Optimization method for data processing in digital twin-enabled computing power network for industrial production

To meet the demand for real-time sensing, synchronization, and analysis of high-throughput, heterogeneous industrial data in computing power twin networks for industrial scenarios, an optimization method for data processing in industrial computing twin networks was proposed to enhance data synchronization quality. Firstly, an industrial computing twin architecture was proposed, considering industrial production requirements and computing twin services. Secondly, in response to the digital twin’s requirements for low synchronization latency, high data accuracy, and efficient compu-ting resource utilization, a data processing model placement problem was formulated by comprehensively incorporating these factors. To solve this problem, an optimization algorithm based on the harmony search algorithm was developed to efficiently address model placement. Simulation results show that, compared with benchmark algorithms, the proposed algorithm provides better model placement decisions and demonstrates superior global optimum search capabilities.