为实现沿海区域的海上风电场、海上采气平台和陆上热电联供燃气电厂等多种能源生产子单元的协同化运行,考虑可再生能源出力和氢负荷的随机波动,提出沿海区域综合能源生产单元(coastal integrated energy production units,CIEPU)随机优化调度模型.采用参数化代价函数近似(parametric cost function approximation,PCFA)的动态规划算法求解随机优化调度模型.通过一种基于梯度下降的求解方法——Adadelta法,获得策略函数的一阶信息,并计算梯度平方的指数衰减平均值,以更新策略函数的迭代步长;对随机优化调度模型进行策略参数逼近,从而得到近似最优的策略参数,并逐一时段求解出CIEPU的最优调度计划.最后,以某个CIEPU为例,分析计算结果表明,所提出方法获得的优化调度方案可以提高CIEPU运行的经济性并降低碳排放量,验证了所提方法的准确性和高效性.
Stochastic Optimal Dispatch of Coastal Integrated Energy Production Units
In order to realize the coordinated operation of various energy production sub-units such as the offshore wind farms(OWF),the offshore gas production platforms(OGPP)and the onshore heat and power-combined(CHP)plants in the coastal areas,considering the stochastic fluctuation of the renewable energy output and the hydrogen loads,a stochastic optimal dispatch model of the coastal integrated energy production units(CIEPU)is proposed.A parametric cost function approximation(PCFA)dynamic programming algorithm is put forward to solve the stochastic optimal dispatch model.Using a gradient descent-based Adadelta algorithm,the average exponential decay value of the squared gradient is calculated by the first-order information of the policy function to update the iteration steps.The policy parameter approximation is carried out for the stochastic optimal dispatch model,obtaining the approximate optimal policy parameters.Then,the optimal dispatch scheme of the CIEPU is solved in one time period after another.Finally,with the case study on an actual CIEPU,the calculation results show that the optimal dispatch scheme obtained by the proposed method is able to improve the economic operation and reduce the carbon emissions of the CIEPU,which demonstrates the computational accuracy and efficiency of the proposed method.
integrated energy production unitsstochastic optimal dispatchapproximate dynamic programmingparametric cost function approximationAdadelta algorithm