214 results about "Causal graph" patented technology

The invention relates to an electrical load prediction and optimization regulation and control method and system for high-energy-consumption equipment, and solves the problems of inaccurate load prediction, single regulation and control means and difficulty in dynamic adaptation of the high-energy-consumption equipment, and the method comprises the steps: collecting multi-source data of the high-energy-consumption equipment in real time, constructing a dynamic equipment collaborative causal graph after preprocessing, and extracting key constraints; inputting the data and the constraints into the dynamic digital sample model to obtain a system state simulation result; based on the result, a multi-objective optimization regulation and control strategy is generated and executed by using a meta-learning + reinforcement learning decision framework; and collecting actual data comparison deviation, starting hierarchical federated learning when a threshold value is exceeded, grouping and aggregating similar experiences according to a causal graph topology, and dynamically calibrating model parameters and a decision framework. The method has the following effects that accurate load prediction and multi-target cooperative regulation and control of the high-energy-consumption equipment are achieved, working condition changes are dynamically adapted, the cost is reduced, and continuous production and the service life of the equipment are guaranteed.

The invention provides a multivariate time sequence anomaly detection method based on an adaptive causal diagram and spatio-temporal evolution, and belongs to the technical field of time sequence anomaly detection. According to the technical scheme, firstly, unification, missing value filling and Min-Max normalization processing are carried out on multivariate time series data, on this basis, a graph attention network is utilized to construct an adaptive correlation graph, a causal relationship between variables is quantized through Granger causal test, then the correlation graph and a causal graph are fused to generate a causal correlation mixed graph, and then, the causal correlation mixed graph is subjected to data processing. And inputting the mixed graph into a space-time converter to carry out future numerical value and structure prediction, finally calculating a prediction residual error and generating a comprehensive anomaly score, and further judging an abnormal node. According to the method, dynamic detection and interpretable analysis of abnormal events can be realized, and the problems that in the prior art, static state, causality and correlation of a graph structure are not fused, structural evolution modeling is lacked, and the judgment dimension is single are solved. According to the method, the anomaly detection coverage and sensitivity are remarkably improved.

The invention relates to a grid-connected scheduling management method and device constructed in combination with a knowledge graph, equipment and a medium. According to the method, a comprehensive data set is constructed by integrating multi-source data such as new energy output, power grid topology, load, weather and historical fault records, and then a dynamic knowledge graph is formed by using entity recognition and relation extraction technologies; a probability causal graph model is constructed by extracting a causal path and adding probability parameters, and uncertainty propagation intensity is quantified in combination with a sequence diagram neural network; on the basis of a propagation model, risk index conditional probability is calculated by adopting probability causal reasoning, and a fault propagation sequence is simulated through a cascade failure theory to realize multi-level risk assessment; based on a multi-objective optimization model and deep reinforcement learning, an adaptive scheduling strategy is generated, a complete technical closed loop from data fusion and causal reasoning to intelligent decision is realized, and the technical effects of describing a new energy uncertainty propagation path, prospectively evaluating a power grid risk situation and dynamically generating an optimal grid-connected scheduling scheme are achieved.

The invention provides a PCB manufacturability intelligent analysis and early warning method and system based on artificial intelligence, and the method comprises the steps: collecting and marking the multi-source time sequence process parameter data in the PCB design and manufacturing process under working conditions, building a dynamic causal graph model with direction and time lag marks through a sliding window and standardization processing by applying a causal discovery algorithm, and carrying out the calculation of the dynamic causal graph model. Dynamic expression of causal relationships among process variables is realized; when manufacturing abnormity is detected, abnormity attribution is carried out by combining a Bayesian back propagation algorithm, high-contribution-degree root dependent variables are screened, the effectiveness of root causes is verified through virtual intervention simulation and statistical test, and finally verification results and a causal mode are stored in a knowledge base to support subsequent rapid matching and reasoning. According to the method, the accuracy, efficiency and interpretability of PCB manufacturing abnormity attribution are improved, and process optimization and preventive intervention are facilitated.

The invention provides a power grid mountain fire prediction method based on causal driving and a space-time diagram convolutional network, and belongs to the technical field of mountain fire prediction. A dynamic feature encoder and a static feature encoder are designed to extract high-dimensional spatial-temporal features of meteorological time sequence information, geographic space environment and power transmission line distribution multi-source heterogeneous data, a causal discovery algorithm is adopted to construct a dynamically evolved causal graph topology, a real causal driven relationship between variables is identified, a causal intensity matrix is decoupled into positive and negative adjacent matrixes, and the dynamic evolved multi-source heterogeneous data is obtained. And designing a causal constrained graph convolution module to aggregate and propagate high-order information, and finally realizing accurate prediction of the power grid forest fire. According to the method, the PCMCI causal discovery algorithm is introduced, so that the real causal driven relationship among multivariate time sequence factors is effectively identified; a causal GCN layer in the CSTGCN model extracts spatial dependence features by using an adjacent matrix constrained by a causal structure, the CSTGCN model embeds causal structure information into a spatio-temporal feature learning framework, and higher prediction precision and generalization performance are achieved.

The invention provides a multi-modal neural causal inference micro-service fault positioning method and system, and the method comprises the steps: accessing observability data in a service operation process, and representing the tracking information of each request as a directed acyclic graph of a multi-modal feature; performing multi-modal feature coding and graph self-coding anomaly detection on the calling graph, and identifying an abnormal node through a reconstruction error; based on service topology prior, learning a sparse causal relationship graph between services by adopting a multi-scale neural causal inference method; calculating a node root cause score according to the causal relationship graph and the abnormal score, and executing causal path search to generate a fault propagation path; and marking the potential root cause according to the path weight of the propagation graph and the node popularity, and outputting a visual diagnosis result. According to the method, the system operation state is comprehensively described by fusing three kinds of micro-service system multi-modal data of logs, indexes and Trace in the micro-service system, and the structure-perceived causal diagram is constructed, so that accurate and explainable root cause positioning is realized.

The invention provides a gait analysis method and an early warning system based on multi-source data fusion, three types of original signals are collected through a wearable inertial sensor, a plantar pressure insole and an edge calculation camera, and multi-dimensional motion characteristic parameters are fused through wavelet denoising, low-pass filtering, interpolation alignment and data consistency correction, so that the gait analysis method based on multi-source data fusion is realized. And constructing a user individualized gait feature prototype library. The method further adopts a dynamic causal graph network to model a motion relation between joints, determines an abnormal coupling mode based on causal weight and historical reference, and combines activity context and Bayesian rules to generate a dynamic threshold value to realize risk grading early warning, so that heterogeneous data collaborative analysis precision and abnormal gait detection sensitivity are improved, and the method is suitable for large-scale popularization and application. And support is provided for gait health monitoring and personalized risk management and control under multiple scenes.

The invention provides an industrial defect detection and root cause tracing integrated method, and relates to the technical field of industrial intelligent detection and quality control. The method comprises the following steps: constructing a cross-link structured data set containing images, process parameters and defect causes; based on the data set, a multi-modal large model is trained through a course learning strategy, so that the multi-modal large model synchronously outputs a defect positioning result and cause semantic description for the input image; constructing a structured cause-process-control parameter causal map to characterize the causal relationship in the production process; and performing traceability reasoning on the cause semantic description based on a graph neural network, automatically identifying key processes and control parameters causing defects, and outputting a complete traceability path. According to the method, integrated spanning from defect perception to root cause cognition is achieved, and the problems that in a traditional method, detection and traceability are disjointed, manual experience is relied on, efficiency is low, and interpretability is poor are solved.

The invention discloses an automatic causal structure generation method based on semantic representation and logical reasoning of a large language model. The method comprises the following steps: acquiring an input text; performing semantic coding and clustering on the obtained input text by utilizing a large language model, and establishing a candidate causal variable set; causal relationship detection is carried out on the established candidate causal variable set based on anti-fact intervention and do-calculation; performing causal direction judgment, and generating a directed acyclic causal graph meeting logic consistency; and on the basis of the generated directed acyclic causal graph, natural language interpretation is generated by using a large language model, and logic consistency closed-loop verification is carried out. According to the method, automatic generation from the natural language to the causal structure is realized, the causal variable set is automatically extracted and constructed from the unstructured natural language text, the defects that variables need to be manually defined and modeling depends on field experts in the existing causal modeling process are avoided, and the labor cost and professional threshold of causal structure construction are remarkably reduced.

The invention provides a system-level fault analyzing and tracing method and system based on multi-layer causal diagram extraction, and belongs to the technical field of fault diagnosis. Using a multi-level convolutional neural network to convert the time sequence monitoring data features into a feature matrix; by introducing a hierarchical adjacency pruning algorithm and an elastic network regularization constraint, sparse modeling of a multi-level causal matrix is realized, and a causal matrix graph, namely a prediction contribution matrix graph, is obtained; according to a proposed score quantization algorithm, direct propagation and indirect propagation effects are comprehensively considered, prediction information provided by each variable is quantified, a reason score is provided, and a fault reason variable is determined. According to the method, multi-dimensional feature information of the system-level fault can be compared, all useful information is fully utilized, the contribution degree of the system variable fault is accurately evaluated, and the high-level fault reason detection rate is obtained.

The invention discloses a video reasoning training data generation method based on key frame causal chain extraction, and the method comprises the steps: obtaining a to-be-processed video data set, and decoding the to-be-processed video data set to obtain a frame sequence; performing redundancy elimination screening on the frame sequence to obtain a key frame sequence; performing semantic structured representation on the key frame sequence to generate key frame description and entity information, and generating an event candidate set based on the key frame description and the entity information; executing multi-event causal discovery based on the event candidate set to obtain an event-level causal graph, and extracting at least one key frame causal chain from the event-level causal graph; generating a video reasoning training sample based on the key frame causal chain; and performing quality verification and screening on the video reasoning training sample, and outputting a screened training data set. According to the method, through offline key frame index multiplexing and causal structure constraint, the frame-by-frame calculation and labeling cost is reduced, data illusion is inhibited, and the verifiability of reasoning training signals is improved.

The invention provides a database fault root cause positioning method and device based on causal discovery, and belongs to the technical field of database operation and maintenance and fault diagnosis. Comprising the following steps: generating a two-dimensional data table and a statistical information set by using a multi-source operation log of a database system; constructing a causal discovery algorithm set A, and training the gradient boosting tree model by using the training set and the path combination to obtain an agent model Fb; establishing a Monte Carlo tree, selecting child nodes of root nodes according to performance expectation and exploration rewards, and expanding the child nodes to leaf nodes layer by layer; running the Monte Carlo tree, and obtaining a causal graph G * and a performance score by using a voting mode according to all causal relationships in the causal graph obtained by each node; calculating an exploration reward and a performance expectation of each node in a complete algorithm path, and returning the exploration reward and the performance expectation upwards to a root node from a leaf node along the path; updating Fb based on the searched path; and processing the new task by using a Monte Carlo tree, and identifying an affected processing variable according to a path pointing to an abnormal result variable in the causal graph.

The invention discloses a science and technology financial risk identification method and system based on multi-modal data fusion, and the method comprises the following steps: collecting multi-source heterogeneous data, and carrying out the preprocessing; performing modal space alignment and scale normalization on the data to generate a preliminary fusion representation vector; performing semantic association modeling and cross-modal feature collaborative enhancement by adopting a cross attention mechanism to generate a fusion representation vector; cross-modal semantic projection optimization and high-order semantic completion are executed through an improved AlignXpert algorithm, and final fusion representation is generated; based on the final fusion representation, constructing a time causal graph taking the risk identification object entity as the center; inputting the time causal graph into a graph neural network model, and outputting a risk level, an evolution trend and a key causal chain; and continuously updating by adopting a graph structure increment construction and model parameter fine adjustment mechanism. According to the method, the accuracy and interpretability of science and technology financial risk identification under multi-source data are improved, and the method has relatively high intelligence, expandability and practical application value.

The invention relates to the technical field of intelligent futures decision, discloses an intelligent futures decision agent method based on NLP and multi-modal data fusion, and aims to solve the problems of representation decoupling, causal chain deficiency, decision response lag and the like caused by separation and shallow fusion of multi-modal data processing in the prior art. According to the method, a multi-modal data acquisition and preprocessing module, a dynamic heterogeneous causal atlas construction module, a causal conduction space-time diagram neural network reasoning module and a futures decision signal generation and interpretation module are adopted, so that a system for uniformly representing dynamic endogenesis of macroscopic events, industrial logic and microcosmic prices is constructed, and cross-modal is realized.

The invention relates to an artificial intelligence diagnosis auxiliary method and device based on a medical image, equipment and a medium. According to the method, standardized images and interested area masks are extracted from medical images to serve as basic data, in combination with a pathological causal atlas matrix constructed by medical domain knowledge, the atlas matrix is utilized to guide an attention mechanism in a deep neural network to generate a causal-associated weighted feature map and attention distribution; further eliminating the influence of confusion variables through adversarial training and causal intervention loss processing so as to obtain a robust diagnosis model, and finally performing path search and confidence calculation on attention distribution and a pathological causal map based on an analysis result of the model on a target image. And a diagnosis decision path and a visual interpretation report conforming to clinical causal logic are generated, so that the false correlation feature interference is effectively inhibited while the diagnosis accuracy is ensured, and the transparency and clinical credibility of a model decision process are remarkably improved.

The embodiment of the invention provides a broadband oscillation source positioning method and device, electronic equipment, a computer readable storage medium and a program product, and relates to the technical field of electric power. The method adopts a data driving strategy, does not need to depend on accurate physical modeling of a power system, and is suitable for a modern power system with a complex operation state and a changeable topological structure. The causal influence intensity between generator sets is analyzed by introducing transfer entropy, an adjacent matrix is constructed, and reasonable graph structure modeling under the condition of no topological prior is realized. Deep feature extraction is carried out on active power time sequence data of the generator set through the gating circulation unit, dynamic evolution characteristics in the broadband oscillation process are fully captured, and the representation capacity of nodes is enhanced. A causal graph structure and enhanced features are fused in a graph neural network model, space-time joint modeling is realized, the description capability of a disturbance energy propagation path is effectively improved, the spatial positioning precision of a forced oscillation disturbance source is greatly improved, and the method has good engineering applicability and popularization value.

The invention relates to an AI-driven intelligent work reporting method, system and device and a medium. The method comprises the steps of collecting original operation data of production equipment and performing anomaly detection to generate a standardized anomaly event stream; carrying out unified representation on the abnormal event and the production context thereof, and generating a space-time aligned feature sequence; calculating causal intensity between abnormal events based on a preset model and constructing a dynamic time sequence causal graph; back diffusion iteration is carried out by simulating abnormal influence to trace root cause nodes, and a root cause score sorting vector is generated; positioning a key root cause and tracing a propagation path of the key root cause, analyzing backlog data for root cause types such as network interruption, and performing compensation work reporting; and finally, optimizing causal intensity model parameters by utilizing artificial feedback. The method overcomes the defects that in the prior art, abnormal events are analyzed in an isolated mode, and causes and effects cannot be correlated, intelligent root cause diagnosis and work report data self-correction of a complex abnormal chain can be achieved, and therefore production data reliability and operation and maintenance efficiency are improved.

The invention provides a dental implant data monitoring analysis method based on deep learning, and the method comprises the steps: collecting the multi-modal structural data of a patient, such as preoperative images, clinical examination and surgical parameters, carrying out the standardized preprocessing, and inputting the data into a deep neural network model with a double-branch structure, thereby achieving the extraction of the implant failure risk score and local causal association strength; on the basis of model output, a dynamic individualized causal map is constructed, key clinical intervention variables serve as regulation and control nodes, anti-fact disturbance simulation is carried out, variable intervention sensitivity is quantified, and a key causal path is analyzed; according to the method, a priority intervention suggestion list and an auxiliary decision report are generated, the causal atlas can be adaptively optimized according to actual follow-up feedback, long-term closed-loop evolution is realized, the accuracy of implant failure risk prediction and the pertinence and operability of an intervention scheme are improved, and clinical individualized treatment decision support is enhanced.

The invention discloses a new energy power generation optimization control method and system based on multi-source heterogeneous data, and the method comprises the steps: obtaining the multi-source heterogeneous data through a data collection end, and carrying out the preprocessing of the multi-source heterogeneous data, and forming a standardized multi-dimensional data flow; inputting the standardized multi-dimensional data stream into a causal discovery algorithm module, constructing a causal graph representing the causal relationship among the variables in the station by analyzing the conditional independence among the variables, and extracting a causal feature vector having strong causal association with an optimization control target based on the causal graph; inputting the causal feature vector into a time sequence prediction model to obtain a key operation condition parameter sequence of the new energy station in a future rolling time window; inputting the key operation condition parameter sequence into a differentiable optimization layer, and solving a convex optimization problem of embedded equipment physical constraint; according to the method, the problem of massive errors caused by prediction and optimization module target splitting is solved, and intelligent optimization control of the new energy power generation system is realized.

The invention relates to the technical field of artificial intelligence analysis and optimization of a power system, and discloses a line loss variable separation method, system and device based on dynamic time sequence vector fitting, and a medium, and the method comprises the steps: fusing multi-source heterogeneous data, and constructing an enhanced feature vector sequence through an attention mechanism and a physical equation constraint; establishing a time-varying causal graph model based on a dynamic Bayesian network, introducing an attention mechanism to calculate time sequence feature similarity, calculating time-varying causal strength in combination with physical constraint loss and topological distance, and deducing dynamic causal; a theoretical line loss curve is dynamically fitted in a physical-data double-track modeling mode, and key nodes in a causal graph model are used as correction factors; the contribution degree of each factor to the line loss difference is quantified based on a Shapley value algorithm, and accurate separation and attribution of line loss components are realized.

The invention relates to an express cabinet equipment fault intelligent early warning and remote operation and maintenance method. The method comprises the following steps: obtaining a physical topology connection relationship of a target express cabinet as a domain knowledge constraint; performing time sequence causal discovery based on the constraint to generate an initial health causal graph; according to the deployment scene, migrating parameters from the pre-constructed knowledge base for fine tuning to obtain a scene adaptive health causal graph; collecting data in real time and generating a real-time causal graph based on the same constraint; abnormal recognition and early warning are performed by comparing graph structure differences of the real-time graph and the scene-adaptive health causal graph; after early warning, virtual intervention simulation can be carried out in digital twinning, a remote instruction is issued according to a result, and the model is updated according to feedback. According to the method, the early warning accuracy and the positioning precision of composite and intermittent faults are effectively improved, predictive maintenance and remote closed-loop operation and maintenance are realized, and the operation and maintenance cost and the equipment downtime are remarkably reduced.

The invention discloses a simulation model training method and device, a sewage treatment method, equipment and a medium, and is applied to the field of environmental engineering, state nodes are constructed based on state parameters, action nodes are constructed based on action parameters, and intermediate nodes are constructed based on exogenous variables and constants. Constructing a causal graph based on the state nodes, the action nodes and the intermediate nodes, and determining decision nodes and reasoning nodes based on a node relationship in the causal graph; constructing a generator based on the inference model of the inference node and the decision model of the decision node, and constructing a value model for scoring the output of the generator as a discriminator; adversarial learning is carried out based on the generator and the discriminator, and causal decoupling and contrast learning strategies are introduced into adversarial learning. The invention provides a sewage treatment simulation scheme of contrast-adversarial imitation learning based on causal feature decoupling, a virtual environment with high authenticity and generalization ability is constructed, and a reliable basis is provided for intelligent control of the whole flow of sewage treatment.

The application relates to a causal driving unsupervised multi-modal small sample double cycle data fusion method. Each modality data of each lesion sample is respectively preprocessed and feature dimension alignment is performed to obtain an aligned feature matrix. An undirected graph is constructed based on the aligned feature matrix. A causal consistency sample pair set is constructed by using kernel independent component analysis, medical priori and causal consistency test technology, comparative learning of a feature extraction network is performed on the set, causal enhancement features are output, and an updated causal graph is constructed. The updated causal graph is input into a dynamic fusion weight calculation model to calculate dynamic fusion weights, the causal enhancement features after weighted summation and standardization are calculated, preliminary fusion features are generated, the preliminary fusion features are screened, the screened effective fusion features are input into a downstream diagnosis model to correct and update the causal graph, and global fine tuning is performed on the feature extraction network, the dynamic fusion weight calculation model and the downstream diagnosis model to construct an unsupervised small sample fusion model.

The invention discloses a positioning method and device for poor indoor network quality and electronic equipment, and the method comprises the steps: firstly obtaining a network data source, carrying out the first preprocessing of the network data source, obtaining restoration data, carrying out the time granularity unification processing of the restoration data, obtaining unified data, carrying out the multi-source data aggregation of the unified data, obtaining associated data, and carrying out the positioning of the network data source. The method comprises the steps of obtaining associated data, processing the associated data based on a time dimension algorithm to obtain a first causal graph, processing the associated data based on a static dimension algorithm to obtain a second causal graph, finally obtaining a third causal graph according to the first causal graph and the second causal graph, and automatically positioning indoor network poor quality based on the third causal graph. Therefore, waste of human resources is avoided.

The invention discloses an unbalanced discrete data generation method fusing causal constraints and potential variables, and the method comprises the steps: firstly collecting environmental factors and accident severity of a road traffic accident occurrence point, recording the environmental factors and accident severity as road traffic accident data D, building a causal graph G of the road traffic accident data D, and obtaining an adjacent matrix A; secondly, performing encoder mapping calculation on the original data feature X through a variational auto-encoder VAE to obtain a potential variable z; taking the potential variable z and the category label Y as the input of a WGAN-GP network generator to obtain a virtual potential variable, and distinguishing the potential variable z and the virtual potential variable in combination with the authenticity score Di output by a discriminator to obtain a high-fidelity potential variable; and finally, taking the high-fidelity potential variable as the input of a VAE decoder, and generating few-class sample synthesis data in combination with causal constraints. According to the invention, the urban traffic accident data monitoring quality can be improved, and the road safety management and decision-making accuracy can be improved.

The invention discloses a causal graph inference method and a causal graph inference system which are oriented to strong nonlinear continuous variable distribution and based on RealNVP and micro NOTEARS constraints. According to the method, a structure parameter matrix is constructed to represent candidate causal connection, parent variable condition input is constructed for each variable based on structure parameters, and accurate likelihood modeling is performed on the condition density of each variable by adopting a condition RealNVP normalization flow model. By constructing an objective function containing a conditional log-likelihood item, a sparse regular item and a NOTEARS style differentiable acyclic constraint item, a sparse causal structure meeting acyclic constraint is obtained by utilizing gradient optimization, and a causal graph is output. Further, based on the learned causal graph and a conditional RealNVP model, anti-fact inference is executed under a given intervention condition, and an anti-fact result is output. The method is suitable for strong nonlinear relation and complex continuous distribution scenes, and has the advantages of being stable in structure inference, high in interpretability and capable of supporting anti-fact analysis.

The invention discloses a multi-level risk management and control digital safety monitoring management method. The method comprises the following steps: firstly, collecting system and supervision index data and determining a risk level; secondly, extracting supervision historical data to form an operation sequence, converting the operation sequence into a dynamic directed graph, constructing a cause and effect graph model to output a baseline, and setting a threshold value; triggering deep temporary verification according to different levels of user super-operation times, detecting an artifact attack, and closing a session and performing early warning if an exception occurs; after verification is passed, behavior and supervision features are captured at the key business process node for secondary authority authentication; after the authentication is passed, performing short-time dynamic authorization and recording changes; according to the scheme, through acquisition of insurance data hierarchical management and control, a model is constructed in combination with historical operation to identify abnormity, attack is blocked by using a dynamic threshold value and deep verification, risks are prevented and controlled through secondary authentication of key nodes, short-time dynamic authorization gives consideration to flexibility and security, the real-time defense capability of a system for sensitive data leakage and other risks is improved, the false alarm rate is reduced, and the safety of the system is improved. And a full-life-cycle intelligent safety protection system is formed.

The invention discloses an advertisement attribute consistency monitoring method, system and device and a medium, and the method specifically comprises the steps: collecting snapshots and metadata of advertisement attributes in each processing link of a creation process, and constructing an advertisement attribute consanguinity map; extracting time sequence data of attribute values based on the advertisement attribute blood relationship map, modeling a normal fluctuation mode of attributes by using an anomaly detection algorithm, and giving risk early warning in advance for abnormal fluctuation deviating from the normal mode; analyzing conduction paths with inconsistent attributes and positioning root cause nodes by utilizing a pre-trained cause and effect graph model and combining current inconsistent node information and the state of the advertisement attribute blood relationship map; and inputting the alarm event into an alarm strategy model based on reinforcement learning training, and dynamically deciding alarm sending, priority and notification channels. According to the invention, efficient, accurate and real-time advertisement attribute consistency monitoring is realized, the accuracy and efficiency of advertisement putting are improved, and the requirement of continuous development of advertisement services can be met.

The invention relates to the technical field of automatic driving, in particular to a causal effect-based robust automatic driving trajectory prediction method, which comprises the following steps of: establishing a causal graph of a vehicle trajectory prediction model in an attack scene, and analyzing a causal relationship among nodes in the causal graph; building the fact prediction of the vehicle track in the attack scene according to the historical track, leading anti-fact intervention on the historical track, and building the anti-fact prediction of the vehicle track in the attack scene; and calculating a direct total effect by subtracting the anti-fact prediction from the fact prediction, and taking the direct total effect as a final prediction result. According to the method, the direct total effect in causal reasoning is used for defending the adversarial attack, and compared with an existing defending method, the adversarial robustness of the trajectory prediction model under the attack scene is effectively improved at the cost of sacrificing small performance on a clean data set.

The invention discloses a sewage treatment lift pump AI intelligent regulation and control method and system based on working condition perception, and relates to the field of sewage treatment.The sewage treatment lift pump AI intelligent regulation and control method comprises the steps that multi-source heterogeneous working condition data are collected and preprocessed, and a regular real-time perception data sequence is output; carrying out calculation and feature extraction on the regular real-time sensing data sequence, and outputting a structured feature vector; inputting the structured feature vector into a preset lightweight causal graph model for causal inference, and outputting a causal task descriptor; taking a causal task descriptor as a retrieval key, performing matching in a historical operation case library, and outputting a similar historical strategy-result pair set; and jointly inputting the causal task descriptor, the similar historical strategy-result pair set and recent real-time interaction data into a meta-learner online adaptation module. According to the method, asynchronous updating is carried out on a historical operation case library, a lightweight causal graph model and a meta knowledge base of a meta learning device based on full-process data of a current control period.