To solve the chaotic synchronization problem of fractional-order permanent magnet synchronous motor (PMSM), the master and the slave system of fractional-order PMSM model are firstly constructed. Then, according to the stability theory of fractional-order system, a chaotic synchronization controller based on hybrid projection synchronization method is designed, which achieves chaotic synchronization of fractional-order PMSM with different orders. Numerical simulation results proved the correctness and effectiveness of the proposed control method. fractional-order PMSM achieves synchronization in 2 s and the error system is stable. The research results have important significance for the synchronous operation of fractional-order PMSM system.

In order to solve the problem that the gradient descent algorithm has a slow convergence speed, uses all the sample points each time to update the gradient, and has a large amount of calculation and high complexity, a novel dispersion compensated FIR (fine length impulse response) filter for coherent optical receiver is proposed. Through building coherent optical communication system baseband multi-rate transmission model in Matlab, using the Mini-Batch Gradient Descent method with 1% of the total samples to update the weights of filter tap h(n) every time calculating the mean square error and updating the filter gradient, the computational complexity can be reduced to 1% of the original. When the learning efficiency is 0.05, the convergence value reaches 0.005 5, which has a smaller mean square error convergence value. When the mean square error converges, the weight of the filter tap is obtained. The filter weights obtained by this method has a better filtering effect and lower bit error in the whole frequency band. After using baseband pulse shaping to limit the effective bandwidth of the baseband signal, the number of filter taps is 197, which is 37.5% less than the number of taps in the full frequency band. Therefore, an optimal filter with fewer taps and better filtering effects can be designed in a narrow frequency band.

Aiming at the very short-term electric load forecasting, a combined model is proposed, which uses EEMD(ensemble empirical mode decomposition )and SE(sample entropy) to preprocess the original data, and then applies SA(simulated annealing) to optimize the deep belief network for forecasting. In order to reduce the time-series data of the predicted value behind the real value caused by the autocorrelation of the data, the original sequence is decomposed by EEMD, the sequence is reconstructed according to the SE of each decomposed sequence, and the reconstructed sequence is predicted separately by SA-DBN model composed of SA optimizing the number of nodes in each hidden layer of DBN, and the predicted results of each sequence are superimposed to obtain the final prediction curve. The experimental results show that compared with other prediction models, this model can eliminate the lag of prediction curve, the predicted error indexes MAPE,MAE and RMSE are reduced to 1.9 592%, 9.423 9 and 11.977 1 respectively, and the prediction accuracy of the model is increased to 96.435%.

In order to improve the detection accuracy of controlled items, a target detection algorithm that combines Receptive Field Block (RFB) network structure and feature fusion is proposed. First, the collected security inspection data are eliminated and filtered, then, the security inspection data are manually labeled and the data are enhanced. Based on the MobileNetV3-SSD algorithm, the RFB network is introduced to improve its network structure to strengthen the network’s feature extraction capabilities, and the feature fusion method is used to improve the model’s small target detection capability. Finally, a security inspection dataset SCCI2020 is constructed to verify the performance of the algorithm, where the dataset contains 91 767 images. Experimental results show that the detection accuracy of this algorithm on the security inspection dataset SCCI2020 is 87.0%, which is 2.7 percentage point higher than the detection accuracy of the MobileNetV3-SSD algorithm. The detection accuracy on the COCO2014 general dataset and COCO2017 general dataset are 21.9% and 23%. Compared with VGG16-SSD and MobileNetV3-SSD algorithm, the detection accuracy is improved to a certain extent.

Knowledge compilation, as a key direction of artificial intelligence, plays an important role in on-line requirement and reasoning. The ordered binary decision diagram (OBDD) is a main target language in knowledge compilation, which has been widely used to compile many real-world satisfiability (SAT) problems. The construction algorithms for OBDDs are deeply studied in the past years, and the purposes are to decrease the size of target OBDD and shorten the compilation time. The construction for OBDD is an important factor affecting the efficiency of compilation. To improve the time efficiency, an improved construction algorithm for OBDDs is proposed. The algorithm compiles the causes in SAT problem into OBDDs and then combines them into a whole OBDD. Different from the traditional algorithm which combines these OBDDs into a target OBDD one by one, this approach first combines some OBDDs into larger OBDDs and then they are merged into a garget OBDD. The experiments on randomly generated SAT instances and configuration problems indicate that the proposed OBDD construction algorithm performs better than the original algorithm.

In order to effectively improve the convergence speed and solution accuracy of grey wolf optimization algorithm, this paper combines the elite reverse learning strategy to increase the diversity of population, changes the convergence factor from linear to nonlinear, redesigns the position update formula to improve the convergence accuracy of the algorithm, and proposes an elite learning gray wolf optimization algorithm with nonlinear parameters. The experimental results of 8 groups of typical test functions show that the convergence speed and accuracy of the algorithm are improved to different degrees. When solving the optimization design problem of IIR digital filter, it shows excellent performance.

This paper studies the daily GARCH model estimation by introducing the intraday high frequency data. Existing results assume that the constant term in GARCH equation is given, which restricts the extensive applications of the method. In this paper, based on the framework of GARCH(1, 1) model, two estimators for all model parameters, together with the according asymptotic properties, are discussed. A criterion is given to choose the optimal volatility proxy. Simulation studies show that the estimators have smaller asymptotic standard deviation and the empirical study gives a specific application.

For a three-degree-of-freedom vibro-impact system, the dynamic behavior near the grazing periodic orbit is discussed by using discontinuous mapping method. The existence conditions of saddle-node bifurcation and period-doubling bifurcation in 1/n impact periodic motion are deduced theoretically. It is concluded that co-dimension-two bifurcations occur when saddle-node bifurcation, period-doubling bifurcation and grazing bifurcation are detected simultaneously and the numerical simulation results are consistent with the theoretical results above. The bifurcation and chaotic motions of the system near co-dimension-two bifurcation points are studied by combining Lyapunov exponent and the local bifurcation diagram and the periodic motion and chaotic motion of the system appear in turn within a certain range of parameters. Then, using pulse control of chaos, the chaotic motions of the system can be suppressed to the stable periodic orbit. The effectiveness of the control method is verified by comparing the control diagram.

This paper studies the question of decompositions of λ-fold complete bipartite 3-uniform hypergraphs λK⁽³⁾_n,n into hypergraph triangular bipyramid. This kind of decomposition is denoted by S_λ(3,TB,n,n). Firstly, it is proved that the necessary condition of its existence is 6|λn²(n-1),2|λn and n≥3. Then, it is proved that the necessary conditions are also sufficient except for two possible exceptions of n=10,14.

Using the images with spatial resolution better than 10 m as the data source, the information of the continental coastline of Guangxi in six periods of 1960, 1973, 1989, 2001, 2008, and 2018, are extracted, and the temporal and spatial changes of the coastline of Guangxi under the background of human activities characteristics and drivers are comprehensively analyzed. The results show that: 1) The coastline of the Guangxi continent changed drastically from 1960 to 2018, and the total length showed a trend of first decrease and then increase in 1989 as the node, and the overall reduction was 53.45 km. The period of drastic changes in the length of the coastline of Guangxi occurred from 1960 to 1973. 2) During the study period, the fractal dimension of the coastline in Guangxi showed a downward trend and then an upward trend in 1989 as the node, which was consistent with the trend of the length of the coastline. The complexity of the coastline in each period was in Qinzhou City> Fangchenggang City> Beihai City. 3) Human activity has been the main reason for the significant changes in the coastline of the Guangxi mainland in the past 58 years, including salt fields and culture ponds, seawall construction, sea reclamation (harbors, industry, towns, construction, etc.), road and bridge construction, etc.

The nutritional composition of food plays an important role in primate selection. In order to study the nutritional composition of the food for white-headed langur and its influence on food selection, 14 main food and 5 non-main food of white-headed langur were collected from October 2010 to March 2011 in Fusui Nature Reserve, Guangxi, China. The contents of water, crude protein, crude fat, crude fiber and mineral elements such as Ca, Fe, Zn and Mn in each independent sample were determined by atmospheric pressure and constant temperature drying method. CP/CF was calculated, and the nutrient content of each independent sample was tested by Mann-Whitney U test. The result showed: 1) There is no significant difference between the composition of water, crude protein, crude fat, crude fibre and mineral elements such as Ca, Fe, Zn, Mn, and CP/CF in the main food for white-headed langur and that of the non-main food. 2) There was no significant correlation between the nutrient composition, mineral element content and the ratio of crude protein and crude fiber of the main food and that of the non-main food for white-headed langur when choosing food. Therefore, the results showed that the food selection of Fusui white-headed langur was not affected by water, crude protein, crude fat, crude fiber, mineral elements such as Ca, Fe, Zn, Mn and the ratio of crude protein and crude fiber.

Through a comprehensive review of the literatures, it was found that 328 species and subspecies belonged to 8 subfamilies, 77 genera had been recorded from Guangxi so far, with 26 species endemic to Guangxi. By comparing the component of ant species of Guangxi with that of Yunnan Province with a similar area, it is found that the number of species and endemic species of Guangxi are only 63.39% and 12.20% of the latter, respectively, indicating that there are still many species to be discovered and described in Guangxi. In this paper, a catalogue of Formicidae of Guangxi is provided alphabetically based on the hierarchy of subfamily, genus and species to present a reference for scientists.

In order to explore the role of modified calcium peroxide (CaO₂) in the process of organic carbon mineralization in gleyed paddy soil, CaO₂ was selected as raw material and coated with ethyl cellulose dissolved in ethanol to prepare granulated calcium peroxide (GAO), set GAO and calcium oxide (CAO, containing 95.0% of CaO mass fraction), calcium peroxide (PAO, containing 75% of CaO₂ mass fraction) 3 substances each with 3 different oxygen content concentrations (low concentration 0.011 g/kg, medium concentration 0.111 g/kg and high concentration 1.111 g/kg). At the same time, no substance was added as a blank control (CK), and the method of indoor cultivation was used to study and analyze its effect on the organic carbon mineralization of cultivable paddy soil, the influence of dissolved organic carbon (DOC) and microbial biomass carbon (MBC). The results showed that under the condition of high oxygen concentration, after the incubation period, PAO and GAO inhibited the CO₂ emission from the cultivating paddy soil compared with CK and CAO. The cumulative CO₂ emission from the soil treated with PAO were reduced by 50.86% and 46.33% compared with CK and CAO treatments, respectively, the cumulative CO₂ emissions of soil treated with GAO were reduced by 22.49% and 15.35% compared with CK and CAO treatments, respectively. Compared with CK, the addition of CAO, PAO, and GAO promoted CH₄ emissions from cultivable paddy soil. Its cumulative CH₄ emissions are 2.07 times, 1.22 times, and 1.10 times that of CK treatment, respectively. Compared with CK treatment, the addition of medium and high concentrations of oxidants significantly increased the content of MBC and DOC in cultivable paddy soil, and the addition of GAO delayed the peak period of soil MBC content, indicating that GAO is due to its special coating effectively reduces the oxygen release rate of the oxidant, and can still effectively release oxygen in the later stage of the culture. This study shows that the improved oxidant GAO has a good application potential in improving the microbial activity of cultivable paddy soil and promoting the turnover of organic carbon.

In order to study the biomass model construction and distribution characteristics of different layers and organs (by leaf, branch, trunk and above-ground) of Pinus massoniana plantations in the near natural recovery process. Young stand, middle-aged stand, mature stand and over mature stand of P. massoniana plantations in Zhenlong forest farm of Hengxian county in southern subtropical area of Guangxi were selected as the research subjects. The results showed that: 1) The trunk and above-ground biomass estimation results were better, followed by leaves and branches. Overall, Model Ⅱ (W=a×D^b×H^c) had higher goodness of fit and smaller standard error, and the accuracy of model checking was higher, which could be used as the optimal model for biomass model fitting of different organs in shrub layer at different restoration stages; 2) With the near natural recovery, the order of the branch and trunk biomass of the woody community of P. massoniana plantations and the above-ground biomass of the community was overripe forest > mature forest > middle-aged forest > young forest, and the order of leaf biomass was middle-aged forest > overripe forest > mature forest > young forest. The biomass of each organ in the same community had significant difference among different ages of the same community (P < 0.05). The leaf biomass of woody communities of different forest ages were (2.65±0.01)~(7.77±0.35) t/hm², the branch biomass were (4.24±0.00)~(24.49±0.11) t/hm², the trunk biomass were (24.32±0.01)~(236.51±1.22) t/hm², and the above-ground biomass were (42.73±0.67)~(268.03±1.25) t/hm², in which the trunk was dominant. Within each forest age in P. massoniana plantations, the order of the above-ground biomass of each layer was arbor layer > shrub layer > herb layer, and the arbor layer was the main body of the above-ground biomass of the P. massoniana populations; 3) With the progress of near natural restoration, the order of above-ground biomass of P. massoniana populations were as follows: overripe forest > mature forest > middle-aged forest > young forest, and the proportion in the community increased first and then decreased. The proportion of leaf, branch and trunk biomass of P. massoniana populations in different restoration stages showed a downward trend. The results provide a scientific basis for the dynamic change researches of community structure of P. massoniana plantations in the near natural recovery process and are helpful to improve the stand productivity and carbon sink capacity.

Based on the theory of space syntax, a line segment model is established for syntactic calculation. Then the POI data are used to check the model. The accessibility of park green space in Nantai Island is comprehensively evaluated from the four dimensions: global accessibility, local accessibility, perceived accessibility and psychological accessibility. Finally, the optimization strategies of park green space accessibility are put forward from three aspects: road improvement, reasonable greening and quality improvement. The results show that: 1) the global accessibility, local accessibility and psychological accessibility of Nantai Island park green space are average, while the overall intelligibility of perceived accessibility is lower than 0.5, and the perceived accessibility is poor. 2) The layout of the park green space in Nantai Island is unreasonable and the regional development is not coordinated. The layout of the park green space has certain spatial “dislocation” with the overall road, local accessibility and regional population density. 3) There are differences between the scale of park green space and the psychological accessibility, so the scale and quality of park green space should be paid attention to in later construction. 4) After optimization, the global and local accessibility of Nantai Island are increased by 2 times, and the perception and psychological accessibility are also improved.