Aquaculture has long been a critical economic sector in Taiwan.Since a key factor in aquaculture production efficiency is water quality,an effective means of monitoring the dissolved oxygen content(DOC)of aquaculture ...Aquaculture has long been a critical economic sector in Taiwan.Since a key factor in aquaculture production efficiency is water quality,an effective means of monitoring the dissolved oxygen content(DOC)of aquaculture water is essential.This study developed an internet of things system for monitoring DOC by collecting essential data related to water quality.Artificial intelligence technology was used to construct a water quality prediction model for use in a complete system for managing water quality.Since aquaculture water quality depends on a continuous interaction among multiple factors,and the current state is correlated with the previous state,a model with time series is required.Therefore,this study used recurrent neural networks(RNNs)with sequential characteristics.Commonly used RNNs such as long short-term memory model and gated recurrent unit(GRU)model have a memory function that appropriately retains previous results for use in processing current results.To construct a suitable RNN model,this study used Taguchi method to optimize hyperparameters(including hidden layer neuron count,iteration count,batch size,learning rate,and dropout ratio).Additionally,optimization performance was also compared between 5-layer and 7-layer network architectures.The experimental results revealed that the 7-layer GRU was more suitable for the application considered in this study.The values obtained in tests of prediction performance were mean absolute percentage error of 3.7134%,root mean square error of 0.0638,and R-value of 0.9984.Therefore,thewater qualitymanagement system developed in this study can quickly provide practitioners with highly accurate data,which is essential for a timely response to water quality issues.This study was performed in collaboration with the Taiwan Industrial Technology Research Institute and a local fishery company.Practical application of the system by the fishery company confirmed that the monitoring system is effective in improving the survival rate of farmed fish by providing data needed to maintain DOC higher than the standard value.展开更多
Water resources are an indispensable and valuable resource for human survival and development.Water quality predicting plays an important role in the protection and development of water resources.It is difficult to pr...Water resources are an indispensable and valuable resource for human survival and development.Water quality predicting plays an important role in the protection and development of water resources.It is difficult to predictwater quality due to its random and trend changes.Therefore,amethod of predicting water quality which combines Auto Regressive Integrated Moving Average(ARIMA)and clusteringmodelwas proposed in this paper.By taking thewater qualitymonitoring data of a certain river basin as a sample,thewater quality Total Phosphorus(TP)index was selected as the prediction object.Firstly,the sample data was cleaned,stationary analyzed,and white noise analyzed.Secondly,the appropriate parameters were selected according to the Bayesian Information Criterion(BIC)principle,and the trend component characteristics were obtained by using ARIMA to conduct water quality predicting.Thirdly,the relationship between the precipitation and the TP index in themonitoring water field was analyzed by the K-means clusteringmethod,and the random incremental characteristics of precipitation on water quality changes were calculated.Finally,by combining with the trend component characteristics and the random incremental characteristics,the water quality prediction results were calculated.Compared with the ARIMA water quality prediction method,experiments showed that the proposed method has higher accuracy,and its Mean Absolute Error(MAE),Mean Square Error(MSE),and Mean Absolute Percentage Error(MAPE)were respectively reduced by 44.6%,56.8%,and 45.8%.展开更多
The quality of a water body is usually characterized by sets of physical, chemical, and biological parameters, which are mutually interrelated. Since August 1997, monthly records of 33 parameters, monitored at 102 loc...The quality of a water body is usually characterized by sets of physical, chemical, and biological parameters, which are mutually interrelated. Since August 1997, monthly records of 33 parameters, monitored at 102 locations on the Nile Delta drainage system, are stored in a National Database operated by the Drainage Research Institute (DRI). Correlation patterns may be found between water quantity and water quality parameters at the same location, or among water quality parameters within a monitoring location or among locations. Serial correlation is also detected in water quality variables. Through the investigation of the level of information redundancy, assessment and redesign of water quality monitoring network aim to improve the overall network efficiency and cost effectiveness. In this study, the potential of the Artificial Neural Network (ANN) on simulating interrelation between water quality parameters is examined. Several ANN inputs, structures and training possibilities are assessed and the best ANN model and modeling procedure is selected. The prediction capabilities of the ANN are compared with the linear regression models with autocorrelated residuals, usually used for this purpose. It is concluded that the ANN models are more accurate than the linear regression models having the same inputs and output.展开更多
Artificial Neural Network(ANN)models have been extensively applied in the prediction of water resource variables,and Geographical Information System(GIS)includes powerful functions to visualize spatial data.In order t...Artificial Neural Network(ANN)models have been extensively applied in the prediction of water resource variables,and Geographical Information System(GIS)includes powerful functions to visualize spatial data.In order to provide an efficient tool for environmental assessment and management that combines the advantages of these two modules,a GIS-based ANN water quality prediction system was developed in the present study.The ANN module and ArcGIS Engine module,along with a dynamic database,were imbedded in the system,which integrates water quality prediction via the ANN model and spatial presentation of the model results.The structure of the ANN model could be modified through the graphical user interface to optimize the model performance.The developed system was applied to a real case study for the prediction of the total phosphorus concentration in the Lake Champlain area.The prediction results were verified with the monitoring data,and the performance of the developed model was further evaluated through graphical techniques and quantitative statistical methods.Overall,the developed system provided satisfactory prediction results,and spatial distribution maps of the predicted results were obtained,which coincided with the monitored values.The developed GIS-based ANN water quality prediction system could serve as an efficient tool for engineers and decision makers.展开更多
The fraction defective of semi-finished products is predicted to optimize the process of relay production lines, by which production quality and productivity are increased, and the costs are decreased. The process par...The fraction defective of semi-finished products is predicted to optimize the process of relay production lines, by which production quality and productivity are increased, and the costs are decreased. The process parameters of relay production lines are studied based on the long-and-short-term memory network. Then, the Keras deep learning framework is utilized to build up a short-term relay quality prediction algorithm for the semi-finished product. A simulation model is used to study prediction algorithm. The simulation results show that the average prediction absolute error of the fraction is less than 5%. This work displays great application potential in the relay production lines.展开更多
Chaos theory was introduced for water quality, prediction, and the model of water quality prediction was established by combining phase space reconstruction theory and BP neural network forecasting method. Through the...Chaos theory was introduced for water quality, prediction, and the model of water quality prediction was established by combining phase space reconstruction theory and BP neural network forecasting method. Through the phase space reconstruction, the one-dimensional water quality time series were mapped to be multi-dimensional sequence, which enriched the spatial information of water quality change and expanded mapping region of training samples of BP neural network. Established model of combining chaos theory and BP neural network were applied to forecast turbidity time series of a certain reservoir. Contrast to BP neural network method, the relative error and the mean squared error of the combined method had all varying degrees of lower. Results indicated the neural network model with chaos theory had the higher prediction accuracy, at the same time, it had better fault-tolerant capability and generalization performance .展开更多
A deep learning method for predicting oil field production at ultra-high water cut stage from the existing oil field production data was presented,and the experimental verification and application effect analysis were...A deep learning method for predicting oil field production at ultra-high water cut stage from the existing oil field production data was presented,and the experimental verification and application effect analysis were carried out.Since the traditional Fully Connected Neural Network(FCNN)is incapable of preserving the correlation of time series data,the Long Short-Term Memory(LSTM)network,which is a kind of Recurrent Neural Network(RNN),was utilized to establish a model for oil field production prediction.By this model,oil field production can be predicted from the relationship between oil production index and its influencing factors and the trend and correlation of oil production over time.Production data of a medium and high permeability sandstone oilfield in China developed by water flooding was used to predict its production at ultra-high water cut stage,and the results were compared with the results from the traditional FCNN and water drive characteristic curves.The LSTM based on deep learning has higher precision,and gives more accurate production prediction for complex time series in oil field production.The LSTM model was used to predict the monthly oil production of another two oil fields.The prediction results are good,which verifies the versatility of the method.展开更多
At present,water pollution has become an important factor affecting and restricting national and regional economic development.Total phosphorus is one of the main sources of water pollution and eutrophication,so the p...At present,water pollution has become an important factor affecting and restricting national and regional economic development.Total phosphorus is one of the main sources of water pollution and eutrophication,so the prediction of total phosphorus in water quality has good research significance.This paper selects the total phosphorus and turbidity data for analysis by crawling the data of the water quality monitoring platform.By constructing the attribute object mapping relationship,the correlation between the two indicators was analyzed and used to predict the future data.Firstly,the monthly mean and daily mean concentrations of total phosphorus and turbidity outliers were calculated after cleaning,and the correlation between them was analyzed.Secondly,the correlation coefficients of different times and frequencies were used to predict the values for the next five days,and the data trend was predicted by python visualization.Finally,the real value was compared with the predicted value data,and the results showed that the correlation between total phosphorus and turbidity was useful in predicting the water quality.展开更多
Neural networks(NNs)have been used extensively in surface water prediction tasks due to computing algorithm improvements and data accumulation.An essential step in developing an NN is the hyperparameter selection.In p...Neural networks(NNs)have been used extensively in surface water prediction tasks due to computing algorithm improvements and data accumulation.An essential step in developing an NN is the hyperparameter selection.In practice,it is common to manually determine hyperparameters in the studies of NNs in water resources tasks.This may result in considerable randomness and require significant computation time;therefore,hyperparameter optimization(HPO)is essential.This study adopted five representatives of the HPO techniques in the surface water quality prediction tasks,including the grid sampling(GS),random search(RS),genetic algorithm(GA),Bayesian optimization(BO)based on the Gaussian process(GP),and the tree Parzen estimator(TPE).For the evaluation of these techniques,this study proposed a method:first,the optimal hyperparameter value sets achieved by GS were regarded as the benchmark;then,the other HPO techniques were evaluated and compared with the benchmark in convergence,optimization orientation,and consistency of the optimized values.The results indicated that the TPE-based BO algorithm was recommended because it yielded stable convergence,reasonable optimization orientation,and the highest consistency rates with the benchmark values.The optimization consistency rates via TPE for the hyperparameters hidden layers,hidden dimension,learning rate,and batch size were 86.7%,73.3%,73.3%,and 80.0%,respectively.Unlike the evaluation of HPO techniques directly based on the prediction performance of the optimized NN in a single HPO test,the proposed benchmark-based HPO evaluation approach is feasible and robust.展开更多
Efficiently predicting effluent quality through data-driven analysis presents a significant advancement for consistent wastewater treatment operations.In this study,we aimed to develop an integrated method for predict...Efficiently predicting effluent quality through data-driven analysis presents a significant advancement for consistent wastewater treatment operations.In this study,we aimed to develop an integrated method for predicting effluent COD and NH3 levels.We employed a 200 L pilot-scale sequencing batch reactor(SBR)to gather multimodal data from urban sewage over 40 d.Then we collected data on critical parameters like COD,DO,pH,NH_(3),EC,ORP,SS,and water temperature,alongside wastewater surface images,resulting in a data set of approximately 40246 points.Then we proposed a brain-inspired image and temporal fusion model integrated with a CNN-LSTM network(BITF-CL)using this data.This innovative model synergized sewage imagery with water quality data,enhancing prediction accuracy.As a result,the BITF-CL model reduced prediction error by over 23%compared to traditional methods and still performed comparably to conventional techniques even without using DO and SS sensor data.Consequently,this research presents a cost-effective and precise prediction system for sewage treatment,demonstrating the potential of brain-inspired models.展开更多
Anaerobic process is regarded as a green and sustainable process due to low carbon emission and minimal energy consumption in wastewater treatment plants(WWTPs).However,some water quality metrics are not measurable in...Anaerobic process is regarded as a green and sustainable process due to low carbon emission and minimal energy consumption in wastewater treatment plants(WWTPs).However,some water quality metrics are not measurable in real time,thus influencing the judgment of the operators and may increase energy consumption and carbon emission.One of the solutions is using a soft-sensor prediction technique.This article introduces a water quality soft-sensor prediction method based on Bidirectional Gated Recurrent Unit(BiGRU)combined with Gaussian Progress Regression(GPR)optimized by Tree-structured Parzen Estimator(TPE).TPE automatically optimizes the hyperparameters of BiGRU,and BiGRU is trained to obtain the point prediction with GPR for the interval prediction.Then,a case study applying this prediction method for an actual anaerobic process(2500 m^(3)/d)is carried out.Results show that TPE effectively optimizes the hyperparameters of BiGRU.For point prediction of CODeff and biogas yield,R^(2)values of BiGRU,which are 0.973 and 0.939,respectively,are increased by 1.03%–7.61%and 1.28%–10.33%,compared with those of other models,and the valid prediction interval can be obtained.Besides,the proposed model is assessed as a reliable model for anaerobic process through the probability prediction and reliable evaluation.It is expected to provide high accuracy and reliable water quality prediction to offer basis for operators in WWTPs to control the reactor and minimize carbon emission and energy consumption.展开更多
水质指标具有多元相关性、时序性和非线性的特点,为有效预测河流水质变化,针对水质数据存在缺失和异常的问题,提出基于灰色关联分析-门控循环单元(Grey Relational Analysis-Gated Recurrent Unit, GRA-GRU)的水质预测模型。以淮河流域...水质指标具有多元相关性、时序性和非线性的特点,为有效预测河流水质变化,针对水质数据存在缺失和异常的问题,提出基于灰色关联分析-门控循环单元(Grey Relational Analysis-Gated Recurrent Unit, GRA-GRU)的水质预测模型。以淮河流域水质数据为样本,使用线性插值修补缺失数据和剔除的异常数据。使用灰色关联分析计算不同水质指标间的相关性,选择高相关性的水质指标以确定输入变量,并使用门控循环单元(Gated Recurrent Unit, GRU)预测不同的水质指标。将GRA-GRU的预测结果与反向传播神经网络(Back Propagation Neural Network, BPNN)、循环神经网络(Recurrent Neural Network, RNN)、长短期记忆神经网络(Long Short Term Memory, LSTM)、GRU及灰色关联分析-长短期记忆神经网络(Grey Relational Analysis-Long Short Term Memory, GRA-LSTM)进行对比分析,结果显示GRA-GRU在不同水质指标预测上具有较好的适应性,可以有效降低预测误差。其中,与其他模型相比,GRA-GRU预测的化学需氧量在均方根误差上分别降低了3.617%、0.681%、0.478%、1.505%和0.471%。展开更多
针对污水处理厂生化池中参数监测智能化水平不高、人力耗费较大的问题,提出基于麻雀算法-长短期记忆神经网络(Sparrow Search Algorithm-Long Short Term Memory Network,SSA-LSTM)的水质参数预测模型。以污水处理过程中好氧区溶解氧(Di...针对污水处理厂生化池中参数监测智能化水平不高、人力耗费较大的问题,提出基于麻雀算法-长短期记忆神经网络(Sparrow Search Algorithm-Long Short Term Memory Network,SSA-LSTM)的水质参数预测模型。以污水处理过程中好氧区溶解氧(Dissolved Oxygen,DO)、好氧区混合液悬浮固体(Mixed Liquid Suspended Solids,MLSS)质量浓度、缺氧区DO、缺氧区氧化还原电位(Oxidation-Reduction Potential,ORP)、厌氧区DO和厌氧区ORP 6个关键指标为数据样本,进行实例研究。将SSA-LSTM的预测结果与长短期记忆神经网络(Long Short-Term Memory Network,LSTM)、粒子群算法(Particle Swarm optimization-Long Short Term Memory Network,PSO-LSTM)、深度森林以及支持向量机进行对比分析,结果显示:SSA-LSTM在6个参数上的均方误差(EMSE)和决定系数(R2)均表现出更好的预测性,预测精度最高。展开更多
基金Publication costs are funded by the Ministry of Science and Technology,Taiwan,under Grant Numbers MOST 110-2221-E-153-010.
文摘Aquaculture has long been a critical economic sector in Taiwan.Since a key factor in aquaculture production efficiency is water quality,an effective means of monitoring the dissolved oxygen content(DOC)of aquaculture water is essential.This study developed an internet of things system for monitoring DOC by collecting essential data related to water quality.Artificial intelligence technology was used to construct a water quality prediction model for use in a complete system for managing water quality.Since aquaculture water quality depends on a continuous interaction among multiple factors,and the current state is correlated with the previous state,a model with time series is required.Therefore,this study used recurrent neural networks(RNNs)with sequential characteristics.Commonly used RNNs such as long short-term memory model and gated recurrent unit(GRU)model have a memory function that appropriately retains previous results for use in processing current results.To construct a suitable RNN model,this study used Taguchi method to optimize hyperparameters(including hidden layer neuron count,iteration count,batch size,learning rate,and dropout ratio).Additionally,optimization performance was also compared between 5-layer and 7-layer network architectures.The experimental results revealed that the 7-layer GRU was more suitable for the application considered in this study.The values obtained in tests of prediction performance were mean absolute percentage error of 3.7134%,root mean square error of 0.0638,and R-value of 0.9984.Therefore,thewater qualitymanagement system developed in this study can quickly provide practitioners with highly accurate data,which is essential for a timely response to water quality issues.This study was performed in collaboration with the Taiwan Industrial Technology Research Institute and a local fishery company.Practical application of the system by the fishery company confirmed that the monitoring system is effective in improving the survival rate of farmed fish by providing data needed to maintain DOC higher than the standard value.
基金funded by the National Natural Science Foundation of China(No.51775185),Natural Science Foundation of Hunan Province(2022JJ90013)Scientific Research Fund of Hunan Province Education Department(18C0003)+1 种基金Research project on teaching reform in colleges and universities of Hunan Province Education Department(20190147)Hunan Normal University University-Industry Cooperation.This work is implemented at the 2011 Collaborative Innovation Center for Development and Utilization of Finance and Economics Big Data Property,Universities of Hunan Province,Open project,Grant Number 20181901CRP04.
文摘Water resources are an indispensable and valuable resource for human survival and development.Water quality predicting plays an important role in the protection and development of water resources.It is difficult to predictwater quality due to its random and trend changes.Therefore,amethod of predicting water quality which combines Auto Regressive Integrated Moving Average(ARIMA)and clusteringmodelwas proposed in this paper.By taking thewater qualitymonitoring data of a certain river basin as a sample,thewater quality Total Phosphorus(TP)index was selected as the prediction object.Firstly,the sample data was cleaned,stationary analyzed,and white noise analyzed.Secondly,the appropriate parameters were selected according to the Bayesian Information Criterion(BIC)principle,and the trend component characteristics were obtained by using ARIMA to conduct water quality predicting.Thirdly,the relationship between the precipitation and the TP index in themonitoring water field was analyzed by the K-means clusteringmethod,and the random incremental characteristics of precipitation on water quality changes were calculated.Finally,by combining with the trend component characteristics and the random incremental characteristics,the water quality prediction results were calculated.Compared with the ARIMA water quality prediction method,experiments showed that the proposed method has higher accuracy,and its Mean Absolute Error(MAE),Mean Square Error(MSE),and Mean Absolute Percentage Error(MAPE)were respectively reduced by 44.6%,56.8%,and 45.8%.
文摘The quality of a water body is usually characterized by sets of physical, chemical, and biological parameters, which are mutually interrelated. Since August 1997, monthly records of 33 parameters, monitored at 102 locations on the Nile Delta drainage system, are stored in a National Database operated by the Drainage Research Institute (DRI). Correlation patterns may be found between water quantity and water quality parameters at the same location, or among water quality parameters within a monitoring location or among locations. Serial correlation is also detected in water quality variables. Through the investigation of the level of information redundancy, assessment and redesign of water quality monitoring network aim to improve the overall network efficiency and cost effectiveness. In this study, the potential of the Artificial Neural Network (ANN) on simulating interrelation between water quality parameters is examined. Several ANN inputs, structures and training possibilities are assessed and the best ANN model and modeling procedure is selected. The prediction capabilities of the ANN are compared with the linear regression models with autocorrelated residuals, usually used for this purpose. It is concluded that the ANN models are more accurate than the linear regression models having the same inputs and output.
基金Supported by the National Natural Science Foundation of China(Nos.41807247,41807229)the Special Fund for Shandong Post-doctoral Innovation Project。
文摘Artificial Neural Network(ANN)models have been extensively applied in the prediction of water resource variables,and Geographical Information System(GIS)includes powerful functions to visualize spatial data.In order to provide an efficient tool for environmental assessment and management that combines the advantages of these two modules,a GIS-based ANN water quality prediction system was developed in the present study.The ANN module and ArcGIS Engine module,along with a dynamic database,were imbedded in the system,which integrates water quality prediction via the ANN model and spatial presentation of the model results.The structure of the ANN model could be modified through the graphical user interface to optimize the model performance.The developed system was applied to a real case study for the prediction of the total phosphorus concentration in the Lake Champlain area.The prediction results were verified with the monitoring data,and the performance of the developed model was further evaluated through graphical techniques and quantitative statistical methods.Overall,the developed system provided satisfactory prediction results,and spatial distribution maps of the predicted results were obtained,which coincided with the monitored values.The developed GIS-based ANN water quality prediction system could serve as an efficient tool for engineers and decision makers.
基金funded by Fujian Science and Technology Key Project(No.2016H6022,2018J01099,2017H0037)
文摘The fraction defective of semi-finished products is predicted to optimize the process of relay production lines, by which production quality and productivity are increased, and the costs are decreased. The process parameters of relay production lines are studied based on the long-and-short-term memory network. Then, the Keras deep learning framework is utilized to build up a short-term relay quality prediction algorithm for the semi-finished product. A simulation model is used to study prediction algorithm. The simulation results show that the average prediction absolute error of the fraction is less than 5%. This work displays great application potential in the relay production lines.
文摘Chaos theory was introduced for water quality, prediction, and the model of water quality prediction was established by combining phase space reconstruction theory and BP neural network forecasting method. Through the phase space reconstruction, the one-dimensional water quality time series were mapped to be multi-dimensional sequence, which enriched the spatial information of water quality change and expanded mapping region of training samples of BP neural network. Established model of combining chaos theory and BP neural network were applied to forecast turbidity time series of a certain reservoir. Contrast to BP neural network method, the relative error and the mean squared error of the combined method had all varying degrees of lower. Results indicated the neural network model with chaos theory had the higher prediction accuracy, at the same time, it had better fault-tolerant capability and generalization performance .
基金Supported by China National Science and Technology Major Project(2016ZX05016-006)
文摘A deep learning method for predicting oil field production at ultra-high water cut stage from the existing oil field production data was presented,and the experimental verification and application effect analysis were carried out.Since the traditional Fully Connected Neural Network(FCNN)is incapable of preserving the correlation of time series data,the Long Short-Term Memory(LSTM)network,which is a kind of Recurrent Neural Network(RNN),was utilized to establish a model for oil field production prediction.By this model,oil field production can be predicted from the relationship between oil production index and its influencing factors and the trend and correlation of oil production over time.Production data of a medium and high permeability sandstone oilfield in China developed by water flooding was used to predict its production at ultra-high water cut stage,and the results were compared with the results from the traditional FCNN and water drive characteristic curves.The LSTM based on deep learning has higher precision,and gives more accurate production prediction for complex time series in oil field production.The LSTM model was used to predict the monthly oil production of another two oil fields.The prediction results are good,which verifies the versatility of the method.
基金the National Natural Science Foundation of China(No.51775185)Natural Science Foundation of Hunan Province(No.2022JJ90013)+1 种基金Intelligent Environmental Monitoring Technology Hunan Provincial Joint Training Base for Graduate Students in the Integration of Industry and Education,and Hunan Normal University University-Industry Cooperation.the 2011 Collaborative Innovation Center for Development and Utilization of Finance and Economics Big Data Property,Universities of Hunan Province,Open Project,Grant Number 20181901CRP04.
文摘At present,water pollution has become an important factor affecting and restricting national and regional economic development.Total phosphorus is one of the main sources of water pollution and eutrophication,so the prediction of total phosphorus in water quality has good research significance.This paper selects the total phosphorus and turbidity data for analysis by crawling the data of the water quality monitoring platform.By constructing the attribute object mapping relationship,the correlation between the two indicators was analyzed and used to predict the future data.Firstly,the monthly mean and daily mean concentrations of total phosphorus and turbidity outliers were calculated after cleaning,and the correlation between them was analyzed.Secondly,the correlation coefficients of different times and frequencies were used to predict the values for the next five days,and the data trend was predicted by python visualization.Finally,the real value was compared with the predicted value data,and the results showed that the correlation between total phosphorus and turbidity was useful in predicting the water quality.
基金financially supported by the National Key R&D Project(No.2022YFC3203203)the Shaanxi Province Science Fund for Distinguished Young Scholars(No.S2023-JC-JQ-0036).
文摘Neural networks(NNs)have been used extensively in surface water prediction tasks due to computing algorithm improvements and data accumulation.An essential step in developing an NN is the hyperparameter selection.In practice,it is common to manually determine hyperparameters in the studies of NNs in water resources tasks.This may result in considerable randomness and require significant computation time;therefore,hyperparameter optimization(HPO)is essential.This study adopted five representatives of the HPO techniques in the surface water quality prediction tasks,including the grid sampling(GS),random search(RS),genetic algorithm(GA),Bayesian optimization(BO)based on the Gaussian process(GP),and the tree Parzen estimator(TPE).For the evaluation of these techniques,this study proposed a method:first,the optimal hyperparameter value sets achieved by GS were regarded as the benchmark;then,the other HPO techniques were evaluated and compared with the benchmark in convergence,optimization orientation,and consistency of the optimized values.The results indicated that the TPE-based BO algorithm was recommended because it yielded stable convergence,reasonable optimization orientation,and the highest consistency rates with the benchmark values.The optimization consistency rates via TPE for the hyperparameters hidden layers,hidden dimension,learning rate,and batch size were 86.7%,73.3%,73.3%,and 80.0%,respectively.Unlike the evaluation of HPO techniques directly based on the prediction performance of the optimized NN in a single HPO test,the proposed benchmark-based HPO evaluation approach is feasible and robust.
基金supported by the National Key R&D Program of China(No.2021YFC1809001).
文摘Efficiently predicting effluent quality through data-driven analysis presents a significant advancement for consistent wastewater treatment operations.In this study,we aimed to develop an integrated method for predicting effluent COD and NH3 levels.We employed a 200 L pilot-scale sequencing batch reactor(SBR)to gather multimodal data from urban sewage over 40 d.Then we collected data on critical parameters like COD,DO,pH,NH_(3),EC,ORP,SS,and water temperature,alongside wastewater surface images,resulting in a data set of approximately 40246 points.Then we proposed a brain-inspired image and temporal fusion model integrated with a CNN-LSTM network(BITF-CL)using this data.This innovative model synergized sewage imagery with water quality data,enhancing prediction accuracy.As a result,the BITF-CL model reduced prediction error by over 23%compared to traditional methods and still performed comparably to conventional techniques even without using DO and SS sensor data.Consequently,this research presents a cost-effective and precise prediction system for sewage treatment,demonstrating the potential of brain-inspired models.
基金This research was supported by the National Natural Science Foundation of China(Nos.41977300 and 41907297)the Science and Technology Program of Guangzhou(China)(No.202002020055)the Fujian Provincial Natural Science Foundation(China)(No.2020I1001).
文摘Anaerobic process is regarded as a green and sustainable process due to low carbon emission and minimal energy consumption in wastewater treatment plants(WWTPs).However,some water quality metrics are not measurable in real time,thus influencing the judgment of the operators and may increase energy consumption and carbon emission.One of the solutions is using a soft-sensor prediction technique.This article introduces a water quality soft-sensor prediction method based on Bidirectional Gated Recurrent Unit(BiGRU)combined with Gaussian Progress Regression(GPR)optimized by Tree-structured Parzen Estimator(TPE).TPE automatically optimizes the hyperparameters of BiGRU,and BiGRU is trained to obtain the point prediction with GPR for the interval prediction.Then,a case study applying this prediction method for an actual anaerobic process(2500 m^(3)/d)is carried out.Results show that TPE effectively optimizes the hyperparameters of BiGRU.For point prediction of CODeff and biogas yield,R^(2)values of BiGRU,which are 0.973 and 0.939,respectively,are increased by 1.03%–7.61%and 1.28%–10.33%,compared with those of other models,and the valid prediction interval can be obtained.Besides,the proposed model is assessed as a reliable model for anaerobic process through the probability prediction and reliable evaluation.It is expected to provide high accuracy and reliable water quality prediction to offer basis for operators in WWTPs to control the reactor and minimize carbon emission and energy consumption.
文摘水质指标具有多元相关性、时序性和非线性的特点,为有效预测河流水质变化,针对水质数据存在缺失和异常的问题,提出基于灰色关联分析-门控循环单元(Grey Relational Analysis-Gated Recurrent Unit, GRA-GRU)的水质预测模型。以淮河流域水质数据为样本,使用线性插值修补缺失数据和剔除的异常数据。使用灰色关联分析计算不同水质指标间的相关性,选择高相关性的水质指标以确定输入变量,并使用门控循环单元(Gated Recurrent Unit, GRU)预测不同的水质指标。将GRA-GRU的预测结果与反向传播神经网络(Back Propagation Neural Network, BPNN)、循环神经网络(Recurrent Neural Network, RNN)、长短期记忆神经网络(Long Short Term Memory, LSTM)、GRU及灰色关联分析-长短期记忆神经网络(Grey Relational Analysis-Long Short Term Memory, GRA-LSTM)进行对比分析,结果显示GRA-GRU在不同水质指标预测上具有较好的适应性,可以有效降低预测误差。其中,与其他模型相比,GRA-GRU预测的化学需氧量在均方根误差上分别降低了3.617%、0.681%、0.478%、1.505%和0.471%。