Research On Data-driven Green Efficiency Technology And Supporting Systems For Large-scale Component Forging Workshops

As the global economy continues to develop,energy consumption has reached unprecedented levels.In the manufacturing industry,improving energy efficiency is crucial for sustainable production and solving energy shortages.Large component forging and processing is a typical high-energy-consuming industry.Traditional forging workshops focused solely on improving productivity and efficiency,and workers often relied on experience for production,resulting in significant energy waste.Furthermore,a significant amount of energy consumption data was generated during the production process,but managers only roughly calculated the relationship between production and cost using data,resulting in a waste of valuable information.To address these issues,it is essential to use data-driven methods to explore energy value information and develop energy-saving and emission reduction optimization strategies.Therefore,this paper focuses on the key green efficiency enhancement technologies by data-driven methods,utilizing the Chongqing Science and Technology Innovation and Application Development Project "Key Technologies and Applications of Data-Driven Green Efficiency Enhancement in Factories" and the Chongqing Economic and Information Commission Project "Construction of a Digital Workshop Cloud Platform for Green Efficiency Enhancement of Large Heterogeneous Components".The research focuses on a forging workshop of a manufacturing enterprise as the object,and the following main contents are studied:(1)This study focuses on developing key technologies for obtaining multi-source energy consumption data and evaluating energy efficiency in large component workshops with complex layouts and harsh environments.A network architecture for multi-source energy consumption data collection based on IoT technology is proposed after analyzing the forging process characteristics,energy flow pattern,and workshop layout.Energy consumption data is used to develop energy consumption calculation models and six energy efficiency evaluation indicators from the equipment,product,and workshop layer perspectives.A multi-scale energy consumption coefficient calculation method based on the energy efficiency indicators is designed to determine the energy efficiency level of the workshop.A data-driven integrated energy efficiency evaluation system is established using these technologies.This system can assist manufacturers in making data-driven decisions to optimize their energy consumption and improve their overall energy efficiency.It has the potential to promote sustainable development in the manufacturing industry.(2)This research develops key technologies for data-driven energy consumption mining.It addresses challenges in large-scale component forging workshops,such as long-time series energy consumption data,unclear potential green value information,and unclear energy usage characteristics and interrelationships among equipment.The energy data is preprocessed in the data acquisition system,and the K-shape clustering algorithm is applied to analyze the energy consumption data of equipment.Daily basic energy consumption patterns of equipment are mined,and energy consumption feature curves are drawn.The FP-growth algorithm is then used for association analysis to mine the energy usage relationship between equipment and total energy consumption,as well as the interrelationships among energy consumption units in the workshop.Using 10 types of equipment in the workshop as an example,it is found that the energy consumption trend of the heat treatment equipment has the greatest impact on the total energy consumption,and the effectiveness of the algorithm is verified.Finally,the reasons for the energy usage model in the forging workshop are explained by combining the results of energy consumption data mining,and directions for energy-saving decision-making are obtained.These technologies can provide insights for manufacturers to make datadriven decisions and optimize their energy consumption,thereby enhancing energy efficiency and promoting sustainable development in the manufacturing industry.(3)This study aims to address the multi-objective decision-making and scheduling problem for green efficiency in large-scale forging workshops.The existing scheduling methods are found to be inadequate for formulating green efficiency strategies and optimizing production processes at the manufacturing level.To overcome this problem,a mathematical model is developed to minimize the maximum makespan and total energy consumption,considering energy consumption indicators,forging production equipment,and process characteristics.A matrix encoding method is designed,along with a decoding mechanism of "first come,first served" and "delayed loading into the furnace," taking into account the characteristics of the heating furnace.The problem is solved using a multi-objective evolutionary algorithm group(MOEAs)to obtain the Pareto optimal set.Testing is performed on three algorithms using a workshop example,and non-dominated sorting genetic algorithm Ⅱ(NSGA-Ⅱ)in MOEAs demonstrates good overall performance.However,it is found that the optimal algorithm for different production scales varies,suggesting that multiple algorithm selection should be considered for practical problem-solving.The key technologies discussed in the previous sections are put into practical use at a large forging enterprise in Chongqing.A green efficiency module is added to the existing energy management system to achieve visualization of the technology.The system is developed based on the proposed design framework and applied to half-year production energy consumption data for 6 types of equipment and actual production information for 4 types of 8 jobs.By applying the key technologies through the system,the decision results for energy efficiency evaluation,data mining,forging sequence,and equipment selection are visually displayed.This provides a basis and direction for formulating energy efficiency optimization strategies.Additionally,after applying the proposed green efficiency decision-making approach,the completion time and total energy consumption of large forging production are optimized by 11.6% and 3.9%,respectively.The results demonstrate the effectiveness of the proposed approach and its potential to be applied to similar industries.