54 results about "Decision table" patented technology

The invention provides an in-service slope inspection method and system based on machine vision, and the method comprises the steps: building a composite data set through obtaining the continuously collected visible light image data and the synchronously collected infrared thermal imaging data of an in-service slope, and the three-dimensional point cloud data, carrying out the multi-dimensional information extraction processing of the composite data set, and carrying out the inspection of the in-service slope. And obtaining the texture change sequence of the in-service slope surface and the temperature gradient distribution corresponding to the infrared thermal imaging data. And inputting the two into a pre-trained transfer learning model, adjusting and optimizing the model in combination with slope instability case data in a historical disaster database, and outputting instability risk probability distribution of the in-service slope. Performing space-time matching on the instability risk probability distribution and meteorological data acquired in real time, quantifying the influence weight of rainfall and temperature change on the stability of the in-service slope through a meteorological geological coupling model, updating a deformation prediction curve of the in-service slope, performing matching analysis with a disaster threshold library in a historical disaster database based on the deformation prediction curve, and determining the stability of the in-service slope. And generating a slope maintenance decision table.

The invention discloses an intelligent decision-making system for regular work execution of hydropower station equipment. The intelligent decision-making system comprises an equipment database, an intelligent decision-making module and an execution feedback module, the equipment database is in data connection with the intelligent decision-making module, the equipment database collects equipment data through a data interface and stores and records the equipment data, the execution feedback module obtains a decision-making table through the intelligent decision-making module and carries out non-execution feedback, and the intelligent decision-making module is connected with the execution feedback module. The intelligent decision-making module generates a regular work decision-making table according to the data of the equipment database and the feedback data of the execution feedback module; the execution feedback module is in feedback connection with the equipment database. According to the method, the operation big data of the equipment is collected in real time and combined with real-time state detection, the equipment database is constructed, a dynamic adjustment mechanism is established, the regular work decision table is automatically generated, and regular work omission caused by human negligence is avoided.

The application provides a privacy protection vertical federated learning method and system based on gradient boosting decision table, and the method is applied to a system comprising a plurality of participating terminals, each participating terminal holds part of feature data of each learning sample, and the categories of the feature data held by each participating terminal are different; in the learning method provided by the application, all intermediate data is encrypted in the whole learning process, each participating terminal can only obtain additive secret sharing shares of the data, and cannot obtain complete plaintext data. The application can realize privacy protection of privacy protection vertical federated learning data based on gradient boosting decision table.

The invention discloses a flexible load identification method based on load area characteristic enhancement, and aims to solve the problems that the traditional flexible load identification method is seriously insufficient in robustness and generalization ability, poor in dynamic adaptability and over-coarse in classification granularity when facing complex and changeable power consumption behaviors, so that the flexible load identification accuracy is low, and the flexible load identification efficiency is low. And the actual requirements of a modern power grid on efficient and accurate development and scheduling of massive flexible load resources are difficult to meet. According to the method, a load area characteristic enhancement expression model and a time sequence characteristic description model with an area label are trained and constructed by using a load basic information set and an urban function area point information set, and the association between the load and the urban function interval is deeply described. And training a decision-making model by using the decision-making table, and judging whether the load is a flexible load or not by the decision-making model according to a splicing vector of an output result of the load region characteristic enhancement expression model and the time sequence characteristic description model with the region label. The invention belongs to the field of power system management and analysis.

The invention discloses a radio frequency treatment comfort optimization method based on multi-modal data driving, and aims at solving the problems that in the prior art, subjective dependence of comfort evaluation is high, and systematic data support and optimization strategies are lacked. A multi-mode sensor is configured through a cloud server, treatment scheme data including thermal imaging, power output and pressure data are collected, and patient condition and environment data obtained from an authorization evaluation platform are combined. A preset comfort level analysis rule set and a multi-dimensional decision table are utilized to perform objective comfort level division and scoring on the treatment process, and based on the division result and the environment data, an ablation system technology is adopted to simulate and calculate the expected influence of different environment parameter combinations so as to determine the comfort level of the treatment process. And generating personalized environmental parameter optimization suggestions under the condition of meeting security constraints. According to the invention, objective and quantitative evaluation and intelligent environment regulation and control of the radio frequency treatment comfort are realized, and the standardization level of treatment and the patient experience are improved.

The present application relates to the technical field of routing context connection, in particular, the present application relates to the zero code visualizing and automatic routing method of large model service arrangement, the present application extracts core entity and intention dependency by analyzing user session text, constructs entity intention graph and compresses state snapshot, and assigns global session ID, binds snapshot to distributed routing decision tree node and stores, establishes cross-node state continuity, when receiving new request, fusion snapshot and new session semantics, calculate routing weight offset correction decision table, through the rule and strategy of drag interface configuration, generate routing decision logic flow, this method solves the problem of session state loss and intention drift, improves service matching degree, reduces development threshold, and is suitable for agile construction of large model collaborative application.

The application provides a POE power supply-based edge computing type cultural relic warehouse environment multi-parameter online monitoring method, which comprises the following steps: performing protocol unification processing, preprocessing and cleaning on multi-source sensing data to obtain clean data; performing multi-source data fusion correlation analysis and constructing a cultural relic warehouse multi-parameter linkage analysis decision table; obtaining a quantitative cultural relic damage risk assessment and early warning level to obtain a cultural relic damage risk index and an early warning level, so as to output a data uploading scheme and a local backup according to a network state; and constructing a multi-gas cross interference model by using standard data flow to output high-precision gas concentration data compensated by cross interference. The application can simplify wiring, improve data processing efficiency, system reliability and maintenance convenience, and provides an efficient, reliable and easy-to-maintain intelligent solution for cultural relic warehouse environment monitoring.

The application discloses a method for obtaining training data of a speech synthesis model, comprising the following steps: screening part of corpus with high frequency syllables from a corpus of a target language; recording and sampling the screened corpus and marking the corpus to obtain an initial training set, and training the speech synthesis model through the initial training set; synthesizing speech of unmarked corpus in the corpus according to the speech synthesis model to obtain quality information of the speech synthesis; determining decision attributes of the unmarked corpus according to the quality information, taking the occurrence times of basic letters in the unmarked corpus as conditional attributes, constructing a decision table, and obtaining a dominant rough set. The application effectively solves the problems of few training data samples, high professional requirement, high cost and heavy workload in the recording and marking of training data samples in the target language synthesis task.

The invention relates to a game confrontation-based large language model dynamic evaluation and optimization method. The method comprises the following steps: acquiring data of a game environment and providing a state input interface and an action execution interface for a large language model agent; analyzing a game environment to obtain a game feature vector; matching the cue word template to generate cue words; inputting the cue word into a large language model agent, and obtaining an action after obtaining a model response and analyzing the model response; forming a global battle log database based on the actions; obtaining a multi-dimensional evaluation result based on the global battle log database; and forming a closed loop of evaluation, optimization and re-evaluation based on a multi-dimensional evaluation result and a global battle log database. By utilizing the game confrontation-based large language model dynamic evaluation and optimization method designed by the invention, the problems in the prior art that the evaluation process is difficult to reflect the decision performance of the model, the evaluation index lacks linkage investigation, the available range of the evaluation result is small, and the application of the evaluation data is small can be solved.

The invention discloses a real-time data feedback film tape identification and distribution method and system, and relates to the technical field of film tape identification. An alignment data set is collected, optical and geometric features of the alignment data set are extracted, defects are identified, a label matrix is generated, a quality score is calculated based on the label matrix, and a distribution decision table is established. According to the invention, integration of film tape quality detection and distribution control is realized, multi-modal images and operation parameters are fused, time synchronization, threshold self-adaption and feedback learning mechanisms are provided, defects can be identified in real time and accurate distribution can be executed in a high-speed production environment, and the film tape quality detection and distribution control system is suitable for large-scale production. Model parameters can be dynamically corrected according to feedback data, the recognition and control precision is continuously optimized, and the yield and operation stability of film tape production are remarkably improved.

The application discloses a vehicle control method and device, electronic equipment and storage medium. The method comprises the following steps: acquiring current running data and a main driving gear of a vehicle; determining a basic gear corresponding to the current running data according to a preset basic gear decision table; when it is detected that a preset gear shifting trigger condition is met, determining a target gear based on the main driving gear, the basic gear and the current running data; the target gear is a gear that approaches the main driving gear under the premise of meeting the power demand of the vehicle; and controlling the vehicle to perform a gear shifting operation based on the target gear. The technical scheme of the application can reduce the gear shifting frequency during vehicle driving and improve the driving safety of the vehicle.

The invention relates to a low-delay high-definition image processing transmission method and system for a patrolling ammunition, and belongs to the technical field of patrolling ammunition-carried image processing and transmission, and the processing transmission method comprises the steps: obtaining multispectral image data synchronously collected by a visible light sensor and an infrared sensor carried by the patrolling ammunition; performing motion vector calculation and temperature feature extraction on the multispectral image data, and outputting a pixel-level value weight matrix; performing differential coding processing on the multispectral image data to generate a layered compressed data stream; the signal-to-noise ratio of a communication channel is measured in real time, the missile-target distance is solved, the residual electric quantity of the patrol missile is monitored, and a channel state parameter vector is output; querying a preset transmission decision table based on the channel state parameter vector and the hierarchical compressed data stream, and generating a transmission strategy instruction set; and according to the transmission strategy instruction set, performing frequency hopping packaging operation on the layered compressed data stream, and outputting an anti-interference data packet to a ground end. According to the invention, intelligent processing and reliable transmission of the image information of the patrolling bomb platform can be realized.

The invention discloses a quick response time sequence control method for a high-speed array type injection air valve, and the method comprises the following steps: S1, collecting and preprocessing multi-source operation data, and constructing a discrete data set; s2, constructing a decision table and performing attribute reduction to generate a blowing control rule set; s3, bee colony algorithm optimization is executed, and an optimal injection control strategy is obtained; s4, a strategy is issued, an air valve is controlled to execute, and injection feedback data is collected; and S5, feeding back the data to update the data set and the rule, and performing closed-loop optimization on the control flow. The method is used for realizing intelligent time sequence scheduling and closed-loop optimization control of the high-speed array type injection air valve so as to improve the response efficiency, reduce the air consumption and enhance the self-adaptive capability of the system.

A complex system control decision method and system fusing WRT and SPT, the method comprising two-layer architecture: the upper layer classifies the rough granulation of uncertain factors according to the structural information and language and other non-structural information of each related field inside and outside the system, and provides the information required for generating training samples to the causal algorithm of the lower layer, including the corresponding prediction working condition, disturbance event and control measures available at the time of each learning sample example, and efficiently organizes the training process of deep reinforcement learning in high-dimensional decision space; the lower layer adopts a causal-driven deterministic analysis algorithm based on WRT to complete the mapping from system state to control decision; the two layers cooperate to realize the pre-training of decision, and periodically update the pre-decision table according to the results. Once the actual disturbance is detected and identified in the objective system, it is matched with the pre-decision table to quickly complete the control decision.

The invention provides a mining wireless communication access security authentication method and system for industrial cloud computing, and the method comprises the steps: obtaining access authentication multi-dimensional data which comprises a current state space and the position information of an edge node, the current state space comprises current environment parameters collected by the access point and current communication link quality parameters of the access point and the equipment terminal; querying a target action decision table according to the current environment parameters and the current communication link quality parameters, and determining whether an effective network access token is generated or not; if the effective network access token is generated, based on an HMAC-SHA256 algorithm, key seeds are generated according to the access authentication multi-dimensional data and are distributed to the edge node and the equipment terminal, so that the edge node and the equipment terminal generate session keys respectively, user plane communication is established based on the session keys, and security judgment is completed before the network access action of the user plane is executed; and the unauthorized access risk is effectively avoided.

The invention discloses a compressed domain video key frame extraction method and system based on a rough set theory. The method comprises the following steps of: extracting DCT (Discrete Cosine Transform) coefficients of all I frames of a video stream in a compressed domain, acquiring DC (Direct Current) coefficients, constructing an information decision table, calculating an average difference value of the DC coefficients between adjacent frames by utilizing a rough set theory, and eliminating redundant frames through attribute reduction to obtain a primary key frame set; and then optimizing the primary set, initializing a reference frame and a subsequent frame, adopting an SURF algorithm to extract local feature points and calculate similarity, judging content redundancy according to a preset threshold value, and further screening out a final key frame set. According to the method, the key frame is efficiently and accurately extracted in the compressed domain, the processing speed and the content representativeness are considered, and the problems that a traditional method is large in calculation overhead and high in redundancy are effectively solved.

The embodiment of the invention provides a Drools rule engine-oriented cross decision table conversion method and device. The method comprises the following steps of: configuring a service cross decision table; converting the service cross decision table into a service line table; according to the method, the business line table is converted according to the Drools rule engine, the target format file is generated, the target format file is applied to the Drools rule engine, a visual cross decision table structure can be automatically converted into a file which can be executed by the rule engine, multi-dimensional visual modeling and efficient configuration of the business rule are achieved, and the method has the advantages of being high in practicability and easy to popularize. The development and management efficiency of complex business rules is remarkably improved, and meanwhile seamless integration and native compatibility with an existing Drools rule engine ecosystem are ensured.

The application discloses a kind of monitoring alarm signal intelligent research and judge method, device, electronic equipment and storage medium, belong to data processing technical field, including steps S1: obtaining abnormal data, abnormal data include the running state of each fault type associated power equipment, and corresponding operation data and result data;Step S2: establish prediction model and auxiliary model, prediction model is divided into multiple prediction groups;Step S3: based on alarm signal type selection corresponding prediction group, prediction group exports fault type to each auxiliary model, and auxiliary model exports operation data and result data;Step S4: the output result is scored, and based on the scoring result, fault type, operation data and result data are filled into the first table to convert the first table into the auxiliary decision table for alarm signal.It can not only identify corresponding fault according to alarm signal according to the application, and can give corresponding solution according to fault.

The present application relates to a kind of air quality prediction method based on rough set and structural risk minimization, comprising: obtaining environmental parameter sample data related to air quality;Establish air quality evaluation system to the environmental parameter sample data obtained from meteorological monitoring station Grade evaluation creates air quality index decision table;According to air quality index decision table, the empirical error of conditional attribute subset is calculated using rough set theory and structural risk minimization theory and the mutual information of conditional attribute subset and decision attribute;And optimal conditional attribute subset is calculated using genetic algorithm, and the conditional attribute in optimal conditional attribute subset is used as the conditional attribute of rough set classifier to predict air quality with the environmental parameter data of target monitoring point, the prediction model of the present application by rough set theory combines structural risk minimization, on the premise of not reducing the prediction accuracy, while taking into account the complexity of model, the stability and robustness of prediction are enhanced.

The application discloses a robot control method and system based on multi-source information perception and electronic equipment, and the method comprises the following steps: acquiring the brain electrical signal, the inertial signal, the electromyographic signal and / or the force tactile signal of a patient with low motor function; using a random forest machine model to recognize the movement intention for each first signal respectively, and obtaining the corresponding movement intention recognition result; acquiring the weight parameter corresponding to each first signal; preliminarily determining the movement intention of the patient according to the movement intention recognition result and the corresponding weight parameter of each first signal; judging whether the preliminarily determined movement intention conforms to the logic according to the logic decision table; if yes, the final movement intention is determined; determining the movement control signal of the upper limb rehabilitation exoskeleton according to the finally determined movement intention; and controlling the upper limb rehabilitation exoskeleton to move according to the movement control signal. The application can improve the movement intention recognition accuracy of the patient with low motor function in the upper limb rehabilitation training process.

The invention discloses an intelligent routing method and system among multiple agents, and the method comprises the steps: obtaining the dialogue input of a user, and judging the continuity intention corresponding to the dialogue input of the user according to a preset judgment rule; if the continuity intention is the continuity feedback of the old task, directly calling a previous round of processing agent recorded in the dialogue history, and executing task reasoning by the agent as a selected agent; if the continuous intention is a new task, selecting an execution agent corresponding to the rapid mode or the Agent mode according to dialogue input of a user, and taking the execution agent as a selected agent to execute task reasoning; performing type judgment on the content output by the execution task reasoning; and executing different routing choices based on the determined type and a preset decision table. Through mode routing and preset decision table routing, the defects of static routing are overcome, dialogue continuity is improved, and the output quality of the intelligent agent in a complex design task is ensured.

This invention relates to the field of edge computing offloading in the Industrial Internet of Things (IIoT), and discloses a computational offloading method based on evolutionary multi-objective optimization in IIoT. The method includes: constructing a system model; constructing a multi-objective optimization model that minimizes latency and energy consumption; representing the offloading decision for each industrial application as an execution position vector and an execution order vector; solving the optimization problem using a two-layer adaptive multi-objective evolutionary algorithm; initializing the population, weight vectors, and neighborhood relationships; calculating the fitness function of each individual and initializing the reference point; iteratively generating offspring individuals through selection, crossover, and mutation operations, updating the reference point and the non-dominated solution set, and outputting the optimal offloading decision when the termination condition is met. The advantages of this invention are: it can achieve coordinated optimization of latency and energy consumption, effectively improving the task offloading performance of the system.

Provided is a remote sensing image feature discretization method based on rough-fuzzy model, comprising the following steps: listing the digital number in each band and category of a selected sample in remote sensing images, and building an image information decision table based on the digital number and category; initializing the class center of each category and the membership degree of a sample example relative to the class center; updating the class center of each category and the membership degree of the sample example relative to the class center iteratively, and obtaining the final value of the class center of each category and the final value of the membership degree; building a rough-fuzzy set, computing the mean approximation accuracy of the rough-fuzzy set, discretizing the image information decision table, evaluating the discretization results based on the mean approximation accuracy and a genetic algorithm, and selecting an optimal discretization solution.

The invention discloses a distributed intelligent management system of household Internet of Things equipment, and particularly relates to the technical field of equipment management. The method comprises the following steps: acquiring behavior instruction records of household Internet of Things equipment in an autonomous operation state, and analyzing to obtain local optimization target representation; target consistency analysis among multiple devices is executed, device pairs with conflicts are identified, and a conflict influence mapping set is generated; analyzing the behavior conflict relationship and generating a coordination decision table in combination with a family optimization target; issuing the adjusted control instruction to the target household Internet of Things equipment according to the coordination decision table; collecting the actual operation state of the target household internet-of-things equipment, comparing the actual operation state with an expected target, and evaluating the offset degree; and when the offset degree exceeds the tolerance range, generating a re-coordination trigger instruction and updating the coordination decision table. Optimization target fusion and cooperative behavior control among multi-brand household Internet of Things equipment are realized, and the overall operation coordination and intelligent level are improved.

The application discloses a power distribution network fault active research and judgment method and a terminal. A corresponding target decision table is constructed according to the topological structure of a power distribution network. The target decision table contains each condition attribute and a corresponding result attribute. When a fault feature of a fault position in the power distribution network is acquired, the fault feature is matched in the target decision table first to determine a first candidate set of fault elements. Then, a fault diagnosis prediction model corresponding to each candidate fault element is acquired based on the first candidate set. The fault diagnosis prediction model is used to determine a fault credibility corresponding to each candidate fault element. Finally, a fuzzy integral value corresponding to each candidate fault element is determined based on a fault credibility set. The element of the power distribution network where the fault occurs is determined based on the fuzzy integral value, and the active research and judgment of the fault of the power distribution network are realized accurately and quickly.

The invention relates to the technical field of medical health information, in particular to a perinatal period grading management system based on AI risk assessment, which comprises a data acquisition layer, a feature engineering layer, an AI risk assessment module, a five-color dynamic grading module and an intervention push management module. According to the scheme, a multi-scale decision table is constructed by adopting a multi-scale data fusion method, modeling is carried out on data of different sources and different scales, information gain is introduced as a scale selection standard, a high-fidelity fusion feature vector is formed, discriminative information of each data source under the optimal scale is reserved, and a high-fidelity fusion feature vector is formed. High-quality input with a clear structure and a uniform scale is provided for a subsequent AI risk assessment module; an improved Transform model is designed, a time dependence matrix and a dynamic weight adjustment mechanism are embedded, deep modeling of time series data and generation of interpretable dynamic rules are achieved, online self-updating of the rules is supported, and the accuracy and real-time performance of risk assessment are improved.

The invention provides an edge calculation type cultural relic storehouse environment multi-parameter online monitoring method based on POE power supply, and the method comprises the steps: carrying out the protocol unification processing, preprocessing and cleaning of multi-source sensing data, and obtaining clean data; performing multi-source data fusion correlation analysis, and constructing a cultural relic storehouse multi-parameter linkage analysis decision table; obtaining quantitative historical relic damage risk assessment and early warning levels, obtaining historical relic damage risk indexes and early warning levels, and outputting a data uploading scheme and a local backup according to a network state; and constructing a multi-gas cross interference model by using the standard data flow, and outputting high-precision gas concentration data after cross interference compensation. Wiring can be simplified, data processing efficiency, system reliability and maintenance convenience are improved, and an efficient, reliable and easy-to-maintain intelligent solution is provided for cultural relic storehouse environment monitoring.

The invention provides a task scheduling optimization method and device and a storage medium, and the method comprises the steps: updating a scheduling task table, a scheduling decision table and an execution result table in a process that a scheduling edge node executes a task of a client; information of the request task is recorded in the scheduling task table; information of selecting edge nodes according to a scheduling strategy is recorded in the scheduling decision table; information of task execution of the edge nodes is recorded in the execution result table; executing a task on the scheduling edge node to determine a to-be-optimized target index; screening sample data related to the target index from the scheduling task table, the scheduling decision table and the execution result table; and calling the large language model to adjust the scheduling strategy according to the sample data and the target index. The embodiment of the invention has a self-evolution capability, and realizes automation to autonomy, so that the dynamic adaptability is improved, the center of gravity of the scheduling strategy is automatically adjusted according to long-term task mode change and infrastructure change, efficient utilization of resources is kept, and the long-term operation and maintenance cost is reduced.

The application discloses a variable gap GMAW welding penetration control method based on rough set knowledge reduction. The control method adopts a visual sensor, a data acquisition card and other components to collect welding experimental data, under a rough set modeling mode, the collected experimental data is processed by a data preprocessing module to construct an original information decision table, and then is processed by an attribute reduction module and a rule reduction module to construct a rough set model. Under an online control mode, the collected experimental data is processed by the data preprocessing module and matched with the rough set model constructed under the rough set modeling mode, a fuzzy membership function is established by using a fuzzy set theory, a current change amount is output by fuzzy logic reasoning to perform penetration control. The application realizes rough modeling of variable gap welding and penetration control by using a fuzzy controller, has simple system composition, strong anti-interference capability and good engineering practicability, and can achieve the purpose of online penetration control under variable gap conditions.

The invention discloses a regulation and control method for machine-made sand concrete construction, belongs to the technical field of building material concrete preparation, and aims at solving the problems that in the prior art, due to characteristic fluctuation of machine-made sand, the working performance of concrete is unstable, regulation and control depend on artificial experience, and systematicness is lacked. The method comprises the following steps: firstly, establishing a reference mix proportion and machine-made sand standard characteristic file, and carrying out initial adaptation by comparing actually measured characteristics; in the production process, the water content of the machine-made sand is continuously monitored, the driving current of a stirrer and the apparent state of a mixture are synchronously monitored, and when the parameters exceed preset threshold values, it is judged that the performance deviation risk exists; then, aiming at the risk indicated by the moisture content fluctuation or the current and apparent state, respectively executing hierarchical regulation and control based on a calculation or pre-stored decision table; and after each production batch is finished, the material consumption compensation relation and the decision parameters are revised according to the recorded production data. The method is mainly used for realizing self-adaptive regulation and control of the working performance of the machine-made sand concrete in the production process.