WO2021114320A1 - 一种oica和rnn融合模型的污水处理过程故障监测方法 - Google Patents

一种oica和rnn融合模型的污水处理过程故障监测方法 Download PDF

Info

Publication number
WO2021114320A1
WO2021114320A1 PCT/CN2019/125888 CN2019125888W WO2021114320A1 WO 2021114320 A1 WO2021114320 A1 WO 2021114320A1 CN 2019125888 W CN2019125888 W CN 2019125888W WO 2021114320 A1 WO2021114320 A1 WO 2021114320A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
monitoring
fault
new
drnn
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2019/125888
Other languages
English (en)
French (fr)
Inventor
常鹏
李泽宇
王凯
丁春豪
金辰
张祥宇
卢瑞炜
王普
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing University of Technology
Original Assignee
Beijing University of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing University of Technology filed Critical Beijing University of Technology
Publication of WO2021114320A1 publication Critical patent/WO2021114320A1/zh
Priority to US17/520,378 priority Critical patent/US20220155770A1/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/18—Water
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00—Testing or monitoring of control systems or parts thereof
    • G05B23/02—Electric testing or monitoring
    • G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
    • G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
    • G05B23/0243—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model
    • G05B23/0254—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00—Testing or monitoring of control systems or parts thereof
    • G05B23/02—Electric testing or monitoring
    • G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
    • G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
    • G05B23/0224—Process history based detection method, e.g. whereby history implies the availability of large amounts of data
    • G05B23/024—Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/0265—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion
    • G05B13/027—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion using neural networks only
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00—Testing or monitoring of control systems or parts thereof
    • G05B23/02—Electric testing or monitoring
    • G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
    • G05B23/0259—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the response to fault detection
    • G05B23/0275—Fault isolation and identification, e.g. classify fault; estimate cause or root of failure
    • G05B23/0281—Quantitative, e.g. mathematical distance; Clustering; Neural networks; Statistical analysis
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Definitions

  • the invention relates to the technical field of fault monitoring based on deep learning, in particular to a fault monitoring technology for complex industrial processes.
  • the deep learning-based method of the present invention is a specific application in the fault monitoring of a typical complex industrial process—a sewage treatment process.
  • the sewage treatment process is a complex dynamic biochemical process with strong external interference, strong time-varying, strong coupling, and nonlinearity. Therefore, the reliability and stability of the control system are particularly important. But for many abnormal changes (faults) that occur in the process, the controller is often powerless. Due to the continuity and irreplaceability of the sewage treatment system, once a failure occurs, it will cause serious impact. Due to the complex characteristics of the sewage treatment process mechanism and serious external environmental interference, the data of the sewage treatment process has obvious characteristics of non-linearity, non-Gaussianness and time correlation. Traditional methods are not effective in monitoring the faults in the sewage treatment process.
  • KPCA Kernel Principal Component Analysis
  • KPLS Kernel Partial Least Squares
  • ICA independent component analysis
  • the neural network Compared with the multivariate statistical method, the neural network has stronger non-linear processing capabilities, but it does not consider the non-Gaussian and time correlation of the data in the process of applying it to sewage monitoring. And the neural network method is supervised monitoring, and the label of the data will impose certain restrictions on the monitoring of the sewage treatment process.
  • This paper establishes an intelligent fault monitoring method based on the recurrent neural network enhanced by high-order information.
  • this paper chooses to use the OICA (Overcomplete Independent Component Analysis) method to extract the original data into high-level information features.
  • the OICA algorithm was proposed by Anastasia et al. of the Massachusetts Institute of Technology. The algorithm does not need to assume that the data obeys Gaussian distribution. The complexity is low, and it is not restricted by the form of the mixed matrix.
  • the feature data extracted by OICA is entered into the multi-layer recurrent neural network DRNN (Deep Recurrent Neural Network) for layer-by-layer training.
  • DRNN Deep Recurrent Neural Network
  • Recurrent neural networks can learn time series information with multiple levels of abstraction in the data, and are more sensitive to changes in the characteristics of the data, making it easier to detect faults.
  • the extracted high-level statistical information directly establishes a monitoring model for monitoring.
  • the method of OICA directly establishes monitoring is an unsupervised monitoring method. The purpose of this is to monitor that there is no existing label information.
  • the existing fault data database can be expanded on the basis of improving the monitoring accuracy, so that the monitoring results will gradually increase with the increase of time.
  • the historical data X is composed of data of the normal operating state of the sewage treatment process obtained by offline testing.
  • the data includes N sampling moments, and J samples are collected at each sampling moment.
  • mapping is a high-order feature matrix S.
  • the high-order features of the mapping can effectively reflect the non-Gaussian features of the data and can provide more fault information.
  • the steps are as follows, calculate the unmixing matrix W through OICA, and then use W to convert the original data Mapping into a high-order feature matrix S. Obtained by W
  • the formula for the high-order feature matrix S of is as follows:
  • the residual E is obtained according to S, and the formula for obtaining the residual is as follows:
  • step 6 Enter the high-order feature matrix S obtained in step 3 and the label data Y obtained in step 5 into the deep recurrent neural network DRNN for supervised training.
  • the input of the deep cyclic neural network is the high-order feature information S obtained by OICA, and the label data corresponding to the network input is the label Y obtained by the fault classification label obtained in step 5.
  • the method based on DRNN can perform supervised classification of faults very well, but when a fault that is not in the training library of the DRNN network occurs, the monitoring performance of the above method may be reduced.
  • the algorithm of the present invention proposes an unsupervised algorithm based on OICA to monitor the above-mentioned faults, so as to calibrate the monitoring results of DRNN.
  • W is the unmixing matrix determined in step 4.
  • the fault data is set up according to offline step 5 and added to the DRNN training database for training. Continuous iterative training enables the DRNN network to learn new fault information all the time.
  • the intelligent fault monitoring method based on the recurrent neural network enhanced with high-order information can handle the non-Gaussian nature of the data, improve the feature extraction ability for the original data, and the fusion recurrent neural network structure can extract different levels of
  • the sequential information of sewage data can effectively improve the accuracy of monitoring in sewage monitoring.
  • the supervised training data of the failure can be continuously improved, and the monitoring accuracy of the overall monitoring model can be improved.
  • Figure 1 is an overall flow chart of the algorithm of the present invention
  • Figure 2 is a monitoring diagram of sewage sludge expansion failure under a sunny day
  • Figure 3 is a monitoring diagram of the toxic impact failure of sewage under a sunny day
  • Figure 4 is a monitoring diagram of sewage sludge expansion failure under rainy weather
  • Figure 5 is the monitoring diagram of the toxic impact failure of sewage under rainy weather
  • Figure 6 The logic block diagram of the hardware system on which this method is based
  • Figure 7 is a schematic diagram of the network structure proposed by the method of the present invention.
  • a method for monitoring faults in the sewage treatment process based on the OICA and RNN fusion model is proposed.
  • the method is based on an online monitoring instrument.
  • the entire device includes an input module, an information processing module, a console module, and an output result visualization module.
  • the proposed method is imported into the information processing module, and then the network monitoring model is established with the process data retained by the actual industry, and the established model is saved for online fault monitoring.
  • the sewage treatment process is extremely complex, including not only various physical and chemical reactions, but also biochemical reactions.
  • various uncertain factors are flooded with it, such as influent flow, water quality and load changes, which give the sewage treatment monitoring model
  • the establishment of has brought huge challenges.
  • the present invention adopts the "benchmark simulation model 1" (benchmark simulation model 1) developed by the International Water Association (IWA) as the actual sewage treatment process for real-time simulation.
  • the model consists of five reaction vessel (5999m3) and a secondary settling tank (6000m 3) consisting, in addition to three aeration tank.
  • the aeration tank has 10 layers, is 4 meters deep, and occupies an area of 1500m 2.
  • the reaction process includes internal reflux and external reflux.
  • the average sewage treatment flow rate is 20 000 m 3 /d, and the chemical oxygen demand is 300 mg/l.
  • the effluent quality indicators of the sewage model are shown in Table 1.
  • the present invention simulates two kinds of faults based on the BSM1 model, sludge expansion fault and toxic shock fault
  • Step 1 The present invention simulates the sludge expansion fault and toxic impact fault of the sewage treatment process to verify the algorithm.
  • the BSM1 model collects data of 14 days of normal weather and heavy rain, with a sampling interval of 15 minutes, and a total of 1344 sampling points for each weather.
  • the experiment uses multiple batches of sludge expansion data and normal data of the same type with different failure degrees for offline training, and then trains a new set of single batch of sludge failure data as a test.
  • the training and test data of the simulated toxic impact failure are the same as those of the normal data.
  • the sludge expansion failure is the same.
  • Step 2 Process the offline data collected under normal working conditions of the sewage treatment process, which includes N sampling moments collected from multiple batches of data, and 16 process variables are collected to form a data matrix
  • x i (x i,1 ,x i,2 ,...,x i,j )
  • x i,j represents the measured value of the j-th variable at the i-th sampling time
  • Step 3 Then standardize the historical data X, where the standardization formula of the j-th variable at the i-th sampling time is as follows:
  • Step 4 Use the OICA algorithm mentioned above to The mapping is a high-order feature matrix S.
  • the high-order features of the mapping can effectively reflect the non-Gaussian features of the data and can provide more fault information.
  • the specific steps are as follows, calculate the unmixing matrix W through OICA, and then use W to convert the original data Mapping into a high-order feature matrix S. Obtained by W
  • the formula for the high-order feature matrix S of is as follows:
  • the residual E is obtained according to S, and the formula for obtaining the residual is as follows:
  • Step 5 Calculate the statistic I 2 of the independent component space and the statistic SPE of the residual space according to S and E respectively, as shown in the following formula:
  • Step 6 Set up label Y for historical data X afterwards. According to the fault type corresponding to X at each time, set it to 1 when the sewage treatment process is normal, and set it to 0 when the process is faulty.
  • Step 7 Enter the high-order feature matrix S obtained in step 3 and the label data Y obtained in step 5 into the deep recurrent neural network DRNN for supervised training.
  • the input of the deep cyclic neural network is the high-order feature information S obtained by OICA, and the label data corresponding to the network input is the label Y obtained by the fault classification label obtained in step 5.
  • After training save the hyperparameters and structure of the neurons in the network after the DRNN has been supervised and trained.
  • the specific neural network structure and parameters of DRNN are shown in the following table.
  • Step 8 The preprocessing method of new data during online monitoring is as offline step 3, to obtain processed new data X new
  • Step 9 Pass the new data X new through the unmixing matrix W obtained in the offline stage to obtain new high-order feature information feature data S new
  • Step 10 Use S new as the input of the network to enter the DRNN deep cyclic neural network with the network parameters trained in the offline stage for calculation.
  • an output y will be obtained.
  • y is the current judgment for us to determine whether it is faulty Indicator data. When y is greater than 0.5, it means that there is a current fault, and when y is less than 0.5, it means that the monitoring result obtained through DRNN is that there is no fault at the current moment.
  • Step 11 The DRNN-based method can perform supervised classification of faults very well, but when a fault that is not in the training library of the DRNN network occurs, the monitoring performance of the above method may be reduced. Further, the algorithm of the present invention proposes an unsupervised algorithm based on OICA to monitor the above-mentioned faults, so as to calibrate the monitoring results of DRNN.
  • OICA unsupervised algorithm based on OICA
  • W is the unmixing matrix determined in step 4.
  • Step 12 Calculate the monitoring statistics of the current sampling time k And SPE k as shown in the following formula:
  • Step 13 Convert the monitoring statistics obtained in the above steps And SPE k and the control limit obtained in step 6) Compare with SPE limit , if any of the above two indicators exceeds the limit, it will be considered as a fault and alarm; otherwise, it will be considered as normal;
  • Step 15 Set up the fault label according to the offline step 5 of the fault data and add it to the training database of the DRNN for training. Continuous iterative training enables the DRNN network to learn new fault information all the time.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Mathematical Physics (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Software Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Chemical & Material Sciences (AREA)
  • Pure & Applied Mathematics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Algebra (AREA)
  • Food Science & Technology (AREA)
  • Medicinal Chemistry (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Testing And Monitoring For Control Systems (AREA)

Abstract

一种基于高阶信息增强的循环神经网络的智能故障监测方法，用于对污水处理过程故障进行实时监测，包括离线训练和在线软测量两个阶段。离线阶段首先采用OICA把原始数据提取成为高维的高阶信息特征，用于有效的处理数据的非高斯性，并解决变量间的相关性。之后将提取的特征通过DRNN进行训练。在线阶段中，将数据直接映射成新的高阶特征分量并且通过离线训练好的DRNN网络进行分类判别。若结果无故障，再进入由单纯的OICA构成的监测模型中进行无监督监测，此时还没有监测出故障则判定为过程无故障，若发生故障则判定过程故障，并把故障信息加入网络的训练数据中进行训练，从而不断提高DRNN的监测精度。

Description

一种OICA和RNN融合模型的污水处理过程故障监测方法 技术领域
本发明涉及基于深度学习的故障监测技术领域，特别是涉及一种针对复杂工业过程的故障监测技术。本发明的基于深度学习的方法即是在典型复杂工业过程——污水处理过程故障监测方面的具体应用。
背景技术
污水处理过程是一个外界干扰强烈、时变性强、耦合性强、非线性的复杂动态生物化学过程，因此控制系统的可靠性和稳定性显得尤其重要。但对于过程中出现的很多异常变化(故障)，控制器往往无能为力.由于污水处理系统的工作连续性和不可替代性，一旦发生故障，就会造成严重影响。由于污水处理过程的处理过程机理特性复杂和外界环境干扰严重等特点，导致污水处理过程的数据带有明显的非线性、非高斯性和时间相关性等特点。传统方法在对于污水处理过程故障监测上效果欠佳。
近年来基于数据驱动的方法得到了广泛的发展，基于数据驱动的方法不需要研究污水处理过程复杂的机理知识，只是通过过程变量的变化就可以实时的得出监控结果，得到了广泛的应用。在传统的基于数据驱动的方法上，以KPCA(Kernel Principal Component Analysis，KPCA)和KPLS(Kernel Partial Least Squares，KPLS)等的多元统计学方法为主，这些方法可以提取过程的潜在特征变量，从而捕获过程变化的信息，从而反映故障的发生。基于KPCA，KPLS等的方法可以有效的处理数据的非线性，但是以上方法均需要假设过程数据服从高斯分布，在实际的工业过程的数据由于复杂环境的干扰大多是不服从高斯分布的，在实际应用中有很多的限制。为了处理数据的非高斯问题，独立成分分析(independent component analysis，ICA)被提出并广泛的应用于数据的非高斯性特征的提取上。ICA可以有效的利用数据的非高斯性提取特征。但ICA在求解的过程中要大量的迭代并且得到的解有高度的不确定性，导致ICA在应用时困难。现在缺乏一种有效的数据处理手段对于污水处理过程进行监测。近年来神经网络的方法也广泛的应用于污水的过程监测上，如BP神经网络，RBF神经网络等。相对于多元统计方法，神经网络的非线性处理能力更强，但是在应用于污水监测过程中没有考虑 数据的非高斯性和时间相关性。并且神经网络的方法为有监督的监测，数据的标签会对于污水处理的过程监测产生一定的限制。
发明内容
为了克服上述所提的两个技术要素的不足。本文建立一种基于高阶信息增强的循环神经网络的智能故障监测方法。在特征提取阶段，本文选择运用OICA(Overcomplete Independent Component Analysis)的方法把原始数据提取成为高阶信息特征，OICA算法由麻省理工学院的Anastasia等人提出，该算法无需假设数据服从高斯分布，计算复杂度低，而且不受混合矩阵形式的限制。之后将OICA提取的特征数据进入到多层的循环神经网DRNN(Deep Recurrent Neural Network)进行逐层的训练。循环神经网络可以学习数据中具有多个抽象层次的时间序列信息，对于数据的特征变化更加敏感，更易监测到故障。在通过DRNN进行监测的同时，提取的高阶统计信息直接建立监测模型进行监测，OICA直接建立监测的方法是无监督的监测方法，这样做的目的是为了监测到在已有的标签信息中没有的故障类型，可以在提高监测准确率的基础上扩充已有的故障数据数据库，使得监测结果随着时间的增加监测能力逐渐增高。
本发明采用了如下的技术方案及实现步骤：
A.离线建模阶段：
1)对采集污水处理过程正常工况下的历史数据，所述的历史数据X由离线测试得到的污水处理过程正常操作状态的数据构成，数据包含N个采样时刻，每个采样时刻采集J个过程变量形成数据矩阵
Figure PCTCN2019125888-appb-000001
Figure PCTCN2019125888-appb-000002
其中，对于每个采样时刻x i＝(x i,1,x i,2,…,x i,j)，x i,j表示第i个采样时刻的第j个变量的测量值；
2)然后对历史数据X进行标准化，其中第i个采样时刻的第j个变量的标准化公式如下：
Figure PCTCN2019125888-appb-000003
其中，i＝1,2,…N,j＝1,2,…J；将步骤2标准化后的数据重新构造成二维矩阵，如下式所示：
Figure PCTCN2019125888-appb-000004
3)利用上述所提到的OICA算法将
Figure PCTCN2019125888-appb-000005
映射为高阶特征矩阵S，映射的高阶特征可以有效的反映数据的非高斯特征，能够提供更多的故障信息。具体
的步骤如下，通过OICA计算出解混矩阵W，之后利用W将原数据
Figure PCTCN2019125888-appb-000006
映射成为高阶特征矩阵S。通过W得到
Figure PCTCN2019125888-appb-000007
的高阶特征矩阵S的公式如下：
Figure PCTCN2019125888-appb-000008
进一步的，根据S得到残差E，求得残差的公式如下所示：
Figure PCTCN2019125888-appb-000009
4)分别根据S和E计算独立成分空间的统计量I 2和残差空间的统计量SPE，如下式所示：
I 2＝S TS
SPE＝E TE
利用核密度估计算法求得上述I 2和SPE统计量在预设置的置信限时的估计值
Figure PCTCN2019125888-appb-000010
和SPE limit，并将其作为后续运用OICA进行故障监测的控制限。
5)之后对于历史数据X设立标签Y。根据每一时刻X对应的故障类型，污水处理过程为正常时设为1，过程为故障设为0。
6)将步骤3得到的高阶特征矩阵S和步骤5得到的标签数据Y进入深度循环神经网络DRNN中进行有监督训练。深度循环神经网络的输入为OICA得到的高阶特征信息S，网络输入对应的标签数据为步骤5得到的故障分类标签为得到的标签Y。经过训练后保存DRNN经过监督训练过后网络中神经元的参数和结构。
B.在线监测阶段：
1)在线监测时新数据的预处理方式如离线的步骤2，得到处理过后的新数据X new
2)将新数据X new通过离线阶段得到的解混矩阵W得到新的高阶特征信息特征数据S new
Figure PCTCN2019125888-appb-000011
3)将S new作为网络的输入进入在离线阶段训练好的网络参数的DRNN深度 循环神经网络中进行运算，数据经过DRNN神经元的运算会得到一个输出y，y为我们得到当前评判是否故障的指标数据。当y大于0.5则表示当前故障，当y小于0.5表示通过DRNN得到的监测结果为当前时刻无故障。
4)基于DRNN的方法可以很好的对于故障进行有监督分类，但是当发生DRNN网络的训练库中没有的故障时，上述的方法的监测性能可能会降低。进一步的，本发明算法提出一种基于OICA的无监督算法对于上述故障进行监测，从而对于DRNN的监测结果进行校准。当DRNN得到的监测结果为正常时，进行二次监测，具体步骤如下，首先通过高阶统计信息S new得到新数据X new的残差E new，如下式所示：
Figure PCTCN2019125888-appb-000012
其中W为步骤4)中所确定的解混矩阵；
5)计算当前采样时刻k的监控统计量
Figure PCTCN2019125888-appb-000013
和SPE k，如下式所示：
Figure PCTCN2019125888-appb-000014
SPE k＝E new′E new
6)将上述步骤得到的监控统计量
Figure PCTCN2019125888-appb-000015
和SPE k与步骤6)得到的控制限
Figure PCTCN2019125888-appb-000016
和SPE limit进行比较，若上述两个指标中其中任意一个指标超限就认为发生故障并报警；否则即认为是正常；
7)将故障数据按照离线步骤5设立故障标签并加入DRNN的训练数据库中进行训练，不断的迭代训练使DRNN网络可以一直学习到新的故障信息。
有益效果
与现有技术相比，基于高阶信息增强的循环神经网络的智能故障监测方法能够处理数据的非高斯性，提高了对于原始数据的特征提取能力，并且融合循环神经网络结构可以提取不同层次的污水数据的时序性信息，在污水的监测上可以有效的提高监测的精确度。并且通过同时进行监测的OICA无监督模型的监测校准，可以不断的提高故障的有监督训练数据，提升整体监测模型的监测精度。
附图说明
图1为本发明算法的整体流程图；
图2为晴天下对于污水污泥膨胀故障的监测图；
图3为晴天下对污水毒性冲击故障的监测图；
图4为雨天下对于污水污泥膨胀故障的监测图；
图5为雨天下对污水毒性冲击故障的监测图；
图6本方法依托的硬件系统逻辑框图；
图7为本发明方法提出的网络结构示意图。
具体实施方式
为了解决上述问题，提出一种OICA和RNN融合模型的污水处理过程故障监测方法，该方法基于一种在线监测仪器设备。整个设备包含输入模块，信息处理模块，控制台模块，输出结果可视化模块。将所提出的方法导入信息处理模块，然后用实际工业所保留的过程数据建立网络监测模型，所建立的模型保存下来，用于在线故障监测。在实际的工业过程在线监测时，首先将工厂数据传感器采集到的实时过程变量连接到输入模块，作为监测设备的输入信息，然后通过控制台选择之前所训练好的模型进行监测，并将监测结果通过可视化模块实时的显示出来，以便现场工作人员能够根据可视化监测结果及时做出相应的措施，以减少由于过程故障带来的经济损失。
污水处理过程极其复杂，不仅包含了各种物理、化学，而且也包括了生化反应，此外，各种不确定性因素充斥其中，如进水流量、水质和负荷变化等，这给污水处理监控模型的建立带来了巨大的挑战。本发明采用国际水协会(IWA)研发的“仿真基准模型”(Benchmark Simulation Model 1)作为实际污水处理过程进行实时仿真。该模型由五个反应池(5999m3)和一个二沉池(6000m 3)组成，此外还有三个曝气池。曝气池有10层，深4米，占地1500m 2，反应过程有内回流和外回流。平均污水处理流量为20 000m 3/d，化学需氧量为300mg/l。污水模型的出水水质指标如表1所示。在模型故障设置上本发明基于BSM1模型模拟两种故障，污泥膨胀故障和毒性冲击故障
表1 污水出水指标
Figure PCTCN2019125888-appb-000017
Figure PCTCN2019125888-appb-000018
本发明在上述BSM1仿真平台的应用过程具体陈述如下：
A.离线建模阶段：
步骤1：本发明模拟污水处理过程的污泥膨胀故障和毒性冲击故障对于算法进行验证。BSM1模型采集到正常天气和暴雨14天的数据，采样间隔为15min，每个天气的总采样点为1344个。实验采用同类型下的多批不同故障度的污泥膨胀数据和正常数据进行离线训练，并再训练一组新的单批污泥故障数据作为测试，模拟的毒性冲击故障的训练和测试数据与污泥膨胀故障相同。
步骤2：对采集污水处理过程正常工况下的离线数据进行处理，其包含多批数据采集到的N个采样时刻，和采集16个过程变量形成数据矩阵
Figure PCTCN2019125888-appb-000019
Figure PCTCN2019125888-appb-000020
其中，对于每个采样时刻x i＝(x i,1,x i,2,…,x i,j)，x i,j表示第i个采样时刻的第j个变量的测量值；
步骤3：然后对历史数据X进行标准化，其中第i个采样时刻的第j个变量的标准化公式如下：
Figure PCTCN2019125888-appb-000021
其中，i＝1,2,…N,j＝1,2,…J；将步骤2标准化后的数据重新构造成二维矩阵，如下式所示：
Figure PCTCN2019125888-appb-000022
步骤4：利用上述所提到的OICA算法将
Figure PCTCN2019125888-appb-000023
映射为高阶特征矩阵S，映射的高阶特征可以有效的反映数据的非高斯特征，能够提供更多的故障信息。具体的步骤如下，通过OICA计算出解混矩阵W，之后利用W将原数据
Figure PCTCN2019125888-appb-000024
映射成为高阶特征矩阵S。通过W得到
Figure PCTCN2019125888-appb-000025
的高阶特征矩阵S的公式如下：
Figure PCTCN2019125888-appb-000026
进一步的，根据S得到残差E，求得残差的公式如下所示：
Figure PCTCN2019125888-appb-000027
步骤5：分别根据S和E计算独立成分空间的统计量I 2和残差空间的统计量SPE，如下式所示：
I 2＝S TS
SPE＝E TE
利用核密度估计算法求得上述I 2和SPE统计量在预设置的置信限时的估计值
Figure PCTCN2019125888-appb-000028
和SPE limit，并将其作为后续运用OICA进行故障监测的控制限。
步骤6：之后对于历史数据X设立标签Y。根据每一时刻X对应的故障类型，污水处理过程为正常时设为1，过程为故障设为0。
步骤7：将步骤3得到的高阶特征矩阵S和步骤5得到的标签数据Y进入深度循环神经网络DRNN中进行有监督训练。深度循环神经网络的输入为OICA得到的高阶特征信息S，网络输入对应的标签数据为步骤5得到的故障分类标签为得到的标签Y。经过训练后保存DRNN经过监督训练过后网络中神经元的超参数和结构。DRNN的具体神经网络结构及参数如下表所示。
表1 DRNN的网络结构及超参数
Figure PCTCN2019125888-appb-000029
Figure PCTCN2019125888-appb-000030
B.在线监测阶段：
步骤8：在线监测时新数据的预处理方式如离线的步骤3，得到处理过后的新数据X new
步骤9：将新数据X new通过离线阶段得到的解混矩阵W得到新的高阶特征信息特征数据S new
Figure PCTCN2019125888-appb-000031
步骤10：将S new作为网络的输入进入在离线阶段训练好网络参数的DRNN深度循环神经网络中进行运算，数据经过DRNN神经元的运算会得到一个输出y，y为我们得到当前评判是否故障的指标数据。当y大于0.5则表示当前故障，当y小于0.5表示通过DRNN得到的监测结果为当前时刻无故障。
步骤11：基于DRNN的方法可以很好的对于故障进行有监督分类，但是当发生DRNN网络的训练库中没有的故障时，上述的方法的监测性能可能会降低。进一步的，本发明算法提出一种基于OICA的无监督算法对于上述故障进行监测，从而对于DRNN的监测结果进行校准。当DRNN预测为正常时，做二次监测，监测步骤如下，首先通过高阶统计信息S new得到新数据X new的残差E new，如下式所示：
Figure PCTCN2019125888-appb-000032
其中W为步骤4)中所确定的解混矩阵；
步骤12：计算当前采样时刻k的监控统计量
Figure PCTCN2019125888-appb-000033
和SPE k，如下式所示：
Figure PCTCN2019125888-appb-000034
SPE k＝E new′E new
步骤13：将上述步骤得到的监控统计量
Figure PCTCN2019125888-appb-000035
和SPE k与步骤6)得到的控制限
Figure PCTCN2019125888-appb-000036
和SPE limit进行比较，若上述两个指标中其中任意一个指标超限就认为发生故障并报警；否则即认为是正常；
步骤15：将故障数据按照离线步骤5设立故障标签并加入DRNN的训练数据库中进行训练，不断的迭代训练使DRNN网络可以一直学习到新的故障信息。
以上即为本发明再BSM1污水仿真平台上的污水处理过程故障监测的具体应用步骤，为了验证本方法的有效性，本发明在污水的晴天和雨水下分别设置污泥膨胀和毒性冲击两种故障，检验本发明在不同天气下的监测准确性。图2-图5分别是晴天和雨天下污泥膨胀的监测图，图中离散化的分类值中1代表故障发生。表1为故障的报警时间及误报率和漏报率。从图2-5和表1中可以看出，本发明方法对于污泥的故障发生可以有效的监测，同时具有较低的漏报率和误报率。并且在雨天较复杂的环境下也有很好的监测性能，表明本发明的鲁棒性较强。
表2 不同条件下本发明的监测性能
故障类型 故障时间 报警时间 误警数 漏警数
晴天污泥膨胀故障 672-864 672 0 1
晴天毒性冲击故障 672-864 672 3 1
雨天污泥膨胀故障 672-864 672 1 2
雨天毒性冲击故障 672-864 672 0 1

Claims (1)

  1. 一种OICA和RNN融合模型的污水处理过程故障监测方法，包括“离线建模”和“在线监测”两个阶段，具体步骤如下：
    A.离线建模阶段：
    1)采集污水处理过程的历史数据，所述的历史数据X由离线测试得到的污水处理过程正常的数据构成，数据包含N个采样时刻，每个采样时刻采集J个过程变量形成数据矩阵
    Figure PCTCN2019125888-appb-100001
    其中，x i＝(x i,1,x i,2,…,x i,j)，x i,j表示第i个采样时刻的第j个变量的测量值；
    2)然后对历史数据X进行标准化，其中第i个采样时刻的第j个变量的标准化公式如下：
    Figure PCTCN2019125888-appb-100002
    其中，i＝1,2,…N,j＝1,2,…J；将步骤2标准化后的数据重新构造成二维矩阵，如下式所示：
    Figure PCTCN2019125888-appb-100003
    3)利用OICA算法将
    Figure PCTCN2019125888-appb-100004
    映射为高阶特征矩阵S，具体的步骤如下，通过OICA计算出解混矩阵W，之后利用W将原数据
    Figure PCTCN2019125888-appb-100005
    映射成为高阶特征矩阵S，通过W得到
    Figure PCTCN2019125888-appb-100006
    的高阶特征矩阵S的公式如下：
    Figure PCTCN2019125888-appb-100007
    进一步的，根据S得到残差E，求得残差的公式如下所示：
    Figure PCTCN2019125888-appb-100008
    4)分别根据S和E计算独立成分空间的统计量I 2和残差空间的统计量SPE，如下式所示：
    I 2＝S TS
    SPE＝E TE
    利用核密度估计算法求得上述I 2和SPE统计量在预设置的置信限时的估计值
    Figure PCTCN2019125888-appb-100009
    和SPE limit，并将其作为后续运用OICA进行故障监测的控制 限；
    5)之后对于历史数据X设立标签Y，即正常、故障两种。
    6)将步骤3得到的高阶特征矩阵S和步骤5得到的标签数据Y输入深度循环神经网络DRNN中进行有监督训练；经过训练后保存DRNN经过监督训练过后网络中神经元的参数和结构。
    B.在线监测阶段：
    1)在线监测时新数据的预处理方式如离线的步骤2，得到处理过后的新数据X new；
    2)将新数据X new通过离线阶段得到的解混矩阵W得到新的高阶特征信息特征数据S new
    Figure PCTCN2019125888-appb-100010
    3)将S new输入离线阶段训练好的DRNN深度循环神经网络中当输出的故障指标数据大于0.5，则表示当前故障，当输出的故障指标数据小于0.5则表示当前正常；
    4)当DRNN深度循环神经网络预测结果为正常时，需要进行二次监测：首先计算数据X new的残差E new，如下式所示：
    Figure PCTCN2019125888-appb-100011
    其中W为离线阶段得到的解混矩阵；
    5)计算当前采样时刻k的监控统计量
    Figure PCTCN2019125888-appb-100012
    和SPE k，如下式所示：
    Figure PCTCN2019125888-appb-100013
    SPE k＝E new′E new
    6)将上述步骤得到的监控统计量
    Figure PCTCN2019125888-appb-100014
    和SPE k与离线监测阶段步骤6)得到的控制限
    Figure PCTCN2019125888-appb-100015
    和SPE limit进行比较，若上述两个指标中其中任意一个指标超限就认为发生故障并报警；否则即认为是正常；
    7)将故障数据按照离线步骤5所述增加故障标签，并加入DRNN的训练数据库，利用更新后的训练数据再次训练DRNN网络，用于不断学习新的故障信息，从而更加准确的进行监测。
    8)根据权利要求1所述的故障监测方法，其特征在于：DRNN深度循环神 经网络的损失函数为交叉熵损失函数。
PCT/CN2019/125888 2019-12-14 2019-12-17 一种oica和rnn融合模型的污水处理过程故障监测方法 Ceased WO2021114320A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/520,378 US20220155770A1 (en) 2019-12-14 2021-11-05 Method of Fault Monitoring of Sewage Treatment Process Based on OICA and RNN Fusion Model

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911298706.X 2019-12-14
CN201911298706.XA CN111122811A (zh) 2019-12-14 2019-12-14 一种oica和rnn融合模型的污水处理过程故障监测方法

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/520,378 Continuation US20220155770A1 (en) 2019-12-14 2021-11-05 Method of Fault Monitoring of Sewage Treatment Process Based on OICA and RNN Fusion Model

Publications (1)

Publication Number Publication Date
WO2021114320A1 true WO2021114320A1 (zh) 2021-06-17

Family

ID=70498124

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/125888 Ceased WO2021114320A1 (zh) 2019-12-14 2019-12-17 一种oica和rnn融合模型的污水处理过程故障监测方法

Country Status (3)

Country Link
US (1) US20220155770A1 (zh)
CN (1) CN111122811A (zh)
WO (1) WO2021114320A1 (zh)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
IT201900015491A1 (it) * 2019-09-03 2021-03-03 St Microelectronics Srl Procedimento di elaborazione dati di serie temporali, sistema di elaborazione, dispositivo e prodotto informatico corrispondenti
CN111901158B (zh) * 2020-07-14 2023-07-25 广东好太太智能家居有限公司 一种智能家居配网故障数据分析方法、设备及存储介质
CN112631255B (zh) * 2020-12-28 2022-10-28 北京工业大学 一种基于变分自编码器模型的污水处理过程故障监测方法
CN113239957A (zh) * 2021-04-08 2021-08-10 同济大学 一种突发水污染事件在线识别方法
CN117194880B (zh) * 2023-09-11 2025-02-11 上海大学 基于rnn的互补双残差生成器的故障监测方法和系统
CN118039027B (zh) * 2024-01-23 2025-05-09 江苏医药职业学院 一种基于循环神经网络的盐酸多西环素废水处理方法
CN118604074B (zh) * 2024-08-07 2024-11-15 上海科泽智慧环境科技有限公司 利用电化学生物传感器阵列的污染物监测系统及方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3301428B2 (ja) * 2000-03-09 2002-07-15 株式会社 小川環境研究所 廃水処理試験方法
CN105740619A (zh) * 2016-01-28 2016-07-06 华南理工大学 基于核函数的加权极限学习机污水处理在线故障诊断方法
CN106056127A (zh) * 2016-04-07 2016-10-26 江南大学 一种带模型更新的gpr在线软测量方法
CN107741738A (zh) * 2017-10-20 2018-02-27 重庆华绿环保科技发展有限责任公司 一种污水处理过程监控智能预警云系统及污水处理监测预警方法
CN110088619A (zh) * 2017-10-09 2019-08-02 Bl科技有限责任公司 用于废水处理厂或饮用水厂的过程和资产健康诊断、异常检测和控制的智能系统和方法
CN110119579A (zh) * 2019-05-16 2019-08-13 北京工业大学 一种基于oica的复杂工业过程故障监测方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2008148075A1 (en) * 2007-05-24 2008-12-04 Alexander George Parlos Machine condition assessment through power distribution networks
CN100565403C (zh) * 2007-09-26 2009-12-02 东北大学 一种非线性过程故障诊断方法
JP2013516258A (ja) * 2010-01-11 2013-05-13 ユニヴェルシテ ドゥ モンス 脳波計信号から人工肢運動を決定する方法
CN102411308B (zh) * 2011-12-24 2013-07-10 北京工业大学 基于递归神经网络模型的溶解氧的自适应控制方法
CN107895224B (zh) * 2017-10-30 2022-03-15 北京工业大学 一种基于扩展核熵负载矩阵的mkeca发酵过程故障监测方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3301428B2 (ja) * 2000-03-09 2002-07-15 株式会社 小川環境研究所 廃水処理試験方法
CN105740619A (zh) * 2016-01-28 2016-07-06 华南理工大学 基于核函数的加权极限学习机污水处理在线故障诊断方法
CN106056127A (zh) * 2016-04-07 2016-10-26 江南大学 一种带模型更新的gpr在线软测量方法
CN110088619A (zh) * 2017-10-09 2019-08-02 Bl科技有限责任公司 用于废水处理厂或饮用水厂的过程和资产健康诊断、异常检测和控制的智能系统和方法
CN107741738A (zh) * 2017-10-20 2018-02-27 重庆华绿环保科技发展有限责任公司 一种污水处理过程监控智能预警云系统及污水处理监测预警方法
CN110119579A (zh) * 2019-05-16 2019-08-13 北京工业大学 一种基于oica的复杂工业过程故障监测方法

Also Published As

Publication number Publication date
CN111122811A (zh) 2020-05-08
US20220155770A1 (en) 2022-05-19

Similar Documents

Publication Publication Date Title
WO2021114320A1 (zh) 一种oica和rnn融合模型的污水处理过程故障监测方法
CN108803520B (zh) 一种基于变量非线性自相关性剔除的动态过程监测方法
CN102705303B (zh) 一种基于残差与双级Elman神经网络的液压伺服系统故障定位方法
CN101446831B (zh) 一种分散的过程监测方法
CN105629958B (zh) 一种基于子时段mpca‑svm的间歇过程故障诊断方法
CN108898215A (zh) 一种基于二型模糊神经网络的污泥膨胀智能辨识方法
CN106769052A (zh) 一种基于聚类分析的机械系统滚动轴承智能故障诊断方法
CN107544477A (zh) 基于核主元分析的非线性工业过程故障检测方法
CN102880809A (zh) 基于关联向量回归模型的聚丙烯熔融指数在线检测方法
CN108921230A (zh) 基于类均值核主元分析和bp神经网络的故障诊断方法
CN108334943A (zh) 基于主动学习神经网络模型的工业过程半监督软测量建模方法
CN103389701B (zh) 基于分布式数据模型的厂级过程故障检测与诊断方法
CN112904810A (zh) 基于有效特征选择的流程工业非线性过程监测方法
CN109298633A (zh) 基于自适应分块非负矩阵分解的化工生产过程故障监测方法
CN114280935A (zh) 一种基于信息熵的半监督fcm和sae的多阶段发酵过程故障监测方法
CN109472097B (zh) 一种输电线路在线监测设备故障诊断方法
CN112651444B (zh) 一种基于自学习的非平稳过程异常检测方法
Wu et al. Process monitoring of nonlinear uncertain systems based on part interval stacked autoencoder and support vector data description
CN115963788A (zh) 多采样率工业过程关键质量指标在线预测方法
CN118094436A (zh) 一种专故障敏感的自监督装备异常检测方法及系统
CN114879612A (zh) 一种基于Local-DBKSSA的高炉炼铁过程监测方法
CN116109039A (zh) 一种数据驱动异常检测与预警系统
CN118941071A (zh) 基于认知计算的工业安全风险评估与管理系统及方法
CN104142680B (zh) 一种基于鲁棒输入训练神经网络的多传感器故障诊断系统及方法
CN117633690A (zh) 一种基于数据驱动的旋转机械健康状态监测方法及设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19955698

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19955698

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 07.02.2023)

122 Ep: pct application non-entry in european phase

Ref document number: 19955698

Country of ref document: EP

Kind code of ref document: A1