92 results about "Fuzzy decision" patented technology

The invention relates to the technical field of building engineering, and discloses a building equipment energy consumption data driven self-adaptive energy-saving regulation and control system and method, and the method comprises global state sensing, energy consumption feature extraction, dynamic modeling prediction, multi-objective optimization solution and hierarchical execution control. Time sequence and space data are fused through Kalman filtering, unified modeling of cross-protocol equipment data is achieved, equipment operation, environment, weather and control instructions are integrated according to a space-time state input matrix, a multi-dimensional decision basis is formed, a Pareto optimal solution set is solved based on an NSGA-II algorithm, a compromise solution is selected through a fuzzy decision method, and the optimal solution set is obtained. According to the method, conflict targets of energy consumption, comfort and equipment life are balanced, a priority arbitration problem is solved by adopting dynamic weight allocation, meanwhile, a security fault-tolerant mechanism is established, a mobile security boundary model and a self-adaptive retry mechanism are introduced, an operation threshold is dynamically adjusted, and the robustness of the system is remarkably improved.

The invention relates to the technical field of lower limb rehabilitation training, and discloses an intelligent load lower limb rehabilitation training system fused with a multi-source sensor. The system comprises a multi-source sensing module, a motion intention recognition module, a self-adaptive track generation module, a real-time load regulation and control module and an actuator driving module. The multi-source sensing module synchronously collects plantar pressure, joint angles and electromyographic signals during training of a user through a plantar pressure sensor array, a knee joint angle sensor and a surface electromyographic sensor; the motion intention recognition module extracts time domain and frequency domain features from the multi-source data, and obtains a real-time motion intention through classification; the self-adaptive track generation module generates an exclusive three-dimensional motion track in combination with the real-time intention and historical rehabilitation data of the user; the real-time load regulation and control module generates a self-adaptive load control signal through a fuzzy decision algorithm; the actuator driving module adjusts the magnetic powder brake through PWM current, outputs corresponding real-time resistance torque, and assists in optimizing the rehabilitation training effect.

The invention relates to the technical field of intelligent window scheduling, in particular to a multi-objective optimization intelligent window dynamic scheduling method and system combined with business rules. The method comprises the following steps: acquiring reservation queuing data, customer demand characteristics and window state data associated with each service window in real time; based on real-time collected data, a multi-objective optimization function including global average waiting time minimization, overall service efficiency maximization of all windows and global resource utilization balance degree maximization is constructed; solving the multi-objective optimization function based on a sorting genetic optimization algorithm; selecting an optimal scheduling strategy through a multi-factor fuzzy decision maker; and dynamically allocating window service resources according to the optimal scheduling strategy, and adjusting a window service queue in real time. By constructing the multi-objective optimization function, the resource waste problems of partial window congestion and partial idle windows caused by traditional fixed rule scheduling are effectively avoided, and the overall rationality and efficiency of window scheduling are improved.

The invention belongs to the technical field of energy storage regulation and control, and particularly relates to an energy storage cluster multi-modal optimization regulation and control method and system based on wide-area measurement data, and the method comprises the steps: collecting the power data of each measurement point, unifying the data of each measurement point to a common time reference system, and then carrying out the preprocessing of the data of each measurement point; aiming at the energy storage system, constructing a ramp rate model, an efficiency characteristic model and a capacity attenuation model, and establishing an energy storage state index system; establishing a multi-mode regulation and control strategy for an energy storage operation state, and dynamically selecting an optimal control mode according to different operation scenes and energy storage states; according to the method, a scheduling framework with day-ahead-intra-day time layering and superior-subordinate space grading is constructed, a day-ahead scheduling model and an intra-day scheduling model are established accordingly, the day-ahead scheduling model and the intra-day scheduling model are rapidly solved based on fuzzy decision, and global optimization of cluster resources is achieved. According to the method, global collaborative optimization of energy storage resources is realized by fusing high-precision wide-area measurement data processing and a multi-mode regulation and control strategy.

The invention discloses a distributed waterlogging and ponding digital map prediction system based on edge calculation, and relates to the technical field of urban flood prevention intelligent decision making. The system comprises a distributed intelligent acquisition module, an edge integrated processing module, a multi-stage collaborative analysis module and a visualization and decision output module. Collecting multi-source heterogeneous data through multiple types of sensing equipment; calling a local digital elevation map slice and a lightweight hydrological model, generating local waterlogging state information and self-calibrating model parameters; key information is screened based on double threshold values of information entropy change and accumulated water depth change rate, multi-monitoring-point data and a drainage pipe network topological relation are fused, and regional-level inland inundation situation analysis and accumulated water and recessive water trend prediction are achieved; and generating and updating a whole-region distributed ponding digital map, and outputting graded early warning and auxiliary decision information. According to the method, distributed acquisition and edge calculation are combined, the monitoring real-time performance and precision are improved, and the problems that traditional monitoring is high in delay, low in precision, fuzzy in decision and the like are effectively solved.

The invention belongs to the technical field of active power distribution network optimization scheduling, and discloses an active power distribution network day-ahead scheduling method, system and device based on an improved NSGA-II algorithm, and a medium, so as to solve the problem that the existing scheduling method is difficult to consider multiple targets such as operation economy, new energy consumption and voltage safety at the same time. The method comprises the following steps: establishing a day-ahead scheduling model taking minimization of the total operation cost of the power distribution network, minimization of the new energy abandoned electric quantity and minimization of the system voltage deviation as optimization objectives, and setting corresponding power grid operation constraint conditions; carrying out iterative solution on the day-ahead scheduling model by adopting an improved NSGA-II algorithm to obtain a Pareto solution set; and on the basis of the Pareto solution set, a compromise solution is selected through a fuzzy decision method to serve as a day-ahead scheduling scheme. Compared with a traditional single-target or empirical scheduling method, on the premise that the voltage safety is ensured, the system operation cost can be remarkably reduced, and the utilization rate of renewable energy sources can be improved.

The invention provides a park integrated energy optimization scheduling method and system, and relates to the technical field of energy scheduling, and the method comprises the steps: obtaining the state information of a park integrated energy system; establishing an energy scheduling model according to the state information; determining a constraint condition and a target function of the energy scheduling model; solving the energy scheduling model through an epsilon-constraint algorithm under the constraint of a constraint condition by taking the minimization of the objective function as a target, and generating a scheduling solution set; processing the scheduling solution set in a fuzzy decision mode to generate an optimal scheduling solution; and taking the optimal scheduling solution as a scheduling scheme, and carrying out optimal scheduling on the park integrated energy system. The energy utilization efficiency can be improved, and the energy safety and flexibility of the park are enhanced.

The invention provides a multi-network collaborative coal mine integrated energy system scheduling method and system under virtual energy storage support, and the method comprises the steps: building a coal mine heat supply network operation model containing multiple heat sources and a coal transportation network operation model, and obtaining a multi-network coupling relation of power supply, heat supply and coal transportation through analysis; constructing a coal mine multi-network collaborative optimization operation model; the method comprises the following steps: analyzing flexible operation characteristics of multiple links of coal mine drainage, coal transportation and heat supply from coal mine production and energy supply link processes, and establishing a multiple virtual energy storage fusion operation model of coal mine drainage, coal transportation and mine heat supply; constructing a multi-objective optimization scheduling model by taking the minimum comprehensive energy consumption cost of the coal mine as an economical objective and the minimum node voltage deviation as a safety objective; and solving the multi-objective optimization scheduling model by using a normalization normal constraint method to obtain a uniform Pareto leading-edge solution, and deciding a compromise solution from the obtained Pareto leading-edge solution through fuzzy decision to obtain an optimal operation scheme of the system.

The invention discloses an automobile low-voltage storage battery intelligent electricity supplement model based on multi-physics field coupling analysis, which mainly realizes accurate state evaluation and dynamic electricity supplement strategy optimization of a low-voltage storage battery by establishing a nonlinear state space equation and a fuzzy decision rule. The system specifically comprises a multi-parameter coupling modeling module used for evaluating the terminal voltage of the storage battery, a battery state evaluation module used for combining the terminal voltage to evaluate the charge state of the storage battery, a fuzzy decision module used for determining the charging voltage and current according to the charge state, and a control module used for determining the charging current and the charging voltage in the charging process through the charging voltage and the charging current. And the self-adaptive correction module is used for identifying the internal resistance of the battery on line and carrying out dynamic voltage compensation. According to the method, all key factors can be comprehensively integrated, accurate estimation of the capacity of the low-voltage storage battery is realized, so that the electricity supplementing accuracy is improved, the service life of the low-voltage storage battery is prolonged while the power consumption is reduced, and the method is suitable for accurate evaluation of the battery state and optimization of the electricity supplementing strategy when the vehicle is in a dynamic working condition.

The present application belongs to the field of comprehensive energy planning, and provides a kind of comprehensive energy system planning robust optimization configuration method and system.The method comprises, obtaining the resource condition and load demand parameter of comprehensive energy system, establish the IES optimization configuration model with the maximum internal rate of return as optimization target;Establish the uncertain set model of wind and light output and cold and heat load demand parameter, based on IGDT combined with IES optimization configuration model, construct IES robust optimization model facing investment income;Simplify IES robust optimization model, solve the maximum uncertainty fluctuation radius and the uncertain scene parameter of the corresponding worst scenario;Under the condition that the uncertain scene parameter is determined, the optimization configuration scheme of maximizing investment income is solved;According to different expected income deviation calculation forms IES optimization configuration scheme set, adopt fuzzy decision, select optimal scheme.Maximize the investment rate of return as optimization configuration target, and based on information gap decision, under the premise of reaching expected income, solve configuration scheme.

The application provides an AC / DC hybrid micro-grid multi-objective scheduling method and system based on digital twinning, relates to the technical field of micro-grid scheduling, and comprises the following steps: establishing an AC / DC hybrid micro-grid digital twinning model and dynamically integrating grid related data; and establishing a comprehensive total cost model of a small AC / DC hybrid micro-grid system; the application combines photovoltaic power generation, photo-thermal power generation and wind power generation, and an energy storage device, establishes a device dynamic efficiency model based on the influence of local load, external grid environment and other factors on the efficiency of the AC / DC hybrid micro-grid equipment, obtains various twinning data such as local load and external grid environment through the digital twinning model, takes economic benefits and the health degree of the energy storage device as the target, proposes a multi-objective particle swarm optimization algorithm based on an improved file maintenance strategy and a fuzzy decision method, and optimizes the output of the energy storage device in real time, thereby improving the multi-energy complementation and conversion capacity of the system.

This invention relates to the field of power load decomposition technology, and discloses a non-intrusive power load decomposition method and system based on multimodal feature learning. The method involves: synchronously collecting power parameter data from smart meters, environmental parameter data from environmental sensors, and user equipment usage behavior data to obtain multimodal load monitoring data; extracting cross-modal features through the non-negative matrix decomposition layer of a first equipment status recognition model to obtain a multimodal fusion feature vector; identifying load patterns through the first decomposition layer of the first equipment status recognition model to obtain a first decomposed load matrix; performing clustering optimization through the second decomposition layer of the first equipment status recognition model to obtain a second load decomposition matrix; and performing dynamic fuzzy decision-making based on the second load decomposition matrix to obtain the equipment operating status identification result. This invention overcomes the limitations of traditional methods that rely solely on a single power signal, achieving high-precision and highly robust non-intrusive power load decomposition.

The invention discloses a small-curve-radius heavy-load steel rail collaborative maintenance method based on abrasion rate dynamic regulation and control, and belongs to the technical field of railway track maintenance. The core of the method lies in that a closed-loop control system with an abrasion rate threshold value (0.5 mm / ten million ton passing total weight) as a decision center is established. According to the method, the side abrasion amount of the steel rail is monitored regularly, the abrasion rate is calculated, a calculation result is compared with a preset threshold value, and a dynamic decision is made and a lubrication strategy is executed; designing an asymmetric polishing target profile based on the lubrication state; performing staged fine grinding; and differential maintenance is carried out in a long-term service period. According to the method, a maintenance decision is upgraded from a fixed period to a precise state response, deep coordination of lubricating and grinding is achieved, the problems that in the prior art, decision basis is fuzzy, and a coordination mechanism is lacked are solved, the service life of the steel rail can be remarkably prolonged, and the maintenance cost of the whole life cycle is reduced.

The application provides a park comprehensive energy optimization scheduling method and system, and relates to the technical field of energy scheduling. The method comprises the following steps: obtaining state information of a park comprehensive energy system; establishing an energy scheduling model according to the state information; determining constraint conditions and an objective function of the energy scheduling model; under the constraint of the constraint conditions, solving the energy scheduling model by an epsilon-constraint algorithm to generate a scheduling solution set, processing the scheduling solution set by a fuzzy decision method to generate an optimal scheduling solution, and taking the optimal scheduling solution as a scheduling scheme to optimize the park comprehensive energy system. The application can improve energy utilization efficiency and enhance park energy security and flexibility.

The present application belongs to the technical field of frequency modulation control, and particularly relates to a heat supply unit integrated auxiliary frequency modulation intelligent control method and system. The method obtains unit and heat supply subsystem operation parameters, combines a seasonal collaborative prediction coupling model with a built-in Bayesian model, obtains a seasonal prediction frequency modulation boundary space and accuracy, generates a seasonal load regulation distribution value based on the prediction result and frequency modulation evaluation rules using a multi-dimensional fuzzy decision optimization algorithm, obtains frequency modulation result deviation and decision success rate through real-time simulation and evaluation, and dynamically feeds back and optimizes the seasonal collaborative prediction coupling model or updates the control strategy list according to the evaluation result. The present application realizes intelligent collaborative control of the unit and the heat supply system, and significantly improves the response accuracy of the AGC load instruction and the system frequency modulation economy.

This invention discloses a method and system for autonomous decision-making control of coal-fired power units under all operating conditions. The method includes: real-time acquisition and intelligent fusion of multi-source heterogeneous data; identification of monitoring blind spots and execution of blind spot data completion or validity verification; construction of an intelligent control system architecture covering all operating conditions of the unit, decentralizing the functions of the plant-level monitoring information system to the production control area; acquisition of electromagnetic radiation interference characteristic parameters and environmental interference characteristic parameters, and implementation of electromagnetic radiation interference suppression or environmental interference correction based on judgment conditions; fusion of a large-scale artificial intelligence model with a dynamic feature decoupling network and a decision knowledge graph to form a quasi-expert fuzzy decision-making system, executing an integrated control strategy of intelligent control, cruise control, and monitoring; and construction of a high-fidelity digital twin parallel system for offline verification and iterative optimization of the control strategy. This invention achieves autonomous decision-making and near-zero human intervention for coal-fired power units under all operating conditions, improving operational safety and economy.

The invention relates to an intelligent decision-making and control method and platform for a multi-energy heterogeneous coupling energy system of a coupling electric power spot market. The method comprises the following steps: step 1, performing multi-source data fusion and multi-time scale prediction; step 2, generating a multi-objective optimization strategy; step 3, performing large model enhanced fuzzy decision; 4, hierarchical real-time coordination control is carried out; and 5, closed-loop feedback iterative optimization is carried out. The platform comprises a data access and prediction module, a multi-objective optimization engine module, a large model decision agent module, a real-time coordination controller module and a feedback learning module. According to the method, a prediction-optimization-decision-control-learning closed-loop mechanism is constructed by utilizing the deep reasoning capability and multi-time scale prediction data of a large model agent and combining real-time information of the power market and the operation state of equipment, so that the limitation of insufficient adaptability of a traditional method to a complex power market environment is effectively solved.

A man-machine collaborative quality detection and early warning system for intelligent factories, comprising an edge visual acquisition terminal, a cognitive anchor generation module and a hybrid intelligent inference engine, the hybrid intelligent inference engine extracts a high-dimensional feature vector of a workpiece image using a residual network and calculates a Mahalanobis distance. When the distance is in a fuzzy decision interval, a man-machine collaboration request is triggered, and a feature heat map is displayed on the terminal. The cognitive anchor generation module receives defect key pixel points labeled by an operator, maps them to a high-dimensional feature space to construct a feature repulsion sphere, and adds the feature repulsion sphere as a regularization term to the loss function of the residual network for online fine-tuning. The scheme corrects the feature boundary in a targeted manner without changing the weight distribution of the model main body, overcomes long-tail defect missed defects, and improves the feature boundary stability of the detection model under continuous production conditions.

The multi-target node load alignment optimization method based on the alternating current power flow establishes an alternating current power flow model to solve a node load alignment, is used for guiding and exciting demand response, and comprises the following steps: 1) establishing the alternating current power flow model to calculate a node load alignment NCDL, and constructing a multi-target optimization model by taking minimum active power loss and voltage offset as target functions, taking a power grid security constraint and a directrix constraint as model constraints; 2) adopting a second-order cone relaxation method to modify a power flow balance constraint in the model, and introducing a method based on boundary correction iteration to reduce a second-order cone error; 3) based on the Pareto optimal set, introducing a multi-target screening mechanism fusing fuzzy decision and TOPSIS to select a solution with optimal global balance; and 4) carrying out demand response according to the optimized and solved NCDL. The invention provides a multi-target node load directrix optimization method based on alternating current power flow, which guides users to actively participate in regulation and control, comprehensively improves electric energy quality, improves system operation economy and ensures safe and stable operation of a power system.

The invention discloses an intelligent monitoring method and device based on double locking, equipment and a medium. The method comprises the following steps: acquiring a double-locking device digital map corresponding to a target production workshop and a target protection result corresponding to a target protective cover; performing risk level identification on the target protection result based on a preset risk level division rule and a preset fuzzy decision method, and determining a target risk level corresponding to the target protection cover; and the target risk level is visually displayed based on a digital map of the double-locking device, so that an intelligent monitoring process is realized. Through the technical scheme of the invention, the problem of difficult supervision caused by potential safety hazards easily generated due to failure of a double-locking device can be overcome, and the characteristic of the intrinsic safety level is remarkably improved.

The invention discloses a water level threshold value dynamic adjustment method and system based on automatic drainage of a transformer substation, and relates to the technical field of transformer substation hydraulic monitoring, and the method comprises the following steps: obtaining water level data of a transformer substation drainage pipe, and classifying water level states based on the water level data to obtain a water level state set; constructing a membership degree input set based on the water level state set, and performing fuzzy logic reasoning on the membership degree input set to obtain a drainage pump control signal; adjusting an interval range of a preset membership function according to the drainage pump control signal to obtain a water level fuzzy decision set; state transition is carried out on a drainage pump control signal based on a water level fuzzy decision set, a transition control threshold sequence is obtained and discretized into a drainage pump start-stop command capable of being recognized by a PLC, and the problem that drainage is not timely or frequent start-stop is possibly caused due to the fact that a drainage threshold value is fixed and inflexible due to water level fluctuation in substation drainage pump control is solved.

This invention relates to a vehicle differential lock control method, terminal device, and storage medium. The method includes: S1: constructing a first fuzzy membership function, a second fuzzy membership function, and a fuzzy rule inference table; S2: real-time acquisition of environmental values, road grade, and road curvature values, obtaining the maximum value of road curvature; S3: obtaining all corresponding first fuzzy intervals and fuzzy values ​​from the first fuzzy membership function based on the road curvature ratio; S4: obtaining all corresponding second fuzzy intervals and fuzzy values ​​from the second fuzzy membership function based on the environmental values; S5: matching the first fuzzy intervals with the second fuzzy intervals to obtain a combined fuzzy value of the fuzzy interval combination; S6: taking the fuzzy interval combination corresponding to the maximum value as the fuzzy decision combination, and obtaining the decision result from the fuzzy rule inference table; S7: controlling the differential lock based on the decision result. This invention achieves timely differential locking of the vehicle in weather conditions such as rain and snow when slight slippage occurs during normal driving.

This invention discloses a method for graded identification and variable spraying of cotton field weeds based on feature weight enhancement, with the following steps: 1. Multi-channel data decoupling and preprocessing: Acquire cotton field images, extract high-frequency edge features using Laplacian transform to construct texture channels, and construct color channels based on RGB components; 2. Heterogeneous feature embedding and dynamic weighting: Map the dual-channel features to a high-dimensional space, and perform nonlinear weighting enhancement on the texture feature matrix based on the difference in edge curvature between cotton seedlings and weeds; 3. Cross-modal feature fusion and accurate identification: Input the enhanced features into an improved Transformer encoder, use a color semantic mask to focus on vegetation areas and suppress background interference, and output a mask for weed category and location from the decoder; 4. Growth parameter extraction and grade evaluation: Calculate the weed pixel coverage per unit area as a growth density index based on the identification results; 5. Graded decision and prescription map generation: Input the weed species, coverage area, and growth density into a fuzzy decision model to determine the disaster level, match the pesticide spraying dosage, and generate a variable spraying prescription map. This invention improves the accuracy of similar weed identification and the system's adaptive capability in complex field environments by using dynamic feature weighting and multimodal fusion.

The application relates to a cross-domain few-shot named entity recognition method based on coupling of Gaussian distribution and Wasserstein metric, which comprises the following steps: obtaining source domain text data and target domain text data to be processed, and respectively performing data preprocessing; obtaining deep hidden layer state features of each word element; generating a multivariate Gaussian distribution representation with position perception capability; constructing a class Gaussian prototype representing the distribution features of each entity category; and finally outputting the predicted entity category of the text sequence to be measured. By establishing the generation condition of the covariance on the basis of the semantic mean position, the application realizes position-perception uncertainty modeling, so that entities close to the fuzzy decision boundary naturally obtain greater variance, accurately cover the effective semantic radius, and even in the case that the feature distribution of the source domain and the target domain is not overlapped, the application can still provide smooth geometric transmission cost, ensures the stability of optimization, and greatly improves the accuracy of cross-domain named entity recognition.

The present application relates to the field of light storage collaborative control technology, and discloses a photovoltaic cluster collaborative scheduling method and system based on multi-source data fusion, obtains light storage cluster operation state data input based on a support vector machine agent model, and outputs active and reactive operation boundaries; monitors working condition mutations, extracts the historical optimal scheduling output corresponding to the support vector as a guide particle, injects the initial population to complete the hot start initialization of the multi-objective particle swarm optimization algorithm; the initial population is iteratively optimized by using the multi-objective particle swarm optimization algorithm, the inertia weight containing the grid feedback component is adaptively updated according to the grid-connected point voltage deviation amplitude, the scheduling instruction candidate set is forcibly constrained by using the operation boundary, and the non-inferior solution set is output; the optimal solution is selected by fuzzy decision and converted into a drive signal to issue the inverter. The present application eliminates blind iteration to establish a deterministic convergence starting point, and simultaneously realizes converter operation boundary protection and grid transient support.

The embodiment of the invention provides a simulation planning execution method of an intelligent agent, which can be applied to the technical field of artificial intelligence. The intelligent agent simulation planning execution method comprises the steps of calling a preset skill library and a preset action library to perform a first simulation planning process in a second planning execution process based on scene understanding information and corresponding execution abnormity feedback information in a first planning execution process, and generating a fuzzy decision path; according to task execution feedback information of the fuzzy decision path, calling a preset skill library and a preset action library to perform a second simulation planning process in a third planning execution process, and generating a target decision path; and through the target decision path, executing a target skill and a target action consistent with the corresponding code of the fuzzy decision path in the first simulation planning process in a third planning execution process, and completing simulation planning execution. The embodiment of the invention further provides a simulation planning execution device and equipment of the intelligent agent, a storage medium and a program product.

The present application relates to a kind of video scene boundary detection method based on multi-modal decision fusion, belong to video scene boundary detection technical field, solve the problem of low accuracy rate of scene boundary detection in prior art while adapting to multiple contents.The method comprises the following steps: carrying out shot boundary detection to the video to be detected, obtain each shot in the video to be detected;In each mode, judge whether each shot is transition boundary, obtain the boundary decision of each mode;Based on the boundary decision of each mode, construct multi-modal fuzzy decision space, based on multi-modal fuzzy decision space, carry out scene boundary decision and obtain the scene boundary of the video to be detected.It realizes more accurate scene boundary detection while facing multiple contents.

The invention discloses a collaborative collision avoidance method, system and device for an automatic driving automobile, and relates to the technical field of automobile collision avoidance, and the method comprises the steps that the automatic driving automobile collects current data in an emergency scene; the current data are input into a fuzzy decision unit and a deep learning decision unit which are generated in advance, and the fuzzy decision unit calculates the fuzzy membership degree of a decision result based on a fuzzy rule base and generates a first collision avoidance decision behavior based on the size of the fuzzy membership degree; the deep learning decision-making unit outputs a second collision avoidance decision-making behavior based on a deep learning model; based on the respective decision weights, performing fusion decision making on the first collision avoidance decision behavior and the second collision avoidance decision behavior, determining a final cooperative collision avoidance behavior so as to control the autonomous vehicle to execute corresponding operation, and sending collision avoidance prompt information to the target object; according to the method, a cooperative collision avoidance mechanism is adopted to effectively solve the special emergency scene that the collision avoidance mechanism of the self-driving automobile cannot avoid collision.

The application provides a multi-modal data security analysis technology based on a deep fuzzy wavelet learning model. The deep fuzzy wavelet learning model is widely used in the field of security analysis due to its self-adaptive ability in complex data environment. The method improves the deep fuzzy wavelet learning framework and applies it to multi-modal data security analysis. The model first uses the method combining wavelet transform and fuzzy clustering to perform feature decomposition and noise reduction processing on multi-modal data (text, speech, video, image), extracts key semantic, timing and spatial features, and provides high-dimensional information representation for subsequent security analysis. Secondly, a deep fuzzy wavelet neural network is constructed, the extracted features are input into the adaptive fuzzy decision layer, and the wavelet learning model is combined to capture the spatio-temporal correlation features, improve the perception ability of the model to complex attack patterns. Finally, the security analysis result is strengthened by using the space search optimization algorithm, so as to effectively detect malicious samples, identify backdoor attacks, and improve the defense ability against attacks, and finally realize accurate multi-modal data security evaluation.

This application relates to a method, system, device, and medium for monitoring user body fluid balance, belonging to the field of biomedical engineering technology. The monitoring method includes: acquiring user body fluid state data collected by sensors in real time and performing timestamp synchronization and drift correction to obtain a calibrated multimodal data stream; constructing a three-dimensional body fluid state tensor from the calibrated multimodal data stream; constructing a body fluid dynamic balance model based on the three-dimensional body fluid state tensor, and performing dynamic balance analysis and anomaly detection to obtain anomaly feature vectors; determining the intervention level based on fuzzy decision rules; generating a corresponding candidate intervention instruction set based on the anomaly feature vectors and the intervention level; inputting the candidate intervention instruction set into the body fluid dynamic balance model to predict changes in user body fluid state, and selecting the optimal intervention instruction based on the prediction results; and driving the execution device corresponding to the optimal intervention instruction to adjust body fluid parameters. This application can achieve comprehensive management and precise regulation of user body fluid state.