WO2022052510A1 - Anomaly detection system and method for sterile filling production line - Google Patents

Anomaly detection system and method for sterile filling production line Download PDF

Info

Publication number
WO2022052510A1
WO2022052510A1 PCT/CN2021/095632 CN2021095632W WO2022052510A1 WO 2022052510 A1 WO2022052510 A1 WO 2022052510A1 CN 2021095632 W CN2021095632 W CN 2021095632W WO 2022052510 A1 WO2022052510 A1 WO 2022052510A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
mahalanobis distance
production line
information
filling production
Prior art date
Application number
PCT/CN2021/095632
Other languages
French (fr)
Chinese (zh)
Inventor
彭力
李贝贝
李稳
朱凤增
张连富
何子琎
Original Assignee
江南大学
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 江南大学 filed Critical 江南大学
Publication of WO2022052510A1 publication Critical patent/WO2022052510A1/en

Links

Images

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00Programme-control systems
    • G05B19/02Programme-control systems electric
    • G05B19/418Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • G05B19/41875Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by quality surveillance of production
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00Program-control systems
    • G05B2219/30Nc systems
    • G05B2219/31From computer integrated manufacturing till monitoring
    • G05B2219/31356Automatic fault detection and isolation
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02PCLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Definitions

  • the invention relates to a detection system and method, in particular to a system and method for abnormality detection of an aseptic filling production line.
  • the technical problem to be solved by the present invention is to provide a system and method for abnormality detection of aseptic filling production line, which combines neural network and Internet of Things data, greatly improves the accuracy of abnormality detection, and has good stability and robustness. A more excellent fault warning effect is achieved.
  • the present invention provides an abnormality detection system for an aseptic filling production line, including front-end equipment for collecting and uploading information; Front-end equipment management platform; an information service platform for judging information travel through the self-coding network and finding abnormal information.
  • the front-end equipment further includes a plurality of sensors distributed on the aseptic filling production line and set for the production environment.
  • the sensors are connected in series in the form of an RS485 bus.
  • the front-end equipment management platform further includes a sensor information processing module, a sensor association module, a sensor fault self-diagnosis module and a network communication module;
  • the sensor information processing module is used to help the sensor to follow the correct Frequency acquisition data, filter the data and eliminate blank data;
  • the sensor association module is used to integrate the data collected by each sensor, add time and space tags to it, and form the same set of data;
  • the sensor fault self-diagnosis module is used for When the sensor itself fails, an alarm signal is generated, and the information service platform is notified to give an alarm;
  • the network communication module is used for sending data to the information service platform.
  • it further includes enterprise customers for displaying data collected by sensors, querying and displaying historical data, viewing sensor operating status, viewing the operating status of various functions of the aseptic filling production line, and alerting abnormal data. end.
  • the enterprise client is a web display page
  • the information service platform is deployed on a cloud server
  • the web display page is communicatively connected to the information service platform.
  • a method for detecting abnormality in an aseptic filling production line comprising the following steps:
  • S2 preprocess the collected historical data, divide the collected historical data into normal data and obvious abnormal data according to the actual production line operation status, and complete normalization and feature construction for the normal data;
  • the judgment threshold of the Mahalanobis distance is obtained from normal data. If the Mahalanobis distance is not within the threshold interval, the data is determined to be abnormal data; if the Mahalanobis distance is within the threshold interval, the data is added with its Mahalanobis distance feature and recorded as Uncertain data is input to the auto-encoder, and the sparsity limit is added for unsupervised training, and the parameters and dimensions of each layer are adjusted to obtain the optimal hidden layer expression;
  • S4 combine the optimal hidden layer expression with the Sigmoid classifier to construct an auto-encoding network, and perform supervised fine-tuning after inputting the labeled data into the self-encoding network to obtain the optimal parameters and complete the construction of the self-encoding network;
  • step S3 the method for obtaining the discrimination threshold of the Mahalanobis distance is:
  • ⁇ -1 is the inverse of the covariance matrix
  • the covariance matrix is ⁇ N
  • the Mahalanobis distance of each data in the normal data set calculated according to the above formula is recorded as:
  • M N (M 1 ,M 2 ,...M q ,...,M l )
  • M q represents the Mahalanobis distance of the qth normal data in the normal data set
  • the mean and covariance matrix used in calculating the Mahalanobis distance of all data are still ⁇ N and ⁇ N in the normal data set, then the Mahalanobis distance of the data Xi is recorded as:
  • the Mahalanobis distance of most normal data is distributed in In the interval, the Mahalanobis distance of some abnormal data is not in the above interval, so the initial detection of abnormal data can be carried out through the Mahalanobis distance of the data, and the Mahalanobis distance discrimination thresholds T up and T low are set, and the Mahalanobis distance is not in (
  • the data in the interval T low , T up ) is determined as abnormal data, and the expressions of T up and T low are as follows:
  • the working process of the self-encoder includes an encoding process and a decoding process.
  • n is the number of data
  • the dimension of each data is m
  • each data X i is expressed in the hidden layer through the encoding process.
  • the encoding process can be described as:
  • W and b are the coding weights and biases, and ⁇ e is the activation function of the coding layer;
  • the hidden layer expresses the reconstructed data X i ' through the decoding process, and the decoding process can be described as:
  • W' and b' are decoding weights and biases
  • W' W T
  • ⁇ d is the activation function of the decoding layer.
  • step S3 the method for obtaining the hidden layer expression in step S3 is further included:
  • the weights and biases are adjusted by a layer-by-layer greedy algorithm to minimize the reconstruction error.
  • the cost function for the entire training dataset is:
  • L is the loss function of a single data
  • L is the mean square error loss function
  • the uncertain data is used as the input layer data of the autoencoder for training, so that the value of the formula loss function is the smallest, and then the best hidden layer expression is obtained.
  • the front-end equipment can collect the production environment information of each node of the aseptic filling production line, and upload it to the front-end equipment management platform. , screening, and finally find out the abnormal information by the self-encoding network constructed in the information service platform.
  • the auto-encoding network can reconstruct the data in the process of image processing, anomaly detection, fault prediction, data classification, etc., and perform feature extraction to obtain the hidden layer. It has the characteristics of better representative data when it is expressed, which can greatly improve the accuracy, stability and robustness of abnormal detection of aseptic filling production line, and achieve more excellent fault warning effect.
  • This method solves the linkage problem between neural network and IoT data.
  • a sparse auto-encoder is constructed to obtain a better hidden layer expression, and then combined with the Sigmoid classifier, an auto-encoder network is constructed for supervised fine-tuning, and a complete anomaly detection model is obtained. .
  • a complete anomaly detection model is obtained.
  • accurate detection results are obtained.
  • this method combines the self-encoding network and the Internet of Things data for intelligent analysis and detection, which can achieve the characteristics of strong comprehensiveness, high accuracy, and strong robustness.
  • FIG. 1 is a schematic structural diagram of an abnormality detection system for an aseptic filling production line in a preferred embodiment of the present invention
  • Fig. 2 is the schematic flow chart of the abnormality detection method of aseptic filling production line
  • FIG. 3 is a schematic structural diagram of an autoencoder
  • Fig. 4 is the structural representation of self-encoding network
  • Figure 5 is a schematic diagram of the fine-tuning of the autoencoder network.
  • the invention discloses an abnormality detection system for an aseptic filling production line, comprising front-end equipment for collecting and uploading information; a front-end equipment management platform for receiving information, realizing preliminary filtering and processing of information, and eliminating problem information; An information service platform that judges the information travel through the self-encoding network and finds out abnormal information.
  • the front-end equipment can collect the production environment information of each node of the aseptic filling production line and upload it to the front-end equipment management platform.
  • the self-encoding network built in the platform finds out abnormal information.
  • the auto-encoding network can reconstruct the data in the process of image processing, anomaly detection, fault prediction, data classification, etc., and perform feature extraction to obtain the hidden layer. It has the characteristics of better representative data when it is expressed, which can greatly improve the accuracy, stability and robustness of abnormal detection of aseptic filling production line, and achieve more excellent fault warning effect.
  • the above-mentioned front-end equipment includes several sensors distributed on the aseptic filling production line and set for the production environment.
  • the sensors include at least one or more of vibration sensors, voltage and current sensors, temperature sensors, humidity sensors and noise sensors.
  • vibration sensors For a normally operating aseptic filling production line, its production environment is stable, that is, vibration, voltage, current, temperature, humidity, noise and other data are kept within a normal range, and once the production environment is abnormal, these data must also be Corresponding changes will occur.
  • the front-end device management platform is connected in communication with the front-end device.
  • the front-end equipment management platform includes a sensor information processing module, a sensor association module, a sensor fault self-diagnosis module and a network communication module.
  • the sensor information processing module can help the sensor to collect data according to the correct frequency, filter the data and eliminate blank data, and enhance the anti-interference ability.
  • the sensor association module can integrate the data collected by each sensor, add time and space tags to it, and form the same set of data.
  • the sensor fault self-diagnosis module can generate an alarm signal when the sensor itself fails, and notify the information service platform to give an alarm.
  • the network communication module can send data to the information service platform.
  • the network communication module can transmit data to the information service platform through wired Ethernet data transmission.
  • the above-mentioned information service platform is in communication connection with the front-end equipment management platform.
  • the information service platform includes an IoT data processing module, an abnormal data detection module and an abnormal alarm module.
  • the IoT data processing module can receive and process the data sent by the front end, providing a data basis for the abnormal data detection module.
  • the abnormal data detection module can judge the data through the self-encoding network model, and realize the information linkage abnormal detection function.
  • the abnormal alarm module can issue an alarm when the abnormal data detection module determines abnormal data.
  • the information service platform is mainly responsible for processing the IoT data collected by the sensors, and detecting abnormal data through the IoT data combined with the self-encoding network.
  • the information service platform In order to facilitate the access of the information service platform, it can be deployed on the cloud server, which can realize access anytime and anywhere, and greatly enhance the convenience of the system.
  • an abnormality detection system for an aseptic filling production line using self-encoding network and Internet of Things data further includes an enterprise client.
  • the enterprise client communicates with the information service platform.
  • the enterprise client is used to display the data collected by the sensor, query and display historical data, check the operation status of the sensor, check the operation status of each function of the aseptic filling production line, and alarm abnormal data.
  • the information service platform detects abnormal data, it will send the alarm data to the enterprise client, so that the enterprise can know the abnormality in time and observe the data in real time.
  • Enterprise clients can display pages for the web. It communicates and connects with the information service platform on the cloud server. Using web components, functions such as data display and alarm are realized. At the same time, the web page does not require complicated installation and debugging, which further improves the convenience and intuitiveness of the system. In addition, the use of web pages for data display and alarm improves the robustness of the entire system, and is flexible and convenient to use, reflecting a user-friendly Internet thinking, which is the general trend of production line anomaly detection systems in the Internet era.
  • the enterprise client can be composed of data display interface and alarm mechanism.
  • the invention discloses a method for detecting abnormality of an aseptic filling production line, comprising the following steps:
  • S1 collect IoT data through front-end equipment. In order to collect more abnormal data and build a better self-encoding network, it is possible to collect data for a long time, or even artificially cause abnormality for a period of time.
  • S2 Preprocess the collected historical data, divide the collected historical data into normal data and obvious abnormal data according to the actual production line operation state, and complete normalization and feature construction for the normal data.
  • Obvious outliers include blank data and obviously wrong data.
  • the judgment threshold of the Mahalanobis distance is obtained from normal data. If the Mahalanobis distance is not within the threshold interval, the data is determined to be abnormal data; if the Mahalanobis distance is within the threshold interval, the data is added with its Mahalanobis distance feature and recorded as The uncertain data is input to the autoencoder, and the sparsity restriction is added for unsupervised training, and the parameters and the dimensions of each layer are adjusted to obtain the optimal hidden layer expression.
  • the optimal hidden layer expression is combined with the Sigmoid classifier to construct an auto-encoding network.
  • supervised fine-tuning is performed to obtain the optimal parameters, and the construction of the self-encoding network is completed. It solves the problem of neural network parameter initialization, shortens the training times of the classifier, and improves the accuracy of anomaly detection.
  • This method solves the linkage problem between neural networks and IoT data.
  • a sparse auto-encoder is constructed to obtain a better hidden layer expression, and then combined with the Sigmoid classifier, an auto-encoder network is constructed for supervised fine-tuning, and a complete anomaly detection model is obtained. .
  • a complete anomaly detection model is obtained.
  • accurate detection results are obtained.
  • this method combines the self-encoding network and the Internet of Things data for intelligent analysis and detection, which can achieve the characteristics of strong comprehensiveness, high accuracy, and strong robustness.
  • the Mahalanobis distance when combining IoT data with an auto-encoding network, considering the generalized distance of the correlation between variables, the Mahalanobis distance can be represented by the covariance matrix between vectors.
  • ⁇ -1 is the inverse matrix of the covariance matrix
  • the Mahalanobis distance can be regarded as the distance between the data and the mean of the overall data.
  • the Mahalanobis distance considers the correlation between data features.
  • the Mahalanobis distance of a data is smaller, it means that it is more similar to the mean data in the data set.
  • the Mahalanobis distance is more suitable for the distance of the aseptic filling production line data.
  • the Mahalanobis distance of the data is close to the Mahalanobis distance of the normal data, it means that the data is more similar to the normal data, and the data is probably normal data; if the Mahalanobis distance of the data is far from the Mahalanobis distance of the normal data, it means If the similarity between the data and normal data is small, the data is likely to be abnormal data. Therefore, the Mahalanobis distance of the data can be used to judge the possibility that the data is abnormal data.
  • the judgment threshold can be obtained from the Mahalanobis distance of the normal data samples, the Mahalanobis distance between the data and the average value of the normal data is calculated, and the data that does not exceed the judgment threshold is recorded as uncertain data and This part of the data is used for the training of autoencoders and neural networks.
  • M N (M 1 ,M 2 ,...M q ,...,M l ) (2)
  • M q represents the Mahalanobis distance of the qth normal data in the normal data set.
  • the Mahalanobis distance of all data in the entire data set is calculated according to formula (1). Since the change of a certain data will affect the change of the mean value of the data set, the Mahalanobis distance exaggerates the effect of the small change vector, thus affecting the calculation of the Mahalanobis distance of other data.
  • the mean and covariance matrix used in calculating the Mahalanobis distance of all data are still ⁇ N and ⁇ N in the normal data set, obviously independent ⁇ N and ⁇ N are not affected by vector changes ;
  • the Mahalanobis distance of a certain data can be regarded as the distance between the data and the mean of the normal data set. Then the Mahalanobis distance of the data Xi is written as:
  • T up and T up The expression for T low is as follows:
  • the data whose Mahalanobis distance is in the interval of (T low , T up ) is determined as uncertain data, which is denoted as X U , and then the self-encoding network is trained with the uncertain data set X U to complete the model construction.
  • a part of abnormal data is detected by Mahalanobis distance, and the remaining uncertain data is input into the self-encoding network.
  • the training data for the self-encoding network is reduced, and a part of abnormal data can be quickly identified according to the discrimination threshold of the Mahalanobis distance, which improves the detection efficiency.
  • the Mahalanobis distance of the aseptic filling production line data can be used to judge the possibility that the data is abnormal data, so consider the Mahalanobis distance of the data as a feature of the data. Since the Mahalanobis distance between normal data and abnormal data is quite different, in the training process of the neural network, adding the Mahalanobis distance feature as an important feature is beneficial to improve the abnormal detection effect of the neural network. Therefore, the Mahalanobis distance feature of the data is added to the data feature for the training of the self-encoding network.
  • the autoencoder mainly includes encoding and decoding stages, and the structure is symmetrical, that is, if there are multiple hidden layers, the number and structure of hidden layers in the encoding and decoding stages are the same.
  • the main structure consists of input layer, hidden layer and output layer.
  • the hidden layer encodes the input layer data
  • the output layer decodes the hidden layer expression to reconstruct the original data, minimizing the reconstruction error to obtain the best hidden layer expression.
  • the goal is to fit an identity function such that each output value is as equal as possible to the corresponding input value.
  • n is the number of data
  • the dimension of each data is m.
  • Each data X i is expressed in the hidden layer through the encoding process, and the encoding process can be described as:
  • W and b are coding weights and biases
  • ⁇ e is the activation function of the coding layer, which can be Sigmoid, Tanh, Relu, etc.
  • the hidden layer expresses the reconstructed data X i ' through the decoding process, and the decoding process can be described as:
  • W' and b' are decoding weights and biases
  • W' W T
  • ⁇ d is the activation function of the decoding layer.
  • the weights and biases are adjusted by a layer-by-layer greedy algorithm to minimize the reconstruction error.
  • the cost function for the entire training data set is:
  • L is the loss function of a single data
  • L is the mean square error loss function
  • an L2 regularization weight decay term is added to the cost function, and ⁇ is a penalty factor to control the degree to which the regularization term affects the weight decay.
  • KL divergence is added as a constraint on the basis of the autoencoder, and a sparse penalty term is added to its cost function to form a sparse autoencoder.
  • the final loss function is obtained as:
  • the uncertain data is used as the input layer data of the autoencoder for training, so that equation (9) is minimized and the best hidden layer expression is obtained.
  • an auto-encoding network is constructed, and the data is judged as normal data and abnormal data.
  • the expression of the hidden layer is obtained through the training of the autoencoder, combined with the Sigmoid classifier, the anomaly detection accuracy can be improved, and since the parameters of the hidden layer have been obtained through the training of the autoencoder, the parameter initialization problem of the neural network is solved, so that the automatic The number of training times of the encoding network is reduced, and the construction efficiency of the model is improved.
  • labeled data is then required to be input for the training process of the self-encoding network for supervised fine-tuning.
  • the error is back-propagated into the hidden layer and the Sigmoid classifier. Since the autoencoder solves the problem of parameter initialization of the classifier, the training times of the autoencoder network are reduced and the training efficiency is improved.

Landscapes

  • Engineering & Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • General Factory Administration (AREA)
  • Alarm Systems (AREA)
  • Image Analysis (AREA)

Abstract

Provided are an anomaly detection system and method for a sterile filling production line. The detection system comprises a front-end device for collecting and uploading information; a front-end device management platform for receiving the information, realizing preliminary filtering and processing of the information, and eliminating problem information; and an information service platform for determining the information by means of an autoencoder network, and finding anomalous information. The detection method comprises Internet of Things data collection (S1), data pre-processing (S2), autoencoder network construction (S4), and final differentiation of normal data and anomalous data (S5). The accuracy, stability and robustness of the system are good, and a more excellent fault early-warning effect is achieved.

Description

一种无菌灌装生产线异常检测系统和方法A system and method for abnormality detection of aseptic filling production line 技术领域technical field
本发明涉及检测系统和方法,具体涉及一种无菌灌装生产线异常检测系统和方法。The invention relates to a detection system and method, in particular to a system and method for abnormality detection of an aseptic filling production line.
背景技术Background technique
无菌灌装生产线生产过程中的异常检测一直以来是无菌灌装生产线工作时的重要工作。随着物联网技术的飞速发展,各传感器技术的日益完善,越来越精确的数据可以在无菌灌装生产设备上实时采集到,通过生产线上的各种数据进行异常检测一直是研究应用的重要方向。各种故障检测和预测方法大都基于通过物联网技术在无菌灌装生产设备上采集到的数据,而如何利用这些数据提高异常检测准确性成为重要研究课题。Abnormal detection in the production process of aseptic filling production line has always been an important task in the working of aseptic filling production line. With the rapid development of Internet of Things technology and the improvement of various sensor technologies, more and more accurate data can be collected in real time on aseptic filling production equipment. Anomaly detection through various data on the production line has always been an important research and application. direction. Various fault detection and prediction methods are mostly based on the data collected on aseptic filling production equipment through the Internet of Things technology, and how to use these data to improve the accuracy of abnormality detection has become an important research topic.
发明内容SUMMARY OF THE INVENTION
本发明要解决的技术问题是提供一种无菌灌装生产线异常检测系统和方法,结合了神经网络与物联网数据,大大提高了异常检测的准确度,且稳定性和鲁棒性良好,实现了更为优异的故障预警效果。The technical problem to be solved by the present invention is to provide a system and method for abnormality detection of aseptic filling production line, which combines neural network and Internet of Things data, greatly improves the accuracy of abnormality detection, and has good stability and robustness. A more excellent fault warning effect is achieved.
为了解决上述技术问题,本发明提供了一种无菌灌装生产线异常检测系统,包括用于采集和上传信息的前端设备;用于接收信息并实现信息的初步滤波、处理,以及剔除问题信息的前端设备管理平台;用于通过自编码网络对信息行进判断并找出异常信息的信息服务平台。In order to solve the above technical problems, the present invention provides an abnormality detection system for an aseptic filling production line, including front-end equipment for collecting and uploading information; Front-end equipment management platform; an information service platform for judging information travel through the self-coding network and finding abnormal information.
本发明一个较佳实施例中,进一步包括所述前端设备包括若干个分布在无 菌灌装生产线上并针对生产环境设置的传感器。In a preferred embodiment of the present invention, the front-end equipment further includes a plurality of sensors distributed on the aseptic filling production line and set for the production environment.
本发明一个较佳实施例中,进一步包括所述传感器采用RS485总线形式串联。In a preferred embodiment of the present invention, it further includes that the sensors are connected in series in the form of an RS485 bus.
本发明一个较佳实施例中,进一步包括所述前端设备管理平台包括传感器信息处理模块、传感器关联模块、传感器故障自诊断模块和网络通讯模块;所述传感器信息处理模块用于帮助传感器按照正确的频率采集数据,并对数据进行滤波以及剔除空白数据;所述传感器关联模块用于整合各传感器采集的数据,将其添加上时间和空间标签,形成同一组数据;所述传感器故障自诊断模块用于在传感器自身发生故障时,产生报警信号,并通知信息服务平台进行报警;所述网络通信模块用于将数据发送到信息服务平台。In a preferred embodiment of the present invention, the front-end equipment management platform further includes a sensor information processing module, a sensor association module, a sensor fault self-diagnosis module and a network communication module; the sensor information processing module is used to help the sensor to follow the correct Frequency acquisition data, filter the data and eliminate blank data; the sensor association module is used to integrate the data collected by each sensor, add time and space tags to it, and form the same set of data; the sensor fault self-diagnosis module is used for When the sensor itself fails, an alarm signal is generated, and the information service platform is notified to give an alarm; the network communication module is used for sending data to the information service platform.
本发明一个较佳实施例中,进一步包括包括用于展示传感器采集的数据、查询历史数据并展示、查看传感器运行状态、查看无菌灌装生产线各功能运行状态以及对异常数据进行报警的企业客户端。In a preferred embodiment of the present invention, it further includes enterprise customers for displaying data collected by sensors, querying and displaying historical data, viewing sensor operating status, viewing the operating status of various functions of the aseptic filling production line, and alerting abnormal data. end.
本发明一个较佳实施例中,进一步包括所述企业客户端为web展示页面,所述信息服务平台布署在云服务器上,所述web展示页面与信息服务平台通信连接。In a preferred embodiment of the present invention, it further includes that the enterprise client is a web display page, the information service platform is deployed on a cloud server, and the web display page is communicatively connected to the information service platform.
一种无菌灌装生产线异常检测方法,包括以下步骤:A method for detecting abnormality in an aseptic filling production line, comprising the following steps:
S1,通过前端设备采集物联网数据;S1, collect IoT data through front-end equipment;
S2,对采集到的历史数据进行预处理,根据实际生产线运行状态,将采集的历史数据分为正常数据和明显的异常数据,并对正常数据完成归一化和特征构建;S2, preprocess the collected historical data, divide the collected historical data into normal data and obvious abnormal data according to the actual production line operation status, and complete normalization and feature construction for the normal data;
S3,通过正常数据得到马氏距离的判别阈值,若马氏距离不在阈值区间内,则判定该数据为异常数据;若马氏距离在阈值区间内,则将数据添加其马氏距 离特征后记为不确定数据,输入自编码器,并加入稀疏性限制进行无监督的训练,调节参数和各层维度取得最优的隐层表达;S3, the judgment threshold of the Mahalanobis distance is obtained from normal data. If the Mahalanobis distance is not within the threshold interval, the data is determined to be abnormal data; if the Mahalanobis distance is within the threshold interval, the data is added with its Mahalanobis distance feature and recorded as Uncertain data is input to the auto-encoder, and the sparsity limit is added for unsupervised training, and the parameters and dimensions of each layer are adjusted to obtain the optimal hidden layer expression;
S4,将最优的隐层表达结合Sigmoid分类器构建自编码网络,在将带标签的数据输入到自编码网络后进行有监督的微调,得到最优参数,完成自编码网络的构建;S4, combine the optimal hidden layer expression with the Sigmoid classifier to construct an auto-encoding network, and perform supervised fine-tuning after inputting the labeled data into the self-encoding network to obtain the optimal parameters and complete the construction of the self-encoding network;
S5,对马氏距离在阈值区间内的不确定数据,将其输入到构建完成的自编码网络,并由自编码网络判定其是正常数据或异常数据。S5, for the uncertain data whose Mahalanobis distance is within the threshold interval, input it into the constructed auto-encoding network, and the auto-encoding network determines whether it is normal data or abnormal data.
本发明一个较佳实施例中,进一步包括步骤S3中,得到马氏距离的判别阈值的方法为:In a preferred embodiment of the present invention, in step S3, the method for obtaining the discrimination threshold of the Mahalanobis distance is:
计算n个数据、每个数据维度为m的数据集X=(X 1,X 2,X 3,...,X n)的马氏距离,其中均值为μ=(μ 123,...,μ m) T,协方差矩阵为Σ,则对任一数据x=(x 1,x 2,x 3,...,x m) T,则其马氏距离如下所示: Calculate the Mahalanobis distance of n data sets X=(X 1 , X 2 , X 3 ,..., X n ) with each data dimension m, where the mean is μ=(μ 1 , μ 2 , μ 3 ,...,μ m ) T , and the covariance matrix is Σ, then for any data x=(x 1 ,x 2 ,x 3 ,...,x m ) T , the Mahalanobis distance is as follows shown:
Figure PCTCN2021095632-appb-000001
Figure PCTCN2021095632-appb-000001
其中Σ -1为协方差矩阵的逆矩阵; where Σ -1 is the inverse of the covariance matrix;
对于采集到的正常数据集X N=(X 1,X 2,X 3,...,X l),其正常数据均值为μ N=(μ 123,...,μ m) T,协方差矩阵为Σ N,根据上式计算得到正常数据集中每个数据的马氏距离记为: For the collected normal data set X N =(X 1 ,X 2 ,X 3 ,...,X l ), the normal data mean is μ N =(μ 123 ,..., μ m ) T , the covariance matrix is Σ N , and the Mahalanobis distance of each data in the normal data set calculated according to the above formula is recorded as:
M N=(M 1,M 2,...M q,...,M l) M N =(M 1 ,M 2 ,...M q ,...,M l )
其中,M q表示在正常数据集中第q个正常数据的马氏距离; Among them, M q represents the Mahalanobis distance of the qth normal data in the normal data set;
在计算所有数据的马氏距离时使用的均值和协方差矩阵仍然为正常数据集中的μ N和Σ N,则数据X i的马氏距离记为: The mean and covariance matrix used in calculating the Mahalanobis distance of all data are still μ N and Σ N in the normal data set, then the Mahalanobis distance of the data Xi is recorded as:
Figure PCTCN2021095632-appb-000002
Figure PCTCN2021095632-appb-000002
将数据集M N的均值记为
Figure PCTCN2021095632-appb-000003
标准差记为
Figure PCTCN2021095632-appb-000004
根据统计学中的3σ准则,大部分正常数据的马氏距离分布于
Figure PCTCN2021095632-appb-000005
区间中,部分异常数据的马氏距离不在上述区间内,因此可通过数据的马氏距离来进行异常数据的初步检测,设置马氏距离判别阈值T up和T low,将其马氏距离不在(T low,T up)区间内的数据判定为异常数据,T up和T low的表达式如下所示:
Denote the mean of the dataset MN as
Figure PCTCN2021095632-appb-000003
Standard deviation is recorded as
Figure PCTCN2021095632-appb-000004
According to the 3σ criterion in statistics, the Mahalanobis distance of most normal data is distributed in
Figure PCTCN2021095632-appb-000005
In the interval, the Mahalanobis distance of some abnormal data is not in the above interval, so the initial detection of abnormal data can be carried out through the Mahalanobis distance of the data, and the Mahalanobis distance discrimination thresholds T up and T low are set, and the Mahalanobis distance is not in ( The data in the interval T low , T up ) is determined as abnormal data, and the expressions of T up and T low are as follows:
Figure PCTCN2021095632-appb-000006
Figure PCTCN2021095632-appb-000006
Figure PCTCN2021095632-appb-000007
Figure PCTCN2021095632-appb-000007
本发明一个较佳实施例中,进一步包括步骤S3中,自编码器的工作过程包括编码过程和解码过程,对于数据集X=(X 1,X 2,X 3,...,X n),n为数据个数,每个数据维度为m,每一个数据X i经过编码过程得到隐层表达,编码过程可描述为: In a preferred embodiment of the present invention, it further includes that in step S3, the working process of the self-encoder includes an encoding process and a decoding process. For the data set X=(X 1 , X 2 , X 3 ,...,X n ) , n is the number of data, the dimension of each data is m, and each data X i is expressed in the hidden layer through the encoding process. The encoding process can be described as:
h i=σ e(WX i+b) h ie (WX i +b)
其中,W和b为编码权重和偏置,σ e为编码层激活函数; Among them, W and b are the coding weights and biases, and σ e is the activation function of the coding layer;
隐层表达经解码过程得到重构数据X i',解码过程可描述为: The hidden layer expresses the reconstructed data X i ' through the decoding process, and the decoding process can be described as:
X i'=σ d(W'h i+b') X i '=σ d ( W'hi +b')
其中,W'和b'为解码权重和偏置,取W'=W T,σ d为解码层激活函数。 Among them, W' and b' are decoding weights and biases, W'=W T , and σ d is the activation function of the decoding layer.
本发明一个较佳实施例中,进一步包括步骤S3中隐层表达的获得方法为:In a preferred embodiment of the present invention, the method for obtaining the hidden layer expression in step S3 is further included:
首先,通过逐层贪婪算法调节权重和偏置使重构误差最小,对于整个训练数据集的代价函数为:First, the weights and biases are adjusted by a layer-by-layer greedy algorithm to minimize the reconstruction error. The cost function for the entire training dataset is:
Figure PCTCN2021095632-appb-000008
Figure PCTCN2021095632-appb-000008
其中,L为单个数据的损失函数,L为均方误差损失函数;Among them, L is the loss function of a single data, and L is the mean square error loss function;
然后,给代价函数添加一个L2正则化权重衰减项,λ为惩罚因子,添加KL散度作为约束条件,在其代价函数上加入稀疏惩罚项,使其形成稀疏自编码器, 得到最终的损失函数为:Then, an L2 regularization weight decay term is added to the cost function, λ is the penalty factor, KL divergence is added as a constraint, and a sparse penalty term is added to the cost function to form a sparse autoencoder, and the final loss function is obtained. for:
Figure PCTCN2021095632-appb-000009
Figure PCTCN2021095632-appb-000009
其中,
Figure PCTCN2021095632-appb-000010
为正则化项,
Figure PCTCN2021095632-appb-000011
为KL散度的约束条件,k为隐层神经元数量;
in,
Figure PCTCN2021095632-appb-000010
is the regularization term,
Figure PCTCN2021095632-appb-000011
is the constraint condition of KL divergence, and k is the number of neurons in the hidden layer;
最后,经过马氏距离判别出部分异常数据后,将不确定数据作为自编码器输入层数据,进行训练,使得式损失函数的值最小,进而得到最佳隐层表达。Finally, after identifying some abnormal data through Mahalanobis distance, the uncertain data is used as the input layer data of the autoencoder for training, so that the value of the formula loss function is the smallest, and then the best hidden layer expression is obtained.
本发明的有益效果:Beneficial effects of the present invention:
本发明的无菌灌装生产线异常检测系统,前端设备能够采集无菌灌装生产线各节点生产环境的信息,并上传至前端设备管理平台,前端管理平台将分立的信息融合后对信息进行降噪、筛选,最后由信息服务平台中构建的自编码网络找出异常信息。自编码网络作为一种使得输出等于输入的特殊的神经网络结构,在利用其进行图像处理、异常检测、故障预测、数据分类等过程中,能够重构数据,且在进行特征提取从而得到隐层表达时具有更好的代表数据的特征,能够大大提高无菌灌装生产线异常检测的准确性、稳定性和鲁棒性,实现了更为优异的故障预警效果。本方法解决了神经网络与物联网数据的联动问题,首先通过构建稀疏自编码器得到较好的隐层表达,随后结合Sigmoid分类器,构建出自编码网络进行有监督地微调,得到完整异常检测模型。将数据输入自编码网络异常检测模型中后得到准确的检测结果。相较于传统异常检测方法,本方法结合自编码网络与物联网数据进行智能分析检测,可以达到综合性强、准确度高、鲁棒性强等特点。In the abnormal detection system of the aseptic filling production line of the present invention, the front-end equipment can collect the production environment information of each node of the aseptic filling production line, and upload it to the front-end equipment management platform. , screening, and finally find out the abnormal information by the self-encoding network constructed in the information service platform. As a special neural network structure that makes the output equal to the input, the auto-encoding network can reconstruct the data in the process of image processing, anomaly detection, fault prediction, data classification, etc., and perform feature extraction to obtain the hidden layer. It has the characteristics of better representative data when it is expressed, which can greatly improve the accuracy, stability and robustness of abnormal detection of aseptic filling production line, and achieve more excellent fault warning effect. This method solves the linkage problem between neural network and IoT data. First, a sparse auto-encoder is constructed to obtain a better hidden layer expression, and then combined with the Sigmoid classifier, an auto-encoder network is constructed for supervised fine-tuning, and a complete anomaly detection model is obtained. . After inputting the data into the self-encoding network anomaly detection model, accurate detection results are obtained. Compared with the traditional anomaly detection method, this method combines the self-encoding network and the Internet of Things data for intelligent analysis and detection, which can achieve the characteristics of strong comprehensiveness, high accuracy, and strong robustness.
附图说明Description of drawings
图1为本发明优选实施例中无菌灌装生产线异常检测系统的结构示意图;1 is a schematic structural diagram of an abnormality detection system for an aseptic filling production line in a preferred embodiment of the present invention;
图2为无菌灌装生产线异常检测方法的流程示意图;Fig. 2 is the schematic flow chart of the abnormality detection method of aseptic filling production line;
图3为自编码器的结构示意图;3 is a schematic structural diagram of an autoencoder;
图4为自编码网络的结构示意图;Fig. 4 is the structural representation of self-encoding network;
图5为自编码网络的微调示意图。Figure 5 is a schematic diagram of the fine-tuning of the autoencoder network.
具体实施方式detailed description
下面结合附图和具体实施例对本发明作进一步说明,以使本领域的技术人员可以更好地理解本发明并能予以实施,但所举实施例不作为对本发明的限定。The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.
实施例Example
本发明公开一种无菌灌装生产线异常检测系统,包括用于采集和上传信息的前端设备;用于接收信息并实现信息的初步滤波、处理,并剔除问题信息的前端设备管理平台;用于通过自编码网络对信息行进判断并找出异常信息的信息服务平台。以上优化的结构,前端设备能够采集无菌灌装生产线各节点生产环境的信息,并上传至前端设备管理平台,前端管理平台将分立的信息融合后对信息进行降噪、筛选,最后由信息服务平台中构建的自编码网络找出异常信息。自编码网络作为一种使得输出等于输入的特殊的神经网络结构,在利用其进行图像处理、异常检测、故障预测、数据分类等过程中,能够重构数据,且在进行特征提取从而得到隐层表达时具有更好的代表数据的特征,能够大大提高无菌灌装生产线异常检测的准确性、稳定性和鲁棒性,实现了更为优异的故障预警效果。The invention discloses an abnormality detection system for an aseptic filling production line, comprising front-end equipment for collecting and uploading information; a front-end equipment management platform for receiving information, realizing preliminary filtering and processing of information, and eliminating problem information; An information service platform that judges the information travel through the self-encoding network and finds out abnormal information. With the above optimized structure, the front-end equipment can collect the production environment information of each node of the aseptic filling production line and upload it to the front-end equipment management platform. The self-encoding network built in the platform finds out abnormal information. As a special neural network structure that makes the output equal to the input, the auto-encoding network can reconstruct the data in the process of image processing, anomaly detection, fault prediction, data classification, etc., and perform feature extraction to obtain the hidden layer. It has the characteristics of better representative data when it is expressed, which can greatly improve the accuracy, stability and robustness of abnormal detection of aseptic filling production line, and achieve more excellent fault warning effect.
具体而言,参照图1所示,上述前端设备包括若干个分布在无菌灌装生产线上并针对生产环境设置的传感器。传感器至少包括振动传感器、电压电流传感器、温度传感器、湿度传感器和噪声传感器中的一种或多种。对于正常运行的无菌灌装生产线而言,其生产环境是稳定的,即振动、电压电流、温度、湿度、噪声等数据保持在正常的范围内,而一旦生产环境发生异常,这些数据也必定会发生相应的变化。Specifically, as shown in FIG. 1 , the above-mentioned front-end equipment includes several sensors distributed on the aseptic filling production line and set for the production environment. The sensors include at least one or more of vibration sensors, voltage and current sensors, temperature sensors, humidity sensors and noise sensors. For a normally operating aseptic filling production line, its production environment is stable, that is, vibration, voltage, current, temperature, humidity, noise and other data are kept within a normal range, and once the production environment is abnormal, these data must also be Corresponding changes will occur.
在无菌灌装生产车间,影响网络通信的因素较多,为了降低这些因素的影响,上述各传感器采用RS485总线形式串联。In the aseptic filling production workshop, there are many factors that affect the network communication. In order to reduce the influence of these factors, the above sensors are connected in series in the form of RS485 bus.
上述前端设备管理平台与前端设备通信连接。前端设备管理平台包括传感器信息处理模块、传感器关联模块、传感器故障自诊断模块和网络通讯模块。传感器信息处理模块能够帮助传感器按照正确的频率采集数据,并对数据进行滤波以及剔除空白数据,增强抗干扰能力。传感器关联模块能够整合各传感器采集的数据,将其添加上时间和空间标签,形成同一组数据。传感器故障自诊断模块能够在传感器自身发生故障时,产生报警信号,并通知信息服务平台进行报警。网络通信模块能够将数据发送到信息服务平台。The front-end device management platform is connected in communication with the front-end device. The front-end equipment management platform includes a sensor information processing module, a sensor association module, a sensor fault self-diagnosis module and a network communication module. The sensor information processing module can help the sensor to collect data according to the correct frequency, filter the data and eliminate blank data, and enhance the anti-interference ability. The sensor association module can integrate the data collected by each sensor, add time and space tags to it, and form the same set of data. The sensor fault self-diagnosis module can generate an alarm signal when the sensor itself fails, and notify the information service platform to give an alarm. The network communication module can send data to the information service platform.
为了降低生产车间对通信的影响,网络通讯模块可以将数据通过有线以太网数据发送传输至信息服务平台。In order to reduce the impact of the production workshop on communication, the network communication module can transmit data to the information service platform through wired Ethernet data transmission.
上述信息服务平台与前端设备管理平台通信连接。信息服务平台包括物联网数据处理模块、异常数据检测模块和异常报警模块。物联网数据处理模块能够接收前端发送来的数据并进行处理,为异常数据检测模块提供数据基础。异常数据检测模块能够对数据通过自编码网络模型进行判断,实现信息联动异常检测功能。异常报警模块能够在异常数据检测模块判断出异常数据时进行报警。信息服务平台主要负责对传感器采集到的物联网数据进行处理,并且通过物联网数据结合自编码网络进行异常数据检测。The above-mentioned information service platform is in communication connection with the front-end equipment management platform. The information service platform includes an IoT data processing module, an abnormal data detection module and an abnormal alarm module. The IoT data processing module can receive and process the data sent by the front end, providing a data basis for the abnormal data detection module. The abnormal data detection module can judge the data through the self-encoding network model, and realize the information linkage abnormal detection function. The abnormal alarm module can issue an alarm when the abnormal data detection module determines abnormal data. The information service platform is mainly responsible for processing the IoT data collected by the sensors, and detecting abnormal data through the IoT data combined with the self-encoding network.
为了方便信息服务平台的访问,其可以布署在云服务器上,其能够实现随时随地的访问,极大的增强了系统的便利性。In order to facilitate the access of the information service platform, it can be deployed on the cloud server, which can realize access anytime and anywhere, and greatly enhance the convenience of the system.
在本发明一些优选的实施例中,一种自编码网络与物联网数据的无菌灌装生产线异常检测系统还包括企业客户端。企业客户端与信息服务平台通信连接。企业客户端用于展示传感器采集的数据、查询历史数据并展示、查看传感器运行状态、查看无菌灌装生产线各功能运行状态、对异常数据进行报警等。一旦 信息服务平台检测到异常数据,其会将报警数据发送到企业客户端,企业可以在企业客户端及时得知异常的发生,并实时观测数据。In some preferred embodiments of the present invention, an abnormality detection system for an aseptic filling production line using self-encoding network and Internet of Things data further includes an enterprise client. The enterprise client communicates with the information service platform. The enterprise client is used to display the data collected by the sensor, query and display historical data, check the operation status of the sensor, check the operation status of each function of the aseptic filling production line, and alarm abnormal data. Once the information service platform detects abnormal data, it will send the alarm data to the enterprise client, so that the enterprise can know the abnormality in time and observe the data in real time.
企业客户端可以为web展示页面。其与云服务器上的信息服务平台通信连接。利用web组件,实现了数据展示、报警等功能,同时web页面不需要繁锁的安装和调试,进一步提高了系统的便利性和直观性。另外,使用web页面进行数据展示和报警,提高了整个系统的鲁棒性,并且使用灵活方便,体现了用户友好型的互联网思维,是互联网时代下生产线异常检测系统的大趋势。Enterprise clients can display pages for the web. It communicates and connects with the information service platform on the cloud server. Using web components, functions such as data display and alarm are realized. At the same time, the web page does not require complicated installation and debugging, which further improves the convenience and intuitiveness of the system. In addition, the use of web pages for data display and alarm improves the robustness of the entire system, and is flexible and convenient to use, reflecting a user-friendly Internet thinking, which is the general trend of production line anomaly detection systems in the Internet era.
企业客户端可以由数据展示界面和报警机制组成。The enterprise client can be composed of data display interface and alarm mechanism.
本发明公开一种无菌灌装生产线异常检测方法,包括以下步骤:The invention discloses a method for detecting abnormality of an aseptic filling production line, comprising the following steps:
S1,通过前端设备采集物联网数据。为采集到较多的异常数据从而构建更好的自编码网络,可以进行长时间的数据采集,甚至人为造成一段时间的异常。S1, collect IoT data through front-end equipment. In order to collect more abnormal data and build a better self-encoding network, it is possible to collect data for a long time, or even artificially cause abnormality for a period of time.
S2,对采集到的历史数据进行预处理,根据实际生产线运行状态,将采集的历史数据分为正常数据和明显的异常数据,并对正常数据完成归一化和特征构建。明显的异常数据包括空白数据和明显错误数据。S2: Preprocess the collected historical data, divide the collected historical data into normal data and obvious abnormal data according to the actual production line operation state, and complete normalization and feature construction for the normal data. Obvious outliers include blank data and obviously wrong data.
S3,通过正常数据得到马氏距离的判别阈值,若马氏距离不在阈值区间内,则判定该数据为异常数据;若马氏距离在阈值区间内,则将数据添加其马氏距离特征后记为不确定数据,输入自编码器,并加入稀疏性限制进行无监督的训练,调节参数和各层维度取得最优的隐层表达。S3, the judgment threshold of the Mahalanobis distance is obtained from normal data. If the Mahalanobis distance is not within the threshold interval, the data is determined to be abnormal data; if the Mahalanobis distance is within the threshold interval, the data is added with its Mahalanobis distance feature and recorded as The uncertain data is input to the autoencoder, and the sparsity restriction is added for unsupervised training, and the parameters and the dimensions of each layer are adjusted to obtain the optimal hidden layer expression.
S4,将最优的隐层表达结合Sigmoid分类器构建自编码网络,在将带标签的数据输入到自编码网络后进行有监督的微调,得到最优参数,完成自编码网络的构建。其解决了神经网络参数初始化的问题,并且缩短了分类器的训练次数,提高了异常检测的准确度。S4, the optimal hidden layer expression is combined with the Sigmoid classifier to construct an auto-encoding network. After inputting the labeled data into the auto-encoding network, supervised fine-tuning is performed to obtain the optimal parameters, and the construction of the self-encoding network is completed. It solves the problem of neural network parameter initialization, shortens the training times of the classifier, and improves the accuracy of anomaly detection.
S5,对马氏距离在阈值区间内的不确定数据,将其输入到构建完成的自编 码网络,并由自编码网络判定其是正常数据或异常数据。S5, for the uncertain data whose Mahalanobis distance is within the threshold interval, input it into the constructed self-encoding network, and the self-encoding network determines whether it is normal data or abnormal data.
本方法解决了神经网络与物联网数据的联动问题,首先通过构建稀疏自编码器得到较好的隐层表达,随后结合Sigmoid分类器,构建出自编码网络进行有监督地微调,得到完整异常检测模型。将数据输入自编码网络异常检测模型中后得到准确的检测结果。相较于传统异常检测方法,本方法结合自编码网络与物联网数据进行智能分析检测,可以达到综合性强、准确度高、鲁棒性强等特点。This method solves the linkage problem between neural networks and IoT data. First, a sparse auto-encoder is constructed to obtain a better hidden layer expression, and then combined with the Sigmoid classifier, an auto-encoder network is constructed for supervised fine-tuning, and a complete anomaly detection model is obtained. . After inputting the data into the self-encoding network anomaly detection model, accurate detection results are obtained. Compared with the traditional anomaly detection method, this method combines the self-encoding network and the Internet of Things data for intelligent analysis and detection, which can achieve the characteristics of strong comprehensiveness, high accuracy, and strong robustness.
具体而言,参照图2所示,在将物联网数据和自编码网络结合时,考虑到各变量之间相关性的广义距离,可以利用向量间的协方差矩阵来表示马氏距离。Specifically, referring to Fig. 2, when combining IoT data with an auto-encoding network, considering the generalized distance of the correlation between variables, the Mahalanobis distance can be represented by the covariance matrix between vectors.
对于包含n个数据、每个数据维度为m的数据集X=(X 1,X 2,X 3,...,X n),其中均值为μ=(μ 123,...,μ m) T,协方差矩阵为Σ,其中任一数据为x=(x 1,x 2,x 3,...,x m) T,则其马氏距离如下所示: For a dataset X = ( X 1 , X 2 , X 3 , . ...,μ m ) T , the covariance matrix is Σ, and any data is x=(x 1 ,x 2 ,x 3 ,...,x m ) T , then its Mahalanobis distance is as follows:
Figure PCTCN2021095632-appb-000012
Figure PCTCN2021095632-appb-000012
其中Σ -1为协方差矩阵的逆矩阵,马氏距离可看作数据与总体数据均值的距离。 Among them, Σ -1 is the inverse matrix of the covariance matrix, and the Mahalanobis distance can be regarded as the distance between the data and the mean of the overall data.
由于马氏距离的计算需要使用数据集的协方差矩阵,所以较欧式距离等其他距离的最大优势为马氏距离考虑数据特征之间的相关性。在数据集中,如果一个数据的马氏距离越小,则说明其与数据集中均值数据的相似度越大。在采集到的无菌灌装生产线数据中,由于工艺流程和采集设备等缘故,数据的每个特征之间有着不可忽视的相关性,马氏距离更适合用于无菌灌装生产线数据的距离表达。Since the calculation of Mahalanobis distance needs to use the covariance matrix of the data set, the biggest advantage of other distances such as Euclidean distance is that the Mahalanobis distance considers the correlation between data features. In a data set, if the Mahalanobis distance of a data is smaller, it means that it is more similar to the mean data in the data set. In the collected aseptic filling production line data, due to the technological process and collection equipment, there is a non-negligible correlation between each feature of the data, and the Mahalanobis distance is more suitable for the distance of the aseptic filling production line data. Express.
考虑在数据异常检测中,假设某数据为正常数据,使用正常数据集的均值和协方差矩阵根据式(1)计算其马氏距离。若该数据马氏距离与正常数据马氏距离接近,说明该数据与正常数据相似度较大,该数据大概率为正常数据;若该数据马氏距离与正常数据马氏距离相差较远,说明该数据与正常数据相似度较 小,则该数据大概率为异常数据。所以能够使用数据的马氏距离来判断数据为异常数据的可能性。Considering that in the data anomaly detection, assuming that a certain data is normal data, use the mean and covariance matrix of the normal data set to calculate its Mahalanobis distance according to formula (1). If the Mahalanobis distance of the data is close to the Mahalanobis distance of the normal data, it means that the data is more similar to the normal data, and the data is probably normal data; if the Mahalanobis distance of the data is far from the Mahalanobis distance of the normal data, it means If the similarity between the data and normal data is small, the data is likely to be abnormal data. Therefore, the Mahalanobis distance of the data can be used to judge the possibility that the data is abnormal data.
由于数据量庞大,在自编码器和神经网络的训练过程中,过大的数据量使得训练效率较低并且检测准确度无法得到有效提升。为了减少自编码器和神经网络的训练数据量,可以通过正常数据样本的马氏距离得到判定阈值,计算数据与正常数据均值的马氏距离,将没有超过判定阈值的数据记为不确定数据并将该部分数据用于自编码器和神经网络的训练。Due to the huge amount of data, in the training process of the autoencoder and neural network, the excessively large amount of data makes the training efficiency low and the detection accuracy cannot be effectively improved. In order to reduce the amount of training data for the autoencoder and neural network, the judgment threshold can be obtained from the Mahalanobis distance of the normal data samples, the Mahalanobis distance between the data and the average value of the normal data is calculated, and the data that does not exceed the judgment threshold is recorded as uncertain data and This part of the data is used for the training of autoencoders and neural networks.
假设采集到的数据集记为X=(X 1,X 2,X 3,...,X n),数据维度为m。其中正常数据集记为X N=(X 1,X 2,X 3,...,X l),其中正常数据均值为μ N=(μ 123,...,μ m) T,协方差矩阵为Σ N,根据式(1)计算得到正常数据集中每个数据的马氏距离记为: It is assumed that the collected data set is denoted as X=(X 1 , X 2 , X 3 , . . . , X n ), and the data dimension is m. The normal data set is denoted as X N =(X 1 ,X 2 ,X 3 ,...,X l ), and the normal data mean is μ N =(μ 123 ,...,μ m ) T , the covariance matrix is Σ N , and the Mahalanobis distance of each data in the normal data set calculated according to formula (1) is recorded as:
M N=(M 1,M 2,...M q,...,M l)      (2) M N =(M 1 ,M 2 ,...M q ,...,M l ) (2)
其中,M q表示在正常数据集中第q个正常数据的马氏距离。 Among them, M q represents the Mahalanobis distance of the qth normal data in the normal data set.
接下来根据式(1)计算整个数据集中所有数据的马氏距离。由于某一数据的变化会影响到数据集均值的变化,马氏距离夸大了微小变化向量的作用,从而影响其他数据的马氏距离计算。为了改善上述马氏距离的缺点,在计算所有数据的马氏距离时使用的均值和协方差矩阵仍然为正常数据集中的μ N和Σ N,显然独立的μ N和Σ N不受向量变化影响;得到的某个数据的马氏距离可看作该数据与正常数据集均值的距离。则数据X i的马氏距离记为: Next, the Mahalanobis distance of all data in the entire data set is calculated according to formula (1). Since the change of a certain data will affect the change of the mean value of the data set, the Mahalanobis distance exaggerates the effect of the small change vector, thus affecting the calculation of the Mahalanobis distance of other data. In order to improve the shortcomings of the above Mahalanobis distance, the mean and covariance matrix used in calculating the Mahalanobis distance of all data are still μ N and Σ N in the normal data set, obviously independent μ N and Σ N are not affected by vector changes ; The Mahalanobis distance of a certain data can be regarded as the distance between the data and the mean of the normal data set. Then the Mahalanobis distance of the data Xi is written as:
Figure PCTCN2021095632-appb-000013
Figure PCTCN2021095632-appb-000013
根据统计学相关知识以及后续对振动数据的实验分析可知:如果数据X i为正常数据,则其马氏距离D M(X i)应符合正常数据马氏距离数据集即M N的统计分布;如果X i为异常数据,则D M(X i)不符合M N的统计分布。 According to relevant statistical knowledge and subsequent experimental analysis of vibration data, it can be known that if the data X i is normal data, its Mahalanobis distance D M ( X i ) should conform to the statistical distribution of the normal data Mahalanobis distance data set, namely MN; If X i is abnormal data, then D M ( X i ) does not conform to the statistical distribution of MN.
将数据集M N的均值记为
Figure PCTCN2021095632-appb-000014
标准差记为
Figure PCTCN2021095632-appb-000015
根据统计学中的3σ准则,大部分正常数据的马氏距离分布于
Figure PCTCN2021095632-appb-000016
区间中,部分异常数据的马氏距离不在上述区间内。因此可通过数据的马氏距离来进行异常数据的初 步检测。设置马氏距离判别阈值T up和T low,将其马氏距离不在(T low,T up)区间内的数据判定为异常数据,结合实际数据分析,为保证异常检测的准确性,T up和T low的表达式如下所示:
Denote the mean of the dataset MN as
Figure PCTCN2021095632-appb-000014
Standard deviation is recorded as
Figure PCTCN2021095632-appb-000015
According to the 3σ criterion in statistics, the Mahalanobis distance of most normal data is distributed in
Figure PCTCN2021095632-appb-000016
In the interval, the Mahalanobis distance of some abnormal data is not within the above interval. Therefore, the initial detection of abnormal data can be carried out through the Mahalanobis distance of the data. Set the Mahalanobis distance discrimination thresholds T up and T low , and determine the data whose Mahalanobis distance is not in the interval (T low , T up ) as abnormal data. Combined with actual data analysis, in order to ensure the accuracy of abnormal detection, T up and T up The expression for T low is as follows:
Figure PCTCN2021095632-appb-000017
Figure PCTCN2021095632-appb-000017
Figure PCTCN2021095632-appb-000018
Figure PCTCN2021095632-appb-000018
将马氏距离在(T low,T up)区间内的数据判定为不确定数据,记为X U,然后使用不确定数据集X U训练自编码网络以完成模型构建。 The data whose Mahalanobis distance is in the interval of (T low , T up ) is determined as uncertain data, which is denoted as X U , and then the self-encoding network is trained with the uncertain data set X U to complete the model construction.
于是,通过马氏距离检测出了一部分异常数据,将剩余的不确定数据输入自编码网络中。对于数据量大的无菌灌装生产线数据而言,减少了用于自编码网络的训练数据,并且能够根据马氏距离的判别阈值快速判别出一部分异常数据,提高了检测效率。Therefore, a part of abnormal data is detected by Mahalanobis distance, and the remaining uncertain data is input into the self-encoding network. For the aseptic filling production line data with a large amount of data, the training data for the self-encoding network is reduced, and a part of abnormal data can be quickly identified according to the discrimination threshold of the Mahalanobis distance, which improves the detection efficiency.
另外,无菌灌装生产线数据的马氏距离可以用来判断该数据为异常数据的可能性,故考虑将数据的马氏距离作为数据的一个特征。由于正常数据和异常数据的马氏距离有较大的差别,在神经网络的训练过程中,加入马氏距离特征作为重要特征,有利于提升神经网络的异常检测效果。所以将数据的马氏距离特征加入到数据特征中,用于自编码网络的训练。In addition, the Mahalanobis distance of the aseptic filling production line data can be used to judge the possibility that the data is abnormal data, so consider the Mahalanobis distance of the data as a feature of the data. Since the Mahalanobis distance between normal data and abnormal data is quite different, in the training process of the neural network, adding the Mahalanobis distance feature as an important feature is beneficial to improve the abnormal detection effect of the neural network. Therefore, the Mahalanobis distance feature of the data is added to the data feature for the training of the self-encoding network.
参照图3所示,自编码器主要包括编码和解码阶段,且结构对称,即若存在多个隐层时,编码和解码阶段的隐层数量及结构相同。主要结构由输入层、隐层和输出层组成。隐层对输入层数据进行编码,输出层对隐层表达进行解码重构原始数据,最小化重构误差以获得最佳的隐层表达。其目标是拟合一个恒等函数,使得每个输出值尽可能等于相对应的输入值。Referring to FIG. 3 , the autoencoder mainly includes encoding and decoding stages, and the structure is symmetrical, that is, if there are multiple hidden layers, the number and structure of hidden layers in the encoding and decoding stages are the same. The main structure consists of input layer, hidden layer and output layer. The hidden layer encodes the input layer data, and the output layer decodes the hidden layer expression to reconstruct the original data, minimizing the reconstruction error to obtain the best hidden layer expression. The goal is to fit an identity function such that each output value is as equal as possible to the corresponding input value.
对于数据集X=(X 1,X 2,X 3,...,X n),n为数据个数,每个数据维度为m。每一个数据X i经过编码过程得到隐层表达,编码过程可描述为: For data set X=(X 1 , X 2 , X 3 ,...,X n ), n is the number of data, and the dimension of each data is m. Each data X i is expressed in the hidden layer through the encoding process, and the encoding process can be described as:
h i=σ e(WX i+b)      (6) h ie (WX i +b) (6)
其中,W和b为编码权重和偏置,σ e为编码层激活函数,可以为Sigmoid、 Tanh、Relu等。然后隐层表达经解码过程得到重构数据X i',解码过程可描述为: Among them, W and b are coding weights and biases, and σ e is the activation function of the coding layer, which can be Sigmoid, Tanh, Relu, etc. Then the hidden layer expresses the reconstructed data X i ' through the decoding process, and the decoding process can be described as:
X i'=σ d(W'h i+b')      (7) X i '=σ d ( W'hi +b') (7)
其中,W'和b'为解码权重和偏置,取W'=W T,σ d为解码层激活函数。通过逐层贪婪算法调节权重和偏置使重构误差最小,对于整个训练数据集的代价函数为: Among them, W' and b' are decoding weights and biases, W'=W T , and σ d is the activation function of the decoding layer. The weights and biases are adjusted by a layer-by-layer greedy algorithm to minimize the reconstruction error. The cost function for the entire training data set is:
Figure PCTCN2021095632-appb-000019
Figure PCTCN2021095632-appb-000019
其中,L为单个数据的损失函数,在式(8)中L为均方误差损失函数。Among them, L is the loss function of a single data, and in Equation (8), L is the mean square error loss function.
为了防止出现过拟合,给代价函数添加一个L2正则化权重衰减项,λ为惩罚因子,控制正则化项影响权重衰减的程度;为了提高自编码器学习数据特征的能力,本发明优选的实施例中在自编码器基础上添加KL散度作为约束条件,在其代价函数上加入稀疏惩罚项,使其形成稀疏自编码器。得到最终的损失函数为:In order to prevent over-fitting, an L2 regularization weight decay term is added to the cost function, and λ is a penalty factor to control the degree to which the regularization term affects the weight decay. In the example, KL divergence is added as a constraint on the basis of the autoencoder, and a sparse penalty term is added to its cost function to form a sparse autoencoder. The final loss function is obtained as:
Figure PCTCN2021095632-appb-000020
Figure PCTCN2021095632-appb-000020
其中,
Figure PCTCN2021095632-appb-000021
为正则化项,
Figure PCTCN2021095632-appb-000022
为KL散度的约束条件,k为隐层神经元数量。
in,
Figure PCTCN2021095632-appb-000021
is the regularization term,
Figure PCTCN2021095632-appb-000022
is the constraint condition of KL divergence, and k is the number of neurons in the hidden layer.
将传感器采集到的物联网数据,经过马氏距离判别出部分异常数据后,将不确定数据作为上述自编码器输入层数据,进行训练,使得式(9)最小,得到最佳隐层表达。After some abnormal data are identified by the Mahalanobis distance from the IoT data collected by the sensor, the uncertain data is used as the input layer data of the autoencoder for training, so that equation (9) is minimized and the best hidden layer expression is obtained.
参照图4所示,接下来,通过将得到的隐层作为Sigmoid分类器的输入层,构建自编码网络,将数据判断为正常数据和异常数据。这样,通过自编码器训练得到隐层表达,结合Sigmoid分类器,能够提高异常检测准确度,且由于隐层的参数通过自编码器的训练已得到,解决了神经网络的参数初始化问题,使得自编码网络的训练次数减少,提高模型的构建效率。Referring to Figure 4, next, by using the obtained hidden layer as the input layer of the Sigmoid classifier, an auto-encoding network is constructed, and the data is judged as normal data and abnormal data. In this way, the expression of the hidden layer is obtained through the training of the autoencoder, combined with the Sigmoid classifier, the anomaly detection accuracy can be improved, and since the parameters of the hidden layer have been obtained through the training of the autoencoder, the parameter initialization problem of the neural network is solved, so that the automatic The number of training times of the encoding network is reduced, and the construction efficiency of the model is improved.
参照图5所示,然后,需要输入带标签的数据,进行自编码网络的训练过 程,以进行有监督地微调。微调时,将误差进行反向传播,使其进入到隐层中和Sigmoid分类器中去。由于自编码器解决了分类器的参数初始化问题,所以减少了自编码网络的训练次数,提高了训练效率。Referring to Figure 5, labeled data is then required to be input for the training process of the self-encoding network for supervised fine-tuning. During fine-tuning, the error is back-propagated into the hidden layer and the Sigmoid classifier. Since the autoencoder solves the problem of parameter initialization of the classifier, the training times of the autoencoder network are reduced and the training efficiency is improved.
最终,完成自编码网络异常检测模型的构建,其主要步骤总结如下:Finally, the construction of the self-encoding network anomaly detection model is completed. The main steps are summarized as follows:
S31,计算正常数据的马氏距离,根据式(4)、式(5)得到马氏距离判别阈值。S31, calculate the Mahalanobis distance of the normal data, and obtain the Mahalanobis distance discrimination threshold according to formula (4) and formula (5).
S32,根据式(3)得到所有数据的马氏距离,将马氏距离超过阈值的数据判定为异常数据,没有超过阈值的数据记为不确定数据,并将数据的马氏距离加入数据特征中。S32, obtain the Mahalanobis distance of all the data according to formula (3), determine the data whose Mahalanobis distance exceeds the threshold as abnormal data, and record the data that does not exceed the threshold as uncertain data, and add the Mahalanobis distance of the data to the data feature .
S33,将加入马氏距离特征的不确定数据集作为自编码器的输入,以无监督的方式训练自编码器,得到最优的编码层输出和参数。S33, take the uncertain data set added with Mahalanobis distance feature as the input of the autoencoder, train the autoencoder in an unsupervised manner, and obtain the optimal output and parameters of the encoding layer.
S34,将自编码器编码层作为Sigmoid分类器的输入层,使用带标签的不确定数据作为输入,进行有监督地微调以得到整个网络的最优参数,完成自编码网络的构建。S34, take the self-encoder coding layer as the input layer of the Sigmoid classifier, and use the labeled uncertain data as the input to perform supervised fine-tuning to obtain the optimal parameters of the entire network, and complete the construction of the self-encoding network.
S35,对于马氏距离在阈值区间内的数据,将其放入训练完成的自编码网络中得到判定结果。S35, for the data whose Mahalanobis distance is within the threshold interval, put it into the trained self-encoding network to obtain a judgment result.
以上所述实施例仅是为充分说明本发明而所举的较佳的实施例,本发明的保护范围不限于此。本技术领域的技术人员在本发明基础上所作的等同替代或变换,均在本发明的保护范围之内。本发明的保护范围以权利要求书为准。The above-mentioned embodiments are only preferred embodiments for fully illustrating the present invention, and the protection scope of the present invention is not limited thereto. Equivalent substitutions or transformations made by those skilled in the art on the basis of the present invention are all within the protection scope of the present invention. The protection scope of the present invention is subject to the claims.

Claims (10)

  1. 一种无菌灌装生产线异常检测系统,其特征在于:包括用于采集和上传信息的前端设备;用于接收信息并实现信息的初步滤波、处理,以及剔除问题信息的前端设备管理平台;用于通过自编码网络对信息行进判断并找出异常信息的信息服务平台。An aseptic filling production line anomaly detection system is characterized in that: it includes front-end equipment for collecting and uploading information; a front-end equipment management platform for receiving information, realizing preliminary filtering and processing of information, and eliminating problem information; It is an information service platform that judges and finds abnormal information through the self-encoding network.
  2. 如权利要求1所述的无菌灌装生产线异常检测系统,其特征在于:所述前端设备包括若干个分布在无菌灌装生产线上并针对生产环境设置的传感器。The abnormality detection system for an aseptic filling production line according to claim 1, wherein the front-end equipment includes a plurality of sensors distributed on the aseptic filling production line and set for the production environment.
  3. 如权利要求2所述的无菌灌装生产线异常检测系统和方法,其特征在于:所述传感器采用RS485总线形式串联。The abnormality detection system and method of the aseptic filling production line according to claim 2, wherein the sensors are connected in series in the form of an RS485 bus.
  4. 如权利要求1所述的无菌灌装生产线异常检测系统,其特征在于:所述前端设备管理平台包括传感器信息处理模块、传感器关联模块、传感器故障自诊断模块和网络通讯模块;所述传感器信息处理模块用于帮助传感器按照正确的频率采集数据,并对数据进行滤波以及剔除空白数据;所述传感器关联模块用于整合各传感器采集的数据,将其添加上时间和空间标签,形成同一组数据;所述传感器故障自诊断模块用于在传感器自身发生故障时,产生报警信号,并通知信息服务平台进行报警;所述网络通信模块用于将数据发送到信息服务平台。The abnormality detection system for an aseptic filling production line according to claim 1, wherein the front-end equipment management platform comprises a sensor information processing module, a sensor correlation module, a sensor fault self-diagnosis module and a network communication module; the sensor information The processing module is used to help the sensor collect data according to the correct frequency, filter the data and remove blank data; the sensor association module is used to integrate the data collected by each sensor, add time and space tags to it, and form the same set of data The sensor fault self-diagnosis module is used to generate an alarm signal when the sensor itself fails, and notify the information service platform to give an alarm; the network communication module is used to send data to the information service platform.
  5. 如权利要求1所述的无菌灌装生产线异常检测系统,其特征在于:包括用于展示传感器采集的数据、查询历史数据并展示、查看传感器运行状态、查看无菌灌装生产线各功能运行状态以及对异常数据进行报警的企业客户端。The abnormality detection system of the aseptic filling production line according to claim 1, characterized in that it includes the functions for displaying the data collected by the sensor, querying and displaying historical data, checking the operation status of the sensor, and checking the operation status of each function of the aseptic filling production line. And enterprise clients that alert on abnormal data.
  6. 如权利要求5所述的无菌灌装生产线异常检测系统,其特征在于:所述企业客户端为web展示页面,所述信息服务平台布署在云服务器上,所述web展示页面与信息服务平台通信连接。The abnormality detection system for an aseptic filling production line according to claim 5, wherein the enterprise client is a web display page, the information service platform is deployed on a cloud server, and the web display page and information service Platform communication connection.
  7. 一种无菌灌装生产线异常检测方法,其特征在于:包括以下步骤:A method for detecting abnormality in an aseptic filling production line, comprising the following steps:
    S1,通过前端设备采集物联网数据;S1, collect IoT data through front-end equipment;
    S2,对采集到的历史数据进行预处理,根据实际生产线运行状态,将采集的历史数据分为正常数据和明显的异常数据,并对正常数据完成归一化和特征构建;S2, preprocess the collected historical data, divide the collected historical data into normal data and obvious abnormal data according to the actual production line operation status, and complete normalization and feature construction for the normal data;
    S3,通过正常数据得到马氏距离的判别阈值,若马氏距离不在阈值区间内,则判定该数据为异常数据;若马氏距离在阈值区间内,则将数据添加其马氏距离特征后记为不确定数据,输入自编码器,并加入稀疏性限制进行无监督的训练,调节参数和各层维度取得最优的隐层表达;S3, the judgment threshold of the Mahalanobis distance is obtained from normal data. If the Mahalanobis distance is not within the threshold interval, the data is determined to be abnormal data; if the Mahalanobis distance is within the threshold interval, the data is added with its Mahalanobis distance feature and recorded as Uncertain data is input to the auto-encoder, and the sparsity limit is added for unsupervised training, and the parameters and dimensions of each layer are adjusted to obtain the optimal hidden layer expression;
    S4,将最优的隐层表达结合Sigmoid分类器构建自编码网络,在将带标签的数据输入到自编码网络后进行有监督的微调,得到最优参数,完成自编码网络的构建;S4, combine the optimal hidden layer expression with the Sigmoid classifier to construct an auto-encoding network, and perform supervised fine-tuning after inputting the labeled data into the self-encoding network to obtain the optimal parameters and complete the construction of the self-encoding network;
    S5,对马氏距离在阈值区间内的不确定数据,将其输入到构建完成的自编码网络,并由自编码网络判定其是正常数据或异常数据。S5, for the uncertain data whose Mahalanobis distance is within the threshold interval, input it into the constructed auto-encoding network, and the auto-encoding network determines whether it is normal data or abnormal data.
  8. 如权利要求7所述的无菌灌装生产线异常检测方法,其特征在于:步骤S3中,得到马氏距离的判别阈值的方法为:The method for detecting abnormality of aseptic filling production line as claimed in claim 7, characterized in that: in step S3, the method for obtaining the discrimination threshold of the Mahalanobis distance is:
    计算n个数据、每个数据维度为m的数据集X=(X 1,X 2,X 3,...,X n)的马氏距离,其中均值为μ=(μ 123,...,μ m) T,协方差矩阵为Σ,则对任一数据x=(x 1,x 2,x 3,...,x m) T,则其马氏距离如下所示: Calculate the Mahalanobis distance of n data sets X=(X 1 , X 2 , X 3 ,..., X n ) with each data dimension m, where the mean is μ=(μ 1 , μ 2 , μ 3 ,...,μ m ) T , and the covariance matrix is Σ, then for any data x=(x 1 ,x 2 ,x 3 ,...,x m ) T , the Mahalanobis distance is as follows shown:
    Figure PCTCN2021095632-appb-100001
    Figure PCTCN2021095632-appb-100001
    其中Σ -1为协方差矩阵的逆矩阵; where Σ -1 is the inverse of the covariance matrix;
    对于采集到的正常数据集X N=(X 1,X 2,X 3,...,X l),其正常数据均值为 μ N=(μ 123,...,μ m) T,协方差矩阵为Σ N,根据上式计算得到正常数据集中每个数据的马氏距离,记为: For the collected normal data set X N =(X 1 ,X 2 ,X 3 ,...,X l ), the normal data mean is μ N =(μ 123 ,..., μ m ) T , the covariance matrix is Σ N , and the Mahalanobis distance of each data in the normal data set is calculated according to the above formula, which is recorded as:
    M N=(M 1,M 2,...M q,...,M l) M N =(M 1 ,M 2 ,...M q ,...,M l )
    其中,M q表示在正常数据集中第q个正常数据的马氏距离; Among them, M q represents the Mahalanobis distance of the qth normal data in the normal data set;
    在计算所有数据的马氏距离时使用的均值和协方差矩阵仍然为正常数据集中的μ N和Σ N,则数据X i的马氏距离记为: The mean and covariance matrix used in calculating the Mahalanobis distance of all data are still μ N and Σ N in the normal data set, then the Mahalanobis distance of the data Xi is recorded as:
    Figure PCTCN2021095632-appb-100002
    Figure PCTCN2021095632-appb-100002
    将数据集M N的均值记为
    Figure PCTCN2021095632-appb-100003
    标准差记为
    Figure PCTCN2021095632-appb-100004
    根据统计学中的3σ准则,大部分正常数据的马氏距离分布于
    Figure PCTCN2021095632-appb-100005
    区间中,部分异常数据的马氏距离不在上述区间内,因此可通过数据的马氏距离来进行异常数据的初步检测,设置马氏距离判别阈值T up和T low,将其马氏距离不在(T low,T up)区间内的数据判定为异常数据,T up和T low的表达式如下所示:
    Denote the mean of the dataset MN as
    Figure PCTCN2021095632-appb-100003
    Standard deviation is recorded as
    Figure PCTCN2021095632-appb-100004
    According to the 3σ criterion in statistics, the Mahalanobis distance of most normal data is distributed in
    Figure PCTCN2021095632-appb-100005
    In the interval, the Mahalanobis distance of some abnormal data is not in the above interval, so the initial detection of abnormal data can be carried out through the Mahalanobis distance of the data, and the Mahalanobis distance discrimination thresholds T up and T low are set, and the Mahalanobis distance is not in ( The data in the interval T low , T up ) is determined as abnormal data, and the expressions of T up and T low are as follows:
    Figure PCTCN2021095632-appb-100006
    Figure PCTCN2021095632-appb-100006
    Figure PCTCN2021095632-appb-100007
    Figure PCTCN2021095632-appb-100007
  9. 如权利要求7所述的无菌灌装生产线异常检测方法,其特征在于:步骤S3中,自编码器的工作过程包括编码过程和解码过程,对于数据集X=(X 1,X 2,X 3,...,X n),n为数据个数,每个数据维度为m,每一个数据X i经过编码过程得到隐层表达,编码过程可描述为: The abnormality detection method for an aseptic filling production line according to claim 7, characterized in that: in step S3, the working process of the self-encoder includes an encoding process and a decoding process, and for the data set X=(X 1 , X 2 , X 3 ,...,X n ), n is the number of data, the dimension of each data is m, and each data X i is expressed in the hidden layer through the encoding process, and the encoding process can be described as:
    h i=σ e(WX i+b) h ie (WX i +b)
    其中,W和b为编码权重和偏置,σ e为编码层激活函数; Among them, W and b are the coding weights and biases, and σ e is the activation function of the coding layer;
    隐层表达经解码过程得到重构数据X′ i,解码过程可描述为: The hidden layer expresses the reconstructed data X′ i through the decoding process, and the decoding process can be described as:
    X′ i=σ d(W′h i+b') X' id (W'h i +b')
    其中,W'和b'为解码权重和偏置,取W'=W T,σ d为解码层激活函数。 Among them, W' and b' are decoding weights and biases, W'=W T , and σ d is the activation function of the decoding layer.
  10. 如权利要求9所述的无菌灌装生产线异常检测方法,其特征在于:步骤S3中隐层表达的获得方法为:The abnormality detection method of aseptic filling production line as claimed in claim 9, wherein the method for obtaining the hidden layer expression in step S3 is:
    首先,通过逐层贪婪算法调节权重和偏置使重构误差最小,对于整个训练数据集的代价函数为:First, the weights and biases are adjusted by a layer-by-layer greedy algorithm to minimize the reconstruction error. The cost function for the entire training dataset is:
    Figure PCTCN2021095632-appb-100008
    Figure PCTCN2021095632-appb-100008
    其中,L为单个数据的损失函数,L为均方误差损失函数;Among them, L is the loss function of a single data, and L is the mean square error loss function;
    然后,给代价函数添加一个L2正则化权重衰减项,λ为惩罚因子,添加KL散度作为约束条件,在其代价函数上加入稀疏惩罚项,使其形成稀疏自编码器,得到最终的损失函数为:Then, add an L2 regularization weight decay term to the cost function, λ is a penalty factor, add KL divergence as a constraint, add a sparse penalty term to its cost function to form a sparse autoencoder, and get the final loss function for:
    Figure PCTCN2021095632-appb-100009
    Figure PCTCN2021095632-appb-100009
    其中,
    Figure PCTCN2021095632-appb-100010
    为正则化项,
    Figure PCTCN2021095632-appb-100011
    为KL散度的约束条件,k为隐层神经元数量;
    in,
    Figure PCTCN2021095632-appb-100010
    is the regularization term,
    Figure PCTCN2021095632-appb-100011
    is the constraint condition of KL divergence, and k is the number of neurons in the hidden layer;
    最后,经过马氏距离判别出部分异常数据后,将不确定数据作为自编码器输入层数据,进行训练,使得式损失函数的值最小,进而得到最佳隐层表达。Finally, after identifying some abnormal data through Mahalanobis distance, the uncertain data is used as the input layer data of the autoencoder for training, so that the value of the formula loss function is the smallest, and then the best hidden layer expression is obtained.
PCT/CN2021/095632 2020-09-09 2021-05-25 Anomaly detection system and method for sterile filling production line WO2022052510A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010943208.2 2020-09-09
CN202010943208.2A CN112015153B (en) 2020-09-09 2020-09-09 System and method for detecting abnormity of sterile filling production line

Publications (1)

Publication Number Publication Date
WO2022052510A1 true WO2022052510A1 (en) 2022-03-17

Family

ID=73521259

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/095632 WO2022052510A1 (en) 2020-09-09 2021-05-25 Anomaly detection system and method for sterile filling production line

Country Status (2)

Country Link
CN (1) CN112015153B (en)
WO (1) WO2022052510A1 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114694357A (en) * 2022-03-31 2022-07-01 深圳市佳运通电子有限公司 Heating furnace management and control device and abnormal alarm system based on Internet of things
CN115034304A (en) * 2022-06-09 2022-09-09 广东技术师范大学 Graphormer-based industrial control system anomaly detection method
CN116523466A (en) * 2023-05-06 2023-08-01 福建凯邦锦纶科技有限公司 Production data tracing system and method based on big data

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112015153B (en) * 2020-09-09 2021-06-22 江南大学 System and method for detecting abnormity of sterile filling production line
CN113409105B (en) * 2021-06-04 2023-09-26 山西大学 Method and system for detecting abnormal users of e-commerce network
US20230297073A1 (en) * 2022-03-16 2023-09-21 International Business Machines Corporation Prediction and operational efficiency for system-wide optimization of an industrial processing system
CN114462895A (en) * 2022-04-12 2022-05-10 青岛华正信息技术股份有限公司 Digital transformation management method and system for enterprises
CN116341885B (en) * 2023-05-31 2023-07-25 一汽东机工减振器有限公司 Automatic production management and control system and method based on big data
CN117011111A (en) * 2023-10-07 2023-11-07 华联(天津)创新科技有限公司 Medical waste intelligent tracking management system and method

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109086999A (en) * 2018-08-02 2018-12-25 东南大学 Filling production lines remote data acquisition analysis system and its exception analysis method
CN109613428A (en) * 2018-12-12 2019-04-12 广州汇数信息科技有限公司 It is a kind of can be as system and its application in motor device fault detection method
CN110297469A (en) * 2019-05-17 2019-10-01 同济大学 The production line fault judgment method of Ensemble feature selection algorithm based on resampling
US20190327160A1 (en) * 2018-04-18 2019-10-24 ProKarma System and method for scheduling and executing automated tests
CN111259730A (en) * 2019-12-31 2020-06-09 杭州安脉盛智能技术有限公司 State monitoring method and system based on multivariate state estimation
CN112015153A (en) * 2020-09-09 2020-12-01 江南大学 System and method for detecting abnormity of sterile filling production line

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190327160A1 (en) * 2018-04-18 2019-10-24 ProKarma System and method for scheduling and executing automated tests
CN109086999A (en) * 2018-08-02 2018-12-25 东南大学 Filling production lines remote data acquisition analysis system and its exception analysis method
CN109613428A (en) * 2018-12-12 2019-04-12 广州汇数信息科技有限公司 It is a kind of can be as system and its application in motor device fault detection method
CN110297469A (en) * 2019-05-17 2019-10-01 同济大学 The production line fault judgment method of Ensemble feature selection algorithm based on resampling
CN111259730A (en) * 2019-12-31 2020-06-09 杭州安脉盛智能技术有限公司 State monitoring method and system based on multivariate state estimation
CN112015153A (en) * 2020-09-09 2020-12-01 江南大学 System and method for detecting abnormity of sterile filling production line

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114694357A (en) * 2022-03-31 2022-07-01 深圳市佳运通电子有限公司 Heating furnace management and control device and abnormal alarm system based on Internet of things
CN115034304A (en) * 2022-06-09 2022-09-09 广东技术师范大学 Graphormer-based industrial control system anomaly detection method
CN116523466A (en) * 2023-05-06 2023-08-01 福建凯邦锦纶科技有限公司 Production data tracing system and method based on big data
CN116523466B (en) * 2023-05-06 2023-11-03 福建凯邦锦纶科技有限公司 Production data tracing system and method based on big data

Also Published As

Publication number Publication date
CN112015153B (en) 2021-06-22
CN112015153A (en) 2020-12-01

Similar Documents

Publication Publication Date Title
WO2022052510A1 (en) Anomaly detection system and method for sterile filling production line
US20200064822A1 (en) Unsupervised anomaly detection, diagnosis, and correction in multivariate time series data
CN110213244A (en) A kind of network inbreak detection method based on space-time characteristic fusion
CN111428789A (en) Network traffic anomaly detection method based on deep learning
CN113255848B (en) Water turbine cavitation sound signal identification method based on big data learning
CN110263846A (en) The method for diagnosing faults for being excavated and being learnt based on fault data depth
CN113822421B (en) Neural network-based anomaly locating method, system, equipment and storage medium
CN115688035A (en) Time sequence power data anomaly detection method based on self-supervision learning
CN111504676A (en) Equipment fault diagnosis method, device and system based on multi-source monitoring data fusion
CN113419519B (en) Electromechanical product system or equipment real-time fault diagnosis method based on width learning
CN110163075A (en) A kind of multi-information fusion method for diagnosing faults based on Weight Training
US20220118987A1 (en) Method and system for fault diagnoses of intelligent vehicles
CN114118225A (en) Method, system, electronic device and storage medium for predicting remaining life of generator
CN114004331A (en) Fault analysis method based on key indexes and deep learning
CN113497725A (en) Alarm monitoring method, alarm monitoring system, computer readable storage medium and electronic equipment
CN111949496B (en) Data detection method and device
CN116684878A (en) 5G information transmission data safety monitoring system
CN117421684B (en) Abnormal data monitoring and analyzing method based on data mining and neural network
CN114707577A (en) Anomaly detection method and system based on self-confrontation variational self-encoder
CN114037001A (en) Mechanical pump small sample fault diagnosis method based on WGAN-GP-C and metric learning
CN115460061A (en) Health degree evaluation method and device based on intelligent operation and maintenance scene
KR102611399B1 (en) Abnormal detection device using 1-demension autoencoder
CN116956197B (en) Deep learning-based energy facility fault prediction method and device and electronic equipment
CN117093919B (en) Geotechnical engineering geological disaster prediction method and system based on deep learning
CN117828280B (en) Intelligent fire information acquisition and management method based on Internet of things

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: 21865571

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: 21865571

Country of ref document: EP

Kind code of ref document: A1