Search

In the electronic nose (E-nose) systems, gas type recognition and accurate concentration prediction are some of the most challenging issues. This study introduced an innovative pattern recognition method of time–frequency attention convolutional neural network (TFA-CNN). A time–frequency attention block was designed in the network, aiming to excavate and effectively integrate the temporal and frequency domain information in the E-nose signals to enhance the performance of gas classification and concentration prediction tasks. Additionally, a novel data augmentation strategy was developed, manipulating the feature channels and time dimensions to reduce the interference of sensor drift and redundant information, thereby enhancing the model's robustness and adaptability. Utilizing two types of metal-oxide-semiconductor gas sensors, this research conducted qualitative and quantitative analysis on five target gases. The evaluation results showed that the classification accuracy could reach 100%, and the coefficient of the determination (R2) score of the regression task was up to 0.99. The Pearson correlation coefficient (r) was 0.99, and the mean absolute error (MAE) was 1.54 ppm. The experimental test results were almost consistent with the system predictions, and the MAE was 1.39 ppm. This study provides a method of network learning that combines time–frequency domain information, exhibiting high performance in gas classification and concentration prediction within the E-nose system. [ABSTRACT FROM AUTHOR]

In order to narrow the difference of cigarette weight between the front and rear tobacco rod passages, the secondary stem separator in ZJ116B cigarette maker was modified as following: 1) Designing two independent mechanisms to control the stem separation of the front and rear passages independently. 2) Changing the straight-through separation passage into a Z-shaped one. The test results showed that: 1) The average SD of individual cigarette weight from the front passage decreased from 14.40 to 13.16 mg and that from the rear passage reduced from 16.08 to 13.44 mg. 2) The difference of average SD lowered from 1.68 to 0. 28 mg, and the fluctuation of individual cigarette weight reduced obviously. 3) The content of fiber in stems decreased from 18.4% to 9.0%, it meaned better stem separation and lower tobacco consumption. [ABSTRACT FROM AUTHOR]

In order to conduct the online monitoring and fault diagnosis of the countershaft bearing of cutoff blade carrier in cigarette maker, an intelligent fault diagnosis method was established with SVM on the basis of frequency domain indexes after analyzing the influences of statistical indexes in time domain and frequency domain on SVM classification results. Testing was conducted on the FAG 2206-2RS-TVH bearings with prefabricated faults, the vibration data collected from the bearings with faults of different types and degrees were subjected to preprocessing, characteristic extraction, SVM intelligent training and fault diagnosis testing. The results showed that: 1) At different rotational speeds and fault degrees, the SVM classification based on frequency domain statistic indexes presented higher accuracies of 92% and over, it was able to classify the status of bearings within a specified rotational speed range at different fault degrees. 2) The accuracy based on time domain statistic indexes was 77.6% compared to that based on frequency domain statistic indexes of 95.6%, which suggested that the later was more accurate. This method provides a support for the intelligent fault diagnosis of bearing in cigarette maker. [ABSTRACT FROM AUTHOR]

Deformation behavior and texture evolution of an extruded Mg 3Li sheet with a weak transverse direction (TD)-split texture during tension along different directions were systematically examined. Prismatic 〈a〉 slip was the dominant deformation mode along the extrusion direction (ED), while basal 〈a〉 slip and {10−12} twinning played key roles at the initial stage of deformation and prismatic 〈a〉 slip was also activated at the middle stage of deformation along the TD. The activation of prismatic 〈a〉 slip contributed to the enhanced 〈10−10〉 // ED component along the ED. However, the {10−12} twinning and basal 〈a〉 slip led to the formation of a new ED-texture component and a gradual rotation of c-axes towards the normal direction (ND) along the TD, respectively. Additionally, the cooperative activation of the {10−12} twinning and prismatic 〈a〉 slip were responsible for the decreased 〈10−10〉 // ED component along the TD. Unlabelled Image • Extruded Mg 3Li sheet exhibited a weak TD-split texture. • The room-temperature deformation modes of Mg 3Li sheet were investigated. • Prismatic 〈a〉 slip played a dominant role during tension along the ED. • Li additions improved the activation of prismatic 〈a〉 slip, under a lower strain. [ABSTRACT FROM AUTHOR]