11results about How to "Maintain order" patented technology

The invention provides a network node classification method based on a multilayer graph neural network model, and the method comprises the steps: constructing a graph according to network nodes, carrying out the preprocessing of initial node features in the graph, and obtaining an initial embedded vector; according to the initial embedding vectors of the node and the neighbor node thereof, updating the node representation by adopting a layered message passing mechanism; according to the initial embedded vectors of the nodes and the neighbor nodes thereof, the graph attention network uses an attention mechanism to calculate weights for the neighbor nodes of each node and then dynamically updates feature vectors of the nodes; a multi-head attention mechanism is adopted to splice and merge the feature vectors of the nodes output by the graph attention network and the updated node representation, and the comprehensive feature vector of each node is obtained; the comprehensive feature vectors are further optimized through graph contrast learning and a node embedding module, and optimized features are obtained; and obtaining predicted node classification according to the optimization features. According to the method, complex features of heterogeneous graph data can be effectively captured, the method has good robustness for an over-smoothing problem, more accurate heterogeneous network representation learning is realized, and the method has advanced performance and robustness in a node classification task.

The utility model relates to the technical field of entrance guard, and discloses a parking lot entrance guard control device which comprises a first control box, a second control box is arranged on the right side of the first control box, motors are fixedly connected to the rear side of the first control box and the rear side of the second control box, and gears are fixedly connected to the output ends of the two motors. Lifting rods are slidably connected to the interiors of the first control box and the second control box, rack plates are fixedly connected to the opposite faces of the two lifting rods, sliding plates are fixedly connected to the bottom ends of the middles of the two lifting rods, and a brake bar is arranged between the two lifting rods. According to the utility model, through the mutual cooperation of the control box I, the control box II, the motor, the gear, the lifting rod, the rack plate, the sliding plate and the sliding chute, the lifting control of the brake bar is realized, the entering height of the parking lot can be limited, the mistaken entering of ultrahigh vehicles can be prevented, only vehicles meeting the standard enter the parking lot, and the order in the parking lot is maintained; and confusion caused by over-high vehicles is avoided.

The invention relates to the technical field of intelligent power grids, and provides an electric energy metering box with a security structure, which comprises a box body, the box body is provided with a through hole, a first clamping piece is arranged on the through hole, the first clamping piece is provided with a plurality of first threading holes, the first threading holes are arranged in an arrayed manner, and cables can pass through the first threading holes; the second clamping piece is rotatably arranged on the first clamping piece, the second clamping piece blocks or cancels part of the threading hole after rotating, the locking piece is slidably and rotatably arranged on the first clamping piece, and the locking piece presses or cancels pressing the second clamping piece after sliding and rotating. According to the technical scheme, the problems that in the prior art, an electric energy metering box is prone to electricity stealing, and an electricity stealer is prone to electric shock are solved.

The invention provides a short video tracing method and system, a storage medium and electronic equipment, and the method comprises the steps: obtaining an access request, and judging whether the access request is legal or not; if the access request is illegal, extracting a target URL feature from the access request, and traversing all URL honey points in a preset URL knowledge base according to the target URL feature to obtain a similarity value between the target URL feature and the URL feature of any URL honey point; screening out a target URL honey point from a preset URL knowledge base according to the similarity value, and obtaining IP address information, port information and URL path information of the target URL honey point; and modifying the access request according to the IP address information, the port information and the URL path information of the target URL honey point. According to the method and the device, some specific access requests are forwarded, so that the host deployed with the target URL honey point can attract an attacker to actively attack the target URL honey point by using the access request, the target URL honey point is easily and successfully trapped, and the network defense effect is improved.

The invention provides a fluorescent VPCR identification method for a pinellia ternata adulterated counterfeit product, and belongs to the technical field of drug detection. According to the identification method established by the invention, the extraction time of the DNA of the pinellia ternate medicinal material can be remarkably shortened, one or more of four traditional Chinese medicinal materials such as pinellia ternate, brooklet anemone root, rhizoma typhonii flagelliformis and rhizoma arisaematis and decoction pieces can be identified at the same time, the identification duration is controlled within 1 hour, and meanwhile, the method has excellent specificity, sensitivity, accuracy and applicability and can be widely applied to the field of clinical application. The method can identify 5% of mixed products adulterated in Pinellia ternate, and can provide a new means for quality monitoring of Pinellia ternate medicinal materials and decoction pieces thereof, thereby ensuring lawful rights and interests of consumers and maintaining the normal order of the traditional Chinese medicinal material market.

The invention relates to the technical field of industrial park inspection, and discloses an anomaly detection method and system for an industrial park unmanned inspection machine and a storage medium, and the method comprises the steps: generating continuous frame images through a camera module, recognizing a temporary vehicle, judging whether the temporary vehicle is a moving or static vehicle according to a pixel total displacement value, and recognizing the information of a road where the vehicle is located; the temporary distance between the dynamic vehicle and the nearest static vehicle is calculated when the dynamic vehicle is on the non-traffic road, and first early warning information is sent if the temporary distance does not reach the standard; and sound parameters are generated through the pickup module, and second early warning information is sent after the preset characteristic frequency is matched. And obtaining a thermal image, calculating a heat value of a pixel area to identify a thermal abnormal state, and if the position is matched with the temporary vehicle, sending third early warning information. And calculating early warning data by integrating the three kinds of early warning information, and sending an assistance prompt to the background when the threshold value is exceeded. Through multi-dimensional data detection, early warning data is comprehensively generated, abnormal conditions of vehicles can be timely and accurately found, assistance prompts can be sent to the background according to needs, and the safety of the park is guaranteed.

The invention relates to the technical field of vehicles, in particular to a vehicle sales place identification method and device, electronic equipment, a medium and a product, and the method comprises the steps: obtaining the sales data of a vehicle and the data of a vehicle built-in Internet of Things card; determining one or more target to-be-identified vehicles according to the sales data and the vehicle machine built-in Internet of Things card data, and obtaining vehicle network data of each target to-be-identified vehicle; and calculating a local rate of the target to-be-identified vehicle according to the vehicle network data, and generating an identification result of the target to-be-identified vehicle according to the local rate. Therefore, the problem that the vehicle sales area cannot be accurately identified in the prior art is solved.

The utility model discloses a kind of blanking die of motor connecting plate, it relates to blanking die technical field, including lower mould and fixed mechanism, the fixed mechanism includes the first slide bar welded in the top of lower mould, the first slide bar is slidably connected with mobile clamp, the mobile clamp is fixedly connected with L-shaped rod, the L-shaped rod is threadedly connected with bidirectional threaded rod, the lower mould is fixedly installed with first motor, the lower mould is fixedly connected with support rod, the support rod is fixedly connected with base. By setting fixed mechanism, not only can realize accurate positioning and stable clamping to motor connecting plate workpiece, but also can flexibly adapt to different size sheet material, whether smaller size or larger specification motor connecting plate sheet material, position adjustment can be carried out when stamping, so that different size workpiece is firmly constrained in predetermined position, ensure that connecting plate blank is accurately separated along preset contour line.

The present invention relates to the technical field of port management, and in particular to a port data analysis, prediction and management method based on a cloud platform, comprising: obtaining historical port data on a cloud platform and classifying them; training all port data to obtain an analysis and prediction model, and determining the influence coefficient of each single variable; obtaining application data of ships to be port-entered, simulating and predicting the expected arrival date of the ships according to the analysis and prediction model, determining the ship berthing type according to the ship cargo data to determine the ship berthing area; determining the management type of the ship in the port according to the information of the ships to be port-entered; determining the ship management method according to the ship management type in the port; determining whether to update the analysis and prediction model according to the expected arrival date of the ship and the actual arrival date of the ship; the present invention manages the port by intelligently identifying the matching of the ship unloading and the cargo to be loaded, and predicting the arrival time of the ship according to the model, thereby optimizing the port operation process and improving the port operation efficiency.

The utility model provides an anti-cheating system used for transportation vehicle weighing, relates to the industrial technology field, especially relates to the wagon balance weighing technology field, comprises a wagon balance and an echelette grating device, and is characterized in that the ground at four corners of the wagon balance is respectively provided with a support rod, the support rods are used for fixing the echelette grating device, and the support rods are used for fixing the echelette grating device. The echelette grating device comprises an echelette grating transmitting end and an echelette grating receiving end; the echelette grating transmitting end and the echelette grating receiving end are respectively fixed on two supporting rods on the same side of the wagon balance; the system also comprises an alarm device which is connected to the alarm signal output end of the echelette grating device. The supporting rod is fixed to the ground through a locking mechanism. The structure of the supporting rod and the locking mechanism is adopted, the echelette grating device is effectively fixed, and the problem that a system is unstable and unreliable due to movement or external interference is avoided.

The invention relates to the technical field of boiler equipment, in particular to an ultra-low smoke emission device for a large boiler. The boiler is installed on the support, a desulfurization tower used for flue gas desulfurization is further arranged on the support, and a denitration furnace, an oxidation absorption chamber and a pair of dust removal chambers are sequentially communicated between the boiler and the desulfurization tower from left to right; the proportioning assembly is arranged on the bracket and is used for fusing and proportioning the urea solution; the spraying assembly is arranged on the denitration furnace and used for spraying the urea solution. According to the invention, a denitration mode combining SNCR and a low-temperature oxidation denitration process is adopted, and a plurality of links such as denitration, dust removal and desulfurization are integrated, so that flue gas emission can reach an ultra-low emission standard, the overall adaptability and cost effectiveness of the device are remarkably improved, and moreover, the device is suitable for industrial production. According to the invention, by using the atomizing ejector in rotary spraying operation in the denitration process, not only is the denitration efficiency improved, but also an efficient and economical solution is provided for the field of environmental protection.