186 results about "Fuzzy clustering" patented technology

The invention relates to the technical field of instrument multi-task optimization, in particular to an intelligent instrument multi-task real-time optimization method and system based on dynamic resource scheduling. The optimization method comprises the following steps: acquiring a target item of each task in real time through a sensor array, constructing a multi-dimensional feature vector, dividing each task into task categories by using a fuzzy clustering algorithm, and presetting an initial priority for the task categories for multi-task feature parameter acquisition and classification modeling. According to the method, the multi-dimensional feature vectors including the task urgency degree, the calculation complexity and the data interaction frequency are constructed, and the fuzzy clustering algorithm of the task dependency constraint is introduced, so that the task categories are accurately divided, the cross-category interaction overhead of the dependency task is effectively reduced, and the compatibility of a scheduling strategy is improved from the source.

The invention discloses a smelting waste gas recycling optimization system based on fuzzy control, and relates to the technical field of automatic control. The problems of low control precision and response lag caused by the fact that an existing static membership function cannot adapt to multi-scale working condition changes in real time are solved. Comprising a sensing acquisition module, a membership reconstruction module, a feature decoupling module, a coupling compensation module, a rule optimization module and an execution control module. Standardized data are acquired through sensing acquisition and filtering, an adaptive membership function is constructed by adopting incremental fuzzy clustering, and closed-loop self-correction of a multi-modal control instruction is realized by combining wavelet packet decoupling, feed-forward compensation and double-depth Q network reinforcement learning online optimization; the waste gas desulfurization efficiency, energy consumption optimization and treatment system stability of waste gas treatment can be effectively improved.

The invention discloses a circuit breaker image edge detection method fusing spatial constraint fuzzy clustering and lightweight network optimization, and the method carries out the local adaptive threshold calculation through combining spatial constraint FCM and Otsu algorithms, and optimizes the edge detection process of a Canny operator. According to the method, fuzzy classification is carried out on a circuit breaker image by adopting spatial constraint FCM to obtain a strong marginal probability graph; the circuit breaker image is subjected to block processing through local adaptive threshold calculation, a global threshold is generated through integration, and then the global threshold is input into a Canny operator for accurate edge extraction. In order to further improve the detection effect, a lightweight neural network PiDiNet is used to correct a Canny output image. According to the method, the edge detection precision of the circuit breaker image can be effectively improved, and the method is suitable for edge extraction tasks in high-noise and complex background environments.

The invention discloses a micro-seismic signal arrival time pickup method based on fuzzy clustering and variational mode decomposition. The method comprises the following steps: firstly, calculating characteristic functions of attribute characteristics such as a micro-seismic signal mean value, power and kurtosis and normalizing the characteristic functions to obtain a characteristic matrix, then primarily picking up a micro-seismic P-wave initial movement position and a first arrival moment through a fuzzy clustering algorithm, and extracting an effective time window by taking the first arrival moment as a reference point; a variational mode decomposition algorithm is utilized to decompose signals in a time window into K intrinsic mode function components, AIC function values of the components are calculated by means of an akaike information criterion algorithm, a minimum value point is picked up to serve as first arrival time, energy ratios of all the components are calculated, and final micro-seismic P-wave first arrival time is obtained through weighted calculation. The method effectively deals with the low signal-to-noise ratio environment of the underground coal mine, has higher pickup precision and reliability compared with a traditional pickup method, can provide accurate micro-seismic occurrence time and position information for mine safety early warning, and powerfully guarantees the safety production of the coal mine.

The invention discloses a park digital twinning modeling method based on a generative AI technology, and relates to the technical field of digital twinning modeling, and the method comprises the steps: carrying out the alignment of point cloud, image and state data, completing the preprocessing through topological adaptive filtering and multi-resolution voxelization, frequency domain harmonic fusion and hypersurface texture excitation and time delay coupling fuzzy clustering, and obtaining a digital twinning modeling result; constructing a cross-modal spine network driven by a holographic entropy film, generating a hierarchical token through graph attention, and inputting a surge tuned diffusion converter model for iterative denoising and focus decoding to obtain a three-dimensional fragment; and fusing the fragments in a voxel space by using a cross attention kernel, and adaptively updating parameters through an entropy pulse closed loop until errors converge, so as to generate a high-precision digital twinborn model. A cross-modal semantic network is constructed by constructing spectral mapping and a holographic entropy film, a three-dimensional fragment is efficiently generated in a self-adaptive surge tuned diffusion converter framework through focus fusion, and the cross-modal semantic coupling efficiency, the generative reasoning convergence speed and the multi-fragment voxel fusion continuity are improved.

The invention discloses a distributed photovoltaic optimization coordination control system and method, and relates to the technical field of distributed photovoltaic power generation, and the control system comprises a multi-dimensional data collection module, a self-adaptive cluster division module, a robust dominant node election module, a three-dimensional voltage state evaluation module, and a hierarchical coordination control execution module. According to the distributed photovoltaic optimization coordination control system and method, real-time operation data, meteorological prediction data and power grid load prediction data of each node are continuously monitored, comprehensive information support is provided for a control strategy, and when it is detected that voltage abnormity is caused by sudden change of illumination intensity in a certain area, the real-time operation data of each node, meteorological prediction data and power grid load prediction data are monitored. And immediately finishing cluster structure adjustment in an extremely short time by adopting a fuzzy clustering algorithm with constraint conditions based on the electrical distance and the voltage fluctuation trend characteristic value, and ensuring that the node voltage change characteristics in the same cluster are highly consistent.

The application discloses a kind of big data-based optimization analysis method and system of territorial space planning, it is related to big data and optimization analysis technical field of territorial space planning, including real-time data acquisition and dynamic preprocessing, dynamic regional division, multi-model integrated space-time prediction, multi-objective dynamic optimization planning, real-time feedback and adaptive adjustment.The big data-based optimization analysis method of territorial space planning provided in the application adopts space-time weighted fuzzy clustering method, divides the territorial space into several sub-regions according to the spatial coordinates and time attributes of data, constructs the weighted distance regular objective function between data points and regional center, and updates the membership and the position of regional center iteratively, realizes the self-adaptation and space-time smoothing of regional segmentation, effectively captures the dynamic change characteristics in region.

The invention provides a lithium ion battery safety valve opening and failure early warning method based on expansive force, and belongs to the technical field of lithium ion batteries. Battery expansive force and cycle data under different pre-tightening force conditions are collected, statistical features are extracted to construct a state feature set, health state groups are divided by adopting a fuzzy clustering algorithm, and the early warning result is obtained. Establishing a segmented nonlinear mapping model of the expansive force and the internal pressure, performing wavelet denoising and robust differential calculation on expansive force signals, and optimizing an initial expansive force derivative threshold value by analyzing time dispersion at different heating rates; a multi-scale feature fusion algorithm based on hierarchical attention aggregation is utilized to construct a state self-adaptive early warning model to correct a threshold value, and a four-stage early warning mechanism is set to monitor the opening and failure states of the safety valve. The technical problem that the opening time of the safety valve cannot be accurately predicted and self-adaptive early warning cannot be realized under different battery health states and pretightening force working conditions is solved.

The invention discloses a low-voltage distribution box temperature rise abnormity identification method based on an improved fuzzy clustering algorithm. The method comprises the following steps: S1, generating a temperature-current synchronous acquisition data set according to a time sequence; s2, constructing a load-temperature difference weight matrix in combination with the current current amplitude; s3, obtaining a current membership matrix and a cluster center set; s4, calculating the membership slip amount of each measuring point according to the current membership matrix and the membership matrix at the previous moment, and generating a temperature rise trend state sequence; s5, judging whether the temperature rise trend state sequence has a measuring point which continuously slides from the normal cluster to the risk cluster or not; if yes, corresponding cluster centers in the corresponding cluster center sets are extracted and mapped to the physical coordinates according to the measuring point index table, and positioning information containing the measuring point numbers, the types of the electrical parts and the risk levels is generated. According to the invention, the fault positioning accuracy and the operation and maintenance efficiency are greatly improved.

The invention relates to the technical field of submarine cable fault detection, in particular to a submarine cable fault detection system. The method comprises the steps that an adjustable low-frequency pulse injection unit injects a low-frequency pulse excitation signal and collects a response signal generated by a submarine cable fault; the multi-source signal fusion acquisition unit acquires a reflected voltage signal, an acoustic signal and a leakage magnetic field signal, and constructs a multi-modal fusion response characteristic matrix; the intelligent broadband noise suppression unit is used for performing noise suppression processing on each modal sub-matrix to obtain a de-noised multi-modal fusion response characteristic matrix; and the intelligent fault positioning interpretation unit is used for constructing a three-dimensional signal fingerprint spectrogram, performing classification analysis on fault events based on a fuzzy clustering algorithm and a depth feature matching model to obtain a fault type, and performing calculation through a path residual fitting method to obtain a fault position. The invention realizes a detection system which does not depend on a special cable structure, can actively excite and fuse multi-source signals, and can position submarine cable faults at high precision.

The invention discloses an automatic warehousing system and a sorting method thereof, and the system comprises an execution module, a sensing module and a digital twin management module: the execution module is composed of a vertical lifting container, an autonomous mobile robot and a sorting machine; the sensing module collects cargo full life cycle data through a visual sensor, an RFID tag and a reader-writer. The digital twinborn management module constructs twinborn bodies of the cargos and the equipment, dynamically classifies the cargos based on real-time data, automatically responds to inventory abnormity and simulates and generates an optimal warehousing and sorting route. According to the sorting method, data are collected through the Internet of Things and the RFID technology, goods are classified in real time through a fuzzy clustering algorithm, the digital twinborn body serves as a training environment, a multi-device collaborative optimal sorting path is generated through a reinforcement learning model, and dynamic adjustment is conducted according to the real-time state. According to the scheme, intelligent management and efficient sorting of the warehousing system are achieved, the inventory management precision and the equipment cooperation efficiency are improved, and the system is suitable for high-frequency and high-flexibility warehousing scenes.

The invention relates to the technical field of power distribution network fault monitoring, in particular to a fault recognition method and system based on an intelligent fusion terminal, and the method comprises the steps: obtaining an instantaneous multi-dimensional electrical data set at each moment, and constructing an input sample with the current moment as an end point, inputting the input sample into the trained long-short-term memory model to calculate a first fault probability of the input sample; inputting the input sample into a trained optimal fuzzy clustering model to calculate a second fault probability of the input sample; and distributing respective weights for an output result of the long and short term memory model and an output result of the optimal fuzzy clustering model, carrying out weighted fusion on the first fault probability and the second fault probability to obtain a comprehensive fault probability of the input sample, and judging whether the power distribution network has a fault according to the comprehensive fault probability. According to the invention, through multi-source information fusion, model collaborative optimization and dynamic weight distribution, the fault identification precision and response speed of the power distribution network under complex conditions are effectively improved.

The invention provides a construction safety risk grading method and system combined with fuzzy clustering, and belongs to the technical field of building construction safety risk management.The method comprises the steps that firstly, a real-time risk feature set of a construction scene is obtained, and environmental influences, equipment operation and personnel operation features of a construction area are covered; secondly, fuzzy clustering preprocessing is conducted on the real-time risk feature set, fuzzy membership degree distribution and a feature correlation degree matrix of all risk features are obtained, a risk transmission network is constructed based on the result, nodes are risk features, edges are correlation degree parameters between the features, and risk diffusion coefficients of all the nodes are calculated through the risk transmission network; and generating a risk grade division result according to a preset grading rule, and finally outputting a construction safety grading instruction containing the risk area identifier and the corresponding management and control strategy, thereby dynamically and accurately evaluating the construction safety risk.

The invention relates to a converter transformer external interference source positioning method based on ultrasonic wave and electromagnetic wave fusion, and belongs to the field of partial discharge detection and positioning. Performing layered identification on the time-frequency structures of the two signals through a T-F fuzzy clustering method, and distinguishing internal discharge and external interference signals of the transformer; inputting the external interference signal into a dual-channel neural network for feature extraction, wherein the dual-channel neural network comprises a first channel based on CNN and a second channel based on BiLSTM; the first channel is used for extracting local frequency change characteristics of an external interference signal; the second channel is used for capturing the dependency relationship and global time sequence characteristics of the external interference signal in the time dimension; a self-attention mechanism is further introduced into the network for weighting processing; and carrying out splicing or weighted fusion on the extracted high-dimensional features through a feature fusion module, and inputting the high-dimensional features into a full-connection layer to realize three-dimensional space positioning of the partial discharge interference source.

The invention discloses a power system abnormal data clustering and vulnerability detection method based on a fuzzy mean value algorithm, and belongs to the field of power system security. Aiming at the defects of a traditional method in the aspects of data fuzziness, multi-dimensional association and dynamic adaptability, a'data preprocessing-dynamic fuzzy clustering-vulnerability diagnosis' closed-loop process is constructed. 12-dimensional feature vectors including electrical quantities, equipment states and environmental parameters are constructed, and Z-score standardization is combined with a sliding window updating mechanism to carry out data preprocessing; a feature weighted fuzzy C-means algorithm with entropy regularization is used, and a time decay factor and space-time correlation modeling are fused to realize dynamic fuzzy clustering; abnormal clusters are screened through double thresholds, and vulnerability diagnosis and scoring are completed in combination with an association rule base and a three-level vulnerability scoring model. According to the method, through verification of power grids with different voltage levels, the anomaly detection recall rate reaches 95% or above, the operation and maintenance efficiency is improved by 30%, and the accuracy and real-time performance of anomaly detection of the power system are effectively improved.

The invention provides a rolling quality parameter uncertainty quantification method based on mixed entropy-fuzzy clustering. The method comprises the following steps: S1, multi-source data fusion: integrating rolling compaction parameters, meteorological data and real-time monitoring data in an engineering construction process, and constructing a multi-dimensional feature matrix; s2, calculating mixed entropy, namely quantifying the randomness and fuzziness of parameter distribution in combination with information entropy and fuzzy entropy; s3, performing dynamic fuzzy clustering, namely, optimizing a clustering center based on an improved firefly algorithm, and dividing parameter uncertainty levels; step S4, uncertainty contribution degree analysis: quantifying the influence weight of each parameter on the rolling quality through an entropy weight-grey correlation method; according to the method, the information entropy and the fuzzy entropy are fused, and dynamic clustering and an intelligent optimization algorithm are combined, so that precise quantification and hierarchical management and control of the rolling parameter uncertainty are realized.

The invention provides an intelligent cabinet heat dissipation control method and system, and belongs to the technical field of cabinet heat dissipation control. According to the method, data items all have acquisition time and cabinet area identifiers; the method comprises the following steps: dividing a cabinet hot area by using fuzzy clustering based on a historical sample, dynamically modeling by combining an adaptive PID control algorithm, optimizing control parameters through a particle swarm optimization algorithm, adjusting a hot area control model weight coefficient according to real-time feedback, generating a hot area-equipment association map, and constructing an adaptive heat dissipation control network; predicting a temperature trend through a decision tree model by using a correlation map and real-time operation characteristics generated by the network, and generating a fan rotating speed adjusting strategy, an air conditioner power distribution scheme and a multi-stage heat dissipation control strategy according to the temperature trend; inputting a heat dissipation control strategy and an execution result as feedback into the self-adaptive heat dissipation control network, and dynamically optimizing a correlation map and control parameter modeling; and multi-cycle feedback closed-loop optimization control is realized through multi-time feedback and parameter adjustment.

The invention discloses a high arch dam operation modal parameter automatic identification method and system based on discharge excitation. The method comprises the following steps: 1) obtaining a vibration displacement response signal: determining a penalty factor and an optimal decomposition layer number based on an adaptive multivariate variational mode decomposition algorithm to obtain an optimal IMF component of each sensor channel signal, and performing IMF component screening and signal reconstruction through a frequency domain cross correlation coefficient to realize adaptive noise reduction of the signal; 2) establishing a Monte Carlo three-dimensional stability diagram based on a random subspace recognition algorithm driven by a covariance matrix; and 3) modal parameter automatic identification based on an intelligent clustering algorithm. According to the method, the noise is suppressed by automatically optimizing the modal component reconstruction signal of the multi-sensor vibration signal; a Monde-Carlo three-dimensional stability diagram is established in combination with a Monde-Carlo theory and a covariance driven random subspace method to determine a model order, and automatic interpretation of the stability diagram is realized by applying improved fuzzy clustering, so that operation modal parameters of the high arch dam are accurately identified.

The invention relates to the technical field of fault diagnosis, and provides a self-adaptive diagnosis system and evaluation method for faults of aluminum processing equipment, which accurately senses a logic chain of multi-level feature intelligent reconstruction and multi-dimensional health dynamic evaluation through a multi-modal signal. A complete technical system for self-adaptive diagnosis of aluminum processing equipment faults is constructed, the system breaks through the one-sidedness of traditional single-sensor monitoring, through deep fusion and dynamic modeling of multi-dimensional information, the equipment health state is judged from fuzzy qualitative judgment to accurate quantitative evaluation, through benchmark dynamic matching to difference intelligent judgment, the equipment health state is accurately and quantitatively evaluated, and the fault diagnosis accuracy is improved. According to the method, a signal reliability guarantee mechanism is established through a logic chain of multiple calibration iterations, common features are mapped to health levels through fuzzy clustering, a high-dimensional feature space is converted into interpretable health state probability distribution, and the assessment complexity is reduced.

The invention discloses a power renewable resource recovery carbon footprint modeling method and system, relates to the technical field of power renewable resource recovery, is used for solving the problems of reduced regional load capacity and lack of dynamic matching mechanism fusion, and constructs a carbon emission factor library and calculates carbon emission intensity through carbon emission data of each link of power waste and old materials; in combination with the carbon emission intensity, the logistics distance and the processing complexity, a fuzzy clustering algorithm is adopted to carry out risk division on a recovery path; logistics nodes and energy consumption parameters of the high-risk area are collected, and the carbon emission bearing capacity of the area is evaluated; according to the method, the actual carbon emission amount is detected, the carbon footprint weight is calculated through the analytic hierarchy process, weighted summarization is carried out, the total carbon footprint amount of the whole life cycle is obtained, dynamic modeling and accurate evaluation of the whole carbon emission process are achieved, the green management level of a recovery system is improved, and scientific support is provided for an emission reduction strategy.

The invention relates to the technical field of unmanned aerial vehicle group networks, in particular to a dynamic clustering routing method for a high-dynamic unmanned aerial vehicle group network, which is characterized by comprising the following steps of: dividing the whole unmanned aerial vehicle group network into a centralized-distributed hierarchical domain architecture, dividing the network into a plurality of clusters, and enabling each cluster to consist of a cluster head and a plurality of cluster members, the cluster head can be recombined into a higher-level cluster; according to the method, a layered and domain-divided network architecture is provided, cluster heads are dynamically elected in combination with a grey wolf optimization algorithm, real-time self-adaptive clustering and load balancing routing of the unmanned aerial vehicle group in a high-speed mobile environment are realized through a double-cluster-head switching and reselection mechanism, the cluster heads and paths are ensured to be optimal, routing is optimized through a genetic algorithm, relay forwarding is reduced, and the routing efficiency is improved. The communication overhead is reduced, the service life of the network is prolonged, the robustness of the system is enhanced by the alpha-beta-delta leader model and the iterative surrounding strategy, the time delay is reduced and the transmission success rate is improved by the fuzzy clustering, and the real-time reliable communication requirement in a high dynamic scene is met.

The invention provides a financial risk control and anomaly detection method and system based on a graph neural network, and relates to the field of financial risk control, and the method comprises the steps: constructing a financial transaction graph network, constructing a feature graph in combination with transaction time sequence information, extracting a time sequence feature vector, calculating a risk propagation feature vector, and determining a node embedding vector and a fusion node vector. And constructing a transaction flow diagram and carrying out community division, and finally determining an abnormal transaction community through fuzzy clustering and carrying out early warning.

The invention discloses a long-term and short-term cloud energy storage optimal configuration method considering a lease market, and relates to the field of energy storage optimal configuration, and the method comprises the steps: carrying out the clustering of typical scenes of offshore wind power through employing a semi-supervised multi-kernel fuzzy clustering algorithm; according to the characteristics of hydrogen energy storage and battery energy storage and comprehensive performance and economic factors, a long-short-term cloud energy storage coordinated optimization operation strategy including hydrogen energy storage and battery energy storage operation is output; establishing a long and short term cloud energy storage double-layer optimization configuration model considering two-system lease prices and typical scenes; converting the long and short term cloud energy storage double-layer optimization configuration model into a single-layer optimization model based on an optimality condition; and solving the single-layer optimization model by adopting a fungus growth optimization algorithm improved based on a memory mechanism. According to the method, the long-term and short-term cloud energy storage double-layer optimization configuration model considering the lease market is constructed, and theoretical support and decision basis are provided for large-scale application and sustainable development of cloud energy storage through collaborative optimization processing.

The invention discloses a power distribution network small current grounding fault diagnosis positioning method and system based on edge calculation, and belongs to the field of power system automation, and the method comprises the steps: extracting and recognizing transient features, if a suspected grounding fault is detected, carrying out the multi-source feature fusion analysis through a regional main node, and constructing a transient response time sequence map; a weighted fuzzy clustering algorithm is combined with a multi-channel criterion to determine a fault channel, and a specific fault branch is further positioned through a topology tracking algorithm; introducing a multi-scale wavelet packet energy analysis method into the regional main node, performing energy inversion calculation on the transient current signal under multiple frequency bands, and identifying an energy concentration region near the grounding point through an energy distribution gradient; and constructing a robustness data redundancy model based on the historical data of the edge nodes and the topological relation, and identifying and eliminating abnormal feature data containing noise or distortion. According to the method, transient characteristics are rapidly extracted and preliminary judgment is made after a fault occurs, so that the real-time performance of fault response is greatly improved.

The invention relates to an automatic instrument parameter self-tuning method and device based on deep learning. According to the method, real-time operation data of an instrument is collected through a multi-source sensor, time sequence alignment is carried out to form a multi-dimensional data tensor, and then local time sequence features and a long-term dependency relationship are respectively captured by using a convolutional layer and a bidirectional long-short-term memory network in a depth feature extraction network; the two types of features are fused through an attention mechanism to form a depth feature vector, on this basis, the vector is mapped into a working condition membership degree vector by adopting a differential fuzzy clustering method, and finally, the working condition membership degree is non-linearly mapped into a PID parameter adjustment amount through a multi-layer perceptron network, and the PID parameter adjustment amount is superposed to a basic parameter to realize parameter self-tuning. Therefore, under the condition that manual intervention is not needed, an automatic instrument can automatically adapt to complex and changeable operation conditions, the control precision and the system stability are remarkably improved, and the problem of adaptability of traditional PID control in a time-varying nonlinear system is effectively solved.

The invention provides an electric vehicle ordered charging method and system based on an improved whale optimization algorithm, and belongs to the technical field of electric vehicle ordered charging, and the method comprises the steps: constructing a fuzzy clustering model based on threshold value optimization according to the electrical load prediction data of a micro energy grid user side, an improved particle swarm algorithm is adopted to carry out optimal division of the peak and valley periods of the micro-energy network on the fuzzy clustering model based on threshold optimization; according to historical electric vehicle charging load data, a charging load prediction method based on variable-variable variational mode decomposition and a long-short-term memory neural network is adopted; according to the electrical load prediction data, the photovoltaic output prediction data, the time-of-use electricity price and the electric vehicle charging load prediction data of the micro energy grid user side, the double-layer multi-target optimization model is solved by adopting an improved whale optimization algorithm based on a hybrid reverse learning strategy, and an optimal ordered charging strategy of the electric vehicle is obtained. According to the method, the electric vehicle can be charged more orderly, and the influence on a power grid is smaller.

The present application relates to the technical field of rail transit device fault monitoring, and in particular to a turnout power curve state monitoring method and apparatus. The method comprises: acquiring first feature information of a real-time turnout action power curve; acquiring second feature information of turnout action power curves in different operating states; on the basis of a grey relational analysis method and / or a fuzzy clustering method, calculating the similarity between the first feature information and second feature information corresponding to each fault type; and on the basis of the similarity, determining an operating state corresponding to the real-time turnout action power curve. The present application provides an intelligent railway device electrical characteristic curve abnormality diagnosis method and system, which have a higher accuracy, automation degree and robustness. By using a fault diagnosis method based on digital signal processing, fault features of turnout action power curve data are effectively extracted, thereby meeting various fault detection requirements, simplifying the structure of a classifier, and improving the precision and efficiency of fault diagnosis.

The invention discloses a news event prediction method based on heterogeneous evolutionary event clustering, which comprises the following steps: generating event representations based on preliminarily updated entity representations and relationship representations in a constructed entity graph, and constructing an event graph by taking events as nodes and taking heterogeneous relationships between the events as edges; obtaining event clusters through fuzzy clustering and constructing an event cluster graph; using a self-supervised optimization algorithm to optimize the event cluster representation according to the distance and similarity between the event clusters on the event cluster graph; capturing implicit correlation among the event clusters by using an implicit relation encoder, and sequentially updating representation of the event clusters, representation of events, and representation of entities and relations after sparsification and information aggregation; and predicting through the news event model based on convolution. According to the method, the pairwise correlation, the high-order correlation and the multi-step time sequence evolution of the events are effectively modeled, and the method has important application value in the aspects of international situation analysis, social governance, intelligent decision support and the like.

The invention provides a concrete waste crushing particle size distribution analysis method and device, and the method comprises the steps: obtaining concrete waste image data, extracting a particle contour and geometric parameters through image recognition, determining a particle contact point displacement feature, and recognizing a stress concentration region migration path. Obtaining a particle morphology data set containing stress transfer path deflection information; performing gray value processing on the particle surface image, identifying a cement paste stripping area caused by shear stress component change through brightness difference to obtain interface stripping stress state distribution information, and calculating shear breaking ultimate strength distribution characteristics based on stress transfer path deflection information; and fuzzy clustering processing is carried out on the particle data set, if boundary fuzzy particles exist, crushing force transmission characteristics of the boundary fuzzy particles are obtained according to mechanical response time-frequency analysis, and the crushing risk level of the boundary particles is identified in combination with a particle internal crack initiation force threshold.

The invention discloses a hydrogen liquefaction control method and system based on AI decision making. The method comprises the steps that real-time operation data and parameters of a hydrogen liquefaction device are obtained; based on the real-time operation data and the parameters, a fuzzy clustering algorithm is utilized to identify the current working condition so as to determine the working condition type to which the current working condition belongs; based on the working condition type, a reinforcement learning algorithm is utilized to generate a target control strategy, and the target control strategy comprises at least one control strategy of the yield, the energy consumption and the equipment service life early warning of the hydrogen liquefaction device; and according to the target control strategy, utilizing a particle swarm optimization algorithm to adjust PID parameters of the hydrogen liquefaction device so as to control at least one of yield, energy consumption and equipment life early warning of the hydrogen liquefaction device. The response speed, stability and anti-disturbance capability of the control loop of the hydrogen liquefaction device can be improved, dependence on experience of operators is reduced, performance degradation of the hydrogen liquefaction device is avoided, and efficient operation of the hydrogen liquefaction device is guaranteed.