为优化嵌入式软件可靠性预测智慧可控感知机制,构建了基于连续协同机器学习算法的嵌入式软件可靠性预测模型。构建连续协同机器学习算法机制实现嵌入式软件可靠性精准预测,利用深度LSTM构建时间正序下的嵌入式软件核心要素样本精准预测机制,利用DCNN对数据池后置测试集进行隐性知识感知并输出最优预测结果。最后,对模型开展了工程应用实践验证,结果表明,模型满足嵌入式软件可靠性预测智慧化改造需求,大幅度优化了嵌入式软件可靠性预测智慧可控感知机制。
Abstract
In order to optimize the intelligent controllable perception mechanism of embedded software reliability prediction, an embedded software reliability prediction model based on continuous cooperative machine learning algorithm is constructed.The continuous collaborative machine learning algorithm mechanism is constructed to realize the accurate prediction of embedded software reliability.The deep LSTM is used to construct the accurate prediction mechanism of embedded software core element samples in positive time order, DCNN is used to perceive the tacit knowledge of the post test set of the data pool and output the best prediction results.The results show that the model meets the needs of intelligent transformation of embedded software reliability prediction, and greatly optimizes the intelligent controllable perception mechanism of embedded software reliability prediction.
关键词
嵌入式软件 /
可靠性预测 /
连续协同机器学习 /
深度LSTM /
DCNN算法
Key words
embedded software /
reliability prediction /
continuous collaborative machine learning algorithm /
deep LSTM /
DCNN algorithm
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 曲以堃, 张伟.嵌入式软件系统体系结构可靠性分析方法[J].计算机应用研究, 2020, 37(10):30533057.
[2] 崔三俊, 张贵斌, 韩明.基于代码路径的嵌入式软件可靠度评估[J].微电子学与计算机, 2020, 37(11):7982.
[3] 马振宇, 张威, 吴纬, 等.基于优化SVR的军用软件可靠性预测方法[J].火力与指挥控制, 2019, 44(11):151155.
[4] Diwaker C, Tomar P, Solanki A, et al.A New Model for Predicting ComponentBased Software Reliability Using Soft Computing[J].IEEE Access, 2019, 180(7):147191147203.
[5] 武晔卿, 刘凯强, 彭耀光.基于隐患概率分析的嵌入式系统可靠性设计方法[J].单片机与嵌入式系统应用, 2020, 20(9):1416.
[6] 郭一帆, 唐家银.基于机器学习算法的寿命预测与故障诊断技术的发展综述[J].计算机测量与控制, 2019, 27(3):1319.
[7] 曾路, 汪浩.基于机器学习的虚拟仪器软件缺陷预测模型研究[J].自动化与仪器仪表, 2020, 247(5):6568.
[8] KaraAhmet.Multistep influenza outbreak forecasting using deep LSTM network and genetic algorithm[J].Expert Systems With Applications, 2021, 180(4):8187.
[9] 李梅.基于级联神经网络组合模型的软件可靠性预测研究[J].自动化与仪器仪表, 2017(7):1415.
[10] 杜瑞超, 华继学, 翟夕阳, 等.基于改进Real AdaBoost算法的软件可靠性预测[J].空军工程大学学报(自然科学版), 2018, 19(1):9196.
[11] Jabeen G, Luo P, Afzal W.An improved software reliability prediction model by using high precision error iterative analysis method[J].Software Testing, Verification and Reliability, 2019, 29(6):97108.
[12] 亓慧, 史颖, 李灯熬, 等.基于连续型深度置信神经网络的软件可靠性预测[J].计算机科学, 2021, 48(5):8690.
基金
*国家自然科学基金面上项目(61671375);陕西省自然科学基础研究计划项目(2019JQ736);西安市科技计划项目(GXYD14.13)。
PDF(1666 KB)
