ARM+FPGA双核计算的配电自动化终端设计

Design of Distribution Automation Terminal with ARM+FPGA Dual Core Computing

为了提高配电自动化终端数据信息自动化分析能力,设计了基于ARM+现场可编程门阵列(FPGA)双核计算的配电自动化终端。为了提高模块计算能力,在模块中构建了堆叠式自动编码器-神经网络(SAE-NN)深度学习算法模型。在常规堆叠式自动编码器(SAE)深度学习模型基础上融合神经网络(NN)模型,应用过程中改善传统NN对分层节点数目的限制。试验结果表明,所设计终端随着系统运行能达到95%以上的精度,而现有SAE模型仅达到85%左右的精度。通过与文献[1]和文献[2]方法的对比可知,所设计终端有较高的调度能力。该设计显著提高了配电网数据信息的分析精度,大幅提升了电网应用对数据信息处理的准确度和效率。

To improve the automated analysis capability of data information of distribution automation terminal,a distribution automation terminal based on ARM+field programmable gate array(FPGA)dual core computing is designed.To improve the computational capability of the module,a stacked autoencoder-neural network(SAE-NN)deep learning algorithm model is constructed in the module.The neural network(NN)model is fused based on the conventional stacked autoencoder(SAE)deep learning model,and the limitation of the traditional NN on the number of layered nodes is improved in the application process.The test results show that the designed terminal can eventually achieve more than 95%accuracy as the system runs,while the existing SAE model only achieves about 85%accuracy.By comparing with the literature[1]and literature[2]method,the designed terminal has a higher scheduling capability.The design significantly improves the accuracy of distribution network data and information,and greatly enhances the accuracy and efficiency of grid applications for data and information processing.