21 results about "Preliminary analysis" patented technology

The invention discloses an intelligent report generation system, method and device and a storage medium, and relates to the technical field of data processing, and the system comprises a main research agent which is used for multi-source data information collection and preliminary analysis; the report writing agent is used for converting the research result into a professional report; and the review agent is used for reviewing the professional report. A multi-agent collaborative architecture and a domain knowledge enhancement technology are deeply fused, a full-link quality assurance system including data credibility evaluation, multi-source cross validation and dynamic knowledge updating is established, and collaborative optimization of professional report generation in three dimensions of timeliness, economy and specialty is realized.

The invention discloses a construction potential safety hazard identification method and system based on multi-agent cooperation. The method comprises the steps that an intelligent agent network is constructed, the intelligent agent network comprises a plurality of intelligent agents responsible for different tasks, and the intelligent agents cooperatively work in a serial mode and a parallel mode; obtaining construction related data; performing preliminary identification of potential safety hazards to obtain a preliminary analysis result; verifying and optimizing the potential safety hazard based on the preliminary analysis result in combination with an intelligent agent network to obtain a verification result; correcting the preliminary analysis result based on the verification result to obtain a conclusion of hidden danger identification and risk assessment; and generating a standardized security management document based on the conclusion according to the selected document type. By implementing the method provided by the invention, more accurate hidden danger identification, risk assessment and regulation matching can be realized, standardized and high-quality safety management documents can be generated, the working efficiency can be improved, and the professionality and accuracy of the documents can be ensured, so that the construction safety management practice is effectively supported.

The invention discloses a dynamic adaptive operation, maintenance and metering optimization method for an electric energy metering box, and the method comprises the steps: collecting electrical parameters, environmental parameters and box body state parameters in real time, constructing a three-dimensional data set through preprocessing, and carrying out the local storage and preliminary analysis through edge calculation; the condensation occurrence probability is calculated based on the environmental parameters, and self-adaptive regulation and control are triggered and a ventilation structure is linked; abnormal features are extracted, abnormity is recognized in a rule judgment and trend analysis double-engine mode, and graded early warning is carried out; performing comprehensive state scoring in combination with the three-dimensional data set, and matching a differentiated inspection mechanism; presetting a metering error reference value of a temperature and humidity interval, calculating a compensation value in real time and remotely correcting parameters of the electric energy meter; and establishing a shared data platform of the power supply and the power consumption, and realizing full-process data tracing and automatic report generation. According to the method, the condensation prevention and control accuracy is improved, the abnormal false alarm rate is reduced, the operation and maintenance resource allocation is optimized, the metering accuracy is improved, and the data collaboration and tracing capability is enhanced.

The invention relates to the field of rainfall pollution detection, in particular to an automatic rainfall sampling method and device for a high-density built-up area. Geographic information parameters and rainfall dynamic parameters of the high-density built-up area are collected; performing format unification, space-time alignment and normalization processing on the obtained multi-source data, and constructing a standardized multi-dimensional feature data set; inputting the standardized multi-dimensional feature data set into a pollution risk prediction model, and dynamically calculating and outputting a pollution risk level of a rainfall event and a time window of a key pollution stage; generating a composite dynamic sampling strategy based on the pollution risk level and the time window; the method comprises the following steps: collecting and storing a water sample according to a composite dynamic sampling strategy, and carrying out preliminary analysis on the collected water sample by using an integrated water quality sensor to generate actually measured concentration data; and carrying out weighted compensation on the generated actually measured concentration data by utilizing geographic information parameters of the high-density built-up area, and calculating the equivalent pollution concentration representing the comprehensive pollution level of the high-density built-up area.

The invention discloses a large model training method for generating corn growth guidance. The method belongs to the technical field of agricultural intelligent monitoring and prediction. The method solves the problem that when applied to a corn growth management scene, a large model method depends on multi-source heterogeneous information such as remote sensing images and meteorological data, although preliminary analysis and prediction based on the crop growth state can be achieved, the large model method still has limitation in the aspects of accuracy, interpretability and actual adaptability of decision suggestions. During data set construction, besides fusion of remote sensing data and multi-dimensional meteorological information, expert suggestions are added, more multivariate fusion features are constructed, and due to similarity between the expert suggestions and farmland management suggestions, correspondence between the features and tags is more accurate; according to the method, corn growth guidance based on a large model architecture can realize a technical span from'visible 'to'understandable' to'guidance ', and a scientific basis and decision support are provided for precise management of corn production.

The invention relates to an energy consumption analysis and energy-saving report generation method and device, and belongs to the technical field of energy consumption analysis and energy-saving report generation. The method comprises the steps that a planning agent analyzes user requirements and formulates an analysis process; the time and space analysis agents analyze the energy consumption data according to the analysis process to obtain a preliminary analysis result; the preliminary analysis results are fed back to the planning agent, and the planning agent combines the preliminary analysis results and submits the results to the questioning agent; the questioning agent questions the preliminary analysis result in combination with the content concerned by the user and the common pain point of energy consumption and feeds back the questions to the time and space analysis agent for enhanced analysis to obtain enhanced analysis results, and the planning agent combines the enhanced analysis results and submits the enhanced analysis results to the questioning agent; the questioning agent questions the enhanced analysis result until there is no problem, and the planning agent submits the final analysis result to the summarizing agent; and the summarizing agent generates an energy-saving report according to the final analysis result and the typical case data.

A large model training method for generating corn growth guidance. It belongs to the field of agricultural intelligent monitoring and prediction technology. It solves the problem that when the large model method is applied to the corn growth management scene, it depends on remote sensing images and meteorological data and other multi-source heterogeneous information. Although it can realize preliminary analysis and prediction based on the growth state of crops, there are still limitations in the accuracy, explainability and actual adaptability of decision-making suggestions. When constructing the data set, in addition to the fusion of remote sensing data and multi-dimensional meteorological information, expert suggestions are also added to construct more diverse fusion features. Due to the similarity between expert suggestions and farmland management suggestions, the correspondence between features and labels is more accurate, which can make the corn growth guidance based on the large model architecture realize the technical leap from "seeing" to "understanding" to "guiding", and provide scientific basis and decision support for the precision management of corn production.

The invention relates to the technical field of new energy, in particular to a hydrogen production system sensitivity analysis method, system and equipment and a medium. According to the method, the Tornado graph is taken as preposed analysis, the layered sensitivity analysis is performed by combining the Morris method and the Sobol method, and the method is used for the input variable influence analysis of the capacity optimization model of the hydrogen production system of the new energy base. The method can systematically and qualitatively identify the variables having great influence on the objective function, then performs fine evaluation, further depicts the interaction effect between the key variables and the proportion of the interaction effect in the overall output variance, provides a basis for subsequent model simplification and parameter setting, and improves the modeling efficiency and stability. The method adapts to a complex engineering model, and is especially suitable for a new energy system planning model with many input variables and high calculation cost.

The present application relates to the technical field of data processing, and more particularly to an airport security data intelligent analysis method and system. The method comprises the following steps: obtaining airport multi-source security data; performing preliminary analysis on the security data according to the airport multi-source security data to obtain security characteristic sequence data; performing behavior mutation feature correlation according to the security characteristic sequence data to obtain security mutation feature correlation data, and performing cross-modal feature correlation according to the security characteristic sequence data to obtain security cross-modal feature correlation data; performing mutation behavior threat level evaluation according to the security mutation feature correlation data to obtain first security danger data, and performing cross-modal threat level evaluation according to the security cross-modal feature correlation data to obtain second security danger data; and performing security decision auxiliary work according to the first security danger data and the second security danger data. The present application greatly improves the intelligent level and efficiency of airport security through quantitative threat and intelligent decision support.

The invention relates to the field of multi-modal model and multi-agent cooperation, and discloses a multi-agent typesetting generation system based on rule formalization injection, which comprises the following core steps: a rule formalization generation agent can receive a design specification input by a natural language to generate a formalization rule base; the VLM multi-mode intention understanding agent receives user materials and design requirements, then carries out preliminary analysis, distributes tasks to downstream professional agents for processing, and creates an initial layout. Then, the formalized rule checking agent calls the rule base to carry out compliance checking; and if the layout is not compliant, the system automatically triggers an iteration correction loop until the layout passes the verification or reaches an iteration upper limit. And the user can carry out manual refined modification or version rollback on the basis. According to the method, subjective design specifications are converted into objective machine instructions, and a multi-agent cooperation process is established, so that the problem of specification conformity in automatic typesetting is solved.

The invention discloses an ammeter connector health state management method and a storage medium. The method comprises the following steps: constructing and binding an ammeter connector digital twin of a power grid topological relation; multi-source operation data such as temperature of a key connection point, electrical data flowing through a loop and environmental data are collected in real time through a sensor network; inputting the data into a health state evaluation model to output a health state grade, fusing preliminary analysis based on a preset rule base and deep analysis based on a data-driven model in the evaluation process of the model, and generating a final grade by adopting a dynamic weight fusion mechanism according to the uncertainty of the results of the preliminary analysis and the deep analysis; generating and triggering a differentiated maintenance decision instruction according to the grade; and optimizing and updating the evaluation rule base and / or the data driving model based on result data fed back by field maintenance. According to the invention, comprehensive perception, accurate evaluation, intelligent decision and closed-loop optimization of the health state of the electric meter connector are realized, and the intelligent level of state management and the reliability of power supply are improved.

The invention relates to the field of automatic control, and particularly discloses an automatic quality control system based on a multi-modal large language model, which comprises the following steps of: firstly, constructing a multi-modal vector knowledge base by using scattered quality documents, CAD (Computer Aided Design) files and standard legend; during detection, the system firstly performs preliminary analysis on a product image to generate a defect hypothesis, and starts a coarse-to-fine two-stage dynamic knowledge retrieval by taking the defect hypothesis as an index. The retrieval process not only matches text semantics, but also finely aligns recalled knowledge with an actual abnormal region in an image through visual positioning, so that a highly focused and credible enhanced context is constructed for a multi-modal large language model. Based on the context, the model can perform deterministic and explainable decision, and finally the general reasoning ability of the model is combined with the harsh standard of a specific industrial scene, so that accurate, flexible and traceable automatic quality control is realized.

The invention discloses a hydropower station trend analysis and professional report generation system and method based on an intelligent agent and a large language model, and belongs to the technical field of hydropower station intelligent operation and maintenance and data analysis. The system comprises a data perception and execution layer, an agent scheduling and analysis layer, a professional report generation layer and a dynamic causal knowledge base. The method comprises the following steps: analyzing a user instruction through a scheduling agent and scheduling a tool to perform preliminary analysis; the analysis agent queries a dynamic causal knowledge base based on the preliminary anomaly to generate a cross-device causal hypothesis, and autonomously schedules a tool to perform data verification; and finally, synthesizing a structured diagnosis report containing a core diagnosis conclusion, an evidence chain and operation and maintenance suggestions by a professional report generation layer by utilizing a large language model. According to the method, the defect that only isolated data description can be carried out in the prior art is overcome, the crossing from'phenomenon description 'to'disease cause diagnosis' is realized, and the analysis depth and decision support value of the report are remarkably improved.

The invention discloses a workshop logistics sorting virtual debugging system based on artificial intelligence and an implementation method, and particularly relates to the field of industrial control, and the system comprises a multi-source cooperative sensing module which obtains original sensing data through a multi-source sensor array; the streaming feature construction module is used for performing data cleaning and feature enhancement on the original sensing data to obtain a multi-dimensional basic feature vector, and performing dimension reduction processing on the multi-dimensional basic feature vector to obtain a multi-dimensional feature vector; the space-time prediction module is used for extracting space-time dependence from the multi-dimensional feature vector to obtain a preliminary analysis result; the enhanced decision module is used for generating an optimal control strategy based on the preliminary analysis result; and the virtual execution and verification module loads the control strategy into a virtual workshop model for simulation operation, and finally generates a control instruction and a debugging analysis report. The accuracy and stability of the sorting process are remarkably improved, meanwhile, the cost and risk of actual debugging are reduced, and the efficient sorting requirement of a modern workshop is met.

The invention discloses a fault root cause determination method and device and electronic equipment. The method comprises the following steps: acquiring abnormal information of multi-source data from an event sensing layer; according to a preset arrangement strategy, calling atomic power to analyze the abnormal information to obtain a preliminary analysis result; analyzing the service damage degree according to the preliminary analysis result to obtain the service damage level; and under the condition that the service damage level reaches a preset level, executing a root cause positioning process according to a preliminary analysis result to obtain a fault root cause description corresponding to the abnormal information. According to the invention, the technical problem of poor fault processing accuracy and timeliness caused by long fault positioning time when a large amount of alarm information exists in a network in the prior art is solved.

According to the system performance inspection method and device, the storage medium and the computer equipment provided by the invention, firstly, the performance index data is preliminarily analyzed through the first model, original fixed rule judgment is replaced, and preliminary understanding of a performance index trend is realized. Then, a matched historical optimal case and an analysis framework are retrieved in the knowledge base according to the preliminary analysis trend, and empirical contexts and judgment standards are injected for the current decision; and finally, inputting the performance index data, the optimal case and the analysis framework into a second model, and simulating a decision-making process of performing comprehensive research and judgment by an expert in combination with experience, a method and real-time data through the second model. And a highly accurate inspection report is obtained. In the process, in combination with deep analysis of the deep learning model and the knowledge base, the ability of understanding and judging performance index fluctuation is improved, so that the misjudgment probability is reduced, and the accuracy and reliability of performance inspection are improved.

The application relates to the field of multi-modal model and multi-agent cooperation, and discloses a multi-agent layout generation system based on rule formalization injection, which comprises the following core steps: a design specification capable of being received by an agent in a natural language input is generated to generate a formal rule library; a VLM multi-modal intention understanding agent performs preliminary analysis after receiving user materials and design requirements, and distributes tasks to downstream professional agents for processing to create an initial layout; subsequently, a formal rule checking agent calls the rule library to perform compliance checking; if the layout is not compliant, an iteration correction cycle is automatically triggered until the layout passes the checking or the iteration upper limit is reached; and the user can manually fine-tune or version rollback on this basis. The subjective design specification is converted into objective machine instructions, and a multi-agent cooperation process is established, so that the specification compliance problem in automatic layout is solved.

The invention relates to the technical field of industrial energy efficiency regulation and control, and discloses a production line energy efficiency analysis and regulation and control system based on a super-brain industrial vertical type large model. A data acquisition module of the system acquires energy consumption data, equipment state parameters and environment monitoring data of a production line in real time, and maps the data into structured input; the energy efficiency preliminary analysis module extracts key energy efficiency indexes from input data as reference entries, retrieves a related analysis data set through a super-brain industrial vertical class large model, divides the related analysis data set into a preliminary analysis data set and a disposable data set, and discards the disposable data set; the energy efficiency comparison module receives the preliminary analysis data set, sorts the preliminary analysis data set in a descending order according to the number of data subsets, selects a plurality of first subsets as an optimal data set, calculates the proportion of the data sets, compares the proportion with a preset screening proportion threshold value, generates a first analysis report and sends an energy efficiency check instruction; and after receiving the instruction, the regulation and control optimization module performs secondary classification on the optimized data set according to a database classification standard, and outputs an optimized regulation and control instruction in combination with a historical regulation and control record.

The invention provides a firework and cracker production and sales whole process risk factor analysis model construction method, which comprises the steps of establishing a business process preliminary analysis framework and business process analysis factors, and listing risk subsystems and risk factors; a TF-IDF algorithm is used to extract the risk factors, and through similarity calculation, seed factors required by model construction are obtained; performing iterative divergence on the seed factors by adopting literature data, and performing divergence by an expert group Delphi method to achieve risk factor expansion so as to form a convergence factor set; adopting a DM-AHP hybrid algorithm to carry out secondary Delphi method operation on the confluence factor set, and endowing each element in the factor set with a weight; establishing hierarchical holographic modeling, and displaying related factors of the risk system for the factor set endowed with the weight data; risk factors of a risk system are analyzed from a mesoscopic level, modeling is performed by using a main logic diagram in a microscopic level, and a risk factor analysis subsystem is established. According to the method, risk factors are analyzed from the whole firework and firecracker production and sales process, and overall system analysis is facilitated.

The invention belongs to the technical field of artificial intelligence and law crossing, and particularly provides a multi-agent law intelligent question answering method and system, electronic equipment and a storage medium, and the method comprises the steps: evaluating an original legal knowledge question of a user, and judging the original legal knowledge question as a simple legal question or a complex legal question; performing structured decomposition on the complex legal problem based on a decomposition reasoning cognitive framework to obtain a plurality of to-be-retrieved sub-problems; generating a preliminary analysis answer for each to-be-retrieved sub-question; all the analysis answers are integrated to form a structured analysis abstract; constructing a multi-agent collaborative generation framework; and based on the structured analysis abstract and the original legal problem, a final legal consultation reply is generated by adopting a multi-agent collaborative generation architecture. According to the method, an agent-type workflow is taken as a core and is responsible for systematically converting an unstructured original legal problem into a professional reply which is clear in structure and can be based on data.