As an important branch of production scheduling, flexible job-shop scheduling problem (FJSP) is difficult to solve and is proven to be NP-hard. Many intelligent algorithms have been proposed to solve FJSP, but their key parameters cannot be dynamically adjusted effectively during the calculation process, which causes the solution efficiency and quality not being able to meet the production requirements. Therefore, a self-learning genetic algorithm (SLGA) is proposed in this paper, in which genetic algorithm (GA) is adopted as the basic optimization method and its key parameters are intelligently adjusted based on reinforcement learning (RL). Firstly, the selflearning model is analyzed and constructed in SLGA, SARSA algorithm and Q-Learning algorithm are applied as the learning methods at initial and later stages of optimization, respectively, and the conversion condition is designed. Secondly, the state determination method and reward method are designed for RL in GA environment. Finally, the learning effect and performance of SLGA in solving FJSP are compared with other algorithms using two groups of benchmark data instances with different scales. Experiment results show that the proposed SLGA significantly outperforms its competitors in solving FJSP.
Keywords: Flexible job-shop scheduling problem (FJSP) | Self-learning genetic algorithm (SLGA) | Genetic algorithm (GA) | Reinforcement learning (RL)

Mie lidar is a powerful tool for detecting the optical properties of atmospheric aerosols. However, there are two unknown parameters in the Mie lidar equation: the extinction coefficient and the backscattering coefficient. In the common methods for solving the equation, it is necessary to make assumptions about the relationship between the two unknown parameters. These assumptions will reduce the detection precision of extinction coefficient. In view of this, the back propagation (BP) neural network is used to retrieve extinction coefficient from the Mie scattering signal of lidar. Firstly, the structure and main parameters of the BP network are designed according to the practical application. In order to improve the convergence speed and prevent falling into local minima, the initial weights and thresholds of BP network are optimized by genetic algorithm (GA). Then the GA-BP network is trained with Mie scattering signal and the extinction coefficient retrieved by Raman method. Thus the mathematical relationship between Mie scattering signal and the extinction coefficient is stored in the BP network. The trained GA-BP network is then used to retrieve the extinction coefficient from Mie scattering signal in different conditions and the applicability of the GA-BP network is researched. The research will promote the development of Mie lidar retrieving algorithm.
Keywords: Aerosol | Mie scattering | Lidar | Extinction coefficient | BP network | Genetic algorithm

Under the background of current environmental pollution and serious shortage of fossil energy, the development of electric vehicles driven by clean new energy is the key to solve this problem, especially the hybrid electric vehicle driven by fuel cell is the most effective solution. Many scholars have found that the output performance of hybrid system is an important reason to determine the life of fuel cell. Unreasonable output will affect the control characteristics of the drive system, resulting in a series of serious consequences such as the reduction of the life of fuel cell hybrid power system. Therefore, the energy management strategy and performance optimization of hybrid system is the key to ensure the normal operation of the system. At present, many excellent researchers have carried out relevant research in this field. Genetic algorithm is a heuristic algorithm, which has better optimization performance. It can easily choose satisfactory solutions according to the optimization objectives, and make up for these shortcomings by using its own characteristics. These characteristics make genetic algorithm have outstanding advantages in the iterative optimization of energy management strategy. This paper analyzes and summarizes the optimization effect of genetic algorithm in various energy management strategies, aiming to analyze and select the optimization rules and parameters, optimization objects and optimization objectives. This paper hopes to provide guidance for the optimal control strategy and structural design of the fuel cell hybrid power system, contribute to the research on improving the energy utilization efficiency of the hybrid power system and extending the life of the fuel cell, and provide more ideas for the optimization of energy management in the future.
Keywords: Fuel cell hybrid electric vehicle | Energy management strategy | Hybrid power system | Genetic algorithm |Optimization parameters and objectives

Optimization of energy management strategy (EMS) for fuel cell/battery/ultracapacitor hybrid electrical vehicle (FCHEV) is primarily aimed on reducing fuel consumption. However, serious power fluctuation has effect on the durability of fuel cell, which still remains one challenging barrier for FCHEVs. In this paper, we propose an optimized frequency decoupling EMS using fuzzy control method to extend fuel cell lifespan and improve fuel economy for FCHEV. In the proposed EMS, fuel cell, battery and ultracapacitor are employed to supply low, middle and high-frequency components of required power, respectively. For accurately adjusting membership functions of proposed fuzzy controllers, genetic algorithm (GA) is adopted to optimize them considering multiple constraints on fuel cell power fluctuation and hydrogen consumption. The proposed EMS is verified by Advisor-Simulink and experiment bench. Simulation and experimental results confirm that the proposed EMS can effectively reduce hydrogen consumption in three typical drive cycles, limit fuel cell power fluctuation within 300 W/s and thus extend fuel cell lifespan.
Keywords: Fuel cell electrical hybrid vehicle | Energy management strategy | Frequency decoupling | Fuzzy control | Genetic algorithm

This paper proposes a novel methodology for redesigning a micro-grid characterized by a heavy reliance on diesel generators due to receiving power supply from an unreliable grid. The new design aims at phasing out the diesel generators and replacing them with a hybrid energy system composed of photovoltaics and a battery storage system. Two optimization approaches are adopted, a heuristic genetic algorithm approach is used to achieve sub-optimal sizing of the hybrid system sources and a rules-based dynamic programming approach to ensure optimal power flow. In order to reduce the computation time, a novel combinational approach employing genetic algorithm, dynamic programming and rules-based algorithm is proposed. The intervention of the dynamic programming for optimal power flow is restricted to certain active hours within a given day, while the rules-based power flow algorithm runs only outside those hours. The study demonstrates that the application of the hybrid system yields minimal operational cost by almost entirely phasing out the diesel generators and significantly reducing the energy purchased from the grid during peak hours. The micro-grid of a university campus is used as a case study where energy and economic indicators are derived to prove the superiority of the proposed techniques.
Keywords: Microgrid optimal design | Energy management system | Genetic algorithm | Dynamic programming | Energy economics