub-decoder is relatively low, the complexity of model training can be reduced, resource consumption can be saved, and different encoder-sub-decoders can be trained in parallel, so that the model training time can be shortened; and furthermore, by utilizing a dual decoding mechanism of the sub-decoder and the global decoder, parameters of the encoder can be continuously adjusted through local optimization and global optimization, the performance of the model is optimized, the accuracy of a reasoning result is improved, the convergence speed of the model is accelerated, the model training efficiency is further improved, and the model training time is saved.

The invention discloses an environmental benefit and resource load evaluation method for an urban sewage treatment plant, and belongs to the field of sustainable evaluation. The implementation method comprises the following steps: training a random forest model by taking pollutant effluent concentration, energy consumption intensity and medicament consumption intensity of a national town sewage treatment plant as output variables and taking other historical operation information as input variables; on the basis, quantile regression is used for calculating quantiles of actual values of three operation indexes of each factory under the original processing condition; constructing an environmental benefit evaluation index system represented by pollutant removal efficiency and a resource load evaluation system represented by consumption efficiency of resources such as energy and chemicals; through obtaining environmental benefit and resource load grading benchmark values of the town sewage treatment plant, core functions and necessary input of each plant are accurately quantified, and real environmental contribution of the town sewage treatment industry is clarified, so that evaluation of sustainability of the town sewage treatment plant is more objective and fair, and sustainable development of urban environment infrastructures is facilitated.

The invention discloses a military strength resource scheduling method based on multi-objective optimization, which comprises the following steps: analyzing a strategy selection problem in a confrontation environment through a game theory model according to a resource demand matching degree and an enemy action prediction result, and if an enemy threat mode changes, triggering a self-adaptive adjustment mechanism specified by scheduling, dynamically correcting resource allocation strategy parameters by adopting a reinforcement learning algorithm to obtain an optimal scheduling rule adapted to the current confrontation situation; and calculating a resource consumption rate and supplement demand prediction according to an accurate resource allocation result, predicting a consumption trend of various resources in a future time period through a time sequence analysis algorithm, if the predicted consumption rate exceeds the supplement capability, triggering a resource allocation plan in advance, and determining an optimal resource reserve and supplement strategy by adopting an inventory optimization model. According to the method, the military strength resource scheduling efficiency and the combat effectiveness in a complex confrontation environment are effectively improved.

The invention discloses a gypsum dehydration control method and system based on multi-parameter collaborative optimization, belongs to the technical field of gypsum dehydration control, and solves the problem of how to improve gypsum dehydration efficiency under a dynamic working condition. Comprising the steps of collecting key operation parameters, a parallel operation mechanism model and a data driving model in the gypsum dehydration process in real time, constructing a weighted multi-objective optimization function based on constraint conditions, performing real-time optimization by adopting model prediction control, and performing global multi-objective optimization by adopting a periodic evolutionary algorithm. Generating a final control instruction based on the fusion strategy and issuing the instruction to an execution mechanism; by combining multivariable real-time prediction, closed-loop optimization control and a self-adaptive learning mechanism, intelligent management is carried out on the dehydration process of the wet desulphurization byproduct gypsum, accurate control over the water content of a filter cake, optimization of washing water and energy consumption and high adaptability of a control system to dynamic working conditions are achieved, and the dehydration efficiency of the wet desulphurization byproduct gypsum is improved. Therefore, on the premise of ensuring the quality of the gypsum, the dehydration efficiency is maximized and the resource consumption is minimized.

The application discloses a display method and device of an animation frame, equipment and a storage medium, and belongs to the technical field of computers. The method comprises the following steps: acquiring a resource consumption index based on an initial animation update frequency of at least one virtual model, wherein the resource consumption index is used to indicate the amount of resources consumed by animation update according to the initial animation update frequency of the at least one virtual model; acquiring a target animation update frequency of the at least one virtual model based on the resource consumption index and the initial animation update frequency; and displaying an animation frame corresponding to the at least one virtual model according to the target animation update frequency. In this way, the target animation update frequency of the at least one virtual model is determined on the basis of the initial animation update frequency by considering the resource consumption index, the reliability is relatively high, the resource consumption can be effectively and globally controlled according to the target animation update frequency of the at least one virtual model, the display fluency of the animation frame is improved, and the human-computer interaction rate is improved.

The invention discloses a PH electrode intelligent calibration method and system based on a self-adaptive calibration algorithm, and relates to the technical field of intelligent calibration control, and the method comprises the steps: carrying out the preprocessing of an original potential signal of an electrode, and obtaining a standardized potential; constructing a two-channel Gaussian process model, and outputting a calibrated pH value and uncertainty in real time; inputting a reinforcement learning decision network, automatically selecting an execution action according to a comprehensive reward function of a prediction error and resource consumption, and obtaining a new sample increment; and the new sample increment is updated to the Gaussian process and reinforcement learning model, and data-driven closed-loop adaptive calibration is completed. According to the method, the calibration pH value and the uncertainty thereof are output in real time through the two-channel Gaussian process model, and automatic balance is performed between prediction errors and resource consumption in combination with the reinforcement learning decision network, so that lag and waste of fixed period calibration are avoided, reagent, energy consumption and manual maintenance cost are remarkably reduced, and the overall operation efficiency of the system is improved.

The invention belongs to the field of deep learning prediction, and particularly relates to a deep learning prediction method for patient medical experience. A deep learning prediction method for patient medical experience comprises the following steps: acquiring sample data containing patient medical images and doctor-patient dialogue audio, converting the sample data into scale data in combination with a stored scale, and unifying data types of the scale data; screening the scale data through a preset missing rate threshold value; performing data amplification according to the score of each label in the scale data; a prediction model is constructed based on a multi-task deep learning model, and the bottom layer of the prediction model is composed of a linear layer and a convolution auto-encoder. According to the scheme, through multi-modal data processing, a multi-task model and edge calculation, the recognition accuracy is improved, the load of the server is reduced, and the problem that the economic burden of a hospital is increased due to the high-performance requirement of the server is solved; meanwhile, resource consumption is balanced, and data security is guaranteed.

The invention discloses a high-concurrency virtual service transfer scheduling method and system based on a cloud edge collaborative architecture, and relates to the technical field of cloud computing and edge computing collaboration, and the method comprises the steps: constructing a service gene analysis model at an edge node, responding to a virtual service access request, extracting the resource consumption characteristics of a service flow in real time, and calculating a service separation potential energy index. The virtual business is dynamically decoupled into a connection anchoring phase and a computing power free phase based on the index; establishing a load inertia monitoring mechanism, modeling resource consumption of edge nodes into a kinematics system, collecting speed and acceleration of load change in real time, calculating system pressure momentum, and generating a graded unloading instruction when the momentum exceeds a preset critical collapse threshold value; starting a convergent differential injection mechanism, pre-constructing a logic shadow at a cloud end, and calculating a convergence ratio of a memory dirty page generation rate to a network transmission rate in real time; the method has the advantages of being high in practicability, having the pre-judgment capacity and being capable of achieving lossless switching.

The invention provides a non-tampering operation and maintenance auditing system based on a block chain, and the system comprises a multi-source auditing data collection module, a block chain evidence storage engine, an intelligent contract auditing module, a cross-scene cooperative auditing module and a visual auditing platform, and all the modules cooperate according to a preset logic to form a closed-loop auditing system. According to the method, the problems of uncredible data, incomplete coverage, poor compliance adaptation, early warning lagging, insufficient cross-scene consistency and the like of a traditional operation and maintenance auditing system are solved, and through deep fusion of a block chain technology and an intelligent algorithm, auditing data full-link credible evidence storage, risk real-time perception, compliance automatic adaptation and cross-scene consistent tracing are realized; on the premise of guaranteeing the security and compliance of an information system, the auditing efficiency is improved, the operation and maintenance interference and resource consumption are reduced, and the operation and maintenance auditing requirements under complex IT architectures in multiple industries such as finance and government affairs are met.

The present invention relates to the technical field of building information modeling (BIM). Disclosed is a BIM lightweight method based on fine-grained geometric objectification, the method comprising the following steps: S1, creating a mapping table T to store data of Face objects and Mesh objects in a model; S2, calling an IExportContext interface in a Revit SDK to extract data of components in the model; S3, checking whether the processing of the components in the model is currently completed; S4, if the processing is completed, ending the process; and if the processing is not completed, acquiring the value of a current flag; S5, if flag=1, calling a Face object lightweight processing process and returning to S2; and S6, if flag=0, calling a Mesh object lightweight processing process and returning to S2. The present invention can effectively simplify model data and realize model lightweighting; moreover, obtained data supports LOD technology, which can greatly reduce resource consumption during front-end rendering and improve rendering efficiency.

The invention discloses a mine ecological restoration effect evaluation method based on a life cycle evaluation model, and relates to the technical field of ecological restoration effect evaluation, and the method comprises the steps: defining a mine restoration effect evaluation target, an evaluation range and a function unit, and carrying out the evaluation of the mine ecological restoration effect based on the evaluation range and the function unit; collecting resource consumption data, environmental emission data and ecological state data in the restoration process, and constructing a dynamic life cycle list database; and carrying out characterization calculation on data in the dynamic life cycle list database to obtain a preliminary environmental influence index value, establishing a recognition rule of a life cycle evaluation model based on sensitivity analysis of the soil microbial community in the mine restoration area to heavy metal stress, and recognizing the ecological toxicity potential of fresh water from the preliminary environmental influence index value. According to the method, the decision support report is generated visually, and multi-dimensional, dynamic and accurate evaluation of the mine restoration effect is realized.

The invention discloses a multi-agent-based recreational vehicle travel agent interaction service method, which comprises the following steps: collecting and preprocessing multi-source information data of a recreational vehicle, and generating a standardized input data set; constructing a self-evolution compliance-facility-experience coupling graph, and dynamically updating node weights; establishing a multi-agent cluster based on the coupling graph, and realizing communication and task allocation; constructing a dual-channel decision model, and fusing semantics and constraint channel output; establishing a three-cavity time topology system, and monitoring resource consumption and supply states; and executing comprehensive optimization and path planning, outputting an optimal travel and re-planning. According to the method, intelligent planning, resource optimization and dynamic compliance control in the touring process of the motor home are realized through multi-agent collaborative decision and self-evolution coupling modeling.

The invention relates to the technical field of computer load balancing, in particular to a large model reasoning load balancing method and system based on a dynamic resource grading threshold value. The method comprises the following steps: performing resource consumption estimation on a user reasoning request to obtain a request score value; the servers are divided into two server groups with different computing capabilities; differentiated scheduling is carried out based on the request score value and the current dynamic resource grading threshold value; and collecting the load state of the server cluster, calculating a load difference value, and dynamically adjusting the dynamic resource grading threshold value through a feedback control algorithm to realize load balancing. According to the method, the resource consumption of the reasoning request of the user is accurately estimated, and the efficient load balancing of the large-model reasoning task is realized in combination with the hierarchical management of the server, the real-time load state evaluation of the server cluster and the dynamic threshold adjustment mechanism.

The invention discloses a GIS-based agricultural data dynamic monitoring and management method, which relates to the technical field of agricultural geographic information and comprises the steps of collecting farmland multi-source monitoring data in real time, constructing a cross-system data fusion model, intelligently analyzing and executing an agricultural decision instruction, dynamically evaluating a regulation and control effect, feeding back an optimization model and the like. The problems that in the prior art, the data collection dimension is single, multi-source information fusion is insufficient, decision hysteresis is high, and resource distribution is extensive are solved. The monitoring accuracy is improved through full-dimensional data acquisition, decision scientificity is optimized through multi-source data fusion analysis, system adaptability is enhanced through dynamic closed-loop management, resource consumption and risks are reduced through intelligent and accurate regulation and control, multi-module collaboration and technology integration are achieved, and the system is suitable for large-scale popularization and application. A standardized, extensible, sustainable and optimized technical framework is provided for intelligent agriculture, and the utilization efficiency of agricultural resources and the intelligent level of production management are remarkably improved.

According to the factoring service hierarchical access method and device provided by the invention, multi-source heterogeneous data of a target enterprise is collected through a standardized interface, dimensions such as financial indexes, system performance, public opinion dynamics and compliance texts are covered, and data quality and safety are ensured through a professional data preprocessing process. A risk grading model of a fusion framework is adopted: text semantic features are extracted by utilizing a BERT model subjected to field fine tuning, structured data are analyzed in combination with the numerical understanding capability of a GPT series model, multi-source feature alignment is realized through a cross-modal attention mechanism, a comprehensive feature vector is generated based on a multi-head attention network, and the risk grading model of the fusion framework is obtained. And finally, outputting a risk level by the lightweight full-connection network, and forming a corresponding access strategy based on the risk level. According to the method, actual measurement-free evaluation is realized, a traditional test period needing a plurality of weeks is compressed to a plurality of hours, the model adaptability is continuously optimized through an online learning mechanism, multi-dimensional risk evaluation is realized, and test resource consumption and time cost are greatly reduced.

The invention discloses a multi-angle model operation automatic configuration optimization method and system based on big data, and relates to the technical field of big data, and the method comprises the steps: a cloud platform collects data from an industrial sensor and an equipment monitoring platform, calculates and standardizes a data quality index and a resource consumption feature, and constructs a standard feature vector and a data quality weight; based on a feature vector evaluation function and a resource matching degree, combining with a data quality weight to generate a candidate configuration scheme; taking the candidate scheme as an initial strategy, performing optimization through a composite reward function in a dynamic reinforcement learning environment, and outputting a global optimal parameter set; the parameter set is applied to job execution, and online fine adjustment is triggered through real-time monitoring and predicted value comparison; and starting a closed-loop updating mechanism containing strategy rollback and incremental learning based on full-cycle efficiency evaluation. According to the invention, the problem of low resource configuration efficiency in the industrial Internet of Things is effectively solved, and the system stability and the resource utilization rate are improved.