Syngas kinetics inside steady Perfectly Stirred Reactor

<p>We present the dataset and python scripts used in our&nbsp;autoencoder (AE) neural network (NN)-based reduced chemistry work (<a href="https://www.dl.begellhouse.com/journals/558048804a15188a,22e553d25a1b5ff0,65e7b0537eb64be1.html">Zhang and Sankaran, 2022</a>).</p> <p>1. To train the AE NN, run&nbsp;&quot;<strong>python Train_PSR_AE_PCA.py</strong> &quot; with Keras&nbsp;</p> <p>2.&nbsp;The dataset is about syngas combustion inside 0-D steady perfectly stirred reactor (PSR) at a wide range of parameter conditions.</p> <ol> <li>The fuel is CO, H2, N2 with a volume ratio 5:1:4. The oxidizer is O2 and N2 mixed in 1:3 by volume. The inflow temperature is 500 K and combustion occurs at atmospheric pressure.</li> <li>In total, there are 1.63 million samples with equivalence ratio varying from 0.09 to 20.0 and residence time scale varying to cover the entire S-curve.</li> <li>The dataset is in hdf5 format and can be loaded with the python script,&nbsp;<strong>load_data_h5.py</strong>. Inside the dataset, there 12 entries.</li> </ol> <ul> <li>1. asciiListtmp = h5f[&#39;vars_name&#39;][()] ##name of the 12 thermochemical state variables</li> <li>2. para_Phi = h5f[&#39;parameters_Phi&#39;][()] ##equivalence ratio, varying from 0.09 to 20.0</li> <li>3. para_Tin = h5f[&#39;parameters_Tin&#39;][()] ##inflow temperature, constant=500 [K]</li> <li>4. para_inv_tau_res = h5f[&#39;parameters_inv_tau_res&#39;][()] ##inverse of residence time, varying from 4.53e-09 to 1.54e+04 [1/s]</li> <li>5. x_train_min = h5f[&#39;trainset_min&#39;][()] ##minimum value of training set</li> <li>6. x_train_max = h5f[&#39;trainset_max&#39;][()] ##maximum value of training set</li> <li>7. x_data = h5f[&#39;dataset&#39;][()] ##Thermochemical state variables (temperature, mass fractions of chemical species), normalized with x_train_min and x_train_max to be [-1,1]</li> <li>8. x_src_data = h5f[&#39;dataset_src&#39;][()] ##source term * 2/(x_train_max-x_train_min)</li> <li>9. x_del_data = h5f[&#39;dataset_del&#39;][()] ##(xinflow-x)* 2/(x_train_max-x_train_min)</li> <li>10. train_ind = h5f[&#39;train_dataset_indices&#39;][()] #0.7, sample index of training set&nbsp;</li> <li>11. test_ind = h5f[&#39;test_dataset_indices&#39;][()] #0.3*0.5, sample index of test set</li> <li>12. vali_ind = h5f[&#39;valid_dataset_indices&#39;][()] #0.3*0.5, sample index of validation set</li> <li>The training/test/validation splitting is used in our reduced chemistry work.&nbsp;</li> </ul> <p>&nbsp;</p>