CN113257368B - Gas equivalence ratio prediction method, system and processing terminal - Google Patents

Abstract

The present invention belongs to the field of gasThe technical field of equivalence ratio prediction discloses a method, a system and a processing terminal for predicting the gas equivalence ratio, wherein the method for predicting the gas equivalence ratio comprises the following steps: measuring by using a multi-frequency excitation ion current sensor to obtain a multi-frequency excitation ion current time-varying signal in the measured combustor; non-whole period online demodulation is carried out on the multi-frequency excitation time-varying signal to obtain ion signal amplitudes corresponding to different excitation frequencies at different moments, and then discrete ion current amplitude spectrums S corresponding to different moments are obtained_A^t (ii) a Discrete amplitude spectrum S at different moments by principal component analysis_A^t Performing characteristic dimension reduction and denoising to obtain dimension reduction characteristic matrix X at different moments_t (ii) a Mixing X_t The ion current amplitude spectrum-gas equivalence ratio nonlinear prediction model established in the early-stage literature is input to realize the real-time prediction of the gas equivalence ratio at different moments. The method combines the multi-frequency excitation demodulation and the nonlinear prediction model, is favorable for accurately describing the real-time change condition of the working condition parameters, reduces the experiment times and realizes the real-time prediction of the gas equivalence ratio.

Description

Translated fromChinese

一种燃气当量比预测方法、系统及处理终端A gas equivalence ratio prediction method, system and processing terminal

技术领域technical field

本发明属于燃气当量比预测技术领域，尤其涉及一种燃气当量比预测方法、系统及处理终端。The invention belongs to the technical field of gas equivalence ratio prediction, and in particular relates to a gas equivalence ratio prediction method, system and processing terminal.

背景技术Background technique

目前，燃气当量比表征了参加燃烧反应的组分配比情况，它决定了燃烧反应能否启动及燃烧反应的剧烈程度，同时也影响着燃烧的完全性、热声振荡程度、共振频率及污染物的排放水平等。因此，燃气当量比是燃烧过程中的关键状态参数。At present, the gas equivalence ratio characterizes the proportion of components participating in the combustion reaction. It determines whether the combustion reaction can be started and the intensity of the combustion reaction. It also affects the completeness of the combustion, the degree of thermoacoustic oscillation, the resonance frequency and pollutants. emission levels, etc. Therefore, the gas equivalence ratio is a key state parameter in the combustion process.

目前，主要使用两种方法来获得当量比。最基本的方法是通过安装在燃烧系统入口上的流量计测量入口燃料和空气流量，进而计算燃气当量比。该方法广泛应用于大多数入口流量稳定的燃烧系统中。然而，对于入口流量随时间变化的燃烧系统而言，通过测定入口流量来确定当量比的方法将受到限制。另一种方法则是使用化学平衡方程从测量的CO，CO₂和O₂浓度计算当量比，目前最流行的方式是使用安装在排气流中的氧气传感器来测量氧气浓度，氧气传感器主要包括加热废气氧(HEGO)传感器和商用通用废气氧(UEGO)传感器；但HEGO传感器无法给出当量比的确切值，UEGO传感器虽然能提供准确的当量比值，但价格昂贵，且大多局限于实验室应用。感兴趣燃烧产物的浓度还可以采用激光诱导击穿光谱(LIB)，激光诱导荧光(LIF)和红外吸收光谱技术等先进的光谱诊断技术获得。但光学诊断技术测量过程中需要开设光学窗口，在实际应用中受到限制。此外，研究人员提出了多种基于压力、化学发光等燃烧参数来预测当量比的方法，并结合神经网络(ANN)及支持向量机(SVM)等人工智能算法进行特征提取和当量比估计。火花塞等多种形式的离子传感器也被成功用于多种发动机燃烧室中的燃气当量比预测。研究表明，离子电流信号的特征值(如最大值、积分值等)与燃气当量比之间存在直接的关系，离子电流传感器结构简单、安装方便、灵敏度高、环境适应性强，可以放置在燃烧器中的任意感兴趣区域，在变流量燃烧系统的燃气当量比测定中具有很好的应用前景。Currently, two methods are mainly used to obtain the equivalence ratio. The most basic method is to calculate the gas equivalence ratio by measuring the inlet fuel and air flow through a flow meter installed on the combustion system inlet. This method is widely used in most combustion systems with stable inlet flow. However, for combustion systems where the inlet flow varies with time, the method of determining the equivalence ratio by measuring the inlet flow will be limited. Another method is to use the chemical equilibrium equation to calculate the equivalence ratio from the measured CO,_CO2 and_O2 concentrations. The most popular way is to use an oxygen sensor installed in the exhaust stream to measure the oxygen concentration. The oxygen sensor mainly includes Heated Exhaust Gas Oxygen (HEGO) sensors and commercial Universal Exhaust Gas Oxygen (UEGO) sensors; however, HEGO sensors cannot give exact values of equivalence ratios, UEGO sensors can provide accurate equivalence ratios but are expensive and mostly limited to laboratory applications . The concentrations of combustion products of interest can also be obtained using advanced spectroscopic diagnostic techniques such as laser-induced breakdown spectroscopy (LIB), laser-induced fluorescence (LIF) and infrared absorption spectroscopy. However, the optical window needs to be opened in the measurement process of optical diagnostic technology, which is limited in practical application. In addition, the researchers proposed a variety of methods to predict the equivalence ratio based on combustion parameters such as pressure and chemiluminescence, and combined with artificial intelligence algorithms such as neural network (ANN) and support vector machine (SVM) for feature extraction and equivalence ratio estimation. Various forms of ion sensors, such as spark plugs, have also been successfully used for gas equivalence ratio prediction in various engine combustion chambers. Studies have shown that there is a direct relationship between the characteristic values of the ion current signal (such as the maximum value, integral value, etc.) and the gas equivalence ratio. The ion current sensor has simple structure, convenient installation, high sensitivity, and strong environmental adaptability. It has a good application prospect in the determination of the gas equivalence ratio of the variable flow combustion system.

离子电流传感器的测量原理是：燃烧过程中会由于化学电离而产生离子和电子等带电粒子，这些带电粒子的存在使得燃烧介质具有电特性。由于带电粒子的存在，燃烧介质呈现电中性弱等离子体特性，当加入外加电场后，带电粒子在电场力作用下定向移动形成电流，称为离子电流。离子电流能够反映燃烧过程中特征参数的变化情况，通过测量离子电流信号的大小及变化情况可以获得燃烧场中离子浓度的变化情况，进而间接获得燃烧运行状况。The measurement principle of the ion current sensor is that charged particles such as ions and electrons are generated due to chemical ionization during the combustion process. The existence of these charged particles makes the combustion medium have electrical properties. Due to the existence of charged particles, the combustion medium exhibits the characteristics of electrically neutral and weak plasma. When an external electric field is added, the charged particles move directionally under the action of the electric field force to form a current, which is called ionic current. The ion current can reflect the change of characteristic parameters in the combustion process. By measuring the magnitude and change of the ion current signal, the change of the ion concentration in the combustion field can be obtained, and then the combustion operation state can be obtained indirectly.

然而，以往的离子电流传感器均采用直流或者单频交流激励测量，单次测量得到是离子电流信号随时间的变化曲线，包含的特征信息有限；采用多频激励方式对脉动燃烧中的离子浓度变化进行测量时，传感器两级间的离子电流响应特性随着激励频率的变化而变化，可以得到离子电流信号随激励频率变化的响应曲线，即离散离子电流谱信息。对于给定的外部电场，离子电流幅值谱的特征取决于燃烧状况，而燃烧状况与燃气当量比密切相关。因此，当量比可以视为离子电流幅值谱的函数，该函数本质上是非线性的，可以使用非线性预测模型来逼近它。前期文献中建立了当量比与离散离子电流幅值谱的非线性预测模型，但在获取离散离子电流幅值谱的过程中采用了基于傅里叶变换的频谱分析方法，该方法属于时均处理方法，无法获取不同时刻下的实时幅值，进而无法实现对当量比的实时预测。However, the previous ion current sensors all use DC or single-frequency AC excitation for measurement, and a single measurement obtains the curve of the ion current signal with time, which contains limited characteristic information; the multi-frequency excitation method is used to measure the change of ion concentration in pulsating combustion During the measurement, the response characteristics of the ion current between the two stages of the sensor change with the change of the excitation frequency, and the response curve of the ion current signal with the change of the excitation frequency, that is, the discrete ion current spectrum information, can be obtained. For a given external electric field, the characteristics of the ionic current amplitude spectrum depend on the combustion conditions, which are closely related to the gas equivalence ratio. Therefore, the equivalence ratio can be viewed as a function of the ionic current magnitude spectrum, which is inherently nonlinear and can be approximated using nonlinear predictive models. In the previous literature, a nonlinear prediction model of the equivalence ratio and the discrete ion current amplitude spectrum was established, but in the process of obtaining the discrete ion current amplitude spectrum, a spectrum analysis method based on Fourier transform was used, which belongs to the time-averaged processing. method, it is impossible to obtain the real-time amplitude at different times, and thus cannot realize the real-time prediction of the equivalence ratio.

通过上述分析，现有技术存在的问题及缺陷为：Through the above analysis, the existing problems and defects in the prior art are:

(1)现有通过安装在燃烧系统入口上的流量计测量入口燃料和空气流量，进而计算燃气当量比的方法，对于入口流量随时间变化的燃烧系统而言，通过测定入口流量来确定当量比的方法将受到限制。(1) The existing method of measuring the inlet fuel and air flow through a flowmeter installed on the inlet of the combustion system, and then calculating the gas equivalence ratio, for a combustion system in which the inlet flow changes with time, the equivalence ratio is determined by measuring the inlet flow method will be limited.

(2)现有使用安装在排气流中的氧气传感器来测量氧气浓度的方法中，HEGO传感器无法给出当量比的确切值；UEGO传感器虽然能提供准确的当量比值，但价格昂贵，且大多局限于实验室应用。(2) In the existing method for measuring the oxygen concentration using the oxygen sensor installed in the exhaust flow, the HEGO sensor cannot give the exact value of the equivalence ratio; although the UEGO sensor can provide an accurate equivalence ratio, it is expensive and most Limited to laboratory applications.

(3)光学诊断技术测量过程中需要开设光学窗口，在实际应用中受到限制。(3) Optical windows need to be opened in the measurement process of optical diagnostic technology, which is limited in practical application.

(4)以往的离子电流传感器大多采用直流或者单频交流激励测量，单次测量得到是离子电流信号随时间的变化曲线，包含的特征信息有限。(4) Most of the previous ion current sensors use DC or single-frequency AC excitation measurement, and a single measurement obtains the curve of the ion current signal with time, which contains limited characteristic information.

(5)以往的多频激励离子电流传感器采用具有时均特性的傅里叶变换获取离散幅值信息，无法实现对不同时刻下当量比的实时预测。(5) The previous multi-frequency excitation ion current sensor uses Fourier transform with time-average characteristics to obtain discrete amplitude information, which cannot realize real-time prediction of equivalence ratio at different times.

解决以上问题及缺陷的难度为：本发明突破传统方法，创新性地提出多频激励离子传感器及多频激励信号的在线解调，其中多频激励信号的优化及不同时刻下离散幅值谱的获取是难点。The difficulty of solving the above problems and defects is as follows: the present invention breaks through the traditional method and innovatively proposes the multi-frequency excitation ion sensor and the online demodulation of the multi-frequency excitation signal, wherein the optimization of the multi-frequency excitation signal and the analysis of the discrete amplitude spectrum at different times Obtaining is difficult.

解决以上问题及缺陷的意义为：本发明所采用的传感器及预测方法安装方便、环境适应性强、实时性好、成本低，为变流量系统中当量比的实时预测提供了技术支撑。The significance of solving the above problems and defects is that the sensor and prediction method adopted in the present invention are easy to install, have strong environmental adaptability, good real-time performance and low cost, and provide technical support for real-time prediction of equivalence ratio in variable flow systems.

发明内容SUMMARY OF THE INVENTION

针对现有技术存在的问题，本发明提供了一种燃气当量比预测方法、系统及处理终端，尤其涉及一种基于多频激励离子信号幅值谱的燃气当量比预测方法、系统及处理终端。Aiming at the problems existing in the prior art, the present invention provides a gas equivalence ratio prediction method, system and processing terminal, in particular, a gas equivalence ratio prediction method, system and processing terminal based on the amplitude spectrum of multi-frequency excitation ion signals.

本发明是这样实现的，一种燃气当量比预测方法，所述燃气当量比预测方法，包括：利用多频激励离子电流传感器测量得到被测燃烧器不同入口流量下的多频激励离子电流时变信号；对多频激励时变信号进行非整周期在线解调，获得不同时刻下、不同激励频率对应的离子信号振幅，进而得到不同时刻对应的离散离子电流幅值谱；利用主成分分析对不同时刻下离散幅值谱进行特征降维及去噪，得到不同时刻下的降维特征矩阵；将不同时刻下的降维特征矩阵输入到文献中所建立的离子电流幅值谱-燃气当量比非线性预测模型中，最终实现对不同时刻下燃气当量比的实时预测。The present invention is realized in this way, a method for predicting the gas equivalence ratio, the method for predicting the gas equivalence ratio includes: using a multi-frequency excitation ion current sensor to measure and obtain the time-varying multi-frequency excitation ion current under different inlet flow rates of the tested burner signal; perform non-integral period on-line demodulation of multi-frequency excitation time-varying signal to obtain ion signal amplitudes corresponding to different excitation frequencies at different times, and then obtain discrete ion current amplitude spectra corresponding to different times; Perform feature dimension reduction and denoising on discrete amplitude spectra at different times, and obtain dimension reduction characteristic matrices at different times; input the dimension reduction characteristic matrices at different times into the ion current amplitude spectrum-gas equivalence ratio equation established in the literature. In the linear prediction model, the real-time prediction of the gas equivalence ratio at different times is finally realized.

进一步，所述燃气当量比预测方法包括以下步骤：Further, the gas equivalence ratio prediction method includes the following steps:

步骤一，控制多频激励信号源，使产生一个具有平坦功率谱及较优峰值因数CF的多频叠加交流激励信号y(t)，将其施加在离子电流传感器两极间；(积极作用：保证多频离子信号的产生)Step 1: Control the multi-frequency excitation signal source to generate a multi-frequency superimposed AC excitation signal y(t) with a flat power spectrum and a better crest factor CF, and apply it between the two poles of the ion current sensor; (positive effect: ensure that Generation of multi-frequency ion signals)

步骤二，利用交流电桥及测量电路将多频激励下的离子电流时变信号转换为电压时变信号，并输入到数据采集系统进行采集；(积极作用：保证多频离子信号的正常采集)Step 2, using an AC bridge and a measuring circuit to convert the ion current time-varying signal under the multi-frequency excitation into a voltage time-varying signal, and input it to the data acquisition system for collection; (positive effect: ensure the normal collection of the multi-frequency ion signal)

步骤三，对所采集的离子电压信号进行非整周期在线解调，获得不同时刻下的离散离子电流幅值谱S_A^t；利用PCA对不同时刻下的离散离子电流幅值谱特征S_A^t进行特征降维及去噪，得到不同时刻下的降维特征矩阵X_t；(积极作用：保证预测方法的实时性)Step 3: Perform non-integral period online demodulation on the collected ion voltage signal to obtain discrete ion current amplitude spectrum S_A^t at different times; use PCA to analyze the discrete ion current amplitude spectrum characteristics S_A^t at different times Perform feature dimensionality reduction and denoising, and obtain the dimensionality reduction feature matrix X_t at different times; (positive effect: ensure the real-time performance of the prediction method)

步骤四，将降维后的特征矩阵X_t输入到前期所建立的离子电流幅值谱-燃气当量比非线性预测模型中进行燃气当量比Ф的预测，从而最终实现对不同时刻下当量比的实时预测。(积极作用：保证预测的准确性)Step 4: Input the reduced dimensionality matrix X_t into the ion current amplitude spectrum-gas equivalence ratio nonlinear prediction model established in the previous stage to predict the gas equivalence ratio Ф, so as to finally realize the equivalence ratio at different times. Real-time predictions. (positive effect: ensure the accuracy of the forecast)

步骤一，控制多频激励信号源，使产生一个具有平坦功率谱及较优峰值因数(CF)的多频叠加交流激励信号y(t)，将其施加在离子电流传感器两极间；Step 1, control the multi-frequency excitation signal source to generate a multi-frequency superimposed AC excitation signal y(t) with a flat power spectrum and a better crest factor (CF), and apply it between the two poles of the ion current sensor;

步骤二，利用交流电桥及测量电路将多频激励下的离子电流信号转换为电压信号，并输入到数据采集系统进行采集；对所采集的离子电压信号进行频谱分析，获得不同激励频率f_i下的离散离子电流幅值谱S_A；In step 2, the ion current signal under the multi-frequency excitation is converted into a voltage signal by using an AC bridge and a measurement circuit, and is input to the data acquisition system for collection; the collected ion voltage signal is subjected to spectrum analysis to obtain different excitation frequencies f_i . The discrete ion current amplitude spectrum S_A ;

步骤三，获取不同当量比目标值下的离散离子电流幅值谱，组成特征样本矩阵S_A^Ф；利用PCA对离散离子电流幅值谱特征矩阵S_A^Ф进行特征降维及去噪，得到降维后的特征矩阵X，用于后续非线性预测模型的训练；Step 3: Obtain the discrete ion current amplitude spectrum under different equivalence ratio target values, and form a characteristic sample matrix S_A^Ф ; use PCA to perform feature dimension reduction and denoising on the discrete ion current amplitude spectrum characteristic matrix S_A^Ф to obtain a reduced The dimensioned feature matrix X is used for the training of the subsequent nonlinear prediction model;

步骤四，将降维后的特征矩阵X输入到SVM模型中进行燃气当量比Ф的预测，选取一部分样本数据进行模型训练，训练好模型后，对剩余样本进行主成分分析，并将得到的主成分数据作为测试集测试所得预测模型的性能，从而最终获得当量比-离散离子幅值谱间的非线性预测模型。Step 4: Input the dimension-reduced feature matrix X into the SVM model to predict the gas equivalence ratio Ф, select a part of the sample data for model training, and after the model is trained, perform principal component analysis on the remaining samples, and use the obtained principal component analysis. The composition data is used as a test set to test the performance of the obtained prediction model, thereby finally obtaining a nonlinear prediction model between the equivalence ratio-discrete ion amplitude spectrum.

进一步，步骤一中，所述y(t)表示为几个正弦波分量的总和，每个正弦波分量都有各自的幅度、频率及相位，记A_i，ω_i和

分别为第i个谐波分量的幅度，角频率和初始相位，M是多频激励信号中谐波分量的数量，则有：Further, instep 1, the y(t) is expressed as the sum of several sine wave components, each sine wave component has its own amplitude, frequency and phase, denoted A_i , ω_i and

are the amplitude, angular frequency and initial phase of the i-th harmonic component, respectively, and M is the number of harmonic components in the multi-frequency excitation signal, then:

为了优化多频激励信号的峰值因数，使用下式对

进行优化：To optimize the crest factor of the multi-frequency excitation signal, use the following equation for

optimize:

进一步，步骤一中，所述多频叠加交流激励信号施加在离子电流传感器电极间，使得两级间的产生特定的离子电流-频率响应曲线。Further, instep 1, the multi-frequency superimposed AC excitation signal is applied between the electrodes of the ion current sensor, so that a specific ion current-frequency response curve is generated between the two stages.

进一步，步骤二中，时刻点数为N时，所述S_A和S_A^t分别表示为：所述S_A表示为：Further, in step 2, when the number of time points is N, the_SA and_SA^t are respectively expressed as: the_SA is expressed as:

S_A＝{A(f_i)|i＝1,2,...,M},S_A^t＝{S_A(t_k),|k＝1,2,...,N}。S_A ={A(f_i )|i=1,2,...,M}, S_A^t ={S_A (t_k ),|k=1,2,...,N}.

本发明的另一目的在于提供一种应用所述的燃气当量比预测方法的燃气当量比预测系统，所述燃气当量比预测系统包括：多频激励信号源、测量探针、基于交流电桥的测量电路、在线解调电路及数据采集系统。其中，多频激励信号源用于产生多频激励信号，测量探针作为传感器的电极，测量电路将离子电流信号转化为电压信号便于采集，在线解调电路用于多频离子电压信号的实时解调，数据采集系统用于解调后幅值数据的采集与保存。Another object of the present invention is to provide a gas equivalence ratio prediction system using the gas equivalence ratio prediction method. The gas equivalence ratio prediction system includes: a multi-frequency excitation signal source, a measurement probe, and an AC bridge-based measurement Circuit, online demodulation circuit and data acquisition system. Among them, the multi-frequency excitation signal source is used to generate multi-frequency excitation signals, the measurement probe is used as the electrode of the sensor, the measurement circuit converts the ion current signal into a voltage signal for easy acquisition, and the online demodulation circuit is used for real-time solution of the multi-frequency ion voltage signal. The data acquisition system is used to collect and save the amplitude data after demodulation.

进一步，所述燃气当量比预测系统，还包括：Further, the gas equivalence ratio prediction system also includes:

利用多频激励信号源产生一个具有平坦功率谱及较优峰值因数的交流分量叠加激励信号，并将所述信号施加在离子电流传感器测量探针的两极间；A multi-frequency excitation signal source is used to generate an AC component superimposed excitation signal with a flat power spectrum and a better crest factor, and the signal is applied between the two poles of the measuring probe of the ion current sensor;

利用交流电桥及测量电路将多频激励下的离子电流信号转换为电压信号，并输入到数据采集系统进行采集；对采集的多频激励离子电压时变信号进行在线非整周期解调，获得不同时刻下的离散离子电流幅值谱；The ion current signal under the multi-frequency excitation is converted into a voltage signal by using an AC bridge and a measurement circuit, and then input to the data acquisition system for acquisition; the collected multi-frequency excitation ion voltage time-varying signal is demodulated on-line and aperiodic to obtain different The discrete ion current amplitude spectrum at time;

利用PCA对不同时刻的离散离子电流幅值谱特征进行降维及去噪，降维后的特征数据最终输入到前期文献建立的非线性预测模型中进行燃气当量比的实时预测。PCA is used to reduce the dimension and denoise the spectral features of discrete ion current amplitudes at different times. The feature data after dimension reduction is finally input into the nonlinear prediction model established in the previous literature for real-time prediction of gas equivalence ratio.

本发明的另一目的在于提供一种计算机设备，所述计算机设备包括存储器和处理器，所述存储器存储有计算机程序，所述计算机程序被所述处理器执行时，使得所述处理器执行如下步骤：Another object of the present invention is to provide a computer device, the computer device includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the following step:

利用多频激励离子电流传感器测量得到被测燃烧器不同入口流量下的多频激励离子电流信号；Using the multi-frequency excitation ion current sensor to measure the multi-frequency excitation ion current signal under different inlet flow rates of the tested burner;

对得到的多频激励离子信号进行在线非整周期解调，获得不同时刻下的离散离子电流幅值谱；Perform on-line aperiodic demodulation of the obtained multi-frequency excitation ion signal to obtain discrete ion current amplitude spectra at different times;

利用主成分分析PCA来减小不同时刻离子电流幅值谱的数据维数和噪声；Principal component analysis (PCA) is used to reduce the data dimension and noise of the ion current amplitude spectrum at different times;

降维后的特征数据最终输入到前期文献所建立的支持向量机SVM模型中实现对燃气当量比的实时预测。The feature data after dimension reduction is finally input into the support vector machine SVM model established in the previous literature to realize real-time prediction of gas equivalence ratio.

本发明的另一目的在于提供一种计算机可读存储介质，存储有计算机程序，所述计算机程序被处理器执行时，使得所述处理器执行如下步骤：Another object of the present invention is to provide a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, causes the processor to perform the following steps:

利用多频激励离子电流传感器测量得到被测燃烧器不同入口流量下的多频激励离子电流信号；Using the multi-frequency excitation ion current sensor to measure the multi-frequency excitation ion current signal under different inlet flow rates of the tested burner;

对得到的多频激励离子信号进行在线非整周期解调，获得不同时刻下的离散离子电流幅值谱；Perform on-line aperiodic demodulation of the obtained multi-frequency excitation ion signal to obtain discrete ion current amplitude spectra at different times;

利用主成分分析PCA来减小不同时刻离子电流幅值谱的数据维数和噪声；Principal component analysis (PCA) is used to reduce the data dimension and noise of the ion current amplitude spectrum at different times;

降维后的特征数据最终输入到前期文献所建立的支持向量机SVM模型中实现对燃气当量比的实时预测。The feature data after dimension reduction is finally input into the support vector machine SVM model established in the previous literature to realize real-time prediction of gas equivalence ratio.

本发明的另一目的在于提供一种信息数据处理终端，所述信息数据处理终端用于实现所述的燃气当量比预测系统。Another object of the present invention is to provide an information data processing terminal for implementing the gas equivalence ratio prediction system.

结合上述的所有技术方案，本发明所具备的优点及积极效果为：本发明提供的燃气当量比预测方法，可应用于各类入口流量变化的燃烧器中燃气当量比的预测。本发明首先利用多频激励离子电流传感器测量得到被测燃烧器不同入口流量下的多频激励离子电流时变信号；其次对多频激励时变信号进行非整周期在线解调，获得不同时刻下、不同激励频率对应的离子信号振幅，进而得到不同时刻对应的离散离子电流幅值谱；利用主成分分析对不同时刻下离散幅值谱进行特征降维及去噪，得到不同时刻下的降维特征矩阵；将不同时刻下的降维特征矩阵输入到文献中所建立的离子电流幅值谱-燃气当量比非线性预测模型中，最终实现对不同时刻下燃气当量比的实时预测。Combining all the above technical solutions, the advantages and positive effects of the present invention are: the gas equivalence ratio prediction method provided by the present invention can be applied to the prediction of the gas equivalence ratio in various types of burners with varying inlet flow rates. The invention firstly uses the multi-frequency excitation ion current sensor to measure the multi-frequency excitation ion current time-varying signal under different inlet flow rates of the burner to be tested; , the ion signal amplitudes corresponding to different excitation frequencies, and then obtain the discrete ion current amplitude spectra corresponding to different times; use principal component analysis to perform feature dimension reduction and denoising on the discrete amplitude spectra at different times, and obtain the dimension reduction at different times. Characteristic matrix; input the dimensionality reduction characteristic matrix at different times into the nonlinear prediction model of ion current amplitude spectrum-gas equivalence ratio established in the literature, and finally realize the real-time prediction of the gas equivalence ratio at different times.

同时，本发明利用多频激励信号激励离子电流传感器，与传统的直流或者单频交流激励方式相比，多频激励方式单次测量得到的幅值谱特征信息丰富，同时，结合非整周期在线解调及非线性预测模型，有利于准确刻画工况参数的实时变化，并减少了实验次数，在变流量燃烧系统燃气当量比的测定中具有重要的使用价值和广阔的应用前景。At the same time, the present invention uses the multi-frequency excitation signal to excite the ion current sensor. Compared with the traditional DC or single-frequency AC excitation methods, the multi-frequency excitation method can obtain rich amplitude spectrum characteristic information obtained by a single measurement. The demodulation and nonlinear prediction model is beneficial to accurately describe the real-time changes of operating parameters, and reduces the number of experiments.

附图说明Description of drawings

为了更清楚地说明本发明实施例的技术方案，下面将对本发明实施例中所需要使用的附图做简单的介绍，显而易见地，下面所描述的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下还可以根据这些附图获得其他的附图。In order to illustrate the technical solutions of the embodiments of the present invention more clearly, the following will briefly introduce the accompanying drawings that need to be used in the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without creative effort.

图1是本发明实施例提供的燃气当量比预测方法流程图。FIG. 1 is a flowchart of a method for predicting a gas equivalence ratio provided by an embodiment of the present invention.

图2是本发明实施例提供的测量系统图。FIG. 2 is a diagram of a measurement system provided by an embodiment of the present invention.

图3是本发明实施例提供的施加在离子电流两极间的多频激励信号示意图。FIG. 3 is a schematic diagram of a multi-frequency excitation signal applied between two poles of an ion current according to an embodiment of the present invention.

图4是本发明实施例提供的测量得到的时域离子信号示意图。FIG. 4 is a schematic diagram of a time-domain ion signal obtained by measurement according to an embodiment of the present invention.

图5是本发明实施例提供的不同时刻对应的离散幅值谱曲线示意图。FIG. 5 is a schematic diagram of discrete amplitude spectrum curves corresponding to different times provided by an embodiment of the present invention.

图6是本发明实施例提供的燃气当量比实时预测结果示意图。FIG. 6 is a schematic diagram of a real-time prediction result of a gas equivalence ratio provided by an embodiment of the present invention.

具体实施方式Detailed ways

为了使本发明的目的、技术方案及优点更加清楚明白，以下结合实施例，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.

针对现有技术存在的问题，本发明提供了一种燃气当量比预测方法、系统及处理终端，下面结合附图对本发明作详细的描述。In view of the problems existing in the prior art, the present invention provides a method, system and processing terminal for predicting the gas equivalence ratio. The present invention will be described in detail below with reference to the accompanying drawings.

如图1所示，本发明实施例提供的燃气当量比预测方法包括以下步骤：As shown in FIG. 1 , the gas equivalence ratio prediction method provided by the embodiment of the present invention includes the following steps:

S101：控制多频激励信号源，使产生一个具有平坦功率谱及较优峰值因数CF的多频叠加交流激励信号y(t)，将其施加在离子电流传感器两极间；S101: Control the multi-frequency excitation signal source to generate a multi-frequency superimposed AC excitation signal y(t) with a flat power spectrum and a better crest factor CF, and apply it between the two poles of the ion current sensor;

S102：利用交流电桥及测量电路将多频激励下的离子电流时变信号转换为电压时变信号，并输入到数据采集系统进行采集；S102: using an AC bridge and a measuring circuit to convert the ion current time-varying signal under multi-frequency excitation into a voltage time-varying signal, and input it to the data acquisition system for acquisition;

S103：对所采集的离子电压信号进行非整周期在线解调，获得不同时刻下的离散离子电流幅值谱S_A^t；利用PCA对不同时刻下的离散离子电流幅值谱特征S_A^t进行特征降维及去噪，得到不同时刻下的降维特征矩阵X_t；S103: Perform non-integral period online demodulation on the collected ion voltage signal to obtain discrete ion current amplitude spectrum S_A^t at different times; use PCA to perform discrete ion current amplitude spectrum characteristics S_A^t at different times Feature dimensionality reduction and denoising to obtain dimensionality reduction feature matrix X_t at different times;

S104：将降维后的特征矩阵X_t输入到前期所建立的离子电流幅值谱-燃气当量比非线性预测模型中进行燃气当量比Ф的预测，从而最终实现对不同时刻下当量比的实时预测。S104: Input the dimensionality-reduced feature matrix X_t into the ion current amplitude spectrum-gas equivalence ratio nonlinear prediction model established in the previous stage to predict the gas equivalence ratio Ф, so as to finally realize the real-time equivalence ratio at different times. predict.

下面结合实施例对本发明的技术方案作进一步描述。The technical solutions of the present invention will be further described below in conjunction with the embodiments.

实施例1Example 1

本发明提出一种基于多频激励离子信号幅值谱的当量比预测方法，利用多频激励离子电流传感器测量得到不同燃气供应压力下的多频激励离子电流信号；对得到的多频激励离子信号进行在线解调，获得不同时刻、不同激励频率下的离散离子电流幅值谱；此后，利用PCA对不同时刻下的离子电流幅值谱进行特征降维及去噪；降维后的特征数据最终输入到非线性预测模型中进行燃气当量比的实时预测。The invention proposes an equivalence ratio prediction method based on the amplitude spectrum of the multi-frequency excitation ion signal, and uses the multi-frequency excitation ion current sensor to measure the multi-frequency excitation ion current signals under different gas supply pressures; Perform online demodulation to obtain discrete ion current amplitude spectra at different times and different excitation frequencies; after that, PCA is used to perform feature dimension reduction and denoising on the ion current amplitude spectra at different times; the feature data after dimension reduction is finally Input into nonlinear prediction model for real-time prediction of gas equivalence ratio.

本发明提出的基于多频激励离子信号幅值谱的当量比预测方法中，所用元件包括：多频激励信号源、测量探针、基于交流电桥的测量电路、多频信号解调电路及数据采集模块。其中，多频激励信号源用于产生多频激励信号，测量探针作为传感器的电极，测量电路将离子电流信号转化为电压信号便于采集，在线解调电路用于多频离子电压信号的实时解调，数据采集系统用于解调后幅值数据的采集与保存。In the equivalence ratio prediction method based on the amplitude spectrum of the multi-frequency excitation ion signal proposed by the present invention, the components used include: a multi-frequency excitation signal source, a measurement probe, a measurement circuit based on an AC bridge, a multi-frequency signal demodulation circuit and data acquisition. module. Among them, the multi-frequency excitation signal source is used to generate multi-frequency excitation signals, the measurement probe is used as the electrode of the sensor, the measurement circuit converts the ion current signal into a voltage signal for easy acquisition, and the online demodulation circuit is used for real-time solution of the multi-frequency ion voltage signal. The data acquisition system is used to collect and save the amplitude data after demodulation.

本发明利用多频激励信号源产生一个具有平坦功率谱及较优峰值因数的交流分量叠加激励信号，并将其施加在离子电流传感器测量探针的两极间。利用交流电桥测量电路测量得到多频激励下离子电流所对应的电压信号，并对得到的多频激励离子电压信号进行在线解调，获得不同时刻、不同激励频率下的离散离子电流幅值谱；利用PCA对离散离子电流幅值谱所组成的特征矩阵进行降维及去噪；降维后的特征数据最终输入到非线性预测模型中进行燃气当量比的实时预测。The invention utilizes a multi-frequency excitation signal source to generate an AC component superimposed excitation signal with a flat power spectrum and a better crest factor, and applies it between the two poles of the measuring probe of the ion current sensor. The voltage signal corresponding to the ion current under the multi-frequency excitation is measured by the AC bridge measurement circuit, and the obtained multi-frequency excitation ion voltage signal is online demodulated to obtain the discrete ion current amplitude spectrum at different times and different excitation frequencies; PCA is used to reduce the dimension and denoise the characteristic matrix composed of the discrete ion current amplitude spectrum; the dimension-reduced characteristic data is finally input into the nonlinear prediction model for real-time prediction of the gas equivalence ratio.

本发明利用多频激励信号激励离子电流传感器，与传统的直流或者单频交流激励方式相比，多频激励方式单次测量得到的幅值谱特征信息丰富，结合非整周期在线解调及非线性预测模型，有利于准确刻画工况参数的实时变化，有利于准确刻画工况参数的变化，并减少了实验次数。在变流量燃烧系统的燃气当量比测定中具有很好的应用前景。The invention uses multi-frequency excitation signal to excite the ion current sensor. Compared with the traditional direct current or single-frequency alternating current excitation method, the multi-frequency excitation method can obtain rich amplitude spectrum characteristic information obtained by a single measurement. The linear prediction model is beneficial to accurately describe the real-time changes of the parameters of the working conditions, which is beneficial to accurately describe the changes of the parameters of the working conditions, and reduces the number of experiments. It has a good application prospect in the determination of gas equivalence ratio of variable flow combustion system.

实施例2Example 2

在图2中，利用多频激励信号源产生一个具有平坦功率谱及较优峰值因数的多频叠加交流激励信号，并将其施加在离子电流传感器测量探针的电极间。利用交流电桥及测量电路将多频激励下的离子电流信号转换为电压信号，并输入到数据采集系统进行采集；对多频激励时变信号进行非整周期在线解调，获得不同时刻下、不同激励频率对应的离子信号振幅，进而得到不同时刻对应的离散离子电流幅值谱。In Figure 2, a multi-frequency excitation signal source is used to generate a multi-frequency superimposed AC excitation signal with a flat power spectrum and a better crest factor, and it is applied between the electrodes of the ion current sensor measurement probe. The ion current signal under the multi-frequency excitation is converted into a voltage signal by using an AC bridge and a measuring circuit, and is input to the data acquisition system for acquisition; the multi-frequency excitation time-varying signal is demodulated on-line with a non-integral period to obtain different time and different conditions. The ion signal amplitude corresponding to the excitation frequency was obtained, and then the discrete ion current amplitude spectrum corresponding to different times was obtained.

步骤一：控制多频激励信号源，使产生一个具有平坦功率谱及较优峰值因数(CF)的多频叠加交流激励信号y(t)(如图3所示)，将其施加在离子电流传感器两极间。y(t)可以表示为几个正弦波分量的总和，每个正弦波分量都有各自的幅度、频率及相位，记A_i，ω_i和

分别为第i个谐波分量的幅度，角频率和初始相位，M是多频激励信号中谐波分量的数量，则有：Step 1: Control the multi-frequency excitation signal source to generate a multi-frequency superimposed AC excitation signal y(t) with a flat power spectrum and a better crest factor (CF) (as shown in Figure 3), and apply it to the ion current. between the sensor poles. y(t) can be expressed as the sum of several sine wave components, each sine wave component has its own amplitude, frequency and phase, denoted A_i , ω_i and

are the amplitude, angular frequency and initial phase of the i-th harmonic component, respectively, and M is the number of harmonic components in the multi-frequency excitation signal, then:

为了优化多频激励信号的峰值因数，使用下式对

进行优化：To optimize the crest factor of the multi-frequency excitation signal, use the following equation for

optimize:

步骤二：利用交流电桥及测量电路将多频激励下的离子电流信号转换为电压信号，并输入到数据采集系统进行采集；对所采集的离子电压信号(如图4所示)进行实时在线解调，获得不同时刻下、不同激励频率f_i下的离散离子电流幅值谱S_A^t(如图5所示)，S_A和S_A^t可记为：Step 2: Convert the ion current signal under multi-frequency excitation into a voltage signal by using an AC bridge and a measuring circuit, and input it to the data acquisition system for collection; perform a real-time online solution of the collected ion voltage signal (as shown in Figure 4). The discrete ion current amplitude spectrum S_A^t (as shown in Figure 5) at different times and different excitation frequencies f_i is obtained. S_A and S_A^t can be recorded as:

S_A＝{A(f_i)|i＝1,2,...,M} (3)S_A ={A(f_i )|i=1,2,...,M} (3)

S_A^t＝{S_A(t_k)|k＝1,2,...,N} (4)S_A^t = {S_A (t_k )|k=1,2,...,N} (4)

步骤三：利用PCA对离散离子电流幅值谱特征矩阵S_A^t进行特征降维及去噪，得到降维后的特征矩阵X_t。Step 3: Use PCA to perform feature dimension reduction and denoising on the discrete ion current amplitude spectrum characteristic matrix S_A^t to obtain a dimension-reduced characteristic matrix X_t .

步骤四：将降维后的特征数据X_t输入到前期建立的当量比-离散幅值谱非线性预测模型中进行燃气当量比Ф的实时预测。Step 4: Input the dimension-reduced feature data X_t into the equivalence ratio-discrete amplitude spectrum nonlinear prediction model established in the previous stage to perform real-time prediction of the gas equivalence ratio Ф.

本发明实施例提供的燃气当量比实时预测结果如图6所示。The real-time prediction result of the gas equivalence ratio provided by the embodiment of the present invention is shown in FIG. 6 .

在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用全部或部分地以计算机程序产品的形式实现，所述计算机程序产品包括一个或多个计算机指令。在计算机上加载或执行所述计算机程序指令时，全部或部分地产生按照本发明实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(DSL)或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输)。所述计算机可读取存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质，(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或者半导体介质(例如固态硬盘SolidState Disk(SSD))等。In the above-mentioned embodiments, it may be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented in whole or in part in the form of a computer program product, the computer program product includes one or more computer instructions. When the computer program instructions are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer may be a general purpose computer, special purpose computer, computer network, or other programmable device. The computer instructions may be stored in or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions may be downloaded from a website site, computer, server or data center Transmission to another website site, computer, server, or data center by wire (eg, coaxial cable, fiber optic, digital subscriber line (DSL), or wireless (eg, infrared, wireless, microwave, etc.)). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, a data center, or the like that includes an integration of one or more available media. The usable media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVD), or semiconductor media (eg, Solid State Disk (SSD)), among others.

以上所述，仅为本发明的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，都应涵盖在本发明的保护范围之内。The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited to this. Any person skilled in the art is within the technical scope disclosed by the present invention, and all within the spirit and principle of the present invention Any modifications, equivalent replacements and improvements made within the scope of the present invention should be included within the protection scope of the present invention.

Claims (9)

Translated fromChinese

1.一种燃气当量比预测方法，其特征在于，所述燃气当量比预测方法利用多频激励离子电流传感器测量得到被测燃烧器不同入口流量下的多频激励离子电流时变信号；对多频激励时变信号进行非整周期在线解调，获得不同时刻下、不同激励频率对应的离子信号振幅，进而得到不同时刻对应的离散离子电流幅值谱；利用主成分分析对不同时刻下离散幅值谱进行特征降维及去噪，得到不同时刻下的降维特征矩阵；将不同时刻下的降维特征矩阵输入到所建立的离子电流幅值谱-燃气当量比非线性预测模型中，最终实现对不同时刻下燃气当量比的实时预测。1. a gas equivalence ratio prediction method, it is characterized in that, described gas equivalence ratio prediction method utilizes multi-frequency excitation ion current sensor to measure and obtain the multi-frequency excitation ion current time-varying signal under different inlet flow rates of the burner under test; The frequency excitation time-varying signal is demodulated on-line with a non-integral period, and the ion signal amplitude corresponding to different excitation frequencies at different times is obtained, and then the discrete ion current amplitude spectrum corresponding to different times is obtained. The feature dimensionality reduction and denoising of the value spectrum are performed to obtain the dimensionality reduction feature matrix at different times; the dimensionality reduction feature matrix at different times is input into the established ion current amplitude spectrum-gas equivalence ratio nonlinear prediction model, and finally Real-time prediction of gas equivalence ratio at different times.2.如权利要求1所述的燃气当量比预测方法，其特征在于，所述燃气当量比预测方法包括以下步骤：2. The gas equivalence ratio prediction method according to claim 1, wherein the gas equivalence ratio prediction method comprises the following steps:步骤一，控制多频激励信号源，使产生一个具有平坦功率谱及较优峰值因数CF的多频叠加交流激励信号y(t)，将其施加在离子电流传感器两极间；Step 1: Control the multi-frequency excitation signal source to generate a multi-frequency superimposed AC excitation signal y(t) with a flat power spectrum and a better crest factor CF, and apply it between the two poles of the ion current sensor;步骤二，利用交流电桥及测量电路将多频激励下的离子电流时变信号转换为电压时变信号，并输入到数据采集系统进行采集；Step 2, using the AC bridge and the measuring circuit to convert the ion current time-varying signal under the multi-frequency excitation into a voltage time-varying signal, and input it to the data acquisition system for collection;步骤三，对所采集的离子电压信号进行非整周期在线解调，获得不同时刻下的离散离子电流幅值谱S_A^t；利用PCA对不同时刻下的离散离子电流幅值谱S_A^t进行特征降维及去噪，得到不同时刻下的降维特征矩阵X_t；时刻点数为N时，所述S_A和S_A^t分别表示为：Step 3: Perform non-integral period online demodulation on the collected ion voltage signal to obtain discrete ion current amplitude spectrum S_A^t at different times; use PCA to perform discrete ion current amplitude spectrum S_A^t at different times. Feature dimensionality reduction and denoising, to obtain the dimensionality reduction feature matrix X_t at different times; when the number of time points is N, the S_A and S_A^t are respectively expressed as:S_A＝{A(f_i)|i＝1,2,...,M}，S_A^t＝{S_A(t_k)|k＝1，2，...，N}；S_A ={A(f_i )|i=1,2,...,M}, S_A^t ={S_A (t_k )|k=1,2,...,N};步骤四，将降维后的特征矩阵X_t输入到前期所建立的离子电流幅值谱-燃气当量比非线性预测模型中进行燃气当量比Ф的预测，从而最终实现对不同时刻下当量比的实时预测。Step 4: Input the reduced dimensionality matrix X_t into the ion current amplitude spectrum-gas equivalence ratio nonlinear prediction model established in the previous stage to predict the gas equivalence ratio Ф, so as to finally realize the equivalence ratio at different times. Real-time predictions.3.如权利要求2所述的燃气当量比预测方法，其特征在于，步骤一中，所述y(t)表示为几个正弦波分量的总和，每个正弦波分量都有各自的幅度、频率及相位，记A_i，ω_i和

分别为第i个谐波分量的幅度，角频率和初始相位，M是多频激励信号中谐波分量的数量，则有：3. The gas equivalence ratio prediction method according to claim 2, wherein in step 1, the y(t) is expressed as the sum of several sine wave components, and each sine wave component has its own amplitude, frequency and phase, denote A_i , ω_i and

are the amplitude, angular frequency and initial phase of the i-th harmonic component, respectively, and M is the number of harmonic components in the multi-frequency excitation signal, then:

为了优化多频激励信号的峰值因数，使用下式对

进行优化：To optimize the crest factor of the multi-frequency excitation signal, use the following equation for

optimize:

4.如权利要求2所述的燃气当量比预测方法，其特征在于，步骤一中，所述多频叠加交流激励信号施加在离子电流传感器电极间，使得两级间的产生特定的离子电流-频率响应曲线。4. The gas equivalence ratio prediction method according to claim 2, wherein in step 1, the multi-frequency superimposed AC excitation signal is applied between the electrodes of the ion current sensor, so that a specific ion current - frequency response curve.5.一种实施权利要求1～4任意一项所述的燃气当量比预测方法的燃气当量比预测系统，其特征在于，所述燃气当量比预测系统包括：多频激励信号源、测量探针、基于交流电桥的测量电路、在线解调模块及数据采集系统；其中，多频激励信号源用于产生多频激励信号，测量探针作为传感器的电极，测量电路将离子电流信号转化为电压信号便于采集，在线解调电路用于多频离子电压信号的实时解调，数据采集系统用于解调后幅值数据的采集与保存。5. A gas equivalence ratio prediction system for implementing the gas equivalence ratio prediction method according to any one of claims 1 to 4, wherein the gas equivalence ratio prediction system comprises: a multi-frequency excitation signal source, a measurement probe , AC bridge-based measurement circuit, online demodulation module and data acquisition system; wherein, the multi-frequency excitation signal source is used to generate multi-frequency excitation signals, the measurement probe is used as the electrode of the sensor, and the measurement circuit converts the ionic current signal into a voltage signal Easy to collect, the online demodulation circuit is used for real-time demodulation of multi-frequency ion voltage signals, and the data acquisition system is used to collect and save the amplitude data after demodulation.6.如权利要求5所述的燃气当量比预测系统，其特征在于，所述燃气当量比预测系统，还包括：利用多频激励信号源产生一个具有平坦功率谱及较优峰值因数的交流分量叠加激励信号，并将所述信号施加在离子电流传感器测量探针的两极间；6. The gas equivalence ratio prediction system according to claim 5, wherein the gas equivalence ratio prediction system further comprises: using a multi-frequency excitation signal source to generate an AC component with a flat power spectrum and a better crest factor superimposing the excitation signal and applying the signal between the two poles of the measuring probe of the ion current sensor;利用交流电桥及测量电路将多频激励下的离子电流信号转换为电压信号，并输入到数据采集系统进行采集；对采集的多频激励离子电压时变信号进行在线非整周期解调，获得不同时刻下的离散离子电流幅值谱；时刻点数为N时，S_A和S_A^t分别表示为：The ion current signal under the multi-frequency excitation is converted into a voltage signal by using an AC bridge and a measurement circuit, and then input to the data acquisition system for acquisition; the collected multi-frequency excitation ion voltage time-varying signal is demodulated on-line and aperiodic to obtain different The discrete ion current amplitude spectrum at time; when the number of time points is N, S_A and S_A^t are respectively expressed as:S_A＝{A(f_i)|i＝1,2,...,M}，S_A^t＝{S_A(t_k)|k＝1，2，...，N}；S_A ={A(f_i )|i=1,2,...,M}, S_A^t ={S_A (t_k )|k=1,2,...,N};其中，S_A^t为不同时刻下的离散离子电流幅值谱，利用PCA对不同时刻的离散离子电流幅值谱特征进行降维及去噪，降维后的特征数据最终输入到建立的非线性预测模型中进行燃气当量比的实时预测。Among them, S_A^t is the discrete ion current amplitude spectrum at different times. PCA is used to reduce the dimension and denoise the characteristics of the discrete ion current amplitude spectrum at different times. The feature data after dimension reduction is finally input to the established nonlinear Real-time prediction of gas equivalence ratio in the prediction model.7.一种计算机设备，其特征在于，所述计算机设备包括存储器和处理器，所述存储器存储有计算机程序，所述计算机程序被所述处理器执行时，使得所述处理器执行如下步骤：7. A computer device, characterized in that the computer device comprises a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to perform the following steps:利用多频激励离子电流传感器测量得到被测燃烧器不同入口流量下的多频激励离子电流信号；Using the multi-frequency excitation ion current sensor to measure the multi-frequency excitation ion current signal under different inlet flow rates of the tested burner;对得到的多频激励离子信号进行在线非整周期解调，获得不同时刻下的离散离子电流幅值谱；时刻点数为N时，S_A和S_A^t分别表示为：Perform on-line aperiodic demodulation of the obtained multi-frequency excitation ion signal to obtain discrete ion current amplitude spectra at different times; when the number of time points is N, S_A and S_A^t are respectively expressed as:S_A＝{A(f_i)|i＝1,2,...,M}，S_A^t＝{S_A(t_k)|k＝1，2，...，N}；S_A ={A(f_i )|i=1,2,...,M}, S_A^t ={S_A (t_k )|k=1,2,...,N};其中，S_A^t为不同时刻下的离散离子电流幅值谱，利用主成分分析PCA来减小不同时刻离子电流幅值谱的数据维数和噪声；Among them, S_A^t is the discrete ion current amplitude spectrum at different times, and PCA is used to reduce the data dimension and noise of the ion current amplitude spectrum at different times;降维后的特征数据最终输入到支持向量机SVM模型中实现对燃气当量比的实时预测。The feature data after dimension reduction is finally input into the support vector machine SVM model to realize real-time prediction of gas equivalence ratio.8.一种计算机可读存储介质，存储有计算机程序，所述计算机程序被处理器执行时，使得所述处理器执行如下步骤：8. A computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the following steps:利用多频激励离子电流传感器测量得到被测燃烧器不同入口流量下的多频激励离子电流信号；Using the multi-frequency excitation ion current sensor to measure the multi-frequency excitation ion current signal under different inlet flow rates of the tested burner;对得到的多频激励离子信号进行在线非整周期解调，获得不同时刻下的离散离子电流幅值谱；时刻点数为N时，S_A和S_A^t分别表示为：Perform on-line aperiodic demodulation of the obtained multi-frequency excitation ion signal to obtain discrete ion current amplitude spectra at different times; when the number of time points is N, S_A and S_A^t are respectively expressed as:S_A＝{A(f_i)|i＝1,2,...,M}，S_A^t＝{S_A(t_k)|k＝1，2，...，N}；S_A ={A(f_i )|i=1,2,...,M}, S_A^t ={S_A (t_k )|k=1,2,...,N};其中，S_A^t为不同时刻下的离散离子电流幅值谱，利用主成分分析PCA来减小不同时刻离子电流幅值谱的数据维数和噪声；Among them, S_A^t is the discrete ion current amplitude spectrum at different times, and PCA is used to reduce the data dimension and noise of the ion current amplitude spectrum at different times;降维后的特征数据最终输入到所建立的支持向量机SVM模型中实现对燃气当量比的实时预测。The feature data after dimension reduction is finally input into the established support vector machine SVM model to realize real-time prediction of gas equivalence ratio.9.一种信息数据处理终端，其特征在于，所述信息数据处理终端用于实现如权利要求5～6任意一项所述的燃气当量比实时预测系统。9 . An information data processing terminal, characterized in that, the information data processing terminal is used to implement the real-time prediction system for gas equivalence ratio according to any one of claims 5 to 6 .

Images