This project focuses on the classroom interaction of college English and proposes a framework for optimizing college English classroom interaction by integrating big data. Taking the behavioral analysis layer as the entry point, using PSO’s improved K-mean clustering algorithm, we focus on analyzing the specific application of data mining technology in students’ learning behavior. Then we conduct experiments on two classes of students in a university, design classroom behavioral coding to analyze classroom interaction behavior, and explore the application effect of this English classroom interaction optimization pathway. The students were divided into six categories through cluster analysis, with focused learners (22%) and continuous learners (36%) having the highest fidelity scores and the largest proportion, and the analysis of students’ learning behaviors can provide a reference for teachers’ classroom teaching. The composition of the English interaction optimization classroom changes from teacher-led to student-led in the traditional classroom, the teacher-student speech curves intersect each other and both appear four peaks, showing good classroom scope and teacher-student interaction effect, and the path of interaction optimization in the English classroom based on big data is practicable.

Vocal singing is a key art form of many stage singing arts, specifically including acting and singing. The study firstly is to introduce the detection principle of YOLOv5 target detection algorithm, on the basis of which the original YOLOv5 algorithm is improved by reconstructing the backbone network with the use of SENet and GhostNet, then the original YOLOv5 algorithm and the improved YOLOv5 algorithm are tested for comparison, and the test results show that on the target detection dataset Precision, Recall and mAP values reach 85.75%, 72.34% and 78.48% respectively, which are all improved compared with the original algorithm. Secondly, a high-resolution human posture estimation network incorporating multiple attention mechanisms is proposed to further extract multi-scale feature information and global feature information, and validated on publicly available datasets, CDLNet has an AP value of 0.662 and an AR value of 0.731 on the vocal singing posture estimation dataset, comparing with similar methods, the method has an MPJPE in Human3.6M The lowest is 44.6, which is suitable for use in vocal singing posture estimation in vocal singing scenarios. Finally, an action recognition model based on multi-granularity spatio-temporal graph convolutional neural network designed in this paper is used to analyze the singing gesture action recognition for singing action categories, and experiments show that the average recognition rate of MGstgcn can reach 86.5% on the HSiPu2 dataset, which meets the demand of vocal singing gesture action recognition analysis.

With the rapid development of the regional economy, the urbanization process is gradually accelerated, and the ecological safety problems of the urban water body network gradually appear, so this paper is based on linear planning to optimize the ecological landscape water body network. The study first gives a detailed description of the linear planning theory and highlights the gray linear planning model used. Based on the ecological constraints of the landscape pattern quantity optimization research, the “top-down” gray linear planning model from six aspects to build ecological constraints and objective function, through the simplex method to solve, resulting in different scenarios of the total amount of control of the optimization program. Three practicable optimization scenarios are obtained through repeated debugging of the optimization results, and the three scenarios achieve different results in terms of economic and ecological values. In this paper, effective optimization schemes are proposed for different optimization purposes, which on the one hand make the optimization results more realistically reflect the changes of the ecological landscape water body network, and on the other hand provide an optimal model for the management and development of the ecological landscape water body network, and promote the sustainable development of the region.

Distributed energy storage technology can effectively solve the load peak-to-valley difference and voltage quality problems faced by distribution networks. Reasonable and efficient scheduling of distributed energy storage in distribution networks is an important means to play its role. The study proposes a power prediction-based optimized scheduling strategy for distributed energy storage in distribution grids with hierarchical zoning. Firstly, power prediction is carried out using GWO-EEMDBP neural network. Then partition optimization is carried out according to distributed power and load distribution, and the energy storage scheduling strategy is formulated based on the energy storage power prediction interval. Finally, experiments and arithmetic examples are analyzed based on the data related to the distribution system of the IEEE-33 distribution node. The predicted SOC values based on GWO-EEMD-BP neural network are basically consistent with the real SOC values. After applying the energy storage scheduling strategy designed in this paper, the system power loss decreased by 260.86 kW∙h and the load volatility decreased by 67.5%. In addition, this strategy has significant advantages in terms of system operation economic efficiency and voltage quality improvement, and it is capable of scheduling distributed energy storage in the distribution network in a reasonable manner.

Ethnic folk dance in Southwest China is known for its unique regional characteristics and cultural background, and the optimization of its movement choreography strategy is especially crucial for the inheritance and development of this artistic influence. In this study, an optimized graph neural network model is used to choreograph the movements of folk dances in Southwest China. The model is equipped with multi-feature fusion, spatial modeling and temporal modeling modules, which can maximize the recognition performance of the graph neural network model. Based on the model, a framework for automatic generation of folk dance movements is designed, and the model is trained and validated using Laban-16 and Laban-48 dance movement datasets. The experimental results show that the method of this paper is well tested, and the loss value and accuracy convergence algebra of the training set and the test set are basically the same, reaching 0.25 and 96%, respectively. The lower limb motion recognition rate on Laban-16 dataset is improved by 5.21%~15.81% compared with the comparison model. Under the music of different rhythms, a variety of dance movements can be reasonably choreographed to, and the feasibility score of the model by experts is between 85 and 95, indicating that the model in this paper has practical value.

This paper constructs a three-dimensional model of energy storage power station through threedimensional visualization technology, and builds a virtual simulation environment of energy storage power station by inputting realistic environmental parameters. Four different energy storage technology routes, namely lithium-electronic battery energy storage, lead-acid battery energy storage, pumping energy storage and air compression energy storage, are selected, and the energy storage performance of the four technology routes is explored in depth based on the constructed virtual environment. At the same time, the energy storage performance of four different technology routes in the virtual environment of the energy storage power station is compared using the energy storage capacity and energy storage efficiency as the measurement indexes, and the energy storage technology routes suitable for the environment of this paper are highlighted based on the comparison results. In the energy storage simulation, the net energy storage capacities of the four technology routes in the virtual environment of this paper are 728.99MW, 724.18MW, 461.50MW and 393.45MW, respectively. Compared with the other three energy storage technology routes, the lead-acid battery energy storage capacity fluctuation is smaller, and the energy storage capacity is higher, with a higher degree of adaptability to the virtual simulation environment in this paper. At the same time, the average energy storage efficiency of lead-acid battery in four quarters is 99.71%, compared with the next highest efficiency of lithium-electronic battery energy storage efficiency increased by 14.29%, which further indicates that the lead-acid battery energy storage technology route in this paper builds the best performance of the virtual simulation environment of the energy storage power station.

Inheriting national sports culture plays an important role in promoting national culture and national spirit. The digitization technology’s has played a role in promoting the development and dissemination of traditional sports culture. Based on the digitization model of traditional sports and related research materials and audience comments, the article extracts effective data to construct a model of the influence factors of digitization of traditional sports. Trust, cultural confidence, inheritance willingness, perceived usefulness, perceived ease of use, inheritance resistance and technology anxiety are taken as eight latent variables. Relevant hypotheses are proposed for the relationship among the eight variables. Through the questionnaire method, 321 valid data collected were validated and analyzed. Including the use of structural theory model for reliability analysis, correlation analysis, path coefficient test, etc., it is finally concluded that the mediating effect of perceived usefulness and willingness to pass on is obvious, while the mediating effect of perceived ease of use is relatively insignificant, and it cannot play a significant role of acceptance between resistance to passing on to audience acceptance. Trust, cultural confidence, perceived usefulness, perceived ease of use and inheritance willingness are positive feedback relationships to audience acceptance, and inheritance resistance to inheritance willingness and technology anxiety to audience acceptance are negative feedback relationships.

The stability of the supply chain has a significant impact on both the strategic deployment and operational efficiency of the enterprise, in order to optimize the supply chain management model for the enterprise and resolve the major supply chain risks, this paper realizes the optimization management and risk assessment of the supply chain through Monte Carlo simulation algorithm and SVM. Taking the newsboy problem as an entry point, a supply chain management optimization model is constructed, and Monte Carlo simulation algorithm is used to solve it. Using SVM regression and assessment ideas, supply chain risk regression assessment is carried out by C-SVR. Applying the supply chain management optimization model, it is concluded that when the optimal inventory of prefabricated components of the selected construction unit is 2.55×10³m³, the enterprise profit is the largest, which is 902.31×10³m³ yuan. Supply chain risk assessment of a port, the training error and prediction error of the assessment model in this paper are only 0.043% and 1.76%, which are significantly better than the BP neural network assessment method. Therefore, it proves that the work in this paper achieves the optimization and risk assessment of enterprise supply chain management model through simulation algorithm.

This study aims to explore how Hunan higher vocational colleges and universities can build a new ecosystem of industry-education integration through linear programming optimization strategies under the guidance of the strategy of developing the country through science and technology. The article evaluates and analyzes the ecosystem of industry-teaching integration in Hunan higher vocational colleges under the strategy of developing the country through science and education using linear programming method, and proposes relevant optimization strategies using the dyadic model of linear programming. The main factors affecting the efficiency of industry-teaching integration are identified through multiple regression analysis, including industry-teaching resources, incentive mechanism and management system. According to the linear programming model for maximizing the efficiency of industry-teaching integration in higher vocational colleges and universities, it is calculated that the efficiency of industry-teaching integration is maximized when Hunan higher vocational colleges and universities invest 3.6 million yuan, 0.3 million yuan and 0.5 million yuan in the resource consumption of industry-teaching resources, incentive mechanism and management system respectively. And it is proposed to build a new ecology of industry-education integration from three aspects of platform construction, cooperation docking and parenting path, respectively.

In this paper, the evaluation system of college students’ innovation and entrepreneurship education is constructed and the indexes are assigned by combining the hierarchical analysis method. After that, PSO algorithm is introduced in the optimization of weights and thresholds of BP neural network, the neural network model using particle swarm optimization (PSO-BP) is constructed, and the process of PSO algorithm optimization of BP neural network is described. It was found that the combined weight of five indicators, namely, “examination results of innovation and entrepreneurship courses, entrepreneurial experience, participation in centralized entrepreneurship training camps, obtaining financial support from entrepreneurship funds, and participation in innovation or entrepreneurship clubs”, accounted for more than 10%, while the combined weight coefficients of the rest of the indicators were all below 0.1. Compared to the BP model, the PSO-BP model has better network performance and its training samples have higher correlation with the test samples. In addition, the PSO-BP model can be used for predicting data prediction after 9 iterations of training, and the maximum relative error between the actual value and the expected value of the model network test output is very small (<1.4272%), which makes the model ideal. After PSO optimization PSO-BP model has almost no prediction error (<0.34%), which can improve the evaluation efficiency and accuracy.