94 results about "Decision table" patented technology

A multi-fault diagnosis method for a power distribution network is provided. The method includes: performing short-circuit fault analysis on a line of a power distribution network by using a MATLAB platform, so as to obtain an electrical quantity-based fault information decision table; performing modeling and simulation on the fault information decision table by using a Simulink platform; denoising and collecting output training data of a neural network of the power distribution network by using a wavelet transformation method, and forming a related fault information decision table as a training sample of the neural network; and optimizing weights and thresholds of the neural network by using an improved Artificial Tree intelligent optimization algorithm, selecting some of the data as fault data, and using the neural network trained to perform fault detection.

The invention relates to the technical field of carrying equipment. The multi-platform compatibility optimization method and system for the carrying equipment controller are provided, and the method comprises the following steps: carrying out dynamic monitoring and topology analysis processing on a network node connection relationship of a carrying equipment cluster, and generating a dynamic network topological graph and a link state data set; performing training processing on the mapping relation between the network state and the protocol parameter, and generating a protocol parameter dynamic adjustment strategy and a protocol decision table; performing real-time switching processing on the cross-platform communication protocol stack to generate a control instruction data stream compatible with multiple platforms; performing dynamic updating processing on a reward function of the reinforcement learning model, triggering protocol parameter optimization to be executed again, and generating an iterative optimization protocol parameter dynamic adjustment strategy; the problems that topology awareness and dynamic response depended on by static routing are missing, protocol switching and compression ratio self-adaptive adjustment of network parameter solidification are invalid, and cross-platform instruction conflicts and transmission efficiency degradation of heterogeneous protocol adaptation are caused are solved.

The invention relates to the technical field of data processing, in particular to an automatic conference recording and abstract generating method for an intelligent conference, which comprises the following steps of: capturing voice streams of multiple speakers in real time through a directional microphone array, generating an original text stream with a time sequence mark based on real-time voiceprint clustering, and synchronously extracting intention intensity parameters in the voice streams; performing intention-driven dynamic segmentation processing on the original text stream; performing agenda-perceived abstract block generation on the segmented text, wherein decision expressions meeting a semantic density threshold value are extracted as abstract core blocks; and assembling the abstract core blocks into a structured abstract document according to the hierarchical structure of the agenda template. Compared with a traditional linear transcription mode based on voice recognition, the method has the advantages that the mapping precision between the speech content and the identity of the participant is remarkably improved, and a solid foundation is provided for semantic segmentation and decision tracing.

The invention discloses an active real-time interaction system based on LLM. The active real-time interaction system comprises a self-cognition module, a real-time perception module, an active planning module, a situation decision module, a dynamic execution module, an emotion engine module, an adaptive interaction module and a safety management and control module. The system adopts a multi-dimensional vector to represent role attributes, and constructs a hierarchical memory storage structure to record interaction experience; generating an environment state vector through a multi-modal data processing technology; executing the target task decomposition algorithm to generate a multi-path execution plan; making a real-time decision based on the multi-dimensional decision factor; mapping the abstract decision into an instruction sequence and monitoring an execution process; simulating a system emotional state and influencing decision expression; dynamically adjusting an interaction strategy according to the user characteristics; evaluating decision rationality and starting a corresponding intervention mechanism. All the modules form a closed-loop workflow through a standardized data interface, the problems that a traditional AI system is passive in response, lacks self-cognition, is limited in environmental perception, is single in planning capability, lacks flexibility in decision making and the like are solved, and active service, self-evolution and safety controllability of the system are achieved.

Systems and methods improve performance of a classifier, which comprises a neural network and is trained through machine learning. First and second scores are computed, by the classifier, for each a multiple data examples from a generator. The first score is indicative of whether the data example belongs to a first data cluster and the second score is indicative of whether the data example belongs to a second data cluster. The generator is trained with an objective such that, for each data example generated by the generator, the first and second scores computed by the classifier are equal. Partial derivatives from the classifier are back-propagated for multiple data examples generated by the generator, to obtain a vector, for each data example, that is orthogonal to a decision surface for the classifier. A problem with the classifier is detected based on changes in directions of the vectors. Upon detecting a problem, the classifier is adjusted to reduce errors by the classifier caused by overfitting training data.

The invention relates to the technical field of routing context connection, in particular to a zero-code visualization and automatic routing method for large-model servitization orchestration, which comprises the following steps of: analyzing a user session text, extracting a core entity and an intention dependency relationship, constructing an entity intention graph, compressing to generate a state snapshot, and distributing a global session ID (Identity); the method comprises the steps that a snapshot is bound to a distributed routing decision tree node and stored, cross-node state continuity is established, when a new request is received, the snapshot and new session semantics are fused, a routing weight offset correction decision table is calculated, and a routing decision logic flow is generated by dragging an interface configuration rule and strategy. The method improves the service matching degree, reduces the development threshold, and is suitable for agile construction of large-model collaborative application.

The invention discloses a hybrid vehicle design method and system based on grey correlation, and belongs to the technical field of vehicle optimization design, and the method comprises the steps: collecting an emotional attribute set of a consumer for a hybrid vehicle, and building a corresponding product attribute set; thirdly, dimension reduction is carried out on the emotion attribute set, and the product attribute set is decomposed; then, constructing a decision table, taking the reduced emotional attribute set as a decision attribute set, taking the product sub-attribute set as a condition attribute set, and calculating the importance degree of each condition attribute to each decision attribute by utilizing an importance degree algorithm; then, a quality house matrix is constructed, the importance degree is imported, and the incidence relation is obtained; and finally, taking the reduction emotional attribute set as a reference sequence set, carrying out grey relational degree analysis on the product sub-attribute set to obtain grey relational degrees and sequences of the product sub-attributes, and carrying out design optimization on the hybrid vehicle according to the grey relational degrees and sequences.

The invention relates to the field of multi-energy coupling complementary scheduling, and discloses a successive approximation random dynamic planning method for a watershed water-wind-light integrated system, and the method comprises the steps: 1, inputting the description of uncertainty; a plurality of characteristic discrete values are sampled from theoretical probability distribution functions of runoff, photovoltaic output and wind power output to represent known continuous probability distribution, and uncertainty of input of the watershed water-wind-light integrated system is described; 2, solving a transition probability matrix; describing the continuity of an input variable time sequence by adopting a simple discrete time Markov chain; 3, considering an objective function and constraint conditions of the watershed water-wind-light integrated system, and constructing a complementary optimization scheduling model; 4, solving by adopting a successive approximation-random dynamic programming method (SDP-SA), and deducing a scheduling rule, namely a scheduling decision table, of the watershed water-wind-light integrated system by taking the time period initial storage capacity, runoff, photovoltaic output and wind power output scenes of the adjustable reservoir in each stage of sub-complementary system as multi-dimensional state variables and the time period final storage capacity as a decision variable; and 5, when an actual inflow condition occurs, looking up the scheduling decision table to perform interpolation calculation according to the reservoir capacity and the inflow condition of the current sub-complementary system and the wind power output and the photovoltaic output which are obtained by forecasting in the period of time, obtaining the reservoir capacity at the end of the period of time, and completing an optimal scheduling decision.

The invention provides a message backlog processing system and a message backlog processing method. The message backlog processing system includes a storage device and a processor. The storage device is used for storing the processor, is coupled with the storage device, and is used for executing the data detection module, the decision module and the calculation module. The data detection module collects a production log, a consumption log, message queue data and a processing data set in the processor. In response to the fact that the message backlog amount is larger than the message backlog threshold value, the decision module obtains a processing strategy according to the decision weight in the decision table and the processing data set. The computing module computes a suitable performance value according to the processing policy and updates the message processing settings according to the suitable performance value.

The invention relates to the technical field of data processing, and discloses a communication unit data transmission optimization method and system for large-scale electric energy data. The method comprises the steps that electric energy data features are extracted through a space-time correlation analysis algorithm, and an electric energy data time sequence dependency matrix and a space correlation matrix are constructed; constructing an EQ-learning intelligent scheduling model based on the matrix, and generating a Q value decision table fused with power grid constraints; performing multi-communication-unit path selection by applying the decision table to obtain an optimal transmission path set; a P-DPDK processor is adopted to realize multi-network card concurrent transmission; and performing dynamic load balancing adjustment according to the transmission result, and optimizing the electric energy data transmission performance. According to the invention, the problem that the electric energy data lacks intelligent transmission scheduling and dynamic resource configuration of spatial-temporal correlation perception in the prior art is solved, and the real-time performance, the reliability and the resource utilization efficiency of large-scale electric energy data transmission are improved.

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 provides a decision routing configuration method and device, a storage medium and electronic equipment, and the method comprises the steps: obtaining a new decision link to be verified, the new decision link comprises at least one new decision expression, and the new decision link is used for carrying out the risk assessment of a credit activity transaction, performing grammar verification on each new decision expression of the new decision link based on a preset grammar rule, and if the new decision link passes the grammar verification, performing semantic verification on each new decision expression and at least one effective decision expression to obtain a verification result corresponding to the new decision link, the valid decision expressions are expressions corresponding to valid decision links included in the decision link set, and if the verification result indicates that semantic overlapping does not exist between all the new decision expressions and all the valid decision expressions, it is determined that the new decision expressions take effect. And the routing configuration and distribution accuracy of the transaction decision link is improved.

The invention belongs to the technical field of wargame deduction, and discloses an emergency wargame deduction system capable of editing total elements. Comprising an operator editing subsystem, a decision table editing subsystem, a map editing subsystem, a chess editing subsystem, a scenario editing subsystem, a training organization subsystem, a deduction work subsystem, a guidance control subsystem, a deduction observation subsystem, a deduction replaying subsystem, a deduction engine subsystem and a system management subsystem. The system can meet the requirements of relevant professional operational scientific research activities on the sea. According to requirements, a deduction scenario is compiled, deduction conditions are set, and scientific research activities are carried out by selecting different weapons, combat deployment, combat actions, guarantee conditions, battlefield environments and the like and adopting a deduction-while-research or deduction playback mode. The war game system can be used as an explanation tool to provide an analysis and research environment, deduce and research are carried out on combat scenarios, classic combat examples and new combat styles and battlemen, and valuable reference data and experience are obtained.

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 present invention relates to the technical field of nuclear fusion tokamak control systems, and in particular to a plasma control system health assessment method based on a decision table and an analytic hierarchy process. The technical solution comprises the following steps: assessing the plasma control system health status based on a decision table; assessing the plasma control system health status based on an analytic hierarchy process; and combining a decision table-based system health assessment method with an analytic hierarchy process-based health assessment method to comprehensively assess the plasma control system health status. By combining the advantages of decision tables and the analytic hierarchy process, the present invention achieves real-time health assessment and precise fault diagnosis for complex distributed control systems, significantly improving the stability, reliability, and safety of the plasma control system and providing strong support for the efficient operation of the system.

The invention provides a cloud platform system based on an intelligent decision chain and a resource availability method. The cloud platform system comprises a monitoring module, a feature library, a decision table and a decision tree, the monitoring module is used for collecting resource data, performing path matching detection in a decision according to a decision table and triggering a decision behavior of a decision chain when monitoring data is abnormal; the feature library is used for recording resource fault feature information; the decision table is used for recording a weight value of each feature point in the feature library, performing priority ranking according to the weight values, and constructing a decision chain in the decision tree; the decision-making tree is used for recording all decision-making links and decision-making selections corresponding to the decision-making links, each leaf node of the decision-making tree corresponds to one decision-making behavior, and when the decision-making path walks to the leaf node of the decision-making tree, the decision-making behavior corresponding to the leaf node is executed. The overall high availability of the cloud platform system is adjusted in time through state monitoring and feedback of each unit and the load condition.

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 complex system control decision method and system fusing WRT and SPT, and the method comprises two layers of architectures: an upper layer carries out the coarse graining classification of uncertain factors according to the structural information and language and other non-structural information of the related fields inside and outside the system, and provides the information needed by the generation of a training sample for a causal algorithm of a lower layer; comprising prediction working conditions corresponding to each learning sample example, disturbance events, control measures which can be called at that time and the like, and a training process of deep reinforcement learning in a high-dimensional decision space is efficiently organized; a lower layer adopts a cause and effect driven deterministic analysis algorithm based on WRT to complete mapping from a system state to a control decision; the two layers cooperate to realize pre-training of decision, and a pre-decision table is periodically updated according to the result. Once the actual disturbance is detected and recognized in the objective system, the actual disturbance is matched with the pre-decision table, and the control decision is quickly completed.

The present invention relates to the field of data processing technology, and specifically to a method for generating automated meeting records and summaries for intelligent conferences, comprising the following steps: real-time capture of multi-speaker voice streams through a directional microphone array, generation of a time-stamped original text stream based on real-time voiceprint clustering, and simultaneous extraction of intent intensity parameters from the voice stream; performing intent-driven dynamic segmentation processing on the original text stream; performing agenda-aware summary block generation on the segmented text, including extracting decision statements that meet a semantic density threshold as summary core blocks; and assembling the summary core blocks into a structured summary document based on the hierarchical structure of the agenda template. Compared with traditional linear transcription methods based on speech recognition, the present invention significantly improves the mapping accuracy between speech content and participant identity, providing a solid foundation for semantic segmentation and decision tracing.

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 embodiment of the present application provides a message forwarding method, device, equipment and medium for a C2F scenario, which belongs to the field of industrial message processing technology. The method includes: sending a first detection short aggregation message to other routing nodes that are topologically associated with the current routing node, receiving a feedback short aggregation message sent back by other routing nodes based on the first detection short aggregation message, and generating a corresponding path decision table based on the feedback short aggregation message; determining the next-hop initial forwarding routing node to which the target message needs to be sent under the initial setting from multiple other routing nodes, and determining the next-hop update forwarding routing node to which the target message needs to be sent in real time; if the initial forwarding routing node and the update forwarding routing node are different, updating the path identifier, and forwarding the target message with the updated path identifier to the update forwarding routing node. The present application can improve the accuracy of determining the current optimal forwarding path of the target message, thereby improving the quality of network communication.

The invention relates to a communication resource scheduling method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the steps that data of all queues and calculation data of all calculation units are obtained in real time, and a global state matrix is generated; detecting the global state matrix, and judging whether a queue of which the load value exceeds a preset threshold value exists or not; if the judgment result is that the queue with the load value exceeding the preset threshold value exists, establishing a resource scheduling optimization function based on the global state matrix; generating an optimal scheduling matrix based on the global state matrix, the resource scheduling optimization function and a preset scheduling optimization algorithm; and based on the optimal scheduling matrix, generating a scheduling decision table, and based on the scheduling decision table, distributing the data packets in the queues with the load values exceeding a preset threshold value to a target calculation unit. By adopting the method, the data packets of different queues can be dynamically scheduled.

A smart streetlight escort method and system for major events, relating to the field of the Internet of Things. Applied to an urban streetlight management platform, the system includes: obtaining real-time road condition and traffic flow data uploaded by smart streetlights on a target road section, where the real-time road condition information includes multiple real-time sub-information; analyzing the influence coefficient of the real-time road condition data on the traffic flow data; determining the information processing ratio of the multiple real-time sub-information based on the influence coefficient; extracting the data volume corresponding to each of the multiple real-time sub-information from the real-time road condition data based on the information processing ratio; adjusting the traffic flow data using the data volume corresponding to each of the multiple real-time sub-information to generate target traffic flow data; matching the target traffic flow data with a preset guidance decision table to obtain a target guidance decision, and sending the target guidance decision to the smart streetlights on the target road section, thereby resolving the issue of smart streetlights being unable to update traffic guidance information in a timely manner.

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 relates to the technical field of road disease detection and maintenance, in particular to an enhanced road disease knowledge extraction method and system based on large model self-learning, and the method comprises the following steps: carrying out the entity and relation extraction of a road disease field standard document through a pre-trained large model BERT, and generating an initial knowledge triple; a road disease knowledge extraction decision table is constructed, a hierarchical relation coding rule is defined, an inter-entity relation strength value is quantized, and the relation strength value is calculated through dynamic fusion of an expert initial score, a data-driven co-occurrence frequency and an attention weight; the method has the beneficial effect that the propagation risk of the error relationship is effectively reduced through the constraint of the relationship weight table and an expert rechecking mechanism. In 1000 pieces of test data, the conflict detection module successfully recognizes and corrects 83.2% of semantic conflicts (such as'seam strip 'error association' subgrade subsidence '), and the false alarm rate is lower than 5%.

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.