2739 results about "Smart manufacturing" patented technology

The invention discloses an intelligent industrial security cloud gateway equipment system and method. The industrial cloud security gateway equipment and method is characterized in that upper cloud data is encrypted and added with data classification identifiers and time stamps, and decryption, security scanning and multiple security authentication access control are carried out on data on downlink intelligent equipment, so that interconnection, protocol conversion, and multi-directional transparent and real-time transmission among cloud equipment, enterprise datacenter equipment and various intelligent equipment on an industrial field can be achieved. An intelligent industrial cloud security gateway is embedded intelligent equipment comprising various serial ports, communication interfaces for a field bus and field intelligent equipment such as the industrial Ethernet, communication interfaces for wireless mobile communication and wired Internet cloud equipment, wireless or wired local area network interfaces for Bluetooth, broadband and the like, an Ethernet switch, a GPS module, and embedded software. The intelligent industrial cloud security gateway can systematically prevent user data leakage and network virus attacks, is core equipment of a secure and reliable intelligent industrial cloud service system, and can serve intelligent industrial manufacturing.

An automated artificial intelligence method and system for the design and implementation of engineering and manufacturing design services.

A system and method for using near-infrared or short-wave infrared (SWIR) light sources for identification of counterfeit drugs may perform spectroscopy using a super-continuum laser to provide detection in a non-contact and non-destructive manner at stand-off or remote distances with minimal sample preparation. Also, near-infrared or SWIR light may penetrate through plastic containers and packaging, permitting on-line inspection and rapid scanning. The near-infrared or SWIR spectroscopy may also be used to detect illicit drugs and their chemical composition. Moreover, the spectroscopic techniques may also be applied to quality assessment and control in pharmaceutical manufacturing, thus permitting the implementation of smart manufacturing with feedback control. Fiber super-continuum lasers may emit light in the near-infrared or SWIR between approximately 1.4-1.8 microns, 2-2.5 microns, 1.4-2.4 microns, 1-1.8 microns. In particular embodiments, the detection system may be a dispersive spectrometer, a Fourier transform infrared spectrometer, or a hyper-spectral imaging detector or camera.

The invention discloses an intelligent manufacturing plant virtual debugging and virtual monitoring method and system based on digital twinning. In the method, real-time cooperation between a digitalmodel of a product and a field device is realized through an intelligent manufacturing plant platform based on digital twinning. A product manufacturer/product user can grasp the use state of the product in real time on a product digital model of a control center and perform remote monitoring and fault diagnosis on the product/equipment, thereby realizing predictive maintenance and avoiding unplanned downtime of the equipment, and meanwhile the operation data of the product are fed back into the product digital model to help the manufacturing enterprise to continually optimize and improve theproducts and to provide basic data for the intelligent manufacturing at the same time.

The invention discloses a method for uniquely solving the inverse kinematics numerical value of the joint type mechanical arm, belongs to the technical field of modern intelligent manufacturing, relates to the technical field of industrial robots, relates to a method for uniquely solving the inverse kinematics numerical value of a joint type six-degree-of-freedom mechanical arm with shoulder joints facing forward. According to the method, a mechanical arm joint coordinate system is built according to an improved DH parameter method, four structural geometrical parameters between adjacent joints of the mechanical arm are determined, a homogeneous coordinate transformation matrix of two adjacent coordinate systems is calculated. For a given pose matrix of a tail end coordinate system O6, animproved newton iteration method, namely a Levenberg marquardt iterative algorithm is adopted, the inverse kinematics solution of the joint coordinate system is calculated by using a jacobian matrix J, and the six joint rotation angle values theta i which correspond to the pose matrix and meet the precision requirement are obtained. The method overcomes the requirement that a traditional newton iteration method needs to fully rank the jacobian matrix j, the modeling method is simple, clear and effective, the method has the characteristics of being high in solving precision, high in solving speed and simple and feasible in solving process.

The invention discloses an intelligent manufacturing special network data acquisition scheduling system. On the basis of conception of forwarding and control plane separation, a 4G special line intelligent scheduling system facing intelligent manufacturing is constructed, moreover, the system further satisfies concentrated scheduling and decentralized domain management functions, flexible network resource calling and rapid disposition capability are realized, standard northwards API interfaces are provided for the external, and concentrated management of wireless access equipment, on-demand scheduling and rapid security and other value-added service business opening are realized through SDN and NFV technologies. The system comprises five parts, including an internal network connected with manufacturing equipment in enterprises, a wireless access network based on 4G, a novel wireless access gateway connected with the internal network and an external network, a cloud platform used for carrying out concentrated network scheduling management, and various applications for intelligent manufacturing. The full-network configuration is realized through employing the cloud platform, network interconnection is realized through the novel access gateway, for network element functions, an SDN method is employed to separate a control plane from a data plane in an adaptive mode.

The invention discloses a point cloud based quick reconstruction STL digital model generation method. The method includes performing simplified filtering and characteristic extraction on input point cloud; performing curve reconstruction on surface point cloud on a digital model; establishing a model topologic relation on the STL digital model and performing grid simplification; performing hole detection and repair on the simplified STL digital model so as to remove defects in the model. According to the invention problems of laser scanning point cloud based STL model reconstruction can be solved effectively. The STL model is generated through steps including point cloud characteristic extraction, model reconstruction and the like and automatic repair of surface defects of the generated STL model is realized through grid simplification and hole repair technology, so that quick reconstruction of the STL digital model is realized and the availability of the reconstructed STL digital model is improved in the premise of keeping basic characteristics of point cloud, and great demands in different fields such as robot offline programming technology, automatic track planning technology and the like in a smart manufacture process can be met.

The present invention belongs to the related technical field of intelligent manufacturing, and discloses a production site data online collection analysis and fault prediction system. The system is constructed based on B/S configuration, and comprises a bottom-layer sensing network unit, a lower computer collection state unit and an upper computer server unit, wherein the lower computer collectionstate unit is in signal connection with the bottom-layer sensing network unit, and is configured to obtain various real-time data comprising a device state, a production process, product quality andlogistics information of an intelligent manufacturing workshop in real time and extract feature data; the upper computer server unit is configured to perform visual display of workshop poses and operation states based on the real-time data, predict and obtain faults of each device based on the feature data and give a targeted fault maintenance measure. According to the invention, the uniform networking management of the intelligent manufacturing workshop devices can be achieved to thoroughly eliminate an information island and mainly give a targeted solution means for the device faults and energy consumption.

The invention discloses a three-dimensional convolutional neural network-based workflow identification method. Only different process tasks are divided in advance and different action behaviors are tagged artificially in a video analysis process, so that the automation demand of intelligent manufacturing is not met. An inter-frame difference method with an adaptive threshold is proposed firstly and is mainly used for segmenting out a region of a moving object from a complex background, so that the time complexity of subsequent feature extraction and model training is lowered; secondly, a 3D convolutional neural network is improved to fully adapt to a factory environment with a plurality of monitoring devices, and for different views, a view pooling layer is adopted for fusing views of different angles according to weights; and finally, a new action division method is proposed for performing automatic division on continuous production actions in a video, so that an automated workflow identification process is realized.

The invention discloses a three-dimensional vision accurate guiding and positioning method for automobile intelligent manufacturing automatic feeding. A camera is used for conducting photographing and image collecting on a component, and the position and the angle deviation of the component under a fixed coordinate system are calculated through image processing and in combination with the vision detection algorithm; and a robot is used for transmitting the position and angle information, the track of the robot is corrected, the robot is guided to arrive at the correct work position, and accordingly the robot automatically corrects the track for workpiece grabbing and feeding. According to the three-dimensional vision accurate guiding and positioning method, the component grabbing precision is high, the precision can reach +/-0.5 mm, accurate positioning on a material box is not needed, and the cost and maintaining expense for accurately positioning the material box are omitted; and the later maintaining and adjusting technology is facilitated.

The invention relates to the design of bearing assembly lines, in particular to a bearing assembly line based on industrial cloud platform intelligent manufacturing, comprising a cloud design and manufacture system, an MES (manufacturing execution system), an intelligent production system capable of automatically assembling bearings, and a parts WMS (warehouse management system); the cloud designand manufacture system provides collaborative design and inputs an optimized process to the intelligent production system; the MES generates a planning scheduling program according to acquired information of the WMS and input order information; the intelligent production system schedules production according to the planning scheduling program and performs automatic assembly and warehousing of bearings. The cloud design and manufacture system, the MES, the intelligent production system capable of automatically assembling bearings and a parts WMS are used herein to provide assembly, detection, warehousing and the like for bearings, the cloud design and manufacture system is effectively combined with the production line, an intelligent factory is achieved, and production efficiency is effectively improved.

The invention discloses a construction method of an intelligent workshop digital twin system, and belongs to the technical field of intelligent manufacturing. The method comprises the following steps: aiming at physical workshop production elements, constructing a workshop total element multi-level twin model to obtain a digital twin workshop model; constructing a data acquisition system to acquire multi-source heterogeneous data of a physical workshop, preprocessing and standardizing the multi-source heterogeneous data, and storing the preprocessed and standardized multi-source heterogeneous data into a database; by adopting the digital twin workshop model as a client and by combination with a server, achieving interaction between the digital twin workshop client and the database so that bidirectional mapping between the digital twin workshop client and the physical workshop is realized; based on three-dimensional visualization system integration, subjecting the physical workshop model and the digital twin workshop model to deep fusion interaction to finish construction of the intelligent workshop digital twin system. Therefore, the functions of real-time state display, historical state reproduction, remote control, fault alarm, state prediction and the like are achieved, and the problems that existing digital workshop data are monotonous in presentation and poor in interaction and immersion are solved.

The invention provides an intelligent manufacturing-oriented edge computing gateway and system. The computing gateway comprises a communication unit, a scheduling management unit, an execution unit and an interface unit, wherein the communication unit is an interface assembly for interaction between edge computing gateways, and sends a request to other edge computing gateways or outputs a result;the scheduling unit is used for managing a service request; the execution unit is used for executing a logic of a specific task; the scheduling unit and the execution unit are used for jointly executing decisions to realize corresponding control functions and logics and constrain behaviors and actions of the edge computing gateways; and the interface unit processes interactions between edge computing control and hardware, provides an access mechanism for manufacturing resources, monitors resource data and executes commands of the manufacturing resources. By a modularized architecture design and internal implementation structures of all the units, inter-module communication and module functional logic are effectively separated, so that the configurability of the system and the reusability of functions are improved.

The invention discloses a double-front-split-platform type industrial internet of things and a control method, and belongs to the field of intelligent manufacturing. The industrial internet of things comprises a user platform, a service platform, a management platform, a sensing network platform and an object platform which are interacted in sequence; the service platform is arranged in a centralized mode, and the management platform and the sensing network platform are arranged in a front-divided platform mode. The control method is applied to the industrial Internet of Things. In order to coordinate product manufacturing parameters of all production line equipment, a sensing network platform and a management platform are arranged in a front sub-platform manner, data are uniformly sorted, received and transmitted through a main platform, and different sub-platforms are used as data processing or transmission channels of different production line equipment, so that a large amount of data are stored and processed in a classified manner; the overall data processing amount and calculation pressure of the platform are reduced, data are processed on the sensing network platform, and safe and rapid product manufacturing control and data receiving and transmitting are achieved.

The invention relates to a composite type flexible pressure and displacement sensitive element and a preparation method thereof, and belongs to the technical field of measurement. The composite type flexible pressure and displacement sensitive element comprises a displacement sensitive layer, a pressure sensitive layer and an insulated encapsulation layer, wherein the displacement sensitive layer is formed by a copper foil coil covering a polyimide film; the pressure sensitive layer is formed by a conductive polymer composite material which is embedded in Polydimethyl Siloxane; the insulated encapsulation layer is formed by a flexible polymer material which is vulcanized at the outer parts of the displacement sensitive layer and the pressure sensitive layer; information of pressure and displacement can be obtained through detecting the impedance change of the copper foil coil. The composite type flexible pressure and displacement sensitive element prepared by the invention has dual functions of pressure measurement and non-contact displacement measurement and can be applied to the fields of multi-sensitive electronic skin preparation or pressure measurement and non-contact displacement measurement between small narrow curved layers of a whole machine and complete equipment in an intelligent manufacturing complete process.

The invention discloses a light metal matrix composite component and a preparation method thereof and belongs to the technical field of metal composite preparation and molding. According to the lightmetal matrix composite component and the preparation method thereof, the non-melt electrode inert gas shielded arc welding electric arc is adopted as a heat source to conduct material additive manufacturing on a light metal base plate subjected to surface pretreatment to obtain the light metal matrix composite component. A feeding method for material additive manufacturing comprises the steps of blowing zero-dimensional particle reinforcements into a liquid-state molten pool in the axial direction of a non-melt electrode welding gun through inert shielding gas to achieve coaxial powder feeding; and meanwhile, continuously feeding combined welding wires containing one-dimensional/two-dimensional linear/flaky reinforcements from the advancing side of the welding gun to the liquid-state molten pool to achieve paraxial wire feeding. Through the preparation method, the big-size composite component with the complex structure can be directly manufactured without subsequent processing or withlittle processing amount, and integration of material preparation and molding is realized. By adopting mixed reinforcements with different dimensions, the more superior mechanical properties of the material can be realized, the raw material cost is reduced, technological parameters are fewer, and automatic and intelligent manufacturing can be achieved easily.

The invention discloses a production operation behavior identification method, device, equipment and system based on a structured double-flow convolutional neural network and a storage medium, and themethod comprises the following steps: obtaining a monitoring video of a production environment, and decomposing the monitoring video into an image frame sequence and an optical flow sequence; inputting the image frame sequence and the optical flow sequence into a structured double-flow convolutional neural network to obtain an action category and timestamp information of each frame of image; introducing an attention mechanism into the structured double-flow convolutional neural network; and according to the action category and the timestamp information, obtaining a production action instancethrough calculation. According to the production operation behavior identification method provided by the invention, the production operation behavior in the video can be rapidly and accurately identified, and the intelligent manufacturing requirement is met.

The embodiment of the invention disclosesa performance degradation analysis based intelligent manufacturing device reliability assessment method. The method comprises the steps that 1, analysis is performed: a target performance degradationindex identifying and representing the usage reliability of an intelligent manufacturing device and target environment and working condition stress factors influencing the reliability are analyzed; 2, information acquisition is performed: target performance degradationindex data and target environment and working condition stress factor data are acquired; 3, assessment is performed: modeling analysis is conducted on the acquired target performance degradationindex data, and modeling introduction is conducted on the acquired target environment and working condition stress factor data, and reliability assessment is established by applying a bayesian method. Compared with the prior art, the performance degradation analysis based intelligent manufacturing device reliability assessment method can achieve real-time and efficient assessment of the usage reliability of the intelligent manufacturing device.

The invention discloses an intelligent manufacturing technology application training platform, relates to a system platform used in intelligent production and especially relates to the training platform suitable for automobile model production and assembling application. The intelligent manufacturing technology application training platform comprises a stereoscopic warehouse, a stacker crane, an AGV dolly robot, a wheel hub tire and wheel assembling system, a vehicle body assembling device, a vehicle body conveying line, a finished product transfer conveying mechanism, a magnetic guide rail and a control center. The stereoscopic warehouse and the stacker crane are arranged adjacently. A first conveying line is arranged between the stacker crane and the wheel hub tire and wheel assembling system. A second conveying line is arranged between the wheel hub tire and wheel assembling system and the finished product transfer conveying mechanism. The vehicle body assembling device is fixed on a vehicle body assembling bench and then is placed near the second conveying line. The vehicle body conveying line is arranged at a parallel position near the vehicle body assembling device. The magnetic guide rail is arranged on the ground. The design of the system is reasonable, the structure is compact, labor saving and time saving are achieved and the platform is high-efficient and intelligent.

The invention relates to an IM oriented edge computing controller and a work method thereof. The controller comprises a firmware layer, middleware and an application layer; the firmware layer drives abottom layer device, gathers and classifies collected data information, and sends the obtained data to an intermediate layer; the middleware provides the operating and development environment, constructs an equipment model according to the data, and provides equipment state data, logic control, control algorithms and the equipment model for the application layer; and the application layer receives program configuration of a host computer, controls equipment according to the equipment state data, logic control, control algorithms and the equipment model provided by the middleware, and realizesinteraction with an edge server via an onsite bus or real-time Ethernet. The controller is not only suitable for a robot or PLC in certain model, and standardized and modularized design ideas of thecontroller are suitable for control application of different types of industrial equipment. The interface and computing capabilities of industrial production equipment can be improved, and the applicability of the industrial equipment is improved.

The invention discloses a reinforcement learning method for dynamically optimizing logistics scheduling and path planning in a machining process, and belongs to the field of intelligent manufacturing.The method comprises the following steps: constructing a total state matrix of logistics scheduling at a certain moment as the input of the neural network; constructing a reward and punishment valueaccording to the total time length of all operation units waiting for materials in the primary processing process and the total number of conflict times of the AGVs, establishing a loss function of aneural network according to the reward and punishment value, training the neural network to maximize the reward and punishment value obtained in the final primary processing process, and obtaining a reinforcement learning decision model; and then for a new processing task, establishing a total state matrix at the current moment in real time, and inputting the total state matrix into the reinforcement learning decision model to obtain an action needing to be executed by the AGV at the current moment. According to the method, autonomous optimal path planning of the AGV in the machining process can be realized, and the technical problem that the logistics scheduling scheme cannot be dynamically and automatically adjusted in real time in the machining process in the prior art is solved.