430 results about "Continuous optimization" patented technology

The invention relates to the technical field of mining industry, and discloses an intelligent prediction method for gold ore beneficiation process parameters based on cloud and edge fusion, which realizes space-time correlation modeling of beneficiation process parameters and accurately depicts dynamic interaction influence among equipment. The cloud edge collaborative architecture considers global optimization and real-time response requirements, and the prediction stability under complex working conditions is effectively improved. The introduction of physical constraints enhances the applicability of the model in an actual production environment, a bidirectional feedback mechanism ensures the adaptive ability of the system in a dynamic change environment, and through the joint reasoning of a knowledge graph and a neural network, the consistency of a prediction result and a process principle is enhanced, and the risk of misjudgment under an abnormal working condition is reduced; the man-machine cooperation mechanism significantly improves the labeling efficiency of high-value samples, shortens the model iteration period, and ensures the continuous optimization capability of the prediction system in the actual production environment.

The invention discloses a low-code development automatic generation method based on a large language model, which comprises the following steps: collecting natural language description information input by a user, and preprocessing; inputting the standardized demand corpus set into a large language model, and executing semantic understanding and context modeling; matching the low-code component library based on the structured semantic representation to generate component assembly description information; generating and verifying an engineering skeleton according to the assembly description information, and outputting an executable low-code application initial version; running the executable low-code application initial version, and monitoring and analyzing execution difference to generate an increment adjustment instruction; and inputting the increment adjustment instruction into the large language model, performing reconstruction and adaptive optimization, and outputting an executable low-code application final version. According to the method, large language model semantic understanding and adaptive optimization technologies are fused, automatic generation and continuous optimization of low-code applications are realized, and the method has the advantages of intelligence, high precision and engineering reliability.

The invention relates to an automatic driving risk quantification method based on a conflict risk field. Comprising the steps of 1, constructing a basic risk field; step 2, under a basic risk field framework, constructing a conflict risk field by taking an ADV as a center; step 3, regarding the conflict risk field as a repulsive force acting on the ADV, and using the repulsive force to quantify the dynamic influence of the environmental elements and various TP states on the driving safety of the ADV; the method comprises the following steps: firstly, constructing a basic risk field on the basis of traffic vehicle distribution and motion characteristics, further forming a vehicle forward, lateral, backward and traffic regulation constraint multi-dimensional conflict risk field on the basis, and converting an abstract traffic conflict relationship into a computable repulsive force model; dynamic quantitative characterization of driving risks in a state dimension, a space dimension and a time dimension is realized, and continuous optimization and evolution of an end-to-end automatic driving algorithm are supported by means of a repulsive force model constructed based on a conflict risk field.

The invention relates to a steel structure engineering welding quality defect analysis method based on voiceprint monitoring, and the method comprises the steps: carrying out the multi-channel voiceprint synchronous collection, time-frequency feature fusion, wavelet packet analysis and Mel-frequency cepstral coefficient extraction for a plurality of defect features fused in voiceprint data in a welding process; a hierarchical semantic concept space and a dynamic causal relationship generation model are established in combination with a welding physical knowledge base, a causal knowledge graph is constructed, causal association between semantic concepts is deduced through a gating circulation unit and a graph neural network, anti-fact disturbance and path aggregation analysis is carried out on a causal graph structure, and a result is obtained. And finally, defect category probability output and causal traceability graph visual display are realized. According to the scheme, the accuracy, traceability and result interpretability of welding defect recognition are effectively improved, and data support is provided for intelligent diagnosis and continuous model optimization in the welding process.

The invention relates to the technical field of distributed energy aggregation cooperative regulation and control, and discloses a virtual power plant hierarchical cooperative intelligent scheduling method and system based on multi-objective optimization, and the method comprises the steps: obtaining the operation parameters and prediction data of each subsystem of a virtual power plant, generating an energy storage system hour-level charge state target trajectory, and carrying out the prediction of the target trajectory; establishing an energy storage system charge state constraint budget pool; generating a budget allocation table; monitoring a short-term budget allowance state, when the short-term budget allowance state is lower than a safety threshold value, borrowing a part of budget from a long-term budget for redistribution, and dynamically adjusting a power amplitude limit value of deviation correction according to charge state constraint tensity; and when the accumulated deviation exceeds the autonomous correction capability or the constraint tensity reaches a critical value, generating an interlayer deviation report and triggering global re-optimization, and adjusting the budget distribution proportion of the next period according to the actual budget consumption condition at the end of the hour-level period. According to the invention, the continuous optimization and self-learning capability of the hierarchical scheduling strategy are realized.

The invention relates to the technical field of intelligent education, and discloses a teaching programming system based on large language model assistance and a teaching recommendation method. According to the system, a multi-dimensional ability graph and a knowledge point topology network are constructed, and a large language model is utilized to carry out multiple rounds of intention analysis and semantic reasoning to generate a personalized learning path sequence. The system continuously tracks a learning track, realizes continuous optimization of a teaching strategy by dynamically calibrating strategy parameters, and actively configures special reinforcement resources based on track prediction. According to the system, a breakthrough from static recommendation to dynamic adaptation is realized, the accuracy of learning path planning and the timeliness of teaching intervention are improved through deep semantic understanding and a closed-loop optimization mechanism, so that the programming teaching system can really understand learning requirements and adapt to changes of learning states in real time, and the teaching efficiency is improved. And the knowledge mastering firmness and the learning efficiency are improved.

The invention discloses an active defense method and system based on a large model, and relates to the technical field of security protection, and the method comprises the steps: intercepting a malicious request of an external attacker, cleaning sensitive information and adversarial samples in the malicious request, and outputting standardized data; injecting the standardized data as training data into a training confrontation sample to optimize a protection model, and ensuring the leakage traceability of the protection model by embedding a digital watermark; and trapping an attacker by deploying a honey spot interface and triggering a countering strategy, generating a dynamic defense rule by using the protection model, and updating the training confrontation sample in real time for continuous optimization of the protection model. Active attack sensing and advanced attack blocking are achieved through malicious request interception cleaning and honey spot trapping countering, dynamic defense rule generation and protection model continuous optimization are combined to adapt to attack iteration, a digital watermark tracing mechanism is matched, an'interception-protection-optimization 'closed-loop full link is constructed, and the security defense capability of the protection model is improved.

The invention discloses a middle and primary school test question intelligent generation method based on an education big model. The method comprises the steps of generating a structured education data set; constructing an education large model specially used for intelligent generation of middle and primary school test questions; inputting target knowledge points, question types, difficulty and grade parameters, and generating a candidate test question set; setting a test question fitness function, comprehensively evaluating the knowledge coverage rate, difficulty distribution and language expression indexes of the candidate test questions, screening and optimizing the candidate test question set, and generating an optimized candidate test question set; detecting semantic accuracy, logic preciseness and teaching conformity of the test questions by using an automatic quality evaluation mechanism, and outputting test question evaluation results; and continuous optimization and self-adaptive updating of test question generation are realized. According to the method, the optimal multi-question type test question set can be automatically output on the premise of ensuring comprehensive coverage of knowledge points, reasonable difficulty distribution and excellent language quality, and the quality, balance and diversity of automatically generated test questions are greatly improved.

The invention provides a plane layout optimization method and device based on CAD drawing automatic identification, relates to the field of plane layout optimization, and solves the technical problems that only non-overlapping constraints among equipment are considered, actual space obstacles cannot be automatically avoided, and a generated layout scheme cannot be implemented. The method comprises the following steps: extracting barrier data in a layer; and through a parameter configuration mechanism, generating an equipment parameter information structural body containing the mapping relationship between the equipment and the process. And establishing a continuous optimization model containing an IPOPT solver, and obtaining an initial layout solution containing the relative orientation of each device. And based on the relative orientation of each device in the initial layout solution, introducing an obstacle rejection constraint and a device non-overlapping constraint to construct a discrete optimization model, obtaining the center position coordinates of each device through a Gurobi solver, and integrating a dynamic adjustment mechanism and an interlayer relationship optimization mechanism. And generating a new layout map for visual output. The method and the device are used in a plane layout optimization process.

The invention relates to an APP intelligent marketing service method based on intelligent routing and multi-agent cooperation. The method comprises the following steps: receiving input information, wherein the input information comprises user active inquiry information or trigger event information generated based on user behavior monitoring; semantic analysis and intention recognition are carried out on the input information, a task planning directed acyclic graph is generated based on a recognition result, and the task planning directed acyclic graph comprises a plurality of subtask nodes and dependency relationships among the nodes; based on the task planning directed acyclic graph, the state of each functional agent and the historical performance index, executing dynamic routing so as to dispatch the plurality of sub-tasks to the corresponding functional agents; and in the execution process of the plurality of subtasks, performing dependency scheduling and state consistency management on an external tool call chain which is initiated by the functional agent and comprises a plurality of steps, and updating the shared memory associated with the user based on an execution result. By adopting the method, self-adaptive planning, robust execution and continuous optimization of marketing tasks can be realized.

The invention relates to the technical field of intelligent fertilization, and discloses an agricultural precise fertilization method which comprises the following steps: step 1, synchronously collecting multi-modal real-time data of single wheat plants, and automatically identifying the current growth stage of the wheat plants; 2, searching historical fertility data of the position of the wheat; 3, fusing the real-time data, the historical fertility data and the growth stage label to generate an enhanced fusion feature vector; step 4, selecting a corresponding causal diagram model according to the current growth stage, and estimating an average processing effect of different fertilization schemes based on the enhanced fusion feature vector; and 5, inputting the enhanced fusion feature vector, the average processing effect estimated value and the growth stage label into the prediction model and the adjustment model. According to the system and the method, the fertilization accuracy and the crop response effect are improved, continuous optimization of the system is realized through a closed-loop intelligent decision framework, and a landing and evolvable efficient fertilization solution is provided for intelligent agriculture.

The invention relates to the technical field of artificial intelligence learning, and discloses an AI-based scene decision interface dynamic generation and continuous optimization method. The method comprises the following steps: acquiring a user real-time interaction behavior sequence, task context information and an interface component library; fusing the codes to generate a real-time task state vector; predicting the probability distribution of the interface component required by the next decision step through the sequence prediction model; rendering a target interface according to the probability distribution and layout constraint dynamic combination; and calculating a reward value based on user feedback, and continuously optimizing the prediction model by utilizing reinforcement learning. According to the method, dynamic evolution and self-adaptive optimization of the interface along with user intention are realized, and the interaction efficiency and the user experience are improved.

The invention relates to the technical field of intelligent operation and maintenance, in particular to a fault insight and self-healing method and system based on chain control. According to the chain control-based fault insight and self-healing method, indexes, logs and tracking data of a micro-service environment are dynamically collected and fused; the method comprises the following steps: dynamically constructing and visualizing a service dependent topology, realizing dynamic topology construction and intelligent visual rendering, performing intelligent root cause analysis and predictive diagnosis based on multi-dimensional data, realizing self-adaptive early warning, generating an intelligent decision, performing closed-loop learning optimization based on an execution result, and realizing automatic fault repair and continuous optimization. According to the fault insight and self-healing method and system based on chain control, a predictive early warning mechanism is established, automatic fault self-healing is realized, and through intelligent and automatic means, manual intervention is reduced, the operation and maintenance cost is reduced, the average repair time is shortened, the service continuity is guaranteed, and the system availability is improved.

The invention discloses a health data automatic processing and precise health management method, system and device based on artificial intelligence and a medium. The method solves the problems that existing health data collection is fragmented, risk identification is not accurate, and management lacks individuation and continuous optimization, and comprises the steps that multi-source heterogeneous health data is acquired, and data acquisition is deepened in combination with edge calculation and an NLP technology; performing standardized preprocessing and fusion on the data; a multi-modal fusion deep learning model is constructed, the health risk and the prediction trend are accurately identified, and interpretability is provided; a personalized health management decision model based on reinforcement learning is constructed, and customized suggestions are generated; and the causal graph model is utilized to carry out iterative optimization on the feedback data, so that the data processing efficiency and the risk identification accuracy can be improved, thereby realizing highly personalized and adaptive precise health management, and effectively reducing the disease risk and the medical cost.

The invention discloses a city management AI dispatch algorithm and system based on historical mining and responsibility matching, and the method comprises the steps: building and dynamically updating a city management element evolution graph through obtaining the multi-mode description information of a city management case and the real-time state data of disposal resources; calculating potential disposal effects of different candidate dispatching schemes by using a causal inference engine, and generating a comprehensive efficiency estimation vector; on the basis, a multi-target reinforcement learning strategy is adopted to generate an optimal dispatch instruction, and system parameters are continuously optimized through online element learning during execution; cooperative processing network analysis is activated for sudden complex events, and responsibility atlas reconstruction is triggered when the matching efficiency is low. According to the method, the accuracy and efficiency of case disposal are remarkably improved, disposal timeliness optimization, resource load balancing and improvement of the first solution rate are realized, and meanwhile, the adaptive capacity and continuous optimization efficiency of the system to complex scenes are enhanced.

The invention discloses a multi-agent collaborative short play digital production system, which belongs to the technical field of artificial intelligence, and comprises a data and state management module driven by a digital thread, which uniformly manages and synchronizes data, states and decisions of all production links through a throughout digital thread; the full-process intelligent production module takes an intelligent control center as a core, decomposes and dispatches a complex task to five professional intelligent agents for cooperative execution, and covers a complete production link from script generation, visual preview, video shooting and special effects to intelligent editing and final sound and picture synthesis; the man-machine collaborative evaluation and feedback closed-loop module provides an interactive interface for human experts to supervise, guide and correct key production nodes and records artificial decisions as feedback to form a closed loop of a continuous optimization system; according to the multi-agent collaborative short play digital production system provided by the invention, the digital thread penetrates through multi-agent collaboration, so that data tracing and state synchronization of the whole process of short play production can be realized, challenges such as narrative continuity, cross-modal consistency and physical authenticity are effectively solved, and the quality of generated content and the production efficiency are remarkably improved.

The invention discloses a thermal power plant DCS fault prediction and diagnosis method and system, and relates to the technical field of fault prediction and diagnosis, and the method comprises the steps: obtaining equipment information and operation parameter reference values, and generating a to-be-detected equipment list; target equipment is determined based on the importance degree, a communication channel is established to collect real-time data, a parameter fluctuation characteristic matrix is constructed, a reference value is dynamically calibrated, and a calibration parameter set is generated; historical fault feature data are collected, and a fault feature library is established; generating a state vector matrix based on the calibration parameter set and the real-time data, and calculating an early warning index in combination with the fault feature library; calculating a risk coefficient during monitoring, dynamically optimizing an early warning threshold value, and completing early warning configuration; on the basis, diagnosis is executed, a standardized diagnosis process is generated, and a diagnosis knowledge base is formed; and when an exception occurs, triggering hierarchical rollback and updating the knowledge base to realize continuous optimization. The system comprises a master control module, a configuration module, a prediction diagnosis module, a transmission module and a display module.

The invention discloses a ship intelligent fault diagnosis method and system based on open label space identification. The method comprises the following steps: obtaining a multi-source sensor time sequence signal of a ship system and constructing a training sample set; carrying out feature extraction on the training sample set by utilizing a deep neural network model, and strengthening the clustering characteristics of the features by adopting a center loss function in the training process; training a classifier on the basis of feature extraction and introducing an open set loss function to form a comprehensive objective function; extracting features from a to-be-diagnosed sample, embedding the features, calculating the distance between the to-be-diagnosed sample and a known fault category feature center, and judging an unknown fault through comparison between the minimum distance and a preset threshold value; and outputting a known fault category or triggering an unknown fault alarm according to a judgment result, and dynamically expanding and updating a model knowledge base based on accumulated unknown fault samples. The method can break through the limitation of the traditional closed set hypothesis, effectively identifies the unknown fault type, and achieves the self-adaptive learning and continuous optimization of a ship fault diagnosis system.

The invention relates to a traffic simulation agent system construction method based on a large model, and the method comprises the steps: constructing a simulation tool library, and achieving the precise evaluation and continuous optimization of a simulation result through the fusion of multi-source heterogeneous traffic data, the construction of standardized input, and the establishment of a quantitative evaluation system. A large language model is utilized to understand a natural language instruction of a user, tasks are intelligently disassembled, an execution process is planned, dependence management and parallel scheduling are carried out in combination with a directed acyclic graph, and professional tools are driven to automatically execute. And performing evaluation, problem diagnosis and adaptive re-planning on an execution result through a large language model reflection mechanism to form an understanding-planning-execution-reflection closed loop. According to the method, the problems of how to assist a user to interact with a traffic system by utilizing an agent technology driven by a large language model, reducing the technical threshold of traffic simulation software use and saving time cost and labor cost are solved, the traffic simulation use threshold is reduced, the automation and intelligence level is improved, and efficient and accurate traffic system interaction and optimization are realized.

The invention relates to the field of distributed information retrieval, and particularly discloses a memory architecture vector approximate retrieval method based on a graph neural network, which comprises the following steps of: constructing and dynamically maintaining a historical query-hit vector association graph for modeling a deep semantic relationship between a historical query and a successful retrieval result; inputting the features of the current query vector, the information of the current query vector subjected to neighborhood sampling and feature aggregation in the graph and the service scene label into a lightweight graph neural network, and predicting the approximate retrieval tolerance level of the query; on the basis of the prediction result, an optimal retrieval strategy is generated in a self-adaptive mode; and in combination with asynchronous result return and a progressive refinement mechanism based on residual error reordering, a user is responded at the first time, and continuous optimization and pushing of a better result are realized. According to the method, a query-level personalized retrieval strategy is realized, the retrieval precision, the response delay and the system resource consumption are effectively balanced, and the method is suitable for a large-scale high-dimensional vector retrieval scene.

The embodiment of the invention discloses a strategy model training method and device, a strategy generation method and device, a storage medium and a terminal. Firstly, a set of sample answers are generated for a target question through a preset large language model to serve as references. And during training, inputting an answer generated by the current strategy model and a sample answer into the evaluation model for comparison, and outputting a good and bad sorting result. The ranking is quantized as a reward value whose size is positively correlated with the degree to which the current answer is superior to the sample answer. And finally, the system adjusts strategy model parameters according to the reward value and guides the strategy model parameters to be continuously optimized. According to the method, scoring according to standard answers is replaced with relative sorting, and the problem that open questions lack clear judgment standards is solved. The method has the beneficial effects that the training data cost and labeling dependence are remarkably reduced, so that the model can realize stable and autonomous efficiency improvement in the vertical field, and a self-driven benign evolution cycle is formed.

The invention relates to the field of defect detection, in particular to a multi-dimension-based defect detection labeling quality automatic evaluation method and system, and the method comprises the following steps: constructing an initial domain knowledge base, initializing a severity weight, a dynamic reliability weight and a multi-dimensional smoothing coefficient, and loading a pre-training defect detection model; inputting a defect sample batch to be evaluated, iteratively training the defect detection model, and calculating a multi-dimensional original evaluation index; obtaining an evaluation dimension set through dimension reduction and standardization processing, and generating a dynamic tracking result of each dimension index by adopting an index moving average algorithm in combination with a smoothing coefficient; in combination with the severity weight and the dynamic reliability weight, a comprehensive mark quality score is obtained through weighted fusion, and suspected error mark samples are screened out; and performing iterative training on the defect detection model, and outputting a final result. According to the invention, continuous optimization of evaluation parameters is realized through a man-machine cooperative feedback closed loop, and the practicability and stability of the technical scheme are further enhanced.

The invention relates to a network fault root cause positioning system based on multi-modal learning and causal inference, and belongs to the technical field of network fault diagnosis and positioning. The system comprises a data processing and association mining module, an intelligent fault diagnosis and evaluation module, a root cause positioning module and a system optimization module which are connected in sequence. The data processing module collects data through edge nodes, constructs a hierarchical knowledge graph and outputs a feature matrix; the diagnosis module performs fault detection by adopting a multi-modal model of a fusion graph neural network, integrates a small sample learning mechanism and outputs a fault event with confidence; the positioning module constructs a causal graph based on a knowledge graph, fuses multi-source evidences and realizes root cause tracing through a random walk algorithm; and the optimization module adjusts diagnosis parameters by utilizing reinforcement learning, expands a sample set based on the generative adversarial network, and realizes continuous optimization of the model through an automatic assembly line. Closed-loop self-optimization from fault sensing to root cause positioning is realized, and the network fault management capability is improved.

The invention discloses an enterprise operation management intelligent optimization system based on business rules, and belongs to the technical field of enterprise operation management. The system comprises core modules such as a multi-source data fusion center, a dynamic business rule engine, an intelligent optimization decision module and a closed-loop feedback unit, wherein the fusion center realizes privacy protection type multi-source data fusion by adopting federal learning; the rule engine realizes real-time adjustment of business rules without interrupting the system through rule atlas construction and incremental updating; and the optimization module is combined with an analytic hierarchy process and a Q-learning algorithm to generate a real-time optimization decision under business rule constraints. The system can improve business rule adaptation flexibility and data security, realizes decision real-time and optimization continuation, adapts to multiple industries of manufacturing, retail and logistics, and reduces enterprise operation cost and system operation threshold.

The invention discloses a multi-modal sensing edge intelligent monitoring system, which relates to the field of edge intelligent monitoring and comprises a multi-modal data acquisition module for acquiring multi-modal data; the edge intelligent processing module performs space-time registration, feature extraction and feature level fusion on the multi-modal data to generate a multi-modal feature vector; performing joint reasoning on the multi-modal feature vector by adopting an intelligent analysis model, outputting a structured analysis result, and generating a local early warning instruction when the danger level indicated by the structured analysis result exceeds a threshold value; and the central control module carries out early warning information issuing, data archiving and model iteration training according to the structured analysis result. Localized feature level fusion and intelligent analysis are carried out on multi-modal data through an edge intelligent processing module, and low-delay risk identification and early warning are realized. Meanwhile, continuous optimization and global situation awareness of the model are realized through a cloud edge coordination mechanism, and the defects of the traditional architecture in real-time performance, reliability and intelligent level are effectively overcome.

The invention relates to the technical field of software engineering and automatic testing, in particular to an E2E scene test optimization method and system based on target man-machine cooperation. According to the scheme, an optimization target is manually set, an initial test case and a business process are input, and a test resource library is constructed; the system automatically analyzes a service path, generates a use case dependency graph and a key path, and determines a core scene through manual verification; a candidate optimization strategy is generated by the system, and a target strategy is determined through manual screening according to resources and priorities; the system carries out simplification, merging and scheduling optimization on the use cases, and deploys and executes the use cases after manual sampling verification of coverage; the system collects execution data in real time and generates an evaluation report, and whether a use case is supplemented or a strategy is adjusted is determined after manual analysis; and finally updating the resource library and the strategy model, establishing a dynamic feedback mechanism, and triggering continuous optimization during system iteration. According to the method, the E2E test efficiency and the software delivery quality are remarkably improved through man-machine cooperation and closed-loop optimization.

The invention relates to the technical field of environment monitoring, in particular to an environment regulation system and method based on an environment monitoring technology, and the system comprises a region division module which is used for dividing monitoring sub-regions based on the spatial distribution features of a heterogeneous data set, and distributing a corresponding data feature extraction rule for each monitoring sub-region; and the feature extraction module is used for extracting abnormal fluctuation indexes and region association features from the environmental data of each monitoring sub-region based on the data feature extraction rule to obtain a sub-region feature set. According to the method, the parameters of the anomaly analysis model are updated by generating the specific environment optimization strategy and real-time feedback data, and continuous optimization and accurate execution of the regulation and control strategy are ensured; meanwhile, through staged regulation and control and cooperative work of equipment, the environment regulation effect can be improved to the maximum extent.

The invention provides a wire harness diameter simulation calculation method based on a simulated annealing algorithm, and relates to the field of wire harness design, initial layouts are randomly generated for wires and wire harnesses, the wires are small circles, and the wire harnesses are large circles; obtaining a local optimal layout and a given illegal layout by using a continuous optimization method; judging whether the dynamic adjustment amount of the large circle is smaller than set precision or not, if yes, outputting a layout result, meeting constraint conditions, of each circle in the current layout, and if yes, reducing the radius of the large circle by adopting a dichotomy method, and obtaining a new layout from the layout with the reduced diameter of the large circle through a simulated annealing algorithm; new layouts obtained through the simulated annealing algorithm are divided into legal layouts and illegal layouts, the legal layouts are processed according to local optimal layouts, the illegal layouts output through simulated annealing adopt a dichotomy method to increase the radius of a large circle, and iteration is conducted again through the simulated annealing algorithm. The method is high in accuracy and wide in applicability, and the wire arrangement condition of the cross section of the wire harness can be obtained.

The invention discloses a multi-objective continuous optimization method for a chemical reaction kettle. The method comprises the following steps: S1, initializing a population and a trainable parameter set; s2, calculating an individual target value and a gradient, and obtaining a performance steepest improvement direction after standardization; s3, constructing a conjugate direction in combination with historical gradients, and identifying sensitive key dimensions of the target function; s4, fusing the gradient and the conjugate direction to generate a main direction candidate solution, and applying refined disturbance to the sensitive dimension to generate a structural disturbance candidate solution; s5, updating the Pareto optimal solution set, and if a termination condition is met, outputting a multivariable optimization parameter set such as temperature-pressure-flow of the reaction kettle for actual operation; otherwise, returning to S2; according to the method, through gradient correction, conjugate direction fusion, sensitive dimension directional disturbance and parameter adaptive adjustment, the efficiency, precision and stability of multivariable optimization of the reaction kettle are remarkably improved.

The invention provides a business process intelligent configuration and security execution method, system and device, and the method comprises the steps: converting a natural language demand into a BPMN 2.0 process model based on a large language model and an enterprise knowledge base; the service rule description is converted into a Grouping script through multi-level security compiling; the process and the script are executed in the Flowable process engine integrated with the security sandbox, and execution data are collected; and calculating a process health score based on the data, generating and implementing an optimization suggestion, and feeding back a result to the knowledge base. The problems that a traditional process system depends on manual coding, the script safety is poor, and continuous optimization is lacked can be solved. The full-link automation and intelligentization of the business process from design, execution to optimization can be realized, the development efficiency and the system security are greatly improved, and the process has the self-evolution capability through closed-loop feedback.