Prediction of gas‐liquid‐solid product distribution after solid waste pyrolysis process based on artificial neural network model

Abstract

Four solid wastes including sludge, watermelon rind, corncob, and eucalyptus and their demineralized samples were selected to conduct pyrolysis experiments under different experimental conditions, including temperature, residence time, carrier gas flow rate, and heating rate, respectively. The copyrolysis experiment was carried out after mixing different types and different ratios of solid wastes to investigate the influence of different factors on yield of char, tar, and gas. A three-layer artificial neural network (ANN) based on back propagation (BP) algorithm was developed and trained to simulate and predict the yield of products. The experimental conditions and characteristic parameters of samples, including the content of C, H, K, Ca, Mg, Fe, volatile, ash, and fixed carbon, were selected as input parameters while the yields of char, tar, and gas were selected as output parameters. The effect of input parameters including residence time, carrier gas flow rate, heating rate, and the content of metal elements on the network performance was investigated in detail to optimize the ANN model. It was found that the metal element has the greatest influence, and its importance to the yield of each product exceeds 25%. The pyrolysis experimental data of single-component solid waste were selected as the model training set for model learning and training and then use the trained model to predict the co-pyrolysis experimental data of multicomponent solid waste. Good agreement was achieved between experimental and predicted results; the correlation coefficient was 0.9836, and the root mean

A Critical Review of Data Science Applications in Resource Recovery and Carbon Capture from Organic Waste

“…Other studies compared between MLR and MPR to predict bio-oil and biochar yield from the dry feedstocks using secondary data with previously mentioned model inputs. , Studies using secondary data (with operational parameters and feedstock properties) further demonstrated that MLR models of pyrolysis were generally outperformed by different ML models, namely ANN, RT, XGBoost, RFR, , and SVR . Among the different ML models of pyrolysis, ANN and ANFIS were developed with operational parameters and feedstock properties using both primary data and secondary data. − In comparative studies, ANNs developed with just operational parameters of pyrolysis seemingly provided better predictions of bio-oil and biochar yield than MPR when using secondary data. , SVRs developed with secondary data representing operational parameters and feedstock properties provided relatively improved performance compared to the ANNs for both dry and wet feedstocks (animal waste, agricultural and forestry residue). − …”

“… 227 Among the different ML models of pyrolysis, ANN and ANFIS were developed with operational parameters and feedstock properties using both primary data and secondary data. 229 − 234 In comparative studies, ANNs developed with just operational parameters of pyrolysis seemingly provided better predictions of bio-oil and biochar yield than MPR when using secondary data. 235 , 236 SVRs developed with secondary data representing operational parameters and feedstock properties provided relatively improved performance compared to the ANNs for both dry and wet feedstocks (animal waste, agricultural and forestry residue).…”

Artificial Neural Networks for Pyrolysis, Thermal Analysis, and Thermokinetic Studies: The Status Quo

“…In more recent studies, the idea has been extended to employ ANNs for predicting the conversion values at various moments, i.e., a TA signal itself. Such studies differ drastically in experimental systems ( Tithonia diversifolia weed biomass [ 63 ], low-density polyethylene [ 64 ], high-density polyethylene [ 68 ], safflower seed press cake [ 65 ], durian rinds [ 69 ], rape straw [ 70 ], coal gangue and peanut shell [ 71 ], pet coke [ 72 ], sewage sludge and peanut shell [ 73 ], sewage sludge and coffee grounds [ 74 ], vegetable fibers [ 75 ], rice husk and sewage sludge [ 45 ], sludge, watermelon rind, corncob, and eucalyptus [ 76 ], Sargassum sp. seaweed [ 77 ], cotton cocoon shell, tea waste, and olive husk [ 66 ], mechanoactivated coals [ 78 ], cattle manure [ 79 ], lignocellulosic forest residue and olive oil residue [ 80 ], cotton cocoon shell, fabricated tea waste, olive husk, and hazelnut shell [ 81 ]) and in some minor details, but the general concept remains the same.…”

Facile and Accurate Calculation of the Density of Amino Acid Salt Solutions: A Simple and General Correlation vs Artificial Neural Networks

“…Artificial neural networks (ANNs) are the most widely utilized AI models to address engineering problems including property estimation. − ANNs include computational procedures inspired by the function of biological neural networks. These methods have thus far gained popularity as powerful tools in a variety of studies; − however, the performance of ANNs in density modeling in AASs has not been yet assessed. A facile and accurate calculation of the densities of AASs can be managed by employing ANNs.…”