Rapid and accurate assessment of power network loss in the power system has become a key research topic for the vast and diverse dataset of power grid operation. This study integrates data mining techniques with typical scenario modeling concepts and innovatively designs a distribution area power network loss rate multi population parallel spectral clustering evaluation strategy that incorporates distribution characteristics. Firstly, clustering attributes are determined for power network loss evaluation, and a power network loss evaluation framework based on clustering algorithms is proposed. Based on power flow calculation, the distribution characteristics and indicator system of each node’s output are analyzed; Secondly, in order to improve the clustering accuracy of power network loss evaluation, spectral clustering algorithm is introduced, and automatic algorithm design is carried out to address the issue of manually setting the initial number of clusters and cluster centers. Then, multi cluster partitioning and parallel computing methods are used to significantly improve the computational efficiency of spectral clustering algorithm; Finally, to verify the practicality of this method, a provincial power grid was selected as a case study. The results showed that this method not only has high accuracy in evaluating power network loss, but also has excellent computational efficiency, demonstrating good feasibility in practical engineering applications.

Transportation electromechanical engineering has an important role in the process of transportation engineering construction. After studying the basic situation and problems of the current transportation electromechanical engineering, the author selects the Q city subway line 1 as a test of the effect of quality control of its transportation electromechanical engineering construction, so as to assess its performance. Ten monitoring points are randomly selected, and four groups of control quantities between the monitoring points are detected, and their qualification rate is judged by the gap between the detection results and the standard value. Then, optimize the transportation electromechanical engineering by using big data and emerging detection technology, etc. Construct the performance evaluation index system of traffic electromechanical engineering to evaluate the performance of the optimized traffic electromechanical engineering scheme. The pass rate of the preoptimized traffic electromechanical engineering in the four groups of control volume testing is 100%, 22%, 80% and 70%, respectively, and the construction performance is poor. The scores of the equipment layer indicators of the optimized traffic engineering scheme were all above 80 points, and 19 indicators scored more than 90 points. The subsystem index scores are between 87.1 and 96.2 points, and the comprehensive score of traffic E&E engineering performance is 92.2 points, which shows that the optimized traffic E&E engineering has achieved more excellent performance evaluation results.

In the face of the requirements of the financial management system, enterprises need to accelerate the digital transformation of finance and realize the “data-driven” management decision-making operation mechanism. The article constructs a new quality productivity-based finance-driven enterprise digital transformation path, and makes it clear that enterprises need to play a new type of labor objects, labor materials and laborers to achieve digital transformation. Based on this transformation framework, a system dynamics approach is used to construct an enterprise financial dynamic planning system, which consists of five parts: a financial analysis subsystem, a target gap adjustment subsystem, an income statement subsystem, a balance sheet subsystem, and a production and operation subsystem, and analyzes the driving factors that affect value growth. The feasibility of the model is determined through the methods of structure test and sensitivity test on the dynamic financial planning model. Taking Group A as a case study object, the financial data for the five-year period from 2019-2023 are analyzed, and the operation of the enterprise is reflected through the financial indicators of each system, which proves the validity of the model and promotes the realization of the digital transformation of the enterprise, which contributes to the management of enterprise value.

Encouraged by the strategy of rural revitalization, rural areas in many places are exploring the development path of characteristic industries. The article embeds the multi-objective optimization model into the development of rural characteristic industries to optimize the current rural industrial development path. The multi-objective optimization model of rural characteristic industry development is constructed, and the ACO-PSO algorithm is used to solve the model, in order to realize the organic unity of economic, social and ecological benefits of rural industry development. The multi objective optimization model is used to optimize the industrial development of village S. The total regional output value of village S in 2035 is 2.08 times of that in 2025. The proportion of output value of primary industry and secondary industry decreases by 20.29% and 18.50% respectively. The proportion of tertiary industry output value increases by 38.79%, and the industrial structure becomes more and more reasonable. After the multi-objective optimization, Village S changes the development mode at the expense of resources and the environment, and maintains the survival of the ecological environment by appropriately slowing down the economic development. After the multi objective optimization, the total output value of the primary industry and the per capita income of farmers in Village S increased by 17,412 and 205.76 yuan respectively. The total output of tourism in the tertiary industry is 465,222,000 yuan, which is 126% higher than that before optimization.

As the most popular mode of out-of-school education in recent years, study tour plays an important role in the comprehensive ability improvement and overall development of students. Based on the path planning problem of study tour, this paper proposes a travel route optimization model with time optimization as the goal orientation, aiming to plan the time-optimal path for students in the study tour process. The particle swarm algorithm is used to improve the genetic algorithm for solving the travel route optimization model. The effectiveness of the optimization model and the hybrid algorithm is verified through the analysis of an actual case of a study tour, and the experimental results are substantially optimized compared with the traditional planning path, reducing the time spent by 2.2 days. Then we use qualitative comparative analysis method to explore the efficiency improvement of the curriculum of study tours, and obtain four grouping paths, which can cover more than 85% of the cases. The research in this paper not only helps to enrich the academic research of cross disciplines in the form of “travel + education”, but also provides theoretical basis and practical reference for the development of study tours to a certain extent.

Particle swarm algorithm, as a kind of population intelligent optimization algorithm, shows great potential in solving multivariate and nonlinear optimization problems due to its simple and efficient characteristics. The article constructs a concrete ratio optimization model in construction engineering technology, which is supported by particle swarm algorithm as the main technology. The model also integrates the least squares support vector regression algorithm, which makes it not only simple ratio optimization, but also has the function of concrete performance prediction. The relative error of the model in predicting the physical properties of concrete is small, less than 5%, which improves the reliability of concrete proportioning. The concrete samples generated by the model with five different ratios have better physical properties for daily needs. In the durability test, the concrete sample with proportion 4 showed the best performance in terms of mass loss rate and impermeability, which were 3.52% (after 400 cycles) and 156.44C (after 56d), respectively. And all the concrete samples used were in the range of proportional qualification and the cost was 5.99% to 28.61% lower than the comparison method.

Blockchain theory and its key technologies are developing rapidly, and the industrial internet combined with blockchain technology is driving the realization of safe and reliable comprehensive connectivity in multiple fields. In this paper, we propose a resource optimization allocation method for industrial internet that integrates edge computing and blockchain to reduce the task computing energy consumption and computational overhead of the system while improving the efficiency of the consensus process. This optimization problem is constructed as a Markov decision process, and a deep reinforcement learning algorithm is used to solve the optimal resource scheduling strategy under a single edge node. The effectiveness of the proposed resource optimization allocation method for industrial internet fusing edge computing and blockchain is verified by simulation validation. The method is able to obtain better and smoother convergence under the premise of harvesting high total rewards, effectively reduces the computational energy consumption and computational overhead of the device, and at the same time effectively improves the consensus efficiency of the blockchain.

This paper firstly constructs a coupled evaluation index system based on three primary indicators, seven secondary indicators and 29 tertiary indicators for agriculture, culture and tourism. Then the entropy weight-TOPSIS method and the coupling coordination degree model are selected to measure the development of agriculture, culture and tourism industry and the coupling level in Yijun County respectively. Finally, 23 spatially related villages in Yijun county area are selected to reveal the reasons for their spatial differences with spatial measurement model, and analyze the factors affecting the coupled and coordinated development of agricultural, cultural and tourism industries in Yijun county area. From the comprehensive evaluation results, the trend of the development level of agriculture, culture and tourism in 2017-2023 was generally upward, in which the agriculture industry had the highest growth between 2022 and 2023, with an increase of 0.0445. After analyzing the factors influencing the development of the coupled agriculture, culture and tourism industries in the Yijun county region by applying the spatial Durbin model, it was found that the general budget expenditures, human capital, infrastructure construction, fixed asset investment and education investment in the region at a significant level of 0.01 correlation of 0.211, 0.03, 0.082 and 0.085, and education investment in the region at a significant level of 0.1 correlation of 0.211. These five factors significantly affect the Yijun County region agriculture, culture and tourism industry, and deepen the development of integration of tourism industry.

In recent years, intelligent control has realized rapid development in the field of electrical engineering, the article initially studied the principle of electrical intelligent control, accordingly built the electrical intelligent control system, and designed the system hardware, the system module is divided into the main control module, the expansion module, the digital input and output module and the mounting rail. Based on the working principle of fuzzy control, design the software of the electrical intelligent control system, and optimize the traditional fuzzy controller by using fuzzy adaptive hybrid genetic algorithm, so as to improve the fuzzy control accuracy of the electrical intelligent control system in this paper. The electrical control system of this paper is applied to greenhouse greenhouse temperature and humidity control, substation air conditioning energy consumption control and subway station illumination control, and the control effect of the electrical intelligent control system of this paper is known through three experimental data. The system of this paper can effectively deal with the dissimilar data in the greenhouse temperature control experiment. Under the steady state environment, the temperature deviation of this paper’s fuzzy control method and conventional single structure fuzzy control is within 0.1℃ and 1℃ respectively, and the humidity deviation is within 5%RH and 10%RH respectively. Obviously, the fuzzy control method in this paper has higher control accuracy. In the substation air conditioning energy consumption experiment, the annual power consumption of this paper’s electrical intelligent control system and the traditional ventilation and air conditioning system are 32,660 degrees and 45,620 degrees, respectively. The electrical intelligent control system in this paper can save 22,000 yuan per year. The output illuminance of the subway station of the fuzzy control system in this paper increases with the comfort of the light environment and the density of the crowd, which achieves the expected effect.

In today’s rapid development of information technology, the big data industry has ushered in explosive growth, and big data analysis has become an important research topic in the cross-cutting field of computing. This study constructs a big data prediction base model based on deep learning, and uses the improved butterfly optimization algorithm with OGRU model to realize feature selection and classification processing of big data. Then the Adam algorithm is used to optimize the parameters of the model, and finally the classification and prediction model of big data based on deep learning is constructed. Simulation and empirical analysis results show that the model proposed in this paper has excellent classification and prediction performance, and can meet the efficiency requirements of big data classification and prediction. The prediction errors of distribution network load data and smart charging pile operation data are lower than 9% and 16%, respectively, which have high practical application value. This study is of great significance to the research related to big data classification and prediction in different fields, and provides an effective method for data prediction in complex scenarios such as industrial as well as power grid scheduling.