基于改进自适应最稀疏时频分析的小型航空点燃式重油活塞发动机微弱爆震特征提取

Strong knock easily causes damage to the engine. Weak knock can increase the power output and reduce the fuel consumption. The extraction of the weak knock characteristics is an urgent problem to be solved in the knock control of the light-weight spark ignition heavy-oil aeroengine. At present, the empirical mode decomposition method is a new time frequency analysis method, which can be used to extract the weak knock characteristics of the engine. But this method has the problem of modal confusion. The Adaptive and Sparsest Time-Frequency Analysis Method can effectively solve the problem of modal confusion, but it will result in the distortion of the result due to the improper selection of the initial phase. For this reason, the penalty function particle swarm optimization algorithm is introduced into the initial phase optimization selection. An improved adaptive least sparse time-frequency analysis method is built to analyze the combustion pressure signal and cylinder vibration signal of the light-weight spark ignition heavy-oil aeroengine. In this way, the weak knock signal will be separated from the strong noise background and the weak knock feature frequency will be extracted. And a method based on dual harmonic component analysis and evaluation model is established to evaluate the decomposition ability of the improved adaptive least sparse time-frequency analysis method. The absolute error of the decomposed signal component and the real value is used as the evaluation index to analyze the inhibition ability of modal confusion. The results in this project are expected to provide important theoretical basis and technical support for the knock control and the safe operation of the Light-weight Spark Ignition Heavy-oil Aeroengine.

强烈爆震易造成发动机损毁，微弱爆震可提高动力输出及降低燃油消耗量，微弱爆震特征提取是小型航空点燃式重油活塞发动机爆震控制急需解决的问题。目前，经验模态分解方法是一种新型的时频分析方法可用于提取发动机微弱爆震特征，但存在模态混淆问题；自适应最稀疏时频分析方法可有效解决模态混淆问题，却会因初始相位选择不当而导致结果失真。为此，本项目拟将罚函数粒子群算法引入到初始相位的优化选择中，构建改进自适应最稀疏时频分析方法对小型航空点燃式重油活塞发动机的燃烧压力信号和缸体振动信号进行分析，将微弱爆震信号从强噪声背景中分离，并提取微弱爆震特征频率；建立基于双谐波分量信号分析评估模型的方法对改进自适应最稀疏时频分析方法的分解能力进行评价，以分解信号分量与真实值的绝对误差为评价指标，对其抑制模态混淆能力进行分析。本研究将为小型航空点燃式重油活塞发动机爆震控制及安全运行等提供良好基础和有利保障。

微弱爆震特征提取是小型航空点燃式重油活塞发动机爆震控制的关键问题，自适应最稀疏时频分析方法作为一种新型的时频分析方法，虽然可有效解决模态混淆问题，但会因初始相位选择不当而导致结果失真。因此本项目将罚函数粒子群算法引入到初始相位的优化选择中，构建改进自适应最稀疏时频分析方法对小型航空点燃式重油活塞发动机进行微弱爆震特征提取，开展了以下工作：. 在对自适应最稀疏时频分析方法基本原理的研究基础上，探讨了自适应最稀疏时频分析方法与经验模态分解方法等非参数化时频分析方法的理论共性及缺陷；并分析了初始相位对自适应最稀疏时频分析方法的影响。将罚函数粒子群算法引入到初始相位的优化选取中，以分解得到的分量最小误差平方和为优化目标函数，自适应地选取了最佳的初始相位。采用基于双谐波分量信号分析评估模型的方法研究了频率比、幅值比对改进自适应最稀疏时频分析方法分解能力的影响。搭建了小型航空点燃式重油活塞发动机微弱爆震信号采集实验台架，采集了燃烧压力信号和缸体振动信号，采用改进自适应最稀疏时频分解方法结合内禀模态函数能量法实现了微弱爆震特征频率的提取。以分解结果的精确性作为评价指标，与经验模态分解以及自适应最稀疏时频分析比较，验证了改进自适应最稀疏时频分析方法的抑制模态混淆能力。. 改进自适应最稀疏时频分析方法可自适应地选取最佳的初始相位，提高了抑制模态混淆能力和分解结果的精确性，本研究将为小型航空点燃式重油活塞发动机爆震控制及安全运行等提供了良好基础和有利保障，具有一定的理论意义和工程价值。