CN113485260A - Operation optimization control system for complex industrial process - Google Patents
Operation optimization control system for complex industrial process Download PDFInfo
- Publication number
- CN113485260A CN113485260A CN202110680748.0A CN202110680748A CN113485260A CN 113485260 A CN113485260 A CN 113485260A CN 202110680748 A CN202110680748 A CN 202110680748A CN 113485260 A CN113485260 A CN 113485260A
- Authority
- CN
- China
- Prior art keywords
- data
- industrial
- set value
- control
- value
- 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.)
- Granted
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 100
- 238000005457 optimization Methods 0.000 title claims abstract description 41
- 238000000034 method Methods 0.000 claims abstract description 141
- 230000008569 process Effects 0.000 claims abstract description 115
- 238000001514 detection method Methods 0.000 claims abstract description 75
- 238000003556 assay Methods 0.000 claims abstract description 54
- 239000003513 alkali Substances 0.000 claims description 82
- 239000003518 caustics Substances 0.000 claims description 79
- PNEYBMLMFCGWSK-UHFFFAOYSA-N aluminium oxide Inorganic materials [O-2].[O-2].[O-2].[Al+3].[Al+3] PNEYBMLMFCGWSK-UHFFFAOYSA-N 0.000 claims description 50
- 230000029087 digestion Effects 0.000 claims description 21
- 238000001704 evaporation Methods 0.000 claims description 20
- 230000008020 evaporation Effects 0.000 claims description 20
- 238000012360 testing method Methods 0.000 claims description 18
- 238000001914 filtration Methods 0.000 claims description 17
- 238000004090 dissolution Methods 0.000 claims description 15
- 238000007781 pre-processing Methods 0.000 claims description 15
- 239000002585 base Substances 0.000 claims description 13
- 239000007788 liquid Substances 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 8
- 238000002156 mixing Methods 0.000 claims description 6
- 239000012452 mother liquor Substances 0.000 claims description 6
- 230000008447 perception Effects 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 239000011550 stock solution Substances 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 230000006870 function Effects 0.000 claims description 5
- 238000002360 preparation method Methods 0.000 claims description 3
- 238000007599 discharging Methods 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims description 2
- 238000012797 qualification Methods 0.000 abstract description 7
- 230000008859 change Effects 0.000 description 13
- 238000012544 monitoring process Methods 0.000 description 11
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 10
- 238000010586 diagram Methods 0.000 description 10
- 229910052710 silicon Inorganic materials 0.000 description 10
- 239000010703 silicon Substances 0.000 description 10
- 230000005540 biological transmission Effects 0.000 description 9
- 230000007423 decrease Effects 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 238000009776 industrial production Methods 0.000 description 6
- 239000000243 solution Substances 0.000 description 5
- 238000005259 measurement Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 2
- 229910052782 aluminium Inorganic materials 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000002386 leaching Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 239000012670 alkaline solution Substances 0.000 description 1
- CSDREXVUYHZDNP-UHFFFAOYSA-N alumanylidynesilicon Chemical compound [Al].[Si] CSDREXVUYHZDNP-UHFFFAOYSA-N 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- OSMSIOKMMFKNIL-UHFFFAOYSA-N calcium;silicon Chemical compound [Ca]=[Si] OSMSIOKMMFKNIL-UHFFFAOYSA-N 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- TWNQGVIAIRXVLR-UHFFFAOYSA-N oxo(oxoalumanyloxy)alumane Chemical compound O=[Al]O[Al]=O TWNQGVIAIRXVLR-UHFFFAOYSA-N 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
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
- G05B19/418—Total 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], computer integrated manufacturing [CIM]
- G05B19/41865—Total 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], computer integrated manufacturing [CIM] characterised by job scheduling, process planning, material flow
-
- 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
- G05B2219/30—Nc systems
- G05B2219/33—Director till display
- G05B2219/33273—DCS distributed, decentralised controlsystem, multiprocessor
-
- 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]
Abstract
The embodiment of the application discloses operation optimization control system of complicated industrial process, and the system comprises: the DCS control system is used for acquiring process data of an industrial field in a current control period in a complex industrial process; the industrial cloud server is deployed with a DCS (distributed control system) database and an MES (manufacturing execution system) database, the DCS database is used for storing process data, and the MES database is used for storing assay result data and production report data; the edge industrial server is used for acquiring process data, assay result data and production report data from the industrial cloud server and obtaining a set value of each bottom layer loop through a pre-established index detection model and an operation control model; and the DCS is used for controlling the operation process of the next control period of the complex industrial process according to the set value of each bottom layer loop. The system realizes intelligent detection and intelligent operation optimization control of the operation indexes, thereby improving the control precision of the control system and the qualification rate of the production indexes.
Description
Technical Field
The application belongs to the technical field of industrial control, and particularly relates to an operation optimization control system for a complex industrial process.
Background
The complex industrial process is characterized by long flow, large lag, strong nonlinearity, complex chemical reaction and physical change, unknown interference, change of dynamic characteristics, difficulty in establishing an accurate model and the like. In the process, the control of a bottom loop is frequently interfered in an unknown large range and the set value is frequently changed, and the control target is difficult to meet by adopting a conventional PID control system; complex physical change and chemical reaction exist in the materials, the proportion of mixed components is unknown and changes along with production, the operation index is difficult to detect on line, and the operation index still depends on the laboratory value; the test value period is long, the test value period is delayed from the production process seriously, and adjacent processes in the process industrial process interfere with each other, so that frequent fluctuation of working conditions is caused, and an operator is difficult to timely and effectively give a set value of a bottom layer loop, thereby causing large fluctuation of operation indexes in the production process.
Disclosure of Invention
Technical problem to be solved
In view of the above-mentioned shortcomings and drawbacks of the prior art, the present application provides an operation optimization control system for a complex industrial process.
(II) technical scheme
In order to achieve the purpose, the technical scheme is as follows:
in a first aspect, the present application provides an operation optimization control system for a complex industrial process, the system including a DCS control system, an edge industrial server, and an industrial cloud server;
the DCS is used for acquiring process data of an industrial field in a current control period in a complex industrial process through an industrial sensor assembly and storing the acquired data to the industrial cloud server;
the industrial cloud server is provided with a DCS (distributed control system) database and an MES (manufacturing execution system) database, the DCS database is used for storing the process data, and the MES database is used for storing the test result data and the production report data;
the edge industrial server is used for acquiring the process data, the assay result data and the production report data from the industrial cloud server, acquiring a set value of each bottom loop in the next control period in the DCS control system through a pre-established index detection model and an operation control model, and sending the acquired set value to the DCS control system;
and the DCS is used for controlling the operation process of the next control period of the complex industrial process according to the set value of each bottom layer loop.
Optionally, the edge industrial server comprises:
the data acquisition module is used for acquiring process data, assay result data and production report data of the complex industrial process in the current control period from the industrial cloud server as data to be analyzed;
the operation index intelligent detection module is used for taking the process data as input and obtaining an index detection prediction value of an operation index through a pre-established index detection model, wherein the index detection model is an index detection model obtained by taking the process data as input, taking an index detection result as output and performing system identification by adopting a least square method;
and the intelligent operation optimization control module is used for taking the index detection prediction value and the data to be analyzed as input, obtaining the set value of each bottom loop in the next control period in the DCS through a pre-established operation control model, and the operation control model is a control model for performing feedforward compensation and feedback control by adopting a rule-based reasoning algorithm.
Optionally, the edge industrial server further comprises:
and the error compensation module is used for updating the model parameters of the index detection model based on the assay result data and the index detection forecast value.
Optionally, the edge industrial server further comprises:
and the data preprocessing module is used for preprocessing the abnormal value in the data to be analyzed.
Optionally, the method of preprocessing the outliers comprises first order inertial filtering, min-max normalization and 3 σ principle.
Optionally, the edge industrial server further comprises:
the system comprises a model base, a data processing module and a data processing module, wherein the model base stores an index detection model and an operation control model which are pre-established for different operation indexes;
and the working condition identification and perception module is used for identifying the working conditions according to the collected process data and selecting a corresponding index detection model and an operation control model according to the current working conditions.
Optionally, the complex industrial process comprises an alumina evaporation process, an alumina lye blending process and an alumina dissolution process;
in the alumina evaporation process, the process data comprise feeding flow loop data, refractive index and temperature, the operation indexes are the concentration of evaporation discharging caustic alkali, and the set value is the set value of a feeding flow loop;
in the alumina alkali liquor blending process, the process data comprises loop data of alkali liquor stock solution, mother liquor and high-concentration liquid alkali, the operation index is caustic alkali concentration, and the set value comprises a set value of an alkali liquor stock solution flow loop, a set value of a mother liquor flow loop and a set value of a high-concentration liquid alkali flow loop;
in the alumina digestion process, the process data comprises the alkali adding flow loop data, the conductivity and the temperature in the digestion process, the operation indexes are the digestion caustic ratio, and the set value comprises the set value of the alkali adding flow loop.
Optionally, in the alumina evaporation process, the alumina lye blending process and the alumina dissolution process, the data preprocessing method comprises:
performing first-order inertial filtering on the time sequence data in the data to be analyzed in a time window, wherein a first-order inertial filtering function is as follows:
Y(n)＝αX(n)+(1-α)Y(n-1)
wherein, α is a filter coefficient, and is 0.05, x (n) is a current sampling value, Y (n-1) is a last filtering output value, and Y (n) is a current filtering output value.
Optionally, the system further comprises:
and the assay system is used for receiving the assay value of each operation index and storing the assay value into an MES database of the industrial cloud server.
(III) advantageous effects
The beneficial effect of this application is: the application provides an operation optimization control system of complicated industrial process, and the system not only improves the control accuracy of the control system to the production process, but also improves the qualification rate of production indexes through operation index intelligent detection, intelligent operation optimization control and automatic issuing of control instructions.
Drawings
The application is described with the aid of the following figures:
FIG. 1 is a schematic diagram of an operational optimization control system for a complex industrial process according to an embodiment of the present application;
FIG. 2 is a schematic diagram of an edge industrial server architecture in one embodiment of the present application;
FIG. 3 is a schematic flow chart of a method for controlling the optimization of the operation of an alumina evaporation process according to another embodiment of the present application;
FIG. 4 is a schematic diagram illustrating the operation of an alumina evaporation process according to another embodiment of the present application;
FIG. 5 is a schematic diagram illustrating the operation of the alumina digestion process according to yet another embodiment of the present application;
FIG. 6 is a schematic diagram of an operational optimization control system for a complex industrial process according to yet another embodiment of the present application.
Detailed Description
For the purpose of better explaining the present invention and to facilitate understanding, the present invention will be described in detail by way of specific embodiments with reference to the accompanying drawings. It is to be understood that the following specific examples are illustrative of the invention only and are not to be construed as limiting the invention. In addition, it should be noted that, in the case of no conflict, the embodiments and features in the embodiments in the present application may be combined with each other; for convenience of description, only portions related to the invention are shown in the drawings.
In view of the disadvantages of the prior art, the present application provides an operation optimization control system for a complex industrial process, and the present application is described in detail by specific embodiments below.
Example one
Fig. 1 is a schematic diagram of an architecture of an operation optimization control system of a complex industrial process in an embodiment of the present application, and as shown in fig. 1, the system includes a DCS control system 100, an industrial cloud server 200, and an edge industrial server 300;
the DCS control system 100 is configured to acquire process data of an industrial field in a current control period in a complex industrial process through an industrial sensor assembly, and store the acquired data in an industrial cloud server 200;
the industrial cloud server 200 is deployed with a DCS (distributed control system) database and an MES (manufacturing execution system) database, wherein the DCS database is used for storing process data, and the MES database is used for storing assay result data and production report data;
the edge industrial server 300 is used for acquiring process data, assay result data and production report data from the industrial cloud server 200, acquiring set values of each bottom loop in the next control period in the DCS control system 100 through a pre-established index detection model and an operation control model, and sending the acquired set values to the DCS control system 100;
and the DCS control system 100 is used for controlling the operation process of the next control period of the complex industrial process according to the set value of each bottom layer loop.
The control system of the embodiment not only improves the control precision of the control system to the production process, but also improves the qualification rate of the production index through intelligent detection of the operation index, intelligent operation optimization control and automatic issuing of the control instruction.
The following describes each unit of the method of the present embodiment.
Fig. 2 is a schematic diagram of an edge industrial server architecture according to an embodiment of the present invention, and as shown in fig. 2, an edge industrial server 300 in the embodiment includes:
the data acquisition module 310 is configured to acquire process data, assay result data, and production report data in the current control cycle of the complex industrial process from the industrial cloud server 200, and use the process data, the assay result data, and the production report data as data to be analyzed;
the operation index intelligent detection module 320 is used for taking the process data as input, obtaining an index detection prediction value of the operation index through a pre-established index detection model, taking the process data as input and taking an index detection result as output, and obtaining the index detection model by performing system identification by adopting a least square method;
the intelligent operation optimization control module 330 is configured to use the index detection prediction value and the data to be analyzed as input, and obtain a set value of each bottom loop in the next control period in the DCS control system 100 through a pre-established operation control model, where the operation control model is a control model that performs feed-forward compensation and feedback control by using a rule-based inference algorithm.
In some other optional embodiments, the edge industrial server may further include:
and the error compensation module is used for updating the model parameters of the index detection model based on the assay result data and the index detection forecast values.
In some other optional embodiments, the edge industrial server further includes a data preprocessing module, configured to preprocess an abnormal value in the data to be analyzed, so as to remove measurement noise and data outliers, and to complement missing data with the filtered data.
Specifically, the method for preprocessing the abnormal value comprises first-order inertia filtering, min-max standardization and a 3 sigma principle.
In some other optional embodiments, the edge industrial server further comprises:
the model library is stored with an index detection model and an operation control model which are pre-established for different operation indexes;
and the working condition identification and perception module is used for identifying the working conditions according to the collected process data and selecting a corresponding index detection model and an operation control model according to the current working conditions.
Since the production data is reflected in different working conditions, the present embodiment adopts an event triggering manner, for example, when the heating steam pressure reaches a certain level, it is considered that a certain type of working condition is triggered, so that the corresponding index detection model and the operation control model under the working condition are selected in the library. Different production condition changes and production states correspond to different operation indexes, so that a corresponding index detection model and an operation control model are selected, the index detection model detects to obtain a detection value, and the operation control model obtains a set value of each bottom loop in the next control period in the DCS based on the detection value.
In some other optional embodiments, the operation optimization control system may further include an assay system, and the assay system is configured to receive the assay value of each operation index and save the assay value in an MES database of the industrial cloud server.
In this embodiment, the complex industrial process includes an alumina evaporation process, an alumina lye preparation process, and an alumina dissolution process;
in the alumina evaporation process, the process data comprises feed flow loop data, refractive index and temperature, the operation index is the concentration of the evaporation discharge caustic alkali, and the set value is the set value of the feed flow loop;
in the process of preparing the alumina alkali liquor, the process data comprises the loop data of the alkali liquor stock solution, the mother liquor and the high-concentration liquid alkali, the operation index is the concentration of the caustic alkali, and the set value comprises the set value of the flow loop of the alkali liquor stock solution, the set value of the flow loop of the mother liquor and the set value of the flow loop of the high-concentration liquid alkali;
in the alumina digestion process, the process data comprises the alkali adding flow loop data, the conductivity and the temperature in the digestion process, the operation indexes are the digestion caustic ratio, and the set value comprises the set value of the alkali adding flow loop.
In the alumina evaporation process, the alumina alkaline solution preparation process and the alumina dissolution process, the data preprocessing method comprises the following steps:
performing first-order inertial filtering on time sequence data in the data to be analyzed in a time window, wherein a first-order inertial filtering function is as follows:
Y(n)＝αX(n)+(1-α)Y(n-1) (A1)
wherein, α is a filter coefficient, and is 0.05, x (n) is a current sampling value, Y (n-1) is a last filtering output value, and Y (n) is a current filtering output value.
It should be noted that the production report data in this embodiment may include raw ore analysis data, such as an aluminum-silicon ratio and a calcium-silicon ratio of the ore; assay values for key operational indicators, such as caustic ratio assay values, caustic concentration assay values, and the like, may also be included.
Example two
The implementation example is an evaporation process of a certain large-scale alumina plant in China, the evaporation process is a long-flow process, the feeding flow is frequently interfered in an unknown large range and the set value is frequently changed, the control target is difficult to meet by adopting a conventional PID control method, and the operation index is difficult to detect and automatically set on line because the feed liquid components are complex, the proportion of mixed components is unknown and changes along with the production.
Fig. 3 is a flow chart of an operation optimization control method of an alumina evaporation process in another embodiment of the present application, and the following description is made in conjunction with the steps in the embodiment of fig. 3.
Step 1: acquiring process data, assay system data and production report data in a PHD database and a MongoDB of an industrial cloud server through a data transmission module;
step 2: data storage, namely storing the data acquired by the industrial cloud server into a database of the edge industrial server for subsequent intelligent detection of operation indexes and intelligent operation optimization control;
and step 3: the acquired data are screened through a characteristic variable selection module, relevant variables contributing to the intelligent setting model are selected, and the complexity and the training time of the model are reduced on the premise of not influencing the precision requirement of the model;
and 4, step 4: filtering unavoidable measurement noise and data outliers in industrial production data through a data preprocessing module, completing missing data, judging whether a model exists in edge equipment or not by adopting different processing methods aiming at different production data, and executing the step 6 if the model exists, or executing the step 5 if the model does not exist;
and 5: establishing intelligent operation control models corresponding to the intelligent detection models of the indexes aiming at different operation indexes in the edge equipment, and storing the models into a model library;
step 6: judging whether each operation index intelligent detection model and each intelligent operation control model meet the precision requirement, if so, executing the step 7, otherwise, executing the step 3;
and 7: the working condition identification and sensing module identifies working conditions according to data collected in the production process and selects an intelligent detection and intelligent operation control model suitable for each operation index according to the current working condition;
and 8: judging whether a control period is reached, if the control period is reached, executing to a step 9, and if the control period is not reached, executing to a step 7;
and step 9: the current working condition recognition sensing result and each operation index intelligent detection and intelligent operation control model are combined to give a set value of each bottom layer loop in the control system;
step 10: and sending the set value of each bottom layer loop to the real-time system of the edge controller through the data transmission module.
In this embodiment, the development of the industrial software of the complex industrial process intelligent optimization operation control system defined by the software mainly includes database communication connection, data reading and writing, intelligent detection of operation indexes, software implementation of an intelligent operation optimization control algorithm, and process monitoring and manual operation of a corresponding human-computer interface, and the software communicates with a bottom Distributed Control System (DCS) to implement reading of loop data and issuing of control instructions. The bottom layer algorithm and data communication in the software are developed by adopting Python language, and the front-end operator monitoring interface is developed by adopting Foxdraw configuration software.
In this embodiment, the data transmission adopts an OPC protocol and a TCP/IP protocol, the production process data is stored in the PHD database in the industrial cloud server, and the assay system data and the production report data are stored in the MongoDB in the industrial cloud server.
In this embodiment, the feature selection method adopts a wrapping method, the data preprocessing method includes first-order inertial filtering, min-max standardization and 3 σ principle processing abnormal values, the operation index intelligent detection main model adopts least square identification, and the intelligent operation control adopts a rule-based reasoning system.
In this embodiment, only carry out intelligent operation control to a operating mode, so do not relate to operating mode discernment and perception process.
Fig. 4 is a schematic diagram of an operation optimization control of an alumina evaporation process according to another embodiment of the present application, and as shown in fig. 4, the operation optimization control system of the alumina evaporation process includes an index intelligent detection module, an intelligent feedback control module, an intelligent feedforward compensation module, a high-performance controller, and the like. The index intelligent detection module utilizes the original data of the detection instrument and the industrial assay data to realize the online intelligent detection of the operation index of the industrial process. The intelligent feedback controller uses the intelligent detection index as a feedback signal to give a feedback set value of the industrial process, the intelligent feedback controller also gives a proper set value by combining the current operation target value when the test value of the operation index approaches the upper limit and the lower limit or exceeds the upper limit and the lower limit according to the upper limit and the lower limit of the operation index input by the plant operator, and gives an alarm prompt to the operator, and the intelligent feedback controller also gives a set value of the alkali adding flow according to the difference value between the test value and the target value of the operation index. The intelligent feedforward compensator combines the working condition change of the production process to give a feedforward compensation value. And adding the feedback set value and the feedforward compensation value to obtain a final set value. The intelligent controller generates a control quantity every other control period, and the control quantity is a discrete signal; the retainer is used for converting discrete signals into continuous signals and inputting the continuous signals into a controller of the bottom loop. The bottom layer loop realizes high-precision tracking control on an intelligent set value through a high-performance controller, so that intelligent operation optimization control on operation indexes is realized. In addition, the index intelligent detection model can update the model parameters based on the test values and the index detection results.
The software-defined intelligent operation optimization control technology for the complex industrial process designed by the technical scheme is successfully applied to the alumina evaporation process. The industrial application result shows that the control effect of the control system is obviously superior to that of manual setting. Table 1 shows the caustic ratio intelligent detection performance evaluation table, and table 2 shows the intelligent setting performance evaluation table.
TABLE 1
TABLE 2
As can be seen from tables 1 and 2, in industrial experiments, the concentration of the evaporated caustic alkali can be controlled within a target range by adopting the intelligent detection and intelligent operation optimization control technology of the operation index, wherein the intelligent detection of the operation index is improved by 86 percent compared with the precision (mean square error, MSE) of the existing instrument, the qualification rate of a caustic alkali concentration interval is improved by 31.6 percent compared with the manually set value, the MSE is reduced by 35.4 percent, and the average absolute error (MAE) is reduced by 23.1 percent.
EXAMPLE III
The embodiment is a digestion process of a certain large-scale alumina plant in China, the digestion process in alumina production is a key process for carrying out digestion reaction on alumina in high-concentration alkali liquor and ore pulp under the conditions of high temperature and high pressure, and the operation index of the process is caustic ratio (ak) and reflects the digestion quality.
The control method of the aluminum oxide dissolution process specifically comprises the following steps:
step 101, determining initial set values of all operation parameters in an alkali adding flow loop when a current alumina dissolution working condition is started based on a pre-established domain knowledge base of an alumina dissolution process;
102, collecting data in the alkali adding flow loop, wherein the data comprises loop data, process data and assay data which are related to caustic ratio in the alumina digestion process;
103, adjusting the current set value of the alkali adding flow loop by adopting a data backtracking mode according to the assay data in the collected data, a preset target value and the field knowledge base;
and/or compensating the current set value of the alkali adding flow loop by adopting a feedforward compensation mode according to loop data, process data and the field knowledge base in the acquired data.
In step 103, an automatic control method of the caustic ratio of the operation index of the alumina digestion process is proposed by combining feedback control and feedforward compensation, and a final set value is calculated from the set value of the adjusted alkali addition flow rate loop and/or the set value of the compensated alkali addition flow rate loop in step 103.
The knowledge base establishing method comprises the following steps: performing data mining on historical production data in a preset time period by adopting a decision tree regression method to obtain calculation knowledge; summarizing and summarizing the expert experience of field operators in the alumina dissolution process to obtain expert knowledge; integrating the calculation knowledge and the expert knowledge, and establishing a domain knowledge base of the alumina dissolution process; wherein the computational knowledge and the expert knowledge are both stored in the domain knowledge base as a set of rules in the form of IF-THEN.
Fig. 5 is a schematic diagram of the operation optimization control of the alumina digestion process in another embodiment of the present application, and the principle of the intelligent feedforward compensator and the intelligent feedback controller will be described below with reference to fig. 5.
Firstly, the method comprises the following steps: feedback controlled process
Aiming at the problem of large time lag in the dissolution process of alumina, the current set value of the alkali adding flow loop is adjusted by adopting a data backtracking mode according to the assay data in the collected data, a preset target value and the field knowledge base, and the method comprises the following steps:
calculating the test value r (k) and the target value r of the caustic ratio in the alumina dissolution process by adopting a data backtracking mode*The difference e (k) and the change rate nk _ t of the caustic ratio online intelligent detection;
according to the rule of the IF-THEN form in the domain knowledge base, when the difference e (k) or the change rate nk _ t meets the condition, the current set value of the alkali adding flow loop is adjusted.
Wherein the tested value r (k) and the target value r of the caustic ratio of the digestion process*The difference e (k) r (k) -r*Divided into five intervals, b1-b50.01, 0.03, 0.04, 0.05, respectively, and the adjustment unit o of the alkali addition flow circuit1～o510, 15, 20, 30, 50 respectively.
The target value r*Given by the operator during the dissolution process.
Calculating the caustic ratio value on-line intelligent detection in the time window t according to the formula (A2)fInternal rate of change, caustic ratio, over time window tfInner rate of change, in this example, tfIt can be taken for 15 min.
Where up _ t is a time window tfThe sum of the change rates of which the inside is larger than zero, down _ t is the sum of the change rates of which the inside is smaller than zero in a time window, and th is a threshold value of the sum of the change rates;
ak _ t ═ 1 denotes the time window tfThe internal caustic ratio increased, and ak _ t is 0, which indicates the time window tfThe internal caustic ratio decreased, and ak _ t-1 indicates the time window tfThe internal caustic ratio was unchanged.
The specific inference rule is as follows:
step A: and in the event triggering mode, the testing value of the operation index ak is used as a feedback signal to adjust the set value, the current moment is k, the testing value of the operation index is r (k), and the corresponding historical set value of the alkali adding flow is y (k-T)σ-Tδ) Wherein T isσLag time, T, for complex industrial processesδThe time interval from ak sampling to result, in this embodiment, TσTaking for 130min, TδTaking for 40 min.
Wherein, the test value can be obtained by sampling at a pipeline sampling port of the alumina dissolution system by an operator and sending the sample to a laboratory for testing.
(1) When the assay value fluctuation of the caustic ratio is small, the set value of the current alkali adding flow loop is maintained to be unchanged:
Rule1:IF|e(k)|≤b1 THEN y1sp(k)＝y1sp(k-Tσ-Tδ) (3)
the subscript sp denotes the abbreviation of the set point (setpoint), the subscript y1spIs the set point of the alkali addition flow loop.
(2) When the assay value of the caustic ratio rises, the error is in the interval (b)1,b2]In order to ensure that the caustic ratio is maintained within the target range, the caustic addition flow rate needs to be reduced on the basis of the historical set value of the caustic addition flow rate loop:
Rule2:IF b1＜e(k)≤b2 THEN y1sp(T1)＝y1sp(k-Tσ-Tδ)-o1 (4)
(3) when the assay value of the caustic ratio rises, the error is in the interval (b)2,b3]In order to ensure that the caustic ratio is maintained within the target range, the caustic addition flow rate needs to be reduced on the basis of the historical set value of the caustic addition flow rate loop:
Rule3:IFb2＜e(k)≤b3 THEN y1sp(k)＝y1sp(k-Tσ-T)-o2 (5)
(4) when the assay value of the caustic ratio rises, the error is in the interval (b)3,b4]In order to ensure that the caustic ratio is maintained within the target range, the caustic addition flow rate needs to be reduced on the basis of the historical set value of the caustic addition flow rate loop:
Rule4:IFb3＜e(k)≤b4 THEN y1sp(k)＝y1sp(k-Tσ-T)-o3 (6)
(5) when the assay value of the caustic ratio rises, the error is in the interval (b)4,b5]In order to ensure that the caustic ratio is maintained within the target range, the caustic addition flow rate needs to be reduced on the basis of the historical set value of the caustic addition flow rate loop:
Rule5:IFb4＜e(k)≤b5 THEN y1sp(k)＝y1sp(k-Tσ-T)-o4 (7)
(6) when the assay value of the caustic ratio rises, and e (T)1)＞b5The assay value of the caustic ratio is higher, and in order to ensure that the caustic ratio is maintained in the target interval, the alkali addition flow rate needs to be reduced on the basis of the historical set value of the alkali addition flow rate loop:
Rule6:IFe(k)＞b5 THEN y1sp(k)＝y1sp(k-Tσ-T)-o5 (8)
(7) when the assay value of the caustic ratio decreases, the error is in the range [ -b ]2,-b1) In order to ensure that the caustic ratio is maintained within the target range, the caustic addition flow rate needs to be increased on the basis of the historical set value of the caustic addition flow rate loop:
Rule7:IF-b2≤e(k)＜-b1 THEN y1sp(k)＝y1sp(k-Tσ-T)+o1 (9)
(8) when the assay value of the caustic ratio decreases, the error is in the range [ -b ]3,-b2) In order to ensure that the caustic ratio is maintained within the target range, the caustic addition flow rate needs to be increased on the basis of the historical set value of the caustic addition flow rate loop:
Rule8:IF-b3≤e(k)＜-b2 THEN y1sp(k)＝y1sp(k-Tσ-T)+o2 (10)
(9) when the assay value of the caustic ratio decreases, the error is in the range [ -b ]4,-b3) In order to ensure that the caustic ratio is maintained within the target range, the caustic addition flow rate needs to be increased on the basis of the historical set value of the caustic addition flow rate loop:
Rule9:IF-b4≤e(k)＜-b3 THEN y1sp(k)＝y1sp(k-Tσ-T)+o3 (11)
(10) when the assay value of the caustic ratio decreases, the error is in the range [ -b ]5,-b4) In order to ensure that the caustic ratio is maintained within the target range, the caustic addition flow rate needs to be increased on the basis of the historical set value of the caustic addition flow rate loop:
Rule10:IF-b5≤e(k)＜-b4 THEN y1sp(k)＝y1sp(k-Tσ-T)+o4 (12)
(11) when the assay value of the caustic ratio decreases, and e (T)1)＜-b5The test value of the caustic ratio is low, and in order to ensure that the caustic ratio is maintained in a target interval, the alkali adding flow rate needs to be increased on the basis of the historical set value of the alkali adding flow rate loop:
Rule11:IFe(k)＜-b5 THEN y1sp(k)＝y1sp(k-Tσ-T)+o5 (13)
and B: in the time triggering mode, caustic ratio ak online intelligent detection data is used as a feedback signal to finely adjust a set value of an alkali adding flow loop in a feedback control period, the current time is k, the ak online intelligent detection index is q (k), and the corresponding historical set value is y (k-T)σ) In the feedback control period TfThe setting value is adjusted, in this embodiment, TfTaking for 15 min.
(1) In a time window tfThe line intelligent detection data of the internal caustic ratio value is not changed and reaches the control period, and the set value of the alkali adding flow loop keeps the historical set value unchanged:
Rule12:IFt＞Tf and ak_t＝-1THEN y1sp(k)＝y1sp(k-Tσ) (14)
where T is the sampling time, TfIndicating a feedback control period.
(2) In a time window tfThe internal caustic ratio is trending upward and reaches a control period, reducing the caustic addition flow rate based on the historical setpoint of the caustic addition flow loop:
Rule13:
(3) in a time window tfThe internal caustic ratio is in a descending trend and reaches a control period, and the alkali adding flow rate is increased on the basis of the historical set value of the alkali adding flow rate loop:
Rule14:
second, feed forward compensation process
Compensating the set value of the alkali adding flow loop according to the change conditions of A/S before silicon, A/S of red mud, the concentration Nk of caustic alkali of the circulating alkali liquor and the caustic ratio ak of the circulating alkali liquor, wherein A/S represents the ratio of aluminum to silicon:
(1) silicon front feed-forward compensator
The pre-silicon A/S reflects the ore components entering the leaching process, directly influences the alkali blending in the leaching process and the final alumina yield, when the pre-silicon A/S is increased, the alkali adding flow needs to be increased, when the pre-silicon A/S is reduced, the alkali adding flow needs to be reduced, the difference of the A/S between two adjacent pre-silicon is recorded as delta1In the feed forward compensation period Tb1The setting value is adjusted, in this embodiment, Tb1Taking for 120 min.
The specific setting rules are as follows:
1) as the pre-silicon a/S increases, compensation for the set point of the addition of caustic flow loop needs to be added.
Rule15:
2) As the pre-silicon a/S decreases, less compensation is required to the set point of the addition base flow loop.
Rule16:
φ1(k) Is the flow compensation value of the feed-forward compensator before silicon.
(2) Red mud A/S feedforward compensator
The red mud A/S reflects the digestion effect in the digestion process, if the red mud A/S is too high, a large amount of aluminum ore is lost into the red mud, and the resource waste is caused, so the red mud A/S needs to be controlled within a limit, and the A/S difference value of two adjacent red mud is recorded as delta2In the feed forward compensation period Tb2The setting value is adjusted, in this embodiment, Tb2Taking for 120 min.
The specific setting rules are as follows:
1) when the red mud A/S is increased, the compensation of the set value of the alkali adding flow loop needs to be increased.
Rule17:
2) When the red mud a/S decreases, compensation for the set point of the caustic addition flow loop needs to be reduced.
Rule18:
φ2(k) Flow compensation value of red mud feedforward compensator
3) Circulating alkali liquor caustic alkali concentration Nk feedforward compensator
The dissolved alkali solution is the circulating alkali solution, therefore, the alkali adding flow needs to be reduced when the circulating alkali solution Nk is increased, the alkali adding flow needs to be increased when the circulating alkali solution Nk is reduced, and the difference value of two adjacent circulating alkali solutions Nk is recorded as delta3In the feed forward compensation period Tb3The setting value is adjusted, in this embodiment, Tb3Taking for 120 min.
The specific setting rules are as follows:
1) when the circulating lye NK decreases, an additional compensation of the setpoint of the lye flow circuit is required.
Rule19:
2) When the circulating lye NK increases, it is necessary to reduce the compensation for the setpoint of the lye flow circuit.
Rule20:
φ3(k) Is the flow compensation value of the caustic concentration Nk feed forward compensator.
(3) Circulating alkali liquor caustic ratio ak feedforward compensator
Dissolving out alkali, namely circulating alkali liquor, when the circulating alkali liquor ak is increased, the alkali adding flow needs to be reduced, when the circulating alkali liquor ak is decreased, the alkali adding flow needs to be increased, and the difference value of two adjacent circulating alkali liquor ak is recorded as delta4In the feed forward compensation period Tb4The setting value is adjusted, in this embodiment, Tb4Taking for 120 min.
The specific setting rules are as follows:
1) when the circulating lye ak decreases, an additional compensation of the set point of the lye flow circuit is required.
Rule21:
2) When the circulating lye ak increases, the compensation of the setpoint of the lye flow circuit needs to be reduced.
Rule22:
φ4(k) Is the flow compensation value of the caustic ratio ak feed forward compensator.
Thirdly, calculating a final set value of the alkali adding flow loop according to the adjusted alkali adding flow set value and the compensated alkali adding flow set value, namely:
ysp(k)＝y1sp(k)+φ1(k)+φ2(k)+φ3(k)+φ4(k) (25)
and finally, limiting the set value of the alkali adding flow circuit, and downloading the set value to the bottom layer flow control circuit.
In the formula, ymin＝80m3/h，ymax＝300m3The flow rate/h is respectively the lower limit value and the upper limit value of the alkali adding flow rate.
In the embodiment, by combining the operation condition of the dissolution process, the current set value of the alkali adding flow loop control is adjusted in a variable feedback control period by utilizing loop data, process data, assay data and rule reasoning which are collected in real time on the basis of the initial set value; thereby realizing the automatic control of the operation index in the alumina dissolving process and effectively improving the qualification rate of caustic ratio.
Example four
In a third aspect of the present application, a complex industrial process intelligent operation optimization control system is provided through yet another embodiment, fig. 5 is a schematic structural diagram of an operation optimization control system of a complex industrial process in yet another embodiment of the present application, as shown in fig. 5, the system includes a control system, an edge industrial server, an industrial cloud server, and an assay system, and the following description is made on each part of the system of this embodiment.
The control system comprises a DCS control system, an operating computer, an industrial sensor and the like, and is used for realizing data acquisition of the industrial production sensor, reading and downloading of control quantity, monitoring of the production process and remote operation of an executing mechanism.
The DCS control system realizes distributed control and centralized management of industrial production equipment and production processes, transmits related signal data to the operation computer and realizes monitoring of the production equipment and the processes.
The operation computer is used for remotely monitoring the production process and the industrial equipment, comprises the liquid level of the equipment, the opening degree of a valve, the start and stop of the industrial equipment and the like, and is also used for monitoring the running state of the intelligent running optimization control software and adjusting the control instruction. In the manual control stage, an operator can issue a control instruction to the DCS control system through operating the computer.
The industrial sensor is used for online detection of relevant variables such as production equipment, production materials and the like, and transmits a measurement signal to the DCS server for equipment monitoring and online detection of process variables.
The industrial cloud server is a private cloud built in a factory, and a DCS (distributed control system) database and an MES (manufacturing execution system) database are deployed in the industrial cloud server and used for collecting and centrally storing industrial field process data and assay data.
The DCS database in the industrial cloud server is used for storing process data and control loop data in the industrial production process, and related control instructions can be written into the DCS database to serve as data transfer between the upper computer and the lower computer; and deploying an MES database on the industrial cloud server for storing partial data, assay system data, production report data and the like in the DCS database and using the MES database as a background database of the MES platform.
Specifically, the PHD database in the industrial cloud platform realizes communication between the industrial server and the DCS server through the OPC server, stores process data and control loop data in the industrial production process, can write related control instructions into the PHD database, and sends the related control instructions to the DCS control system through the OPC server to serve as data transfer between the upper computer and the lower computer; the MongoDB is deployed on the industrial cloud platform and used for storing partial data, assay system data, production report data and the like in the PHD database, and the MongoDB is used as a background database of the MES platform.
Aiming at the problem of data isolated island in the complex industrial process, instrument data, assay data and production report data in the production process are stored in a database of an industrial cloud server, and relevant data of the whole plant are concentrated in the industrial cloud server; aiming at the problems that a large amount of data exist in an industrial field and the production working conditions are variable and difficult to judge, industrial software is deployed in an edge industrial server, so that the intelligent detection of operation indexes, the identification and perception of the working conditions, the intelligent operation optimization control and the automatic issuing of control instructions are realized, the control precision is improved, and the cost and the delay of data transmission are reduced.
The testing system is used for carrying out online input of a laboratory index testing value, and storing the testing value into an MES database of the industrial cloud server, so that data intercommunication between a testing workshop and a production workshop is realized.
The edge industrial server virtualizes a real-time system and a plurality of non-real-time systems in the edge industrial server by utilizing a software definition technology, wherein the real-time system is used for realizing high-performance control of a bottom layer loop, and the non-real-time systems are used for realizing intelligent detection and intelligent operation optimization control of operation indexes. Table 3 is a table of performance parameters of the edge industrial server in this embodiment.
TABLE 3
The real-time system is provided with an operation module for operating the binary control program compiled by the compiler and a data module for writing data into the real-time database, and in order to ensure the real-time performance, the real-time system continuously operates the program according to an operation period and writes an operation result into the database under the unified scheduling of the operating system.
Specifically, the real-time system includes an operation module for executing a control program and a data communication module written in a real-time database. The compiler firstly compiles a program written by the PLC into a machine language and sends the program to the running module, the running module obtains the program which can be directly run after compiling and starts to execute under the scheduling of the operating system after verification, the data communication module writes operation data into the real-time database according to the point table after execution is finished, and sends a signal to the external device to output the calculation result to the outside. In order to ensure stability and execution speed, the real-time system is configured according to the requirement of light weight, and system-level isolation is realized among a plurality of real-time systems and real-time and non-real-time systems, so that mutual execution is not influenced.
And deploying a database in the non-real-time system for storing algorithm-related data as a background database of the intelligent operation optimization control system. The database here may be an SQL Server database.
The method is characterized in that a software definition technology is adopted, and the industrial application software functions are developed in a modularized manner on an edge server, and the modularized development comprises a data transmission module, a characteristic variable selection module, a data preprocessing module, a timestamp alignment module, an operation index intelligent detection module, a working condition identification and sensing module, an intelligent operation control module and a production index monitoring module.
The data transmission module is used for data transmission between the edge industrial server and the industrial cloud server, reading process data, assay system data and production and management data in the PHD database and the MongoDB, and issuing online detection indexes and control instructions in the edge industrial server to the real-time system.
The SQL Server database is deployed in the edge industrial Server and used for storing relevant data of a complex industrial application algorithm, is used as a background database of the intelligent operation optimization control system, and is used for monitoring production indexes and intelligent operation optimization control.
The characteristic variable selection module is used for screening the acquired data, selecting related variables contributing to the intelligent setting model, and reducing the complexity and training time of the model on the premise of not influencing the precision requirement of the model.
The data preprocessing module is used for processing inevitable measurement noise and data outliers in industrial production data, completing missing data, improving model precision and anti-interference performance, and adopting different processing methods for different production data.
And the timestamp alignment module is used for aligning the sampling time of the assay data and the process data in the production process.
The intelligent detection module for the operation indexes consists of a main model and an error compensation model, and updates the model parameters in real time by using process data and assay data acquired in an industrial field to realize real-time online detection of the operation indexes.
And the working condition identification and perception module is used for analyzing data of related procedures in the complex industrial process and identifying the current working condition by identifying the change of production conditions of adjacent procedures.
The intelligent operation optimization control module adopts relevant data and control loop data of the production process, gives a set value of the control loop by combining a complex industrial process intelligent operation optimization control algorithm and the current production working condition.
The production index monitoring module is used for monitoring the operation indexes in the production process and providing functions of checking historical trends, counting qualification rate, adjusting production indexes and the like.
The control system, the edge industrial server, the industrial cloud server and the assay system realize data intercommunication through communication equipment such as an industrial switch and a field bus.
For example, various sensors, flowmeters and the like in the industrial field are accessed into an IO board card of the DCS through hard wires to read process data, data transmission between the DCS control system and a PHD database in an industrial cloud platform can be realized through an OPC server, the edge industrial server and the industrial cloud server are in the same local area network, and data transmission between the edge industrial server and the industrial cloud server is realized through a TCP/IP protocol.
The embodiment provides an end-edge-cloud cooperative implementation technology consisting of a control system, an edge industrial server, an industrial cloud server and a test system, high-performance control of a bottom loop in a complex industrial process is realized, intelligent detection and intelligent operation optimization control of operation indexes are realized, the labor intensity of workers is reduced, and the qualification rate of the operation indexes is improved. In addition, the system adopts a software definition control technology to decouple software and hardware, thereby improving the self-defining degree of the equipment, allowing an operator to modify the configuration of the control system at any time according to the field condition, reducing the hardware quantity of the control system, improving the stability and reducing the maintenance cost of the equipment due to the addition of a virtualization technology.
In the above embodiments disclosed in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described apparatus and method embodiments are illustrative only, and it will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention should also include such modifications and variations.
Claims (9)
1. The operation optimization control system of the complex industrial process is characterized by comprising a DCS control system, an edge industrial server and an industrial cloud server;
the DCS is used for acquiring process data of an industrial field in a current control period in a complex industrial process through an industrial sensor assembly and storing the acquired data to the industrial cloud server;
the industrial cloud server is provided with a DCS (distributed control system) database and an MES (manufacturing execution system) database, the DCS database is used for storing the process data, and the MES database is used for storing the test result data and the production report data;
the edge industrial server is used for acquiring the process data, the assay result data and the production report data from the industrial cloud server, acquiring a set value of each bottom loop in the next control period in the DCS control system through a pre-established index detection model and an operation control model, and sending the acquired set value to the DCS control system;
and the DCS is used for controlling the operation process of the next control period of the complex industrial process according to the set value of each bottom layer loop.
2. The system of claim 1, wherein the edge industrial server comprises:
the data acquisition module is used for acquiring process data, assay result data and production report data of the complex industrial process in the current control period from the industrial cloud server as data to be analyzed;
the operation index intelligent detection module is used for taking the process data as input and obtaining an index detection prediction value of an operation index through a pre-established index detection model, wherein the index detection model is an index detection model obtained by taking the process data as input, taking an index detection result as output and performing system identification by adopting a least square method;
and the intelligent operation optimization control module is used for taking the index detection prediction value and the data to be analyzed as input, obtaining the set value of each bottom loop in the next control period in the DCS through a pre-established operation control model, and the operation control model is a control model for performing feedforward compensation and feedback control by adopting a rule-based reasoning algorithm.
3. The system of claim 2, wherein the edge industrial server further comprises:
and the error compensation module is used for updating the model parameters of the index detection model based on the assay result data and the index detection forecast value.
4. The system of claim 2, wherein the edge industrial server further comprises:
and the data preprocessing module is used for preprocessing the abnormal value in the data to be analyzed.
5. The system of claim 4, wherein the method of preprocessing the outliers comprises first order inertial filtering, min-max normalization, and 3 σ principle.
6. The system of claim 2, wherein the edge industrial server further comprises:
the system comprises a model base, a data processing module and a data processing module, wherein the model base stores an index detection model and an operation control model which are pre-established for different operation indexes;
and the working condition identification and perception module is used for identifying the working conditions according to the collected process data and selecting a corresponding index detection model and an operation control model according to the current working conditions.
7. The system of claim 1, wherein the complex industrial process comprises an alumina evaporation process, an alumina lye blending process, and an alumina digestion process;
in the alumina evaporation process, the process data comprise feeding flow loop data, refractive index and temperature, the operation indexes are the concentration of evaporation discharging caustic alkali, and the set value is the set value of a feeding flow loop;
in the alumina alkali liquor blending process, the process data comprises loop data of alkali liquor stock solution, mother liquor and high-concentration liquid alkali, the operation index is caustic alkali concentration, and the set value comprises a set value of an alkali liquor stock solution flow loop, a set value of a mother liquor flow loop and a set value of a high-concentration liquid alkali flow loop;
in the alumina digestion process, the process data comprises the alkali adding flow loop data, the conductivity and the temperature in the digestion process, the operation indexes are the digestion caustic ratio, and the set value comprises the set value of the alkali adding flow loop.
8. The system of claim 1, wherein in the alumina evaporation process, the alumina lye preparation process and the alumina dissolution process, the data preprocessing method comprises the following steps:
performing first-order inertial filtering on the time sequence data in the data to be analyzed in a time window, wherein a first-order inertial filtering function is as follows:
Y(n)＝αX(n)+(1-α)Y(n-1)
wherein, α is a filter coefficient, and is 0.05, x (n) is a current sampling value, Y (n-1) is a last filtering output value, and Y (n) is a current filtering output value.
9. The system of claim 1, further comprising:
and the assay system is used for receiving the assay value of each operation index and storing the assay value into an MES database of the industrial cloud server.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110680748.0A CN113485260B (en) | 2021-06-18 | 2021-06-18 | Operation optimization control system for complex industrial process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110680748.0A CN113485260B (en) | 2021-06-18 | 2021-06-18 | Operation optimization control system for complex industrial process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113485260A true CN113485260A (en) | 2021-10-08 |
| CN113485260B CN113485260B (en) | 2022-10-28 |
Family
ID=77935533
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110680748.0A Active CN113485260B (en) | 2021-06-18 | 2021-06-18 | Operation optimization control system for complex industrial process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113485260B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040075689A1 (en) * | 2002-10-22 | 2004-04-22 | Duncan Schleiss | Smart process modules and objects in process plants |
| CN101358869A (en) * | 2008-03-17 | 2009-02-04 | 西安艾贝尔科技发展有限公司 | Material level measuring device for steel ball coal mill and optimization and control method of powder-making process by bin type steel ball coal mill |
| CN101477367A (en) * | 2009-01-23 | 2009-07-08 | 广东工业大学 | Energy-saving consumption reduction oriented process efficiency optimization control method and use thereof |
| CN104571068A (en) * | 2015-01-30 | 2015-04-29 | 中国华电集团科学技术研究总院有限公司 | Optimized operation control method and system of distributed energy system |
| CN109709891A (en) * | 2018-12-28 | 2019-05-03 | 西安交通大学 | A kind of direct-drive high-speed feed system servo parameter Multipurpose Optimal Method |
-
2021
- 2021-06-18 CN CN202110680748.0A patent/CN113485260B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040075689A1 (en) * | 2002-10-22 | 2004-04-22 | Duncan Schleiss | Smart process modules and objects in process plants |
| CN101358869A (en) * | 2008-03-17 | 2009-02-04 | 西安艾贝尔科技发展有限公司 | Material level measuring device for steel ball coal mill and optimization and control method of powder-making process by bin type steel ball coal mill |
| CN101477367A (en) * | 2009-01-23 | 2009-07-08 | 广东工业大学 | Energy-saving consumption reduction oriented process efficiency optimization control method and use thereof |
| CN104571068A (en) * | 2015-01-30 | 2015-04-29 | 中国华电集团科学技术研究总院有限公司 | Optimized operation control method and system of distributed energy system |
| CN109709891A (en) * | 2018-12-28 | 2019-05-03 | 西安交通大学 | A kind of direct-drive high-speed feed system servo parameter Multipurpose Optimal Method |
Non-Patent Citations (4)
| Title |
|---|
| 柴天佑: "《复杂工业过程运行优化与反馈控制》", 《自动化学报》 * |
| 柴天佑: "《生产制造全流程优化控制对控制与优化理论方法的挑战》", 《自动化学报》 * |
| 王永刚等: "《氧化铝蒸发过程的混合智能优化设定控制》", 《化工学报》 * |
| 薛美盛等: "工业过程的先进控制", 《化工自动化及仪表》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113485260B (en) | 2022-10-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11487252B2 (en) | Process model identification in a process control system | |
| CN107632592B (en) | Nonlinear time-varying process fault monitoring method based on efficient recursion kernel principal component analysis | |
| CN104345726B (en) | Noninvasive data analysis in Process Control System | |
| Hongm et al. | Data driven fault diagnosis and fault tolerant control: some advances and possible new directions | |
| CN110032799A (en) | A kind of the angle similarity divided stages and monitoring method of microbiological pharmacy process | |
| CN115390459A (en) | Model prediction control method and device | |
| CN113485260B (en) | Operation optimization control system for complex industrial process | |
| CN111914886A (en) | Nonlinear chemical process monitoring method based on online brief kernel learning | |
| CN110687885A (en) | Fault diagnosis method and system for regulating valve of first-order constant value control system | |
| CN113467295B (en) | Automatic control method for aluminum oxide dissolution process | |
| CN117076260B (en) | Parameter and equipment abnormality detection method and device | |
| CN113408591A (en) | Process trend analysis and prediction method based on intelligent instrument | |
| Li et al. | Qualitative/quantitative simulation of process temporal behavior using clustered fuzzy digraphs | |
| Arzac-Garmendia et al. | Industrial Data Science for Batch Manufacturing Processes | |
| Ling et al. | A soft sensor based on influent mode discrimination neural network for a wastewater treatment process | |
| Wang et al. | On-line modeling and monitoring for multi-operation batch processes with infinite data types | |
| CN117422441A (en) | Electromechanical device detection and maintenance system | |
| Sánchez et al. | Design and implementation of a Real-Time control platform for the testing of advanced control systems and data quality management in the wastewater industry | |
| Tong et al. | A novel adaptive extreme learning machine-autoencoder algorithm for online monitoring and fault detection | |
| Luhui et al. | Fault prediction based on data-driven technique | |
| Rao | Fault Diagnosis of CNC Machine Tools Based on Neural Network Optimization | |
| CN117744476A (en) | Real-time dosage monitoring method and system for finished product slurry tank | |
| Madhavan et al. | Control of bioconversion processes | |
| CN114943175A (en) | Method and system for predicting health reference value of measuring point of power plant equipment | |
| CN117352762A (en) | Control system of battery manufacturing equipment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20211008 Assignee: Shanxi John Fenlei Design Engineering Co.,Ltd. Assignor: Northeastern University Contract record no.: X2024210000014 Denomination of invention: Optimization Control System for Complex Industrial Processes Granted publication date: 20221028 License type: Common License Record date: 20240308 |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20211008 Assignee: ZHEJIANG DAFENG TECHNOLOGY Co.,Ltd. Assignor: Northeastern University Contract record no.: X2024210000018 Denomination of invention: Optimization Control System for Complex Industrial Processes Granted publication date: 20221028 License type: Common License Record date: 20240321 |