CN111626532A - Intelligent scheduling method for steelmaking production plan based on big data rule self-learning - Google Patents
Intelligent scheduling method for steelmaking production plan based on big data rule self-learning Download PDFInfo
- Publication number
- CN111626532A CN111626532A CN201910151534.7A CN201910151534A CN111626532A CN 111626532 A CN111626532 A CN 111626532A CN 201910151534 A CN201910151534 A CN 201910151534A CN 111626532 A CN111626532 A CN 111626532A
- Authority
- CN
- China
- Prior art keywords
- disturbance
- plan
- production
- rule
- time
- 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.)
- Pending
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 89
- 238000000034 method Methods 0.000 title claims abstract description 72
- 238000009628 steelmaking Methods 0.000 title claims abstract description 25
- 238000009749 continuous casting Methods 0.000 claims abstract description 37
- 230000008707 rearrangement Effects 0.000 claims abstract description 29
- 238000005065 mining Methods 0.000 claims abstract description 17
- 238000005516 engineering process Methods 0.000 claims abstract description 12
- 238000013135 deep learning Methods 0.000 claims abstract description 3
- 230000008569 process Effects 0.000 claims description 48
- 229910000831 Steel Inorganic materials 0.000 claims description 44
- 239000010959 steel Substances 0.000 claims description 44
- 238000005266 casting Methods 0.000 claims description 32
- 238000005457 optimization Methods 0.000 claims description 14
- 230000008859 change Effects 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 8
- 238000010586 diagram Methods 0.000 claims description 5
- 238000011156 evaluation Methods 0.000 claims description 5
- 238000005098 hot rolling Methods 0.000 claims description 5
- 238000012423 maintenance Methods 0.000 claims description 5
- 238000005096 rolling process Methods 0.000 claims description 5
- 230000007613 environmental effect Effects 0.000 claims description 4
- 238000010801 machine learning Methods 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 238000004092 self-diagnosis Methods 0.000 claims description 2
- 230000002159 abnormal effect Effects 0.000 claims 2
- 238000013439 planning Methods 0.000 abstract description 32
- 230000008901 benefit Effects 0.000 abstract description 4
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 16
- 230000003068 static effect Effects 0.000 description 10
- 229910052742 iron Inorganic materials 0.000 description 8
- 238000007670 refining Methods 0.000 description 8
- 239000002436 steel type Substances 0.000 description 7
- 238000012384 transportation and delivery Methods 0.000 description 5
- 230000002776 aggregation Effects 0.000 description 4
- 238000004220 aggregation Methods 0.000 description 4
- 230000037430 deletion Effects 0.000 description 4
- 238000012217 deletion Methods 0.000 description 4
- 230000009467 reduction Effects 0.000 description 4
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 238000003723 Smelting Methods 0.000 description 2
- 230000003139 buffering effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000003780 insertion Methods 0.000 description 2
- 230000037431 insertion Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 239000002893 slag Substances 0.000 description 2
- 239000000243 solution Substances 0.000 description 2
- 229910001208 Crucible steel Inorganic materials 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000009529 body temperature measurement Methods 0.000 description 1
- 235000009508 confectionery Nutrition 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000006477 desulfuration reaction Methods 0.000 description 1
- 230000023556 desulfurization Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000033764 rhythmic process Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- 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
- Y02P80/00—Climate change mitigation technologies for sector-wide applications
- Y02P80/10—Efficient use of energy, e.g. using compressed air or pressurized fluid as energy carrier
-
- 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
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Manufacturing & Machinery (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- General Factory Administration (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a steel-making production plan intelligent scheduling method based on big data rule self-learning, which mainly comprises the following steps: 1) the decision table is used for aligning time, scenes and scheduling operation according to the historical scheduling records to form a scheduling decision table; 2) rule mining, namely generating a rule base from historical records by adopting rule mining algorithms such as a rough set and deep learning; 3) planning, namely planning a continuous casting machine based on actual data, scheduling a bottom layer device based on a container tracking technology, and planning a production operation based on a time parallel backward-pushing algorithm; 4) disturbance rearrangement, namely intelligent rearrangement is realized based on a rule description mode of 'disturbance + scene description + adjustment strategy' and an expert rule base; 5) and (4) plan display, namely displaying the operation plan through a Gantt chart and providing interface dynamic adjustment. The invention ensures the efficient and orderly operation of the steelmaking-continuous casting production process, reduces the production and operation cost, saves energy, reduces consumption, improves the production efficiency and realizes the maximization of enterprise benefit.
Description
Technical Field
The invention belongs to the technical field of intelligent scheduling of a steel-making production process, and particularly relates to a rearrangement method of a steel-making production operation plan under disturbance based on big data rule self-learning.
Background
In the steelmaking production process, molten iron, molten steel and scrap steel flow among all processes through a container, the molten iron enters a steelmaking workshop from the molten iron to a molten steel injection continuous casting machine, the whole process needs to pass through a plurality of processes, each process is provided with a plurality of devices, and the establishment of an operation plan based on a steelmaking process topological graph is the basis for ensuring the normal operation of the steelmaking process. However, the field process is complex, the processes are multiple, the disturbance is frequent, and the well-planned operation plan taking the heat as a unit needs to be adjusted frequently, so that reasonable planning and rearrangement under the disturbance are important guarantees of compact and orderly operation in the steelmaking process, and have important meanings for saving energy and reducing consumption in the steelmaking process, saving production and operation cost of enterprises, and improving production benefits and enterprise competitiveness. At present, the management technology and informatization level of most steel-making-continuous casting production processes in China obviously lags behind the related industry level of developed economy, and an intelligent rearrangement method and an intelligent rearrangement system under practical disturbance are not formed.
Disclosure of Invention
In order to solve the problem that the original plan cannot be smoothly executed due to disturbance in the steel-making and continuous casting production process, the invention aims to provide the intelligent scheduling method which can reduce the burden of scheduling personnel, improve the production efficiency and quickly and effectively realize the dynamic rearrangement of the steel-making and continuous casting production operation plan under different disturbances. FIG. 1 is an intelligent scheduling framework overall framework.
In order to realize the purpose of intelligent scheduling, the technical scheme of the invention mainly comprises the following contents: including production planning and production static job planning, execution equipment planning based on production job planning, and planning rearrangement under disturbance in the production process. FIG. 2 is a flow chart of planning; the method mainly comprises the following steps:
1) and (3) production planning: the production plan is based on the production order in the ERP system of the company and the hot-charging production plan of hot-rolling delivery given by the hot-rolling workshop, the steel type and the steel number of the planned factory production and the planned starting time and the planned delivery time
2) Static planning and arrangement: the static job plan includes: a continuous caster plan, a tundish plan, and a production operation plan. The planning is based on a time parallel backward-pushing algorithm and a heuristic intelligent optimization algorithm of an inverse process path, and determines a production path of a smelting steel type, transportation time among stations of a procedure, processing time, time of a heat in the casting time, finishing time, heat processing equipment and the like on the premise of ensuring continuous casting of a continuous casting machine. Fig. 3 is a flow chart of static planning, which is implemented in the following manner:
a. and planning the continuous casting machine, wherein the planning and arrangement of the continuous casting machine are formed by order preprocessing, order grouping, furnace grouping and furnace grouping, casting time combination and casting time sequencing according to the production order information of the ERP system.
b. And the tundish plan comprises a cast steel grade, the number of furnaces, a tundish casting sequence, a tundish internal furnace sequence, the casting starting time of each tundish on a corresponding continuous casting machine and the casting starting time of each furnace.
c. The production operation plan is to arrange the specific start and end time of all furnaces at a certain station of each process of steel making and continuous casting, and the arrangement basis is as follows: continuous casting operation plan, steel grade process path, field layout topological graph, transportation time matrix and other field industrial rules.
3) And executing an equipment plan, wherein the equipment plan comprises a container plan and a crown block plan, and because the production operation plan is compiled based on the current situation of the production equipment, the smooth implementation of the production plan needs to be ensured, containers such as an iron ladle, a steel ladle, a scrap steel hopper and the like are required to be in place in time besides the normal operation of the production equipment, and the coordination of the crown block trolley carrying of the transport container.
a. The container planning and arrangement comprises a ladle plan, a ladle plan and a scrap steel bucket plan, and the arrangement of the positioning plan of various containers based on the operation plan is extremely important for ordered and compact production of steel making.
b, planning and arranging an overhead traveling crane, wherein the container position is required to be carried and executed by the overhead traveling crane and a trolley, so that an overhead traveling crane operation plan shown in figure 4 is required to be planned according to the container plan, and the overhead traveling crane plan mainly specifies the starting position and the stopping position of each different overhead traveling crane and the time; the overhead traveling crane plan includes: the method comprises the following steps of overhead traveling crane task, overhead traveling crane temporary task, overhead traveling crane path avoidance, overhead traveling crane execution decision and bottom equipment state judgment. The temporary task of the crown block is a temporary task which is not predicted and sporadic and is executed by the crown block, because the temporary task cannot be programmed into a crown block plan; the decision-making executed by the overhead travelling crane is that the overhead travelling crane task has strict conditions, such as strict condition limits on lifting, releasing, running and the like, and an overhead travelling crane task execution decision table of whether the overhead travelling crane task can be executed or not, and the task conditions are not specific and the like needs to be established based on the field working conditions;
4) plan rearrangement, if the interference such as order disturbance, equipment disturbance, maintenance plan disturbance, large time delay and the like occurs on the site, the well-programmed operation plan needs to be readjusted according to the site condition; for basic disturbance rearrangement, rearrangement can be performed in a parallel backstepping plus heuristic intelligent calculation mode; for more complicated disturbance rearrangement such as equipment failure, maintenance schedule change, molten steel remelting and the like, a corresponding rule base needs to be established, and the schedule rearrangement is realized in a rule matching mode.
a, establishing a decision table: in order to establish a rule base conveniently and realize rule matching and rule mining, a standard rule description mode needs to be established and consists of three parts of disturbance, environmental conditions and operation sequences.
b, mining industrial big data rules: the manual input rule is simple and convenient, but the manual rule has certain limitation, all knowledge cannot be acquired, and the knowledge cannot be iteratively optimized. For this purpose, an expert unknown rule intelligent mining platform based on historical operation records is proposed, and the framework of the intelligent platform is shown in fig. 6. The specific implementation scheme of the industrial big data rule mining intelligent framework comprises the following steps: 1) a decision record table, 2) data reduction, 3) data preprocessing, 4) data mining based on a rough set and an LSTM deep network, and 5) rule optimization evaluation index and manual adjustment operation interface.
c, rearrangement rule matching: and recording the field working condition and the equipment state, matching rules from the expert good rule base according to the current working condition, and finding out the optimal rule capable of solving the current disturbance for adjustment decision under the fluctuation of the working condition.
The technical scheme adopted by the invention has the following beneficial effects:
1. the invention adopts a time parallel reverse-pushing algorithm of an inverse process path and a heuristic intelligent optimization algorithm to compile an operation plan based on the actual working condition and the process constraint condition of a steelmaking field, and provides a reference basis for ordered production for the steelmaking process;
2. according to the invention, a rough set is adopted to dig expert rearrangement experience and perform online optimization, real-time adjustment can be performed on the operation plan according to the actual field, ordered and compact operation in the steelmaking process is realized, and the defect that the operation plan is continuously adjusted manually according to the field condition is overcome;
3. the method adopted by the invention can maximally accelerate the production rhythm while ensuring the orderly progress of the plan, give positive help and constraint to production operators, save the production cost, reduce the energy consumption and improve the production efficiency of steel enterprises.
Drawings
FIG. 1 is a steel-making-continuous casting intelligent scheduling system framework;
FIG. 2 illustrates a steel-making-continuous casting planning process;
FIG. 3 is a static planning orchestration framework diagram;
FIG. 4 is a Gantt chart of a day car plan;
FIG. 5 a rough set of rules mining algorithm interfaces;
FIG. 6 is an industrial big data rule intelligent mining platform framework;
FIG. 7 is a job plan after a device failure has been rearranged;
FIG. 8 is a time-delayed disturbance pull-down speed adjustment interface;
Detailed Description
Fig. 1 is a framework diagram of the technical solution of the present invention, and the technical solution of the present invention mainly includes five contents: 1) a production plan; 2) static operation planning; 3) performing equipment arrangement; 4) planning interference dynamic rearrangement; 5) gantt chart shows. The following further describes the embodiments of the present invention with reference to the drawings.
1) Acquiring production information, wherein the production information comprises production order information and a slab rolling plan; the production order information comprises the specification requirements of the plate blanks such as order number, length, height, thickness, density, weight and the like, and the steel grade and the steel number which are planned to be produced; a scheduled start time and a scheduled delivery time; the slab rolling plan is generated according to a hot-delivery hot-charging production plan given by a hot rolling workshop.
2) A static job planning, the static job plan comprising: a continuous casting machine, a tundish plan and a production operation plan. Fig. 3 is a frame diagram in the static operation planning process, on the premise of ensuring continuous casting of a continuous casting machine, a production path (production process and sequence to be performed) is determined according to smelting steel types, then according to transportation time between stations of adjacent processes and processing time of different steel types in each process, a parallel back-pushing algorithm is adopted to determine the time when a furnace in each casting process enters each process and the finishing time of each process, the available time of each process is considered, a heuristic intelligent optimization algorithm is adopted to designate processing equipment of each process for each furnace, an operation plan is compiled by using a time parallel back-pushing algorithm of an inverse process path and a heuristic intelligent optimization algorithm, on the premise of ensuring continuous casting of the continuous casting machine, the time when the furnace in each casting process enters each process and the finishing time of each process are determined, and the specific planning comprises the following steps:
3) the planning arrangement of the continuous casting machine is a tundish plan matched with the continuous casting machine formed by performing order preprocessing, order grouping, furnace grouping and casting order combination according to the production order information of the ERP system. The tundish is compiled by production information, and then the scheduled tundish plan needs to be adjusted on line according to the collected production actual performance data of each procedure, and the specific implementation mode is as follows:
a) preprocessing an order, matching the information of steel grade, specification and yield of a customer order with the current stock, and calculating the quantity of slabs to be smelted; secondly, the customer order is decomposed into a plurality of small orders according to the product information, and each small order corresponds to one steel type and one slab specification.
b) The orders are grouped, a plurality of small orders of a plurality of customer orders which are decomposed in the order preprocessing stage are grouped, the orders with the same or similar steel type and plate blank specification are grouped in the same group, the furnace grouping and the group casting in the production process are facilitated, and therefore the equipment utilization rate and the yield are improved. An aggregation is a way of grouping orders, orders with similar attributes are grouped together, AND before the aggregation is used in the scheduling stage of the system, a plurality of rules can be used to define an aggregation, AND only if it satisfies all the rules (AND logic) in the rule set, the orders will fall into the aggregation.
c) The furnace grouping stage can decompose a larger order (more than the amount of molten iron smelted by the converter once) in the grouping into a plurality of times, and simultaneously combine a plurality of small orders (less than the amount of molten iron smelted by the converter once) and the last non-whole furnace of the larger order into a whole furnace according to the same or similar rules of steel types and billet specifications.
d) The pouring times are combined, the furnace number corresponding to each order is subjected to group pouring according to the current production condition of the continuous casting machine, and the most reasonable pouring times can be formed according to the combination of the yield requirement, steel grade requirement and slab specification requirement of the order and the equipment capacity of a steel mill; the production capacity of the tundish of the continuous casting machine in use can be fully utilized, the heat plan is added to the tundish which does not fully utilize the performance, all the heat can be guaranteed to be compiled in a certain casting time, the residual heat does not exist, and the casting time is associated with the continuous casting machine.
e) The module can automatically carry out optimization calculation and solve the sequencing problem of each casting time or tundish according to the difference between the use condition of a continuous casting machine and the delivery date of an order under the support of a casting time sequencing rule base so as to form an optimized and feasible tundish plan.
4) The production operation plan is to arrange the specific start and end time of all furnaces at a certain station of each process of steel making and continuous casting, and the arrangement basis is as follows: continuous casting operation plan, steel grade process path, field layout topological graph, transportation time matrix and other field industrial rules.
5) The container plan, wherein the execution equipment plan comprises a container plan and a crown block plan; the container plan is compiled based on the current situation of the production equipment, so that the smooth implementation of the production plan is ensured, and an iron ladle, a steel ladle and a scrap steel hopper are required to be in place in time besides the requirement on the normal operation of the production equipment. Therefore, arranging the positioning plan of various containers based on the operation plan is extremely important for ordered and compact production of steel-making production; the specific implementation of the container comprises: ladle plans, and scrap box plans.
a) And (3) a ladle plan, based on container tracking of a logistics system, selecting a ladle from online circulating ladles according to a ladle number according to the ladle plan aiming at a specific heat plan, planning the time of arriving an empty ladle and leaving a heavy ladle from a molten iron receiving level, the time of arriving a heavy ladle and leaving a desulfurization station and a slagging-off station, the time of arriving a heavy ladle at a temperature measurement sampling position, the time of arriving a heavy ladle at an inlet position of a converter, the time of leaving an empty ladle at the inlet position of the converter, and online recording and offline deletion of the ladles.
b) And (3) a ladle plan, based on container tracking of a logistics system, selecting ladles from hot-repaired ladles according to ladle numbers by the ladle plan aiming at a specific heat plan, calculating the time when an empty ladle reaches a converter steel receiving position, the time when the molten steel leaves the steel receiving position after receiving, the time when a full ladle reaches and leaves a refining station, the time when the ladle reaches a rotating platform of a continuous casting machine to receive a ladle position after refining, and the time when the empty ladle leaves the rotating platform to receive the ladle position after casting. If the molten steel ladle has a pouring allowance, the temporary task of pouring the residual molten steel, pouring the slag, the time and the hot repair position when the molten steel ladle reaches the steel receiving position of the converter by inserting the molten steel ladle is required.
c) And (3) planning a scrap steel bucket, wherein the scrap steel bucket is selected according to the bucket number aiming at a specific heat based on container tracking of a logistics system, the time of the scrap steel bucket reaching a scrap steel allocation position is planned, a formula of the scrap steel bucket of each heat is given, the time of the scrap steel bucket reaching a scrap steel hanging position through a cross car is planned, the time of the scrap steel bucket reaching a converter scrap steel loading position is planned, the time of an empty scrap steel bucket returning to the cross car is planned, and the time of a slag cross returning through the cross car is planned.
6) Planning and arranging an overhead traveling crane, wherein the planning of the overhead traveling crane shown in fig. 4 refers to that the container position is moved and needs to be carried and executed by the overhead traveling crane and a trolley, so that an operation plan of the overhead traveling crane needs to be arranged according to the container plan, and the plan of the overhead traveling crane mainly specifies the starting position and the stopping position of each overhead traveling crane and the time; the specific implementation plan comprises the following steps:
a) the overhead traveling crane operation task and temporary task, by compiling the overhead traveling crane operation plan according to the container plan, mainly stipulate and each stride different overhead traveling crane to start, stop position and time, except carrying out the task in the plan based on operation plan and container plan, still need carry out the occasional temporary task, because need not disturb the original plan as much as possible, can set for the priority to the temporary task, arrange the tabulation in the low priority task list, wait for the idle time, to urgent high priority task, need to enter the system of the temporary task, give and dodge tactics and new overhead traveling crane execution plan according to the route planning of the temporary task by the intelligent scheduling system.
b) The decision-making executed by the overhead travelling crane needs to give an overhead travelling crane task execution decision table because the overhead travelling crane executes tasks under strict conditions, such as hoisting, releasing, running and the like, which have strict condition limits, and the recommendation information whether the overhead travelling crane tasks can be executed or not is given according to the on-site working condition, so that the on-site timely processing can be reminded by an interface when the task execution time is up and the conditions are not specific
c) The path of the crown block is avoided, and when the crown block is inserted into an unplanned temporary task and the crown block task is rearranged under field disturbance, the path conflict between different crown blocks in the same span area can be caused. Therefore, the overhead traveling crane avoidance algorithm is designed to distribute the priority of the execution task according to the process rules of air avoidance light, light avoidance heavy and heavy avoidance urgent, and finally the operation scheme of avoiding the overhead traveling crane and the final execution scheme of avoiding the task are obtained.
d) Judging the state of the bottom equipment, wherein the formulated operation plan, container plan and crown block plan have certain difference with the actual execution condition of the plan, so that the actual execution time of a signal judgment task of the bottom equipment needs to be acquired, and a difference database is stored to provide data support for the optimization and improvement of the plan;
7) the method comprises the steps of obtaining an expert rule base, wherein a large amount of manpower and material resources are consumed for manual collection of rescheduling rules, so that a standard rule description mode and a rescheduling operation record storage mode need to be established, then, machine learning technologies such as a rough set and a deep learning network are adopted to mine rescheduling rules under various interferences according to manual operation and adjustment experience, a standard rescheduling rule base is formed after optimization selection, a rescheduling algorithm selects proper rescheduling rules according to field working conditions to achieve planning rescheduling, a rescheduling rule input interface is designed, and field scheduling personnel can select the rescheduling rules under various interferences through the interface. The specific implementation method comprises the following steps:
a) the rule description comprises three parts of a disturbance event, an environmental condition and an operation sequence to form a rule, and the most appropriate rearrangement rule can be matched according to the disturbance type, the field working condition and the operation plan execution condition when the disturbance occurs. Table 1 shows the rule description
Table 1 rule description
b) And manual interface operation, which adjusts the manual operation interface and ensures the accuracy and the practicability of the called rule.
c) Although the industrial big data rule mining is simple and convenient for manual entry of the rule, the manual rule has certain limitation, all knowledge cannot be acquired, and the knowledge cannot be iteratively optimized. Therefore, an expert unknown rule intelligent mining platform based on historical operation records is provided, and an intelligent platform framework is shown in fig. 6 and mainly comprises the following research contents:
c (1) industrial production data arrangement, extracting field working conditions and equipment states which have influence on the decision and decision operation under the state, forming a decision record table in the form of working condition combination-operation sequence, establishing a real-time data table of the field working conditions and the equipment states, and matching an optimal rule according to the real-time working conditions.
c (2) data reduction, namely removing redundant records in the historical operation records, removing samples with low differentiation degree, and removing part of state attributes of the samples which do not provide information support for decision, thereby carrying out attribute reduction on the data.
c (3) data preprocessing, wherein the data preprocessing is to carry out denoising, filtering, outlier removing and incomplete information repairing processing on illegal data by using a data reconstruction technology; due to the fact that the field external environment is changeable and the subjective factors of operators are added, the information of the recorded production data has uncertain characteristics such as noise, fuzziness, incomplete knowledge, uncertainty and even contradiction; and the severe field environment such as high temperature, dust and the like makes some key parameters undetectable or partial information unknown. Uncertainty and incompleteness of information (collectively, incomplete information) can overwhelm the inherent regularity embodied in the data. Therefore, data preprocessing technologies aiming at incomplete information, such as filtering and denoising, are researched, a complete production data warehouse is established, and a complete data source is provided for rule discovery.
c (4) a machine learning algorithm, which researches a static process rule intelligent discovery technology based on a rough set algorithm and a support vector machine, and a rule mining algorithm interface based on a rough set is shown in FIG. 5; the intelligent discovery technology of the dynamic adjustment rule based on the combination of the similarity subsequence search, the rough set and the fuzzy matching technology is researched, and the algorithm focuses on the discovery technology of the dynamic adjustment rule of incremental data because the data speed is accelerated very fast, so that the rule mining efficiency is improved.
And c (5) establishing excellent rule evaluation indexes, designing manual regulation and interface addition, deletion, modification and check of the rules, providing a modification interface for manual regulation decision of the rules, and recording the regulation result of the specific rules manually. And then, based on a rule optimization model of the evaluation index, taking a manual adjustment result as a supervised learning guide teacher, and optimally adjusting the original rule into an optimal rule.
c (6) matching rules, wherein the matching is to find out the optimal rule capable of solving the current disturbance according to the matching rules in the good rule base under the current working condition, and the optimal rule is used for adjustment decision under the fluctuation of the working condition.
8) Plan rearrangement under disturbance, when order disturbance, equipment disturbance, maintenance plan disturbance, large time delay and other disturbances occur on site, the programmed operation plan needs to be readjusted according to the site conditions, and plan rearrangement is realized by adopting two modes of an expert rule base and heuristic intelligent algorithm optimization, wherein the specific implementation method comprises the following steps:
a) based on disturbance self-diagnosis and plan rearrangement of working conditions, when disturbance which causes that a planned plan cannot be executed occurs on site, site personnel are generally required to page and tell a scheduling room, and then the scheduling room personnel adjust Gantt charts to realize plan rearrangement according to experience. Under the condition of establishing a rearrangement rule base, the intelligent scheduling system can realize plan rearrangement according to the field working condition and the plan execution condition as long as field disturbance is input.
b) Plan rearrangement under equipment failure, fig. 7 is a sweet chart of an operation plan after rescheduling when the LF3 refining furnace fails and stops, and it can be seen that when the LF3 fails, 1-2,1-4,1-6 arranged above the LF3 are respectively transferred to the LF1 and the LF2 to continue production, and when the 1-2,1-4,1-6 are distributed to other refining furnaces, the operation starting time is obviously lagged behind that of the original furnace, and two treatment methods are provided for ensuring continuous casting of a continuous casting machine: 1. if the delay time caused by the redistribution of 1-2,1-4,1-6 can be realized by reducing the pulling speed, the pulling speed of CC3 is reduced by 2, if the pulling speed can not realize the rearrangement, the pouring time of the tundish containing 1-2,1-4,1-6 needs to be pushed back, and for the heat (1-1) which enters the production flow, the tundish stays in the refining LF4 for a period of time to ensure the sequential production until the delay time is matched with the scheduled starting time of the tundish.
c) When the casting time plan and the change operations of the furnace time plan such as insertion, deletion, exchange sequence and the like in the casting time plan occur, the production operation plan compiled based on the casting time plan needs to be rearranged, the change operations of the casting time and the furnace time such as insertion, deletion, exchange sequence and the like can be implemented by a mouse, and the production operation plan can be automatically rearranged according to the adjusted casting time and furnace time after the operation is completed.
d) Plan adjustment under delay disturbance can automatically adjust the time buffer of the refining furnace to eliminate short time delay caused by general production fluctuation through a system; for larger fault time delay, if the refining buffering can not eliminate the time delay influence, the refining buffering and the proper adjustment of the pulling speed can be adopted to eliminate the time delay disturbance. The drawing speed of a continuous casting machine generally requires that the drawing speed is faster and better on the premise of continuous casting and no drawing leakage, but the delay disturbance can be eliminated by adjusting the drawing speed under the condition of not influencing the productivity. The Gantt chart after the casting speed is adjusted is shown in fig. 8, wherein the casting time on the CC501 is longest, the casting time on the CC503 is shortest, the casting time on the CC502 is in the middle of the two casting times, and the casting time can be adjusted through the casting speed.
e) Station matching, wherein the process path passed by each heat is determined by steel grade, but the station selection relationship between adjacent process paths has two modes: 1. and (2) a cross selection relation, namely the heat finished in the previous process can enter any equipment in the next process for production, specifically selecting which equipment needs to be optimized according to the current situation to determine a one-to-one laminar flow relation, namely a specified corresponding production relation exists between a certain station in the previous process and a certain station in the next process, and the production equipment cannot be changed.
9) Gantt chart shows: in the whole steelmaking continuous casting process, the generated and dynamically adjusted production and production equipment plan is displayed to a dispatcher in a Gantt chart form, and the Gantt chart can be manually adjusted by dragging the Gantt chart through a mouse.
Compared with the prior art, the method has the advantages of good arrangement effect, strong environment change adapting capability, good intelligent rearrangement rapidity group, compact and efficient production in the steelmaking process, production and operation cost reduction, energy conservation, enterprise management enhancement, production efficiency improvement and contribution to promoting the maximization of enterprise benefits.
The foregoing shows and describes the general principles and features of the present invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (5)
1. Rescheduling decision table: based on historical operation records and adjustment experience, time and scenes are aligned again, a scheduling decision table between a field state and disturbance is generated, a basis is provided for rule mining and machine learning, and the method comprises the following specific implementation steps:
(1) recording a production order and a slab rolling plan of a hot rolling plant issued by an ERP system of the steel plant;
(2) recording the specific starting time and the specific ending time of all the heats at the stations of each process of steel making and continuous casting;
(3) tracking and recording ladle scheduling, ladle scheduling and scrap steel bucket scheduling by a container circulation tracking technology based on crown block and trolley positioning technologies;
(4) when various disturbances such as cast-in-place times change, equipment failure, production delay and the like are recorded on site, the original plan cannot be smoothly executed, the process and the result of readjustment of the current production state by manpower or machines are recorded, the records are aligned with the disturbance time, the whole production process scene before disturbance and the manual experience and the result of processing the disturbance, a complete form is formed, and a complete data basis is provided for rule mining.
2. And (3) rule mining and obtaining: firstly, establishing a standard rule description mode based on 'disturbance + scene description + adjustment strategy', converting expert experience into a knowledge rule which can be recognized by a machine, and mining scheduling and rescheduling rules under various interferences by adopting machine learning technologies such as a rough set and a deep learning network according to a rescheduling decision table, wherein the method mainly comprises the following steps:
(1) rule description, mainly including "perturbation", "environmental condition" and "operation sequence"; the disturbance classification comprises order change disturbance, fault disturbance, maintenance plan disturbance, time disturbance, temperature disturbance, steel change disturbance, casting abnormal disturbance and the like, and the description of the environmental conditions comprises the following steps: the equipment state, the logistics information, the processing information and the heat task state information can be matched with the most appropriate operation sequence according to the disturbance type and the scene description when disturbance occurs;
(2) providing an adjustment interface of a field expert for the selected rule under the current working condition, and generating a manual adjustment record;
(3) reducing attributes, namely reducing important influence attributes by adopting a rough set algorithm;
(4) learning expert operation rules from the reduced operation records by adopting a rule mining algorithm;
(5) and establishing a rule optimization evaluation index, establishing a rule optimization model based on the evaluation index, and adjusting records according to the rules of the manual interface to optimize the rules to form a good rule base.
3. The method mainly comprises the following steps of statically arranging an operation plan, establishing a tundish plan based on a production order, a slab rolling plan and actual performance data of each process, establishing a steelmaking production operation plan based on process rules such as a field device topological graph, a process path and the like by adopting a parallel backward-pushing algorithm according to a casting schedule, and mainly comprising the following steps of:
(1) receiving a production order issued by an ERP system of a steel plant and a slab rolling plan which needs to be hot-fed and hot-loaded and is transmitted by a hot rolling plant, and meanwhile, dividing the production order into five parts according to the collected production actual performance data of each process: generating a tundish plan by order preprocessing, order grouping, furnace grouping, casting time combination and casting time sequencing
(2) Constructing a steelmaking-continuous casting process topological diagram according to process parameters such as production processes, station relations, process equipment numbers and the like;
(3) acquiring field process rules such as steel grade process production paths, production time of steel grades in different equipment, transportation time matrixes among different stations of each procedure, the number of devices without faults in each procedure and the like;
(4) calculating the starting time and the ending time of each heat task in each process by adopting an inverse process path direction time parallel backward-pushing algorithm based on a steelmaking-continuous casting process topological diagram;
(5) and determining the specific production station of each heat in each process based on the equipment optimization selection rule, and if time conflicts occur among the heats, adjusting the starting time of the pouring times of the conflicting heats for recalculation.
4. The method comprises the following steps that an operation plan under disturbance is automatically rearranged, when the disturbance such as order change disturbance, fault disturbance, maintenance plan disturbance, time disturbance, temperature disturbance, steel change disturbance, casting abnormal disturbance and the like occurs, and the disturbance needs to be automatically readjusted according to the field condition of the programmed operation plan, a rule base and heuristic intelligent algorithm optimizing method is adopted to readjust the programmed operation plan according to the field condition, based on a formed expert rule base and a decision table, when the disturbance which causes the programmed plan to be incapable of being executed occurs on the field, the intelligent scheduling system can realize plan rearrangement according to the field working condition and the plan execution condition only by recording the field disturbance, so that the unmanned rearrangement of the plan is realized, and the method mainly comprises the following steps:
(1) disturbance self-diagnosis of working conditions is based on a steelmaking Internet of things system, detailed field states can be obtained in real time, various disturbances can be automatically diagnosed in real time based on a decision table, and unmanned rearrangement of plans is achieved;
(2) and rearranging the original plan according to the current state and the working condition according to different disturbances.
5. And (3) displaying an operation plan: and displaying the adjusted plan to a dispatcher in a Gantt chart form, wherein the Gantt chart can be manually adjusted by dragging the Gantt chart with a mouse.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910151534.7A CN111626532A (en) | 2019-02-28 | 2019-02-28 | Intelligent scheduling method for steelmaking production plan based on big data rule self-learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910151534.7A CN111626532A (en) | 2019-02-28 | 2019-02-28 | Intelligent scheduling method for steelmaking production plan based on big data rule self-learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111626532A true CN111626532A (en) | 2020-09-04 |
Family
ID=72271655
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910151534.7A Pending CN111626532A (en) | 2019-02-28 | 2019-02-28 | Intelligent scheduling method for steelmaking production plan based on big data rule self-learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111626532A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112712246A (en) * | 2020-12-25 | 2021-04-27 | 包头钢铁（集团）有限责任公司 | Method for centralized batch production of falling plate blanks |
| CN112712289A (en) * | 2021-01-18 | 2021-04-27 | 上海交通大学 | Adaptive method, system, and medium based on temporal information entropy |
| TWI745256B (en) * | 2021-04-12 | 2021-11-01 | 南亞塑膠工業股份有限公司 | Operation management system |
| CN114386719A (en) * | 2022-03-22 | 2022-04-22 | 宁波钢铁有限公司 | Method and device for optimizing heat batch plan and storage medium |
| CN114662802A (en) * | 2022-05-23 | 2022-06-24 | 宁波钢铁有限公司 | Hot-rolled strip scheduling method and system based on disturbance factors |
| CN114781934A (en) * | 2022-06-17 | 2022-07-22 | 希望知舟技术(深圳)有限公司 | Work order distribution method and related device |
| CN114844783A (en) * | 2021-01-14 | 2022-08-02 | 新智云数据服务有限公司 | Agent starting deployment system based on cloud platform computing decision |
| CN114908210A (en) * | 2021-02-10 | 2022-08-16 | 上海梅山钢铁股份有限公司 | Automatic ladle matching method for converter ladle |
| CN115511292A (en) * | 2022-09-27 | 2022-12-23 | 北京虎蜥信息技术有限公司 | Production scheduling method, system, intelligent terminal and storage medium |
| CN117541042A (en) * | 2024-01-10 | 2024-02-09 | 山东钢铁股份有限公司 | Scheduling method and scheduling system for realizing matching of steelmaking furnace |
| CN117952567A (en) * | 2024-03-25 | 2024-04-30 | 四川多联实业有限公司 | Production management method and system based on MES intelligent manufacturing |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105243512A (en) * | 2015-11-06 | 2016-01-13 | 湖南千盟物联信息技术有限公司 | Dynamic scheduling method of steelmaking operation plan |
| CN105353733A (en) * | 2015-11-09 | 2016-02-24 | 湖南千盟物联信息技术有限公司 | Steel-making production process intelligent scheduling method |
| CN108985537A (en) * | 2018-03-20 | 2018-12-11 | 湖南师范大学 | A kind of steel smelting-continuous casting production plan rearrangement method based on rough set rule digging |
-
2019
- 2019-02-28 CN CN201910151534.7A patent/CN111626532A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105243512A (en) * | 2015-11-06 | 2016-01-13 | 湖南千盟物联信息技术有限公司 | Dynamic scheduling method of steelmaking operation plan |
| CN105353733A (en) * | 2015-11-09 | 2016-02-24 | 湖南千盟物联信息技术有限公司 | Steel-making production process intelligent scheduling method |
| CN108985537A (en) * | 2018-03-20 | 2018-12-11 | 湖南师范大学 | A kind of steel smelting-continuous casting production plan rearrangement method based on rough set rule digging |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112712246B (en) * | 2020-12-25 | 2023-02-21 | 包头钢铁（集团）有限责任公司 | Method for centralized batch production of falling plate blanks |
| CN112712246A (en) * | 2020-12-25 | 2021-04-27 | 包头钢铁（集团）有限责任公司 | Method for centralized batch production of falling plate blanks |
| CN114844783A (en) * | 2021-01-14 | 2022-08-02 | 新智云数据服务有限公司 | Agent starting deployment system based on cloud platform computing decision |
| CN114844783B (en) * | 2021-01-14 | 2024-04-19 | 新智云数据服务有限公司 | Agent starting deployment system based on cloud platform calculation decision |
| CN112712289A (en) * | 2021-01-18 | 2021-04-27 | 上海交通大学 | Adaptive method, system, and medium based on temporal information entropy |
| CN112712289B (en) * | 2021-01-18 | 2022-11-22 | 上海交通大学 | Adaptive method, system, and medium based on temporal information entropy |
| CN114908210A (en) * | 2021-02-10 | 2022-08-16 | 上海梅山钢铁股份有限公司 | Automatic ladle matching method for converter ladle |
| TWI745256B (en) * | 2021-04-12 | 2021-11-01 | 南亞塑膠工業股份有限公司 | Operation management system |
| CN114386719B (en) * | 2022-03-22 | 2022-08-05 | 宁波钢铁有限公司 | Method and device for optimizing heat batch plan and storage medium |
| CN114386719A (en) * | 2022-03-22 | 2022-04-22 | 宁波钢铁有限公司 | Method and device for optimizing heat batch plan and storage medium |
| CN114662802A (en) * | 2022-05-23 | 2022-06-24 | 宁波钢铁有限公司 | Hot-rolled strip scheduling method and system based on disturbance factors |
| CN114781934A (en) * | 2022-06-17 | 2022-07-22 | 希望知舟技术(深圳)有限公司 | Work order distribution method and related device |
| CN114781934B (en) * | 2022-06-17 | 2022-09-20 | 希望知舟技术(深圳)有限公司 | Work order distribution method and related device |
| CN115511292A (en) * | 2022-09-27 | 2022-12-23 | 北京虎蜥信息技术有限公司 | Production scheduling method, system, intelligent terminal and storage medium |
| CN115511292B (en) * | 2022-09-27 | 2023-05-30 | 北京虎蜥信息技术有限公司 | Production scheduling method, system, intelligent terminal and storage medium |
| CN117541042A (en) * | 2024-01-10 | 2024-02-09 | 山东钢铁股份有限公司 | Scheduling method and scheduling system for realizing matching of steelmaking furnace |
| CN117952567A (en) * | 2024-03-25 | 2024-04-30 | 四川多联实业有限公司 | Production management method and system based on MES intelligent manufacturing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111626532A (en) | Intelligent scheduling method for steelmaking production plan based on big data rule self-learning | |
| Tang et al. | Integrated charge batching and casting width selection at Baosteel | |
| CN106779220B (en) | Steelmaking-continuous casting-hot rolling integrated scheduling method and system | |
| Lee et al. | Primary production scheduling at steelmaking industries | |
| Atighehchian et al. | A novel hybrid algorithm for scheduling steel-making continuous casting production | |
| Cui et al. | An improved Lagrangian relaxation approach to scheduling steelmaking-continuous casting process | |
| Hao et al. | A soft-decision based two-layered scheduling approach for uncertain steelmaking-continuous casting process | |
| CN105550751A (en) | Steelmaking-continuous casting scheduling method utilizing priority policy hybrid genetic algorithm | |
| CN105243512A (en) | Dynamic scheduling method of steelmaking operation plan | |
| CN111242414B (en) | Planning and scheduling system applied to steelmaking-continuous casting process in steel industry | |
| US10032128B2 (en) | Yard management apparatus, yard management method, and computer program | |
| CN106647674A (en) | Knowledge representation-based steel production scheduling model matching method | |
| CN105303320A (en) | Intelligent scheduling algorithm for steelmaking | |
| Liu et al. | Ladle intelligent re-scheduling method in steelmaking–refining–continuous casting production process based on BP neural network working condition estimation | |
| CN116909236A (en) | Hot metal ladle logistics simulation and intelligent scheduling method and system based on iron-steel interface | |
| CN105427028A (en) | Intelligent steelmaking bottom layer transportation device scheduling method | |
| Milewska | IT systems supporting the management of production capacity | |
| CN108985537A (en) | A kind of steel smelting-continuous casting production plan rearrangement method based on rough set rule digging | |
| CN111598464A (en) | Steelmaking scheduling method based on process modeling | |
| Tamura et al. | Synchronized scheduling method in manufacturing steel sheets | |
| CN114153185B (en) | Steelmaking-continuous casting elastic scheduling optimization method and system based on dynamic multi-target differential evolution algorithm | |
| Ito et al. | Production planning and scheduling technology for steel manufacturing process | |
| Sun et al. | Crane scheduling of steel-making and continuous casting process using the mixed-timed petri net modelling via CPLEX optimization | |
| Zeng et al. | Intelligent optimization method for the dynamic scheduling of hot metal ladles of one-ladle technology on ironmaking and steelmaking interface in steel plants | |
| de Souza et al. | Models for scheduling charges in continuous casting: application to a Brazilian steel plant |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200904 |