HYDROX is a new kind of torpedo energy. According to its application possibility to thermal power torpedo, the application patterns, supply of constituent element of the energy, adjustment of power system are discussed in the paper. By the conclusion that the paper comes to, it is clear that the using of HYDROX can increase torpedoes’ tactical and technical requirements, improve its concealment performance, reliability and maintainability, and the methods to supply hydrogen and oxygen are simple and feasible, and the security of hydrogen and oxygen can be well ensured, so HYDROX will have a great future. The method and system scheme mentioned in the paper contribute to the design of thermal power system.

Aiming at the fact that the current method of designing elastic coupling in swashplate engine is deficient, and the vibration damping effect is inadequate, a new method for optimization design of coupling is put forward based on back propagation artificial neural network and genetic algorithm(BP-GA). Firstly the dynamics model of swashplate engine shafting is set up, and the samples are gained by numerical simulation, then the non-linear mapping relationship of elastic coupling designed parameters of the objective function is established with BP neural network, finally the trained network is called back by GA to make global optimization. The optimization results show that the global optimal resolution can be searched rapidly and correctly with the method, besides, the method is precise, convenient and applicable.

Based on the analyses of the current solenoid valve driving circuits in marine diesel, a new type of dual-power double-maintain injector driving circuit is designed for marine high-pressure common-rail diesel. The circuit uses BOOST high voltage (80V) and storage battery low voltage (24V) to make up of dual-power time-sharing driver, which achieves automatic PWM feedback modulation of the solenoid valve injection current from the hardware. Experiments were carried out on a certain type of injector, the results showed that: the driving circuit had fast response time, only 0.045ms from zero to 22A of the solenoid valve current was required; the peak value of driving current has a good consistency, parameters including peakvalue current altitude, lasting time, maintain current altitude and maintain current lasting time could be adjusted flexibly. Besides, the circuit could be flexibly configured without occupying MCU resources.

Multi-agent reinforcement learning (MARL) is gradually becoming an attractive research field of adaptive traffic signal control (ATSC). Nevertheless, in a multi-agent environment, some inherent disadvantages exist, such as the partial observability and non-stationarity caused by the constantly changing decision-making strategies of agents, which have been extensively researched but remain challenging. Herein, NCCLight, which is a fully scalable decentralized MARL model built around an independent advantage actor-critic (IA2C) under the background of ATSC, is rationally designed and validated to offer a feasible approach to realizing communication and coordination between multiple agents. In addition, guided by cognitive consistency theory, the constraint of neighborhood cognitive consistency (NCC) is constructed to achieve communication and coordination between multiple agents. More significantly, cognitive consistency theory is employed in MARL for ATSC for the first time, which is validated by a large number of experiments on both real and synthetic data. We hope that this work can serve as a pioneering reference owing to the better performance of NCCLight than of the most advanced ATSC based on MARL.