为提高柴油机装配质量和冷试性能,基于柴油机装配冷试基础数据集,选取加州大学欧文分校(University of California Irvine,UCI)机器学习资料库标准数据集中的Seeds、Wine、Wdbc三种数据集,对比支持向量机(support vector machines,SVM)...为提高柴油机装配质量和冷试性能,基于柴油机装配冷试基础数据集,选取加州大学欧文分校(University of California Irvine,UCI)机器学习资料库标准数据集中的Seeds、Wine、Wdbc三种数据集,对比支持向量机(support vector machines,SVM)模型、组合智能算法改进后SVM模型、Transformer模型应用于冷试异常数据的分析效果。结果表明:SVM、改进后SVM,Transformer模型对正常数据和异常数据分类的准确率分别为85.20%、92.54%、97.94%;相比SVM、改进SVM模型,Transformer模型的分类准确率有较大的提高,可用于分析参数异常;排气压力与转矩关系密切,排气压力较大造成转矩增大;排气门开启时间过长导致进气真空度异常,验证了Transformer模型对发动机装配异常识别方法的有效性。展开更多
针对低信噪比环境下超声细微缺陷特征提取难题,提出一种适用于低信噪比超声信号的门控残差与双级压缩-激励(squeeze and excitation,SE)注意力协同增强网络。该模型以卷积神经网络(convolutional neural network,CNN)为基础,通过残差块...针对低信噪比环境下超声细微缺陷特征提取难题,提出一种适用于低信噪比超声信号的门控残差与双级压缩-激励(squeeze and excitation,SE)注意力协同增强网络。该模型以卷积神经网络(convolutional neural network,CNN)为基础,通过残差块-SE模块-池化级联结构,在残差块内部嵌入普通SE模块进行初步通道筛选,在网络末端利用局部增强SE模块聚焦峰值信号,并采用门控残差连接从而动态保留原始细微特征,实现噪声抑制与特征增强的协同优化。结果显示:改进后模型的均方根误差(root mean square error,RMSE)均值为0.0683、平均绝对误差(mean absolute error,MAE)均值为0.0471,较基准CNN分别降低49.7%、41.7%,且模型显著优于仅使用单一注意力或残差块的改进模型,验证了双机制协同的优越性,且训练稳定性突出,低信噪比环境下仍保持高精度。所提模型的预测精度、抗干扰能力及稳定性显著优于传统方法与现有模型,为钢管超声无损检测提供高效技术方案,具有重要工业应用价值。展开更多
Large language model-based(LLM-based)text-to-SQL methods have achieved important progress in generating SQL queries for real-world applications.When confronted with table content-aware questions in real-world scenario...Large language model-based(LLM-based)text-to-SQL methods have achieved important progress in generating SQL queries for real-world applications.When confronted with table content-aware questions in real-world scenarios,ambiguous data content keywords and nonexistent database schema column names within the question lead to the poor performance of existing methods.To solve this problem,we propose a novel approach towards table content-aware text-to-SQL with self-retrieval(TCSR-SQL).It leverages LLM's in-context learning capability to extract data content keywords within the question and infer possible related database schema,which is used to generate Seed SQL to fuzz search databases.The search results are further used to confirm the encoding knowledge with the designed encoding knowledge table,including column names and exact stored content values used in the SQL.The encoding knowledge is sent to obtain the final Precise SQL following multirounds of generation-execution-revision process.To validate our approach,we introduce a table-content-aware,questionrelated benchmark dataset,containing 2115 question-SQL pairs.Comprehensive experiments conducted on this benchmark demonstrate the remarkable performance of TCSR-SQL,achieving an improvement of at least 27.8%in execution accuracy compared to other state-of-the-art methods.展开更多
文摘为提高柴油机装配质量和冷试性能,基于柴油机装配冷试基础数据集,选取加州大学欧文分校(University of California Irvine,UCI)机器学习资料库标准数据集中的Seeds、Wine、Wdbc三种数据集,对比支持向量机(support vector machines,SVM)模型、组合智能算法改进后SVM模型、Transformer模型应用于冷试异常数据的分析效果。结果表明:SVM、改进后SVM,Transformer模型对正常数据和异常数据分类的准确率分别为85.20%、92.54%、97.94%;相比SVM、改进SVM模型,Transformer模型的分类准确率有较大的提高,可用于分析参数异常;排气压力与转矩关系密切,排气压力较大造成转矩增大;排气门开启时间过长导致进气真空度异常,验证了Transformer模型对发动机装配异常识别方法的有效性。
文摘针对低信噪比环境下超声细微缺陷特征提取难题,提出一种适用于低信噪比超声信号的门控残差与双级压缩-激励(squeeze and excitation,SE)注意力协同增强网络。该模型以卷积神经网络(convolutional neural network,CNN)为基础,通过残差块-SE模块-池化级联结构,在残差块内部嵌入普通SE模块进行初步通道筛选,在网络末端利用局部增强SE模块聚焦峰值信号,并采用门控残差连接从而动态保留原始细微特征,实现噪声抑制与特征增强的协同优化。结果显示:改进后模型的均方根误差(root mean square error,RMSE)均值为0.0683、平均绝对误差(mean absolute error,MAE)均值为0.0471,较基准CNN分别降低49.7%、41.7%,且模型显著优于仅使用单一注意力或残差块的改进模型,验证了双机制协同的优越性,且训练稳定性突出,低信噪比环境下仍保持高精度。所提模型的预测精度、抗干扰能力及稳定性显著优于传统方法与现有模型,为钢管超声无损检测提供高效技术方案,具有重要工业应用价值。
基金supported by the National Key Research and Development Program of China under(Grant 2023YFB3106504)Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies under(Grant 2022B1212010005)+2 种基金the Major Key Project of PCL under(Grant PCL2023A09)Shenzhen Science and Technology Program under(Grants ZDSYS20210623091809029 and RCBS20221008093131089)the project of Guangdong Power Grid Co.Ltd.under(Grants 037800KC23090005 and GD-KJXM20231042).
文摘Large language model-based(LLM-based)text-to-SQL methods have achieved important progress in generating SQL queries for real-world applications.When confronted with table content-aware questions in real-world scenarios,ambiguous data content keywords and nonexistent database schema column names within the question lead to the poor performance of existing methods.To solve this problem,we propose a novel approach towards table content-aware text-to-SQL with self-retrieval(TCSR-SQL).It leverages LLM's in-context learning capability to extract data content keywords within the question and infer possible related database schema,which is used to generate Seed SQL to fuzz search databases.The search results are further used to confirm the encoding knowledge with the designed encoding knowledge table,including column names and exact stored content values used in the SQL.The encoding knowledge is sent to obtain the final Precise SQL following multirounds of generation-execution-revision process.To validate our approach,we introduce a table-content-aware,questionrelated benchmark dataset,containing 2115 question-SQL pairs.Comprehensive experiments conducted on this benchmark demonstrate the remarkable performance of TCSR-SQL,achieving an improvement of at least 27.8%in execution accuracy compared to other state-of-the-art methods.
