基于改进粒子群优化与混合卷积神经网络的受端电网直流闭锁频率紧急控制决策优化

Optimization of Emergency Control Strategy for Frequency of Receiving-end Power Grid Under DC Blocking Based on Improved Particle Swarm Optimization and Hybrid Convolutional Neural Network

  • 摘要: 针对直流闭锁事故后受端电网频率安全问题,提出一种基于改进粒子群优化和混合卷积神经网络的频率紧急控制决策优化方法。首先,协调考虑紧急切负荷和抽蓄切泵控制措施,对受端电网频率紧急控制优化问题进行数学建模。然后,使用粒子群优化算法求解最优控制策略,并基于对立学习机制和混沌Tent映射改进粒子群优化算法,在保证紧急控制策略动态安全可行性前提下提高全局收敛性。最后,在粒子群优化过程中基于混合CNN构建多任务动态安全评估模型,快速判断紧急控制策略是否满足系统动态安全约束,提高频率紧急控制决策优化效率,并以某多直流馈入受端系统为例,验证所提方法有效性。

     

    Abstract: Aiming to address the frequency security risk of receiving-end power grid after DC-blocking, an optimization method of emergency control strategy for transient frequency was proposed based on improved particle swarm optimization (PSO) and hybrid convolutional neural network (CNN). Firstly, the optimization of emergency control strategy for frequency of receiving-end power grid was modeled with emergency load-shedding and cutting pump taken into account. The optimization problem was subsequently solved by PSO algorithm. To enhance the global convergence while ensuring the feasibility of control strategy at the same time, the PSO algorithm was improved based on opposition learning and chaos Tent mapping. Finally, to boost the optimization efficiency, the multiple-task dynamic security assessment model based on hybrid CNN was developed to determine whether the emergency control strategy can satisfy dynamic security constraints or not. Taking a receiving-end power grid with multi-infeed DC as an example, the validity of the proposed method was verified.

     

/

返回文章
返回