针对火电机组SO_(2)排放质量浓度的影响因素众多,难以准确预测的问题,提出一种改进向量加权平均(weighted mean of vectors,INFO)算法与双向长短期记忆(bi-directional long short term memory,Bi-LSTM)神经网络相结合的预测模型(改进IN...针对火电机组SO_(2)排放质量浓度的影响因素众多,难以准确预测的问题,提出一种改进向量加权平均(weighted mean of vectors,INFO)算法与双向长短期记忆(bi-directional long short term memory,Bi-LSTM)神经网络相结合的预测模型(改进INFO-Bi-LSTM模型)。采用Circle混沌映射和反向学习产生高质量初始化种群,引入自适应t分布提升INFO算法跳出局部最优解和全局搜索的能力。选取改进INFO-Bi-LSTM模型和多种预测模型对炉内外联合脱硫过程中4种典型工况下的SO_(2)排放质量浓度进行预测,将预测结果进行验证对比。结果表明:改进INFO算法的寻优能力得到提升,并且改进INFO-Bi-LSTM模型精度更高,更加适用于SO_(2)排放质量浓度的预测,可为变工况下的脱硫控制提供控制理论支撑。展开更多
传统的静态检测恶意JavaScript代码方法十分依赖于已有的恶意代码特征,无法有效提取混淆恶意代码特征,导致检测混淆恶意JavaScript代码的精确率低。针对该问题提出基于双向长短期记忆网络(Bidirectional Long Short-term Memory, Bi-LS...传统的静态检测恶意JavaScript代码方法十分依赖于已有的恶意代码特征,无法有效提取混淆恶意代码特征,导致检测混淆恶意JavaScript代码的精确率低。针对该问题提出基于双向长短期记忆网络(Bidirectional Long Short-term Memory, Bi-LSTM)的恶意代码检测模型。通过抽象语法树将JavaScript代码转化为句法单元序列,通过Doc2Vec算法将句法单元序列用分布式向量表示,将句向量矩阵送入Bi-LSTM模型进行检测。实验结果表明,该方法对于混淆恶意JavaScript代码具有良好的检测效果且检测效率高,准确率为97.03%,召回率为97.10%。展开更多
Machine Learning(ML)algorithms play a pivotal role in Speech Emotion Recognition(SER),although they encounter a formidable obstacle in accurately discerning a speaker’s emotional state.The examination of the emotiona...Machine Learning(ML)algorithms play a pivotal role in Speech Emotion Recognition(SER),although they encounter a formidable obstacle in accurately discerning a speaker’s emotional state.The examination of the emotional states of speakers holds significant importance in a range of real-time applications,including but not limited to virtual reality,human-robot interaction,emergency centers,and human behavior assessment.Accurately identifying emotions in the SER process relies on extracting relevant information from audio inputs.Previous studies on SER have predominantly utilized short-time characteristics such as Mel Frequency Cepstral Coefficients(MFCCs)due to their ability to capture the periodic nature of audio signals effectively.Although these traits may improve their ability to perceive and interpret emotional depictions appropriately,MFCCS has some limitations.So this study aims to tackle the aforementioned issue by systematically picking multiple audio cues,enhancing the classifier model’s efficacy in accurately discerning human emotions.The utilized dataset is taken from the EMO-DB database,preprocessing input speech is done using a 2D Convolution Neural Network(CNN)involves applying convolutional operations to spectrograms as they afford a visual representation of the way the audio signal frequency content changes over time.The next step is the spectrogram data normalization which is crucial for Neural Network(NN)training as it aids in faster convergence.Then the five auditory features MFCCs,Chroma,Mel-Spectrogram,Contrast,and Tonnetz are extracted from the spectrogram sequentially.The attitude of feature selection is to retain only dominant features by excluding the irrelevant ones.In this paper,the Sequential Forward Selection(SFS)and Sequential Backward Selection(SBS)techniques were employed for multiple audio cues features selection.Finally,the feature sets composed from the hybrid feature extraction methods are fed into the deep Bidirectional Long Short Term Memory(Bi-LSTM)network to discern emotions.Since the deep Bi-LSTM can hierarchically learn complex features and increases model capacity by achieving more robust temporal modeling,it is more effective than a shallow Bi-LSTM in capturing the intricate tones of emotional content existent in speech signals.The effectiveness and resilience of the proposed SER model were evaluated by experiments,comparing it to state-of-the-art SER techniques.The results indicated that the model achieved accuracy rates of 90.92%,93%,and 92%over the Ryerson Audio-Visual Database of Emotional Speech and Song(RAVDESS),Berlin Database of Emotional Speech(EMO-DB),and The Interactive Emotional Dyadic Motion Capture(IEMOCAP)datasets,respectively.These findings signify a prominent enhancement in the ability to emotional depictions identification in speech,showcasing the potential of the proposed model in advancing the SER field.展开更多
针对现有模型仅考虑一种内部状态对锂电池性能退化影响的问题,同步建立3个模型分别预测3种时变状态随锂电池性能退化的变化轨迹,并以内阻与温度预测为基础实现锂电池容量的实时更新;针对传统神经网络中的sigmoid与ReLU激活函数存在梯度...针对现有模型仅考虑一种内部状态对锂电池性能退化影响的问题,同步建立3个模型分别预测3种时变状态随锂电池性能退化的变化轨迹,并以内阻与温度预测为基础实现锂电池容量的实时更新;针对传统神经网络中的sigmoid与ReLU激活函数存在梯度消失与神经元坏死问题,在双向长短时记忆(bi-directional long short term memory,Bi-LSTM)网络与全连接网络中引入一种新的Mish激活函数,使模型以平稳的梯度流提取更高质量的特征用于剩余使用寿命(RUL)预测的建模分析。最后利用蒙特卡洛(Monte Carlo,MC)与Dropout技术对锂电池RUL的预测结果不确定性进行分析。在美国Kristen教授课题组所公开的锂电池数据集上进行对比试验的结果表明,所提改进Bi-LSTM模型预测的均方误差(mean squared error,MSE)、平均绝对误差(mean absolute error,MAE)与R^(2)可达9.16×10^(-5)、0.00795、99.794%。随着获取数据量的增加,模型对锂电池RUL预测的精度越高,RUL平均预测误差可达2.3个循环,验证了所提模型能有效地实现锂电池循环RUL的实时更新。展开更多
为快速构建并准确预测温度作用引起的斜拉桥主梁应变用于结构状态评估,基于某大跨度斜拉桥主梁超过1年的温度和应变监测数据,提出了一种基于迁移学习和双向长短时记忆(bi-directional long short-term memory,Bi-LSTM)神经网络的斜拉桥...为快速构建并准确预测温度作用引起的斜拉桥主梁应变用于结构状态评估,基于某大跨度斜拉桥主梁超过1年的温度和应变监测数据,提出了一种基于迁移学习和双向长短时记忆(bi-directional long short-term memory,Bi-LSTM)神经网络的斜拉桥温度-应变映射模型建立方法。首先,利用解析模态分解(analytical mode decomposition,AMD)去噪应变数据,得到仅由温度引起的应变响应;其次,选择温度和某一测点应变数据构成数据集,采用Bi-LSTM神经网络训练该数据集,并通过网络结构和超参数优化建立温度-应变Bi-LSTM基准模型;最后,利用迁移学习方法,将已训练好的基准模型中部分参数迁移到其他温度-应变数据集,建立相应的温度-应变映射被迁移模型,并与未采用迁移学习的神经网络训练方法进行对比。研究结果表明,相比直接建立的温度-应变Bi-LSTM神经网络映射模型,采用迁移学习方法建立的被迁移模型,其拟合精度均高于所用的基准模型,且训练时间短,预测误差小。展开更多
文摘针对火电机组SO_(2)排放质量浓度的影响因素众多,难以准确预测的问题,提出一种改进向量加权平均(weighted mean of vectors,INFO)算法与双向长短期记忆(bi-directional long short term memory,Bi-LSTM)神经网络相结合的预测模型(改进INFO-Bi-LSTM模型)。采用Circle混沌映射和反向学习产生高质量初始化种群,引入自适应t分布提升INFO算法跳出局部最优解和全局搜索的能力。选取改进INFO-Bi-LSTM模型和多种预测模型对炉内外联合脱硫过程中4种典型工况下的SO_(2)排放质量浓度进行预测,将预测结果进行验证对比。结果表明:改进INFO算法的寻优能力得到提升,并且改进INFO-Bi-LSTM模型精度更高,更加适用于SO_(2)排放质量浓度的预测,可为变工况下的脱硫控制提供控制理论支撑。
文摘传统的静态检测恶意JavaScript代码方法十分依赖于已有的恶意代码特征,无法有效提取混淆恶意代码特征,导致检测混淆恶意JavaScript代码的精确率低。针对该问题提出基于双向长短期记忆网络(Bidirectional Long Short-term Memory, Bi-LSTM)的恶意代码检测模型。通过抽象语法树将JavaScript代码转化为句法单元序列,通过Doc2Vec算法将句法单元序列用分布式向量表示,将句向量矩阵送入Bi-LSTM模型进行检测。实验结果表明,该方法对于混淆恶意JavaScript代码具有良好的检测效果且检测效率高,准确率为97.03%,召回率为97.10%。
基金Projects(52078487,U1934207,52178180)supported by the National Natural Science Foundations of ChinaProjects(2022YFB2302603,2022YFC3004304)supported by the National Key Research and Development Program of China+2 种基金Project(2022TJ-Y10)supported by the Hunan Province Science and Technology Talent Lifting,ChinaProject(SKL-IOTSC(UM)-2021-2023)supported by the Science and Technology Development Fund,ChinaProject(SKL-IoTSC(UM)-2024-2026/ORP/GA08/2023)supported by the State Key Laboratory of Internet of Things for Smart City(University of Macao),China。
文摘Machine Learning(ML)algorithms play a pivotal role in Speech Emotion Recognition(SER),although they encounter a formidable obstacle in accurately discerning a speaker’s emotional state.The examination of the emotional states of speakers holds significant importance in a range of real-time applications,including but not limited to virtual reality,human-robot interaction,emergency centers,and human behavior assessment.Accurately identifying emotions in the SER process relies on extracting relevant information from audio inputs.Previous studies on SER have predominantly utilized short-time characteristics such as Mel Frequency Cepstral Coefficients(MFCCs)due to their ability to capture the periodic nature of audio signals effectively.Although these traits may improve their ability to perceive and interpret emotional depictions appropriately,MFCCS has some limitations.So this study aims to tackle the aforementioned issue by systematically picking multiple audio cues,enhancing the classifier model’s efficacy in accurately discerning human emotions.The utilized dataset is taken from the EMO-DB database,preprocessing input speech is done using a 2D Convolution Neural Network(CNN)involves applying convolutional operations to spectrograms as they afford a visual representation of the way the audio signal frequency content changes over time.The next step is the spectrogram data normalization which is crucial for Neural Network(NN)training as it aids in faster convergence.Then the five auditory features MFCCs,Chroma,Mel-Spectrogram,Contrast,and Tonnetz are extracted from the spectrogram sequentially.The attitude of feature selection is to retain only dominant features by excluding the irrelevant ones.In this paper,the Sequential Forward Selection(SFS)and Sequential Backward Selection(SBS)techniques were employed for multiple audio cues features selection.Finally,the feature sets composed from the hybrid feature extraction methods are fed into the deep Bidirectional Long Short Term Memory(Bi-LSTM)network to discern emotions.Since the deep Bi-LSTM can hierarchically learn complex features and increases model capacity by achieving more robust temporal modeling,it is more effective than a shallow Bi-LSTM in capturing the intricate tones of emotional content existent in speech signals.The effectiveness and resilience of the proposed SER model were evaluated by experiments,comparing it to state-of-the-art SER techniques.The results indicated that the model achieved accuracy rates of 90.92%,93%,and 92%over the Ryerson Audio-Visual Database of Emotional Speech and Song(RAVDESS),Berlin Database of Emotional Speech(EMO-DB),and The Interactive Emotional Dyadic Motion Capture(IEMOCAP)datasets,respectively.These findings signify a prominent enhancement in the ability to emotional depictions identification in speech,showcasing the potential of the proposed model in advancing the SER field.
文摘针对现有模型仅考虑一种内部状态对锂电池性能退化影响的问题,同步建立3个模型分别预测3种时变状态随锂电池性能退化的变化轨迹,并以内阻与温度预测为基础实现锂电池容量的实时更新;针对传统神经网络中的sigmoid与ReLU激活函数存在梯度消失与神经元坏死问题,在双向长短时记忆(bi-directional long short term memory,Bi-LSTM)网络与全连接网络中引入一种新的Mish激活函数,使模型以平稳的梯度流提取更高质量的特征用于剩余使用寿命(RUL)预测的建模分析。最后利用蒙特卡洛(Monte Carlo,MC)与Dropout技术对锂电池RUL的预测结果不确定性进行分析。在美国Kristen教授课题组所公开的锂电池数据集上进行对比试验的结果表明,所提改进Bi-LSTM模型预测的均方误差(mean squared error,MSE)、平均绝对误差(mean absolute error,MAE)与R^(2)可达9.16×10^(-5)、0.00795、99.794%。随着获取数据量的增加,模型对锂电池RUL预测的精度越高,RUL平均预测误差可达2.3个循环,验证了所提模型能有效地实现锂电池循环RUL的实时更新。
