Artificial neural networks for water quality soft-sensing in wastewater treatment: a review

Wang, Gongming; Jia, Qing‐Shan; Zhou, Meng Chu; Bi, Jing; Qiao, Junfei; Abusorrah, Abdullah

doi:10.1007/s10462-021-10038-8

Cited by 93 publications

(29 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Prinsip dasar BPNN adalah mempelajari sampel input, menilai kesalahan, memodifikasi bobot dan nilai ambang batas untuk mengurangi kesalahan, dan kemudian mengulangi berkali-kali untuk mendapatkan hubungan pemetaan yang optimal [13]. BPNN terdiri dari banyak lapisan, termasuk lapisan masukan, lapisan tersembunyi, dan lapisan keluaran [14].…”

Section: Pendahuluanunclassified

Prediksi Kuantitas Penggunaan Obat pada Layanan Kesehatan Menggunakan Algoritma Backpropagation Neural Network

Khairati

Putra²

2022

jsisfotek

View full text Add to dashboard Cite

Prediction of the amount of drug use at public health centers is needed to ensure the availability of drugs for patients in service quality management. A good prediction of the amount of medicine needed helps the quality of development planning in the health sector. Scientific developments in the field of Artificial Intelligence (AI) deliver a variety of the best techniques for making predictions. By adopting the workings of neural networks (neurons) in the human brain or Artificial Neural Network (ANN), the Backpropagation Neural Network (BPNN) algorithm is one of the best algorithms in making predictions, including predicting drug use in health services. The problem of this research is how to design the best architectural model such as the number of neurons in the input layer, hidden layer and other parameters so as to produce predictions with optimal accuracy. This study aims to develop an ANN architectural design with the Backpropagation algorithm to predict the need for drug use. The data used is data on drug use reports from 2015 to 2021 at the Andalas Community Health Center (Puskesmas) Padang City. The steps taken to predict are; collect data, pre-process data and perform analysis, design ANN architecture, make predictions. Learning using the backpropagation algorithm through the initial weight initialization process, activation stage, weight training (weight change) and iteration stage. The proportion of the amount of data used for training is 70% data and 30% for testing data. The results of this study indicate that the best ANN architecture is 12-12-1 with an accuracy of predicting the quantity of drug use reaching 97.87% for paracetamol with a Mean Absolute Percentage Error (MAPE) of 2.13%. The prediction results become a reference for the Puskesmas and the Health Office for service planning and development.

show abstract

Section: Pendahuluanunclassified

Prediksi Kuantitas Penggunaan Obat pada Layanan Kesehatan Menggunakan Algoritma Backpropagation Neural Network

Khairati

Putra²

2022

jsisfotek

View full text Add to dashboard Cite

show abstract

“…Most colleges and universities in the United States collect the information data generated in the learning process, mainly to obtain teachers' teaching data and students' feedback data (Wang et al, 2021) [ 13 ]. The above information is used to help teachers optimize their teaching methods.…”

Section: Related Workmentioning

confidence: 99%

Discrete Dynamic Modeling of Learner Behavior Analysis in Physical Education Teaching

Shi

Jun

Zheng

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

With the advent of the big data era, the combination of information technology and education has become an important way for the development of the industry. The large-scale realization of teaching tasks under the background of information data requires the prediction and analysis of learners' characteristics, behavior, and development trend. Based on the above situation, this paper uses discrete dynamic modeling technology in big data environment to study the learners' behavior in physical education teaching. By quantifying the learning process data, the feature points of each learner are extracted to realize the personalized construction of dynamic learning data. Due to the rapid development of network technology, we mainly analyze the online education platform and explore the influencing factors of learners' behavior characteristics from many aspects. Finally, it carries out dynamic modeling and prediction for physical education learners from the aspect of achievement change, uses the grey model to build the achievement change system, and combines the dynamic modeling technology to reflect the development trend of achievement. The results show that the main factor affecting learners' behavior change in physical education is video learning. Most students are passive and lack initiative in the learning process. Discrete dynamic modeling technology can improve the accuracy of predicting student achievement changes and provide effective data for the research content.

show abstract

“…Artificial neural network is composed of simple adaptive units of a wide range of parallel interconnected network. In recent years, deep learning neural network algorithms have shown that compared with traditional neural network methods, deep learning algorithms have better accuracy and generalization ability (Wang et al, 2022). Kwon et al (2021) proposed a prediction model of distillation process based on the long short-term memory (LSTM) algorithm and the Adam optimization method.…”

Section: Introductionmentioning

confidence: 99%

A soft-sensing method for product quality monitoring based on particle swarm optimization deep belief networks

Yang

et al. 2022

Transactions of the Institute of Measurement and Control

View full text Add to dashboard Cite

A novel soft-sensing method for quality parameters of aviation kerosene in atmospheric distillation column based on least absolute shrinkage and selection operator and particle swarm optimization deep belief network (LASSO-PSO-DBN) is proposed. First, to reduce the dimension of the input variables, the least absolute shrinkage and selection operator (LASSO) algorithm is used to select the input variables that are irrelevant to the soft sensor of aviation kerosene quality parameters. Then, to improve the generalization of soft sensor model, a deep learning algorithm, deep belief network (DBN), is proposed for soft sensing of aviation kerosene quality parameters. Considering that the structure characteristics and parameters of DBN algorithm have a great impact on the learning and prediction results, the parameters of DBN are optimized based on particle swarm optimization (PSO) algorithm. The benchmark data sets and the industrial atmospheric distillation column data are used for simulation analysis and evaluation of the soft-sensing performance. The simulation results show that the novel proposed algorithm can effectively reduce the dimension of the input variables and simplify the structure of the soft sensor model. It also has good generalization ability and the predicted value is in good agreement with the actual measured value.

show abstract

Artificial neural networks for water quality soft-sensing in wastewater treatment: a review

Cited by 93 publications

References 59 publications

Prediksi Kuantitas Penggunaan Obat pada Layanan Kesehatan Menggunakan Algoritma Backpropagation Neural Network

Prediksi Kuantitas Penggunaan Obat pada Layanan Kesehatan Menggunakan Algoritma Backpropagation Neural Network

Discrete Dynamic Modeling of Learner Behavior Analysis in Physical Education Teaching

A soft-sensing method for product quality monitoring based on particle swarm optimization deep belief networks

Contact Info

Product

Resources

About