5 results about "SAFE AIR" patented technology

The present application relates to the technical field of aviation safety and artificial intelligence, in particular to an ADS-B time series data adversarial sample generation method based on improved WGAN-GP, comprising the following steps: obtaining ADS-B historical flight data, extracting the longitude, latitude, height, speed and heading five-dimensional features to construct a time series data set; building an adversarial sample generation model based on an improved Wasserstein distance generative adversarial network, and configuring a multi-objective composite loss function for the generator; the present application builds a generator and a discriminator based on LSTM, designs a composite loss function that integrates adversarial loss, concealment loss and time series smoothness loss, adopts gradient penalty and asymmetric update strategy to train the model, and the generated adversarial sample has high concealment and physical consistency, can effectively avoid various unsupervised anomaly detection models, the model training is stable, and technical support is provided for building a safer air traffic control defense mechanism.

The application discloses a kind of user perceivable safe air car structure design method based on AI.First, by structural equation model path analysis, with structural safety as independent variable, willingness to use as dependent variable, determine attitude, initial trust, perceived ease of use as psychological intermediary path, and verify the nonlinear regulation effect of empathy characteristic;Then the psychological path is parameterized as engineering design constraint rules, including basic constraints and adjustment constraints;Again, the constraint rule is encoded as a multi-role collaborative prompt word, input into the generative AI model, sequentially output by psychological semantics, engineering structure, kansei engineering role, generate air car three-dimensional structure form scheme;Finally, the scheme is displayed on the user terminal, and the compliance judgment result input by the user based on the preset engineering logic rule is obtained, and the constraint is updated according to the correction instruction and iteratively optimized.The application embeds user psychological perception into AI design process, and the generated scheme is superior to conventional design without constraints in the evaluation of perceived safety.

This application discloses a grinding head motor, relating to the field of motor technology. It includes a motor body, an output shaft disposed at one end of the motor body, and a controller. The controller is integrated and installed on the port section of the motor body away from the output shaft, and the controller and motor body are designed as a single integrated housing. The surfaces of the motor body housing, output shaft, and exposed connecting parts are coated with an anti-corrosion coating. In this application, when positive pressure is generated inside the motor, the seal is automatically enhanced to prevent internal humid and hot gas from carrying trace amounts of moisture from leaking out. When the motor stops cooling and generates negative pressure, a unique pressure balance channel with a hydrophobic and breathable membrane becomes the only safe air intake path. It effectively blocks external liquid water, water mist, and dust, allowing only dry air to slowly enter to balance the pressure. This fundamentally eliminates the possibility of internal component corrosion and short circuits caused by the intake of moisture and contaminants due to negative pressure.

This invention relates to the interdisciplinary field of aviation safety and artificial intelligence, specifically to an improved WGAN-GP method for generating adversarial examples from ADS-B time-series data. The method includes the following steps: acquiring historical ADS-B flight data and extracting features from five dimensions—longitude, latitude, altitude, speed, and heading—to construct a time-series dataset; building an adversarial example generation model based on an improved Wasserstein distance generative adversarial network and configuring a multi-objective composite loss function for the generator. This invention constructs an LSTM-based generator and discriminator, designs a composite loss function integrating adversarial loss, concealment loss, and temporal smoothness loss, and trains the model using gradient penalty and asymmetric update strategies. The generated adversarial examples possess both high concealment and physical consistency, effectively circumventing various unsupervised anomaly detection models. The model training is stable, providing technical support for building a safer air traffic control defense mechanism.

The application discloses a metal pipe inner diameter laser measuring device, and relates to the technical field of inner diameter laser measuring, comprising a machine table; the hollow compressor and the air inlet pipe of the cleaning assembly are matched, high-pressure gas can be delivered to the base, sufficient power is provided for impurity cleaning, the convenience and completeness of cleaning are improved, and the functions of stable high-pressure gas supply and laying a foundation for inner wall cleaning are realized; the total weight of impurities estimated by the impurity judging module is compared with a preset threshold value through the air injection control module, the minimum effective air pressure, linear air pressure adjustment or the maximum safe air pressure is dynamically selected; the problem that traditional cleaning methods adopt fixed air pressure to cause insufficient cleaning or energy waste is solved, on-demand air supply and accurate cleaning are realized, the cleaning effect is ensured, damage to the pipe caused by excessively high air pressure and energy waste are avoided, and the cleaning efficiency and equipment operation economy are remarkably improved.