165 results about "Black box" patented technology

This application discloses a method, apparatus, equipment, and medium for building load forecasting, belonging to the field of electricity forecasting. The method comprises: collecting first load data of a target building and second load data of several adjacent buildings of the same type corresponding to the target building; inputting the first load data and each of the second load data into a pre-trained black-box adaptation network to obtain input embedding vectors, spatial embedding vectors, and feature embedding vectors, respectively; inputting the input embedding vectors, spatial embedding vectors, and feature embedding vectors into a preset large language model to obtain a load forecast embedding vector representing the load forecasting result; wherein, the large language model is iteratively trained on the black-box adaptation network to avoid access to the internal parameters of the large language model; and determining the load forecasting result of the target building based on the load forecast embedding vector. This application can improve the accuracy of building load forecasting in scenarios with few or zero samples.

The application discloses a kind of general complex system governance architecture based on core anchoring+black box credible traceability, it is related to the bottom layer security and credible management of complex system in all industries.The architecture includes core anchoring layer, black box credible traceability layer, distributed quantization coordination layer, realizes behavior bottom line strong management and control through front rigid constraint, realizes traceable and auditable through tamper-proof record throughout the process, realizes large-scale global coordination and automatic deviation correction through distributed quantization.The application can be reused across industries and all scenarios, solves the global common pain points of system out of control, untraceable, difficult to supervise and difficult to coordinate from the root, and is suitable for artificial intelligence, finance, medical care, transportation, energy, justice, agriculture and all complex systems, with high universality, security and social application value.

The application provides a dense reservoir fluid identification method based on physical data double driving, comprising: performing abnormal value correction and fluid category direct marking on conventional logging data to obtain labeled logging data; performing signal self-adaptive time-frequency conversion on the labeled logging data, and calculating porosity, permeability and water saturation to output physical prior feature channels; inputting two-dimensional multi-channel time-frequency spectrum and the physical prior feature channels into a pre-constructed MDSC-TAM neural network model to extract hidden logging signal frequency domain spectrum series structure and global sedimentary rule to obtain a dense reservoir fluid identification result; and performing physical perception constraint and decision output on the fusion feature representation to obtain a fluid type and confidence. The application solves the data scarcity and model black box bottleneck, improves the engineering practicability of the algorithm, and realizes high-precision identification of the dense reservoir fluid and enhanced physical interpretability.

This invention discloses a method and system for temperature compensation of fiber optic gyroscopes. It acquires the angular velocity output, temperature, and zero-bias drift value under varying temperature conditions, constructs basic features, temporal memory features, and physical interaction feature vectors, and integrates them into a comprehensive physical information feature vector. This comprehensive vector is used as input, and the zero-bias drift value is used as the training label to train the model until convergence. In the online phase, this comprehensive feature vector is reconstructed in real time and input into the model to predict the zero-bias drift, correcting the real-time output angular velocity. The temporal memory feature introduced in this invention quantifies the thermal accumulation history using statistics from a multi-scale sliding window, eliminating the ambiguity problem of thermal hysteresis mapping and significantly reducing computational load. Simultaneously, the physical interaction feature pre-introduces the nonlinear product term of temperature and rate of change into the model, decoupling linear and nonlinear errors, breaking through the generalization bottleneck of traditional black-box models under small sample conditions, and endowing the model with strong physical interpretability and extrapolation capabilities.

This invention belongs to the field of solid oxide fuel cell control and optimization technology, and particularly to a dual-input model predictive control method for solid oxide fuel cells with carbon deposition constraints. It includes the following steps: S1: Constructing a DIR-SOFC controlled object model with 11 state variables; S2: Solving for the optimal control increment through sequential quadratic programming; S3: Constructing a constraint system; S4: Introducing an adaptive setpoint for the output voltage that adjusts linearly with the load current density. The model dimension of this invention is expanded from the existing 3 states to 11 states, fully covering six gas phase partial pressures, bilinear coupling of carbon accumulation and catalyst activity, and three-zone thermal dynamics. The ODE white-box prediction engine accurately reflects MSR nonlinearity and positive feedback of carbon deposition. Compared with ARMAX black-box prediction, the error is reduced from approximately 3% to less than 0.8%, with an average single-step calculation time of 133ms, meeting the real-time constraint of a 1s sampling time.

The application provides an App-assisted remote Internet of Things terminal black box fuzzy test method, which comprises the following steps: extracting an App control command based on a document; identifying a mutation point based on App hybrid analysis; conducting fuzzy test based on side channel information guidance; and monitoring an Internet of Things terminal crash based on network behavior. The application uses the time interval between the sending and receiving responses of network messages as side channel information to infer the server-side black box verification logic, realizes the Internet of Things terminal black box fuzzy test technology supported by the bypass cloud server, and thus automatically mines the vulnerabilities of the Internet of Things terminal remotely. The application effectively solves the challenge of remote Internet of Things device black box fuzzy test and improves the fuzzy test efficiency.

The application relates to the technical field of computer system fault diagnosis, and provides a desktop terminal fault intelligent diagnosis and early warning method and system based on multi-source log analysis, which comprises the following steps: acquiring multi-source log operation data of a desktop terminal, generating a structure prior mask screening feature based on a desktop terminal component dependency relationship and an abnormal distribution; injecting a historical causal structure into a self-attention network as an attention bias, extracting a causal dependency strength matrix based on the screening feature; performing counterfactual intervention on a candidate root cause node based on the matrix, and identifying a node meeting causal effect significance as a fault root cause. Through the collaborative design of a causal gating network, a knowledge graph attention and a double-branch encoder, the application can output a structured root cause reasoning path while maintaining the powerful feature extraction capability of deep learning, effectively eliminates the black box effect, enhances reasoning capability, and significantly improves the trust degree and adoption rate of an operation and maintenance personnel to a diagnosis result.

The present application relates to the technical field of image sensor calibration, and particularly relates to a processing method and system of a hybrid event and frame sensor, computer equipment and a medium; the method comprises the following steps: acquiring original data of the hybrid event and frame sensor; performing a noise calibration operation on an intensity frame to generate an active pixel sensor noise model and a net intensity signal; performing an event probability coupling operation on an event stream to generate an event visual sensor event probability model; and performing unified collaborative processing on the active pixel sensor noise model and the event visual sensor event probability model to output uniformly calibrated sensor data. In this way, a unified and reproducible cross-modal calibration framework can be provided to solve the technical problems of existing calibration technologies, such as weak physical coupling, insufficient calibration granularity and black box implementation, and to provide a unified, reproducible and high-precision sensor data calibration output for high-speed and high-dynamic-range mobile perception applications.

The application discloses a kind of SAR image's adversarial sample generation method and application, belong to synthetic aperture radar automatic target recognition security technical field;The present application is based on the average value of using integral integrated gradient, by randomly selecting part of model in integrated model multiple times, and calculating the difference of the temporary adversarial sample gradient generated in inner and outer loop, to correct the outer loop integrated gradient, so that its update direction is more accurate, can generate high-quality adversarial samples in real black box scene, and then can improve the success rate of attack in black box attack, and then affect the accuracy of security detection, in addition, it can also be used to reinforce depth learning model, such as in the process of training model, by adding the generated adversarial sample in data set can further construct better model with robustness, and then can effectively defend the attack of adversarial sample.

The application discloses a distribution network fault identification method and system based on transient state feature, and belongs to the technical field of power system automation and intelligent operation and maintenance. The method obtains distribution network transient recording data and extracts multi-dimensional features, and constructs a state description vector mapping the underlying physical discharge state. The state description vector is combined with the multi-dimensional features to construct a mechanism collaborative feature subset, and redundant features in a random discrete state are removed in combination with distribution network grounding operation parameters. The remaining subset is projected into a transient latent variable space, and a state membership degree vector of the subset and a preset physical state anchor point is calculated. The state membership degree vector is converted into a control mask to inject a deep neural network, and a joint loss function is used to constrain the network hidden layer feature mapping result to fit the corresponding physical state anchor point, so that an initial classification is output. Finally, the feature marginal contribution degree of an inference link is calculated, and a final diagnosis result is output. The application realizes deep integration of physical mechanism and artificial intelligence, and solves the problem of lack of physical basis of a black box AI model.

This invention discloses a method for predicting the application volume of inter-provincial high-voltage direct current (HVDC) channels based on multi-source fusion and probability, belonging to the field of data processing technology. The method includes the following steps: S1, determining the prediction period and basic parameters; S2, determining the final available capacity based on the basic parameters; S3, calculating energy priority based on the final available capacity; S4, constructing a pass rate prediction model based on the final available capacity; S5, determining the output results for the prediction period based on the pass rate prediction model and energy priority. This invention is the first to model channel capacity calculation as a deterministic optimization problem with physical constraints, combining a probabilistic correction factor for meteorological risk to achieve a hybrid modeling of determinism and stochasticity, supporting risk attribution, outperforming black-box prediction, and improving the application pass rate.

The application provides a smart device abnormal recovery method and device, an electronic device and a device local area network, and relates to the technical field of smart homes. The smart device abnormal recovery method comprises the following steps: determining that a fault occurs; obtaining black box abstract information of fault state data at a current time; broadcasting an abnormal rescue request in the device local area network; after a network interconnection device detects the broadcast of the abnormal rescue request, the network interconnection device establishes a connection with the smart device, reads the black box abstract information from the smart device; receiving an update package sent by the network interconnection device; the update package corresponds to the black box abstract information; the update package is received from a cloud server after the network interconnection device reports the black box abstract information to the cloud server; and performing an online upgrade operation based on the update package. The application can efficiently and reliably recover the fault of the smart device occurring in the running process.

The application discloses a kind of SAR image's adversarial sample generation method and application, belong to synthetic aperture radar automatic target identification security technical field;The application proposes the concept of teacher adversarial sample and student adversarial sample, first using gradient-based iterative attack method to the gradient vector of teacher adversarial sample and teacher adversarial sample is alternately updated, obtains the gradient vector of teacher adversarial sample obtained in last update, then in subsequent iterative process Guiding student adversarial sample gradient vector update;The application considers that SAR image is prone to cause the problem that gradient direction changes too much in gradient calculation process due to the existence of coherent noise, by using the gradient information of teacher adversarial sample, it is fused into the generation process of student adversarial sample, can well fit SAR data characteristics, to generate high-quality adversarial sample in real black box scene.

This invention discloses a pollen identification method based on dual-model decision arbitration and multimodal database association, belonging to the field of intelligent identification technology. The method includes: acquiring pollen images; inputting them into a first model based on Visual Transformer and a second model based on YOLO for parallel inference, outputting candidate categories and predicted probabilities; inputting the results into an arbitration engine; if the highest confidence categories match, a direct determination is made; otherwise, dynamic weighted arbitration is performed to obtain the final determination category; subsequently, the corresponding morphological feature text descriptions are retrieved in real time from the multimodal database, and the determination category, text description, feature heatmap, and alternative references are simultaneously pushed to the interactive interface. This invention overcomes the "black box" defect of models, achieves macro- and micro-feature complementarity and intelligent arbitration, constructs an "instant recognition and interpretation" auxiliary identification system, and improves the accuracy and interpretability of pollen identification.

The application discloses a large language model backdoor attack detection method, and belongs to the field of artificial intelligence security. In view of the problem that existing defense methods are difficult to be applied to generative large language models and have large calculation overhead, the application is based on the first discovered "sequence locking" phenomenon, that is, when generating an attack target, the backdoor model will output an abnormally high and consistent word element confidence, forming a deterministic path without branches. The method detects high-confidence sequences that continuously exceed the length threshold L and have a probability higher than the threshold P by monitoring the Top 1 probability of the model output sequence in real time, thereby determining the backdoor attack. The method only needs to access the Top 1 probability of the model in a black box, does not need complete Logits vectors or auxiliary models, almost does not introduce additional delay and performance overhead, is suitable for real-time interaction and large-scale deployment scenarios, and can efficiently detect various backdoor triggering modes and target types.

The application discloses a kind of virtual power plant dynamic support capacity quantification and boundary measurement method based on inverse optimization, it is related to power system operation and control field, first, the physical model of virtual power plant multi-time period dynamic optimal power flow based on second-order cone relaxation is constructed;Second, with the minimum of virtual power plant grid connection point limited measurement data residual as the goal, a composite inverse optimization model containing feasible region compactness penalty is constructed;Then, using the asymmetric moving average sequential updating algorithm with attenuation, the equivalent power boundary and virtual energy capacity of the system are accurately inversed from multi-scenario operation data;Finally, the target response time length is explicitly introduced, and the state-dependent dynamic flexibility envelope considering instantaneous energy state constraints is constructed.The application overcomes the limitation of bottom-layer parameter black box, and quantifies the dynamic support potential of virtual power plant under multi-time scale in detail, providing a defense over-limit decision basis for safe dispatch.

The application discloses a kind of grey-box isomeric doubly-fed wind farm low-frequency oscillation suppression method based on virtual damping modulation, comprising: 1 grey-box isomeric doubly-fed wind farm is divided into white box and black box unit, and the state space model of all white box units of traditional doubly-fed induction generator, virtual synchronous machine control and installation additional damping controller is established;2 improve subspace identification method to obtain black box unit model;3. white box and black box model are coupled by feedback interconnection, and the overall state space model of wind farm is constructed;4 determine target low-frequency oscillation mode based on coefficient matrix;5 real-time extraction and target mode strong correlation common bus power signal, combined with adaptive time-varying gain generates virtual synchronous machine damping coefficient adjustment amount;6 according to complex mode controllability factor and unit distribution coefficient, allocate the virtual damping modulation instruction of each white box unit.The application realizes the identification and suppression of grey-box isomeric doubly-fed wind farm low-frequency oscillation, and improves the ability of wind farm damping low-frequency oscillation.

The present application relates to the field of resistance spot welding quality monitoring, in particular to a multi-modal time sequence quality identification system and method for electronic spot welding, synchronously collecting three types of multi-modal signals of electric-thermal, acoustics and vision, realizing hardware level time alignment by FPGA, extracting local mutation features by one-dimensional convolutional neural network 1D-CNN, capturing long time sequence dependence by time convolution network TCN, introducing cross-modal attention mechanism, taking electric-thermal features as the leading dynamic fusion auxiliary modal information, constructing electric-thermal-acoustic-vision four-channel parallel collection architecture through edge deployment optimization, solving the problem that the current modal signal information is one-sided and difficult to comprehensively represent the welding physical process; the deep learning model inference delay is high, which cannot meet the real-time requirement of <20ms of the production line, the model decision is "black box", and the defect root cause is not explainable.

This invention discloses a multi-agent collaboration method and system for domestically developed operating systems, belonging to the field of intelligent technology for domestically developed desktop operating systems. It aims to solve the core problems of existing solutions, such as lack of collaboration standards, black-box decision-making, and difficulty in reusing experience. The method establishes a collaboration system by building an MCP protocol stack and a native DBus mapping, generating a structured inference chain through the System-2 inference engine, and achieving a closed loop through dynamic task distribution, three-level memory reuse, and visual auditing. The system includes a main agent, a cluster of sub-agents, and multiple core modules to collaboratively support the implementation of the method. This solution enables plug-and-play intelligent agents, full traceability of decision-making, and improves the execution efficiency of similar tasks by more than 30%. It is deeply adapted to multiple versions of openKylin and domestically developed hardware, meeting the security compliance and efficient collaboration needs of key scenarios such as government affairs and finance.

The present application relates to the technical field of household photovoltaic energy storage collaborative control, in particular to a household photovoltaic storage collaborative method and system based on shared timing diagram network and multi-task learning, which first collects multi-source heterogeneous data to construct a household micro-grid node space-time topology graph, uses a space-time enhancement transformer as a unified shared backbone network to extract global shared space-time features; based on the shared features, multi-task joint decoding is performed to output photovoltaic / load prediction values, SOC trajectories and confidence intervals, and same variance uncertainty is introduced to dynamically adjust the task loss weights; the prediction results are input into a model predictive controller with fused physical hard constraints to solve optimal charging / discharging instructions; finally, through knowledge distillation by adding a physical consistency penalty term, a cloud large model is compressed into an edge lightweight model to realize local safe autonomous scheduling; the present application solves problems such as prediction error cascade amplification, weak resistance to extreme electricity prices, and easy triggering of battery physical over-limit by AI black box, and significantly improves the operation economy and control safety of the system.

The application provides a modulation method and system of an orthogonal frequency division multiplexing signal, and belongs to the technical field of wireless communication. The method comprises the following steps: mapping a frequency domain data signal and a pilot signal to be transmitted into a frequency domain resource grid matrix; performing a frequency domain shift operation on each row vector of the frequency domain resource grid matrix, shifting a zero frequency component to the center of the vector, and obtaining a shifted matrix; and performing an inverse fast Fourier transform on each row vector of the shifted matrix, and obtaining a time domain orthogonal frequency division multiplexing symbol matrix as a modulation result. The application constructs the entire frequency domain resource into a two-dimensional matrix, and all subsequent signal processing operations are based on the row vectors of the two-dimensional matrix as a basic unit, and are completed through a series of explicit matrix and vector operation steps. Thus, the IFFT call for the entire frame, which is a black box, is decomposed into individual steps for a single symbol, which are visible and controllable, and the problem of opaque process is solved.

The application belongs to the technical field of robot dynamics parameter identification, and discloses a double sparse efficient identification method and system for robot dynamics parameters. The method comprises the following steps: S1, collecting joint positions, joint speeds, joint torques and joint accelerations of a robot to be processed at each time; S2, constructing an implicit dynamics equation of the joint torque by taking the joint torque as a target; S3, constructing a base function database, evaluating the importance of each base function in the base function database, thereby completing the first sparse of the base function; and performing sparsification on the coefficient matrix to make some elements in the coefficient matrix 0, thereby completing the second sparse of the base function, determining the base function and the coefficient matrix, and realizing the acquisition of an explicit dynamics equation of the joint torque. Through the application, the technical problems of complex theoretical modeling or low interpretability of a black box model in the prior art are solved.

The application discloses a kind of multi-modal forest fire target detection method and system based on thought chain, belong to forestry intelligent monitoring technical field.It includes the following steps: collecting and pre-processing paired visible light and thermal infrared forestry scene image, after data labeling, pre-training large visual language model is trained using supervised fine-tuning, and the basic perception model is constructed;Composite reward function is constructed, and the reasoning process of the basic perception model is optimized using group strategy optimization reinforcement learning algorithm;By inputting the preset thought chain prompt to the optimized basic perception model, the model synchronously generates structured natural language text.Compared with the prior art, the application has the advantages that the "black box" model is converted into an interpretable and traceable decision report by the thought chain, greatly enhancing the credibility of the system.

The application discloses a kind of reconfigurable intelligent surface auxiliary wireless environment modeling method based on three-dimensional Gaussian splash technology, belong to wireless communication field, this method will three-dimensional Gaussian splash technology from computer graphics field innovatively migrate to RIS auxiliary wireless communication system modeling field, the explicit parameterization thought of 3D-GS is applied to radio frequency electromagnetic field modeling, break through the limitation of traditional deep learning "black box" modeling;Adopt two-stage joint modeling framework: for the characteristics of RIS system, "TX→RIS" and "RIS→RX" cascade modeling strategy is designed, can accurately capture the electromagnetic regulation effect of RIS;With explicit physical parameterization: replace large-scale neural network weight with Gaussian parameter with clear physical meaning, realize light weight;In addition, by constructing complete differentiable rendering process, support efficient gradient descent optimization, ensure training efficiency and convergence.

The application discloses a video super-resolution method based on space-time convolution attention, and is specifically implemented according to the following steps: step 1, converting an input high-resolution video sequence into a low-resolution video sequence, so as to obtain a low-resolution video set; step 2, constructing a deep neural network structure for video super-resolution; step 3, training the network based on paired video data; step 4, video super-resolution reconstruction, so as to realize super-resolution recovery of the input video sequence. The application solves the problem in the prior art that the network is regarded as a black box, performance is improved by increasing the network depth and parameter scale, the rapid growth of the calculation complexity is ignored, and the efficiency and the reconstruction quality are difficult to be considered.

The application belongs to the field of financial data processing, and discloses a kind of financial data processing system and method, comprising: data acquisition module, for collecting the multi-source multi-modal data of listed company, multi-source multi-modal data includes enterprise internal financial data, enterprise internal non-financial data, social external index and macroeconomic variable;Pretreatment module;Characteristic screening module;Model construction and optimization module;Interpretability analysis module;Output module.This scheme constructs a set of multi-source data integration, automation processing, intelligent modeling and decision explanation in one financial early warning system.Through the standardization of module pipeline, the whole process closed loop processing is realized;At the same time, the high precision prediction and SHAP interpretability technology are fused, so that the model can output accurate risk probability, and the influence degree and direction of each risk factor can be clearly quantified, the traditional "black box" prediction is transformed into transparent, reliable and operable decision basis, and the unity of early warning accuracy and decision support is realized.

The application discloses a kind of product service network knowledge graph construction and dynamic evolution method, belong to dynamic knowledge graph and product service system field.Method includes: first, the knowledge graph of multiple product service networks is collected as data source and the ontology of each is extracted;Then attribute, concept and relation similarity function are sequentially constructed;Based on the calculation result of multi-granularity similarity, the ontology of each data source and the corresponding knowledge graph are merged into the unified ontology and the fusion knowledge graph constructed;When data source changes, first determine whether to carry out full update, select full update or incremental update according to the determination result, and finally extract the dynamic evolution information of the fusion knowledge graph through the type-oriented attention mechanism.The method of the application effectively improves the explainability, logical consistency and credibility of product service cross-source data, and overcomes the absurd or contradictory prediction that may be produced by a pure black box model.

This invention belongs to the field of CCUS-EOR technology and discloses an interpretable data-driven prediction and optimization method for CO2 enhanced oil recovery and storage. The method includes the following steps: using a CMG-GEM component simulator, numerical simulation is conducted through Monte Carlo sampling to obtain a multimodal data pipeline; a LightUFTrans hybrid model is built to achieve three-dimensional prediction of CO2-EOR and storage performance; LightUFTrans is embedded as a surrogate model into the LA-MONSGA-II optimization framework to obtain the Pareto optimal solution; and the model decision-making mechanism is analyzed from the dimensions of global features, temporal dependence, and spatial distribution by combining SHAP, attention mechanism, and spatiotemporal verification. This invention uses the above method to construct the LightUFTrans multimodal surrogate model, accurately depicting the dynamic evolution of underground spatiotemporal processes, significantly improving computational efficiency, and greatly reducing engineering design and decision-making costs; through multidimensional interpretability analysis, key engineering control parameters are identified, their temporal dynamic characteristics and spatial migration mechanisms are analyzed, overcoming the limitations of black-box models and improving engineering credibility.