摘要
固体氧化物电解池(SOEC)在高温环境下运行,内部存在复杂的热电相互耦合,温度、电压的控制对电堆平稳、安全运行至关重要。针对电堆复杂非线性、时空分布的特点,基于时空最小二乘支持向量机(LS-SVM)构建了SOEC温度、电压时空分布模型,采用一个核函数来描述电堆流道方向不同位置的空间相关性,采用动态回归方程描述电堆温度、电压分布的时间特性。通过Simulink构建SOEC机理仿真模型,生成样本数据来对时空分布模型进行训练并测试。仿真结果表明该模型能够准确预测SOEC温度、压力在时间及空间维度的分布,具有较优的泛化能力,可为电解系统优化与控制提供参考。
There are complex thermoelectric couplings in solid oxide electrolysis cells(SOEC)operating under high temperature environment.Thus,temperature control and voltage control are crucial to the stable and safe operation of SOEC stacks.In view that SOEC stacks are complex,nonlinear and spatiotemporal distributed systems,a spatiotemporal distributed model for the temperature and voltage of an SOEC is constructed based on the spatiotemporal least square support vector machine(LS-SVM).A kernel function is adopted to represent the spatial correlations at different locations along the flow channel,and dynamic regression equation is used to represent the temporal features of the temperature and voltage of the SOEC stack.A mechanism model of the SOEC stack is established in Simulink to obtain the sample data,which are used for the training and testing of the spatiotemporal distributed model.The simulation results show that the developed model can effectively predict the spatiotemporal distributions of the temperature and voltage in SOEC and has an excellent generalization ability,which can guide the further optimization and control of electrolysis cells.
作者
窦真兰
李佳文
张春雁
蔡祯祺
袁本峰
郏琨琪
肖国萍
王建强
DOU Zhenlan;LI Jiawen;ZHANG Chunyan;CAI Zhenqi;YUAN Benfeng;JIA Kunqi;XIAO Guoping;WANG Jianqiang(State Grid Shanghai Power Supply Company,Shanghai 200122,China;Shanghai Institute of Applied Physics,CAS,Shanghai 201800,China)
出处
《综合智慧能源》
CAS
2024年第7期53-62,共10页
Integrated Intelligent Energy
基金
国家电网有限公司科技项目(520911220003)。
关键词
固体氧化物电解池
温度控制
时空LS-SVM
分布式动态预测模型
电解系统
电解制氢
solid oxide electrolysis cell
temperature control
spatiotemporal LS-SVM
distributed dynamic prediction model
electrolysis system
water electrolysis for hydrogen production
