Dynamically constrained multi-resource scheduling method and system based on decision threshold

Abstract

The invention belongs to the technical field of resource scheduling, and discloses a dynamically constrained multi-resource scheduling method and system based on a decision threshold, and the method comprises the steps: obtaining a real-time resource state and a current environment constraint state, and calculating a value function under the current resource state and the current environment constraint state; constructing a multi-source dynamic resource coupling system model, establishing a cross-state optimization equation set based on the model and the value function, and solving the cross-state optimization equation set to obtain a decision threshold; constructing a two-state environment constraint indicator, judging an executable state of an environment constraint state based on the indicator, calculating a resource proportion according to a real-time resource state, and obtaining a relative state index; and based on the decision threshold and the executable state, constructing a resource scheduling dynamic decision engine, and performing real-time decision on multi-resource scheduling. According to the method, the dynamic decision engine is constructed to process the resource state and the constraint state in real time, a scientific threshold type decision signal is output, and the decision efficiency of resource scheduling with constraints is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of resource scheduling technology, and in particular to a dynamically constrained multi-resource scheduling method and system based on decision thresholds. Background Technology

[0002] Dynamically constrained multi-resource scheduling refers to a strategy that adjusts resource allocation in real time based on system operating status, task requirements, and environmental changes during the resource allocation process, while simultaneously considering the optimization of various constraints (such as load, power, and bandwidth). Common scenarios involving dynamically constrained multi-resource scheduling include cloud resource release and reclamation, and power grid peak scheduling.

[0003] For the above scenarios, traditional multi-resource scheduling methods struggle to address two core challenges: 1. Real-time coupling of environmental constraints and resource dynamics: External constraints (such as temporary resource unavailability, market time window) appear in the form of random switching (two-state Markov process), while the states of multiple resources (equipment load, price signals, etc.) fluctuate continuously driven by environmental noise. The asynchronous changes of the two lead to misjudgment of decision timing.

[0004] 2. High-dimensional decision lag: The coupling relationship between resources (such as energy consumption ratio and supply-demand ratio) cannot be resolved in real time in multidimensional stochastic systems, resulting in delay in strategy generation and significantly reducing scheduling benefits. Summary of the Invention

[0005] To address the aforementioned issues, this invention proposes a dynamically constrained multi-resource scheduling method and system based on decision thresholds. It constructs a dynamic decision engine to process resource and constraint states in real time and outputs scientific threshold-based decision signals, significantly improving the efficiency of constrained resource scheduling decisions.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a dynamically constrained multi-resource scheduling method based on a decision threshold, comprising the following steps: Obtain real-time resource status and current environmental constraint status, and calculate the value function under the current resource status and environmental constraint status; A multi-source dynamic resource coupling system model constrained by the actual available resource window is constructed. Based on this model and the value function, a cross-state optimization equation system integrating resource dynamics and constraint states is established. The cross-state optimization equation system is solved, and the decision threshold is obtained through partitioned simultaneous equations and smooth fitting. Construct a dual-state environment constraint indicator, determine the executable state of the environment constraint state based on the indicator, calculate the ratio of the amount of resources to be recycled to the amount of resources to be released based on the real-time resource state, and obtain the relative state index; Based on the executable state of decision thresholds and environmental constraints, a dynamic decision engine for resource scheduling is constructed to make real-time decisions on multi-resource scheduling.

[0007] As an alternative implementation, when constructing a two-state environment constraint indicator, specifically: Collect observable measurements related to environmental constraints, construct environmental feature vectors, and calculate the environmental constraint intensity index based on the environmental feature vectors; Based on a preset threshold or hysteresis mechanism, an instantaneous environmental executable decision signal is generated from the environmental constraint strength index. The state switching rate of the dual-state Markov chain is dynamically adjusted based on the instantaneous environment executable decision signal. Random state transitions are performed based on the adjusted switching rate to obtain a final environmental constraint state indication with time consistency.

[0008] As an alternative implementation method, the multi-source dynamic resource coupling system model constrained by the actual available resource window is as follows: ; in, n For the number of resource states, For environmental noise, , i =1,2,..., n This represents the expected direction of change in resource status. The impact intensity and correlation of environmental noise on resource status.

[0009] As an alternative implementation, the multi-resource scheduling method further includes using relative state indices to reduce the dimensionality of the established cross-state optimization equation set, thereby obtaining a cross-state optimization equation set characterized by relative state indices. In this equation set, the cross-state optimization rules of the hybrid dynamic system consist of two parts: waiting rules and triggering rules.

[0010] As an alternative implementation, the cross-state optimization equations characterized by relative state indices are as follows: .

[0011] As an alternative implementation, the waiting rule is determined by... Confirmed, the triggering rule is determined by trigger.

[0012] Secondly, the present invention provides a dynamically constrained multi-resource scheduling system based on a decision threshold, comprising: The data acquisition module is configured to: acquire real-time resource status and current environmental constraint status, and calculate the value function under the current resource status and environmental constraint status; The decision threshold calculation module is configured to: construct a multi-source dynamic resource coupling system model constrained by the actual available resource window; based on the model and the value function, establish a cross-state optimization equation system that integrates resource dynamics and constraint state; solve the cross-state optimization equation system; and obtain the decision threshold through partitioned simultaneous equations and smooth fitting. The decision parameter calculation module is configured to: construct a dual-state environmental constraint indicator, determine the executable state of the environmental constraint state based on the indicator, calculate the ratio of the amount of resources to be recycled to the amount of resources to be released based on the real-time resource state, and obtain the relative state index. The resource scheduling module is configured to: build a dynamic decision engine for resource scheduling based on the executable state of decision thresholds and environmental constraints, and make real-time decisions on multi-resource scheduling.

[0013] Thirdly, the present invention provides an electronic device including a memory and a processor, and computer instructions stored in the memory and running on the processor, wherein the computer instructions, when executed by the processor, perform the method described in the first aspect.

[0014] Fourthly, the present invention provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, perform the method described in the first aspect.

[0015] Fifthly, the present invention provides a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention proposes a dynamically constrained multi-resource scheduling method based on decision thresholds. It considers a high-order dynamic resource model driven by Brownian motion and couples it with a two-state Markov chain to characterize the state constraints. This transforms the complex policy-solving problem into a quasi-variational inequality problem for a hybrid diffusion system, successfully deriving an executable solution with clear signal meaning. This method provides scientific threshold-based decision signals, significantly improving the efficiency of constrained resource scheduling decisions. It is particularly suitable for intelligent decision-making regarding optimal scheduling timing in dynamic systems constrained by external environments.

[0017] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0018] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0019] Figure 1 This is a flowchart of the dynamically constrained multi-resource scheduling method based on decision thresholds according to the present invention; Figure 2 This is a flowchart of the determination and triggering process of the dual-state environment constraint indicator of the present invention; Figure 3 Flowchart for deriving cross-state optimization rules for hybrid dynamic systems; Figure 4 A flowchart for the algorithm that generates decision thresholds; Figure 5 A flowchart for the real-time decision-making of the resource scheduling dynamic decision engine; Figure 6 Sensitivity analysis of the target operation threshold with respect to resource trend parameters; Figure 7 Sensitivity analysis of the target operating threshold with respect to environmental disturbance parameters; Figure 8 Analysis of the asymptotic properties of the target operation threshold with respect to constraint parameters. Detailed Implementation

[0020] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0021] It should be noted that the following detailed description is exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0022] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments of the invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form as well. Furthermore, it should be understood that the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but includes other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0023] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0024] Example 1 like Figure 1 As shown, this embodiment provides a dynamically constrained multi-resource scheduling method based on decision thresholds, including the following steps: Obtain real-time resource status and current environmental constraint status, and calculate the value function under the current resource status and environmental constraint status; A multi-source dynamic resource coupling system model constrained by the actual available resource window is constructed. Based on this model and the value function, a cross-state optimization equation system integrating resource dynamics and constraint states is established. The cross-state optimization equation system is solved, and the decision threshold is obtained through partitioned simultaneous equations and smooth fitting. Construct a dual-state environment constraint indicator, determine the executable state of the environment constraint state based on the indicator, calculate the ratio of the amount of resources to be recycled to the amount of resources to be released based on the real-time resource state, and obtain the relative state index; Based on the executable state of decision thresholds and environmental constraints, a dynamic decision engine for resource scheduling is constructed to make real-time decisions on multi-resource scheduling.

[0025] The specific solution of the present invention is as follows: This invention provides an intelligent algorithm and system implementation scheme for handling multi-source dynamic resource scheduling constrained by an actual available resource window. The core of this system lies in constructing a dynamic decision engine (solver), which can process resource and constraint states in real time and output precise decision signals of "execute immediately" or "wait continuously". See also... Figure 5 The core flowchart of the solver shown below illustrates the steps of this method: Step 1: The hybrid dynamic system receives real-time status data streams from multiple resources. (These resources are not limited to: device sensor readings, market price signals, network load values, etc.) The hybrid dynamic system integrates physical resource dynamics (a multi-state stochastic process driven by Brownian motion, including trend direction and noise intensity parameters) and constraint dynamics (executable state switches controlled by a two-state Markov chain). A coupled dynamic system model is used to describe the evolution of resource states: ; in, n For resource status quantity, environmental noise For m-dimensional Brownian motion, , i =1,2,..., n This refers to the expected direction of change in resource status (such as equipment wear rate and resource demand growth rate). The impact and correlation of environmental noise on resource status are investigated. To address real-world constraints such as inconsistencies between resource status and execution timescales, and insufficient resource acquisition, a bi-state environmental constraint indicator is constructed. The core setting of the environmental constraint indicator is that it only applies when... The system can trigger target operations (such as resource release or task execution). This indicates that the environment does not meet the execution conditions, and the system should either wait or prohibit execution. Estimation is performed using observable environmental data, and temporal consistency is maintained through a random switching mechanism. The measurement and estimation steps are as follows: Figure 2 As shown, it includes: Step 1.1: Collect observables related to environmental constraints.

[0026] The following environmental constraint-related input data are collected in real time by the time system: Time window availability flag This indicates whether the current period is within the allowed execution timeframe; Resource availability margin This represents the proportion of the system's remaining resources (bandwidth, power, computing power, etc.) relative to its maximum capacity. External mandatory constraint mark This is used to indicate whether there are external restrictions on operations, compliance, or regulation that prohibit their execution; Environmental fluctuation indicators It is used to reflect the strength of market fluctuations, load jitter, or noise disturbances.

[0027] The above quantities are combined to form an environmental feature vector: ; Step 1.2: Calculate the environmental constraint intensity index .

[0028] To provide a unified measure for multidimensional environmental variables, the system employs a linear aggregation approach to construct the environmental constraint strength index. ; in This is the environmental sensitivity weight, used to balance the influence of different constraint factors.

[0029] The larger the value, the stronger the environmental constraints, making it less likely to trigger execution.

[0030] Step 1.3: Obtain instantaneous executability determination based on the threshold. .

[0031] Set the environment's executable threshold The system is based on the constraint index Perform instantaneous determination: ; When it is necessary to avoid frequent switching, the system can adopt a dual-threshold hysteresis mechanism:

[0032] in .

[0033] Step 1.4: Adjust the Markov switching rate based on instantaneous determination.

[0034] To characterize the randomness and persistence of the execution window, the system uses a bi-state Markov chain description. The evolution of , its generator matrix is: ; in This indicates the switching rate from executable state 1 to non-executable state 0. This represents the switching rate from state 0 to state 1. Based on the instantaneous determination result. Dynamically adjust the switching rate: when At time (environmentally determined to be executable): ; when At the time (the environment determined it to be unexecutable): ; The "high / low" values ​​can be set based on experience or obtained through training.

[0035] Step 1.5: Obtain the final environmental constraint indicator through random transitions. .

[0036] at discrete time step The system then calculates the state transition probability based on the current switching rate: ; Update the environment constraint indicator according to the following rules:

[0037] This yields a time-consistent and noise-robust final environmental constraint state. .

[0038] Step 2: Define the restricted value function within the set S of allowed execution times. : ; in, The net revenue coefficient after deducting scheduling costs from the unit price of resource release. This is the net expenditure coefficient after adding scheduling costs to the unit price of resource recovery. τ is the continuous-time discount factor, and τ is the random triggering time that satisfies the environmental state constraints in steps 1.1-1.5.

[0039] The above formula represents the current resource state. and constraint state The expected long-term benefits of the system implementing the optimal operating strategy are estimated below. Figure 3The derivation of the quasi-variable inequality for the hybrid diffusion system shown is essentially based on maximizing the long-term net benefits (such as reduced energy consumption and increased revenue) of the target operation by constructing the Hamilton-Jacobi-Bellman (HJB) equation through the time value factor. Calculate the expected value. Specifically: First, using the principles of dynamic programming, a system of cross-state optimization equations is established that integrates resource dynamics and constraint states. , ; in, , , , .

[0040] Then, extract the relative status indicators among the key resources. (e.g., energy consumption ratio, supply-demand ratio) dimensionality reduction is performed, simplifying multidimensional problems into univariate decision models, and then... This yields a set of cross-state optimization equations characterized by relative state indices: ;in, .

[0041] Finally, in this system of equations, the cross-state optimization rule for the hybrid dynamic system consists of two parts: the waiting rule is composed of... Confirmed, the triggering rule is determined by trigger.

[0042] Step 3: Perform the following steps on the cross-state optimization equations constructed in Step 2, characterized by relative state indices: Figure 4 The solution process shown determines the target operation threshold for performing multi-resource scheduling under dynamic constraints. Specifically: First, in the dual-state environment constraint indicator Based on this, an additional relative state index is introduced. y Target operation threshold k A cross-state optimization equation executor based on dual judgment of resource dynamics and constraint state, as shown in Table 1, is constructed. Specifically, a threshold exists. k When y < k When the resource release action is triggered, y > k The resource is kept in reserve.

[0043] Table 1. Cross-state optimization equation executor based on dual judgment of resource dynamics and constraint state;

[0044] Then, by jointly executing the equations under different constraint states, the undetermined coefficient solution of the cross-state optimization equation system is derived. .

[0045] in, These are characteristic roots. These are the coefficients to be determined.

[0046] Finally utilize At the target operation threshold boundary point of the relative state index The continuous differentiability at a given point allows for the analytical determination of the coefficients to be determined using a smoothing fitting technique. and target operation threshold : ; in, ; .

[0047] The target operation threshold k This is the core output of the intelligent decision-making algorithm, used in the resource scheduling dynamic decision engine to determine whether the resource dynamics have met the execution criteria.

[0048] Step 4: Based on the intelligent algorithm for multi-resource scheduling thresholds with dynamic constraints, construct a dynamic decision engine for resource scheduling and perform multi-dimensional validity verification to achieve a complete closed loop for intelligent system decision-making.

[0049] First, based on the target operation threshold and the dual-state environment constraint indicator, a dual-drive decision-maker is constructed as follows: Figure 4 The example shown is a resource scheduling dynamic decision engine that transforms resource dynamics and environmental constraints into real-time operation commands. Specifically: (1) Input: Enter the current environmental constraint state The data is input into the constraint state determiner, along with the real-time resource status. and the calculated relative state index The input relative state judge is a target operation threshold k dynamically calculated by an intelligent algorithm. (2) After verification by the dual-drive detector, a decision command is output: .

[0050] Then, a multi-dimensional effectiveness verification was performed, demonstrating the specific implementation effect of our method in this embodiment for handling two types of heterogeneous resources, as well as the analysis of the changes in threshold k with respect to model parameters and constraint parameters. Specifically: (1) Sensitivity to trend parameters: such as Figure 6This shows the threshold k for handling two types of heterogeneous resources, release resource A and recovery resource B, in this embodiment, with respect to the resource trend parameter (resource gain coefficient or recovery loss coefficient). Sensitivity analysis. When the gain coefficient of resource A to be released increases or the loss coefficient of resource B to be recycled decreases, the threshold k decreases, i.e., the release area shrinks. The reason is that the increase in the future value of the resource to be released or the decrease in the recycling cost both increase the marginal benefit of continuing to hold it. The system rationally postpones the trigger to lock in higher overall benefits, such as delaying shutdown when the CPU utilization of a cloud host is expected to rise.

[0051] (2) Sensitivity to fluctuation parameters: such as Figure 7 This shows the threshold k for handling two types of heterogeneous resources, resource A and resource B, in this embodiment with respect to the environmental disturbance parameter. Sensitivity analysis. When the environmental disturbance parameter increases, the threshold k decreases. The reason is that high disturbance amplifies the uncertain gains of future states, and the system requires more stringent immediate gain compensation before giving up the wait-and-see approach, such as the "wait-and-see-re-scale-down" strategy when network traffic jitter intensifies.

[0052] (3) Asymptotic characteristics of constraint parameters: such as Figure 8 Further analysis was conducted on the threshold k with respect to the constraint parameters. and The asymptotic nature of [the property]. First, we fix [the property]. Unchanged, with Increasing k gradually decreases this because the scheduling window appears more frequently, giving the system more opportunities to execute and thus making it more willing to wait for a better resource ratio to trigger the operation; simultaneously, when hour ,in This value coincides precisely with the threshold calculated using the method in step 3 for the unconstrained cross-state optimization equations, verifying the model's self-consistency in the case of constraint disappearance. Specifically, This means the constraints disappear, at which point it degenerates into an unconstrained resource scheduling problem. Then we fix... Unchanged, with As k gradually increases, it does so because the scheduling window is compressed, the executable interval is shortened, and the system is forced to trigger earlier to avoid errors; simultaneously, when hour ,in The limit value of k approaches the boundary strategy of "execute as soon as the window is seen".

[0053] The above analysis fully demonstrates the intrinsic influence of resource trends and environmental disturbances on optimal scheduling timing, fully verifying the analytical accuracy of the intelligent solver constructed in this invention in characterizing dynamic resource coupling relationships, and providing a robust and practical decision-making basis for constrained multi-resource dynamic scheduling. (Threshold) kThe monotonic response to the constraint frequency is in complete agreement with the engineering optimization logic of the system’s “waiting cost-opportunity trade-off”, proving the consistency of the invention in a constrained scheduling environment.

[0054] Example 2 This embodiment provides a dynamically constrained multi-resource scheduling system based on decision thresholds, including: The data acquisition module is configured to: acquire real-time resource status and current environmental constraint status, and calculate the value function under the current resource status and environmental constraint status; The decision threshold calculation module is configured to: construct a multi-source dynamic resource coupling system model constrained by the actual available resource window; based on the model and the value function, establish a cross-state optimization equation system that integrates resource dynamics and constraint state; solve the cross-state optimization equation system; and obtain the decision threshold through partitioned simultaneous equations and smooth fitting. The decision parameter calculation module is configured to: construct a dual-state environmental constraint indicator, determine the executable state of the environmental constraint state based on the indicator, calculate the ratio of the amount of resources to be recycled to the amount of resources to be released based on the real-time resource state, and obtain the relative state index. The resource scheduling module is configured to: build a dynamic decision engine for resource scheduling based on the executable state of decision thresholds and environmental constraints, and make real-time decisions on multi-resource scheduling.

[0055] It should be noted that the above modules correspond to the steps in Embodiment 1, and the examples and application scenarios implemented by the above modules and their corresponding steps are the same, but are not limited to the content disclosed in Embodiment 1. It should also be noted that the above modules can be executed in a computer system as part of the system.

[0056] In further embodiments, the following is also provided: An electronic device includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, perform the method described in Embodiment 1. For brevity, further details are omitted here.

[0057] It should be understood that in this embodiment, the processor can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0058] A computer-readable storage medium for storing computer instructions that, when executed by a processor, perform the method of Embodiment 1.

[0059] The method in Example 1 can be directly executed by a hardware processor, or it can be executed by a combination of hardware and software modules within the processor. The software modules can reside in readily available storage media in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, a detailed description is not provided here.

[0060] A computer program product includes a computer program that, when executed by a processor, implements the method in Embodiment 1.

[0061] The present invention also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, which execute in a device on a target real or virtual processor to perform the processes / methods described above. Typically, program modules include routines, programs, libraries, objects, classes, components, data structures, etc., that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of program modules can be combined or divided among program modules as needed. The machine-executable instructions for the program modules can execute within a local or distributed device. In a distributed device, the program modules can reside in both local and remote storage media.

[0062] The computer program code used to implement the methods of the present invention may be written in one or more programming languages. This computer program code may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the computer or other programmable data processing device, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a computer, partially on a computer, as a stand-alone software package, partially on a computer and partially on a remote computer, or entirely on a remote computer or server.

[0063] In the context of this invention, computer program code or related data may be carried by any suitable carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and the like. Examples of signals may include electrical, optical, radio, sound, or other forms of propagation signals, such as carrier waves, infrared signals, etc.

[0064] Those skilled in the art will recognize that the units and algorithm steps described in conjunction with the embodiments herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0065] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A dynamically constrained multi-resource scheduling method based on decision thresholds, characterized in that, Includes the following steps: Obtain real-time resource status and current environmental constraint status, and calculate the value function under the current resource status and environmental constraint status; A multi-source dynamic resource coupling system model constrained by the actual available resource window is constructed. Based on this model and the value function, a cross-state optimization equation system integrating resource dynamics and constraint states is established. The cross-state optimization equation system is solved, and the decision threshold is obtained through partitioned simultaneous equations and smooth fitting. Construct a dual-state environment constraint indicator, determine the executable state of the environment constraint state based on the indicator, calculate the ratio of the amount of resources to be recycled to the amount of resources to be released based on the real-time resource state, and obtain the relative state index; Based on the executable state of decision thresholds and environmental constraints, a dynamic decision engine for resource scheduling is constructed to make real-time decisions on multi-resource scheduling.

2. The multi-resource scheduling method based on decision thresholds under dynamic constraints as described in claim 1, characterized in that, Construct a two-state environment constraint indicator, specifically as follows: Collect observable measurements related to environmental constraints, construct environmental feature vectors, and calculate the environmental constraint intensity index based on the environmental feature vectors; Based on a preset threshold or hysteresis mechanism, an instantaneous environmental executable decision signal is generated from the environmental constraint strength index. The state switching rate of the dual-state Markov chain is dynamically adjusted based on the instantaneous environment executable decision signal. Random state transitions are performed based on the adjusted switching rate to obtain a final environmental constraint state indication with time consistency.

3. The multi-resource scheduling method based on decision thresholds under dynamic constraints as described in claim 1, characterized in that, The multi-source dynamic resource coupling system model constrained by the actual available resource window is as follows: ； in, n For the number of resource states, For environmental noise, , i =1,2,..., n This represents the expected direction of change in resource status. The impact intensity and correlation of environmental noise on resource status.

4. The dynamically constrained multi-resource scheduling method based on decision thresholds as described in claim 1, characterized in that, The multi-resource scheduling method further includes using relative state indices to reduce the dimensionality of the established cross-state optimization equation set, thereby obtaining a cross-state optimization equation set characterized by relative state indices. In this equation set, the cross-state optimization rules of the hybrid dynamic system consist of two parts: waiting rules and triggering rules.

5. The dynamically constrained multi-resource scheduling method based on decision thresholds as described in claim 4, characterized in that, The cross-state optimization equations characterized by relative state indices are as follows: 。 6. The dynamically constrained multi-resource scheduling method based on decision thresholds as described in claim 4, characterized in that, Waiting rules are determined by Confirmed, the triggering rule is determined by trigger.

7. A dynamically constrained multi-resource scheduling system based on decision thresholds, characterized in that, include: The data acquisition module is configured to: acquire real-time resource status and current environmental constraint status, and calculate the value function under the current resource status and environmental constraint status; The decision threshold calculation module is configured to: construct a multi-source dynamic resource coupling system model constrained by the actual available resource window; based on the model and the value function, establish a cross-state optimization equation system that integrates resource dynamics and constraint state; solve the cross-state optimization equation system; and obtain the decision threshold through partitioned simultaneous equations and smooth fitting. The decision parameter calculation module is configured to: construct a dual-state environmental constraint indicator, determine the executable state of the environmental constraint state based on the indicator, calculate the ratio of the amount of resources to be recycled to the amount of resources to be released based on the real-time resource state, and obtain the relative state index. The resource scheduling module is configured to: build a dynamic decision engine for resource scheduling based on the executable state of decision thresholds and environmental constraints, and make real-time decisions on multi-resource scheduling.

8. An electronic device, characterized in that, It includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, perform the method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, Used to store computer instructions, which, when executed by a processor, perform the method described in any one of claims 1-6.

10. A computer program product, characterized in that, Includes a computer program, which, when executed by a processor, implements the method described in any one of claims 1-6.

Images