153 results about "Adaptive system" patented technology

The invention relates to the technical field of thermal management control, and particularly discloses a thermal management controller regulation and control method and system based on dynamic thermal inertia prediction. According to the method, the dynamic thermal inertia model is constructed through system identification, and the future temperature field evolution and thermal accumulation risks of the system are accurately predicted. On the basis, a multi-layer nested compensation strategy composed of pre-compensation, main compensation and residual compensation is generated in three dimensions of a compensation angle, a compensation order and a compensation domain, and prospective accurate intervention is achieved. In order to evaluate the control effect, an artifact intensity entropy value is introduced as a quantitative index, and a model and a strategy are corrected in real time through an outer ring optimization mechanism to form self-correction. When continuous thermal unbalance caused by hardware design defects is detected, the system can further start hardware cooperative regulation and control, local heat capacity and a heat dissipation path are optimized fundamentally, cross-level cooperative optimization from a control algorithm to a physical structure is achieved, and finally high-precision, low-energy-consumption and self-adaptive system-level thermal management is achieved.

The invention discloses a multi-sensor fusion sensing precision self-adaptive system, and relates to the technical field of radio positioning and multi-sensor fusion sensing, the system integrates multi-source sensing units such as a laser radar and a GPS, establishes a unified reference by means of a space-time synchronization module to complete data alignment, and optimizes data quality through preprocessing means such as filtering and denoising; a dynamic weight fusion module allocates weights according to the real-time quality scores, and a unified sensing result is output by fusing the positioning and speed measuring results; through multi-dimensional index evaluation precision grade, self-adaptive adjustment of sensor parameters and fusion strategies, and balance of precision and energy consumption, a result is output in a standardized format, an operation state is fed back, and an external system is linked. According to the invention, through multi-sensing fusion and dynamic adaptive adjustment, the sensing precision and stability are significantly improved; the method can intelligently adapt to multiple scenes, resists interference, has fault-tolerant capability, adapts to new requirements through self-learning evolution, and provides reliable perception support for automatic driving, unmanned aerial vehicle surveying and mapping and the like.

The invention discloses an IMU online adaptive calibration method and calibration device and an automatic driving system, and relates to the technical field of deep learning, automatic driving and navigation. Comprising the following steps: acquiring original measurement data from an IMU in real time; inputting the original measurement data into an error prediction model to output predicted IMU error parameters, wherein the error prediction model adopts a hybrid neural network architecture, is trained and updated based on an end-to-end framework containing a differentiable physical model, and is configured to execute offline training in an offline stage and execute adaptive updating in an online application stage; and correcting the original measurement data by using the IMU error parameter and executing dead reckoning to generate a calibrated motion state track. According to the invention, a static off-line calibration system is improved into a dynamic on-line adaptive system, so that the long-term robustness and reliability are remarkably improved. A differentiable physical model is introduced as a training bridge to realize an end-to-end physical model training normal form, so that the training precision can be improved and the training convergence can be accelerated.

The invention discloses an upper limb rehabilitation intelligent self-adaptive system based on multi-modal fusion and MRD-FES cooperation. The system comprises a multi-modal signal acquisition module, a dynamic intention analysis module, a variable damping precise execution module, a neural remodeling strengthening module and a quantitative effect evaluation module. The tDCS-MRD-FES time sequence cooperation protocol is provided, the cortical excitability is improved through tDCS preprocessing, MRD drives training based on intention, FES closed loop strengthens muscle contraction, and vibration tactile feedback is assisted to strengthen sensory movement integration. And the quantitative effect evaluation module further combines a neuroplasticity index and clinical scale digitalization to realize dynamic optimization of training parameters.

The present invention provides a computer system and a computer-implemented method for adaptive quality management in artificial intelligence (AI) processing. The system comprises a memory unit storing machine-readable instructions and a processor configured to execute the instructions to store, in a persistent correction memory, records of prior corrections to AI outputs; retrieve a correction for a current input based on similarity between the input and stored records; generate, by an audit-trail module, a tamper-evident record of the correction and its application; schedule, by a predictive resource manager, AI workloads in dependence on service-level constraints, carbon-intensity data, and quality metrics; and update, by a feedback controller, retrieval and scheduling parameters based on effectiveness information. The processor thereby enables the persistent correction memory, audit-trail module, predictive resource manager, and feedback controller to operate as a closed-loop adaptive system that reduces recurrence of AI errors, ensures verifiable auditability, and optimizes computational efficiency.

The invention provides a big data advertisement putting method and system based on a cloud platform. The method comprises the following steps of performing multi-source data collection and preprocessing; dynamically constructing a user portrait; performing multi-dimensional analysis on the behavior pattern; target audience subdivision is based on a Gaussian mixture model; dynamically generating personalized advertisement creativity; a step of optimizing a multi-target delivery strategy; a step of real-time bidding and bidding optimization; a step of putting execution and real-time monitoring; a step of effect evaluation and multi-dimensional feedback; and carrying out adaptive system adjustment and iterative optimization. The big data advertisement putting method based on the cloud platform has the advantages that putting precision and efficiency are improved, audiences are subdivided through complex statistics, dynamic creativity is generated, advertisement correlation is enhanced, the predicted click rate is increased by 15-20%, and budget waste is reduced. And the cloud platform supports stream processing and automatic adjustment, so that the system can quickly respond to market changes, and the delay is reduced to a millisecond level.

The invention belongs to the field of artificial intelligence, particularly relates to an intelligent sound field adaptive system of digital professional sound equipment, and aims to solve the problems of inaccurate sound field regulation and control, response lag, dependence on manual tuning and the like in a complex acoustic environment. The system comprises a sound field sensing module, an acoustic modeling and analysis module, a self-adaptive sound field regulation and control engine, a multi-channel digital audio processing unit and a feedback optimization module, and dynamic sound field modeling, real-time audio processing and environment self-adaptive regulation and control are realized through distributed sensing, hybrid modeling, multi-target optimization and closed-loop feedback. The system supports rapid re-calibration, multi-scene memory and user preference learning, ensures voice clarity and music fidelity, balances full-field hearing consistency, and significantly reduces manual intervention requirements.

The invention relates to the field of natural language processing, in particular to a low-rank adaptation system capable of differential polarization rank search, which comprises an initialization module, a warm-up training module, a joint optimization module, a sparse pruning module and a final fine tuning module. According to the method, dynamic rank allocation and continuous search are realized by introducing trainable rank architecture parameters, and model parameter waste and calculation overhead are remarkably reduced by combining polarization regularization and L2 regularization strategies. According to the method, the performance of the large model in the downstream task can be effectively improved, meanwhile, the method is suitable for scenes with multi-task concurrence, resource-constrained device deployment and high real-time requirements, and a technical solution is provided for efficient adaptation of the large-scale pre-training model.

The invention discloses a network security policy full-life-cycle adaptive tuning system based on intelligent computation driving, relates to the technical field of network security, and is used for solving the problems of policy adaptation and efficiency balance under resource limitation. The method comprises the following steps: firstly, predicting an attack probability and mapping a topological weight by using an LSTM combined with an attention mechanism, and generating a risk prediction vector; then, performing weighted pruning on the rule based on the computing power constraint and the flow priority, and generating an execution state strategy packet; after issuing, monitoring a delay increment and an interception hit rate during operation, and generating a regular degradation instruction for a low-efficiency and high-consumption rule; and finally, regular scanning parameters are adjusted through gradient optimization, a self-adaptive system from risk pre-judgment to closed-loop optimization is constructed, and intelligent support is provided for network security defense.

The present disclosure relates to a multidimensional adaptive sensing system and method for event detection with improved accuracy. The system dynamically adjusts, using baseline and compound models, a triggering conditional threshold of each event based on real-time variations in environmental parameters and sensor health. The adaptive system allows for fine-tuning of sensitivity to environmental changes, seasonal variations, and sensor decay, thereby reducing false positives and negatives. Installation flexibility is achieved by selecting and deploying a subset of models tailored to available data at the sensor's location, ensuring optimal operation regardless of specific environmental conditions or sensor configurations. The adaptive sensing system is used in environmental control systems for dynamic routine management, sensor health management, a triage system for complex event analysis, and multifaceted compliance assurance solution. Further, the adaptive sensing system is used for automated configuration, model deployment, and maintenance by leveraging edge computing units and a central control unit.

The invention relates to the field of mine intelligent tunneling, and discloses a hard rock lithology identification and cutting parameter self-adaption system based on machine vision, which comprises an image acquisition module, a lithology identification module, a rock hardness mapping module and a self-adaption control module. The image acquisition module is integrated with an air curtain cleaning and dual-mode light supplementing device to obtain a rock surface image of the driving working face; the lithology identification module extracts lithology features by using an adaptive enhancement algorithm and an improved deep learning network; the rock hardness mapping module quantifies and outputs a rock hardness value through a regression model; and the self-adaptive control module dynamically adjusts the rotating speed and the feeding speed of the cutting head according to the parameters, and stabilizes mechanical shock through a parameter smoothing unit. In addition, a feedback optimization module performs closed-loop verification and Kalman filtering fusion on a visual result by using cutting load data. According to the method, the problems of difficulty in lithology identification and lagging of cutting parameter adjustment in the underground high-dust environment are solved.

The invention relates to the technical field of inverter control, in particular to a self-adaptive MPPT inversion control method and system based on virtual inertia dynamic coupling. According to the scheme, the theoretical maximum power (Pmpp) and the dynamic reduction margin (delta P) are calculated by monitoring the output of a photovoltaic array and the frequency change of a power grid in real time, and an adjustable power target Pref is generated. And according to a frequency change rate (RoCoF), dynamically calculating virtual inertia power Pinertia, coupling the virtual inertia power Pinertia with Pref to form an active instruction Pcmd, and controlling an inverter to dynamically adjust power generation power on the premise of ensuring rapid inertia response to form a closed-loop adaptive system. According to the scheme, the power margin is dynamically established, the frequency change rate is mapped into the virtual inertia demand, the virtual inertia demand is coupled to the active instruction after slope limitation, and the problems of inertia support and power sudden change suppression of the photovoltaic grid-connected system are solved.

This invention relates to the field of intelligent molding technology, specifically to an intelligent adjustable medical cotton ball molding method and system, comprising the following steps: collecting pressure data through a distributed pressure sensor array and generating a pressure feature vector through Gaussian filtering for noise reduction; analyzing the pressure change trend based on the feature vector and outputting a gradient adjustment command; controlling a valve to adjust the air pressure to obtain compensation parameters; driving a servo motor to adjust the mold gap and collecting offset data to generate a feedback signal; and combining the feedback and adjustment command inputs to a neural network to evaluate the molding result. In this invention, data is collected through a distributed sensor array, multi-dimensional noise reduction enhances feature extraction, dynamic functions analyze pressure trends, offset adjustment parameters are identified, collaborative control achieves closed-loop compensation, displacement sensors provide feedback on gap changes, parameter modeling optimizes accuracy and consistency, convolutional networks extract temporal features, and fusion analysis constructs an adaptive system to achieve intelligent prediction and optimization of molding quality.

The invention provides a ground fault two-stage positioning detection method and device under a power-off state of a power distribution network, computer equipment and a medium, and the method combines a coordinate adaptive Transform system with StarNet-G, and is used for positioning a ground fault under the power-off state of the power distribution network. According to the method, firstly, a coordinate adaptive Transform system in the method fuses the advantages of a Swin adaptive coordinate Transform, an IC-DarkNet and a NasNet through integrated learning, and is used for accurately positioning a fault area; secondly, in the method, a weighted global feature module is introduced into StarNet-G, so that multiple faults in an area can be finely positioned under the condition that the data volume is small. According to the ground fault two-stage positioning detection method based on the coordinate adaptive Transformer system and the StarNet-G in the power-off state of the power distribution network, the problem that due to the fact that characteristic signals of multiple fault sources in a fault area are mutually overlapped, the fault in the area cannot be accurately positioned can be solved, and after switching-on of the power distribution network fails, the fault in the area can be accurately positioned. And the grounding fault of the power distribution network in the power-off state can be quickly and accurately positioned.

Systems, computer program products, and methods are described herein for dynamic security exposure assessment. The present disclosure is configured to query a variety of public and private sources to capture relevant exposure information about a third party. Machine learning models analyze this information to identify key exposure vectors. Based on these vectors, an exposure assessment model is dynamically generated and deployed within the third party's network environment to evaluate their security posture. This approach enables real-time, comprehensive assessment of third-party security exposures by integrating advanced data aggregation and machine learning techniques.

This invention provides a keyboard protocol adaptive system and method for embedded graphics, belonging to the field of embedded graphics display and control. It includes an application layer, a graphics system layer, a plug-in layer, a driver layer, and a hardware layer. The hardware layer is connected to the driver layer for unidirectional information transmission; the driver layer is connected to the plug-in layer for unidirectional information transmission; the plug-in layer is connected to the graphics system layer for bidirectional information transmission; and the graphics system layer is connected to the application layer for unidirectional information transmission. The driver layer includes independently configured USB 1.0 / 2.0 keyboard device driver modules and PS2 keyboard device driver modules. The plug-in layer includes a keyboard multi-protocol adaptive processing module, and the graphics system layer includes a keyboard event processing module. This invention also provides a method for operating the above system. This invention achieves adaptive support for USB 1.0, USB 2.0, and PS2 protocol type keyboard devices in a domestically produced embedded graphics system.

The application provides a rehabilitation training difficulty self-adaptive system based on motion capture, and relates to the technical field of rehabilitation training.The rehabilitation training difficulty self-adaptive system based on motion capture comprises a signal processing module, a muscle movement acquisition module, an electric stimulation module, a self-adaptive adjustment module and a neuromuscular evaluation module.The output end of the signal processing module is connected with the self-adaptive adjustment module and the neuromuscular evaluation module.The output end of the muscle movement acquisition module is connected with the input end of the signal processing module.The output end of the electric stimulation module is connected with the input end of the muscle movement acquisition module.The output end of the self-adaptive adjustment module is connected with the input end of the electric stimulation module.Through the signal processing module, the muscle movement acquisition module, the electric stimulation module, the self-adaptive adjustment module and the neuromuscular evaluation module, the motion trajectory and the motion speed of the rehabilitation training can be conveniently acquired, and the difficulty of the rehabilitation training can be self-adapted, so that the switching is accurate.

This invention relates to a MTPA control method based on a permanent magnet synchronous motor (PMSM), and pertains to the field of motor control. The invention addresses the problem in existing MTPA control methods that neglect parameter variations, making precise MTPA control difficult. This invention obtains the direct-axis inductance L based on a model reference adaptive system. d And stator resistance R; based on direct-axis inductance L d A torque equation is established with respect to the stator resistance R; a virtual high-frequency signal is injected to obtain the current angle at the MTPA operating point; based on the current angle at the MTPA operating point and the stator current vector, the quadrature-axis current and direct-axis current during MTPA operation are obtained. This invention constructs a new torque estimation equation with higher accuracy by using these parameters and high-frequency signal injection, thus achieving high-precision motor control.

This invention discloses a sensorless switching control method based on a variable parameter model reference adaptive system, relating to the field of permanent magnet synchronous motor (PMSM) control technology. Traditional switching control methods for PMSMs suffer from problems such as large speed fluctuations, severe voltage and current overshoot, high computational resource consumption, and complex engineering implementation when the position sensor fails. This invention employs a sensorless algorithm based on a variable parameter model reference adaptive system and designs a staged switching control strategy. By optimizing key parameters such as the adaptive law PI parameters, demodulation value weights, and switching delay time, a smooth switching between sensor-on and sensorless control is achieved. This invention significantly reduces speed fluctuations, voltage and current overshoot, and improves motor operation stability and reliability while requiring relatively little processor memory and computational resources. It is applicable to PMSM drive systems in civilian and military fields.

The application relates to a belt conveyor surface tearing detection method based on image processing, which comprises the following steps: setting hardware, obtaining a mask through reference images, processing the to-be-detected images by using the mask, and analyzing whether defects exist in the to-be-detected images. The belt conveyor surface tearing detection method based on image processing can identify various defect types such as cracks and scratches in real time. Compared with a traditional manual detection method, the adaptive system can process a large amount of data in a shorter time, reduces human errors, and improves the reliability of detection.

The invention relates to the technical field of commercial vehicle electric drive axles, and discloses an electric drive axle flow self-adaption system which comprises a lubricating part, a main oil way, a plurality of branches and an electronic control unit, the main oil way is sequentially connected with an oil pan, an electronic oil pump and an oil cooler through an oil way, the branches are led out of an outlet of the oil cooler, and the branches are arranged in parallel. Each branch is sequentially connected with an electromagnetic valve, a pressure compensation valve, a throttling hole and a corresponding lubricating part through an oil duct, and the electronic control unit is connected with the oil temperature sensor, the oil pressure sensor, the electromagnetic valves and the electronic oil pump. The flow self-adaptive system of the electric drive axle performs hierarchical control on different working conditions, adjusts the rotating speed of the electronic oil pump in real time, and performs self-adaptive compensation on the power of the electronic oil pump; each loop can be controlled independently, a valve back compensation structure is adopted, it is guaranteed that flow is accurately distributed in proportion and is not affected by loads, throttling and overflow losses are reduced, and system power consumption is reduced.

This invention discloses an island / reef adaptive system and its installation method, comprising: a floating breakwater energy storage system, an energy storage tank structure, and a control structure. The floating breakwater energy storage system includes a semi-submersible tank, and a first power generation module and a first energy storage device respectively disposed within the semi-submersible tank, the first power generation module and the first energy storage device being electrically connected. The energy storage tank structure includes a water storage tank, an energy storage tank structure body, and a second energy storage device, the water storage tank and the second energy storage device being respectively disposed within the energy storage tank structure body. The water storage tank includes a second power generation module and a water storage tank body connected to the semi-submersible tank, the second power generation module being disposed within the water storage tank body, the second energy storage device being electrically connected to the first energy storage device, and the water storage tank being electrically connected to the second energy storage device. The control structure includes a control center and an intermediate station for connecting the power consumption area and the energy storage tank structure, the intermediate station being electrically connected to the control center. The structural design of this solution effectively improves space utilization efficiency and long-term stability.

According to the stirring and mixing device capable of improving the uniformity, a periodic sweeping track is formed in the axial lifting process of the flexible stirring belt, a laminar flow dead zone of a traditional stirrer is broken through, radial cutting of the spiral blades and axial vortex superposition of the stirring belt enable materials to form complex convection in a three-dimensional space, and the uniformity of the materials is improved. The traditional'pre-grinding and mixing 'two-step process is integrated to be completed by single equipment, the component segregation risk caused by material transfer is reduced, and through a vortex-lifting-grinding triple coupling mechanism, remarkable advantages are formed in the three aspects of mixing uniformity, process efficiency and equipment integration degree; and the method is particularly suitable for the fields of nano materials, biological agents and high-added-value chemicals with strict requirements on component consistency. The core value of the system is that crushing, dispersing and mixing processes of traditional separation are integrated into a dynamic self-adaptive system, and an innovative normal form is provided for process-equipment collaborative optimization in intelligent manufacturing.

The present application relates to the field of mine intelligent tunneling, and discloses a hard rock lithology identification and cutting parameter adaptive system based on machine vision, which comprises an image acquisition module, a lithology identification module, a rock hardness mapping module and an adaptive control module. The image acquisition module is integrated with air curtain cleaning and double-mode light supplementing devices to obtain rock surface images of the tunneling working face; the lithology identification module extracts lithology features by using an adaptive enhancement algorithm and an improved deep learning network; the rock hardness mapping module quantitatively outputs rock hardness values through a regression model; and the adaptive control module dynamically adjusts the rotating speed and feeding speed of the cutting head according to the rock hardness values, and smoothes mechanical impact through a parameter smoothing unit. In addition, a feedback optimization module uses cutting load data to perform closed-loop verification and Kalman filter fusion on the visual results. The present application solves the problems of difficult lithology identification and lagging cutting parameter adjustment in the high-dust underground environment.

The application provides a kind of urban rainwater system scheduling method and device considering train effect, related to rainwater system scheduling technical field, the dynamic drainage capacity of urban rainwater system is comprehensively evaluated, and according to the characteristic parameters of train effect rainfall event, pre-emptying capacity is calculated in advance, and then adaptive system scheduling instruction is dynamically generated, which can effectively break the paralysis dilemma caused by hydraulic correlation accumulation when traditional rainwater system faces high frequency, short interval strong convective weather, greatly improving the comprehensive resilience of urban flood control system to train effect.

A system and method for Distributed Fiber Optic Sensing (DFOS) recognition and adaptation, which utilizes a memory-augmented neural network (MANN) and a disaggregated computing infrastructure (DCI) to achieve double privacy protection. The system separates the feature extraction encoder on a client-side machine from an external memory bank and similarity-based classification module on a server-side machine. In operation, the client computes an embedding from raw sensing data and transmits only the embedding vector to the server. The server performs classification and, during model fine-tuning, calculates and returns a gradient vector with respect to the embedding. This architecture ensures proprietary training data on the server is never exposed, while client raw sensing data remains private. The system supports continuous model adaptation and robust class-incremental learning.

The invention discloses a refrigerant dynamic charging process based on complete machine simulation operation, and belongs to the technical field of air conditioners and heat pumps. The problem that existing refrigerant charging depends on a static target value and cannot adapt to system individual differences and working condition changes in a self-adaption mode is solved. According to the process, a digital twinborn simulation model of a to-be-charged air conditioning system is constructed, the system is started to operate under a preset simulation working condition, and a first operation parameter is collected to initialize and calibrate the model; dynamically filling the refrigerant according to a preset step length, continuously collecting a real-time second operation parameter, and feeding the real-time second operation parameter back to the model; on the basis of real-time data, the model performs online fitting on a changing curve of parameters such as the supercooling degree and the superheat degree along with the charging amount, and predicts the refrigerant amount when the energy efficiency ratio of the system reaches the optimal as the optimal charging target; and when the parameters collected in real time and the extreme value region predicted by the model converge, it is judged that filling is completed. According to the system and the method, 'tailoring 'type dynamic accurate charging aiming at each specific air conditioning system is realized, and the charging precision and the system operation energy efficiency are remarkably improved.

The invention relates to the technical field of peak regulation of thermal power generation equipment, in particular to a wide-load adaptive system for deep peak regulation feed water temperature of a large circulating fluidized bed boiler. Comprising a water supply main pipeline, a multi-source heat exchange unit (a low-pressure economizer, an auxiliary electric heater and a waste heat exchanger), a parameter acquisition unit (a temperature sensor, a load sensor, a bed temperature sensor and a pressure sensor), a self-adaptive control unit (a fuzzy PID + load prediction algorithm) and a redundancy execution unit. The self-adaptive control unit dynamically dispatches the heat exchange priority of the multi-source heat exchange unit according to the boiler load change trend acquired by the parameter acquisition unit (high load preferentially utilizes waste heat / low-pressure coal economizer and low load starts auxiliary electric heating), fault seamless switching is achieved in combination with the redundancy execution unit, and meanwhile the self-adaptive control unit has the coking early warning function. According to the system, the feed water temperature can be accurately controlled at + / -2 DEG C under the load of 30%-100%, the energy consumption is reduced by more than 15%, the fault-free operation time is prolonged by 30%, and the boiler peak regulation reliability and economical efficiency are effectively improved.

The invention discloses a dynamic scene image domain adaptation system and method based on four-dimensional Gaussian sputtering and a computer storage medium, and relates to the field of computer vision and computer graphics, and the method comprises the steps: obtaining dynamic images captured at different time points from a plurality of visual angles; generating a four-dimensional Gaussian model based on the dynamic image; decomposing the four-dimensional Gaussian model into a conditional three-dimensional Gaussian model and an edge one-dimensional time component; extracting a target domain embedding vector; based on the extracted target domain embedding vector and the embedding vector of the three-dimensional Gaussian model, performing affine transformation on each Gaussian point, and mapping the Gaussian representation to the distribution of the target domain; and the multi-view consistency is maintained by predicting the corresponding relationship between different training views. According to the dynamic scene image domain adaptation method based on four-dimensional Gaussian sputtering provided by the embodiment of the invention, the four-dimensional Gaussian is decomposed into the conditional three-dimensional Gaussian and the one-dimensional Gaussian based on time distribution by utilizing extension, so that the multi-view visual consistency can be ensured.