A method for selecting a ladle in a steel plant based on rule-based reasoning

Abstract

The application provides a steel ladle matching method based on rule reasoning in a steel plant, and belongs to the field of metallurgy.The method comprises the following steps: constructing a rule base of reasoning rules based on actual steel ladle matching rules;obtaining a current steelmaking production schedule and an information table of all available steel ladles, and constructing a fact base based on known facts; a reasoning engine performs forward reasoning based on the formed rule base and fact base, generates new facts, and updates the fact base until no new facts are generated; and based on the final fact base, the optimal matching steel ladle based on the rules in the rule base is reasoned out. By using the application, the steel ladle matching problem in the steel plant can be solved by the method of rule reasoning, and the application has high flexibility and high accuracy.

Description

Technical Field

[0001] This invention relates to the field of metallurgy, and in particular to a method for selecting and matching steel ladles in steel plants based on rule-based reasoning. Background Technology

[0002] Ladle selection is an essential part of the steelmaking process. Determining the optimal ladle selection based on the current production plan and ladle information is crucial for achieving optimal energy consumption and lowest cost. Many factors need to be considered in ladle selection, such as whether the ladle meets hot repair time limits and residual element constraints. Relying on manual ladle selection may overlook some rules or facts, and is prone to subjectivity and inadequacy. Summary of the Invention

[0003] This invention provides a rule-based reasoning method for selecting steel ladles in steel plants, which can solve the problem of steel ladle selection in steel plants through rule-based reasoning, and is highly flexible and accurate.

[0004] The steel ladle selection method for steelmaking plants based on rule-based reasoning provided in this embodiment of the invention includes:

[0005] A rule base for inference rules is constructed based on actual ladle selection rules;

[0006] Obtain the current steelmaking production plan and all available steel ladle information, and construct a fact base based on known facts;

[0007] The inference engine performs forward reasoning based on the established rule base and fact base, generates new facts, and updates the fact base until no new facts are generated.

[0008] Based on the final fact base, the optimal steel ladle is deduced based on the rules in the rule base.

[0009] Furthermore, the rule base consists of several rules in the form of foreach…assert…

[0010] Furthermore, the rule base includes: filtering rules, sorting rules, and transmission rules; wherein,

[0011] Filtering rules: If the time interval between the offline time of ladle α in the ladle information table and the tapping time in the current production plan is less than T0 minutes, then ladle α will not be included in the subsequent ladle selection; if the time interval between the offline time of ladle β in the ladle information table and the tapping time in the selected production plan is greater than or equal to T0 minutes, then ladle β will be included in the subsequent ladle selection; where T0 is the hot repair and auxiliary transportation time;

[0012] Sorting rule: If ladle θ and ladle γ in the ladle information table both meet the filtering rule, and if the time interval Tθ from the offline time of ladle θ in the ladle information table to the tapping time in the selected production plan is less than the time interval Tγ from the offline time of ladle γ in the ladle information table to the tapping time in the selected production plan, that is: T0 < Tθ < Tγ, then θ has priority over γ in ladle selection;

[0013] Transmission rule: For ladle ζ, ladle θ and ladle γ in the ladle information table, if the time interval Tζ from the offline time of ladle ζ to the tapping time in the selected production plan is less than the time interval Tθ from the offline time of ladle θ in the ladle information table to the tapping time in the selected production plan, that is Tζ < Tθ; and at the same time, the time interval Tθ from the offline time of ladle θ in the ladle information table to the tapping time in the selected production plan is less than the time interval Tγ from the offline time of ladle γ in the ladle information table to the tapping time in the selected production plan, that is Tθ < Tγ, then there is Tζ < Tθ < Tγ, and it can be deduced that: Tζ < Tγ.

[0014] Furthermore, the production plan table includes: tapping time;

[0015] The ladle information table includes: ladle number and ladle offline time.

[0016] Furthermore, the fact library is composed of known facts and is used as evidence to test whether other statements are true; among them, each time ladle selection is performed, the fact library needs to be refreshed in real time.

[0017] Furthermore, the fact library includes: filtering facts, sorting facts and transmission facts;

[0018] Filtering facts: If the time interval Tα from the offline time of ladle α in the ladle information table to the tapping time in the selected production plan is less than T0 minutes, then the filtering fact of ladle α is generated and recorded as: less_than(α, T0), which means that ladle α does not meet the hot repair and auxiliary transportation time limit; if the time interval Tβ from the offline time of ladle β in the ladle information table to the tapping time in the selected production plan is greater than or equal to T0 minutes, then the filtering fact of ladle β is generated and recorded as: over_than(β, T0), which means that ladle β meets the hot repair and auxiliary transportation time limit; where T0 is the hot repair and auxiliary transportation time;

[0019] In order to distinguish the new filtered facts generated in subsequent reasoning from the initial known facts, it is stipulated that: if there is less_than(α,T0), then a new fact sort_less_than(T0,α) is inferred, which has the same meaning as the former; if there is over_than(β,T0), then a new fact sort_over_than(β,T0) is inferred, which has the same meaning as the former.

[0020] Sort Fact: If ladle θ and ladle γ in the ladle information table satisfy the following condition: the time interval Tθ between the time ladle θ is offline and the time slot Tγ in the selected production plan is less than the time interval Tγ between the time ladle γ is offline and the time slot Tγ in the selected production plan, then a sort fact for ladle θ and ladle γ is generated and recorded as: greater_than(γ,θ), which means: Tθ <Tγ;

[0021] In order to distinguish the new sorted facts generated by subsequent reasoning from the initial known facts, it is stipulated that: if there is greater_than(γ,θ) and sort_over_than(θ,T0), then a new fact sort_better(θ,γ) is inferred, which means that the θ ladle meets the hot repair and auxiliary transportation time T0 limit, and the θ ladle is superior to the γ ladle in the selection sequence.

[0022] Transmitting facts: If sort_better(θ,γ) and sort_better(γ,α) exist in the fact base, then infer the new fact sort_better(θ,α);

[0023] The transitive fact sort_better(θ,α) is formally identical to the newly generated sorted fact, and the transitive fact will be treated as a sorted fact in the subsequent diagram description.

[0024] Furthermore, the step of obtaining the current steelmaking production plan and all available steel ladle information, and constructing a fact base based on known facts, includes:

[0025] Obtain the current steelmaking production plan and all available steel ladle information tables, and dynamically construct filtered and sorted facts for all steel ladles in the ladle information table.

[0026] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0027] In this embodiment of the invention, a rule base for inference rules is constructed based on actual ladle selection rules; a fact base based on known facts is constructed by obtaining the current steelmaking production plan and all available ladle information; the inference engine performs forward inference based on the formed rule base and fact base to deduce the optimal ladle based on known facts and rules, thus enabling the steel plant to select ladles according to the production plan. In this way, the ladle selection problem in steel plants is solved through rule-based inference. Introducing a rule-based ladle selection method allows the inference process to be activated simply by applying a rule base containing different rules when the rules in the rule base change, thereby achieving the inference goal. This provides great flexibility and avoids the cumbersome modification of underlying code as requirements change. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 A flowchart illustrating the rule-based reasoning-based ladle selection method for steel plants provided in this embodiment of the invention. Figure 1 ;

[0030] Figure 2 A flowchart illustrating the rule-based reasoning-based ladle selection method for steel plants provided in this embodiment of the invention. Figure 2 . Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0032] like Figure 1 As shown, this embodiment of the invention provides a method for selecting and matching steel ladles in steel mills based on rule-based reasoning, including:

[0033] S101, a rule base for constructing inference rules based on actual ladle selection rules;

[0034] In this embodiment, in actual steelmaking production, the ladle selection and matching rule considering time factors is as follows: After the ladle is taken offline from the continuous caster, it enters the hot repair station and then receives molten steel under the converter as specified in the current production plan. The hot repair and auxiliary transportation time is recorded as T0 minutes (the hot repair time accounts for the main part). It is required that the time interval from the ladle offline time to the steelmaking time of the selected production plan shall not be less than T0 minutes. At the same time, when this interval is not less than T0 minutes, the smaller this interval is, the more forward the ladle corresponding to this interval should be in the ladle selection and matching sequence. Less temperature drop and more energy-saving effects can be obtained in the process.

[0035] In this embodiment, the rule base consists of several "foreach...assert...", that is, rules in the form of "For each fact, if... then assert...".

[0036] Such as Figure 2 shown, the rule base includes: filtering rules, sorting rules and transfer rules; among them,

[0037] Filtering rule: If the time interval from the offline time of ladle α in the ladle information table to the steelmaking time in the current production plan is less than T0 minutes, then ladle α does not enter the subsequent ladle selection and matching; if the time interval from the offline time of ladle β in the ladle information table to the steelmaking time in the selected production plan is greater than or equal to T0 minutes, then ladle β enters the subsequent ladle selection and matching; where T0 is the hot repair and auxiliary transportation time, generally composed of hot repair time, the lifting time from ladle offline to hot repair, and the lifting time from hot repair to the converter, and among them, the hot repair time accounts for a large proportion; specifically: determine the hot repair and auxiliary transportation time T0 according to the ladle transportation path;

[0038] Sorting rule: If both ladle θ and ladle γ in the ladle information table meet the filtering rules, and if the time interval Tθ from the offline time of ladle θ in the ladle information table to the steelmaking time in the selected production plan is less than the time interval Tγ from the offline time of ladle γ in the ladle information table to the steelmaking time in the selected production plan, that is: T0 < Tθ < Tγ, then θ has priority over γ in ladle selection and matching; this meets the process requirements of preferentially selecting a ladle with a better temperature state;

[0039] Transfer rule: For ladle ζ, ladle θ and ladle γ in the ladle information table, if the time interval Tζ from the offline time of ladle ζ to the steelmaking time in the selected production plan is less than the time interval Tθ from the offline time of ladle θ in the ladle information table to the steelmaking time in the selected production plan, that is Tζ < Tθ; and at the same time, the time interval Tθ from the offline time of ladle θ in the ladle information table to the steelmaking time in the selected production plan is less than the time interval Tγ from the offline time of ladle γ in the ladle information table to the steelmaking time in the selected production plan, that is Tθ < Tγ, then there is Tζ < Tθ < Tγ, and it can be deduced that: Tζ < Tγ.

[0040] S102, Obtain the current steelmaking production plan and all available steel ladle information, and construct a fact base based on known facts;

[0041] In this embodiment, the production plan includes: steel tapping time; the ladle information table includes: ladle number and ladle off-line time.

[0042] In this embodiment, the fact base consists of a series of known facts, which are used as evidence to verify whether other statements are true; wherein, the fact base needs to be refreshed in real time each time a ladle is selected.

[0043] In this embodiment, the fact base includes: filtering facts, sorting facts, and delivering facts;

[0044] Filtering Facts: If the time interval Tα between the offline time of ladle α in the ladle information table and the tapping time in the selected production plan is less than T0 minutes, a filtering fact for ladle α is generated and recorded as: less_than(α,T0), which means that ladle α does not meet the hot repair and auxiliary transportation time limits; if the time interval Tβ between the offline time of ladle β in the ladle information table and the tapping time in the selected production plan is greater than or equal to T0 minutes, a filtering fact for ladle β is generated and recorded as: over_than(β,T0), which means that ladle β meets the hot repair and auxiliary transportation time limits; where T0 is the hot repair and auxiliary transportation time.

[0045] In order to distinguish the new filtered facts generated in subsequent reasoning from the initial known facts, it is stipulated that: if there is less_than(α,T0), then a new fact sort_less_than(T0,α) is inferred, which has the same meaning as the former; if there is over_than(β,T0), then a new fact sort_over_than(β,T0) is inferred, which has the same meaning as the former.

[0046] Sort Fact: If ladle θ and ladle γ in the ladle information table satisfy the following condition: the time interval Tθ between the time ladle θ is offline and the time slot Tγ in the selected production plan is less than the time interval Tγ between the time ladle γ is offline and the time slot Tγ in the selected production plan, then a sort fact for ladle θ and ladle γ is generated and recorded as: greater_than(γ,θ), which means: Tθ <Tγ;

[0047] In order to distinguish the new sorted facts generated by subsequent reasoning from the initial known facts, it is stipulated that: if there is greater_than(γ,θ) and sort_over_than(θ,T0), then a new fact sort_better(θ,γ) is inferred, which means that the θ ladle meets the hot repair and auxiliary transportation time T0 limit, and the θ ladle is superior to the γ ladle in the selection sequence.

[0048] Transmitting facts: If sort_better(θ,γ) and sort_better(γ,α) exist in the fact base, then infer the new fact sort_better(θ,α);

[0049] The transitive fact sort_better(θ,α) is formally identical to the newly generated sorted fact, and the transitive fact will be treated as a sorted fact in the subsequent diagram description.

[0050] In this embodiment, the current steelmaking production plan and all available steel ladle information are obtained, and the filtered and sorted facts for all steel ladles in the ladle information table are dynamically constructed, such as... Figure 2 As shown.

[0051] S103, the reasoning engine performs forward reasoning based on the formed rule base and fact base, generates new facts, and updates the fact base until no new facts are generated;

[0052] like Figure 2 As shown, in this embodiment, forward reasoning is used for reasoning. Forward reasoning is a method that, based on known facts and rules, continuously asserts new facts based on known facts and rules, and then asserts all the facts that can be inferred based on the updated facts.

[0053] In this embodiment, rule-based reasoning is employed due to its powerful reasoning capabilities based on known facts and rules. It can assert all facts conforming to the rules based on known facts under rule constraints, exhibiting sufficiency, objectivity, and efficiency. Furthermore, different rules can be used for reasoning in different application scenarios to arrive at conclusions conforming to specific rules, demonstrating flexibility in its application.

[0054] In this embodiment, the use of rule-based reasoning technology has at least the following advantages:

[0055] 1) Rule-based reasoning offers high flexibility. In actual industrial production, different combinations of rules may be used based on different needs, and rule-based reasoning supports dynamic adjustment of rules without modifying the code.

[0056] 2) The rule-based reasoning technique has good sufficiency, as it can deduce all factual relationships that conform to the rules based on known rules and facts.

[0057] 3) Rule-based reasoning techniques are highly efficient, enabling them to complete reasoning tasks more effectively than humans within the same timeframe.

[0058] 4) The rule-based inference technique is completely objective. Since it is based entirely on rules and known facts, it does not have the potential subjective characteristics of manual selection.

[0059] S104, based on the final fact base, infers the optimal steel ladle based on the rules in the rule base.

[0060] The steel ladle selection method for steel plants based on rule-based reasoning described in this invention constructs a rule base for reasoning rules based on actual ladle selection rules; obtains the current steel production plan and a table of all available ladles, constructing a fact base based on known facts; the reasoning engine performs forward reasoning based on the formed rule base and fact base to deduce the optimal ladle based on known facts and rules, thus enabling steel plants to select ladles according to the production plan. In this way, the steel plant ladle selection problem is solved through rule-based reasoning. Introducing a rule-based ladle selection method allows the reasoning process to be activated simply by applying a rule base containing different rules when the rules in the rule base change, achieving the reasoning goal. This provides great flexibility and avoids the cumbersome modification of underlying code as requirements change.

[0061] To better understand the rule-based reasoning-based ladle selection method for steel plants described in this embodiment of the invention, a specific ladle selection scenario will be used for illustration. The workflow of the method is as follows:

[0062] Step 1: Rule Base Construction: Based on the actual ladle selection rules, construct the filtering rules, sorting rules, and transmission rules in the rule base. Once constructed, the rule base generally remains unchanged unless rules need to be dynamically added, modified, or deleted according to the production process. The rules constructed in this embodiment include: filtering rules `decide_over_than` and `decide_less_than`, a sorting rule `greater_than`, and a transmission rule `greater_than_transmit`. For a detailed description, please see Step 4.

[0063] Step 2: Obtain the current steelmaking production plan and all available ladle information. The steelmaking production plan must include tapping time information, and the ladle information must include ladle number and ladle completion time.

[0064] The current production plan is shown in Table 1:

[0065] Table 1 Current Production Plan

[0066]

[0067] The ladle information is shown in Table 2:

[0068] Table 2 Ladle Information Table

[0069]

[0070] Step 3: Construction of the fact library: After reading the current steelmaking production schedule and all available ladle information tables, generate the filtering facts and sorting facts for all ladles in the ladle information table. Each time ladle selection and matching are performed, the fact library needs to be refreshed in real time.

[0071] Step 4: The inference engine performs forward inference based on the formed rule library and fact library, generates new facts, and updates the fact library until no new facts are generated; among them, the inference engine used is the Pyke inference engine (python knowledge engine, a knowledge-based inference engine), and Pyke is an open-source inference engine based on Python. The overall knowledge formation process and inference process are implemented by combining python code with the Pyke inference engine.

[0072] For example:

[0073] Suppose in the above Table 1 and Table 2, Tp - T1 > T0, Tp - T2 > T0, Tp - T3 > T0, Tp - T4 < T0, and T1 > T2 > T3 > T4. The rule library includes filtering rules, sorting rules, and transitive rules.

[0074] Among them, the rules for forward inference are implemented through foreach…assert… in Pyke.

[0075] The filtering rule is:

[0076]

[0077] Here, in the above rule example, steel_ladle represents that the name of the fact library is steel_ladle.kfb, and $steel_ladle is a variable representing the time interval from the ladle offline time to the tapping time in the selected production plan.

[0078] For example, if the time interval (subsequently simplified as this time) $steel_ladle from the offline time of ladle α in the ladle information table to the tapping time in the selected production plan is less than $standard minutes, then ladle α does not enter the subsequent ladle selection and matching. If the time $steel_ladle of ladle β in the ladle information table is greater than or equal to $standard minutes, then ladle β enters the subsequent ladle selection and matching.

[0079] Here, decide_over_than and decide_less_than are rule names, over_than and less_than are known facts, sort_over_than and sort_less_than are newly inferred facts, and $steel_ladle and $standard are two variables, identifying the actual time and the standard time respectively. The same applies hereinafter!

[0080] The sorting rule is as follows:

[0081]

[0082] If $steel_ladle2 meets the constraint of the interval time $standard, and the interval time of $steel_ladle1 is greater than that of $steel_ladle2, i.e., the greater_than($steel_ladle1, $steel_ladle2) relationship holds, then there exists sort_better($steel_ladle2, $steel_ladle1), which means that the steel ladle $steel_ladle2 is before the steel ladle $steel_ladle1 in the matching selection sequence.

[0083] The transitive rule is as follows:

[0084]

[0085] If there exist both the facts sort_better($steel_ladle3, $steel_ladle2) and sort_better($steel_ladle2, $steel_ladle1) simultaneously, then it can be considered that sort_better($steel_ladle3, $steel_ladle1). That is, if the steel ladle $steel_ladle2 is superior to the steel ladle $steel_ladle1 in the matching selection and the steel ladle $steel_ladle3 is superior to the steel ladle $steel_ladle2 in the matching selection, then it must be considered that the steel ladle $steel_ladle3 is superior to the steel ladle $steel_ladle1 in the matching selection.

[0086] In this example, TS001 = Tp - T1, simply denoted as S001, and the same applies to the others. According to the assumptions that Tp - T1 > T0, Tp - T2 > T0, Tp - T3 > T0, Tp - T4 < T0, and T1 > T2 > T3 > T4, first convert the facts on the production site into facts in Pyke as follows:

[0087] over_than(S001,T0), over_than(S002,T0);

[0088] over_than(S003,T0), less_than(S004,T0);

[0089] greater_than(S004,S003), greater_than(S003,S002);

[0090] greater_than(S002,S001).

[0091] Next:

[0092] Based on the decide_over_than rule in the filtering rules, the following facts are generated:

[0093] pass:

[0094] over_than(S001,T0), over_than(S002,T0), over_than(S003,T0).

[0095] Generate new facts:

[0096] `sort_over_than(S001,T0)`, `sort_over_than(S002,T0)`, `sort_over_than(S003,T0)`. Based on the `decide_less_than` rule in the filtering rules, the following facts are generated:

[0097] pass:

[0098] less_than(S004,T0).

[0099] Generate new facts:

[0100] sort_less_than(T0,S004).

[0101] Based on the greater_than rule in the sorting rules, the following facts are generated:

[0102] pass:

[0103] greater_than(S004,S003), greater_than(S003,S002), greater_than(S002,S001). and:

[0104] sort_over_than(S001,T0), sort_over_than(S002,T0), sort_over_than(S003,T0); sort_less_than(T0,S004).

[0105] Generate new facts:

[0106] sort_better(S002,S003), sort_better(S001,S002).

[0107] Based on the greater_than_transmit rule in the transit rules, the following facts are generated:

[0108] pass:

[0109] sort_better(S002,S003), sort_better(S001,S002).

[0110] Generate new facts:

[0111] sort_better(S001,S003).

[0112] The final fact base is:

[0113] Known facts:

[0114] over_than(S001,T0), over_than(S002,T0);

[0115] over_than(S003,T0), less_than(S004,T0);

[0116] greater_than(S004,S003), greater_than(S003,S002);

[0117] greater_than(S002,S001)

[0118] Newly deduced facts:

[0119] sort_over_than(S001,T0), sort_over_than(S002,T0), sort_over_than(S003,T0); sort_less_than(T0,S004);

[0120] sort_better(S002,S003), sort_better(S001,S002);

[0121] sort_better(S001,S003).

[0122] Step 5: Based on the final fact base, deduce the optimal steel ladle selection based on the rules in the rule base. In this embodiment, based on known facts and new facts, by scanning the sort_better type facts, steel ladle S001 is determined to be the optimal steel ladle.

[0123] The above examples only demonstrate ladle selection in steelmaking plants considering time constraints. In actual production, ladle selection needs to comprehensively consider time constraints, temperature constraints, product type constraints, etc. Different steel mills will have very different applications of these constraints due to differences in production processes, requiring flexible rule settings. The rule-based reasoning method for ladle selection in steelmaking plants can support flexible dynamic rule adjustments without modifying the software code.

[0124] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for selecting and matching steel ladles in a steel plant based on rule-based reasoning, characterized in that, It includes: Construct a rule base for inference rules based on the actual ladle selection rules; Obtain the current steelmaking production schedule and all available ladle information tables, and construct a fact base based on known facts; The inference engine performs forward inference based on the formed rule base and fact base, generates new facts, and updates the fact base until no new facts are generated; Based on the final fact base, infer the most optimal ladle selection based on the rules in the rule base; Among them, the rule base includes: filtering rules, sorting rules, and transfer rules; among which, Filtering rule: If the time interval from the offline time of ladle α in the ladle information table to the steelmaking time in the current production plan is less than T0 minutes, then ladle α does not enter the subsequent ladle selection; if the time interval from the offline time of ladle β in the ladle information table to the steelmaking time in the selected production plan is greater than or equal to T0 minutes, then ladle β enters the subsequent ladle selection; where, T0 is the hot repair and auxiliary transportation time; Sorting rule: If ladle θ and ladle γ in the ladle information table both meet the filtering rules, and if the time interval Tθ from the offline time of ladle θ in the ladle information table to the steelmaking time in the selected production plan is less than the time interval Tγ from the offline time of ladle γ in the ladle information table to the steelmaking time in the selected production plan, that is: T0 < Tθ < Tγ, then θ takes precedence over γ in ladle selection; Transfer rule: For ladle ζ, ladle θ, and ladle γ in the ladle information table, if the time interval Tζ from the offline time of ladle ζ to the steelmaking time in the selected production plan is less than the time interval Tθ from the offline time of ladle θ in the ladle information table to the steelmaking time in the selected production plan, that is Tζ < Tθ; and at the same time, the time interval Tθ from the offline time of ladle θ in the ladle information table to the steelmaking time in the selected production plan is less than the time interval Tγ from the offline time of ladle γ in the ladle information table to the steelmaking time in the selected production plan, that is Tθ < Tγ, then there is Tζ < Tθ < Tγ, and it can be inferred that: Tζ < Tγ; Among them, the fact base includes: filtering facts, sorting facts, and transfer facts; Filtering fact: If the time interval Tα from the offline time of ladle α in the ladle information table to the steelmaking time in the selected production plan is less than T0 minutes, then a filtering fact of ladle α is generated and recorded as: less_than(α, T0), which means that ladle α does not meet the hot repair and auxiliary transportation time limit; if the time interval Tβ from the offline time of ladle β in the ladle information table to the steelmaking time in the selected production plan is greater than or equal to T0 minutes, then a filtering fact of ladle β is generated and recorded as: over_than(β, T0), which means that ladle β meets the hot repair and auxiliary transportation time limit; where, T0 is the hot repair and auxiliary transportation time; In order to distinguish the new filtered facts generated in subsequent reasoning from the initial known facts, it is stipulated that: if there is less_than(α,T0), then a new fact sort_less_than(T0,α) is inferred, which has the same meaning as the former; if there is over_than(β,T0), then a new fact sort_over_than(β,T0) is inferred, which has the same meaning as the former. Sort Fact: If ladle θ and ladle γ in the ladle information table satisfy the following condition: the time interval Tθ between the time ladle θ is offline and the time slot Tγ in the selected production plan is less than the time interval Tγ between the time ladle γ is offline and the time slot Tγ in the selected production plan, then a sort fact for ladle θ and ladle γ is generated and recorded as: greater_than(γ,θ), which means: Tθ <Tγ; In order to distinguish the new sorted facts generated by subsequent reasoning from the initial known facts, it is stipulated that: if there is greater_than(γ,θ) and sort_over_than(θ,T0), then a new fact sort_better(θ,γ) is inferred, which means that the θ ladle meets the hot repair and auxiliary transportation time T0 limit, and the θ ladle is superior to the γ ladle in the selection sequence. Transmitting facts: If sort_better(θ,γ) and sort_better(γ,α) exist in the fact base, then infer the new fact sort_better(θ,α); The transitive fact sort_better(θ,α) is formally identical to the newly generated sorted fact, and the transitive fact will be treated as a sorted fact in the subsequent diagram description.

2. The steel ladle selection method for steelmaking plants based on rule-based reasoning according to claim 1, characterized in that, The rule base consists of several rules in the form of foreach...assert...

3. The steel ladle selection method for steelmaking plants based on rule-based reasoning according to claim 1, characterized in that, The production schedule includes: steel tapping time; The ladle information table includes: ladle number and ladle production time.

4. The steel ladle selection method for steelmaking plants based on rule-based reasoning according to claim 1, characterized in that, The fact base consists of known facts and is used as evidence to verify whether other statements are true; the fact base needs to be updated in real time each time a ladle is selected.

5. The steel ladle selection method for steelmaking plants based on rule-based reasoning according to claim 1, characterized in that, The process of obtaining the current steelmaking production plan and all available steel ladle information, and constructing a fact base based on known facts, includes: Obtain the current steelmaking production plan and all available steel ladle information tables, and dynamically construct filtered and sorted facts for all steel ladles in the ladle information table.

Images