In this paper, we consider multistage optimal control of bioconversion glycerol to 1,3-propanediol(1,3-PD) in fed-batch fermentation process. To maximize the productivity of 1,3-PD, the whole fermentation process is divided into three stages according to the characteristics of microbial growth. Stages 2 and 3 are discussed mainly. The main aim of stage 2 is to restrict accumulation of 3-hydroxypropionaldehyde and maximize the biomass in the shortest time, and the purpose of stage 3 is to get high productivity of 1,3-PD. With these different objectives, multi-objective optimal control problems are proposed in stages 2 and 3. In order to solve the above optimal control problems, the multi-objective problems are transformed to the corresponding single-objective problems using the mass balance equation of biomass and normalization of the objective. Furthermore, the single-objective optimal control problems are transformed to two-level optimization problems by the control parametrization technique. Finally, numerical solution methods combined an improved Particle Swarm Optimization with penalty function method are developed to solve the resulting optimization problems. Numerical results show that the productivity of 1,3-PD is higher than the reported results.

The application of machine learning and deep learning in the field of vulnerability detection is a hot topic in security research, but currently it faces the problem of lack of dataset. Considering vulnerable code can be obtained from vulnerability fix commits, we propose an automatic vulnerability commit identification tool based on hierarchical attention network (HAN) to expand existing vulnerability dataset. HAN can model the input data at the word and sentence levels respectively and pay attention to the changes in the characteristics of different words in different categories, which improves the classification performance. Experimental results show that the accuracy and F1 of our model both achieve 92%. Through the vulnerability fix commit, researchers can quickly locate the vulnerable code. And extracting vulnerable code from open-source software can effectively expand the current dataset due to the enormous number of open-source software.