The operating conditions of all-electric tugboats are flexible and changeable. They are more complicated than conventional vessels in terms of joint voyages and power generation scheduling. To guarantee the reliable operation of the ship, a new coordinated optimization scheme that combines economy and operational reliability is proposed. It is based on the various operating conditions of the tugboat during its voyage, taking into account the random outages of equipment and load fluctuations due to speed and wave uncertainties. Due to the difficulty of implementing a stochastic sampling method with space-time coupling constraints (e.g., the voyage is related to propulsion load), an analytical approach is needed to transform the model into a readily solvable mixed-integer linear program (MINP) which attributes risk scenarios to load fluctuations under various conditional probabilities. In addition, this paper proposes an improved piecewise linearization method based on a differential evolutionary algorithm to speed up the solution process and improve computational accuracy. Meanwhile, the energy storage loss cost due to battery degradation is added to the optimization target. The battery’s cycle life is extended by rational scheduling of charging and discharging. Simulations validate this paper’s joint scheduling optimization scheme in multiple comparison experiments. The results show that it can effectively balance the economic and reliability levels under various risk scenarios and improve the environmental energy efficiency indicators.

In bandwidth-limited image transmission, maintaining perceptual quality at very low bit rates remains challenging. Conventional compression codecs typically introduce noticeable blocking artifacts and discard fine details when the compression ratio becomes too aggressive. To address this problem, we present BCP-SDXL, an edge-cloud collaborative image compression framework with cloud-side generative reconstruction. Instead of transmitting full-resolution pixels, BCP-SDXL sends two lightweight cues: a compressed Base image and a Canny structural map. On the edge side, an Adaptive Compression Agent, built on a Deep Q-Network (DQN), dynamically selects compression parameters and extracts structural priors with low deployment overhead. Upon reception, the cloud server leverages the Multimodal Large Language Model (Qwen3-VL-2B) to analyze the compressed base image and generate high-fidelity semantic descriptions. These descriptions, together with the base image and edge maps, are then used to condition a Latent Diffusion Model (SDXL) integrated with ControlNet. The resulting multi-condition reconstruction process restores missing high-frequency details while preserving the original scene’s structural layout. Experiments on public datasets show that BCP-SDXL reduces transmission cost by more than 90% and achieves better perceptual quality than conventional codecs and recent deep learning baselines under highly compressed settings.