With advancements in reusable liquid rocket engine technology to meet the diverse demands of space missions, engine systems have become increasingly complex. In most cases, these engines rely on stable open-loop control and closed-loop regulation systems. However, due to the high degree of coupling and nonlinear dynamics within the system, most transient adjustments still depend on open-loop control. Open-loop control often fails to provide the optimal control strategy when encountering external disturbances. To address this issue, we introduce the intrinsically motivated twin delayed deep deterministic (TD3) algorithm, specifically designed for the startup process of LOX/Kerosene high-pressure staged combustion engine. This approach leverages intrinsic motivation to enable the algorithm to adapt to the abrupt parameter changes during the start-up process. A series of comprehensive experiments were conducted to verify the effectiveness of our method. The experimental results demonstrate that our method outperforms both the PID method and previous researchers' reinforcement learning methods based on the TD3 algorithm and DDPG, achieving a faster and more stable start-up process and significantly enhancing engine performance.

Carsharing has become a popular travel mode owing to its convenience of use, easy parking, and low cost of using a car by those who only need it occasionally. However, because of the inadequate location of carsharing stations (station-based systems) or vehicles (free-floating systems), effectively requiring expensive and complex relocation strategies, a number of customers are lost, and some carsharing companies are facing bankruptcy. This study proposes a data-driven, dynamic, multi-company relocation method, which aims to reduce relocation costs and increase profit in one-way carsharing station-based systems through cooperative strategies. The method starts from the prediction of carsharing inflows and outflows at each station throughout the day using a new deep learning algorithm designated as “the attention-enhanced temporal graph convolutional network”. It adopts an encoder-decoder structure to simultaneously capture the temporal and spatial carsharing usage patterns. A two-phase integer programming model is proposed to optimize the process of vehicle relocation and staff rebalancing with cooperative relocation strategies: the sharing of relocation staff, the sharing of vehicles and stations among the different companies. An adaptive large neighborhood search based heuristic approach is implemented to solve the two-phase model. Based on the 6-month travel records from four carsharing companies operating simultaneously in Fuzhou, China, the proposed model and cooperative strategies are assessed. The results show that the total profit of the four carsharing companies can be increased by 25.49% with the cooperation of staff and vehicles. In addition, we prospect the future relocation with automated vehicles, whereby the profit can be increased by 46.69% without the need to employ the relocation staff.

Free-floating bike sharing is an innovative and sustainable travel mode, where shared bikes can be picked up and returned at any proper place on the streets and not just at docking stations. Nevertheless, in these systems, two major problems arise. One is the imbalance of free-floating shared bikes (FFSB) between zones due to one-way trips, the other is the damaged bikes that must be brought for repair. In this study, a modeling framework for dynamic relocating operational and damaged bikes is proposed that starts with predicting the number and location of shared bikes using deep learning algorithms. The demand forecasting model adopts the Encoder-Decoder architecture embedded with the attention mechanism to further enhance the model's prediction ability and flexibility. Then, a data-driven optimization model for FFSB relocations is presented, where the multi-period optimization is applied to dynamically plan the relocation activities throughout the day. A new hybrid metaheuristic algorithm that incorporates variable neighborhood search (VNS) and enhanced simulated annealing (ESA) algorithm is developed for solving the relocating problem, in which satisfactory performance is observed from the numerical example. We test the proposed framework with the real-world FFSB data from Beijing, China. The results show that relocating both operational and damaged bikes timely decreases the probability of users finding damaged bikes in the system, but leads to higher relocation costs. For peak-hours, considering only the operational bikes for relocation is the most effective strategy given the limited relocation resources. It is urgent at those times of the day to focus on providing bikes to clients where they are undersupplied.

This paper proposes a controller design for the electric pump of a deep-throttling rocket engine. The nonlinearity of the system is taken into consideration by analyzing the gap metric. Then, proportional-integral-derivative controller and gain-scheduling linear quadratic regulator are designed. Analyzing the amplitude- and phase-frequency characteristics as well as the pole-zero distribution of the system, the results show that the designed controllers can stabilize the linearized equations in incremental form at different operating points. This indicates that these two controllers are available for the original system in the whole range of working conditions and this is verified in the simulation. Meanwhile, the comparison between proportional-integral-derivative controller and gain-scheduling linear quadratic regulator is presented. It demonstrates that the proportional-integral-derivative controller is better at tracking both step and ramp signals but with worse control signals. It means that the proportional-integral-derivative controller seems less suitable for real use due to severe oscillations. Meanwhile, the parameter tuning of a proportional-integral-derivative controller depends on more extensive manual tuning. Therefore, the gain-scheduling linear quadratic regulator is preferred.