Search

Text emotion recognition (TER) is an important natural language processing (NLP) task which is widely used in human–computer interaction, public opinion analysis, mental health analysis, and social network analysis. In this paper, a deep learning model based on XLNet with bidirectional recurrent unit and attention mechanism (XLNet-BiGRU-Att) is proposed in order to improve TER performance. XLNet is used to build bidirectional language models which can learn contextual information simultaneously, while the bidirectional gated recurrent unit (BiGRU) helps to extract more effective features which can pay attention to current and previous states using hidden layers and the attention mechanism (Att) provides different weights to enhance the ’attention’ paid to important information, thereby improving the quality of word vectors and the accuracy of sentiment analysis model judgments. The proposed model composed of XLNet, BiGRU, and Att improves performance on the whole TER task. Experiments on the Interactive Emotional Dyadic Motion Capture (IEMOCAP) database and the Chinese Academy of Sciences Institute of Automation (CASIA) dataset were carried out to compare XLNet-BiGRU-Att, XLNet, BERT, and BERT-BiLSTM, and the results show that the model proposed in this paper has superior performance compared to the others.

Speech emotion recognition (SER) technology is significant for human–computer interaction, and this paper studies the features and modeling of SER. Mel-spectrogram is introduced and utilized as the feature of speech, and the theory and extraction process of mel-spectrogram are presented in detail. A deep residual shrinkage network with bi-directional gated recurrent unit (DRSN-BiGRU) is proposed in this paper, which is composed of convolution network, residual shrinkage network, bi-directional recurrent unit, and fully-connected network. Through the self-attention mechanism, DRSN-BiGRU can automatically ignore noisy information and improve the ability to learn effective features. Network optimization, verification experiment is carried out in three emotional datasets (CASIA, IEMOCAP, and MELD), and the accuracy of DRSN-BiGRU are 86.03%, 86.07%, and 70.57%, respectively. The results are also analyzed and compared with DCNN-LSTM, CNN-BiLSTM, and DRN-BiGRU, which verified the superior performance of DRSN-BiGRU.

The poor adhesion of TPU films limits their extensive application in lightweight laminated glass (LLG). A simple and effective method is reported in this paper to obtain modified TPU films by silane coupling agent (SCA) treatment. The polar groups (such as urethane groups, carboxyl groups, etc.) on the surface of TPU films reacted chemically with the reactive groups of SCA. Meanwhile, hydrogen bonds were formed between the silicon hydroxyl groups generated by the hydrolysis of SCA. Therefore, the adhesivity and thermal stability of TPU films were improved. Moreover, the surface of TPU films became rough after modification, hence the interfacial bonding area between TPU film and glass increased, which furthered the bonding effect of TPU film with glass plate. Compared with unmodified TPU films, the tensile shear bond strength (TSBS) of modified TPU films increased by nearly 28%, and the initial decomposition temperature increased from 277 °C to a maximum of 295.3 °C. The impact resistance of LLG was significantly improved due to the improvement of the adhesivity of TPU film.

In this paper, a low-voltage and low-voltage power rail-to-rail operational amplifier is designed based on the research progress of low-voltage and low-voltage power rail-to-rail operational amplifier. Based on the level shift technology, the problem of limited common mode input range caused by constant threshold voltage under low voltage is effectively solved. Based on the SMIC 0.18 [Formula: see text]m complementary metal-oxide semiconductor (CMOS) process, the operational amplifier has a slew rate of 9.3 V/[Formula: see text]s and an establishment time of 0.41 [Formula: see text]s. The power consumption of the operational amplifier is 0.52 mW. The effective area of the layout is about [Formula: see text]m2.

This paper constructs and simulates the interface circuit of a temperature sensor based on SMIC 0.18 [Formula: see text]m CMOS. The simulation results show that when the power supply voltage is 1.8 V, the chopper op-amp gain is 89.44 dB, the low-frequency noise is 71.83 nV/Hz,[Formula: see text] and the temperature coefficient of the core temperature sensitive circuit is 1.7808 mV/[Formula: see text]C. The sampling rate of 10-bit SAR ADC was 10 kS/s, effective bit was 9.0119, SNR was 59.3256 dB, SFDR was 68.7091 dB, and THD was −62.5859 dB. The measurement range of temperature sensor interface circuit is −50[Formula: see text]C[Formula: see text]C, the relative temperature measurement error is ±0.47[Formula: see text]C, the resolution is 0.2[Formula: see text]C/LSB, and the overall average power consumption is 434.9 [Formula: see text]W.

Estimating the degree of polarization (DOP) of target signal light ( p targ ) is a critical step in the underwater polarization differential imaging process. It is challenging to accurately estimate the p targ of objects containing multiple attributes in the imaging range. Inspired by the optical polarization characteristics of objects with different DOP, we find a prior of connected domains (CDs) of objects in polarization orthogonal images. Based on this prior, the CDs of low-DOP (LDOP) and high-DOP (HDOP) objects are extracted from the darkest image ( I min ) and brightest image ( I max ) without additional image acquisition. The p targ of LDOP and HDOP objects in the scene can be estimated through the extracted CDs, which realizes the transition from global optimum estimation in the traditional method to local optimum estimation. According to the estimated p targ , the enhanced results of the entire image marked by p targ can be obtained after fusing the enhanced results of the objects in each CD. Experimental results of simple and complex combination scenes of objects show that our method can recognize CDs of LDOP and HDOP objects. We obtained the enhanced results of the entire image, verified the feasibility and robustness of our method, and found that the CDs of LDOP and HDOP objects could be effectively extracted in up to 34.3 NTU of turbid water. The local optimum improves the application ability of underwater polarization imaging in unpredictable scenarios, while the enhanced result of target labeling has potential application value in underwater target recognition and classification.

A low noise interface ASIC for micro gyroscope with ball-disc rotor is realized in 0.5&micro, m CMOS technology. The interface circuit utilizes a transimpedance pre-amplifier which reduces input noise. The proposed interface achieves 0.003 o/s/Hz1/2 noise density and 0.003 o/s sensitivity with &plusmn, 100 o/s measure range. The functionality of the full circuit, including circuit analysis, noise analysis and measurement results, has been demonstrated.

Carbon materials have the characteristics of stable structure, good mechanical properties, large specific surface area, and numerous reactive sites; however, they are characterized by low volumetric specific capacity, poor cycle stability, and poor electrochemical performance, which restricts their application in energy storage. In this study, three-dimensional graphene oxide (3DGO) as a carbon substrate material was combined with Co-MOF precursors to prepare carbon-based composite materials, and the influence of the microstructure of the composite on its electrochemical performance was investigated. The 3DGO/Co3O4 nanosheet array composite material had a radial size of about 2 µm and a longitudinal size of about 20 nm. The arrays were staggered to form a porous structure, and the pore size was 0.2–2 µm. After 30 cycles, the Rct decreased to 50 Ω, from 140 Ω. The specific discharge capacity of the first cycle reached 800 mAh g−1, and the specific discharge capacity remained at 280 mAh g−1 at a high current density of 1000 mA g−1. The 3D porous structure improved the lithium storage capacity and rate performance of the material. After the cycling experiment, a crosslinked structure appeared on the top of the electrode material nanosheets.

At present, the energy-saving application of power system has become an important means to ensure the sustainable development of social economy. However, there are still many deficiencies and problems in energy-saving dispatching and demand side management of power system, which urgently needs innovation and improvement. Based on this, this paper first analyzes the development status of energy-saving power operation mechanism, then studies the power operation optimization based on computer-based energy-saving dispatching, and finally gives the power operation optimization strategy based on power demand side management.

Power demand side management (hereinafter referred to as DSM) is an important part of smart grid construction in China, which can realize scientific energy saving and power saving. Through DSM, the power grid can dispatch power based on user demand, which can support the sustainable development of national power industry. At the same time, through DSM, the power grid can transmit power intelligently, which is of great significance to the energy conservation and emission reduction of the society. With the construction of smart grid, power DSM has formed an important management system, which can better expand the intelligent and scientific power grid. Through the continuous transformation to energy-saving management system, the traditional load management is no longer suitable for the needs of the society, which requires us to comprehensively review the development of smart grid. Through the DSM system, we can expand the load analysis management means in smart grid, which will better tap the demand side resource potential. Through the effective evaluation system, we can provide more comprehensive, scientific and accurate decision-making basis. Firstly, this paper presents the current load forecasting methods. Then, a model is proposed. Finally, some suggestions are put forward.

Underwater polarization differential imaging requires the estimation of different parameters, and the parameters can be accurately obtained by using optical correlation. However, optical correlation as a criterion function to estimate parameters takes a lot of time. To expedite the parameters’ estimation process, we propose two operations to process underwater polarization images. One operation is to update the analyzer angle range to reduce the number of processed images. The other is image downsampling, which reduces the amount of calculation for the corresponding images. In experiments, we confirmed the feasibility of our method. We have obtained an average of 42 times the calculation speed increase under the conditions of updating the analyzer angle range 3 times and reducing the image scale by 16 times. The results of our method are consistent with those of traditional methods. This established method is conducive to the practical application of underwater polarization differential imaging.

Transition metal oxide materials have attracted the attention of researchers due to their advantages, but their development is hindered by structural instability, a small specific surface area and low electrical conductivity [1,2]. Porous carbon materials are often used in combination with transition metal oxide materials since they can produce a synergistic effect [3]. However, the interfacial properties of electrode materials have an important influence on the energy storage characteristics. In this paper, two types of composites were prepared by using three-dimensional graphene foam (3DGF) and three-dimensional porous carbon (3DPC) as the carbon matrix and Co-MOFs as the precursor. The effect of microstructure and contact interfaces on electrochemical properties was analyzed. Besides, non-equilibrium energy band models were established according to the metal semiconductor contact theory and molecular orbital theory, and the carbon-semiconductor and electrode-electrolyte contact interfaces were studied from the energy storage perspective based on the non-equilibrium carrier concentration distribution during battery cycling. Compared with the precursor conversion type (Pcver) 3DGF/Co3O4, the precursor continuation type (Pctin) 3DPC/Co/CoO had better interface compatibility since the energy band at the semiconductor side of the electrode-electrolyte interface bent more slightly. What’s more, the porous channel structure of the Pctin type 3DPC/Co/CoO electrode material endowed it with better cycle stability and more diversified lithium insertion modes. In summary, the Pctin type composite could improve the cycle stability of the electrode and the lithium intercalation model. This study provided insights into the structure design and interface analysis of electrode materials under the non-equilibrium state.

In order to improve the slew rate of rail-to-rail operational amplifiers, this paper presents a novel slew boosting circuit. Two slew-boosting unit circuits are proposed to provide large compensation current by directly detecting the differential level change at the input terminals of the amplifier. Accordingly, the slewing performance is dynamically boosted. In each slew-boosting unit circuit, tunnel diodes are used to replace the conventional MOS transistors as loads. This circuit is designed in 1.0[Formula: see text][Formula: see text]m TVS process. Measurement results show that the rail-to-rail operational amplifier achieves a positive slew rate of 164[Formula: see text]V/[Formula: see text]s and negative slew rate of 185[Formula: see text]V/[Formula: see text]s.

The design of single loop two-order Delta-Sigma modulator with feed forward structure is presented in this thesis. Oversampled analog-to-digital converters based on second-order delta-sigma modulation are attractive for VLSI implementation because they are especially tolerant of circuit nonidealities and component mismatch. The noise shaping and oversampling are used in this modulator, the oversampling rate (OSR) is 256 and the resolution could be reached 14 bits. The integrally design of Delta-Sigma modulator is argued in deep in the paper, since the behavior model is tested efficiently by Simulink, the time of design has been cut to the bone. As the fully differential structure is used, the system to inhibit the ability of common-mode interference improved and the sampling rate is also raised, the fully differential switching capacity structure is used in the circuit design of modulator. And because of the fully differential structure, the input noise is reduced, the output slewrate of op-amp is greatly improved and the whole circuit is more stable. The 0.5 ?m CMOS process is used in simulation of circuit, the results of simulation are shown that the amplifier gain is 82.3 dB, the Phase Margin is 71.34°, the output slewrate of op-amp is 202 V/?s, the unit GainBandWidth is 102 MHz. the whole circuit of modulator is tested with 10.24 MHz of the clock frequency, the SNDR reached to 93.2 dB, the Dynamic Range (DR) is 100 dB. In summary, this Delta-Sigma modulator could be used in Portable, Audio system and so on.