15 results about "Real systems" patented technology

This invention discloses a digital twin method and system for autonomous driving that integrates virtual and real environments based on a full-link interface. The invention includes: converting multi-source perception data from autonomous vehicles in a real physical environment into a real object state dataset; establishing a standardized state mapping between the real system state space and the virtual simulation system state space; converting the real object state dataset into virtual environment state data executable by the simulation engine; constructing a virtual environment; performing virtual perception based on the constructed virtual environment: performing sensor-level sampling on the virtual environment state data to generate virtual raw sensor data, forming a virtual perception result data stream; and verifying the virtual perception result data stream through the twin verification loop. This invention solves the problem of simulation engines being unable to dynamically generate objects during runtime, achieving high-fidelity, low-latency reproduction of real-world traffic flow in a simulation environment.

The present invention provides a method and system for fault detection of a motorhome energy storage system, belonging to the technical fields of new energy vehicles and intelligent detection. The method captures load start / stop operation instructions with timestamps in real time, and synchronously collects voltage and current signals at the output end of the energy storage system, strictly aligning the operation behavior and electrical dynamics on the time axis. Based on the load identifier in the operation instruction, a pre-stored feature template is retrieved, and the transient electrical features extracted in real time are subjected to spatio-temporal consistency verification with it, effectively distinguishing legal load startup from real system faults. When the verification fails, the similarity with the fault feature library is further calculated for risk assessment and hierarchical protection is executed. The corresponding system includes an operation instruction capture module, an electrical signal acquisition module, a central processing and synchronization module, a load fingerprint matching module, a fault risk assessment module, and a hierarchical protection execution module. The present invention significantly reduces the false alarm rate and improves the accuracy of fault detection and the continuity of system power supply.

The present application relates to the technical field of energy storage system, and particularly to a BMS simulation verification method and platform based on fault injection closed-loop joint debugging.A BMS simulation verification method and platform based on fault injection closed-loop joint debugging, the platform comprises the following: a model management layer; a scene generation and scheduling layer; a fault injection layer; a closed-loop joint debugging layer; a result collection and evaluation layer; a report generation and backfilling layer.Compared with the prior art, through the combination of the model library, the scene library and the fault injection library, the virtual BMS is quickly verified under different objects, different working conditions and different fault types; through the closed-loop joint debugging between the battery model, the thermal management model, the actuator model and the communication model, the closeness of the verification result to the real system is improved; through a unified coverage, response quality and protection effect evaluation mechanism, the comparability and screening efficiency between different verification tasks are improved.

Provided is a simulation device capable of presenting, to a user, an item of a parameter that may contribute to improving a simulation result. The simulation device causes a virtual system in a virtual space, the virtual system having virtual components corresponding to components included in a system in a real space, to reproduce an operation of the system in the real space by simulation using real data acquired from the system in the real space. The simulation device comprises a parameter setting processing unit and a simulation unit. The parameter setting processing unit reads data item management information including a value of a parameter to be set for the virtual component, and outputs a parameter setting screen that allows the value of the parameter to be changed. The simulation unit simulates an operation of the virtual system according to content set on the parameter setting screen. The parameter setting screen includes a parameter value type corresponding to the parameter value.

This invention discloses an indoor gait monitoring method and system based on distributed optical fiber sensing. The method utilizes a two-dimensional grid of sensing optical fibers laid beneath the indoor floor as the sensing component of a distributed optical fiber sensing system. The fiber spacing is less than half the average foot length of the monitored object, enabling blind-spot-free vibration sensing across the entire indoor area. Furthermore, a vibration data matrix containing spatial and temporal dimensions is generated. Through numerical simulation of the distributed optical fiber sensing system, a matching channel consistent with the statistical characteristics of the real system is constructed, generating a high-quality training set to train a multi-layer feedforward neural network to obtain a classification model. The real-time vibration data matrix is ​​input into the trained model to identify fall events and trigger an alarm response. This method achieves radiation-free, privacy-free, and wearable-free full-area gait monitoring, effectively solving the problem of scarce real fall samples and significantly improving the accuracy and generalization ability of fall recognition.

This invention belongs to the field of digital twin modeling and intelligent optimization technology for new energy power generation systems, and relates to a method and system for online updating of wind turbine twin parameters based on neural deformability and moss optimization algorithms. The method proposes an online updating method for digital twin model parameters coupled with neural deformability networks and moss optimization algorithms. By constructing a learnable deep elastic time series similarity metric, it solves the evaluation problem of multivariable temporal phase drift and local deformation. Combined with the efficient global search characteristics of the moss optimization algorithm, it achieves high-precision, adaptive, and online updating of key parameters of the wind turbine digital twin model, ultimately improving the consistency of dynamic behavior between the virtual and real systems, and laying a model foundation for the engineering application of digital twins in the wind power field.

This invention provides a method and apparatus for joint system simulation verification based on multi-source data injection. Based on a preset task objective, the invention constructs a virtual training environment model in a simulation platform, including geographical environment, force units, task scenarios, and dynamic interference factors. Realistic simulation data is dynamically generated based on the virtual training environment model. False feature codes and dynamic intensity noise signals are generated, purified, and then injected into the real system under test to generate corresponding control commands. These control commands are fed back to the simulation platform, driving the entity state update in the virtual environment and evaluating performance indicators through a digital twin module. This invention can accurately reproduce the joint system operation state under complex real-world scenarios, comprehensively evaluate system functionality, real-time performance, anti-interference capabilities, and decision optimization, significantly reduce verification costs, and provide technical support for the stability and reliability of multi-system joint operation.

This invention discloses a machine learning-based method and system for configuring parallelization of dense LLM inference. The method receives a query request containing the input and output lengths of a dense large language model to be configured. It enumerates all feasible parallelization strategies for the target model, inputs these strategies into a trained hybrid prediction model, and predicts the throughput of each parallelization strategy under the constraints of the query request. The parallelization strategies include tensor parallelism, pipeline parallelism, and the number of GPUs. The optimal parallelization strategy is selected as the parallelization strategy for the target model's inference. This invention learns performance patterns from real system operation data, avoids complex and imprecise modeling, and supports intelligent decision-making under changing resource constraints, outputting the optimal parallelization strategy for dense large language model inference.

A method for repairing highly reflective areas combining digital twins and transfer learning is proposed. The specific implementation includes the following steps: First, the camera and projector of the real 3D measurement system are calibrated to obtain their intrinsic and extrinsic parameters. Second, a digital twin measurement system corresponding to the real system is constructed using the simulation software Blender, generating a large amount of virtual data. Overexposed images are used as label values, and unsaturated images are used as the network output. A U-net network model is trained using the dataset to obtain a pre-trained network, giving it initial capabilities for repairing overexposed areas. Finally, transfer learning is introduced, and the pre-trained network is fine-tuned using a small amount of real data to improve the model's generalization performance in real-world scenes. This method provides a new approach to solving the problem of stripe information loss caused by high reflectivity, significantly reducing data acquisition costs and time, and can be applied to the reconstruction of highly reflective objects.

The present application relates to the field of complex equipment development in the aviation industry, in particular to a SysML heterogeneous model integration method, aiming to solve the problems of data island and collaboration barriers caused by model heterogeneity. Based on a unified data management platform, the method realizes the normalization of multi-source SysML models, the unified integration and scheduling of heterogeneous models, the high-precision joint simulation of virtual and real systems, and the multi-dimensional intelligent analysis of simulation results through five core steps of intercommunication and synchronization, including SysML model analysis, functional unit conversion, multi-source heterogeneous model integration, virtual-real fusion test simulation, and comprehensive analysis and iterative optimization. The platform is equipped with four special function modules, which are compatible with multiple modeling tools and form a closed-loop design system through technologies such as full-factor mapping, middleware integration, gradient descent method parameter calibration, and three-dimensional visualization analysis. The method has been verified effective in the development of aircraft engines and airborne equipment, significantly improving model reuse rate and cross-disciplinary collaboration efficiency, shortening design optimization cycle, and providing replicable technical support for the implementation of MBSE in the aviation industry.

The application relates to a multi-agent cooperation method based on space perception and multi-modal AI fusion, and belongs to the field of information technology. The method comprises the following steps: through synchronous acquisition of multi-source sensor data, fusion of multi-modal features, establishment of virtual-real system bidirectional mapping and closed-loop control, construction of a digital twin model of a physical entity and generation of an optimal control strategy; through cross-modal feature fusion space perception action, construction of a dynamic knowledge graph, triggering of a semantic rule engine for logical deduction, acquisition of an optimized strategy set and state prediction data; based on the optimized strategy set, an enhanced strategy set is constructed, a two-stage game is adopted to realize Pareto optimization to obtain a consensus strategy, and a reward shaping technology is used to quantize indicators to correct KPI and feed back to the space perception action. Through multi-modal AI fusion and multi-agent collaborative optimization, the application realizes accurate multi-source data acquisition and fusion, virtual-real state synchronization and dynamic closed-loop control, and improves the collaborative efficiency of the physical entity and the digital twin model.

This disclosure provides a method, apparatus, electronic device, and storage medium for generating adversarial examples, relating to the field of artificial intelligence technology, particularly machine learning and natural language processing. The specific implementation involves: obtaining attribute information of fields in a first structured data set; determining configuration information for an adversarial attack based on the attribute information of the fields in the first structured data set; and obtaining adversarial examples of the first structured data set based on the configuration information of the adversarial attack. This disclosure takes into account the diverse field attributes of structured data in real-world system applications, making the adversarial attack more closely resemble real-world system applications, possessing high usability, and applicable to adversarial attacks on structured data.

This application provides a predictive maintenance method for industrial equipment based on digital twins and spatiotemporal data, relating to the field of data processing. In this method, an equipment identifier is constructed and a uniquely bound digital twin object identifier is generated, establishing a maintenance right pool; multi-source spatiotemporal observation data is collected to form a spatiotemporal evidence film, constructing an irreversible response evidence set and a reversible response evidence set, and establishing an irreversible evidence fidelity quota; a set of maintenance right certificates is generated by combining structural topology, control topology, and constraints; target maintenance right certificates are extracted within a rolling cycle and a secondary revocation check is performed; upon successful revocation, maintenance actions are executed, and irreversible evidence is compared and replayed to generate a fidelity assessment result, which is then output as an executable maintenance recommendation in the form of a maintenance right certificate. Implementing the technical solution provided in this application facilitates the prediction model in maintaining its predictive capabilities while suppressing the mutual adaptation deviation between the digital twin and the real system.

This application discloses a data access method, a data receiving method, an apparatus, a system, a device, and a medium. The data access method is applied to a gateway and includes: acquiring configuration information of cloud device components; the cloud device component configuration information includes a system shadow and device component shadows corresponding to the real target system, the shadows containing communication connection information between the device components and the gateway; establishing a communication connection between the gateway and the corresponding device components based on the communication connection information; acquiring measurement data of multiple device components, mapping the measurement data of each device component to the device component shadow, obtaining device component shadow data; integrating the multiple device component shadow data based on the system shadow, and sending the obtained system data to the cloud device. The scheme disclosed in this application integrates the data of each device component in the real system into the real system's system data based on the device component shadow and the system shadow before reporting it to the cloud device, thereby reducing the complexity of system access and management.

A facility for generating and using a digital twin to represent a structure, such as a complex or machine, via mixed reality is described herein. The facility accesses a digital twin simulation model including data indicating a current state of one or more aspects of a structure. The facility receives image data associated with the one or more aspects of the structure that indicates a real-world representation of the one or more aspects of the structure. The facility causes a mixed reality device to present a mixed reality user interface based on the digital twin simulation model and the image data. The mixed reality user interface includes the received image data and a visual representation of the data indicating the current state of one or more aspects of the structure overlayed onto the received image data.