Prediction of sludge volume index bulking using image analysis and neural network at a full-scale activated sludge plant

Boztoprak, Halime; Özbay, Yüksel; Güçlü, Dünyamin; Küçükhemek, Murat

doi:10.1080/19443994.2015.1085909

Cited by 26 publications

(15 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As main advantages, QIA removes the subjectivity of human analysis, enabling to extract quantitative data and avoid tedious and highly time-consuming tasks to human operators [16]. In fact, QIA has been increasingly used for wastewater treatment plants (WWTP) characterization, since the initial studies of Grijspeerdt and Verstraete in 1997 [17] using QIA to relate the sludge characteristics to the settling ability, also studied by [15,[18][19][20]. Nonetheless, most published studies used QIA for the characterization of the structural differences of the aggregated biomass, resulting from different operating conditions, in AS systems rather than AGS systems [16,[21][22][23][24][25].…”

Section: Introductionmentioning

confidence: 99%

Sludge volume index and suspended solids estimation of mature aerobic granular sludge by quantitative image analysis and chemometric tools

Leal

Río

Mesquita

et al. 2020

Separation and Purification Technology

View full text Add to dashboard Cite

Aerobic granular sludge (AGS) is considered a promising technology for wastewater treatment. Furthermore, it is recognized that the stability of the process is related to the balanced growth of the suspended (floccular) and granular fractions. Therefore, the development of adequate techniques to monitor this balance is of interest. In this work the sludge volume index (SVI), volatile suspended solids (VSS) and total suspended solids (TSS) of mature AGS were successfully predicted with multilinear regression (MLR) models using data obtained from quantitative image analysis (QIA) of both fractions (suspended and granular). Relevant predictions were obtained for the SVI (R 2 of 0.975), granules TSS (R 2 of 0.985), flocs TSS (R 2 of 0.971), granules VSS (R 2 of 0.984) and flocs VSS (R 2 of 0.986). The estimation of the granular fraction ratio from the predicted TSS and VSS was also successful (R 2 of 0.985). The predictions help to avoid instability episodes of the AGS system, such as changes in biomass morphology, structure and settling properties.

show abstract

Section: Introductionmentioning

confidence: 99%

Sludge volume index and suspended solids estimation of mature aerobic granular sludge by quantitative image analysis and chemometric tools

Leal

Río

Mesquita

et al. 2020

Separation and Purification Technology

View full text Add to dashboard Cite

show abstract

“…Because of the advantages of ABC algorithm, it has been applied in various research fields. For example, Halime et al adopted the ABC algorithm to train the weights of a neural network, and monitor and predict the sludge volume index using image analysis and the trained neural network at a full-scale activated sludge plant [16]. Santhi et al introduced the crossover operation into the ABC algorithm for more efficient job scheduling [17].…”

Section: Abc Algorithmmentioning

confidence: 99%

Application of an Improved ABC Algorithm in Urban Land Use Prediction

Huo

Zhang

2018

Information

View full text Add to dashboard Cite

Scientifically and rationally analyzing the characteristics of land use evolution and exploring future trends in land use changes can provide the scientific reference basis for the rational development and utilization of regional land resources and sustainable economic development. In this paper, an improved hybrid artificial bee colony (ABC) algorithm based on the mutation of inferior solutions (MHABC) is introduced to combine with the cellular automata (CA) model to implement a new CA rule mining algorithm (MHABC-CA). To verify the capabilities of this algorithm, remote sensing data of three stages, 2005, 2010, and 2015, are adopted to dynamically simulate urban development of Dengzhou city in Henan province, China, using the MHABC-CA algorithm. The comprehensive validation and analysis of the simulation results are performed by two aspects of comparison, the visual features of urban land use types and the quantification analysis of simulation accuracy. Compared with a cellular automata model based on a particle swarm optimization (PSO-CA) algorithm, the experimental results demonstrate the effectiveness of the MHABC-CA algorithm in the prediction field of urban land use changes.

show abstract

“…(3)(4)(5)(6) By extracting the morphological information of flocs and filaments from microscopic images, physical and chemical parameters such as SVI can be effectively measured, and the abnormal operation of a sewage treatment plant can be detected early. (7) Since the collected samples do not require special preparation and filamentous bacteria can be observed at a relatively low magnification, phase contrast microscopy (PCM) is commonly used to observe activated sludge. Over the last two decades, the processing and analyses of PCM images of activated sludge have received considerable attention and widely applied to the measurement of SVI and the early detection and fault diagnosis of sludge bulking in wastewater treatment plants.…”

Section: Introductionmentioning

confidence: 99%

Segmentation of Activated Sludge Phase Contrast Microscopy Images Using U-Net Deep Learning Model

Zhao¹,

Zou²,

Zhang³

et al. 2019

Sensors and Materials

View full text Add to dashboard Cite

For the activated sludge wastewater treatment process, the image segmentation of flocs and filaments has become a crucial component in the successful implementation of a sludge volume index (SVI) sensor and the early fault detection of filamentous bulking. The segmentation of a phase contrast microscopy (PCM) image is a challenging problem because of the weak greyscale distinction between flocs and filaments, as well as the artifacts of halos and shadows. In this work, we proposed an automatic floc and filament segmentation method for PCM images using a U-Net deep learning structure with data augmentation. A loss function combining the binary cross entropy (BCE) function and Dice coefficient is proposed to improve the segmentation accuracy and sensitivity with unbalanced foreground and background samples. The performance of the segmentation algorithm is evaluated by the accuracy, precision, recall, F-measure, and intersection-over-union (IoU) metrics. Lab-scale experiments on the activated sludge process have been carried out to verify the proposed image segmentation method. Our proposed U-Net models with the combined loss function give better results than the U-Net models with BCE, fully convolutional network-VGG16 (FCN-VGG16), and a traditional segmentation method.

show abstract

Prediction of sludge volume index bulking using image analysis and neural network at a full-scale activated sludge plant

Cited by 26 publications

References 31 publications

Sludge volume index and suspended solids estimation of mature aerobic granular sludge by quantitative image analysis and chemometric tools

Sludge volume index and suspended solids estimation of mature aerobic granular sludge by quantitative image analysis and chemometric tools

Application of an Improved ABC Algorithm in Urban Land Use Prediction

Segmentation of Activated Sludge Phase Contrast Microscopy Images Using U-Net Deep Learning Model

Contact Info

Product

Resources

About