一种多模型融合的风电系统永磁同步发电机数字孪生建模方法

Digital twin modeling method of wind power system permanent magnet synchronous generator based on multi model fusion

针对当前风电系统永磁同步发电机(PMSG)建模技术存在设备内部参数理想化和系统耦合单一化的不足,无法满足数字孪生技术对设备虚拟模型要求的问题,提出一种面向数字孪生的风电系统永磁同步发电机建模方法。首先,依据实体样机参数,构建2 MW PMSG本体及电磁模型。之后为实现系统对PMSG的控制,基于联合仿真技术利用数字孪生建模平台搭建风电系统PMSG控制电路,实现多系统耦合仿真建模;在此基础上,充分考虑电机传热散热特性等影响,对PMSG热模型进行了构建与计算,并针对温度场计算流程复杂,不具备实时性等问题,引入一种改进的粒子群优化支持向量机(PSO-SVM)代理模型方法对热模型进行降阶处理并集成;最终完成了基于多模型融合的PMSG数字孪生虚拟模型建模,并采用实体风机运行数据进行验证与测试,结果表明该模型可以有效地反映PMSG真实运行特性。

Aiming at the current modeling technology of permanent magnet synchronous generator(PMSG)in wind power system,there are deficiencies in idealization of internal parameters of equipment and simplification of system coupling,which can not meet the requirements of digital twin technology for virtual model of equipment.A digital twin oriented modeling method of permanent magnet synchronous generator for wind power system was proposed.Firstly,according to the parameters of the solid prototype,the 2 MW PMSG body and electromagnetic model were constructed.Then,in order to realize the control of PMSG by the system,based on the joint simulation technology,the PMSG control circuit of wind power system was built by using the digital twin modeling platform to realize the multi system coupling simulation modeling;On this basis,considering the influence of heat transfer and heat dissipation characteristics of the motor,the PMSG thermal model was constructed and calculated.Aiming at the problems of complex temperature field calculation process and lack of real-time,an improved particle swarm optimization sup-port vector machine(PSO-SVM)proxy model method was introduced to reduce the order of the thermal model and integrate it;Finally,the PMSG digital twin virtual model modeling based on multi model fu-sion was completed,and the physical fan operation data is used to verify and test.The results show that the model can effectively reflect the real operation characteristics of PMSG.