In order to understand the research status and hotspot of aerobic granular sludge (AGS) at home and abroad, bibliometrics was used to review and summarize the global literatures on AGS published between 2008 and 2021 based on the China National Knowledge Network (CNKI) database and the Web of Science core database. At the same time, VosViewer visualization tool was used to conduct keyword co-occurrence cluster analysis, and various knowledge graphs were drawn to show the evolution of knowledge in AGS research field at home and abroad. Key words analysis showed that the application of AGS in the treatment of different types of wastewater, the utilization of AGS multiple biological functions (nitrogen and phosphorus removal), the formation mechanism and conditions of AGS, enhanced granulation and continuous stability cultivation were the main research hotspots of AGS. The research progress of AGS in China has been increasing rapidly, mainly focusing on the formation mechanism of AGS. Overseas research focuses on the types of AGS start-up reactor and disintegration and recovery. The rapid granulation and stable culture of AGS, the synergisation of various functional microorganisms, and the continued expansion of the efficient utilization of AGS in different fields may become the research hotspots in the future.

In recent years, the proportion of large and medium-sized trucks in the traffic flow has gradually increased, and the impact on urban traffic is increasing day by day. Accurate and timely prediction of the arrival of large and medium-sized trucks is of great significance to accurate urban traffic control. To solve this problem, a multi-scale prediction method for the arrival volume of large and medium-sized trucks on expressway based on the system correlation is proposed: the spatio-temporal correlation of the flow of large and medium-sized trucks at the entrance and exit of expressway toll stations is analyzed, and a neural network model is constructed to learn the spatial weight and time weight. In each step of time, the input is fused with spatial weight and time weight, and the offset term is set to obtain the prediction result of the time step after correction.The final prediction result is obtained by summing the prediction results of each step of time. The experimental results show that the prediction accuracy of the method is 90.92%, 92.48% and 94.33% respectively at the time scales of 15, 30 and 60 mins, which is better than other models, and the practicability and effectiveness are guaranteed.

Improving the prediction accuracy of remaining driving range can alleviate the “driving anxiety” of drivers, help vehicle manufacturers develop fine battery management system and improve the acceptance of pure electric vehicles. Based on the improved symbolic regression algorithm, a new data feature field closely related to the label field is automatically generated to expand the data dimension. Then the dimension expanded data is transmitted to the xgboost model optimized by super parameters to predict the remaining driving range. Compared with the original data using only classical feature fields, the maximum relative absolute error of the dimension expanded data in the prediction accuracy decreases by 4.9%, and the average absolute error and root mean square error decrease by more than 20%. With the increase of time, the error of prediction using the dimension expanded data decreases faster. The results show that the proposed method can optimize the quality of the data set, improve the accuracy of the prediction results and reduce the error, which provides a new idea for the prediction of the remaining driving range of pure electric vehicles.

Aiming at the low accuracy of vehicle detection in unmanned system environment perception, a three-dimensional vehicle detection algorithm based on multi-modal feature fusion is proposed. Through the joint calibration of millimeter wave radar and camera, the coordinate relationship between the two sensors is matched and the sampling error is reduced. Statistical filtering is used to eliminate the redundant points of millimeter wave radar data and reduce the interference of outliers. The multi-modal feature fusion module is constructed, and the point cloud and image information are fused by pixel average. Adding the feature pyramid to extract the fused high-level feature information to improve the detection accuracy in complex road scenes, a feature fusion region recommendation structure is established, and the region recommendation is generated according to the advanced feature information. After removing the redundant detection frame, the vehicle detection results are output through the vertex matching of the detection frame. The experimental results on KITTI data set show that the proposed method can realize vehicle detection quickly and accurately. The average detection time is 0.14 s and the average detection accuracy is 84.71%. The algorithm has important theoretical and practical value, and can provide a powerful means for vehicle detection in unmanned system.

To solve the problem that it's difficult for the large amount of network parameters and calculations for existing multi-object tracking algorithm to meet the real-time requirements of mobile devices, a road vehicle multi-object tracking algorithm is proposed by improving the JDE tracking algorithm. Firstly, in order to improve the tracking accuracy of the algorithm and reduce the number of ID switching, the association fusion network is used to solve the competition problem of multi-task learning in the JDE algorithm. Secondly, in order to reduce the complexity of the model and improve the real-time detection speed of the model, the improved EfficientNetV2 is used to rebuild the feature extraction network in YOLOv5. Finally, the improved YOLOv5 detection algorithm is combined with the JDE tracking algorithm to achieve multi-object tracking of road vehicles. The experimental results show that compared with the original JDE tracking algorithm, the proposed method improves MOTA by 0.3percentage point and tracking speed by about 43.2%. It can meet the speed requirements for vehicle tracking in actual autonomous driving scenarios.

In order to improve the accuracy of the Sigma-Delta modulator and reduce its power consumption, an improved second-order single-loop CIFF Sigma-Delta modulator is designed. The additional power consumption caused by the number of noise shaping filters is reduced by using op-amp sharing technology. The idea of floating coefficient iteration is applied to the modeling of the modulator in MATLAB, and the specific values of various parameters that meet the precision requirements are finally determined. Through the introduction of non-ideal factors, the obtained parameters are simulated and verified to meet the minimum performance index, and then the transistor level circuit is designed. The modulator has a signal bandwidth of 8 kHz and a sampling frequency of 4 MHz. The circuit design uses UMC 0.11 μm CMOS process, and the core circuit layout size is 226.8 μm×187.44 μm. The post-simulation results show that when the power supply voltage is 1.2 V, the total power consumption of the modulator is 290 μW. At -40-125 ℃, and the effective bits of each process Angle is more than 15 bits.

The chromatic dispersion in the optical fiber is the direct cause of the inter-symbol interference (ISI) of the transmission signal, and the signal degradation is particularly serious when the transmission environment fluctuates. Traditional adaptive blind equalization algorithms have slow convergence speed and large errors, which are not suitable for high-speed coherent receivers. In order to solve this problem, the momentum factor is introduced to improve the decision-directed least means square (DD-LMS) algorithm to optimize the effect of fiber channel dispersion. Theoretically, the momentum factor is derived and demonstrated on the convergence performance and error function of the blind equalization algorithm. At the same time, the momentum factor of different functions is explored on the optimization performance of the algorithm.A coherent optical transmission system with a single carrier transmission rate of 224 Gib/s is built on the OptiSystem optical simulation software. The results show that, compared with the traditional time domain equalizer (TDE ), adding this adaptive filter at the back end of the system can reduce the average bit error rate of the system by about 2.5 dB. When the optical signal-to-noise ratio is 15 dB and the transmission distance is 500-1 000 km, the error rate of the dispersion compensation module can be stabilized at about 10^-2, and it has strong robustness in the dispersion channel.

Most of the existing algorithms can only enhance the backlight images with specific illumination, but cannot solve the backlight images with various illuminations efficiently. Therefore, an image enhancement algorithm based on convolutional neural network is proposed in this paper, and a new network architecture that integrates decomposition, recovery and adjustment is built at the same time. Using Retinex theory, a decomposition network is designed to decompose the backlight image and its corresponding highlight image into reflectance map and illumination map. The reflectance component of highlight image is used as the denoising reference to repair the dark light defect, and the color saturation module is added to retain the color and other details in the image restoration process. The brightness of the backlight images can be adjusted adaptively according to the user's preference. The enhancement ratio (the ratio between the target light source and the image light source) is set as the adjustment index. When the backlight images are enhanced to the high-light images, the enhancement ratio should be greater than 1. Validated on multiple public datasets (LOL, DICM, NPE), the research shows that this method can effectively enhance the brightness of backlight images, improve image quality, ensure that image details are not lost, and avoid color distortion. It has good effects on backlight images with different illuminations, and the results of subjective and objective evaluation indicators are better than the existing algorithms, which has application value for the development of smart city security and artificial intelligence.

Accurate determination of drug-target interactions is crucial in drug discovery process and repositioning. Traditional methods for DTI prediction are either time-consuming (simulation-based methods) or heavily dependent on domain expertise (similarity-based and feature-based methods). Existing computation-based methods using single data information or sparse data, always suffer from high false positive rates. Although integrating multiple heterogeneous networks has been prevalent for drug target prediction, how to retain as much structural information as possible is still a big challenge. This paper proposes a novel framework NGDTI, which extracts relevant biological properties and association information from the network while maintaining the topology information. Further, the graph neural network is applied to update the extracted feature information. The learned topology-preserving representations of drugs and targets promote DTI prediction. Compared with the state-of-the-art methods, NGDTI increases the AUPR value by nearly 0.01. The results demonstrate that NGDTI is promising for drug development and repositioning.

A kind of prey-predator system with Allee effect and artificially controlled migration of predators is studied. The system has a square root functional response function. Firstly, by qualitative analysis of the model, the boundedness of the solution is proved, and the existence of the equilibrium point is analyzed. Sufficient conditions for the local stability of the equilibrium point of the system are obtained. Then, the existence of the Hopf-bifurcation of the equilibrium point is discussed, and the stability and direction of the equilibrium Hopf-bifurcation are studied by calculating the first Lyapunov coefficient. Finally, the correctness of the conclusion is verified by numerical simulation. The results indicate that the Allee effect and artificially controlled migration rate are important for the survival and extinction of prey and predator populations.

In this paper the nonlinear p-Laplace equation Δ_pu+au^p-1lnu+λu^p-1=0 is studied on complete manifolds with some suitable curvature condition, where a, λ and p>1 are some given constants. Differential inequalities for the p-Laplace equation on compact manifolds with Ricci curvature bounded from below are established, based on the evolution of the geometric quantity along the p-Laplace equation. Similar inequalities can also be established on a noncompact manifold whose sectional curvature is bounded from below, based on the skills of cut off function and the Hessian comparison theorem. As an application, Liouville theorems are obtained.

The large-time behavior of solutions to the Cauchy problem of a 3D hyperbolic-parabolic system modeling chemotaxis is investigated. The optimal time decay rates of the higher-order spatial derivatives of the solutions are obtained. Compared with previous results, the main innovation of this paper is to give the highest order spatial derivative of the solutions which is the same as that of the heat equation. The proof is mainly based on low-frequency and high-frequency decomposition and delicate energy estimates.

In order to study the influence of growing domain on Turing instability of the parasite-host model, this paper takes the parasite-host infectious disease model as the main body and analyzes its Turing instability conditions through linearization and spectral analysis. Numerical simulations are used to verify the theoretical results. The results show that the increase of the diffusion coefficient is benefitial for formation of Turing pattern, but regional growth has a destructive effect on Turing pattern formation.

Phosphomolybdic acid particles have a resonance Rayleigh scattering (RRS) effect, which produces a RRS peak at 450 nm. In the HCOOH-HCOONa buffer solution at pH=3.1, gold nanoparticles (AuNPs) can catalyze the reaction of phosphomolybdic acid-formic acid to produce phosphomolybdenum blue, making the RRS intensity of phosphomolybdic acid linearly decrease at 450 nm. Hg²⁺ can undergo a galvanic replacement reaction with AuNPs, thereby inhibiting the catalytic effect of AuNPs. In the range of 2.5×10^-4-3.5 μmol/L, as the concentration of Hg²⁺ increases, the catalytic effect of AuNPs gradually weakens, the color of the reaction solution gradually changed from blue to colorless, and the resonance Rayleigh scattering peak of the system at 450 nm increases linearly. The regression equation is ΔI=0.32C+46.1, and the detection limit is 0.18 nmol/L. This method is used for the detection of Hg²⁺ in wastewater, and the results are satisfactory.

In order to efficiently utilize the groundwater resources in the Yong'an-Dahu Basin, Fujian, the methods of field investigation, data analysis, numerical simulation and hydrological analysis were used to analyze the basin's water storage capacity and possible environmental geological problems under exploitation conditions. Considering the constraints of reservoir regulation and coordinated control of environmental and geological issues, a suitable exploitation area was delineated with an area of about 37.38 km². The available storage capacity of this area was calculated to be 1.362×10⁷ m³, and the adjustable storage capacity was 8.03×10⁶ m³. On this basis, numerical simulation and hydrological analysis methods were used to calculate that the allowable exploitation quantity were 40 619.45 m³/d and 36 957.48 m³/d, respectively. Four ways (zoned mining, dual water supply, rainwater resource utilization, and water saving)were proposed to improve the utilization rate of groundwater resources realize the balance of comprehensive allocation of water resources, and control environmental geological disasters in Yong'an-Dahu Basin. The research will provide a basis for the efficient use of local water resources, and a reference for water resources exploitation in similar areas.

In this paper, a novel inorganic crystal complex copper phosphate was used as a carrier to immobilize the lytic polysaccharide monooxygenase (cx-LPMO-B), the optimal conditions for immobilization of cx-LPMO-B was investigated, and the enzymatic properties were compared with the free and immobilization of cx-LPMO-B. The results showed that, copper sulfate was added to the pH=7.4 and 0.01 mol/L phosphate buffer solution containing the cx-LPMO-B to form immobilized cx-LPMO-B-copper phosphate nanoflowers, namely cx-LPMO-B-NF. The cx-LPMO-B-NF reached the highest immobilized yield at 25 ℃ for 14 h with the 0.3 g/L cx-LPMO-B. The immobilized cx-LPMO-B-NF retained a residual activity of approximately 60% after 6 cycles reaction. The nano structure of the immobilized cx-LPMO-B-NF was characterized with scanning electron microscope (SEM) and transmission electron microscope (TEM). It was performed that the immobilized cx-LPMO-B-NF was uniformly dispersed and single blooming flower shape. The optimal pH and temperature for immobilized cx-LPMO-B-NF were pH=4.0 and 50 ℃, respectively. The nano flower structure increased surface area of immobilized carrier to enhance cx-LPMO-B immobilization. Moreover, the reusability of immobilized cx-LPMO-B-NF was significantly enhanced.Therefore, the immobilized cx-LPMO-B-NF can promote advantages for cx-LPMO-B industrial application.

In Youjiang River basin, 71 shallow soil samples were collected according to the network density of 10 km×3 km, and the contents of Cd and Pb were analyzed and tested. Cumulative frequency method and fractal analysis method were used to analyze and calculate their environmental geochemical baseline values, so as to assess the Cd and Pb pollution of shallow soil in this area. The results showed that, there was not much difference between the environmental geochemical baseline obtained by the two methods, among which Cd had a difference of 10.3% in upper limit of baseline, 9.4% in baseline value, Pb had a difference of 3.6% in upper limit of baseline, 6.7% in baseline value, indicating that the calculation results of the two methods were reasonable. In this study, the arithmetic average of the two methods was taken as the environmental geochemical baseline values, that was, Pb was 25.44 mg/kg, and Cd was 0.304 mg/kg. The single factor pollution assessment results of Cd showed that the light pollution, intermediate pollution and heavy pollution accounted for 67.61%, 23.94% and 5.63%, respectively, while the single factor pollution assessment of Pb showed that the light pollution, intermediate pollution and heavy pollution accounted for 53.52%, 25.35% and 11.27%, respectively. Nemerow comprehensive pollution assessment showed that light pollution, intermediate polltuion and heavy pollution accounted for 52.11%, 33.88% and 8.45%, respectively. This showed that the pollution in this area was dominated by light pollution, followed by moderate pollution, only a small number of samples were severely polluted. Moderate and severe pollution areas were mainly distributed in the southeast and northwest of the study area, and there was a moderate pollution point in the central and western regions. The geoaccumulation index of Cd and Pb of most samples was less than 1, indicating that the accumulation effect of Cd and Pb in this area was not obvious. However, the geoaccumulation index of Cd and Pb in the southeastern part of the study area and the north-central part of the study area reached the level of moderate pollution. All these indicated that the development of bauxite resources in the east area and the production activities of the thermal power plant in the middle area had a great impact on the pollution of Cd and Pb in the area.

In view of the high concentration of nitrate nitrogen (higher than 600 mg/L) contained in photovoltaic wastewater, continuous flow biofilm method was used for denitrification of wastewater, and continuous flow activated sludge method was set as a comparison. Firstly, the operating conditions of continuous flow denitrification were optimized. The effects of different carbon nitrogen ratios (3∶1, 3.5∶1 and 4∶1) on denitrification were studied by adjusting the carbon source, and the hydraulic retention time (8 h, 10 h and 12 h) of continuous flow was changed to find a sufficient effective reaction time. The ratio of carbon to nitrogen and hydraulic retention time were kept, and the influent nitrate concentration was gradually increased under the optimal operating conditions. The results showed that the best operating conditions of continuous flow denitrification were C/N ratio of 3.5 and HRT of 12 hours. Under this operating condition, the total nitrogen removal rate of wastewater with influent nitrate concentration of 600 mg/L, 1 200 mg/L and 1 500 mg/L reached 96.69%, 92.95% and 90.93%, respectively. The experiment showed that compared with continuous flow activated sludge process, continuous flow biofilm process had higher total nitrogen removal rate and lower nitrite nitrogen accumulation rate for the treatment of wastewater with high nitrate concentration, which can provide not only reference for the optimization of high concentration wastewater treatment, but also basic data for the operation of photovoltaic wastewater with high concentration of nitrate nitrogen in the future.

CD4⁺CD25⁺ cells are a group of T cells with immunosuppressive activity, which are also known as regulatory T cells (Tregs). Tregs suppress T-cell activation and promote tumorigenesis and progression in tumorigenesis, whereas the immunomodulatory effects produced by Tregs of normal physiological origin upon return transfusion are unknown. To investigate the immunomodulatory effect of Tregs under normal physiological conditions, the effects of tumor growth and T-cell production and lung lesions were compared in normal and tumor-burdened mice of CD4⁺CD25⁺ cell transplantation by mouse lung cancer model and cell transplantation. The results showed that mice transplanted with normal CD4⁺CD25⁺ cells had longer survival, slower tumor growth, increased T-cell and their subpopulations in the peripheral blood and spleen, and no significant lesions in the lungs compared with the control group, while mice transplanted with tumor-bearing CD4⁺CD25⁺ cells had shorter survival, faster tumor growth, significantly decreased T-cell and their subpopulations in the peripheral blood and spleen, and more severe diffuse focal lesions in the lungs. These results indicated that normal-derived CD4⁺CD25⁺ cells had antitumor effects in the mouse lung cancer model.

To observe the expression of Sdr9c7 gene in the lung tissue of mice with acute lung injury (ALI) and explore its mechanism. 9 female C57BL/6 mice were randomly divided into blank group, 3rd-day group and 7th-day group, with 3 mice in each group. Acute lung injury model was induced by lipopolysaccharide (LPS). The experiment was terminated on the 3rd day or on the 7th day of LPS induction respectively. PBS solution was intranasal instilled in the blank group as control group. A549 cells were transfected with SDR9C7-siRNA and the effect of knockdown of Sdr9c7 gene expression was verified. A549 cells were divided into control group, induction group, knockdown group and combination group. Hematoxylin eosin (HE) staining was used to observe the morphology of lung tissue and evaluate the pathological score. The relative expression levels of IL-1β, TNF-α, IL-6 and Sdr9c7 genes in lung tissue and cells were detected by real-time quantitative PCR (RT-qPCR). Results showed that compared with the blank group, the pulmonary edema score, inflammation score, total pathological score and the relative expression of Sdr9c7, IL-1β, TNF-α and IL-6 genes in the 3rd and 7th day groups were increased (all P<0.05); compared with the 3rd day group, the inflammation score, total pathological score and the relative expression of Sdr9c7, IL-1β and TNF-α genes in the 7th day group were decreased (all P<0.05). The relative expression of Sdr9c7 gene was positively correlated with pathological score (r=0.964, P<0.01). In the experiment of knockdown of Sdr9c7 gene, the effect of SDR9C7-siRNA3 was more significant. In the experiment of SDR9C7-siRNA3 knockdown in LPS induced A549 cells, the relative expression levels of IL-1β, TNF-α, IL-6 and Sdr9c7 were detected by RT-qPCR. Compared with the control group, the expression levels of IL-1β, TNF-α, IL-6 and Sdr9c7 increased in the induction group (all P<0.05), decreased in the knockdown group and combination group (all P<0.05), and the expression levels of IL-1β, TNF-α, IL-6 and Sdr9c7 in the knockdown group were lower than those in the combination group (all P<0.05). It suggests that the abnormal expression of Sdr9c7 gene in lung tissue of acute lung injury model mice may be related to LPS induced inflammatory response.

White-headed langur (Trachypithecus leucocephalus) and François' langur (Trachypithecus francoisi) are sibling species sharing similarities in habitat type, body size and social structure. Comparion of their allogrooming behavior are essential for understanding the similarities of their behavioral characteristics and social functions. Data regarding allogrooming time budget to various body region and grooming posture of the two langurs were collected in Fusui, China from February 2016 to January 2017 using focal animal sampling and continuous recording. Results showed that in each 1% skin surface, the allogrooming percentage of easy to reach area, inaccessible area and difficult to reach area of white-headed langur were 0.67%±0.21%, 1.01%±0.47% and 1.44%±0.46%, and that of François' langur were 0.44%±0.14%, 1.61%±0.58% and 1.21%±0.54%. The allogrooming percentage of different surface area of the two species supported the hygienic functional hypothesis. In each 1% skin surface, the allogrooming percentage of anogenital area of white-headed langur and François' langur were 7.12%±2.26% and 10.61%±7.68%, showing high selectivity. Percentages of allogrooming time received of groveling of white-headed langur and François' langur in anogenital area were 9.47%±3.12% and 10.61%±7.68%, which supported the distensive functional hypothesis. The grooming dyads of white-headed langur and François' langur adopted postures by avoiding facing each other, which accounted for 47.86%±16.23% in white-headed langur and 63.76%±17.26% for François' langur, indicating that the distensive functional hypothesis was not supported by the allogrooming percentage of white-headed langur and supported by the allogrooming percentage of François' langur.

In order to reveal ant diversity patterns in Sichuan Province, ant communities from middle Daliangshan were surveyed by plot sampling method. The results showed that a total of 115 species belonging to 44 genera and 6 subfamilies were collected from 6 vertical zones in middle Daliangshan. The dominant ant species was not found, and there were 5 common species, 21 relative-common species, 44 relative-rare species and 45 rare species in the ant community. Main indicators of ant communities from different sample-plots were as follow: species number 8-29 (average 17), individual density 25.6-482.4 heads/m² (average 166.8 heads/m²), diversity index 0.954 8-2.499 0 (average 1.752 6), evenness index 0.386 5-0.826 3 (average 0.630 8), dominant index 0.117 5-0.579 9 (average 0.261 2). Similarity coefficients between ant communities were 0.258 1-0.629 0 (average 0.403 5), which showed moderate dissimilar to moderate similar level. It is concluded that the decisive factors of ant diversity were latitude, altitude and temperature. When the altitude raises, the diversity indexes, individual densities and species numbers generally decrease, but the main indicators of ant community commonly show a multi-domain effect phenomenon, which is mainly affected by climate factors and human disturbance factors. Although the vertical zones in middle Daliangshan commonly have been disturbed by human being to different degree, ant communities from different vertical zones are obviously different. The ant fauna and species diversity rank a relatively higher level in Sichuan Province and therefore have higher conservation value.

In order to explore the effects of inoculation of Glomus etunicatum (abbreviated as A, the similar as below) and Sinorhizobium fredii (R) on plant growth of soybean (Glycine max)in kumquat (Fortunella margarita (Lour.) Swingle (F)/soybean (S) intercropping system, a series of field experiments were conducted to analyze the photosynthetic and respiratory metabolism and plant growth of S inoculated with several treatments. The results showed that: 1)The mycorrhizal infection intensity η_m and vesicle intensity η_V in root system of S of all intercropping treatments were lower than those of monocropping treatments respectively, there were significant differences in η_m and no significant differences in η_V among the corresponding treatments (except that η_V increased significantly in intercropping double inoculation treatment compared with monoculture double inoculation treatment). Compared with the corresponding S or F+S treatment, those of all single inoculation A treatments or single inoculation R treatments showed significant or unsignificant promoting effects respectively, while the increase effect of η_m or η_V of double inoculation treatment A+R was the largest and the differences was significant respectively. 2)Compared with the corresponding monocropping treatments, the number of nodule and dry weight of S were decreased in intercropping system. Compared with the corresponding single inoculation A or R treatments, the promotion with A+R treatment were the most significant respectively. 3)The chlorophyll content and net photosynthetic rate of S were decreased with F+S treatment, while increased after inoculated with single A or R treatment. Those of the promotion effect of both A and R treatment was better than single A or R treatment. 4)The activity of succinate dehydrogenase of S in intercropping system were higher than that of monoculture system, and the promoting effect sequence of inoculation treatment was A+R>A>R. 5)The change trend of H⁺-ATPase activity in mitochondrial membrane of S leaves was similar to that of net photosynthetic rate. 6)The soybean biomass and yield of the intercropping inoculation treatment were higher than the monocropping inoculation treatment, and the inoculation treatments showed a decreasing trend of A+R>A>R. Therefore, treatment inoculated with both A and R could promote significantly the growth of S in intercropping system, the promotion effect was better than that of the single inoculation treatment.

A single factor experiment was adopted to investigate the effects of dietary branched-chain Amino Acid ratio flora of 28-63 days old Youxian duck. 504 Youxian ducks of 28 day-old and healthy and similar in weight were randomly allocated to 6 groups with 6 replicates each group and 14 in each replicate. Six diets with leucine, valine and isoleucine ratios of 1∶0.4∶0.3 (group MA), 1∶0.5∶0.4 (group MB), 1∶0.6∶0.5 (group MC), 1∶0.7∶0.6(group MD), 1∶0.8∶0.7(group ME) and 1∶0.9∶0.8 (group MF) were avilable. The experiment lasted for 35 days. At the end of the feeding experiment, one duck was selected from each replicate of each treatment to collect cecal contents, and the cecal microbial structure and flora diversity were detected and analyzed by 16S rDNA sequencing technology. The results show that: 1) There are 344 identical OTUs in the 6 groups, and only one unique OTU existed in group MD. 2) At the phylum level, the core flora are Bacteroidetes, Firmicutes and Proteobacteria. At the genus level, Bacteroides and Desulfovibrio are the dominant genera; 3) The results of comparion among the 6 groups show that, the abundance of Proteobacteria in group MD is significantly higher than that in group ME (P<0.05), and the abundance of Actinomycetes in group MF is significantly higher than that in group MA (P<0.05); 4)There is no significant difference in Ace index and Chao1 index among the 6 groups, however, the Shannon index in group MF is significantly higher than that in group MC (P<0.05), and Simpson index in group MC is significantly higher than that in group MF (P<0.01) and significantly higher than that in group MA (P<0.05). In summary, the core flora of the cecum of 63-day-old Youxian duck is Bacteroides, Firmicutes, and Proteobacteria. The proportion of branched-chain amino acids in diets significantly affects the Proteobacteria and Actinomycetes. When the ratio of mleucine, valine and isoleucine is 1∶0.9∶0.8, the microbial species diversity of the cecum in Youxian duck is the highest.