准确的健康状态(state of health,SOH)估算可以确保锂离子电池安全可靠运行,延长其使用寿命。针对当前许多健康特征无法表征电池老化机理,异常工况时无法准确追踪SOH变化趋势的问题,本文提出一种经验模型与数据驱动相结合的SOH估算方法...准确的健康状态(state of health,SOH)估算可以确保锂离子电池安全可靠运行,延长其使用寿命。针对当前许多健康特征无法表征电池老化机理,异常工况时无法准确追踪SOH变化趋势的问题,本文提出一种经验模型与数据驱动相结合的SOH估算方法。将锂离子电池负极固体电解质界面(SEI)膜增厚机理融入Arrhenius定律中构建经验模型,然后采用最小二乘法进行参数辨识,并分别计算每个参数与容量的Spearman相关系数。结果表明,它们与容量衰退都具有强相关性,可以作为估算SOH的健康特征。此外,为了克服双向长短期记忆(bidirectional long and short term memory,BiLSTM)网络参数较多且容易陷入过拟合的问题,本文使用减平均优化(subtraction average based optimizer,SABO)算法对BiLSTM的超参数进行寻优,建立SOH估算模型。最后,采用实验测试数据与美国航空航天局(National Aeronautics and Space Administration,NASA)数据验证了所提方法的适应性,并与长短期记忆(long and short-term memory,LSTM)网络、双向长短期记忆网络以及粒子群优化(particle swarm optimization,PSO)的双向长短期记忆网络3种算法的估算结果进行对比。结果表明,采用SABO-BiLSTM算法估算4节电池SOH的平均绝对百分比误差分别为0.043%、0.053%、0.259%、0.230%,相较于LSTM降低了94.58%、 92.85%、 88.65%、 90.13%,相较于BiLSTM降低了89.11%、91.60%、77.90%、76.41%,相较于PSO-BiLSTM降低了58.65%、58.91%、65.37%、69.29%。展开更多
DNA N6-甲基腺嘌呤(6mA)是一种重要的表观遗传修饰,参与基因调控、DNA复制和修复等生物过程,对疾病研究也具有重要意义,准确识别DNA 6mA位点对理解其功能和机制至关重要。尽管现有的NA 6mA位点预测方法已取得较大成功,但在预测精度和跨...DNA N6-甲基腺嘌呤(6mA)是一种重要的表观遗传修饰,参与基因调控、DNA复制和修复等生物过程,对疾病研究也具有重要意义,准确识别DNA 6mA位点对理解其功能和机制至关重要。尽管现有的NA 6mA位点预测方法已取得较大成功,但在预测精度和跨物种泛化能力上仍有改进空间。本文提出了一种结合双向长短期记忆网络(BiLSTM)和卷积神经网络(CNN)的混合深度学习模型(BiLSTM→CNN)来提高对DNA 6mA位点预测的能力。模型首先采用one-hot、EIIP和DNA二聚体三种编码方式对DNA序列进行编码,然后在不同网络结构、层数和优化器下优化模型。通过在蔷薇科植物、水稻和拟南芥的数据集上的广泛实验表明,BiLSTM→CNN模型在蔷薇科植物中的准确率(ACC)为94.5%,在水稻中为93.8%,在拟南芥中为86.6%。与其他方法相比,BiLSTM→CNN模型在三个植物物种的6mA位点预测中均展现出良好的性能,并具有出色的跨物种泛化能力。DNA N6-methyladenine (6mA) is an important epigenetic modification involved in biological processes such as gene regulation, DNA replication, and repair, making it significant for disease research. Therefore, accurately identifying DNA 6mA sites is crucial for understanding their functions and mechanisms. Despite notable successes with existing methods, there is still room for improvement in prediction accuracy and cross-species generalization. In this study, we propose a hybrid deep learning model (BiLSTM→CNN) that integrates bidirectional long short-term memory networks (BiLSTM) and convolutional neural networks (CNN). Firstly, the model-encoded DNA sequences employ one-hot encoding, EIIP encoding, and DNA dimer encoding. And then optimized under various network architectures, layer configurations and optimizers. We conducted experiments on datasets from Rosaceae, rice and Arabidopsis thaliana, the results indicate that the BiLSTM→CNNmodel achieves an accuracy (ACC) of 94.5% for Rosaceae, 93.8% for rice, and 86.6% for Arabidopsis. Compared to other methods, BiLSTM→CNNdemonstrates excellent performance in predicting 6mA sites across the three plant species, and exhibits cross-species generalization capabilities.展开更多
文摘准确的健康状态(state of health,SOH)估算可以确保锂离子电池安全可靠运行,延长其使用寿命。针对当前许多健康特征无法表征电池老化机理,异常工况时无法准确追踪SOH变化趋势的问题,本文提出一种经验模型与数据驱动相结合的SOH估算方法。将锂离子电池负极固体电解质界面(SEI)膜增厚机理融入Arrhenius定律中构建经验模型,然后采用最小二乘法进行参数辨识,并分别计算每个参数与容量的Spearman相关系数。结果表明,它们与容量衰退都具有强相关性,可以作为估算SOH的健康特征。此外,为了克服双向长短期记忆(bidirectional long and short term memory,BiLSTM)网络参数较多且容易陷入过拟合的问题,本文使用减平均优化(subtraction average based optimizer,SABO)算法对BiLSTM的超参数进行寻优,建立SOH估算模型。最后,采用实验测试数据与美国航空航天局(National Aeronautics and Space Administration,NASA)数据验证了所提方法的适应性,并与长短期记忆(long and short-term memory,LSTM)网络、双向长短期记忆网络以及粒子群优化(particle swarm optimization,PSO)的双向长短期记忆网络3种算法的估算结果进行对比。结果表明,采用SABO-BiLSTM算法估算4节电池SOH的平均绝对百分比误差分别为0.043%、0.053%、0.259%、0.230%,相较于LSTM降低了94.58%、 92.85%、 88.65%、 90.13%,相较于BiLSTM降低了89.11%、91.60%、77.90%、76.41%,相较于PSO-BiLSTM降低了58.65%、58.91%、65.37%、69.29%。
文摘DNA N6-甲基腺嘌呤(6mA)是一种重要的表观遗传修饰,参与基因调控、DNA复制和修复等生物过程,对疾病研究也具有重要意义,准确识别DNA 6mA位点对理解其功能和机制至关重要。尽管现有的NA 6mA位点预测方法已取得较大成功,但在预测精度和跨物种泛化能力上仍有改进空间。本文提出了一种结合双向长短期记忆网络(BiLSTM)和卷积神经网络(CNN)的混合深度学习模型(BiLSTM→CNN)来提高对DNA 6mA位点预测的能力。模型首先采用one-hot、EIIP和DNA二聚体三种编码方式对DNA序列进行编码,然后在不同网络结构、层数和优化器下优化模型。通过在蔷薇科植物、水稻和拟南芥的数据集上的广泛实验表明,BiLSTM→CNN模型在蔷薇科植物中的准确率(ACC)为94.5%,在水稻中为93.8%,在拟南芥中为86.6%。与其他方法相比,BiLSTM→CNN模型在三个植物物种的6mA位点预测中均展现出良好的性能,并具有出色的跨物种泛化能力。DNA N6-methyladenine (6mA) is an important epigenetic modification involved in biological processes such as gene regulation, DNA replication, and repair, making it significant for disease research. Therefore, accurately identifying DNA 6mA sites is crucial for understanding their functions and mechanisms. Despite notable successes with existing methods, there is still room for improvement in prediction accuracy and cross-species generalization. In this study, we propose a hybrid deep learning model (BiLSTM→CNN) that integrates bidirectional long short-term memory networks (BiLSTM) and convolutional neural networks (CNN). Firstly, the model-encoded DNA sequences employ one-hot encoding, EIIP encoding, and DNA dimer encoding. And then optimized under various network architectures, layer configurations and optimizers. We conducted experiments on datasets from Rosaceae, rice and Arabidopsis thaliana, the results indicate that the BiLSTM→CNNmodel achieves an accuracy (ACC) of 94.5% for Rosaceae, 93.8% for rice, and 86.6% for Arabidopsis. Compared to other methods, BiLSTM→CNNdemonstrates excellent performance in predicting 6mA sites across the three plant species, and exhibits cross-species generalization capabilities.
