Fast characterization of solid organic waste content with near infrared spectroscopy in anaerobic digestion

Research on the chemical oxygen demand spectral inversion model in water based on IPLS-GAN-SVM hybrid algorithm

Spectral collinearity and limited spectral datasets are the problems influencing Chemical Oxygen Demand (COD) modeling. To address the first problem and obtain optimal modeling range, the spectra are preprocessed using six methods including Standard Normal Variate, Savitzky-Golay Smoothing Filtering (SG) etc. Subsequently, the 190-350 nm spectral range is divided into 10 subintervals, and Interval Partial Least Squares (IPLS) is used to perform PLS modeling on each interval. The results indicate that it is best modeled in the 7th range (238~253 nm). The values of Mean Square Error (MSE), Mean Absolute Error (MAE) and R2score of the model without pretreatment are 1.6489, 1.0661, and 0.9942. After pretreatment, the SG is better than others, with MSE and MAE decreasing to 1.4727, 1.0318 and R2score improving to 0.9944. Using the optimal model, the predicted COD for three samples are 10.87 mg/L, 14.88 mg/L, and 19.29 mg/L. To address the problem of the small dataset, using Generative Adversarial Networks for data augmentation, three datasets are obtained for Support Vector Machine (SVM) modeling. The results indicate that, compared to the original dataset, the SVM's MSE and MAE have decreased, while its accuracy has improved by 2.88%, 11.53%, and 11.53%, and the R2score has improved by 18.07%, 17.40%, and 18.74%.

Rapid characterization of sulfur and phosphorus in organic waste by near infrared spectroscopy

Near-infrared spectroscopy (NIRS) has recently emerged as a valuable tool for monitoring organic waste utilized in anaerobic digestion processes. Over the past decade, NIRS has significantly improved the characterization of organic waste by enabling the prediction of several crucial parameters such as biochemical methane potential, carbohydrate, lipid and nitrogen contents, Chemical Oxygen Demand, and kinetic parameters. This study investigates the application of NIRS for predicting the levels of Sulfur (S) and Phosphorus (P) within organic waste materials. The results for sulfur prediction exhibited a high level of accuracy, yielding an error of 1.21 g/Kg[TS] in an independently validated dataset, coupled with an R-squared value of 0.84. Conversely, the prediction of phosphorus proved to be slightly less successful, showing an error of 1.49 g/Kg[TS] with an R-squared value of 0.70. Furthermore, the disparities in performance seem to stem from the inherent correlation between the spectral data and the sulfur or phosphorus contents. Significantly, a variable selection technique known as CovSel was employed, shedding light on the differing approaches used for sulfur and phosphorus predictions. In the case of sulfur, the prediction was achieved through a direct correlation with wavelengths associated with sulfur-related functional groups (such as R - S(=O)2 - OH, -SH, and R-S-S-R) present in the NIR spectra. In contrast, phosphorus prediction relied on an indirect correlation with absorption bands related to organic matter (including CH, CH2, CH3, -CHO, R-OH, C = O, -CO2H, and CONH).

Modifications to the anaerobic digestion model no. 1 (ADM1) for enhanced understanding and application of the anaerobic treatment processes - A comprehensive review

Anaerobic digestion (AD) is a promising method for the recovery of resources and energy from organic wastes. Correspondingly, AD modelling has also been developed in recent years. The International Water Association (IWA) Anaerobic Digestion Model No. 1 (ADM1) is currently the most commonly used structured AD model. However, as substrates become more complex and our understanding of the AD mechanism grows, both systematic and specific modifications have been applied to the ADM1. Modified models have provided a diverse range of application besides AD processes, such as fermentation and biogas upgrading processes. This paper reviews research on the modification of the ADM1, with a particular focus on processes, kinetics, stoichiometry and parameters, which are the major elements of the model. The paper begins with a brief introduction to the ADM1, followed by a summary of modifications, including extensions to the model structure, modifications to kinetics (including inhibition functions) and stoichiometry, as well as simplifications to the model. The paper also covers kinetic parameter estimation and validation of the model, as well as practical applications of the model to a variety of scenarios. The review highlights the need for improvements in simulating AD and biogas upgrading processes, as well as the lack of full-scale applications to other substrates besides sludge (such as food waste and agricultural waste). Future research directions are suggested for model development based on detailed understanding of the anaerobic treatment mechanisms, and the need to recover of valuable products.

Hydrothermal and thermal-alkali pretreatments of wheat straw: Co-digestion, substrate solubilization, biogas yield and kinetic study

Agro-waste having lignocellulosic biomass is considered most effective (heating value 16 MJ/kg) for energy production through anaerobic digestion (AD). However, recalcitrant lignocellulosic fraction in agro-waste obstructs its biotransformation and is a rate-limiting step of the process. This study investigated the effects of hydrothermal and thermal-alkaline pretreatment on anaerobic co-digestion of wheat straw (WS). The hydrothermal pretreatment of WS revealed that 60 min was the best pretreatment time to achieve the highest substrate solubilization. It was employed for thermal-alkali pretreatment at variable temperatures and NaOH doses. Thermal-alkali pretreatment at 125°C-7% NaOH shows the highest (34%) biogas yield of 662 mL/gVS, followed by 646 mL/gVS biogas yield at 150°C-1% NaOH assay (31% higher) over control. Although the 125°C-7% NaOH assay achieved the highest biogas yield, the 150°C-1% NaOH assay was found more feasible considering the cost of a 6% higher chemical used in the earlier assay. The thermal-alkali pretreatment was observed to reduce the formation of recalcitrant compounds (HMF, Furfural) and increase the buffering capacity of the slurry over hydrothermal pretreatment. Principal component analysis (PCA) of the various pretreatment and AD operational parameters was carried out to study their in-depth correlation. Moreover, a kinetic study of the experimental data was performed to observe the biodegradation trend and compare it with the Modified Gompertz (MG) and First Order (FO) models.

Fast and robust NIRS-based characterization of raw organic waste: Using non-linear methods to handle water effects

Fast characterization of organic waste using near infrared spectroscopy (NIRS) has been successfully developed in the last decade. However, up to now, an on-site use of this technology has been hindered by necessary sample preparation steps (freeze-drying and grinding) to avoid important water effects on NIRS. Recent research studies have shown that these effects are highly non-linear and relate both to the biochemical and physical properties of samples. To account for these complex effects, the current study compares the use of many different types of non-linear methods such as partial least squares regression (PLSR) based methods (global, clustered and local versions of PLSR), machine learning methods (support vector machines, regression trees and ensemble methods) and deep learning methods (artificial and convolutional neural networks). On an independent test data set, non-linear methods showed errors 28% lower than linear methods. The standard errors of prediction obtained for the prediction of total solids content (TS%), chemical oxygen demand (COD) and biochemical methane potential (BMP) were respectively 8%, 160 mg(O₂).gTS^-1 and 92 mL(CH₄).gTS^-1_. These latter errors are similar to successful NIRS applications developed on freeze-dried samples. These findings hold great promises regarding the development of at-site and online NIRS solutions in anaerobic digestion plants.

Agronomic characterization of anaerobic digestates with near-infrared spectroscopy

Anaerobic digestion is an increasingly widespread process for organic waste treatment and renewable energy production due to the methane content of the biogas. This biological process also produces a digestate (i.e., the remaining content of the waste after treatment) with a high fertilizing potential. The digestate composition is highly variable due to the various organic wastes used as feedstock, the different plant configurations, and the post-treatment processes used. In order to optimize digestate spreading on agricultural soils by optimizing the fertilizer dose and, thus, reducing environmental impacts associated to digestate application, the agronomic characterization of digestate is essential. This study investigates the use of near infrared spectroscopy for predicting the most important agronomic parameters from freeze-dried digestates. A data set of 193 digestates was created to calibrate partial least squares regression models predicting organic matter, total organic carbon, organic nitrogen, phosphorus, and potassium contents. The calibration range of the models were between 249.8 and 878.6 gOM.kgDM^-1, 171.9 and 499.5 gC.kgDM^-1, 5.3 and 74.1 gN.kgDM^-1, 2.7 and 44.9 gP.kgDM^-1 and between 0.5 and 171.8 gK.kgDM^-1, respectively. The calibrated models reliably predicted organic matter, total organic carbon, and phosphorus contents for the whole diversity of digestates with root mean square errors of prediction of 70.51 gOM.kgDM^-1, 34.84 gC.kgDM^-1 and 4.08 gP.kgDM^-1, respectively. On the other hand, the model prediction of the organic nitrogen content had a root mean square error of 7.55 gN.kgDM^-1 and was considered as acceptable. Lastly, the results did not demonstrate the feasibility of predicting the potassium content in digestates with near infrared spectroscopy. These results show that near infrared spectroscopy is a very promising analytical method for the characterization of the fertilizing value of digestates, which could provide large benefits in terms of analysis time and cost.

Monitor process state of batch anaerobic digestion in reliance on volatile and semi-volatile metabolome

It has been a challenge to recognize appropriate compounds as indicators for monitoring and early-warning of the anaerobic digestion process. A strategy was initiated to explore the evolution of the panorama profile of volatile and semi-volatile metabolites. Non-target analysis using high-resolution gas chromatography coupled with Orbitrap mass spectrometry was applied to construct a time-series molecular fingerprint of 218 metabolites classified in 14 categories. Alkanes accounted for the main part in early and late stages of methanization and aromatic compounds were the major in middle stage. Spearman correlation analysis and partial least squares analysis unwind that Trichococcus (1.49%-83.96%) was positively related to most of metabolites at early and middle stages, while Brevundimonas (0%-24.04%) was positively related to acylamide at late stage. This indicated that microbial volatile organic compounds were possible to serve as biochemical indicators for anaerobic digestion performance and to build nexus of "what" (metabolites), "who" (microorganism), and "how" (kinetics).

Implementation of an early warning system with hyperspectral imaging combined with deep learning model for chlorine in refuse derived fuels

Chlorine content is one of the most important parameters in Refuse Derived Fuels (RDFs) used as a fuel in cement kilns. The main problem with the use of RDF is that chlorine in the waste weakens the cement, increases the risk of corrosion in the kiln and forms toxic gas emissions. Alternative fuels containing high amounts of chlorine, such as plastic waste should be used in limited quantities with the quality of the kiln used and the cement being should be preserved by preparing the appropriate RDF mixture. Analyses conducted on the samples taken before the RDF is given to the furnace are time consuming and costly. Therefore, in this study, the aim is to present a more efficient solution to classify by using chlorine analysis results with hyperspectral imaging and a deep learning model study. For this purpose, a model was created using validated laboratory results and spectral data from samples, the model was tested on a prototype conveyor belt, and was implemented using an online early warning system for high chlorine concentrations. The chlorine content of the RDF samples used in the study ranged from 0.10% to 1.41%, with an average of 0.27%. According to the results, the accuracy, precision, Recall and F1 Score related to the early warning system were found to be 0.8909, 0.8889, 0.8889, 0.8889, respectively. In addition, chlorine measurements were performed at 200, 500 and 1000 mm/s belt speeds and accuracy values of 78.39%, 76.35% and 69.94 %, respectively were obtained.

On-site substrate characterization in the anaerobic digestion context: A dataset of near infrared spectra acquired with four different optical systems on freeze-dried and ground organic waste

The near infrared spectra of thirty-three freeze-dried and ground organic waste samples of various biochemical composition were collected on four different optical systems, including a laboratory spectrometer, a transportable spectrometer with two measurement configurations (an immersed probe, and a polarized light system) and a micro-spectrometer. The provided data contains one file per spectroscopic system including the reflectance or absorbance spectra with the corresponding sample name and wavelengths. A reference data file containing carbohydrates, lipid and nitrogen content, biochemical methane potential (BMP) and chemical oxygen demand (COD) for each sample is also provided. This data enables the comparison of the optical systems for predictive model calibration based for example on Partial Least Squares Regression (PLS-R) [1], but could be used more broadly to test new chemometrics methods. For example, the data could be used to evaluate different transfer functions between spectroscopic systems [2]. This dataset enabled the research work reported by Mallet et al. 2021 [3].

Fast at-line characterization of solid organic waste: Comparing analytical performance of different compact near infrared spectroscopic systems with different measurement configurations

Fast characterization of solid organic waste using near infrared spectroscopy has been successfully developed in the last decade. However, its adoption in biogas plants for monitoring the feeding substrates remains limited due to the lack of applicability and high costs. Recent evolutions in the technology have given rise to both more compact and more modular low-cost near infrared systems which could allow a larger scale deployment. The current study investigates the relevance of these new systems by evaluating four different Fourier transform near-infrared spectroscopic systems with different compactness (laboratory, portable, micro spectrometer) but also different measurement configurations (polarized light, at distance, in contact). Though the conventional laboratory spectrometer showed the best performance on the various biochemical parameters tested (carbohydrates, lipids, nitrogen, chemical oxygen demand, biochemical methane potential), the compact systems provided very close results. Prediction of the biochemical methane potential was possible using a low-cost micro spectrometer with an independent validation set error of only 91 NmL(CH₄).gTS^-1 compared to 60 NmL(CH₄).gTS^-1 for a laboratory spectrometer. The differences in performance were shown to result mainly from poorer spectral sampling; and not from instrument characteristics such as spectral resolution. Regarding the measurement configurations, none of the evaluated systems allowed a significant gain in robustness. In particular, the polarized light system provided better results when using its multi-scattered signal which brings further evidence of the importance of physical light-scattering properties in the success of models built on solid organic waste.

Unveiling non-linear water effects in near infrared spectroscopy: A study on organic wastes during drying using chemometrics

In the context of organic waste management, near infrared spectroscopy (NIRS) is being used to offer a fast, non-destructive, and cost-effective characterization system. However, cumbersome freeze-drying steps of the samples are required to avoid water's interference on near infrared spectra. In order to better understand these effects, spectral variations induced by dry matter content variations were obtained for a wide variety of organic substrates. This was made possible by the development of a customized near infrared acquisition system with dynamic highly-resolved simultaneous scanning of near infrared spectra and estimation of dry matter content during a drying process at ambient temperature. Using principal components analysis, the complex water effects on near infrared spectra are detailed. Water effects are shown to be a combination of both physical and chemical effects, and depend on both the characteristics of the samples (biochemical type and physical structure) and the moisture content level. This results in a non-linear relationship between the measured signal and the analytical characteristic of interest. A typology of substrates with respect to these water effects is provided and could further be efficiently used as a basis for the development of local quantitative calibration models and correction methods accounting for these water effects.

Determination of chemical oxygen demand of agricultural wastes by combining acid hydrolysis and commercial COD kit analysis

Chemical oxygen demand (COD) is an essential parameter in waste management, particularly for monitoring bioprocess such as anaerobic digestion. Indeed, chemical oxygen demand (COD) is a key parameter that can prove useful for the evaluation of waste biodegradability and to evaluate mass and energetic balances of the overall process. In this study, an adapted method to determine the COD of solid agricultural wastes was developed. This method combined a double acid hydrolysis of the solid waste materials followed by commercial COD tubes analysis. This method was compared to direct sampling after a standard dilution (3.5 g TS.L^-1) and analysis in commercial COD tubes. The method developed in this study allowed the COD of nine agricultural wastes to be accurately predicted, with an absolute error of 7% compared to the theoretical COD. In comparison, the method with only a prior water dilution resulted in higher absolute errors of 36% and 31% when sampling was performed with pipette tips and cut pipette tips, respectively.

Mechanical pretreatment of lignocelluloses for enhanced biogas production: Methane yield prediction from biomass structural components

In this study, mechanical pretreatment was applied to six different lignocelluloses in two different treatment phases and the prediction of their methane yield was done from biomass chemical composition. Physicochemical, proximate and microbial analyses were carried out on both pretreated and untreated biomass using standard methods. Mechanical pretreatments caused the breakdown of structural materials in all the used biomass which was characterized by reduction of the lagging time during anaerobic digestion and the subsequent increase in methane yield up to 22%. The different loading rate of biomass had no effect on the overall methane yield increase. Both single and multiple linear regressions models were used in order to correlate the chemical composition of the biomass with their methane potentials and a fairly high correlation (R² = 0.63) was obtained. The study also showed that the pretreatments are economically feasible. Therefore, its further application to other biomass is encouraged.

Fast prediction of organic wastes methane potential by near infrared reflectance spectroscopy: A successful tool for farm-scale biogas plant monitoring

Currently, there is a growing worldwide interest for the treatment of wastes, and especially farm wastes, by anaerobic digestion. Biochemical methane potential is a key parameter for the design, optimisation and monitoring of the anaerobic digestion process, but it is also time consuming (4-7 weeks). Near infrared reflectance spectroscopy seems a promising method to predict the biochemical methane potential of a wide range of organic substrates. This study compares a 'global' predictive model mainly built with biogas plant feedstocks, and a more 'agricultural' specific one built with farm wastes only (e.g. manures and crop residues). The global model was calibrated with 245 samples and the specific one with 171 samples. In parallel, validation sets composed of 36 farm wastes and eight other wastes (sludge, fruit residues and vegetables) were used to evaluate and compare both models. Satisfying results were obtained on the validation sets considering, respectively for the global and the specific models, a root mean square error of prediction of 44 and 34 NL CH₄ kg^-1 volatile solid, a coefficient of determination of 0.76 and 0.83, and a ratio of performance to deviation of 2.0 and 2.4. In general rules, the specific model was better than the global one in the prediction of farm wastes methane potential. However, thanks to its larger sample variability, the global one was more robust, especially towards the 'other' wastes, which can be introduced punctually in agricultural biogas plant.

Mechanical pretreatment for increased biogas production from lignocellulosic biomass; predicting the methane yield from structural plant components

Lignocellulosic substrates are associated with limited biodegradability due to the structural complexity. For that reason, a pretreatment step is mandatory for efficient biomass transformation which will lead to increased bioenergy output. The aim of the present study was to assess the efficiency of two pretreatment machines to enhance the methane yield of meadow grass. Specifically, the application of shearing forces with a rotated plastic sweeping brush against a steel roller significantly increased biomass biodegradability by 20% under relatively gentle operation conditions (600 rpm). The more intense operation (1200 rpm) was not associated with higher methane yield enhancement. Regarding an alternative machine, in which the brush was replaced with a coarse steel roller resulted in a more distinct effect (+27%) despite the lower rotating speed (∼400 rpm). Moreover, the association of the substrate's individual chemical components and the practical methane yield was assessed, establishing single and multiple linear regression models. However, the estimation accuracy was rather low with either single (regressor: lignin, R²: 0.50) or multiple linear regression analyses (regressors: arabinan-lignin-protein, R²: 0.61). Results showed that poorly lignified plant tissue containing relatively high fractions of protein and arabinan is more susceptible to anaerobic digestion.

Anaerobic digestion of food waste: A review focusing on process stability

Food waste (FW) is rich in biomass energy, and increasing numbers of national programs are being established to recover energy from FW using anaerobic digestion (AD). However process instability is a common operational issue for AD of FW. Process monitoring and control as well as microbial management can be used to control instability and increase the energy conversion efficiency of anaerobic digesters. Here, we review research progress related to these methods and identify existing limitations to efficient AD; recommendations for future research are also discussed. Process monitoring and control are suitable for evaluating the current operational status of digesters, whereas microbial management can facilitate early diagnosis and process optimization. Optimizing and combining these two methods are necessary to improve AD efficiency.