Thrust estimation is a significant part of aeroengine thrust control systems.The traditional estimation methods are either low in accuracy or large in computation.To further improve the estimation effect,a thrust esti...Thrust estimation is a significant part of aeroengine thrust control systems.The traditional estimation methods are either low in accuracy or large in computation.To further improve the estimation effect,a thrust estimator based on Multi-layer Residual Temporal Convolutional Network(M-RTCN)is proposed.To solve the problem of dead Rectified Linear Unit(ReLU),the proposed method uses the Gaussian Error Linear Unit(GELU)activation function instead of ReLU in residual block.Then the overall architecture of the multi-layer convolutional network is adjusted by using residual connections,so that the network thrust estimation effect and memory consumption are further improved.Moreover,the comparison with seven other methods shows that the proposed method has the advantages of higher estimation accuracy and faster convergence speed.Furthermore,six neural network models are deployed in the embedded controller of the micro-turbojet engine.The Hardware-in-the-Loop(HIL)testing results demonstrate the superiority of M-RTCN in terms of estimation accuracy,memory occupation and running time.Finally,an ignition verification is conducted to confirm the expected thrust estimation and real-time performance.展开更多
Based on the experimental ignition delay results of n-butane/hydrogen mixtures in a rapid compression machine,a Genetic Algorithm(GA)optimized Back Propagation(BP)neural network model is originally developed for ignit...Based on the experimental ignition delay results of n-butane/hydrogen mixtures in a rapid compression machine,a Genetic Algorithm(GA)optimized Back Propagation(BP)neural network model is originally developed for ignition delay prediction.In the BP model,the activation function,learning rate and the neurons number in the hidden layer are optimized,respectively.The prediction ability of the BP model is validated in wide operating ranges,i.e.,compression pressures from 20 to 25 bar,compression temperatures from 722 to 987 K,equivalence ratios from 0.5 to 1.5 and molar ratios of hydrogen(X_(H2))from 0 to 75%.Compared with the BP model,the GA optimized BP model could increase the average correlation coefficient from 0.9745 to 0.9890,in the opposite,the average Mean Square Error(MSE)decreased from 2.21 to 1.06.On the other hand,to assess the BP-GA model prediction ability in the never-seen-before cases,a limited BP-GA model is fostered in the𝑋X_(H2) range from 0 to 50%to predict the ignition delays at the cases of𝑋X_(H2)=75%.It is found that the predicted ignition delays are underestimated due to the training dataset lacking of“acceleration feature”that happened at𝑋X_(H2)=75%.However,three possible options are reported to improve the prediction accuracy in such never-seen-before cases.展开更多
基金co-supported by the National Natural Science Foundation of China(Nos.61890920,61890921)。
文摘Thrust estimation is a significant part of aeroengine thrust control systems.The traditional estimation methods are either low in accuracy or large in computation.To further improve the estimation effect,a thrust estimator based on Multi-layer Residual Temporal Convolutional Network(M-RTCN)is proposed.To solve the problem of dead Rectified Linear Unit(ReLU),the proposed method uses the Gaussian Error Linear Unit(GELU)activation function instead of ReLU in residual block.Then the overall architecture of the multi-layer convolutional network is adjusted by using residual connections,so that the network thrust estimation effect and memory consumption are further improved.Moreover,the comparison with seven other methods shows that the proposed method has the advantages of higher estimation accuracy and faster convergence speed.Furthermore,six neural network models are deployed in the embedded controller of the micro-turbojet engine.The Hardware-in-the-Loop(HIL)testing results demonstrate the superiority of M-RTCN in terms of estimation accuracy,memory occupation and running time.Finally,an ignition verification is conducted to confirm the expected thrust estimation and real-time performance.
基金The authors would like to acknowledge the financial support to the research provided by the National Natural Science Foundation of China through the Project of 51922076 and 51706140.
文摘Based on the experimental ignition delay results of n-butane/hydrogen mixtures in a rapid compression machine,a Genetic Algorithm(GA)optimized Back Propagation(BP)neural network model is originally developed for ignition delay prediction.In the BP model,the activation function,learning rate and the neurons number in the hidden layer are optimized,respectively.The prediction ability of the BP model is validated in wide operating ranges,i.e.,compression pressures from 20 to 25 bar,compression temperatures from 722 to 987 K,equivalence ratios from 0.5 to 1.5 and molar ratios of hydrogen(X_(H2))from 0 to 75%.Compared with the BP model,the GA optimized BP model could increase the average correlation coefficient from 0.9745 to 0.9890,in the opposite,the average Mean Square Error(MSE)decreased from 2.21 to 1.06.On the other hand,to assess the BP-GA model prediction ability in the never-seen-before cases,a limited BP-GA model is fostered in the𝑋X_(H2) range from 0 to 50%to predict the ignition delays at the cases of𝑋X_(H2)=75%.It is found that the predicted ignition delays are underestimated due to the training dataset lacking of“acceleration feature”that happened at𝑋X_(H2)=75%.However,three possible options are reported to improve the prediction accuracy in such never-seen-before cases.
文摘根据煤的硫分、灰分以及煤自燃过程中的耗氧速率、CO和CO2产生率等随温度变化的序列值与煤自然发火期之间存在的密切对应关系,建立了前向多层人工神经网络模型,用已有的煤自然发火实验数据对网络进行训练,得到了神经元间的联结强度,从而准确地表征这种对应关系.设计了一套油浴程序升温实验装置,确定了实验试管的尺寸和实验条件,从而能够准确测定煤自燃在不同温度下的耗氧速率及气体产生率.将煤样油浴程序升温实验数据及煤质分析数据代入人工神经网络,可算出煤的自然发火期.与煤自然发火实验相比,该方法测定煤样的自然发火期用煤量减少了99%以上,实验耗时缩短了90%以上,二者测试结果的偏差小于3 d.
