摘要
为了提高数字化电厂系统的故障预测精度,提出了一种基于粒子群算法优化的BP神经网络(PSO-BPNN)预测方法。将数字化电厂系统以往的历史故障数据作为输入数据,建立BPNN网络模型,并且为了提高BP神经网络模型精度,通过粒子群算法(PSO)对BPNN中的初始权值阈值进行优化,再对BP神经网络模型进行迭代优化,建立最优预测模型。将提出的PSO-BPNN模型与SVR模型、BPNN模型进行对比,验证了所提出的PSO-BPNN预测模型在数字化电厂系统故障预测中具有更高的预测精度。
In order to improve the accuracy of fault prediction in digital power plant systems,this paper proposes a particle swarm optimization based BP neural network(PSO-BPNN)prediction method.Using historical fault data from the digital power plant system as input data,a BPNN network model is established.In order to improve the accuracy of the BP neural network model,particle swarm optimization(PSO)is used to optimize the initial weight threshold in the BPNN,and then the BP neural network model is iteratively optimized to establish the optimal prediction model.The proposed PSO-BPNN model was compared with SVR model and BPNN model to verify its higher prediction accuracy in digital power plant system fault prediction.
作者
李利民
谢标林
LI Limin;XIE Biaolin(State Energy(Huizhou)Thermal Power Co.,Ltd.,Huizhou,Guangdong 5516082,China)
出处
《自动化与仪器仪表》
2025年第7期61-63,共3页
Automation & Instrumentation
关键词
数字化电厂
粒子群算法
BP神经网络
故障预测
digital power plant
particle swarm optimization algorithm
BP neural network
fault prediction
