CN114995256A - Workshop environment monitoring device and monitoring method based on digital twins - Google Patents
Workshop environment monitoring device and monitoring method based on digital twins Download PDFInfo
- Publication number
- CN114995256A CN114995256A CN202210723205.7A CN202210723205A CN114995256A CN 114995256 A CN114995256 A CN 114995256A CN 202210723205 A CN202210723205 A CN 202210723205A CN 114995256 A CN114995256 A CN 114995256A
- Authority
- CN
- China
- Prior art keywords
- data
- monitoring
- zigbee
- sensing
- module
- 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.)
- Withdrawn
Links
Images
Classifications
-
- 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
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
- G05B19/042—Programme control other than numerical control, i.e. in sequence controllers or logic controllers using digital processors
- G05B19/0423—Input/output
-
- 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
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24215—Scada supervisory control and data acquisition
-
- Y—GENERAL 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a workshop environment monitoring device and a monitoring method based on digital twins, wherein the system comprises a sensing and control module, a centralized module, a cloud platform and an application module; the sensing and control module is responsible for collecting monitoring data and controlling functions, a sensor of the sensing and control module comprises smoke, temperature and humidity, a controller of the sensing and control module comprises a sprayer and a fan, and the ZigBee node comprises a single chip microcomputer system circuit, a battery power supply circuit and a peripheral interface circuit; the centralized module is responsible for collecting and uploading monitoring data, and a ZigBee node is arranged as a ZigBee coordinator to complete the collection and command issuing of each monitoring point data in the ZigBee ad hoc network; the cloud platform completes data storage and management by using the OneNet cloud platform, and the invention provides a method for processing monitoring data by using a deep neural network model in a monitoring system, thereby ensuring the accuracy of alarming.
Description
Technical Field
The invention relates to the field of workshop monitoring, in particular to a workshop environment monitoring device and method based on digital twins.
Background
As is known, the environment of a factory workshop has great influence on the health and the working state of workers, and a good workshop working environment can not only prevent the workers from generating various occupational diseases, but also improve the working state of the workers. Today, most factory floor environments need to be improved from several points:
(1) noise and noisy workshop environment noise can interfere thinking of workers, people can easily lose spirit and cannot concentrate, people can feel worried, work efficiency is influenced, rest and sleep are hindered, hearing loss, deafness and other symptoms can be caused when people are in a noisy environment for a long time, and symptoms such as dizziness, headache, neurasthenia, dyspepsia and the like are also accompanied, so that hypertension and cardiovascular diseases are caused. Stronger noise stimulates the vestibule of the inner ear cavity, so that people feel dizzy, nausea and vomiting, and eyeball vibration, blurred vision, and fluctuation of respiration, pulse, blood pressure and the like can be caused.
(2) The smoke dust, which is in a vehicle environment with large smoke dust for a long time, can destroy the normal defense function of the human body, cause rhinitis, pharyngolaryngitis, lung diseases and the like, and even more serious, can also cause metal poisoning and lung cancer.
(3) Open fire, if not timely extinguished, may cause fire, and bring devastating harm to factories and workers.
(4) The temperature and humidity, the worker is dazzled due to the overhigh temperature in the workshop, and the working efficiency of the worker is affected by the over-dry air and the overhigh humidity.
Therefore, a good workshop working environment is very important, and how to accurately control the change of the workshop environment in real time becomes a difficult problem to be solved by related technicians of enterprises.
Patent CN107168252A discloses a factory workshop environmental monitoring system, including data acquisition terminal, wireless mesh net and monitor terminal, data acquisition terminal is arranged in carrying the environmental data who handles to foretell wireless mesh net, this data acquisition terminal includes data acquisition device, PLC data processing center and route node, data acquisition device is used for measuring workshop environmental parameter, including smoke transducer, temperature and humidity sensor, noise detector and naked light detector, PLC data processing center then is used for carrying out the operation processing to raw data, rethread route node transmission is to foretell wireless mesh net in, monitor terminal is used for the real-time display monitoring workshop environmental quality data. The system can accurately monitor the workshop environment data in real time, find that the environment value exceeds the standard, take measures in time, avoid harming the health of workshop workers and influencing the working efficiency of the workshop workers, and is low in operation cost, convenient to maintain, reliable in monitoring and worthy of application and popularization.
In the prior art, an algorithm for performing what kind of detection data of a workshop is not provided, so that the accuracy of alarming is ensured.
Disclosure of Invention
In order to overcome the defects and shortcomings of the prior art, the invention provides a workshop environment monitoring device and a monitoring method based on a digital twin.
The technical scheme adopted by the invention is that the system comprises a sensing and control module, a centralized module, a cloud platform and an application module;
the sensing and control module is responsible for collecting monitoring data and controlling functions, a sensor of the sensing and control module comprises smoke, temperature and humidity, a controller of the sensing and control module comprises a sprayer and a fan, and the ZigBee node comprises a single chip microcomputer system circuit, a battery power supply circuit and a peripheral interface circuit; the centralized module is responsible for collecting and uploading monitoring data, and a ZigBee node is arranged to serve as a ZigBee coordinator to complete the collection and command issuing of each monitoring point data in the ZigBee ad hoc network; the NB-IoT network is used for carrying out serial port communication with the ZigBee coordinator, and the NB-IoT network is used as a gateway of the ZigBee ad hoc network and is responsible for uploading data to a cloud platform and receiving a control command; the cloud platform completes data storage and management by using the OneNet cloud platform; the application module adopts a WEB framework, and builds an application terminal based on an API and a development tool provided by the OneNet platform, and remotely checks monitoring data and remotely controls the monitoring data.
Furthermore, the sensing and control module comprises a ZigBee node, a sensor and a controller, a temperature and humidity sensor in the sensor utilizes a DHT11 digital temperature and humidity sensor, a smoke sensor in the sensor utilizes an MQ-2 smoke sensor, when the temperature and the humidity of a workshop are too high, a fan is automatically or manually started to cool and dehumidify, when the smoke sensor monitors that the indoor smoke concentration is too high, an alarm signal is sent to the coordinator, and spraying equipment is manually or automatically started.
Furthermore, the centralized module consists of a ZigBee coordinator and an NB-IoT node, completes the creation and networking management of a ZigBee ad hoc network, centralizes information from different ZigBee monitoring nodes, and finally transmits the information including temperature and humidity, smoke amount, battery power and alarm information to the NB-IoT node through a serial port;
further, the NB-IoT node uses a BC28 module, and the coordinator CC2530 performs serial communication with the NBIoT node using a serial port 1(P0_4, P0_5), so as to send an AT command or receive control information.
Furthermore, the cloud platform utilizes a Web architecture, the console automatically generates a Web application link, and the browser or the OneNET mobile phone terminal APP is utilized for remote monitoring and control;
furthermore, the application module utilizes an application management function of the OneNET cloud platform to build a Web application, the workbench provides data management and stores and calls data, different types of controls can be bound with data sources of corresponding equipment, the controls comprise display curves, instruments, pictures, control switches, knobs and command frames, one-to-one or many-to-one correspondence between the states of the controls and the data sources is realized, and the interface design comprises three parts: threshold setting and alarm signs, data display line graphs, equipment and working mode control.
The monitoring method comprises the following steps:
step S1: preprocessing data;
step S2: taking the data as state signals of the workshop environment, and respectively recording the state signals as: [ P1, P2, P3, P4, P5, P6, P7 ];
step S3: selecting a deep neural network model, respectively taking parameters [ P1, P2, P3, P4, P5, P6 and P7] as output, training the deep neural network model by adopting a sparrow search algorithm optimization method, and calculating a residual signal between the output of the model and an actual value;
step S4: computing T of residual sequence 2 Statistics, in terms of [ P1, P2, P3, P4, P5, P6, P7 per sample]Residue of 7 parametersTaking the difference as input, calculating the T of the sample residual error 2 Statistics;
step S5: determining T using kernel density estimation 2 An alarm limit for the statistics; if T of real-time operation data of workshop 2 If the statistic exceeds the control limit, the workshop is abnormal, otherwise, the workshop is in a normal state.
Further, in step S1, the data preprocessing performs normalization on the data, where the expression is:
in the formula: x is the number of ij The ith data corresponding to the jth variable; max (x) j ) Is the maximum value of the jth variable; min (x) j ) Is the minimum value of the jth variable; x' ij is the value after transformation.
Further, in step S3, the residual expression is:
in the formula:
an estimate of the jth parameter for the ith sample; y is ij Representing the actual value of the jth parameter of the ith sample.
Further, in the step S4, T 2 The expression of the statistics is:
T 2 ＝(E i -μ) T ∑ -1 (E i -μ)
in the formula: e i Is the ith sample [ P1, P2, P3, P4, P5, P6, P7]A column vector consisting of residuals of 7 parameters; mu and Σ are the mean vector and covariance matrix, respectively, of the training data residual sequence.
Further, in step S5, the kernel density expression is:
in the formula: x is a sample point; x is the number of i Is a sample observation point; h is the bandwidth; k is a kernel function and satisfies the following formula:
k (x) is not less than 0
Determining T using kernel density estimation 2 And (5) counting the alarm limit.
The invention provides a workshop environment monitoring device and a monitoring method based on digital twins, which are used for processing monitoring data by using a deep neural network model in a monitoring system so as to ensure the accuracy of alarming.
Drawings
FIG. 1 is a block diagram of the system of the present invention;
FIG. 2 is a hardware connection diagram of the sensing and control module of the present invention;
FIG. 3 is a diagram of a centralized module hardware connection of the present invention;
FIG. 4 is a flow chart of the sensing and control module routine of the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments can be combined with each other without conflict, and the present application will be further described in detail with reference to the drawings and specific embodiments.
As shown in fig. 1, the system comprises a sensing and control module, a centralized module, a cloud platform and an application module;
the sensing and control module is responsible for collecting monitoring data and controlling functions, a sensor of the sensing and control module comprises smoke, temperature and humidity, a controller of the sensing and control module comprises a sprayer and a fan, and the ZigBee node comprises a single chip microcomputer system circuit, a battery power supply circuit and a peripheral interface circuit; the centralized module is responsible for collecting and uploading monitoring data, and a ZigBee node is arranged as a ZigBee coordinator to complete the collection and command issuing of each monitoring point data in the ZigBee ad hoc network; the NB-IoT network is used for carrying out serial port communication with the ZigBee coordinator, and the NB-IoT network is used as a gateway of the ZigBee ad hoc network and is responsible for uploading data to a cloud platform and receiving a control command; the cloud platform completes data storage and management by using the OneNet cloud platform; the application module adopts a WEB framework, and builds an application terminal based on an API and a development tool provided by the OneNet platform, and remotely checks monitoring data and remotely controls the application terminal.
The system hardware consists of a sensing and control module and a centralized module. The ZigBee transmission uses a 2.4GHz general frequency band, a plurality of sensing and control modules can be arranged in the range of 10-100 m of the ad hoc network according to the requirements of actual monitoring points, and the system only needs to be provided with one centralized module, so that the design cost of the gateway is reduced.
As shown in fig. 2, the sensing and control module includes three parts, namely a ZigBee node, a sensor and a controller, the ZigBee node uses a CC2530 single chip microcomputer produced by TI corporation as a main control chip, the module is designed for a real object and a main function IO port of the single chip microcomputer, and the hardware structure of the ZigBee node includes a single chip microcomputer main control board, a power supply circuit (using a 3.3V battery for power supply) and a peripheral circuit expansion board, and three contents, namely sensor driving, controller driving and wireless radio frequency driving, are completed.
The temperature and humidity sensor adopts a DHT11 digital temperature and humidity sensor, and has calibrated digital signal output, a humidity range of 5-95% RH and a temperature range of-20-60 ℃, and the sensitivity and reliability of the temperature and humidity sensor meet the requirements of experimental practical training environment monitoring. The CC2530 singlechip uses the IO port P0_7 as a digital signal receiving port, the singlechip needs to perform character string conversion on the received temperature and humidity information, and the temperature and humidity information is stored by using variables temp and humidity for OLED display and singlechip logic control conditions.
The smoke sensor has the requirements of reliability and sensitivity, the MQ-2 smoke sensor can be adaptive, the measurement range is wide and can reach 15000 ppm, the conductivity of the sensor is increased along with the increase of the gas concentration in the environment, the analog output voltage of the sensor is also increased, a CC2530 single chip microcomputer is also increased, a port P0_6 is used as an analog input port to obtain input voltage and carry out AD conversion, and a ring can be calculated by carrying out corresponding proportional operation on the measurement range of the sensorAmbient concentration value (calculated as the maximum value of measurement 20000ppm, input voltage V) in Denotes, the reference voltage V Ref Represents):
the control to fan and equipment that sprays need use two sets of relays, and the singlechip uses two IO mouths of P1_0, P1_1 respectively to control the relay, adopts low level drive, and the relay normally open contact and then the break-make of control function equipment high voltage circuit.
As shown in fig. 3, the centralized module is composed of a ZigBee coordinator and an NB-IoT node, completes creation and networking management of a ZigBee ad hoc network, centralizes information from different ZigBee monitoring nodes, and transmits the information of the ZigBee monitoring nodes including temperature and humidity, smoke amount, battery power and alarm information to the NB-IoT node through a serial port;
the NB-IoT node utilizes the BC28 module, and the coordinator CC2530 uses the serial port 1(P0_4 and P0_5) to carry out serial port communication with the NBIoT node, so as to realize the purpose of sending AT instructions or receiving control information.
Reasonable software design is needed for controlling the ZigBee ad hoc network and controlling the communication between the ZigBee coordinator and the NB-IoT node by the system, and the reasonable software design mainly comprises sensing and control module program design, centralized module program design and energy consumption control.
As shown in fig. 4, in order to meet the power consumption control requirement of the monitoring nodes, on the basis of ensuring that each node of the ZigBee is quickly networked and can transmit data in real time, a system node dormancy and awakening mechanism is introduced, that is, each monitoring node is dormant and awakened by the ZigBee coordinator. The monitoring nodes can immediately search ZigBee channels and request for network access after initialization, after networking is successful, the monitoring nodes of each laboratory can transmit own position mark information together except for transmitting the acquired sensor data to the coordinator, the coordinator distinguishes different laboratory room data according to the received mark information, and after the monitoring nodes are successfully transmitted, the monitoring nodes can enter a sleep mode after acquiring sleep instructions, so that energy consumption control is effectively achieved.
The centralized module consists of a ZigBee coordinator and an NB-IoT node. After each monitoring node is successfully connected and enters the ad hoc network, the coordinator can receive data sent by the monitoring nodes and transmit the data to the BC28 node through a serial port, the BC28 node is used as a gateway to achieve data uploading to a cloud platform, after the data transmission is finished, the ZigBee coordinator immediately broadcasts a sleep command to each monitoring node and the BC28 node at the same time, at the moment, the system immediately enters a sleep mode, and in order to guarantee equipment synchronism, each node must be provided with the same sleep timer.
The cloud platform realizes data management, data display and control command sending, and the system adopts the OneNET cloud platform. The OneNet cloud platform provides rich APIs and various application development templates, can be quickly accessed to various sensors and intelligent hardware, is friendly in development environment and supports multi-protocol access, and enables intelligent application development to be simple. The platform adopts Web framework, the development environment provides necessary interface construction elements, the data binding is concise and visual, the control console can automatically generate Web application links, and the remote monitoring and control can be conveniently realized by using a browser or an OneNet mobile phone APP.
And the data management of the cloud platform realizes data interaction and analysis processing of the Web application and the bottom layer sensing network. The cloud platform is a bridge connecting the Web application and a bottom layer sensing network, receives and stores monitoring data from the sensing network and provides an API for Web application development. Because the environment monitoring system focuses on data real-time processing and does not need huge historical data storage, the system uses a self-contained database of the OneNet platform without additionally building a database, and the access mode of the platform equipment selects an NB-IoT mode system data management process.
In order to guarantee the stability and reliability of the system, a false alarm prevention mechanism is introduced, a monitoring point sensor possibly has data uploading deviation due to signal interference or transmission error, the false alarm prevention design idea is that if the data uploaded by one monitoring point exceeds a threshold value, a data platform cannot give an alarm and control the terminal equipment to drive at once, but firstly sends an instruction to read the monitoring data again, if the data exceeds a set threshold value again, the system can give an alarm and drive the control terminal equipment to operate, otherwise, the system can ignore the alarm, the stability of the monitoring data of the system is improved, and the loss caused by false alarm is avoided.
The application module utilizes the application management function of the OneNET cloud platform to build Web application, the workbench provides data management, stores and calls data, different types of controls can be bound with data sources of corresponding equipment, the controls comprise display curves, instruments, pictures, control switches, knobs and command frames, the state of the controls and the corresponding relation of the data sources are realized in one-to-one or many-to-one mode, and the interface design comprises three parts: threshold setting and alarm signs, data display line graphs, equipment and working mode control.
The system is designed with two working states of an automatic mode and a manual mode, and in the automatic mode, when monitoring data such as temperature, humidity or smoke concentration exceed a threshold value, the system confirms that no false alarm is generated and generates an alarm mark, and automatically drives a functional relay to act, so that functional equipment such as spraying equipment or a fan is started; in the manual mode, when monitoring data such as temperature, humidity or smog concentration exceed a threshold value, the system can also generate an alarm mark after confirming that false alarm does not exist, but the driving of the functional relay requires that an operator manually controls the starting as required, and in the manual mode, if the monitoring data do not exceed the threshold value, the functional equipment can also be manually started and stopped as required.
The monitoring method comprises the following steps:
step S1: preprocessing data;
step S2: taking the data as state signals of the workshop environment, and respectively recording the state signals as: [ P1, P2, P3, P4, P5, P6, P7 ];
step S3: selecting a deep neural network model, respectively taking parameters [ P1, P2, P3, P4, P5, P6 and P7] as output, training the deep neural network model by adopting a sparrow search algorithm optimization method, and calculating a residual signal between the output of the model and an actual value;
step S4: computing T of residual sequence 2 Statistics, in terms of [ P1, P2, P3, P4, P5, P6, P7 per sample]Residual errors of 7 parameters are used as input, and T of sample residual errors is calculated 2 Statistics;
step S5: determining T using kernel density estimation 2 Statistical alarmLimiting; if T of real-time operation data of workshop 2 If the statistic exceeds the control limit, the workshop is abnormal, otherwise, the workshop is in a normal state.
In step S1, the data preprocessing performs normalization processing on the data, where the expression is:
in the formula: x is the number of ij The ith data corresponding to the jth variable; max (x) j ) Is the maximum value of the jth variable; min (x) j ) Is the minimum value of the jth variable; x' ij is the value after transformation.
In step S3, the residual expression is:
in the formula:
an estimate of the jth parameter for the ith sample; y is ij Representing the actual value of the jth parameter of the ith sample.
In step S4, T 2 The expression of the statistics is:
T 2 ＝(E i -μ) r ∑ -1 (E i -μ)
in the formula: e i Is the ith sample [ P1, P2, P3, P4, P5, P6, P7]A column vector consisting of residuals of 7 parameters; mu and Σ are the mean vector and covariance matrix, respectively, of the training data residual sequence.
In step S5, the expression for the kernel density is:
in the formula: x is a sample point; x is the number of i Is a sample observation point; h is the bandwidth; k is a kernel function and satisfies the following formula:
k (x) is not less than 0
Determining T using kernel density estimation 2 And (5) counting the alarm limit.
The invention provides a workshop environment monitoring device and a monitoring method based on digital twins, which are used for processing monitoring data by using a deep neural network model in a monitoring system so as to ensure the accuracy of alarming.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made herein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (9)
1. The workshop environment monitoring device based on the digital twins is characterized by comprising a sensing and control module, a centralized module, a cloud platform and an application module;
the sensing and control module is responsible for collecting monitoring data and controlling functions, and sensors of the sensing and control module comprise smoke, temperature and humidity;
the controller of the sensing and control module comprises a sprayer and a fan;
the ZigBee node comprises a singlechip system circuit, a battery power supply circuit and a peripheral interface circuit;
the centralized module is responsible for collecting and uploading monitoring data, and a ZigBee node is arranged as a ZigBee coordinator to complete the collection and command issuing of each monitoring point data in the ZigBee ad hoc network; the NB-IoT network is used for carrying out serial port communication with the ZigBee coordinator, and the NB-IoT network is used as a gateway of the ZigBee ad hoc network and is responsible for uploading data to a cloud platform and receiving a control command; the cloud platform completes data storage and management by using the OneNET cloud platform; the application module adopts a WEB framework, and builds an application terminal based on an API and a development tool provided by the OneNet platform, and remotely checks monitoring data and remotely controls the monitoring data.
2. The digital twin-based plant environment monitoring device as claimed in claim 1, wherein the sensing and control module comprises a ZigBee node, a sensor and a controller, a temperature and humidity sensor in the sensor utilizes a DHT11 digital temperature and humidity sensor, a smoke sensor in the sensor utilizes an MQ-2 smoke sensor, when the temperature and humidity of the plant are too high, a fan is automatically or manually turned on to cool and dehumidify, when the smoke sensor monitors that the indoor smoke concentration is too high, an alarm signal is sent to a coordinator, and spraying equipment is manually or automatically turned on.
3. The digital twin-based workshop environment monitoring device according to claim 2, wherein the centralized module is composed of a ZigBee coordinator and an NB-IoT node, completes creation and networking management of a ZigBee ad hoc network, centralizes information from different ZigBee monitoring nodes, and finally transmits the information including temperature and humidity, smoke amount, battery power and alarm information to the NB-IoT node through a serial port;
the NB-IoT node uses a BC28 module, and the coordinator CC2530 performs serial communication with the NBIoT node using the serial port 1(P0_4, P0_5) to transmit an AT command or receive control information.
4. The device for monitoring the workshop environment based on the digital twin as claimed in claim 3, wherein the cloud platform utilizes a Web architecture, the console automatically generates a Web application link, and the APP is remotely monitored and controlled by a browser or an OneNet mobile phone terminal;
the application module utilizes an application management function of the OneNET cloud platform to build Web application, the workbench provides data management and stores and calls data, different types of controls can be bound with data sources of corresponding equipment, the controls comprise display curves, instruments, pictures, control switches, knobs and command frames, one-to-one or many-to-one correspondence between the states of the controls and the data sources is realized, and interface design comprises three parts: threshold setting and alarm signs, data display line graphs, equipment and working mode control.
5. The digital twin based plant environment monitoring method according to claim 4, wherein the monitoring method comprises the steps of:
step S1: preprocessing data;
step S2: taking the data as state signals of the workshop environment, and respectively recording the state signals as: [ P1, P2, P3, P4, P5, P6, P7 ];
step S3: selecting a deep neural network model, respectively taking parameters [ P1, P2, P3, P4, P5, P6 and P7] as output, training the deep neural network model by adopting a sparrow search algorithm optimization method, and calculating a residual signal between the output of the model and an actual value;
step S4: computing T of residual sequence 2 Statistics, in terms of [ P1, P2, P3, P4, P5, P6, P7 per sample]Residual errors of 7 parameters are used as input, and T of sample residual errors is calculated 2 Statistics;
step S5: determining T using kernel density estimation 2 An alarm limit for the statistics; t if real-time workshop operation data 2 If the statistic exceeds the control limit, the workshop is abnormal, otherwise, the workshop is in a normal state.
6. The digital twin-based plant environment monitoring method as claimed in claim 5, wherein in step S1, the data preprocessing is performed to normalize the data, and the expression is as follows:
in the formula: x is the number of ij The ith data corresponding to the jth variable; max (x) j ) Is the maximum value of the jth variable; min (x) j ) Is the minimum value of the jth variable; x' ij is the value after transformation.
7. The digital twin-based plant environment monitoring method according to claim 6, wherein in the step S3, the residual expression is as follows:
in the formula:
an estimate of the jth parameter for the ith sample; y is ij Representing the actual value of the jth parameter at the ith sample.
8. The method for monitoring a plant environment based on digital twin as set forth in claim 7, wherein in step S4, T 2 The expression of the statistics is:
T 2 ＝(E i -μ) T ∑ -1 (E i -μ)
in the formula: e i Is the ith sample [ P1, P2, P3, P4, P5, P6, P7]A column vector consisting of residuals of 7 parameters; mu and Σ are the mean vector and covariance matrix, respectively, of the training data residual sequence.
9. The digital twin-based plant environment monitoring method according to claim 8, wherein in the step S5, the expression of the nuclear density is:
in the formula: x is a sample point; x is the number of i Is a sample observation point; h is the bandwidth; k is a kernel function and satisfies the following formula:
k (x) is not less than 0
Determining T using kernel density estimation 2 And (5) counting the alarm limit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210723205.7A CN114995256A (en) | 2022-06-24 | 2022-06-24 | Workshop environment monitoring device and monitoring method based on digital twins |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210723205.7A CN114995256A (en) | 2022-06-24 | 2022-06-24 | Workshop environment monitoring device and monitoring method based on digital twins |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114995256A true CN114995256A (en) | 2022-09-02 |
Family
ID=83037698
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210723205.7A Withdrawn CN114995256A (en) | 2022-06-24 | 2022-06-24 | Workshop environment monitoring device and monitoring method based on digital twins |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114995256A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117111540A (en) * | 2023-10-25 | 2023-11-24 | 南京德克威尔自动化有限公司 | Environment monitoring and early warning method and system for IO remote control bus module |
-
2022
- 2022-06-24 CN CN202210723205.7A patent/CN114995256A/en not_active Withdrawn
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117111540A (en) * | 2023-10-25 | 2023-11-24 | 南京德克威尔自动化有限公司 | Environment monitoring and early warning method and system for IO remote control bus module |
| CN117111540B (en) * | 2023-10-25 | 2023-12-29 | 南京德克威尔自动化有限公司 | Environment monitoring and early warning method and system for IO remote control bus module |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN211503153U (en) | Indoor air quality cooperative monitoring and regulation system based on multiple targets | |
| CN104243248B (en) | wireless automation device | |
| CN201594088U (en) | Novel indoor wireless air quality monitor | |
| CN108490816A (en) | A kind of intelligent electric power cruising inspection system based on wearable technology | |
| CN103389710B (en) | A kind of environment of internet of things control method based on context aware and equipment | |
| CN113804598B (en) | Construction environment monitoring system and monitoring method based on big data | |
| CN112000030A (en) | Environment monitoring control system | |
| CN114995256A (en) | Workshop environment monitoring device and monitoring method based on digital twins | |
| CN111844039A (en) | Wisdom space system based on robot control | |
| CN110108010A (en) | Dense crowd indoor air quality-monitoring and improvement system | |
| CN107045304A (en) | Intelligent hatching system | |
| CN111243120A (en) | Environment inspection system based on big data | |
| CN110134049A (en) | Forestry monitoring system based on Internet of Things | |
| CN112987827A (en) | Intelligent heating and ventilation control system | |
| CN204723028U (en) | Environment and human life parameter monitoring system of internet of things | |
| CN106681161A (en) | Family endowment monitor | |
| CN108828967A (en) | A kind of smart home management and control system based on ROS | |
| CN108958207A (en) | A kind of intelligent substation auxiliary system comprehensive monitoring platform | |
| CN107796085A (en) | A kind of intelligent control purifier system based on NB IOT | |
| CN107063347A (en) | A kind of gas sensor, monitoring system and monitoring method | |
| CN208312636U (en) | Central air-conditioning monitoring system | |
| CN106686118A (en) | Tower type multi-parameter environment monitoring method | |
| CN209247061U (en) | Detection of Air Quality equipment and system | |
| CN206741254U (en) | Intelligent hatching system | |
| CN112486135B (en) | Intelligent home control system based on Internet of things and cloud computing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20220902 |
|
| WW01 | Invention patent application withdrawn after publication |