Scheduling strategy determination method and device, storage medium and electronic equipment

Abstract

The invention discloses a scheduling strategy determination method and device, a storage medium and electronic equipment. The method comprises the following steps: acquiring an initialized scheduling strategy and an auxiliary scheduling strategy; performing genetic reproduction on the initialized scheduling strategy to obtain a first intermediate scheduling strategy; extracting first scheduling strategy knowledge based on the auxiliary scheduling strategy, and updating the first intermediate scheduling strategy based on the first scheduling strategy knowledge to obtain a first scheduling strategy; when the iteration stopping condition is not met, the auxiliary scheduling strategy is updated, the iteration process of the scheduling strategy is executed based on the first scheduling strategy and the updated auxiliary scheduling strategy, a second scheduling strategy is obtained, and when the iteration stopping condition is met, a target scheduling strategy is obtained. By extracting the first scheduling strategy knowledge in the auxiliary scheduling strategy, the first intermediate scheduling strategy obtained by multiplying the initialized scheduling strategy is subjected to diversity updating, the local optimum problem is avoided, the globally optimal target scheduling strategy is obtained, and the accuracy of the target scheduling strategy is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent logistics technology, and in particular to a scheduling strategy determination method, device, storage medium and electronic equipment. Background Art

[0002] Dynamic scheduling problems exist in many fields, such as the scheduling of transport vehicles in the logistics industry, and the scheduling of personnel in the power maintenance industry. Dynamic scheduling problems generally include active methods and passive methods. Active methods can plan the scheduling methods under various scheduling environments in advance, and passive methods can create scheduling strategies to solve problems in real time, which can flexibly respond to the uncertainty in the scheduling process.

[0003] In the process of implementing the present invention, it is found that there are at least the following technical problems in the prior art: Genetic programming, as a passive method for solving dynamic scheduling problems, can provide scheduling strategies for uncertain scheduling environments. However, during the search process, genetic programming is prone to losing the diversity of the population, resulting in falling into a local optimum, which indirectly causes the final scheduling strategy to be non-globally optimal. Summary of the invention

[0004] The present invention provides a scheduling strategy determination method, device, storage medium and electronic device to improve the diversity of scheduling strategies for dynamic scheduling problems in the genetic variation process and solve the problem of local optimality.

[0005] According to one aspect of the present invention, a scheduling strategy determination method is provided, comprising:

[0006] Acquire an initialization scheduling strategy and an auxiliary scheduling strategy, wherein the number of the initialization scheduling strategy and the number of the auxiliary scheduling strategies are multiple, and the initialization scheduling strategy and the auxiliary scheduling strategy do not overlap;

[0007] Performing genetic propagation on the initialization scheduling strategy to obtain a first intermediate scheduling strategy;

[0008] Extracting first scheduling strategy knowledge based on the auxiliary scheduling strategy, and updating the first intermediate scheduling strategy based on the first scheduling strategy knowledge to obtain a first scheduling strategy;

[0009] When the iteration stop condition is not met, the auxiliary scheduling strategy is updated, and the scheduling strategy iteration process is performed based on the first scheduling strategy and the updated auxiliary scheduling strategy to obtain a second scheduling strategy, until the iteration stop condition is met to obtain a target scheduling strategy.

[0010] Optionally, obtaining the initialization scheduling strategy and the auxiliary scheduling strategy includes: obtaining a set of historical scheduling strategies corresponding to historical scheduling requirements respectively; sampling the set of historical scheduling strategies to obtain a first number of initialization scheduling strategies and a second number of auxiliary scheduling strategies.

[0011] Optionally, the genetically propagating the initialization scheduling strategy to obtain the first intermediate scheduling strategy includes: extracting the second scheduling strategy knowledge based on the initialization scheduling strategy; screening the fitness data corresponding to the target scheduling requirements based on the initialization scheduling strategy and the second scheduling strategy knowledge respectively to obtain the scheduling strategy to be propagated; and genetically propagating the scheduling strategy to be propagated to obtain the first intermediate scheduling strategy.

[0012] Optionally, the auxiliary scheduling strategy or the initialization scheduling strategy is in the form of a decision tree;

[0013] The method for extracting the first scheduling strategy knowledge or the second scheduling strategy knowledge includes: splitting the auxiliary scheduling strategy or the initialization scheduling strategy to obtain a candidate knowledge set, wherein the candidate knowledge set includes candidate knowledge, and the candidate knowledge includes one or more of a scheduling strategy decision tree, a scheduling strategy subtree and a common subtree; respectively determining the fitness data corresponding to the target scheduling requirement of the candidate knowledge in the candidate knowledge set, and determining the first scheduling strategy knowledge or the second scheduling strategy knowledge based on the fitness data corresponding to the candidate knowledge relative to the target scheduling requirement.

[0014] Optionally, the method for determining the fitness data includes: determining the fitness data corresponding to the target scheduling requirement of the individual scheduling strategy based on a preset fitness function, wherein the individual scheduling strategy includes one or more of an initialization scheduling strategy, an auxiliary scheduling strategy, candidate knowledge, and scheduling strategy knowledge;

[0015] Alternatively, the fitness data of the individual scheduling strategy corresponding to the target scheduling demand is determined based on a predetermined proxy model, wherein the proxy model is determined based on the corresponding relationship between the individual scheduling strategy and the fitness data.

[0016] Optionally, the updating of the first intermediate scheduling strategy based on the first scheduling strategy knowledge to obtain the first scheduling strategy includes one or more of the following: cross-mutating the first scheduling strategy knowledge with the first intermediate scheduling strategy; performing genetic reproduction based on the first scheduling strategy knowledge and the auxiliary scheduling strategy to obtain the second intermediate scheduling strategy; and cross-mutating based on the first intermediate scheduling strategy and the second intermediate scheduling strategy.

[0017] Optionally, the iteration stop condition includes one or more of the following: the number of iterations meets a preset total number of iterations; the gain of the fitness data for a preset number of iterations is less than a preset gain threshold.

[0018] According to another aspect of the present invention, there is provided a scheduling strategy determination device, comprising:

[0019] An initialization module, used to obtain an initialization scheduling strategy and an auxiliary scheduling strategy, wherein the number of the initialization scheduling strategy and the number of the auxiliary scheduling strategies are multiple, and the initialization scheduling strategy and the auxiliary scheduling strategy do not overlap;

[0020] The iterative module is used to genetically propagate the initialization scheduling strategy to obtain a first intermediate scheduling strategy; extract the first scheduling strategy knowledge based on the auxiliary scheduling strategy, and update the first intermediate scheduling strategy based on the first scheduling strategy knowledge to obtain a first scheduling strategy; when the iteration stop condition is not met, update the auxiliary scheduling strategy, and execute the iterative process of the scheduling strategy based on the first scheduling strategy and the updated auxiliary scheduling strategy to obtain a second scheduling strategy, until the iteration stop condition is met to obtain a target scheduling strategy.

[0021] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0022] at least one processor; and

[0023] a memory communicatively connected to the at least one processor; wherein,

[0024] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the scheduling strategy determination method described in any embodiment of the present invention.

[0025] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the scheduling strategy determination method described in any embodiment of the present invention when executed.

[0026] The technical solution of the embodiment of the present invention sets an auxiliary scheduling strategy on the basis of heritage multiplication based on the initialization scheduling strategy, and performs diversity update on the first intermediate scheduling strategy obtained by multiplying the initialization scheduling strategy by extracting the first scheduling strategy knowledge in the auxiliary scheduling strategy, so as to avoid repeated scheduling strategies in the process of heritage multiplication based on the initialization scheduling strategy, thereby causing local optimal problems, obtaining the globally optimal target scheduling strategy, and improving the accuracy of the target scheduling strategy.

[0027] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0029] Figure 1 is a flow chart of a scheduling strategy determination method provided by an embodiment of the present invention;

[0030] Figure 2 is a flow chart of a scheduling strategy determination method provided by an embodiment of the present invention;

[0031] Figure 3 is a flow chart of a scheduling strategy determination method provided by an embodiment of the present invention;

[0032] Figure 4 It is a structural diagram of a scheduling strategy determination device provided by an embodiment of the present invention;

[0033] Figure 5 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0035] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0036] Figure 1 It is a flowchart of a scheduling strategy determination method provided by an embodiment of the present invention. This embodiment is applicable to the case where the scheduling strategy is determined by genetic programming. The method can be executed by a scheduling strategy determination device. The scheduling strategy determination device can be implemented in the form of hardware and / or software. The scheduling strategy determination device can be configured in mobile terminals such as mobile phones, tablet computers, etc., electronic devices such as computers or servers.

[0037] The dynamic scheduling scenario in this embodiment can be a vehicle scheduling scenario in the logistics business, where different vehicle models result in different vehicle carrying capacities. Different logistics tasks correspond to different task information, and the task information includes but is not limited to information such as the starting point, destination, weight, volume, and cargo type. For multiple logistics tasks, the vehicles corresponding to the logistics tasks are determined respectively. Since logistics tasks can be dynamically changing, for example, logistics tasks can include planned tasks and temporary tasks, the corresponding transport vehicles are called for different types of logistics tasks by determining the target scheduling strategy, so as to realize dynamic scheduling of vehicles. In the above-mentioned vehicle scheduling scenario, different scheduling strategies can be determined according to different scheduling requirements. For example, scheduling requirements can include but are not limited to minimum transportation path and minimum transportation cost.

[0038] In this embodiment, the dynamic scheduling scenario can also be, for example, a personnel scheduling scenario in the power maintenance industry. Specifically, when an abnormality in the power supply equipment is detected or a repair information is received, an electric power maintenance form is generated and transmitted to the task scheduling device, which can be a server or a mobile terminal, etc. The task scheduling device pre-stores maintenance personnel information, where the maintenance personnel information includes but is not limited to the idle status and location of the maintenance personnel, and the electric power maintenance form includes but is not limited to the abnormality type and repair location information. By determining the target scheduling strategy, the corresponding maintenance personnel are determined respectively to achieve personnel scheduling. In the above-mentioned personnel scheduling scenario, different scheduling strategies can be determined according to different scheduling requirements. For example, the scheduling requirements can include but are not limited to the minimum processing time and the minimum scheduling cost.

[0039] It is understandable that the dynamic scheduling scenario is not limited to the vehicle scheduling scenario and the personnel scheduling scenario. In other embodiments, the target scheduling strategy can also be determined for other dynamic scheduling scenarios.

[0040] like Figure 1 As shown, the method includes:

[0041] S110, obtaining an initialization scheduling strategy and an auxiliary scheduling strategy, wherein the number of the initialization scheduling strategy and the number of the auxiliary scheduling strategies are plural respectively, and the initialization scheduling strategy and the auxiliary scheduling strategy do not overlap.

[0042] S120, performing genetic propagation on the initialization scheduling strategy to obtain a first intermediate scheduling strategy;

[0043] S130. Extract first scheduling strategy knowledge based on the auxiliary scheduling strategy, and update the first intermediate scheduling strategy based on the first scheduling strategy knowledge to obtain a first scheduling strategy.

[0044] S140, determining whether the iteration stop condition is met, if not, updating the auxiliary scheduling strategy, returning to step S120 based on the first scheduling strategy and the updated auxiliary scheduling strategy, and outputting the target scheduling strategy if the iteration stop condition is met.

[0045] The initialization scheduling strategy can be understood as the initial scheduling strategy for the genetic programming process of the scheduling strategy, and the auxiliary scheduling strategy can be understood as a scheduling strategy that provides diversity assistance to the genetic programming process of the scheduling strategy. The scheduling strategy here can be understood as a rule for determining the scheduling information corresponding to the task information. The first number of initialization scheduling strategies is at least one, and the second number of auxiliary scheduling strategies is at least one, wherein the first number and the second number may be the same or different. In some embodiments, the first number is greater than the second number, and the first number is a preset multiple of the second number. The number of initialization scheduling strategies and auxiliary scheduling strategies is not limited here, and can be determined according to the requirements of the genetic coding process of the scheduling strategy.

[0046] There are no overlapping scheduling strategies between the initialization scheduling strategy and the auxiliary scheduling strategy. By setting the auxiliary scheduling strategy, the diversity of the population in the genetic encoding process can be maintained. Based on the auxiliary scheduling strategy, the genetic encoding process of the initialization scheduling strategy is provided with diverse survey strategy knowledge, which avoids a large number of repeated individuals in the genetic encoding process of the initialization scheduling strategy, leading to local optimal situations.

[0047] The initialization scheduling strategy and the auxiliary scheduling strategy can be in the form of decision trees respectively. By inputting the task information into the above-mentioned initialization scheduling strategy and auxiliary scheduling strategy, the scheduling information corresponding to the task information can be determined. Taking the vehicle scheduling scenario as an example, the task information can be logistics task information, and the scheduling information can be vehicle information (for example, it can be vehicle identification, etc.).

[0048] Optionally, the initialization scheduling strategy and the auxiliary scheduling strategy can be constructed randomly separately.

[0049] Optionally, the initialization scheduling strategy and the auxiliary scheduling strategy can be set based on the decision tree construction model, and the decision tree construction model can be a machine learning model, which is obtained by training the decision tree creation demand and the constructed historical decision tree. Among them, the decision tree creation demand can include description information of the dynamic scheduling scenario and the target scheduling demand. Taking the vehicle scheduling scenario as an example, the description information of the target scheduling demand can be the minimum scheduling cost. By converting the decision tree creation demand into an embedding vector, the embedding vector is used as the input of the decision tree construction model, and multiple optional scheduling strategies output by the decision tree construction model are obtained, and the first part of the optional scheduling strategy is used as the initialization scheduling strategy, and the second part of the optional scheduling strategy is used as the auxiliary scheduling strategy. Specifically, the optional scheduling strategy can be sampled based on the first quantity and the second quantity, respectively, to obtain the initialization scheduling strategy and the auxiliary scheduling strategy. The decision tree construction model is used to create an initialization scheduling strategy and an auxiliary scheduling strategy that match the target scheduling demand in the dynamic scheduling scenario, which provides a good foundation for the genetic coding process of the scheduling strategy, facilitates the acceleration of the genetic coding process, and improves the efficiency of genetic coding.

[0050] Optionally, the initialization scheduling strategy and the auxiliary scheduling strategy may be historical scheduling strategies. Taking the vehicle scheduling scenario as an example, multiple historical scheduling strategies corresponding to the vehicle scheduling scenario are obtained, and sampling is performed in the multiple historical scheduling strategies respectively to determine the initialization scheduling strategy and the auxiliary scheduling strategy. It is understandable that the historical scheduling strategy may be determined based on other scheduling requirements in the dynamic scheduling scenario. Taking the vehicle scheduling scenario as an example, the target scheduling requirement corresponding to the genetic encoding of the current scheduling strategy may be the minimum scheduling cost, and the historical scheduling requirement corresponding to the historical scheduling strategy may be, for example, the shortest scheduling path, etc. The historical scheduling requirements corresponding to the historical scheduling strategies may be the same or different, and each historical scheduling requirement may correspond to multiple historical scheduling strategies.

[0051] Obtaining the initialization scheduling strategy and the auxiliary scheduling strategy includes: obtaining a set of historical scheduling strategies corresponding to the historical scheduling requirements respectively; sampling in the set of historical scheduling strategies to obtain a first number of initialization scheduling strategies and a second number of auxiliary scheduling strategies. It can be understood that the above sampling process is a sampling process without replacement to avoid overlapping of the initialization scheduling strategy and the auxiliary scheduling strategy. In this embodiment, by taking the historical scheduling strategy corresponding to the historical scheduling requirements as the basis, the initialization scheduling strategy and the auxiliary scheduling strategy of the target scheduling requirements are used to provide high-quality initialization scheduling strategies and auxiliary scheduling strategies for the genetic coding process of the scheduling strategy, simplify the genetic coding process of the scheduling strategy, and accelerate the convergence process of the genetic coding of the scheduling strategy.

[0052] In this embodiment, the first number of initialized scheduling strategies is used as the main group, and genetic reproduction is performed through the main group to obtain the first intermediate scheduling strategy. Exemplarily, the first number of initialized scheduling strategies are genetically reproduced based on at least one genetic operator corresponding to the genetic encoding. For example, the at least one genetic operator corresponding to the genetic encoding may include but is not limited to one or more of a selection operator, a crossover operator, and a mutation operator, wherein the selection operator includes but is not limited to one or more of roulette selection, random traversal sampling, sorting selection, fitness scaling selection, and tournament selection. Crossover operators include but are not limited to one or more of single-point crossover, K-point crossover, uniform crossover, and ordered crossover. Mutation operators include but are not limited to one or more of bit flip mutation, exchange mutation, inversion mutation, and reversal mutation.

[0053] For the second number of auxiliary scheduling strategies, the first scheduling strategy knowledge of the auxiliary scheduling strategies is extracted, wherein the scheduling strategy knowledge can be understood as at least part of the scheduling strategy that is beneficial to the genetic encoding process of the scheduling strategy, and the scheduling strategy knowledge can be the scheduling strategy or part of the scheduling strategy. Taking the scheduling strategy as a decision tree as an example, the scheduling strategy knowledge can be the overall decision tree or a decision tree subtree.

[0054] Optionally, the scheduling strategy knowledge may be at least part of the scheduling strategy that satisfies the fitness condition. The first scheduling strategy knowledge is to extract at least part of the scheduling strategy that is beneficial to the genetic encoding process of the scheduling strategy from the auxiliary scheduling strategy. The first scheduling strategy knowledge may be a scheduling strategy decision tree (i.e., any auxiliary scheduling strategy), a scheduling strategy subtree (i.e., a part of the auxiliary scheduling strategy), and a common subtree of the auxiliary scheduling strategy.

[0055] The first scheduling strategy knowledge can be determined in a manner as follows: splitting the auxiliary scheduling strategy to obtain a candidate knowledge set, wherein the candidate knowledge set includes candidate knowledge, and the candidate knowledge includes one or more of a scheduling strategy decision tree (i.e., an auxiliary scheduling strategy), a scheduling strategy subtree (i.e., a subtree of the auxiliary scheduling strategy), and a common subtree (i.e., a common subtree of the auxiliary scheduling strategy); respectively determining the fitness data corresponding to the target scheduling requirement of the candidate knowledge in the candidate knowledge set, and determining the first scheduling strategy knowledge based on the fitness data corresponding to the candidate knowledge relative to the target scheduling requirement.

[0056] The auxiliary scheduling strategy may be split into random subtrees and / or common subtrees, where the common subtree is a subtree commonly included in at least two auxiliary scheduling strategies. The auxiliary scheduling strategy is added to the candidate knowledge set, and the subtree obtained by splitting the auxiliary scheduling strategy is added to the candidate knowledge set.

[0057] Respectively determine the fitness data corresponding to the target scheduling requirement for each candidate knowledge in the candidate knowledge set, and screen the candidate knowledge based on the fitness data corresponding to the target scheduling requirement for each candidate knowledge, and determine the screened candidate knowledge as the first scheduling strategy knowledge. Exemplarily, the candidate knowledge may be sorted based on the fitness data, and a third number of candidate knowledge may be selected as the first scheduling strategy knowledge based on the sorting. Exemplarily, the fitness data corresponding to the candidate knowledge may be compared based on the fitness threshold, and the candidate knowledge whose architecture fitness data is greater than the fitness threshold may be selected as the first scheduling strategy knowledge.

[0058] Optionally, the fitness data corresponding to each candidate knowledge relative to the target scheduling requirement may be determined based on a preset fitness function and the candidate knowledge. Specifically, the candidate knowledge is introduced into the fitness function to obtain the fitness data corresponding to the candidate knowledge.

[0059] Optionally, the fitness data corresponding to the candidate knowledge relative to the target scheduling demand can also be obtained by processing the candidate knowledge through a proxy model. The proxy model is determined based on the corresponding relationship between the scheduling strategy individual and the fitness data, and the scheduling strategy individual here can be an overall decision tree or a decision tree subtree. For example, the proxy model can be a machine learning model, which uses the scheduling strategy individual as input and obtains the fitness data corresponding to the scheduling strategy individual as a label training, wherein the fitness data corresponding to the scheduling strategy individual can be determined based on a pre-set fitness function. The scheduling strategy individual here can be a scheduling strategy individual in the historical number of iterations during the genetic coding process and / or a scheduling strategy individual in the current number of iterations. In the first round of iterations, a proxy model can be constructed based on the corresponding relationship between the initialization scheduling strategy (at least part of the multiple initialization scheduling strategies) and the fitness data of the initialization scheduling strategy. In the subsequent iterations, the proxy model is updated based on the scheduling strategy in each iteration number and the fitness data of the scheduling strategy to improve the processing accuracy of the proxy model. By setting a proxy model, the calculation process of the fitness data can be simplified and the calculation cost in the genetic coding process can be reduced.

[0060] When the first scheduling strategy knowledge and the first intermediate scheduling strategy are determined, the first intermediate scheduling strategy is updated based on the first scheduling strategy knowledge to obtain the first scheduling strategy. Specifically, the first intermediate scheduling strategy is cross-mutated based on the first scheduling strategy knowledge to obtain the first scheduling strategy.

[0061] Optionally, the first scheduling strategy may be generated by cross-mutating the first scheduling strategy knowledge with the first intermediate scheduling strategy. For example, any first scheduling strategy knowledge may be substituted for a subtree in any first intermediate scheduling strategy to obtain the first scheduling strategy; for example, any first scheduling strategy knowledge may be added to the first intermediate scheduling strategy as a subtree of any first intermediate scheduling strategy to obtain the first scheduling strategy.

[0062] Optionally, the first scheduling strategy may be generated by: performing genetic reproduction based on the first scheduling strategy knowledge and the auxiliary scheduling strategy to obtain the second intermediate scheduling strategy; performing crossover mutation based on the first intermediate scheduling strategy and the second intermediate scheduling strategy. The auxiliary scheduling strategy may be genetically reproduced based on a preset genetic operator, which is not limited here. For example, the first intermediate scheduling strategy and the second intermediate scheduling strategy perform a subtree exchange, that is, the subtree in the first intermediate scheduling strategy replaces the subtree in the second intermediate scheduling strategy to obtain the first scheduling strategy; for example, the subtree in the first intermediate scheduling strategy may be used as a subtree in the second intermediate scheduling strategy, added to the second intermediate scheduling strategy, to obtain the first scheduling strategy; for example, the first intermediate scheduling strategy replaces the subtree in the second intermediate scheduling strategy to obtain the first scheduling strategy; for example, the first intermediate scheduling strategy may be used as a subtree in the second intermediate scheduling strategy, added to the second intermediate scheduling strategy, to obtain the first scheduling strategy.

[0063] Based on the above updating method, the second intermediate scheduling strategy is updated by using the first scheduling strategy knowledge extracted from the auxiliary scheduling strategy to obtain the first scheduling strategy, so as to improve the diversity of the first scheduling strategy and avoid the local optimal situation.

[0064] It can be understood that the genetic encoding process of the scheduling strategy includes multiple iterations. After obtaining the first scheduling strategy, it is determined whether the iteration stop condition is met. If the iteration stop condition is met, the target scheduling strategy is determined based on the first scheduling strategy, and the target scheduling strategy is output. If the iteration stop condition is not met, the auxiliary scheduling strategy is updated, and the next iteration process is performed based on the updated auxiliary scheduling strategy and the first scheduling strategy to obtain the second scheduling strategy. After obtaining the second scheduling strategy, if the iteration stop condition is still not met, the auxiliary scheduling strategy is updated, and the next iteration process is continued based on the updated auxiliary scheduling strategy and the second scheduling strategy, and so on, until the iteration stop condition is met and the target scheduling strategy is output.

[0065] The updating of the auxiliary scheduling strategy may be resampling the historical scheduling strategy to determine a new auxiliary scheduling strategy; or the historical scheduling strategy and the second intermediate scheduling strategy may be combined into a strategy set, and sampling may be performed in the strategy set to determine a new auxiliary scheduling strategy.

[0066] It can be understood that the target scheduling strategy may be the scheduling strategy corresponding to the maximum fitness data among the multiple N-th scheduling strategies obtained in the last iteration process.

[0067] Optionally, the iteration stop condition may include that the number of iterations satisfies a preset total number of iterations; the preset total number of iterations here may be pre-set, for example, may be 1000, which is not limited here.

[0068] Optionally, the iteration stop condition may include that the gain of the fitness data of a preset number of iterations is less than a preset gain threshold. The gain of the fitness data is determined by the fitness data of the Nth scheduling strategy obtained in each iterative process and the fitness data of the N-1th scheduling strategy obtained in the previous iterative process. Wherein, N is an integer greater than 1. Wherein, the gain of the fitness data can be determined based on the difference between the fitness data of the Nth scheduling strategy and the fitness data of the N-1th scheduling strategy, or based on the rate of change of the fitness data of the N-1th scheduling strategy. The gain of the fitness data can be a positive number or a negative number.

[0069] There may be multiple Nth scheduling strategies corresponding to each number of iterations, and there may be multiple fitness data of the corresponding Nth scheduling strategy. The gain of the fitness data may be determined by the maximum value of the fitness data of the Nth scheduling strategy and the maximum value of the fitness data of the N-1th scheduling strategy. N is the number of iterations.

[0070] If the gain of the fitness data for a preset number of consecutive iterations is less than a preset gain threshold, the preset gain threshold here may be 0, indicating that the iterative process for the preset number of consecutive iterations cannot optimize the scheduling strategy, and the iteration stop condition is met.

[0071] The technical solution provided in this embodiment sets an auxiliary scheduling strategy on the basis of legacy multiplication based on the initialization scheduling strategy, and performs diversity update on the first intermediate scheduling strategy obtained by multiplying the initialization scheduling strategy by extracting the first scheduling strategy knowledge in the auxiliary scheduling strategy, so as to avoid repeated scheduling strategies in the process of legacy multiplication based on the initialization scheduling strategy, which leads to the problem of local optimality, obtain the globally optimal target scheduling strategy, and improve the accuracy of the target scheduling strategy.

[0072] Figure 2 This is a flowchart of determining a scheduling strategy provided by an embodiment of the present invention, which is refined on the basis of the above embodiment. The method specifically includes:

[0073] S210: Acquire an initialization scheduling strategy and an auxiliary scheduling strategy, where the number of the initialization scheduling strategy and the number of the auxiliary scheduling strategies are plural respectively, and the initialization scheduling strategy and the auxiliary scheduling strategy do not overlap.

[0074] S220, extracting the second scheduling strategy knowledge based on the initialization scheduling strategy; screening the fitness data corresponding to the target scheduling requirements based on the initialization scheduling strategy and the second scheduling strategy knowledge respectively to obtain the scheduling strategy to be propagated; performing genetic propagation based on the scheduling strategy to be propagated to obtain the first intermediate scheduling strategy.

[0075] S230. Extract first scheduling strategy knowledge based on the auxiliary scheduling strategy, and update the first intermediate scheduling strategy based on the first scheduling strategy knowledge to obtain a first scheduling strategy.

[0076] S240, determine whether the iteration stop condition is met. If the iteration stop condition is not met, update the auxiliary scheduling strategy, and return to execute step S120 based on the first scheduling strategy and the updated second scheduling strategy. If the iteration stop condition is met, output the target scheduling strategy.

[0077] The method for extracting the second scheduling strategy knowledge includes: splitting the initialization scheduling strategy to obtain a candidate knowledge set, wherein the candidate knowledge set includes candidate knowledge, and the candidate knowledge includes one or more of a scheduling strategy decision tree, a scheduling strategy subtree and a common subtree; respectively determining the fitness data corresponding to the target scheduling requirement of the candidate knowledge in the candidate knowledge set, and determining the second scheduling strategy knowledge based on the fitness data corresponding to the candidate knowledge relative to the target scheduling requirement.

[0078] The method for extracting the second scheduling strategy knowledge is the same as the method for extracting the first-degree strategy knowledge, which will not be repeated here.

[0079] It can be understood that the number of initialization scheduling strategies is the first number, and the sum of the number of initialization scheduling strategies and the second scheduling strategy knowledge is greater than the first number. By screening the initialization scheduling strategies and the second scheduling strategy knowledge, the scheduling strategies to be propagated are obtained, and the initialization scheduling strategies or the second scheduling strategy knowledge that interfere with the genetic coding process of the scheduling strategies are removed, thereby reducing the amount of data processing and avoiding interference.

[0080] The fitness data corresponding to the initialization scheduling strategy and the second scheduling strategy knowledge relative to the target scheduling demand are determined respectively, the initialization scheduling strategy and the second scheduling strategy knowledge are sorted according to the fitness data, and the first number of scheduling strategies to be propagated are screened based on the sorting. For example, the initialization scheduling strategy and the second scheduling strategy knowledge are sorted from large to small based on the fitness data, and the first m positions of the sorting are determined as the scheduling strategies to be propagated, where m is the first number.

[0081] Here, the fitness data of the individual scheduling strategy corresponding to the target scheduling requirement can be determined based on a preset fitness function, and the individual scheduling strategy includes one or more of the initialization scheduling strategy and the second scheduling strategy knowledge; or, the fitness data of the individual scheduling strategy corresponding to the target scheduling requirement can be determined based on a predetermined agent model.

[0082] Genetically propagate the scheduling strategy to be propagated based on a preset genetic operator to obtain a first intermediate scheduling strategy, and update the first intermediate scheduling strategy through the first scheduling strategy knowledge extracted from the auxiliary scheduling strategy to obtain the first scheduling strategy.

[0083] In each iteration of the genetic mutation process, before genetic reproduction based on the N-1 scheduling strategy, the second scheduling strategy knowledge of the N-1 scheduling strategy is extracted, and the scheduling strategy to be propagated is selected from the second scheduling strategy knowledge and the N-1 scheduling strategy, which provides a better basis for the genetic reproduction process and is conducive to the rapid convergence of the genetic encoding process. Among them, the N-1 scheduling strategy in the first iteration is the initialization scheduling strategy.

[0084] Based on the above embodiments, the present invention also provides an optimal example of a method for determining a scheduling strategy. Figure 3 , Figure 3 It is a flowchart of a scheduling strategy determination method also provided in an embodiment of the present invention.

[0085] The strategy set of historical scheduling requirements includes historical scheduling strategies, and the historical scheduling strategies, subtrees of historical scheduling strategies, and common subtrees are randomly sampled from the historical scheduling strategies of historical scheduling requirements as initialization scheduling strategies. The historical scheduling strategies, subtrees of historical scheduling strategies, and common subtrees are randomly sampled from the historical scheduling strategies of historical scheduling requirements as auxiliary scheduling strategies, and the initialization scheduling strategies and the auxiliary scheduling strategies may not overlap. Multiple initialization scheduling strategies can be used as a population in the genetic encoding process, and each initialization scheduling strategy can be used as an individual.

[0086] The fitness data of the initialization scheduling strategy is evaluated through a preset fitness function, and a code model is constructed based on the initialization scheduling strategy and the fitness data of the initialization scheduling strategy. The second scheduling strategy knowledge of the initialization scheduling strategy is extracted through the proxy model, and the second scheduling strategy knowledge and the fitness data of the initialization scheduling strategy are evaluated through the proxy model, and the scheduling strategy to be propagated is selected, and the scheduling strategy to be propagated is genetically propagated to obtain the first intermediate scheduling strategy.

[0087] The first scheduling strategy knowledge of the auxiliary scheduling model is extracted through the agent model, and the auxiliary scheduling model is genetically propagated through the first scheduling strategy knowledge to obtain the second intermediate scheduling strategy. The first intermediate scheduling strategy and the second intermediate scheduling strategy are exchanged through crossover mutation to obtain the first scheduling strategy.

[0088] When the maximum number of iterations is reached or there is no incremental benefit after several rounds of iterations, that is, when the iteration notification condition is met, the target scheduling strategy is output. When the iteration notification condition is not met, the next iteration is executed based on the currently obtained Nth scheduling strategy.

[0089] Figure 4 Schematic diagram of a scheduling strategy determination device provided by an embodiment of the present invention. Figure 4 As shown, the device includes: an initialization module 310 and an iteration module 320. Among them:

[0090] An initialization module 310 is used to obtain an initialization scheduling strategy and an auxiliary scheduling strategy, wherein the number of the initialization scheduling strategy and the number of the auxiliary scheduling strategies are multiple, and the initialization scheduling strategy and the auxiliary scheduling strategy do not overlap;

[0091] The iteration module 320 is used to genetically propagate the initialization scheduling strategy to obtain a first intermediate scheduling strategy; extract the first scheduling strategy knowledge based on the auxiliary scheduling strategy, and update the first intermediate scheduling strategy based on the first scheduling strategy knowledge to obtain a first scheduling strategy; when the iteration stop condition is not met, update the auxiliary scheduling strategy, and execute the iteration process of the scheduling strategy based on the first scheduling strategy and the updated auxiliary scheduling strategy to obtain a second scheduling strategy, until the iteration stop condition is met to obtain a target scheduling strategy.

[0092] The technical solution of this embodiment sets an auxiliary scheduling strategy on the basis of heritage multiplication based on the initialization scheduling strategy, and performs diversity update on the first intermediate scheduling strategy obtained by multiplying the initialization scheduling strategy by extracting the first scheduling strategy knowledge in the auxiliary scheduling strategy, so as to avoid repeated scheduling strategies in the process of heritage multiplication based on the initialization scheduling strategy, which leads to the problem of local optimality, obtain the globally optimal target scheduling strategy, and improve the accuracy of the target scheduling strategy.

[0093] Based on the above embodiment, optionally, the initialization module 310 is used to obtain a set of historical scheduling strategies corresponding to historical scheduling requirements respectively; and perform sampling in the set of historical scheduling strategies to obtain a first number of initialization scheduling strategies and a second number of auxiliary scheduling strategies.

[0094] Based on the above embodiment, optionally, the iteration module 320 is used to: extract the second scheduling strategy knowledge based on the initialization scheduling strategy; screen the fitness data corresponding to the target scheduling requirements based on the initialization scheduling strategy and the second scheduling strategy knowledge, respectively, to obtain the scheduling strategy to be propagated; perform genetic propagation based on the scheduling strategy to be propagated, to obtain the first intermediate scheduling strategy.

[0095] Optionally, the auxiliary scheduling strategy or the initialization scheduling strategy is in the form of a decision tree;

[0096] The iteration module 320 is also used to: split the auxiliary scheduling strategy or the initialization scheduling strategy to obtain a candidate knowledge set, wherein the candidate knowledge set includes candidate knowledge, and the candidate knowledge includes one or more of a scheduling strategy decision tree, a scheduling strategy subtree and a common subtree; respectively determine the fitness data corresponding to the target scheduling requirement of the candidate knowledge in the candidate knowledge set, and determine the first scheduling strategy knowledge or the second scheduling strategy knowledge based on the fitness data corresponding to the candidate knowledge relative to the target scheduling requirement.

[0097] Optionally, the iteration module 320 is further used to determine the fitness data corresponding to the target scheduling requirement of the individual scheduling strategy based on a preset fitness function, wherein the individual scheduling strategy includes one or more of the initialization scheduling strategy, the auxiliary scheduling strategy, the candidate knowledge and the scheduling strategy knowledge;

[0098] Optionally, the iteration module 320 is further used to determine the fitness data of the individual scheduling strategy relative to the target scheduling demand based on a predetermined proxy model, wherein the proxy model is determined based on the correspondence between the individual scheduling strategy and the fitness data.

[0099] Optionally, the iteration module 320 is further configured to perform cross-mutation on the first scheduling strategy knowledge and the first intermediate scheduling strategy.

[0100] Optionally, the iteration module 320 is further used to perform genetic reproduction based on the first scheduling strategy knowledge and the auxiliary scheduling strategy to obtain a second intermediate scheduling strategy; and perform crossover mutation based on the first intermediate scheduling strategy and the second intermediate scheduling strategy.

[0101] Optionally, the iteration stop condition includes one or more of the following: the number of iterations meets a preset total number of iterations; the gain of the fitness data for a preset number of iterations is less than a preset gain threshold.

[0102] The scheduling strategy determination device provided in the embodiment of the present invention can execute the scheduling strategy determination method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0103] Figure 51 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0104] like Figure 5 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0105] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0106] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a scheduling strategy determination method.

[0107] In some embodiments, the scheduling policy determination method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the scheduling policy determination method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the scheduling policy determination method in any other appropriate manner (e.g., by means of firmware).

[0108] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0109] The computer program for implementing the scheduling strategy determination method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, partially on the machine as an independent software package and partially on a remote machine, or completely on a remote machine or server.

[0110] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to cause a processor to execute a scheduling strategy determination method, the method comprising:

[0111] An initialization scheduling strategy and an auxiliary scheduling strategy are obtained, wherein the number of the initialization scheduling strategy and the number of the auxiliary scheduling strategies are multiple respectively, and the initialization scheduling strategy and the auxiliary scheduling strategies do not overlap; genetically propagate the initialization scheduling strategy to obtain a first intermediate scheduling strategy; extract first scheduling strategy knowledge based on the auxiliary scheduling strategy, and update the first intermediate scheduling strategy based on the first scheduling strategy knowledge to obtain a first scheduling strategy; when an iteration stop condition is not met, update the auxiliary scheduling strategy, and execute an iterative process of the scheduling strategy based on the first scheduling strategy and the updated auxiliary scheduling strategy to obtain a second scheduling strategy, until the iteration stop condition is met, to obtain a target scheduling strategy.

[0112] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0113] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0114] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0115] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0116] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0117] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for determining a scheduling strategy, characterized in that: include: Acquire an initialization scheduling strategy and an auxiliary scheduling strategy, wherein the number of the initialization scheduling strategy and the number of the auxiliary scheduling strategies are multiple, and the initialization scheduling strategy and the auxiliary scheduling strategy do not overlap; Performing genetic propagation on the initialization scheduling strategy to obtain a first intermediate scheduling strategy; Extracting first scheduling strategy knowledge based on the auxiliary scheduling strategy, and updating the first intermediate scheduling strategy based on the first scheduling strategy knowledge to obtain a first scheduling strategy; When the iteration stop condition is not met, the auxiliary scheduling strategy is updated, and the scheduling strategy iteration process is performed based on the first scheduling strategy and the updated auxiliary scheduling strategy to obtain a second scheduling strategy, until the iteration stop condition is met to obtain a target scheduling strategy.

2. The method according to claim 1, characterized in that The obtaining of the initialization scheduling strategy and the auxiliary scheduling strategy includes: Obtain a set of historical scheduling strategies corresponding to each historical scheduling demand; Sampling is performed in the historical scheduling strategy set to obtain a first number of initialization scheduling strategies and a second number of auxiliary scheduling strategies.

3. The method according to claim 1, characterized in that The step of performing genetic propagation on the initialization scheduling strategy to obtain a first intermediate scheduling strategy includes: Extracting second scheduling strategy knowledge based on the initialization scheduling strategy; Based on the initialization scheduling strategy and the second scheduling strategy knowledge, the fitness data corresponding to the target scheduling requirements are respectively screened to obtain the scheduling strategy to be propagated; Genetic reproduction is performed based on the scheduling strategy to be reproduced to obtain a first intermediate scheduling strategy.

4. The method according to claim 3, characterized in that: The auxiliary scheduling strategy or the initialization scheduling strategy is in the form of a decision tree; The first scheduling strategy knowledge or the second scheduling strategy knowledge is extracted in a manner including: Splitting the auxiliary scheduling strategy or the initialization scheduling strategy to obtain a candidate knowledge set, wherein the candidate knowledge set includes candidate knowledge, and the candidate knowledge includes one or more of a scheduling strategy decision tree, a scheduling strategy subtree, and a common subtree; The fitness data corresponding to the target scheduling requirement of the candidate knowledge in the candidate knowledge set are respectively determined, and the first scheduling strategy knowledge or the second scheduling strategy knowledge is determined based on the fitness data corresponding to the target scheduling requirement of the candidate knowledge.

5. The method according to claim 3 or 4, characterized in that: The method for determining the fitness data includes: Determine the fitness data of the individual scheduling strategy corresponding to the target scheduling requirement based on a preset fitness function, wherein the individual scheduling strategy includes one or more of an initialization scheduling strategy, an auxiliary scheduling strategy, candidate knowledge, and scheduling strategy knowledge; or, The fitness data of the individual scheduling strategy corresponding to the target scheduling demand is determined based on a predetermined proxy model, wherein the proxy model is determined based on the corresponding relationship between the individual scheduling strategy and the fitness data.

6. The method according to claim 1, characterized in that The updating of the first intermediate scheduling strategy based on the first scheduling strategy knowledge to obtain a first scheduling strategy includes one or more of the following: Cross-mutate the first scheduling strategy knowledge and the first intermediate scheduling strategy; Genetic reproduction is performed based on the first scheduling strategy knowledge and the auxiliary scheduling strategy to obtain a second intermediate scheduling strategy; and crossover mutation is performed based on the first intermediate scheduling strategy and the second intermediate scheduling strategy.

7. The method according to claim 1, characterized in that The iteration stop condition includes one or more of the following: the number of iterations meets the preset total number of iterations; The gain of the fitness data for a preset number of iterations is less than a preset gain threshold.

8. A scheduling strategy determination device, characterized in that: include: An initialization module, used to obtain an initialization scheduling strategy and an auxiliary scheduling strategy, wherein the number of the initialization scheduling strategy and the number of the auxiliary scheduling strategies are multiple, and the initialization scheduling strategy and the auxiliary scheduling strategy do not overlap; An iteration module, used for performing genetic propagation on the initialization scheduling strategy to obtain a first intermediate scheduling strategy; Extracting first scheduling strategy knowledge based on the auxiliary scheduling strategy, and updating the first intermediate scheduling strategy based on the first scheduling strategy knowledge to obtain a first scheduling strategy; When the iteration stop condition is not met, the auxiliary scheduling strategy is updated, and the scheduling strategy iteration process is performed based on the first scheduling strategy and the updated auxiliary scheduling strategy to obtain a second scheduling strategy, until the iteration stop condition is met to obtain a target scheduling strategy.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the scheduling strategy determination method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the scheduling strategy determination method according to any one of claims 1 to 7 when executed.

Images