60 results about "Embedded intelligence" patented technology

The invention relates to the technical field of fault diagnosis, and particularly discloses a bearing fault diagnosis method based on adaptive feature migration fusion. According to the method, an adaptive feature migration fusion algorithm is utilized, and multi-mode signals of acceleration, load, temperature and rotating speed are comprehensively processed, so that accurate fault diagnosis of the embedded intelligent bearing under a complex working condition is realized. Aiming at the problems that a traditional diagnosis method is difficult in signal feature extraction and large in data distribution difference under different working conditions, the invention provides a self-adaptive feature migration fusion algorithm, and the robustness and expression ability of fault features in different working conditions and data sets are improved by constructing a migration learning model and combining a feature selection and fusion mechanism. Compared with the dependence of a traditional deep learning algorithm on large-scale labeled samples, the method can efficiently utilize a small amount of label data, the requirements for hardware computing power and data resources are remarkably reduced, and an efficient and flexible embedded bearing fault diagnosis method is provided for online fault diagnosis of an industrial site.

The embodiment of the invention provides a convolution input structure multiplexing method and device based on signal sliding window row block cache, and relates to the technical field of embedded intelligent sensing. The method is applied to a front-end cache of a multiply-accumulate array, and comprises the following steps: determining a cache configuration parameter according to a parameter of an executed convolution operation; establishing a plurality of line caches in the front-end cache according to the cache configuration parameters; newly added data obtained through the sliding window operation is written into a row cache; and reading data from the row cache when the multiply-accumulate array executes a calculation operation. According to the implementation mode provided by the invention, the input redundant loading is remarkably reduced, the reuse rate of the overlapped part of the adjacent sliding windows is increased, and the average access frequency of the SRAM is reduced.

The invention discloses a universal shaped protection and monitoring method based on an Internet of Things intelligent system, and relates to the field of safety protection of building engineering construction. A universal embedded intelligent control box is embedded into various set protection facilities, a matched sensor group is used for collecting data, real-time self-diagnosis and risk assessment of the facilities such as edge guardrails and tool type step ladders are achieved by means of an edge calculation algorithm, then a dynamic risk assessment model is adopted to output a comprehensive risk weight value, and the risk assessment accuracy is improved. Triggering multi-level linkage alarm according to comparison results of the data and different levels of thresholds, and finally uploading the data to a cloud through a communication module for collaborative management and block chain evidence storage. According to the invention, the problems of monitoring lagging, passive early warning and the like of a traditional protection facility are solved, and the intelligent level and management efficiency of construction safety protection are improved.

The invention relates to embedded intelligent equipment, a system starting method and device thereof and a medium, and the method comprises the steps: responding to a system starting signal, and loading and running a system kernel of a pre-configured first operating system to drive hardware resources of the embedded intelligent equipment; in a first user space environment created by a system kernel starting initialization process, independent file system resources are created for a second operating system, and the second operating system and the first operating system are heterogeneous; mapping an access path of a runtime resource of the file system resource to a global file path of the first user space environment by creating a symbolic link, and constructing a second user space environment corresponding to a second operating system; and the initialization process is ended, so that support is provided for application programs of the first user space environment and the second user space environment based on the shared access capability of the system kernel to the hardware resources. According to the method and the device, millisecond-level dynamic switching of the double systems without restarting is realized, and meanwhile, the native performance and efficient resource utilization are ensured.

The utility model relates to the technical field of terminals, in particular to an embedded intelligent elevator number display terminal which comprises a panel, a floor display screen is installed in the middle of the outer end of the panel, a square hole is formed above the floor display screen, a terminal display screen is arranged behind the square hole at the same height, and a limiting plate is installed outside the terminal display screen. Placing grooves with the same length and width as the limiting plates are formed in the periphery of the square hole in the inner end of the panel, and the limiting plates are matched with the placing grooves; the fixing assemblies are arranged on the panel at the two ends of the containing groove, the fixing assemblies are used for fixing the terminal display screen to the panel, and the embedded intelligent elevator number display terminal aims at solving the problems that in the prior art, fixing is not stable, and installation is complex.

The invention discloses a virtual sensor modeling and deployment method and system based on spatio-temporal feature fusion and a gradient boosting strategy, and relates to the technical field of industrial process monitoring and control, embedded intelligent sensing and data-driven modeling. The method comprises the following steps: collecting multi-source sensor data under a multi-environment condition and preprocessing the multi-source sensor data; constructing a spatial-temporal characteristic system which simultaneously represents historical memory, dynamic change and a multivariable coupling relationship, and performing characteristic screening and weight reduction; under a gradient lifting framework, adopting automatic hyper-parameter optimization to obtain a virtual sensor model with compromise between performance and complexity; the trained model is converted into a unified reasoning format irrelevant to a platform, and real-time reasoning is achieved on a resource-limited embedded control unit. Compared with the prior art, the method has higher prediction precision and generalization ability under the complex dynamic working condition, the model size and the single reasoning delay are controllable, and the method is suitable for being deployed and applied in a vehicle-mounted ECU, an industrial controller and an edge node.

The utility model discloses an embedded intelligent control station, comprising a housing, the housing comprises a first side plate, a second side plate and a first top plate, the outer side of the first side plate and the outer side of the second side plate are provided with first heat radiation fins extending along a first direction, the outer side of the first top plate is provided with second heat radiation fins extending along a second direction, the first direction is perpendicular to the second direction; the control main board is mounted in the shell; the main control chip is mounted on one side, far away from the first top plate, of the control mainboard; one side of the heat conduction pad abuts against the main control chip, and the other side of the heat conduction pad abuts against the bottom of the shell; and the cooling fan is mounted on the outer side of the first top plate. According to the embedded intelligent control station, the heat dissipation efficiency of the embedded intelligent control station can be effectively improved, so that the working performance of the embedded intelligent control station is improved.

The application discloses a kind of based on ZYNQ platform's transmission tower steel member weld surface defect detection system and method, belong to industrial vision detection and embedded intelligent identification technical field.Pass through acquisition weld image and construct defect dataset, in YOLOV3 target detection network training stage introduction self-adapting residual network structure and channel attention mechanism, enhance shallow layer detail information transmission and defect relevant feature response, realize to crack, pore, slag inclusion, undercut and weld tumor and so on Defect Effective identification.Training optimized model faces ZYNQ heterogeneous hardware and is deployed, through hierarchical data access, convolution unit reuse, multi-level parallel computing and pipeline processing, reduce data movement overhead, improve convolution operation efficiency and on-chip resource utilization, reduce the hardware scale required for edge deployment, to improve the real-time performance and engineering application capability of weld surface defect detection.

1. The name of this design product: Embedded intelligent door control system. 2. Purpose of this design product: used to control the opening and closing of doors. 3. The key point of the design of this product lies in its shape. 4. The picture or photo that best illustrates the design points: three-dimensional picture.

The application provides an embedded intelligent switch control system based on dual-mode detection, and relates to the technical field of intelligent control, comprising: a driving logic program establishment module, which is responsible for configuring a dual-mode control channel of infrared and sound cooperation; a main control sensing establishment module, which is responsible for writing driving logic into a main control chip and connecting infrared and sound sensors; a signal determination module, which is responsible for driving sensor staggered detection and determining signals to generate switch instructions; and a switch state switching module, which is responsible for controlling a relay to switch a switch state according to the instructions. The application solves the technical problems of high false triggering rate, high energy consumption and poor scene adaptability of the conventional switch control system in the prior art due to a single triggering mode, achieves the technical effects of judging time zones and flexible calibration mechanism through infrared and sound dual-mode detection combination, reducing the false triggering rate and energy consumption in complex environments, and improving the scene adaptability.

1. Name of the product in this design: Embedded Smart Lock (DX40). 2. Purpose of this design: This design is for a smart lock installed on a wall for the purpose of theft prevention. 3. The key design feature of this product is its shape. 4. The image or photograph that best illustrates the design's key points: a 3D model. 5. The back of this product is a part that is not easily seen or cannot be seen during use, so the rear view is omitted.

The invention relates to the technical field of road diseases, in particular to a road disease vehicle-mounted embedded intelligent detection method based on deep learning, and the method comprises the steps: introducing a lightweight GSNet network, integrating a GC2f feature module of a global context attention mechanism, and fusing a dynamic up-sampling module DySample and a VoV-GSCSP module. The parameter quantity and the calculation complexity are obviously reduced; and meanwhile, the high detection precision is kept. In the model evaluation stage, the system analyzes the performance of each type of diseases in a simple scene and a complex scene by utilizing a defect detection result of a test image, respectively calculates the accuracy of a first type and a second type, and dynamically allocates an evaluation weight for each type and generates a weighted comprehensive evaluation result, so that the performance evaluation is more fit with the real road complexity. According to the method, light weight, high precision and scene adaptive evaluation capability are considered, the real-time performance, robustness and deployment reliability of road disease detection are effectively improved, and the method is suitable for application scenes such as intelligent inspection vehicles, road maintenance systems and automatic driving perception.

The invention relates to the technical field of intelligent spectrum sensing, in particular to a pure water quality total chlorine online monitoring method and device based on a spectrophotometric method. The system of the device serves as an integrated intelligent sensing system, and synchronously collects multi-wavelength absorbance, transmission light intensity and light source voltage data of a water sample; analyzing a light intensity fluctuation index based on the light source voltage data to evaluate light source stability; calculating the characteristic deviation degree of the water sample by combining absorbance data and the index; fusing the absorbance, the transmission light intensity and the characteristic deviation degree data to generate a detection credibility index; and finally, independently judging whether the current detection has a misjudgment risk or not according to the credibility. Through multi-dimensional data fusion and embedded intelligent analysis, self-diagnosis and result credibility self-evaluation of the sensing process are realized, and the robustness, reliability and intelligent level of an online monitoring system are remarkably improved.

This invention discloses a pet posture recognition and health monitoring method and collar, belonging to the field of pet wearable devices and embedded intelligent sensing technology. The method includes: collecting raw motion data of the target pet and preprocessing it; extracting acceleration and angular velocity features based on the preprocessed data to obtain temporal fusion features; inputting the temporal fusion features into a preset lightweight classification model to output the pet's posture category; performing periodic quantitative statistics on the posture categories to obtain behavioral statistics data; and comparing the behavioral statistics data with preset health judgment conditions to obtain relevant health status data of the target pet. This invention effectively distinguishes between effective and ineffective motion through temporal fusion features, achieves local offline inference through a lightweight classification model, and realizes pet health monitoring through a complete closed loop from posture recognition to health assessment. It features high recognition accuracy, low power consumption, and long battery life.

1. Name of the product in this design: Embedded Smart Lock (BF-Q9). 2. Purpose of this design: As a security and anti-theft device, it is used to lock and protect doors, boxes, cabinets, etc. 3. The key design feature of this product is its shape. 4. Images or photos that best illustrate the design points: Refer to Figure 1 for the combined state.

The invention relates to an industrial-grade general intelligent test hardware platform, which is characterized in that a traditional fixed function device is split into a plurality of independent detachable and replaceable function modules through modular design, so that the test requirements of various types of signals are flexibly met, free combination and emerging signal expansion are supported, and the problems of single function and expansibility are solved; a unified interface and a protocol adaptation layer are defined, one-button intelligent wiring is matched, the wiring process is simplified, manual errors are avoided, and the problems of interface disorder and operation complexity are solved; test parameters are automatically optimized based on an embedded intelligent control unit and an adaptive algorithm, multi-task parallel processing and real-time data analysis are supported, and efficiency and precision are greatly improved; a hierarchical open software system is constructed, a unified drive API is provided, multi-platform access, user-defined extension and cloud collaboration are supported, the limitation of closed software is broken, and the personalized and long-term development requirements of enterprises are met. The invention is suitable for the technical field of embedded product test systems.

The utility model provides an embedded intelligent lock module. The embedded intelligent lock module comprises a shell; the inner shell is arranged in the outer shell; the main control module is mounted in the inner shell; the face recognition module is in control connection with the main control module insertion fixing structure; the radar module is connected with the main control module in an inserted and fixed control mode, the face recognition module, the radar module and the main control module are arranged in parallel, and the face recognition module and the radar module are arranged on the same side of the main control module. According to the embedded intelligent lock module, the main control module is fixed in the inner shell, and the face recognition module and the radar module are directly connected with the main control module in an inserted mode through the contact pins, so that connection is convenient, the internal installation space is saved, and additional position fastening operation for the face recognition module and the radar module is omitted.

The application provides a liquid strain biomass soft measurement method based on machine vision, comprising the following steps: offline sampling and analyzing liquid strain biomass parameters and evolution rules in the whole fermentation period; establishing a biomass parameter soft measurement model M1 by means of manually measured auxiliary variable characteristics M-Features; collecting visible light images and visible-near infrared I-near infrared II multi-source spectrum data of liquid strains in different fermentation stages, constructing an atlas data set, and building a multi-task machine vision model V1 for extracting auxiliary variable characteristics E-Features based on machine vision; combining the biomass parameter soft measurement model M1 and the multi-task machine vision model V1, establishing the correlation between E-Features and M-Features, and obtaining a cascade machine vision soft measurement model MV; and deploying the MV model on a terminal embedded intelligent device and designing interactive software for task execution. The method can evaluate the trend of the evolution of the biomass parameter of the liquid strain fermentation process in real time, so as to guide the production application of the strain.

The invention discloses a computing power test method and system for an embedded system, an embedded intelligent computing system is tested through theoretical computing power and peak computing power tests, a peak computing method is provided, and a test result is obtained by calculating the peak computing power and the computing power utilization rate of a computing power test model. According to the method, through performance evaluation and benchmark testing, the calculation capability of the system is quantified, a basis is provided for model selection of a core processor of the embedded intelligent system, and through spaceflight testing of the peak calculation capability of the embedded intelligent hardware platform, the calculation capability of the embedded intelligent hardware platform is improved. A design base line is provided for optimization and simplification of a deep learning algorithm, intelligent development of spaceflight embedded equipment is facilitated, development and design of a later algorithm are facilitated, software and hardware in the spaceflight embedded equipment are guided to be optimized, computing resources are reasonably distributed, it is ensured that the embedded equipment stably operates under the condition of high computing load, and the development and design of the spaceflight embedded equipment are facilitated. And support is provided for the design of an intelligent embedded system.

The invention provides an embedded intelligent system for controlling power output of a horticultural tractor, and the system comprises a resistance induction module which is used for collecting and analyzing real-time soil operation environment factors of the horticultural tractor based on a sensor array, and determining target resistance borne by a grooving machine tool in the horticultural tractor; the data analysis module is used for evaluating the target speed of the horticultural tractor according to the target resistance borne by the grooving machine tool, analyzing the target speed according to the reference constant speed and determining a power adjusting mechanism of the horticultural tractor; the intelligent regulation and control module is used for automatically controlling the power output of the horticultural tractor according to a power regulation mechanism until the reference constant speed of the horticultural tractor is met; the intelligent control over power output of the horticultural tractor is guaranteed, so that the stability of the groove depth or the ditch depth is effectively guaranteed, and the effect of accurately controlling the tillage depth is achieved.

This invention discloses a high-efficiency heat exchange air source heat pump system and its control method, belonging to the field of air source heat pump heat exchange control technology. The system includes a heat pump hardware body and an embedded intelligent control software system. The software system constructs a perception-decision-execution-iteration closed-loop architecture, integrating load forecasting, intelligent heat exchange regulation, phase change collaborative control, fault diagnosis, and data interaction modules. Through multi-algorithm fusion, it achieves accurate heat exchange load forecasting, dynamic parameter adjustment, and fault protection under all operating conditions, solving problems such as control lag and poor coordination in traditional systems. This invention significantly improves load forecasting accuracy and reduces regulation lag through innovative software system architecture and multi-module collaboration, achieving precise coordination between phase change components and the main unit, accurate and efficient fault diagnosis, supporting continuous system optimization, significantly improving heat exchange efficiency and operational stability under all operating conditions, reducing energy consumption and maintenance costs, and possessing broad application prospects.

The invention relates to the technical field of DCS (Distributed Control System), and particularly discloses a DCS embedded intelligent monitoring system based on multi-algorithm fusion, which comprises a DCS embedded intelligent alarm analysis platform of a physically isolated control area and a multi-dimensional comprehensive early warning management platform of a management area, the DCS embedded intelligent alarm analysis platform is deployed in a gas power station production control network, is provided with an independent server cluster, and is used for collecting operation parameters and alarm information of DCS main and auxiliary networks and a TCS system in real time, and executing data modeling, alarm optimization, fault playback, video linkage and intelligent control functions; the multi-dimensional comprehensive early warning management platform is deployed in a gas power station management network, and control area data are obtained in a one-way mode through a domestic one-way isolation gatekeeper. According to the system, through embedded deployment and multi-module cooperation, functions of intelligent disk monitoring, alarm optimization, fault playback, intelligent control and the like can be integrated without large-scale transformation of existing DCS hardware.

The invention relates to the technical field of connectors, in particular to a rotary waterproof connector which comprises a female end, a male end and a sealing layer, the female end is connected with the male end, the sealing layer abuts against the position between the female end and the male end, and the rotary waterproof connector further comprises an embedded intelligent management system. Through the embedded intelligent management system, aiming at the problem that early warning of sealing failure is not timely, the system acquires the pressure P (t) of the sealing interface in real time and evaluates based on a dynamic failure threshold model, so that early, dynamic and accurate early warning of the sealing state is realized. Aiming at the problem of insufficient signal transmission reliability, the system monitors the impedance mismatch degree and temperature in real time, and starts adaptive impedance matching and online calibration by using a quantum-classical hierarchical optimization strategy and a quantum annealing algorithm, thereby ensuring signal integrity and transmission reliability in a high dynamic environment. In addition, multi-source data fusion and full-life-cycle data management provide high-quality data basis and continuous optimization capability for the functions.

This utility model relates to microprocessor technology, biometric identification systems, and embedded intelligent devices designed for automatic personal recognition based on multiple biometric characteristics. The proposed utility model is a multimodal microprocessor-based biometric identification device comprising a fingerprint scanner, a voice identification module, a microprocessor computing module, a memory module, a neural network data processing unit, a decision-making unit, a communication interface, and a power source. A distinctive feature of the utility model is its use of a unified microprocessor architecture, enabling local intelligent processing of multiple biometric modalities without the need to transmit data to external computing servers.The microprocessor computing module collects, digitally filters, extracts features, analyzes, and compares user biometric data with templates stored in the device's memory. Fingerprints and voice characteristics are used as biometric features. Combined processing of multiple biometric parameters improves identification accuracy, reduces the likelihood of false positives and false negatives, and enhances the system's resilience to external influences and noise. The technical result of this utility model is increased reliability, speed, and autonomy of biometric identification through the use of integrated microprocessor processing of multimodal biometric data and recognition algorithms.The utility model can be used in access control and management systems, banking and financial systems, mobile devices, medical information systems, critical infrastructure facilities, government personal identification systems, and intelligent IoT systems. The proposed device is compact, capable of autonomous operation, offers high data processing speeds, and can be integrated into modern digital security systems.

1. The name of the design product: light fastener (embedded KTV intelligent type). 2. The use of the design product: for light fastener. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view 1.

The invention discloses a high-flux intelligent spectrophotometric detection system and method, and the system comprises a multi-beam optical subsystem, a high-flux sample detection module, a multi-channel optical fiber collection system, and a spectrum detection and intelligent analysis subsystem. A plurality of samples in the high-flux sample detection module are irradiated at the same time, and generated transmission or reflection light signals are collected by the multi-channel optical fiber collection system and transmitted to the spectrum detection and intelligent analysis subsystem; the system and the method can perform spectrophotometric detection based on high-throughput sample treatment, high-sensitivity spectrum detection and embedded intelligent analysis.

The application relates to the technical field of communication base station monitoring, and discloses an embedded intelligent internet of things edge computing terminal and a monitoring method thereof in a communication base station environment, wherein the application comprises the following steps: identifying a data frame accessed through a serial port to obtain an effective monitoring value, generating an alarm triggering identifier, and selecting a main communication protocol from an IP network, an E1 special line, a PTN transmission network, an L2TP tunnel and a 4G network; obtaining the number of response timeout times after reporting data to a remote monitoring center, switching from a remote centralized mode to a local independent operation and maintenance mode, and creating an operation and maintenance mode identifier; judging whether a triggering identifier is reported according to the operation and maintenance mode identifier, the alarm triggering identifier and the effective monitoring value; if yes, transmitting the triggering identifier through the main communication protocol in a fixed-length frame structure or a fragmentation mode to obtain a reporting response message; the application reduces the false alarm rate caused by measurement noise, reduces network bandwidth occupation under the premise of ensuring the real-time alarm, improves communication efficiency and reduces operation cost.