Emergency decision generation method and device, computer device, storage medium and program product

Abstract

The application relates to an emergency decision generation method and device, computer equipment, a storage medium and a program product. The method comprises the following steps: processing operation data flow of a plurality of business units in a coal supply system to obtain an instantaneous coupling matrix; determining whether the coal supply system is in a disordered resonance mode according to the instantaneous coupling matrix, and obtaining a target emergency decision based on a prediction model and a target function in the case that the coal supply system is determined to be in the disordered resonance mode; and the target emergency decision is used for adjusting the operation rhythm of at least one business unit. By using the method, the risk of the coal supply system before a fault event occurs can be predicted, early intervention before the fault event occurs can be realized, the problem of negative influence diffusion caused by response after the fault event occurs can be effectively avoided, and therefore, the emergency intervention cost can be reduced.

Description

Technical Field

[0001] This application relates to the field of coal supply chain risk assessment technology, and in particular to an emergency decision generation method, apparatus, computer equipment, storage medium and program product. Background Technology

[0002] As a basic energy source, the stability and security of the coal supply chain are of paramount importance. The coal supply chain is a complex system consisting of multiple links, including production, transportation, transshipment, and consumption. These links are closely coupled and influence each other. In actual operation, any disturbance in a single link, such as equipment failure, weather effects, or market fluctuations, can trigger a chain reaction, causing widespread and cross-linked operational disruptions, i.e., cascading risks.

[0003] To address these emergencies, existing technologies typically employ emergency decision support systems. These systems provide managers with decision-making support by monitoring operational parameters of key business units such as coal production, railway capacity, port inventory, and power plant coal consumption. However, these systems usually operate on an event-driven basis, meaning that their emergency response mechanisms are activated only after a specific, occurring failure event or operational anomaly exceeding a preset static threshold is detected.

[0004] This event-driven decision-making model inevitably involves decisions and interventions following a failure event, resulting in an inherent time lag in response. By the time the system takes intervention measures, the negative impacts of the initial disturbance have often already spread, increasing intervention costs and making it more difficult to recover losses. Summary of the Invention

[0005] Therefore, it is necessary to provide an emergency decision generation method, apparatus, computer equipment, storage medium, and program product to address the aforementioned technical problems.

[0006] In a first aspect, this application provides an emergency decision generation method, which includes: processing the operational data streams of multiple business units in a coal supply system to obtain an instantaneous coupling matrix; the value of each element in the instantaneous coupling matrix is ​​used to characterize the coupling strength of the operational rhythm between two business units; determining whether the coal supply system is in an imbalanced resonance mode based on the instantaneous coupling matrix, and, if it is determined that the coal supply system is in an imbalanced resonance mode, obtaining a target emergency decision based on a prediction model and an objective function; the target emergency decision is used to adjust the operational rhythm of at least one business unit.

[0007] In one embodiment, obtaining the target emergency decision based on the instantaneous coupling matrix includes: selecting a target resonant mode from multiple efficient resonant modes in a library of efficient resonant modes; inputting the instantaneous coupling matrix into a prediction model to obtain the predicted system state; solving the objective function based on the target resonant mode and the predicted system state to obtain the optimal intervention waveform; and processing the optimal intervention waveform to obtain the target emergency decision.

[0008] In one embodiment, the objective function includes a risk cost term, a resonance benefit term, and a control cost term, wherein: the risk cost term is used to penalize the predicted system state for approaching any misaligned resonance mode in the misaligned resonance mode library; the resonance benefit term is used to reward the predicted system state for approaching the target resonance mode; and the control cost term is used to penalize the amplitude and drastic change of the intervention waveform.

[0009] In one embodiment, the method further includes: obtaining the actual evolution trajectory based on multiple instantaneous coupling matrices corresponding to a preset time period; obtaining the predicted evolution trajectory based on the predicted system state corresponding to the preset time period; and adjusting the parameters of the prediction model based on the difference between the actual evolution trajectory and the predicted evolution trajectory to obtain an updated prediction model.

[0010] In one embodiment, the operational data streams of multiple business units in the coal supply system are processed to obtain an instantaneous coupling matrix, including: for two business units, performing continuous wavelet transform on the operational data streams of the two business units respectively to obtain two wavelet spectra; calculating a wavelet coherence spectrum based on the two wavelet spectra; integrating the wavelet coherence spectrum over a preset operational frequency interval to obtain the element values ​​corresponding to the two business units; and obtaining an instantaneous coupling matrix based on the element values ​​corresponding to every two business units in the multiple business units.

[0011] In one embodiment, determining whether the coal supply system is in an off-mode resonance based on the instantaneous coupling matrix includes: reducing the dimensionality of the instantaneous coupling matrix to obtain a low-dimensional latent vector corresponding to the instantaneous coupling matrix; determining whether there is an off-mode resonance in the off-mode resonance library that matches the low-dimensional latent vector; if yes, then determining that the coal supply system is in an off-mode resonance; if no, then determining that the coal supply system is not in an off-mode resonance.

[0012] In one embodiment, the method further includes: squaring the values ​​of each element in the instantaneous coupling matrix to obtain multiple squared values; performing a weighted summation of the multiple squared values ​​to obtain a target value; and taking the square root of the target value to obtain a system resonance index; the system resonance index is used to quantify the vulnerability of the coal supply system.

[0013] Secondly, this application also provides an emergency decision generation device, which includes:

[0014] The first determining module is used to process the operational data streams of multiple business units in the coal supply system to obtain an instantaneous coupling matrix; the value of each element in the instantaneous coupling matrix is ​​used to characterize the operational rhythm coupling strength between two business units.

[0015] The second determining module is used to determine whether the coal supply system is in an imbalanced resonance mode based on the instantaneous coupling matrix, and when it is determined that the coal supply system is in an imbalanced resonance mode, to obtain a target emergency decision based on the prediction model and the objective function; the target emergency decision is used to adjust the operating rhythm of at least one business unit.

[0016] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in any one of the first aspects above.

[0017] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the first aspects above.

[0018] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in any one of the first aspects above.

[0019] The aforementioned emergency decision generation method, apparatus, computer equipment, storage medium, and program product process the operational data streams of multiple business units in a coal supply system to obtain an instantaneous coupling matrix. Each element in the instantaneous coupling matrix characterizes the coupling strength of the operational rhythms between two business units. Then, based on the instantaneous coupling matrix, it is determined whether the coal supply system is in an imbalanced resonance mode. If the coal supply system is determined to be in an imbalanced resonance mode, a target emergency decision is obtained based on a prediction model and an objective function. This target emergency decision is used to adjust the operational rhythm of at least one business unit. Thus, by using the instantaneous coupling matrix obtained from the operational data streams of the coal supply system, risks to the coal supply system can be predicted before a failure event occurs. When it is determined that the coal supply system is in an imbalanced resonance mode, a target emergency decision is generated. Intervention based on this target emergency decision allows for early intervention before a failure event occurs, effectively avoiding the negative impact spread caused by responding after a failure event has occurred, thereby reducing emergency intervention costs. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is an application environment diagram of the emergency decision generation method in one embodiment;

[0022] Figure 2 This is a flowchart illustrating an emergency decision generation method in one embodiment;

[0023] Figure 3 This is a flowchart illustrating a method for determining whether a coal supply system is in an off-mode resonance state, as shown in one embodiment.

[0024] Figure 4 This is a schematic diagram of a method for generating target emergency decisions in one embodiment;

[0025] Figure 5 This is a schematic diagram of the update method for the prediction model in one embodiment;

[0026] Figure 6 This is a schematic diagram of the emergency decision generation system in one embodiment;

[0027] Figure 7 This is a structural block diagram of an emergency decision generation device in one embodiment;

[0028] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0030] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0031] As a basic energy source, the stability and security of the coal supply chain are of paramount importance. The coal supply chain is a complex system consisting of multiple links, including production, transportation, transshipment, and consumption. These links are closely coupled and influence each other. In actual operation, any disturbance in a single link, such as equipment failure, weather effects, or market fluctuations, can trigger a chain reaction, causing widespread and cross-linked operational disruptions, i.e., cascading risks.

[0032] To address these emergencies, existing technologies typically employ emergency decision support systems. These systems provide managers with decision-making support by monitoring operational parameters of key business units such as coal production, railway capacity, port inventory, and power plant coal consumption. However, these systems usually operate on an event-driven basis, meaning that their emergency response mechanisms are activated only after a specific, occurring failure event or operational anomaly exceeding a preset static threshold is detected.

[0033] This event-driven decision-making model inevitably involves decisions and interventions following a failure event, resulting in an inherent time lag in response. By the time the system takes intervention measures, the negative impacts of the initial disturbance have often already spread, increasing intervention costs and making it more difficult to recover losses.

[0034] Furthermore, existing technologies, when conducting status assessments, tend to focus on the independent analysis of individual business units, lacking quantitative monitoring and analysis methods for the dynamic coupling relationships between the operational rhythms of various units. Therefore, these systems struggle to identify the cumulative process of systemic vulnerabilities caused by uncoordinated operational rhythms across multiple stages before specific failures occur. This accumulated vulnerability is a breeding ground for cascading risks, but because existing technologies cannot effectively predict it, proactive intervention measures cannot be taken to avoid or mitigate potential systemic risks.

[0035] Therefore, it is necessary to propose effective technical means to solve the above problems. The following will explain in detail how this application solves the above technical problems, with reference to the accompanying drawings.

[0036] The emergency decision generation method provided in this application embodiment can be applied to, for example, Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located on the cloud or other network servers. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, etc. Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0037] In one exemplary embodiment, such as Figure 2 As shown, an emergency decision generation method is provided, which can be applied to... Figure 1 Taking the terminal in the example, the explanation includes the following steps 201 and 202. Wherein:

[0038] Step 201: Process the operational data streams of multiple business units in the coal supply system to obtain an instantaneous coupling matrix; the value of each element in the instantaneous coupling matrix is ​​used to characterize the operational rhythm coupling strength between two business units.

[0039] The coal supply system refers to the coal supply chain, a complex system consisting of multiple links such as production, transportation, transshipment, and consumption. The coal supply system includes various business units, such as coal production units, railway transportation units, port operation units, shipping scheduling units, and power plant consumption units.

[0040] Business units generate operational data in real time, forming operational data streams. Operational data streams are time-series data that characterize the operational rhythm of a business unit, such as production rates or transit volumes sampled at specific time intervals.

[0041] In one possible implementation, the instantaneous coupling matrix is ​​obtained by performing wavelet coherence analysis on the operational data streams of multiple business units.

[0042] Step 202: Determine whether the coal supply system is in an imbalanced resonance mode based on the instantaneous coupling matrix, and if it is determined that the coal supply system is in an imbalanced resonance mode, obtain the target emergency decision based on the prediction model and the objective function; the target emergency decision is used to adjust the operating rhythm of at least one business unit.

[0043] Among them, the misaligned resonance mode refers to the system state that is highly correlated with known faults or inefficiencies.

[0044] In one possible implementation, there exists an offset resonance mode library that stores cluster center vectors corresponding to multiple different offset resonance modes. By converting the instantaneous coupling matrix into a low-dimensional latent vector, the distance between the low-dimensional latent vector and the cluster center vector corresponding to each offset resonance mode is calculated to determine whether the instantaneous coupling matrix is ​​in an offset resonance mode.

[0045] Given that the coal supply system is in an out-of-balance resonance mode, the target emergency decision is obtained based on the prediction model and objective function, resulting in the optimal intervention waveform. This optimal intervention waveform is a continuous vector function, which is decomposed into scalar functions for different controllable business units. Each scalar function is discretized in time to generate a series of specific operation instructions with precise timestamps, which is the target emergency decision.

[0046] For example, if several pre-waveforms involve adjusting the train speed of railway units and the loading and unloading rate of port units, two independent instruction sequences will be generated and issued to the corresponding railway units and port units respectively. The dynamic scheduling mechanism ensures that the execution of these two sets of instructions is synchronized in time or coordinated according to a preset timing sequence, thereby jointly achieving the adjustment of the coal supply system's cycle time.

[0047] The aforementioned emergency decision generation method processes the operational data streams of multiple business units within the coal supply system to obtain an instantaneous coupling matrix. Each element in the instantaneous coupling matrix characterizes the coupling strength of the operational rhythms between two business units. Then, based on the instantaneous coupling matrix, it determines whether the coal supply system is in an imbalanced resonance mode. If the coal supply system is determined to be in an imbalanced resonance mode, a target emergency decision is obtained based on a prediction model and an objective function. This target emergency decision is used to adjust the operational rhythm of at least one business unit. Thus, by using the instantaneous coupling matrix obtained from the operational data streams of the coal supply system, risks to the coal supply system can be anticipated before a failure event occurs. When the coal supply system is determined to be in an imbalanced resonance mode, a target emergency decision is generated. Intervention based on this target emergency decision allows for early intervention before a failure event occurs, effectively avoiding the negative impact spread caused by responding after a failure event has occurred, thereby reducing emergency intervention costs.

[0048] In an exemplary embodiment, the operational data streams of multiple business units in a coal supply system are processed to obtain an instantaneous coupling matrix, including the following steps 2011 and 2012, wherein:

[0049] Step 2011: For the two business units, perform continuous wavelet transform on the operational data streams of the two business units respectively to obtain two wavelet spectra; calculate the wavelet coherence spectrum based on the two wavelet spectra; integrate the wavelet coherence spectrum over a preset operational frequency range to obtain the element values ​​corresponding to the two business units.

[0050] Step 2012: Based on the element values ​​corresponding to every two business units in the multiple business units, obtain the instantaneous coupling matrix.

[0051] In one possible implementation, the operational data flow for any pair of business units and The complex Morlet wavelet was chosen as the mother wavelet function because it has good localization properties in time and frequency. The operational data stream was obtained through continuous wavelet transform. and The corresponding wavelet spectra and Then, the wavelet coherence spectrum is calculated based on the following formula. .

[0052]

[0053] in, For frequency, For time location, This is an operator for smoothing in the time-frequency plane. For complex numbers Taking the conjugate is used to stabilize the coherence calculation results. The range of the coherence spectrum characterizes the linear coupling strength of the two operational data streams at a specific time and frequency point.

[0054] Integrating the wavelet coherence spectrum over a preset operating frequency range yields the instantaneous coupling matrix. elements The operating frequency range corresponds to the cycles with actual business significance in the coal supply system (coal supply chain), such as the daily cycle and shift cycle. This integral operation aggregates the time-frequency related coupling strength information into a single coupling degree index that is only related to time.

[0055] Before explaining how to determine whether a coal supply system is in an out-of-tuned resonance mode and how to generate target emergency decisions, we will first introduce how to construct an efficient resonance mode library and an out-of-tuned resonance mode library.

[0056] The historical instantaneous coupling matrix sequence is calculated by analyzing the historical operational data streams from multiple business units. The historical instantaneous coupling matrix in the historical instantaneous coupling matrix sequence can be referred to as the dynamic fingerprint. Since the historical instantaneous coupling matrix has a high dimension, nonlinear dimensionality reduction techniques, such as variational autoencoders (VAEs), can be used to map it to a low-dimensional latent space. In one example, the historical instantaneous coupling matrix sequence... Each historical instantaneous coupling matrix Input variational autoencoder to obtain the coupling matrix of each historical instant. The low-dimensional latent vector z Multiple historical instantaneous coupling matrices The low-dimensional latent vector z Construct a set of low-dimensional latent vectors .

[0057] Clustering algorithm for low-dimensional latent vector sets Clustering is performed to obtain the clustering results. The clustering algorithm can be a density-based spatial clustering of applications with noise (DBSCAN) algorithm, which can automatically identify several data clusters based on the distribution density of vectors. Each cluster represents a recurring and typical system dynamics pattern.

[0058] The clustering results are correlated with stored historical business performance data (e.g., recorded system failure events, transportation efficiency, operating costs, etc.). If the historical instantaneous coupling matrix in a certain cluster is highly correlated with known failures or inefficiencies in time, then the cluster is labeled and stored in the misalignment resonance mode library. Conversely, if a cluster is associated with a long-term, efficient, and stable operating state, it is stored in the efficient resonant mode library. Thus, the resulting offset resonance mode library stores the center vectors of various clusters highly correlated with known faults or inefficient events. These clusters are called offset resonance modes. Similarly, the resulting high-efficiency resonance mode library stores the center vectors of various clusters related to long-term efficient and stable operating states. These clusters are called high-efficiency resonance modes. Therefore, the offset resonance mode library stores the center vectors corresponding to various offset resonance modes, and the high-efficiency resonance mode library stores the center vectors corresponding to various high-efficiency resonance modes.

[0059] In one exemplary embodiment, such as Figure 3 The diagram illustrates a method for determining whether a coal supply system is in an off-mode resonance state. The method, based on the instantaneous coupling matrix, includes:

[0060] The instantaneous coupling matrix is ​​dimensionality reduced to obtain the corresponding low-dimensional latent vector. Based on the low-dimensional latent vector, it is determined whether there is an offset resonance mode in the offset resonance mode library that matches the low-dimensional latent vector; if so, the coal supply system is determined to be in an offset resonance mode; if not, the coal supply system is determined not to be in an offset resonance mode.

[0061] In one possible implementation, a nonlinear dimensionality reduction technique is used to reduce the dimensionality of the instantaneous coupling matrix, resulting in a low-dimensional latent vector corresponding to the instantaneous coupling matrix. In one example, the instantaneous coupling matrix is ​​input into a variational autoencoder to obtain the low-dimensional latent vector corresponding to the instantaneous coupling matrix.

[0062] Calculate the distance between the low-dimensional latent vector corresponding to the instantaneous coupling matrix and the center vector corresponding to each mismatched resonance mode in the mismatched resonance mode library; if there is a mismatched resonance mode with a distance less than a preset distance threshold, it is determined that there is a mismatched resonance mode in the mismatched resonance mode library that matches the low-dimensional latent vector; if there is no mismatched resonance mode with a distance less than the preset distance threshold, it is determined that there is no mismatched resonance mode in the mismatched resonance mode library that matches the low-dimensional latent vector.

[0063] When there is an imbalanced resonance mode with a distance less than a preset distance threshold, it indicates that the current state of the coal supply system is approaching a known dangerous state. Therefore, it is necessary to trigger the intervention process, that is, to generate a target emergency decision to intervene in the coal supply system.

[0064] In one exemplary embodiment, such as Figure 4 The diagram illustrates a method for generating target emergency decisions, which obtains target emergency decisions based on an instantaneous coupling matrix, including steps 401 to 404, wherein:

[0065] Step 401: Select the target resonant mode from multiple high-efficiency resonant modes in the high-efficiency resonant mode library.

[0066] It should be noted that one or more target resonant modes can be selected from multiple high-efficiency resonant modes, if multiple modes are selected.

[0067] Step 402: Generate the intervention waveform and input the intervention waveform into the prediction model to obtain the predicted system state.

[0068] Step 403: Solve the objective function based on the target resonant mode and the predicted system state. Re-execute the step of generating the intervention waveform based on the solution of the objective function until the stopping condition is met, and obtain the optimal intervention waveform.

[0069] In other words, the goal of generating the optimal intervention waveform by solving an optimal control problem is to find an intervention waveform that minimizes a comprehensive objective function. The optimal intervention waveform can be used within the preset intervention time domain. The system state is guided from the current danger zone to the vicinity of the target resonant mode, while minimizing risk and intervention costs.

[0070] The prediction model can be a surrogate model based on graph neural networks or long short-term memory networks. The predicted system state is actually a vector that can be used to characterize whether the coal supply system is in an off-mode resonance or an efficient resonant mode. Inputting the intervention waveform into the prediction model to obtain the predicted system state is also known as intervention echo prediction.

[0071] objective function The expression is as follows:

[0072]

[0073] in, This function is used to quantify the predicted system state. The degree of danger; This function is used to quantify the predicted system state. The degree of ideality; This function is used to quantize the waveform of the intervention. The inherent costs; risk costs. System state used for penalty prediction The resonance gain term is close to any known mistuned resonant mode (any mistuned resonant mode in the mistuned resonant mode library). The system state used for reward prediction approaches the target resonant mode. Controlling cost items Used to punish the amplitude and drastic changes of the intervention waveform itself; , and These are the weighting coefficients for the risk cost item, the resonance benefit item, and the control cost item, respectively.

[0074] In one example, control cost item The expression is as follows:

[0075]

[0076] in, and This is a preset positive definite weight matrix; It is an intervention waveform; Intervention waveform Transpose of; Intervention waveform The first derivative with respect to time is also a vector.

[0077] The objective function can be solved by treating the intervention waveform as particles and using a particle swarm optimization algorithm; alternatively, it can be solved using numerical optimization algorithms, such as gradient-based optimization methods. The stopping condition can be satisfying the number of iterations or other conditions, which are not limited here.

[0078] Step 404: Process the optimal intervention waveform to obtain the target emergency decision.

[0079] Among them, the optimal intervention wave Since it is a continuous function, it needs to be discretized over time to be converted into a series of timestamped setpoints.

[0080] For example, with a fixed time step For the optimal intervention wave Sampling is performed to generate a sequence of instructions. ,in, It is the corresponding number in the waveform vector. The components of a controllable variable.

[0081] It should be noted that when an optimal intervention waveform involves the coordinated operation of multiple business units, dynamic orchestration is required. This involves generating a unique dynamic behavioral instruction sequence for each relevant business unit and ensuring that these sequences have a unified time base when they are issued. This allows multiple business units to operate synchronously (executing their respective instructions at the same time) or to perform sequential operations according to a preset precise timing sequence, thereby jointly achieving coordinated adjustment of the system's beat.

[0082] In one exemplary embodiment, such as Figure 5 The diagram illustrates a flowchart of a prediction model update method; the method further includes steps 501 to 503, wherein:

[0083] Step 501: Obtain the actual evolution trajectory based on multiple instantaneous coupling matrices corresponding to the preset time period.

[0084] Step 502: Obtain the predicted evolution trajectory based on the predicted system state corresponding to the preset time period.

[0085] It is understandable that executing steps 201 and 202 in real time will yield multiple instantaneous coupling matrices and multiple predicted system states. Of course, the predicted system states here refer to the predicted system states corresponding to the optimal intervention waveform. The instantaneous coupling matrices used here are those of the coal supply system in the misaligned resonance mode.

[0086] It should be noted that the actual evolution trajectory is a series of low-dimensional latent vectors obtained by reducing the dimensionality of multiple instantaneous coupling matrices. As explained above, the predicted system state is also a vector; therefore, the predicted evolution trajectory is also composed of multiple vectors.

[0087] Step 503: Based on the difference between the actual evolutionary trajectory and the predicted evolutionary trajectory, the parameters of the prediction model are adjusted to obtain the updated prediction model.

[0088] In one example, the mean squared error between the actual evolutionary trajectory and the predicted evolutionary trajectory is calculated to obtain the loss value. The prediction model is then tuned based on this loss value using the backpropagation algorithm to improve the prediction accuracy of the prediction model.

[0089] Once the updated prediction model is obtained, it can be used to predict the system state based on the intervention waveform during the next generation of target emergency decisions.

[0090] This implementation allows for continuous optimization of the prediction model, thereby improving its prediction accuracy.

[0091] In an exemplary embodiment, the method further includes: squaring the values ​​of each element in the instantaneous coupling matrix to obtain multiple squared values; performing a weighted summation of the multiple squared values ​​to obtain a target value; and taking the square root of the target value to obtain a system resonance index; the system resonance index is used to quantify the vulnerability of the coal supply system.

[0092] The data can be expressed using formulas as follows:

[0093]

[0094] in, for The system resonance index at a given moment; This represents the total number of business units in the coal supply system. for The instantaneous coupling matrix at time t is the first Line number The elements of the column represent business units. With business units The strength of the coupling between their operational rhythms; The preset weighting coefficients represent the business units. and The degree to which the coupling relationship between the components affects the overall stability of the system.

[0095] By quantifying the vulnerability of the coal supply system through the system resonance index, early warning can be achieved based on the system resonance index, and it can also be used as an indicator to trigger the generation of target emergency decisions.

[0096] In one example, after obtaining the system resonance index, if the system resonance index is greater than or equal to the trigger threshold, it is determined that the coal supply system is in an out-of-balance resonance mode, and an alarm message is output to inform the substation staff that there is a problem with the coal supply system.

[0097] In one embodiment, such as Figure 6 As shown, an emergency decision generation system is provided, which includes multiple intelligent agents, a dynamic state monitoring module, a bimodal knowledge base learning module, an intervention decision module, and a strategy execution module.

[0098] Multiple intelligent agents, each corresponding to a different business unit in the coal supply system, are equipped with the ability to acquire real-time operational data from their corresponding business unit and output operational data streams.

[0099] The dynamic state monitoring module connects to multiple intelligent agents through a communication interface. The dynamic state monitoring module receives the operational data streams output by the multiple intelligent agents; and calculates the instantaneous coupling matrix based on the received operational data streams, and further calculates the system resonance index based on the instantaneous coupling matrix.

[0100] The dual-modal knowledge base learning module is connected to the dynamic state monitoring module. It receives and continuously collects the instantaneous coupling matrix generated by the dynamic state monitoring module. This module processes the historical instantaneous coupling matrix sequence through an unsupervised learning algorithm and associates it with historical business performance data, thereby constructing and storing two types of system dynamic modes: the misaligned resonance mode related to historical system failures or inefficient operation, and the efficient resonance mode related to historical efficient and stable system operation.

[0101] The intervention decision module, connected to the dynamic state monitoring module and the bimodal knowledge base learning module, receives the real-time instantaneous coupling matrix and system resonance index from the dynamic state monitoring module, and obtains the learned misaligned resonance mode and efficient resonance mode from the bimodal knowledge base learning module. Based on these input information, the intervention decision module generates one or more intervention waveforms for adjusting the operational rhythm of at least one agent (i.e., business unit).

[0102] The strategy execution module, connected to the intervention decision module and multiple agents, receives the intervention waveform generated by the intervention decision module. The function of the strategy execution module is to decompose the mathematical intervention waveform into specific, executable dynamic behavior instruction sequences and distribute these instruction sequences to one or more corresponding agents.

[0103] In the overall workflow of the system, the operational data streams generated by multiple agents are first sent to the dynamic state monitoring module for processing to generate a quantitative description of the system state (instantaneous coupling matrix and system resonance index). This quantitative description is sent to the intervention decision module for real-time decision-making on the one hand, and to the bimodal knowledge base learning module for long-term learning on the other hand.

[0104] The intervention decision-making module combines real-time status and historical knowledge to generate an intervention waveform and transmit it to the strategy execution module. The strategy execution module converts the intervention waveform into instructions and issues them to the corresponding intelligent agents for execution. After the intelligent agents execute the instructions, their operating rhythm changes, thereby generating new dynamic data streams and forming a closed-loop monitoring, decision-making, execution and feedback process.

[0105] Based on the above, the entire early warning, decision-making, execution, and outcome event can be recorded as a complete sample and fed into the bimodal knowledge base learning module. The bimodal knowledge base learning module will use these new samples to iteratively update the misaligned resonance mode library and the efficient resonance mode library in subsequent learning cycles, thereby continuously improving the system's knowledge system.

[0106] By calculating the instantaneous coupling matrix that characterizes the coupling strength of the operational rhythm between various business units in real time, and further synthesizing the system resonance index, a quantitative assessment of the system's vulnerability is achieved. This approach can identify potential risks caused by operational rhythm imbalances between multiple links before a specific failure event occurs, realizing a shift from event-driven passive response to state-driven proactive early warning, thus gaining a time window to avoid cascading risks.

[0107] By constructing the generation of intervention strategies as an optimal control problem, the goal is to find an optimal intervention waveform that can guide the system state from the danger zone to the target resonance state. Compared with the extensive adjustment based on static rules or human experience in the existing technology, this approach can generate precise, dynamic and cost-controllable intervention schemes, ensuring the scientific nature and efficiency of the emergency intervention measures themselves, and avoiding secondary disturbances caused by improper intervention.

[0108] By implementing closed-loop adaptive optimization of decision-making and execution, on the one hand, the system can learn autonomously from historical data and continuously improve its understanding of dangerous and ideal states; on the other hand, the dynamic orchestration mechanism ensures the collaborative execution of multiple agents, which enables the entire system to have the ability to continuously learn and self-evolve, and to continuously adapt to changes in the external environment and internal state, so as to achieve long-term stability and efficient operation of the entire supply chain.

[0109] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0110] Based on the same inventive concept, this application also provides an emergency decision generation device for implementing the emergency decision generation method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the emergency decision generation device provided below can be found in the limitations of the emergency decision generation method described above, and will not be repeated here.

[0111] In one exemplary embodiment, such as Figure 7 As shown, an emergency decision generation device 700 is provided, comprising: a first determining module 701 and a second determining module 702, wherein:

[0112] The first determining module 701 is used to process the operational data streams of multiple business units in the coal supply system to obtain an instantaneous coupling matrix; the value of each element in the instantaneous coupling matrix is ​​used to characterize the operational rhythm coupling strength between two business units.

[0113] The second determining module 702 is used to determine whether the coal supply system is in an out-of-balance resonance mode based on the instantaneous coupling matrix, and when it is determined that the coal supply system is in an out-of-balance resonance mode, to obtain a target emergency decision based on the prediction model and the objective function; the target emergency decision is used to adjust the operating rhythm of at least one business unit.

[0114] In one embodiment, the second determining module 702 is specifically used to select a target resonant mode from multiple high-efficiency resonant modes in the high-efficiency resonant mode library; input the instantaneous coupling matrix into the prediction model to obtain the predicted system state; solve the objective function based on the target resonant mode and the predicted system state to obtain the optimal intervention waveform; and process the optimal intervention waveform to obtain the target emergency decision.

[0115] In one embodiment, the objective function includes a risk cost term, a resonance benefit term, and a control cost term, wherein: the risk cost term is used to penalize the predicted system state for approaching any misaligned resonance mode in the misaligned resonance mode library; the resonance benefit term is used to reward the predicted system state for approaching the target resonance mode; and the control cost term is used to penalize the amplitude and drastic change of the intervention waveform.

[0116] In one embodiment, the device further includes a module for obtaining the actual evolution trajectory based on multiple instantaneous coupling matrices corresponding to a preset time period; obtaining the predicted evolution trajectory based on the predicted system state corresponding to the preset time period; and adjusting the parameters of the prediction model based on the difference between the actual evolution trajectory and the predicted evolution trajectory to obtain an updated prediction model.

[0117] In one embodiment, the first determining module 701 is specifically used to perform continuous wavelet transform on the operational data streams of the two business units respectively to obtain two wavelet spectra; calculate the wavelet coherence spectrum based on the two wavelet spectra; integrate the wavelet coherence spectrum over a preset operational frequency range to obtain the element values ​​corresponding to the two business units; and obtain the instantaneous coupling matrix based on the element values ​​corresponding to every two business units in the multiple business units.

[0118] In one embodiment, the second determining module 702 is specifically used to perform dimensionality reduction processing on the instantaneous coupling matrix to obtain the low-dimensional latent vector corresponding to the instantaneous coupling matrix; based on the low-dimensional latent vector, determine whether there is an offset resonance mode in the offset resonance mode library that matches the low-dimensional latent vector; if so, determine that the coal supply system is in an offset resonance mode; if not, determine that the coal supply system is not in an offset resonance mode.

[0119] In one embodiment, the device further includes a third determining module, used to square the values ​​of each element in the instantaneous coupling matrix to obtain multiple squared values; to perform a weighted summation of the multiple squared values ​​to obtain a target value; to take the square root of the target value to obtain a system resonance index; the system resonance index is used to quantify the vulnerability of the coal supply system.

[0120] Each module in the aforementioned emergency decision generation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0121] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 8As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements an emergency decision-making method. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0122] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0123] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any one of the above method embodiments.

[0124] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any of the above method embodiments.

[0125] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the method described in any one of the above method embodiments.

[0126] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0127] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0128] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An emergency decision generation method, characterized in that, The method includes: The operational data streams of multiple business units in the coal supply system are processed to obtain an instantaneous coupling matrix; the value of each element in the instantaneous coupling matrix is ​​used to characterize the operational rhythm coupling strength between two of the business units. The instantaneous coupling matrix is ​​used to determine whether the coal supply system is in an imbalanced resonance mode. If the coal supply system is determined to be in the imbalanced resonance mode, a target emergency decision is obtained based on the prediction model and the objective function. The target emergency decision is used to adjust the operating rhythm of at least one of the business units.

2. The method according to claim 1, characterized in that, The process of obtaining the target emergency decision based on the prediction model and objective function includes: Select the target resonant mode from multiple high-efficiency resonant modes in the high-efficiency resonant mode library; An intervention waveform is generated and input into a prediction model to obtain the predicted system state; The objective function is solved based on the target resonant mode and the predicted system state. The step of generating the intervention waveform is repeated based on the solution of the objective function until the stopping condition is met, and the optimal intervention waveform is obtained. The optimal intervention waveform is processed to obtain the target emergency decision.

3. The method according to claim 2, characterized in that, The objective function includes a risk cost term, a resonance benefit term, and a control cost term, wherein: The risk cost term is used to penalize the predicted system state for being close to any mismatched resonance mode in the mismatched resonance mode library; The resonance reward term is used to reward the predicted system state for approaching the target resonance mode; The control cost item is used to penalize the amplitude and drastic changes of the intervention waveform.

4. The method according to claim 2, characterized in that, The method further includes: The actual evolution trajectory is obtained based on the multiple instantaneous coupling matrices corresponding to the preset time period; Based on the predicted system state corresponding to the preset time period, the predicted evolution trajectory is obtained; Based on the difference between the actual evolutionary trajectory and the predicted evolutionary trajectory, the prediction model is adjusted to obtain an updated prediction model.

5. The method according to claim 1, characterized in that, The process of processing the operational data streams of multiple business units in the coal supply system to obtain an instantaneous coupling matrix includes: For the two business units, continuous wavelet transform is performed on the operational data streams of the two business units respectively to obtain two wavelet spectra; a wavelet coherence spectrum is calculated based on the two wavelet spectra; the wavelet coherence spectrum is integrated over a preset operational frequency range to obtain the element values ​​corresponding to the two business units. The instantaneous coupling matrix is ​​obtained based on the element values ​​corresponding to every two of the plurality of business units.

6. The method according to claim 1, characterized in that, The step of determining whether the coal supply system is in an out-of-mode resonance based on the instantaneous coupling matrix includes: The instantaneous coupling matrix is ​​reduced in dimension to obtain the low-dimensional latent vector corresponding to the instantaneous coupling matrix; Based on the low-dimensional latent vector, determine whether there is an off-mode resonance in the off-mode resonance library that matches the low-dimensional latent vector; If so, then the coal supply system is determined to be in an out-of-tuned resonance mode; If not, then it is determined that the coal supply system is not in an out-of-tuned resonance mode.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: Squaring each element in the instantaneous coupling matrix yields multiple squared values. The target value is obtained by weighted summation of the multiple squared values; The square root of the target value is used to obtain the system resonance index; the system resonance index is used to quantify the vulnerability of the coal supply system.

8. An emergency decision generation device, characterized in that, The device includes: The first determining module is used to process the operational data streams of multiple business units in the coal supply system to obtain an instantaneous coupling matrix; the value of each element in the instantaneous coupling matrix is ​​used to characterize the operational rhythm coupling strength between two business units. The second determining module is used to determine whether the coal supply system is in an imbalanced resonance mode based on the instantaneous coupling matrix, and if it is determined that the coal supply system is in the imbalanced resonance mode, to obtain a target emergency decision based on a prediction model and an objective function; the target emergency decision is used to adjust the operating rhythm of at least one of the business units.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images