Methods and systems for power grid risk assessment and collaborative optimization of emergency resource allocation under extreme operating conditions

By acquiring power grid operation data and meteorological forecast data, an evaluation cloud model was established to quantify the matching degree between emergency resources and risk levels. A multi-objective optimization model was constructed, which solved the problem of quantitative assessment of resource response capabilities and regional risk levels in emergency resource allocation, and realized efficient collaborative allocation and optimized configuration of cross-regional resources.

CN123073808APending Publication Date: 2026-11-10SHANDONG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611226521.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-13
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing technologies lack quantitative assessments of resource response capabilities and regional risk levels in emergency resource allocation, and cross-regional resource coordination lacks systematic optimization, making it difficult to achieve optimal resource allocation in multi-regional scenarios.

Method used

By acquiring power grid operation data and meteorological forecast data, an evaluation cloud model is established to quantify the matching degree between emergency resources and risk levels. A multi-objective optimization model is constructed to optimize the allocation of emergency resources in different regions, meeting the constraint of no mismatch between traffic and normal areas.

Benefits of technology

It enables accurate identification of power grid risks and efficient coordination and allocation of emergency resources under extreme operating conditions, ensuring improved matching in high-risk areas and no mismatch in normal areas, thus optimizing resource allocation efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123073808A_ABST
    Figure CN123073808A_ABST
Patent Text Reader

Abstract

This invention proposes a method and system for power grid risk assessment and collaborative optimization of emergency resource allocation under extreme operating conditions, belonging to the field of power system emergency management technology. The method includes: acquiring power grid operation data and meteorological forecast data for each region to obtain the time-domain trajectory of electrical quantities in each region; conducting a comprehensive assessment of each region based on the acquired time-domain trajectory of electrical quantities to determine the final risk level of each region; establishing an evaluation cloud model; quantitatively assessing the matching degree between emergency resources and risk levels based on the cloud model; calculating the regional risk comprehensive matching degree index; and identifying severely mismatched regions. For regions with severely mismatched emergency resources and risk levels, a multi-objective optimization model is constructed with the objectives of maximizing the matching degree and minimizing the movement distance of emergency resources. Under the premise of satisfying traffic constraints and ensuring no mismatch in normal regions, the model optimizes the allocation of emergency resources among different regions and proposes a hierarchical solution method to determine the type, capacity, and path of resource allocation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of power system emergency management technology, and in particular relates to a method and system for power grid risk assessment and collaborative optimization of emergency resource allocation under extreme operating conditions. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] The increasing frequency of extreme weather events has led to a more complex and severe operating environment for power systems. Under extreme conditions, grid equipment may suffer widespread damage, resulting in voltage drops, frequency deviations, and power angle instability. In severe cases, this can even lead to catastrophic consequences such as system disconnection, generator loss of synchronism, and large-scale load shedding. How to quickly identify sudden risks, scientifically assess risk levels, and efficiently allocate emergency resources has become a critical issue that urgently needs to be addressed in the field of power system emergency management.

[0004] Currently, numerous studies have focused on the issue of power grid emergency resource allocation. However, existing technologies still have the following shortcomings: Insufficient methods for assessing the matching degree between emergency resources and risk levels: Current technologies often rely on fixed resource demand forecasts or simple prioritization, lacking quantitative assessment methods to evaluate the matching degree between resource response capabilities and the actual risk level of a region. Emergency resources in different regions vary significantly in terms of resource type, response time, repair speed, available capacity, and current status. How to scientifically measure the matching degree between emergency resource capabilities and risk levels in each region, and accurately identify under-matched areas, is a problem that current technologies have not adequately addressed.

[0005] The lack of systematic optimization in cross-regional resource coordination and allocation: Existing emergency resource dispatching methods mostly focus on resource allocation within a single region or simple material transportation route planning, lacking a systematic optimization approach that comprehensively considers constraints such as matching high-risk areas to meet standards, avoiding mismatches in normal areas, and transportation timeliness at the cross-regional level. Especially in scenarios where multiple regions simultaneously face different risk levels, existing technologies have yet to provide an effective solution for ensuring the emergency needs of high-risk areas while preventing mismatches in normal areas due to resource outflows, thus achieving optimal global resource allocation. Summary of the Invention

[0006] To overcome the shortcomings of the prior art, this invention provides a method and system for power grid risk assessment and emergency resource coordination and allocation under extreme operating conditions, enabling accurate identification of sudden power grid risks and efficient coordination and allocation of emergency resources under extreme operating conditions.

[0007] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: Firstly, it discloses methods for power grid risk assessment and coordinated optimization of emergency resource allocation under extreme operating conditions, including: By acquiring power grid operation data and meteorological forecast data for each region, the time-domain trajectory of electrical quantities in each region is obtained. Based on the acquired time-domain trajectory of electrical quantities, a comprehensive assessment of each region is conducted to determine the final risk level of each region. Establish an evaluation cloud model, quantitatively assess the matching degree between emergency resources and risk levels based on the cloud model, calculate the regional risk comprehensive matching degree index, and identify areas with serious under-match; For regions where emergency resources and risk levels are severely mismatched, a multi-objective optimization model is constructed with the goals of maximizing the matching degree and minimizing the movement distance of emergency resources. Under the premise of satisfying traffic constraints and ensuring no mismatch in normal areas, the model optimizes the allocation of emergency resources in different regions and proposes a hierarchical solution method to determine the type, capacity, and path of resource allocation.

[0008] As a further technical solution, based on the acquisition of power grid operation data and meteorological forecast data in various regions, when a power grid emergency risk warning signal is detected, extreme operating condition evolution is simulated based on time-domain simulation for different regional power grids to obtain the time-domain trajectory of electrical quantities in each region, including voltage trajectory, frequency trajectory and power angle trajectory. Based on the acquired electrical quantity time-domain trajectory and in conjunction with national standards, a comprehensive assessment of each region is conducted from four dimensions: line overload safety margin, power angle stability level, frequency stability level, and voltage stability level.

[0009] As a further technical solution, during the comprehensive assessment, if any dimension is rated as urgent, the overall risk level is severe risk, corresponding to system disconnection, generator loss of synchronism, system instability, large-scale load loss, severe frequency or voltage deviation, and severe overload of multiple lines. If there is a dimension that is rated as alarm but there is no emergency dimension, the overall risk level is general risk, corresponding to consequences such as severe frequency or voltage deviation, severe overload of multiple lines, etc. If all dimensions are rated as normal, the overall risk level is no risk, corresponding to voltage, frequency, and line load all being within the specified range.

[0010] As a further technical solution, in terms of frequency stability, two forms are used to calculate the power angle stability margin: one based on the ratio of deceleration area to acceleration area and the other based on the difference between deceleration area and acceleration area. These are respectively applicable to comprehensive evaluation scenarios that need to consider both acceleration energy and potential deceleration capability, and to rapid evaluation scenarios that need to directly quantify the net difference between acceleration area and deceleration area.

[0011] As a further technical solution, regarding frequency stability, the stability margins of low-frequency deviation and high-frequency deviation are calculated separately, and the overall frequency stability level is taken as the smaller value between the stability margin of low-frequency deviation and the stability margin of high-frequency deviation, so as to reflect the stability margin of the system in the most unfavorable frequency deviation direction.

[0012] As a further technical solution, in terms of voltage stability level, the voltage stability margin index is used to characterize the degree to which the system node voltage deviates from the reference voltage. The larger the value, the higher the voltage stability level.

[0013] As a further technical solution, when establishing the evaluation cloud model, the response capability of emergency resources in each region is quantified from five dimensions: resource type, action response time, repair operation speed, available capacity, and current status, and various risk levels are established. g The following matching indicators q Evaluation criteria for cloud models: In the formula, g The risk level is represented by a value of severe risk, moderate risk, or no risk. q To match the metrics, five dimensions are used: resource type, action response time, emergency repair speed, available capacity, and current status. Risk level g Lowering the target q The expected value of the cloud model reflects the ideal central value of the indicator at this level; The corresponding entropy reflects the degree of fuzziness in the index value; This corresponds to hyperentropy, reflecting the randomness of the index values. Risk level g Matching Indicators q The evaluation criteria for cloud models.

[0014] As a further technical solution, resource type is used to reflect the completeness of the types of emergency resources allocated in the region, and a score is given based on the degree of matching between the actual resource types allocated in the region and the required resource types. Action response time is used to reflect the degree of match between the actual response time of emergency resources in the area from receiving an instruction to arriving at the scene and the allowed response time; The speed of emergency repair operations reflects the ability of the area's emergency repair teams to complete emergency repair tasks within the permitted time. Available capacity reflects the degree of matching between the available power supply resources and the required power supply capacity in the area; The availability status index is used to reflect the current availability status of regional emergency resources. It takes into account factors such as resource integrity rate, on-duty rate, and call-up status. The higher the value, the better the resource availability.

[0015] As a further technical solution, based on the actual emergency resource data of a certain area, the degree of matching and the certainty of the risk level of that area are calculated, reflecting the degree of risk level of that area. q The measured values ​​of each matching indicator belong to the risk level. G r The degree to which the target cloud is met; the larger the value, the closer the indicator is to meeting the target level. Then, the overall risk matching degree of the region is calculated. The higher the value, the higher the degree of matching between the emergency resources and the risk level of the region. To determine the matching level, four matching cloud levels are defined: severely undermatched, undermatched, basic match, and good match. Calculate the degree of certainty of the overall risk matching degree for each matching level in this region; The matching level determination result for this region is one of the following: severely undermatched, undermatched, basically matched, or well matched. That is, the matching level with the highest certainty is taken as the final matching level for this region.

[0016] As a further technical solution, in the multi-objective optimization model, the first optimization objective is to maximize the matching degree between emergency resources and risk levels in each region after adjustment. Optimization objective two: Minimize the movement path cost of cross-regional emergency resource allocation; The constraints include: adjusted resource capacity constraints, adjusted resource matching degree constraints, matching compliance constraints in high-risk areas, no mismatch constraints in normal areas, allocation time constraints, and linkage constraints between allocation variables.

[0017] Secondly, a system for power grid risk assessment and emergency resource collaborative optimization and allocation under extreme operating conditions was disclosed, including: The module for determining the final risk level of each region is configured to: acquire power grid operation data and weather forecast data for each region, obtain the time-domain trajectory of electrical quantities for each region, conduct a comprehensive assessment of each region based on the acquired time-domain trajectory of electrical quantities, and determine the final risk level of each region; The matching degree quantitative assessment module is configured to: establish an evaluation cloud model, quantitatively assess the matching degree between emergency resources and risk levels based on the cloud model, calculate the regional risk comprehensive matching degree index, and identify severely mismatched areas; The resource allocation module is configured to: for areas where emergency resources and risk levels are severely mismatched, construct a multi-objective optimization model with the objectives of maximizing the matching degree and minimizing the movement distance of emergency resources, optimize the allocation of emergency resources in different areas under the premise of satisfying traffic constraints and ensuring no mismatch in normal areas, and propose a hierarchical solution method to determine the type, capacity and path of resource allocation.

[0018] The above one or more technical solutions have the following beneficial effects: This invention provides a complete emergency decision-making solution that integrates risk identification, level assessment, matching measurement, and coordination optimization. It achieves risk classification by refining the time-domain evolution trajectory of electrical quantities in various regional power grids under extreme conditions, integrating the national standard multi-dimensional stability evaluation system, introducing a cloud model to quantify the matching degree between emergency resource response capabilities and risk levels, and constructing a multi-objective cross-regional resource coordination optimization model based on this.

[0019] 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

[0020] 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.

[0021] Figure 1 Heatmaps showing the initial risk levels for each region; Figure 2 Heatmap of risk levels for each region after risk changes; Figure 3 A flowchart for quantitative evaluation of matching degree based on cloud model; Figure 4 To determine the risk level and overall compatibility of each region before allocation; Figure 5 Flowchart for optimizing the allocation of cross-regional emergency resources; Figure 6 To ensure the matching degree of resources in each region after allocation; Figure 7 To compare the matching degree of resources in each region before and after the allocation. Detailed Implementation

[0022] It should be noted that the following detailed descriptions are 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.

[0023] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0024] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0025] Terminology Explanation: Extreme weather events refer to disastrous weather phenomena that pose a significant threat to the safe and stable operation of the power system, including but not limited to typhoons, rainstorms (snowstorms), ice storms, cold waves, and sandstorms.

[0026] Extreme operating conditions: refers to the state in which the operating parameters of the power system deviate significantly from the normal safe range, specifically manifested as one or more of the following situations: (1) the node voltage drops below 0.8 pu (per unit value) or exceeds the allowable operating range; (2) the system frequency deviates from the rated value by 50 Hz and exceeds the allowable deviation limit; (3) the generator power angle difference approaches or exceeds the static / transient stability limit; (4) the line load rate exceeds the rated current carrying capacity and the overload safety margin is negative; (5) the above-mentioned multiple unfavorable operating conditions occur in combination.

[0027] Risk identification: Based on the extreme operating condition evolution simulation in the time domain, the voltage, frequency, and power angle trajectories are obtained. Then, in combination with national standards, a comprehensive assessment is conducted from four dimensions: line overload safety margin, power angle stability level, frequency stability level, and voltage stability level, and the final risk level of each region is determined.

[0028] "Achievement Cloud" is a visual term for the cloud model of evaluation standards for matching indicators under each risk level.

[0029] Example 1 This embodiment discloses a method for power grid risk assessment and collaborative optimization of emergency resource allocation under extreme operating conditions, including: Step 1: Real-time access to the operation monitoring data and extreme weather forecast information of the power grid in each region. First, the electrical quantity trajectory of each region under extreme operating conditions is simulated in the time domain. Based on national standards, the risk level is assessed from multiple dimensions, and each region is divided into three levels: severe risk, general risk, and no risk.

[0030] Meteorological data serves as the early warning trigger signal and the boundary condition for time-domain simulation, while operational monitoring data provides the initial state of the simulation and directly serves as the core input for multi-dimensional risk level assessment. After being integrated at the time-domain simulation layer, the two sequentially drive risk level determination, resource matching degree assessment (combined with the availability status indicators in the operational data), and cross-regional coordinated optimization and allocation. Furthermore, the entire process is automatically rerun when meteorological data is updated, ensuring that identification and allocation are always based on the latest operating conditions.

[0031] Step 2: Quantify the response capabilities of emergency resources in each region from five dimensions, use cloud models to calculate the comprehensive matching degree between the emergency resource capabilities of each region and its risk level, and identify severely mismatched and undermatched regions.

[0032] Step 3: For the aforementioned under-matched areas, construct a multi-objective optimization model to coordinate and allocate resources across regions. Under the premise of ensuring that high-risk areas meet the matching standards and normal areas do not suffer from mismatch, determine the optimal resource allocation plan.

[0033] Step 4: If large-scale mismatch still exists after coordination, the mechanism for calling upon emergency resources from outside the region will be activated as a supplement.

[0034] In one implementation example, regarding the sudden risk classification step in step one, which is based on multi-dimensional electrical quantity trajectory extrapolation, when a sudden risk warning signal for the power grid is detected, extreme operating condition evolution extrapolation is performed based on time-domain simulation for different regions of the power grid to obtain the time-domain trajectory of electrical quantities in each region, including voltage trajectory. U ( t ), frequency trajectory f ( t ) and work angle trajectory δ ( t Based on the acquired electrical quantity time-domain trajectory, and in accordance with the national standard GB / T 40585-2021, a comprehensive evaluation of each region is conducted from four dimensions: line overload safety margin, power angle stability level, frequency stability level, and voltage stability level.

[0035] The aforementioned power grid emergency risk warning signal is jointly driven by extreme weather forecast information and power grid operation monitoring data: the weather forecast provides advance prediction and is triggered when parameters such as wind speed and icing exceed the limit, while the operation data provides real-time perception and is triggered when parameters such as voltage and frequency exceed the limit. The two types of data work together to determine the generation of the signal. Specifically, the signal is determined by one of the following two methods: (1) when parameters such as wind speed and icing thickness in the extreme weather forecast information exceed the equipment tolerance threshold, it is triggered by weather prediction; (2) when electrical quantities such as voltage, frequency, power angle, or line power flow in the power grid operation monitoring data deviate from the safe operating range, it is triggered by real-time monitoring.

[0036] Obtaining the time-domain trajectory of electrical quantities in each region specifically includes: establishing a simulation model based on the real-time topology of the power grid and operation monitoring data, setting initial operating conditions, and then applying anticipated disturbances (such as short circuits, line breaks, and generator disconnection) to the corresponding regions based on extreme weather forecast information. Through time-domain simulation, the differential equations and algebraic equations describing the dynamic behavior of the power system are solved simultaneously at small time steps. The operating state of the system at each moment during the occurrence, duration, and recovery of the fault is calculated step by step, thereby completely reproducing the evolution process of extreme operating conditions. Finally, the time-domain trajectory of electrical quantities such as node voltage, frequency, and power angle in each region is output, providing a data foundation for the subsequent risk level assessment in four dimensions.

[0037] The voltage trajectory U(t) is used to extract the actual node voltage V and substitute it into the voltage stability margin formula to assess the voltage stability level; the frequency trajectory f(t) is used to extract transient frequency extrema and substitute them into the frequency stability level formula to calculate the comprehensive frequency stability margin; the power angle trajectory δ(t) is used to extract power angle extrema and substitute them into the power angle stability margin formula. These three are used as inputs for three independent dimensions, together with the line overload safety margin, to form the basis for a four-dimensional assessment. Finally, the overall risk level of each area is determined comprehensively according to the principle of the most severe level.

[0038] Regarding the safety margin for line overload, it is calculated using the following formula: (1) In the formula, η line This is for the safety margin of line overload; I real This represents the current load rate of the line. I max This represents the maximum allowable load rate for the line.

[0039] when If the condition is deemed normal, it indicates that the line load rate is within a safe range. If this is detected as an alarm, it indicates that the line load is too high and needs attention. If the situation is deemed an emergency, it indicates that the line is operating beyond its rated capacity and faces the risk of overload tripping.

[0040] Regarding the stability level of the work angle, the work angle stability margin is used as the key factor. η Characterization: Based on the equal area rule, this example uses the following two methods to calculate the power angle stability margin: 1. Based on the ratio of deceleration area to acceleration area: (2) In the formula, η For the stability margin of the work angle; A dec,opt This represents the potential deceleration area. A inc To accelerate the area. Accelerated area A inc and deceleration area A dec,pot Based on the equal area rule, the acceleration area is obtained by integrating the generator's power angle characteristic curve, rather than through direct measurement. The acceleration area is calculated by applying the difference P between mechanical power and electromagnetic power during the period from fault occurrence to fault clearing. m -P e The acceleration energy accumulated in the rotor during the fault is obtained by integrating over the power angle δ; the deceleration area is obtained by integrating over the difference P between electromagnetic power and mechanical power during the period from the fault clearing moment to the system reaching the maximum power angle (or theoretical limit power angle). e -P mIntegrating over the power angle δ yields the maximum deceleration energy consumed or capable of being consumed by the rotor deceleration after a fault. In a multi-machine system, these areas are obtained through time-domain simulation to capture the power angle trajectory δ and electromagnetic power P of each generator. e After (t), numerical integration is used to achieve the result.

[0041] 2. Based on the difference between the deceleration area and the acceleration area: (3) In the formula, η For the stability margin of the work angle; A dec This represents the actual deceleration area; A inc To accelerate the area.

[0042] Two forms can be selected based on the actual application scenario: Form 1 is suitable for comprehensive evaluation scenarios that require simultaneous consideration of acceleration energy and potential deceleration capability; Form 2 is suitable for rapid evaluation scenarios that require direct quantification of the net difference between the acceleration area and the deceleration area. Power angle stability level. η The criteria for judgment are: This is normal, indicating that the system has sufficient power angle stability margin; This is an alarm indicating that the system's power angle stability margin is insufficient and there is a risk of losing synchronization. This is an emergency, indicating that the system has lost its synchronization stability.

[0043] Regarding frequency stability, the stability margins for low-frequency and high-frequency deviations are calculated using the following formulas: 1. Low-frequency deviation stability margin: (4) In the formula, This is for low-frequency deviation stability margin; The system's rated frequency; This is the lowest value reached by the frequency drop during the system's transient process, i.e., the transient low-frequency extreme value; The minimum allowable frequency limit is determined according to the national standard GB / T 40585-2021 and the power grid safety operation regulations.

[0044] 2. High-frequency deviation stability margin: (5) In the formula, This is for high-frequency deviation stability margin; The system's rated frequency; This is the maximum value reached by the frequency drop during the system's transient process, i.e., the transient high-frequency extreme value; The maximum permissible frequency limit for the system is determined in accordance with the national standard GB / T 40585-2021 and the regulations for safe operation of power grids.

[0045] Overall frequency stability level Take low-frequency deviation stability margin and high-frequency deviation stability margin The smaller value in, that is: (6) In the formula, To achieve a comprehensive frequency stability level, the smaller of the two values ​​is chosen to reflect the system's stability margin in the direction of the most unfavorable frequency deviation.

[0046] Frequency stability level The criteria for judgment are: This is considered normal, indicating that the frequency fluctuation is within the allowable range. This is an alarm, indicating a significant frequency offset and a risk of frequency instability. This is an emergency signal, indicating a severe frequency deviation and a risk of system frequency collapse.

[0047] Regarding voltage stability levels, the voltage stability margin index is used as a reference. The characterization and calculation formula are as follows: (7) In the formula, This is the voltage stability margin, used to measure the degree to which the system node voltage deviates from the reference voltage by 0.8 pu. The larger the value, the higher the voltage stability level. V The voltage trajectory is extracted from the actual voltage of the system nodes through time-domain simulation.

[0048] 0.8pu is a widely used engineering empirical threshold in power system voltage stability analysis. When the voltage drops below this value, the positive feedback process of reactive power voltage drop, reactive power deficit widening, and further voltage drop will significantly worsen, and the system will face the risk of voltage collapse. However, this value is not a physical constant or legal standard, and other values ​​such as 0.75pu and 0.85pu can be used in different scenarios. In this patent, 0.8pu serves as a benchmark reference point for the voltage stability margin calculation formula and is a preferred embodiment parameter. Its core function is to provide a unified quantitative scale, which can be replaced with other reasonable values ​​according to the actual power grid characteristics by those skilled in the art.

[0049] This index is based on the "80% voltage threshold" criterion commonly used in power system voltage stability analysis. It uses 0.8 pu as the voltage stability critical point and calculates the deviation margin of the actual voltage relative to this critical point. Voltage stability level The criteria for judgment are: A normal value indicates that the node voltage margin is sufficient and the system has good voltage support capability. An alarm indicates that the node voltage is too low and the system is in a voltage alarm state, requiring close monitoring. This is an emergency, indicating that the node voltage has dropped below 0.8 pu, and the system faces the risk of voltage instability.

[0050] Based on the assessment results across four dimensions, the final risk level for each region was determined according to the principle of the most severe level: 1. If any dimension is assessed as urgent, the overall risk level is severe risk, corresponding to consequences such as system disconnection / generator loss of synchronization / system instability / large-scale load loss, severe frequency or voltage deviation, and severe overload of multiple lines.

[0051] 2. If there is a dimension that is rated as alarm but there is no emergency dimension, the overall risk level is general risk, corresponding to consequences such as severe frequency or voltage deviation, severe overload of multiple lines, etc.

[0052] 3. If all dimensions are rated as normal, the overall risk level is no risk, and the corresponding voltage, frequency, and line load are all within the specified range.

[0053] In one implementation example, regarding step two, there is a quantitative assessment of the matching degree between emergency resources and risk levels based on the cloud model.

[0054] To measure the degree of matching between emergency resources and risk levels in different regions, this example proposes a comprehensive evaluation method based on a cloud model. It quantifies the response capability of emergency resources in each region from five dimensions: resource type, action response time, repair operation speed, available capacity, and current status. An evaluation cloud model is established to calculate the comprehensive risk matching index of the region, thereby accurately identifying severely mismatched regions and guiding the subsequent allocation of emergency resources.

[0055] In terms of quantifying emergency resource response capabilities, regional [measures] are set up. r Emergency resource response capabilities are quantified from the following five dimensions: 1. Resource type indicators This indicator reflects the region. r The completeness of the types of emergency resources allocated is scored based on the degree of matching between the types of resources actually allocated in the region and the types of resources needed. The higher the score, the better the matching between the types of resources and the needs.

[0056] 2. Action response time metric: (8) In the formula, For the region rThe action response time matching index reflects the degree of matching between the actual response time and the allowed response time of emergency resources in the area from receiving an instruction to arriving at the scene; For the region r The maximum allowable response time corresponding to the risk level is determined by the risk level of the area; the higher the risk level, the shorter the allowable time. For the region r The actual response time of emergency resources is the time required from the issuance of an emergency order to the arrival of the first batch of resources on site.

[0057] 3. Repair operation speed indicators: (9) In the formula, For the region r The speed matching index of emergency repair operations reflects the ability of the emergency repair teams in the area to complete emergency repair tasks within the allowed time. For the region r The average work efficiency of the emergency repair team represents the actual repair speed of the team. For the region r The maximum allowable response time corresponding to the risk level; For the region r The estimated workload for emergency repairs is based on the type and scale of the fault.

[0058] This indicator represents the amount of repair work that the emergency repair team can complete within the allowed response time. Compared to the actual amount of repair work required The larger the ratio, the more sufficient the emergency repair capability.

[0059] 4. Available capacity indicators: (10) In the formula, For the region r The available capacity matching index reflects the degree of matching between the available power supply support resources and the required power supply capacity in the region; For the region r Total rated power of available mobile emergency generator sets; For the region r Total rated power of available mobile energy storage systems; For the region r The required power supply capacity under the current risk level is determined comprehensively based on the load gap and the risk level.

[0060] This indicator represents the total available power supply capacity in the region. With required capacity The higher the ratio, the more sufficient the power supply support resources.

[0061] 5. Available Status Indicators This indicator reflects the region. r The current availability status of emergency resources takes into account factors such as resource integrity rate, on-duty rate, and availability status. A higher value indicates better resource availability.

[0062] Based on the quantitative indicators of the above five dimensions, each risk level is established. g The following matching indicators q Evaluation criteria for cloud models: (11) In the formula, g The risk level is represented by a value of severe risk, moderate risk, or no risk. q To match the metrics, five dimensions are used: resource type, action response time, emergency repair speed, available capacity, and current status. Risk level g Lowering the target q The expected value of the cloud model reflects the ideal central value of the indicator at this level; The corresponding entropy reflects the degree of fuzziness in the index value; This corresponds to hyperentropy, reflecting the randomness of the index values. Risk level g Matching Indicators q The evaluation criteria for cloud models.

[0063] The above , , The determination of the three cloud model parameters is based on the experience of domain experts and historical statistical data. The reverse cloud algorithm is used to statistically estimate multiple reasonable scores or numerical samples of each indicator under each risk level given by experts. Specifically, the mean, variance and higher-order central moments of the samples are calculated by the method of moments estimation, maximum likelihood estimation or data fitting method. Among them, the expectation mainly reflects the ideal central value in the expert consensus (estimated by the sample mean), the entropy reflects the dispersion of expert opinions, that is, the fuzzy range of indicator values ​​(estimated by the sample variance), and the hyperentropy reflects the stability of the entropy itself, that is, the randomness in the evaluation (estimated by the higher-order statistics of the samples). Finally, the qualitative level standard is quantified into computable cloud model numerical features. After determination, cloud droplets are generated by the forward cloud generator to verify the parameters and ensure their rationality.

[0064] In subsequent calculations As a quantitative benchmark, its three parameters (expectation, entropy, and hyperentropy) are directly substituted into the certainty formula—using expectation as the ideal center value, generating normal random numbers from entropy and hyperentropy, calculating the certainty of the measured values ​​of each region relative to the "standard cloud", and obtaining the regional comprehensive matching degree after weighted summation of the certainty. The regional comprehensive matching degree is then compared with four matching level cloud models (severe undermatch, undermatch, basic match, and good match), and the level with the highest certainty is taken as the final matching level of the region, thus running through the entire process of quantitative evaluation of matching.

[0065] Targeting the region r Based on actual emergency resource data, calculate the matching degree and risk level of the region. G r Certainty: (12) In the formula, For the region r In terms of indicators q Risk level G r The certainty of issuing the target cloud reflects the region's first q The measured values ​​of each matching indicator belong to the risk level. G r The degree to which the target cloud is met; the larger the value, the closer the indicator is to meeting the target level. For the region r In the q Measured values ​​on each matching metric; For the region r Risk level G r Lowering the target q The expected value corresponding to the cloud model; For the region r Risk level G r Lowering the target q A normal random number corresponding to the cloud model is used to characterize the degree of dispersion of cloud droplets in deterministic calculation.

[0066] The above It is obtained directly from power grid operation monitoring, resource ledgers, and statistical data; Derived from a pre-established "compliance cloud" model of each indicator under each risk level, it represents the ideal compliance center value under that level; This is done during each determination calculation, using For the mean, The standard deviation is a specific random number generated by normal random sampling. It is used to reflect the discrete fluctuation of the fuzziness of the indicator at this level in a single calculation, so that the determination result has randomness to characterize the uncertainty in the evaluation.

[0067] in, Obey For the mean, Normal distribution with standard deviation: (13) In the formula, Risk level G r Lowering the target q The entropy corresponding to the cloud model; This corresponds to hyperentropy. The introduction of this normally distributed random number makes the determination calculation no longer a single value, but a distribution with randomness, thus better characterizing the uncertainty and ambiguity in emergency resource capacity assessment.

[0068] The aforementioned entropy and hyperentropy were both statistically estimated from expert evaluation samples using the inverse cloud algorithm: entropy is the risk level G. r The lower index q corresponds to the entropy of the cloud model, reflecting the degree of ambiguity of the index value at this level. It is estimated by calculating the first-order sample absolute central moment from multiple reasonable scores or numerical samples given by experts. The hyperentropy reflects the stability of the entropy itself, i.e. the randomness in the evaluation, and is calculated by the deviation between the sample variance and the entropy.

[0069] Then calculate the region r Overall risk matching degree: (14) In the formula, For the region r The higher the overall risk matching degree, the higher the degree of matching between emergency resources and risk level in the region. q To match the indicator number, they correspond to five dimensions in sequence: resource type, action response time, emergency repair speed, available capacity, and current status. For the region r In terms of indicators q Certainty on; For the first q The weighting coefficients of each matching indicator are used to reflect the relative importance of different matching indicators in the overall matching degree.

[0070] It is calculated from equation (12); The weights of each indicator are usually determined by combining the analytic hierarchy process (AHP) with expert experience, by constructing a judgment matrix and calculating eigenvectors.

[0071] To determine the matching level, four matching cloud levels are defined: severely undermatched, undermatched, basic match, and good match. The region is calculated using the following formula. r Overall risk matching degree Determinism for each matching level: (15) In the formula, For overall matching degree Matching level l Certainty; l To match the level number, ; For matching level l The expected value corresponding to the cloud model; For matching level l The entropy corresponding to the cloud model.

[0072] area r The matching level is determined by the following formula: (16) In the formula, For the region r The matching level determination result can be one of the following: severely undermatched, undermatched, basically matched, or well matched. That is, the matching level with the highest certainty is taken as the final matching level of the region.

[0073] like If the region is found to be in a state of severe mismatch and the region is still unable to meet the requirements after coordination, then it is determined to be a large-scale mismatch and the mechanism for calling on emergency resources from outside the region will be activated.

[0074] In one implementation example, regarding the coordinated optimization and allocation of cross-regional emergency resources: for regions where emergency resources and risk levels are severely mismatched, a multi-objective optimization model is constructed with the goals of maximizing the matching degree and minimizing the movement distance of emergency resources. Under the premise of satisfying traffic constraints and ensuring no mismatch in normal areas, the model optimizes the allocation of emergency resources among different regions and proposes a hierarchical solution method to determine the type, capacity, and path of resource allocation.

[0075] Optimization Objective 1: Maximize the match between emergency resources and risk levels in each region after adjustments. (17) In the formula, H The set of area IDs with a matching level of "severe undermatch" or "undermatch" is the set of target areas that need to receive emergency resource transfers. r For area code, ; For the region r The risk level is set to "urgent", "critical", or "normal". For the region r Overall matching degree after resource allocation.

[0076] The risk level weighting coefficient is used to characterize the importance of regions with different risk levels in the optimization objective, and is defined as follows: (18) This weighting coefficient indicates that emergency areas should be given priority in obtaining resources to improve matching accuracy; normal areas should not have their matching accuracy reduced due to resource redistribution; and critical areas fall in between.

[0077] The objective function states that after resource allocation is completed, priority should be given to improving the overall matching degree of emergency areas, while avoiding the decline in the matching degree of normal areas due to resource allocation, thereby achieving the optimal distribution of global matching degree.

[0078] Optimization Objective 2: Minimize the movement path cost of cross-regional emergency resource allocation. (19) In the formula, a Assign a region number to the area from which resources are transferred. b This is the resource import region number, i.e., the undermatched region. ; k Emergency resource types are assigned numbers to distinguish different types of resources; p The movement path number indicates the location from the region. a To the area b Optional paths; λ This is the path cost penalty coefficient, used to quantify the distance a resource is moved into economic cost; To the region a To the area b Along the path p Allocation k The unit movement cost of such resources is determined by a combination of factors, including path length and road conditions. To the region a To the area b Along the path p The first allocation k The number of resources of each type.

[0079] The cost is usually calculated by combining GIS road network data with costs such as fuel consumption and toll fees. The unit cost is the same for different resource types (repair team / MEG / MESS) but can be set differently. Its specific value is determined by iterative optimization using a genetic algorithm, provided that six types of constraints are met.

[0080] The objective function aims to minimize the total path cost of resource movement across regions while meeting the emergency needs of each region, thereby improving allocation efficiency and reducing allocation costs.

[0081] Constraint 1: Adjusted resource capacity constraints (20) In the formula, For the region a No. k The original stock of a type of resource, that is, the total number of such resources in the region before the allocation begins; For the region a After allocation k The final holdings of this type of resource; To the region a Move to the undermatched region b The k Number of resource types; To extract from undermatched regions b Transferred to the region a The k Number of resources of each type.

[0082] This constraint ensures that the resource holdings of each region after the allocation are equal to the initial holdings minus the amount transferred out plus the amount transferred in, thus guaranteeing resource conservation.

[0083] Constraint 2: Adjusted resource matching degree constraint.

[0084] (twenty one) In the formula, For the region r Overall matching degree after adjustment; f (·) is the function for calculating the overall matching degree of the cloud model. Its input consists of five dimensions of index values, and its output is the overall matching degree. ; They are respectively regions r The adjusted indicator values ​​are based on five dimensions: resource type, action response time, emergency repair speed, available capacity, and current status.

[0085] Each adjusted indicator value is recalculated based on the adjusted resource availability: resource type and availability indicators generally do not change with changes in resource quantity, remaining unchanged or re-scored according to the new configuration if resource types are added or deleted; action response time is recalculated based on the actual response time of the resource site after adjustment; emergency repair speed is updated based on the average work efficiency of the emergency repair team after adjustment; available capacity is recalculated based on the total rated power of MEG and MESS after adjustment. All adjusted indicator values ​​are then substituted into the cloud model comprehensive matching degree calculation function to obtain the comprehensive matching degree after regional adjustment.

[0086] Constraint 3: High-risk areas must meet the required standards.

[0087] (twenty two) In the formula, For the region r Overall matching degree after adjustment; Risk level The corresponding compliance matching threshold, i.e., the region r The minimum matching degree requirement that should be achieved under the current risk level; This constraint applies to all undermatched regions. H Both must be true.

[0088] This constraint ensures that, after resource allocation, all undermatched areas achieve a comprehensive matching degree that meets at least the minimum compliance level required by their risk level.

[0089] Constraint 4: No mismatch constraint in the normal region.

[0090] (twenty three) In the formula, For the region r Overall matching degree after adjustment; For the region r Current risk level The corresponding matching degree threshold; Table This constraint must hold for all regions that are not in the undermatched set.

[0091] This constraint ensures that regions that did not originally require resource allocation will not see their overall matching degree drop below the threshold due to resource reduction after they become resource providers, thus preventing secondary mismatches.

[0092] Constraint 5: Allocation Time Constraint (twenty four) In the formula, To the region a To the area b Along the path p The first allocation k Number of resource types; The maximum allowable time limit for allocation means that, provided the emergency needs of each region are met, the transportation time for all allocation routes must not exceed this limit. This limit is determined by road conditions, traffic control requirements, and emergency response time limits.

[0093] Constraint 6: Linkage Constraints of Allocation Variables (25) In the formula, To the region a To the area b Along the path p The first allocationk The number of class resources; For the region a No. k The maximum amount of resources that can be retrieved, i.e., within the protected area. a The maximum amount of this type of resource that can be transferred out, assuming the company's own emergency needs are taken into account. For binary decision variables, Indicates selection from region a To the area b Along the path p Allocation k Class resources, This indicates that the allocation path is not selected.

[0094] This constraint states that a positive resource allocation amount can only occur on a selected allocation path; and this allocation amount must not exceed the maximum allocation limit for that type of resource in the originating region. This constraint achieves linked control between "path selection" and "allocation quantity".

[0095] The constructed multi-objective optimization solution model is solved hierarchically using a genetic algorithm, with the specific steps as follows: 1. Input data Enter the risk level of each region Original reserves of various types of resources in various regions Road network distance data Each constraint parameter.

[0096] 2. Chromosome coding The allocation plan is coded, with each chromosome representing all decision variables. and An ordered set, that is, a complete allocation scheme is encoded as a chromosome.

[0097] 3. Initialize the population Randomly generated N Individuals satisfying the basic constraints are used as the initial population, among which N The population size is an integer between 50 and 200.

[0098] 4. Calculate fitness The fitness function for each individual is calculated using a weighted combination of bi-objectives: (26) In the formula, F This represents the fitness value of an individual; a higher value indicates a better allocation plan. The weighting coefficient for the matching degree objective is used to adjust the proportion of the matching degree objective in fitness. The weighting coefficient for the path cost objective is used to adjust the proportion of the path cost objective in the fitness score. 5. Selection, crossover, and mutation The roulette wheel selection method is used to select superior individuals from the current population; crossover operation is performed on the selected individuals with crossover probability to generate offspring individuals; then mutation operation is performed on the offspring individuals with mutation probability to introduce new genetic information and generate a new generation of population.

[0099] 6. Constraint Repair For individuals in the new generation population that do not meet all constraints, constraint repair is performed. Specifically, the decision variables that violate the constraints are adjusted one by one to make them meet all constraints again.

[0100] 7. Termination of judgment Determine if the termination condition is met. If it is, output the optimal allocation scheme; otherwise, return to step 4 to continue iteration. The optimal allocation scheme is the set of resource allocation instructions that maximizes the matching degree and minimizes the path cost, while satisfying all constraints such as high-risk area compliance, no mismatch in normal areas, and allocation time not exceeding limits. In this embodiment, this scheme is represented by the combination of specific allocation records in Table 1.

[0101] If, after the aforementioned cross-regional resource coordination and optimization, some regions still have an overall matching degree lower than their target threshold, resulting in a large-scale mismatch, then the external emergency resource mobilization mechanism will be activated to allocate and supplement resources from the external emergency resource pool to meet the basic emergency support needs of the undermatched regions.

[0102] Regarding the differentiated allocation strategy for emergency resources: This invention divides emergency resources into two categories based on their functional attributes: fault repair resources and power supply support resources, and formulates allocation strategies for each category.

[0103] Fault repair resources are mainly composed of repair teams. Based on the intraday risk classification results, support routes are optimized and traffic constraints, fleet constraints, and power grid operation constraints are met to ensure that repair forces can quickly reach the fault site to carry out restoration work.

[0104] Power supply support resources are mainly mobile emergency generator sets (MEG) and mobile energy storage systems (MESS), and are allocated differently according to the risk level of the area: for emergency areas, large-capacity MEG and MESS are prioritized to provide temporary power supply support to ensure power supply to critical loads; for critical areas, power supply support resources of appropriate capacity are allocated according to the actual power gap; for normal areas, power supply support resources are generally not allocated to ensure that they are not mismatched.

[0105] This embodiment proposes a method for classifying power grid emergencies and coordinating emergency resource allocation under extreme operating conditions, constructing a complete emergency decision-making chain from risk identification, level assessment, matching measurement to coordination optimization. By meticulously extrapolating the time-domain trajectories of multi-dimensional electrical quantities such as voltage, frequency, and power angle of the power grid in various regions under extreme operating conditions, and integrating national standards to comprehensively evaluate from four dimensions—line overload safety margin, power angle stability level, frequency stability level, and voltage stability level—multi-level refined classification of emergencies is achieved. Based on this, a cloud model comprehensive evaluation theory is innovatively introduced to quantify emergency resource response capabilities from five dimensions: resource type, action response time, repair operation speed, available capacity, and current status, accurately identifying severely mismatched areas. Furthermore, with the dual objectives of maximizing matching degree and minimizing movement path cost, a cross-regional resource coordination optimization model considering multiple constraints such as matching compliance in high-risk areas and no mismatch in normal areas is constructed and efficiently solved using a genetic algorithm.

[0106] Emergency resources are categorized into two types based on their functional attributes: fault repair and power supply support. Differentiated allocation is then implemented by combining a hierarchical and progressive strategy driven by matching levels with an external call mechanism. This effectively addresses the prominent issues of existing methods, such as insufficient precision in risk classification, lack of quantitative assessment of matching degree, lack of systematic optimization in cross-regional allocation, and homogenization of resource allocation. As a result, the emergency support capability of the power grid in response to extreme operating conditions and the scientific decision-making level of resource allocation have been significantly improved.

[0107] To address the lack of a refined emergency risk classification mechanism in existing technologies: most existing methods classify risk levels based on a single electrical quantity or a simple threshold, failing to fully consider the comprehensive evolution characteristics of the time-domain trajectories of multi-dimensional electrical quantities such as voltage, frequency, and power angle, and failing to systematically and comprehensively evaluate the line overload safety margin, power angle stability level, frequency stability level, and voltage stability level in conjunction with national standards, resulting in inaccurate and unrefined risk classification results.

[0108] This embodiment's sub-technical solution is based on a refined risk classification mechanism for sudden emergencies, derived from multi-dimensional electrical quantity trajectory extrapolation. Instead of using a single electrical quantity threshold for simple risk assessment, this invention uses time-domain simulation to extrapolate the trajectories of voltage, frequency, and power angle in different regional power grids under extreme conditions. It comprehensively considers line overload safety margin, power angle stability level, frequency stability level, and voltage stability level, and, in accordance with the national standard GB / T40585-2021, classifies regional risks into three levels: emergency, critical, and normal. For each dimension, a clear quantitative calculation formula for stability margin and classification criteria are established. The assessment results for each dimension are comprehensively judged according to the severity level principle, while also associating different consequence types: such as system disconnection, generator loss of synchronism, system instability, large-scale load shedding, severe frequency and voltage deviation, and severe overload of multiple lines are considered severe risks; frequency and voltage deviation and overload of multiple lines are considered general risks; voltage, frequency, and line load within specified ranges pose no risk. This forms a complete mapping chain from electrical quantity trajectory to risk level, ensuring the scientific validity and reliability of the classification results.

[0109] Secondly, this invention employs a quantitative assessment method for the matching degree between emergency resources and risk levels based on cloud models. To address the lack of quantitative measurement methods for the matching between emergency resource capabilities and actual regional risk levels, this invention proposes a comprehensive quantification of the emergency resource response capabilities of each region across five dimensions: resource type, action response time, repair operation speed, available capacity, and current status. It innovatively introduces cloud model comprehensive evaluation theory, establishing evaluation standard cloud models for each matching indicator under each risk level. By calculating the certainty of the regional matching degree against the cloud model at each level, the comprehensive regional risk matching degree is obtained, and the matching results are then divided into four levels: severely under-matched, under-matched, basically matched, and well-matched. This mechanism can accurately identify under-matched regions with severely insufficient emergency resources, providing clear targets for subsequent resource allocation.

[0110] The classification and allocation of emergency resources is not refined enough: existing methods often treat emergency resources as homogeneous materials and allocate them uniformly, failing to classify and manage them differently based on their functional attributes. In reality, there are fundamental differences between fault repair resources and power supply support resources in terms of allocation logic, response characteristics, and constraints, requiring targeted allocation strategies.

[0111] This embodiment's sub-technical solution considers a multi-constraint cross-regional emergency resource coordination and allocation strategy. For identified under-matched areas, this invention constructs a multi-objective optimization model with the dual optimization objectives of maximizing the adjusted matching degree and minimizing the cost of emergency resource movement paths. In terms of constraint settings, it comprehensively incorporates matching compliance constraints for high-risk areas, non-mismatch constraints for normal areas, allocation time constraints, and linkage constraints of allocation variables to ensure the feasibility of the allocation plan in actual implementation. Based on this, a hierarchical solution method based on a genetic algorithm is proposed to determine the type, capacity, and path of resource allocation. Simultaneously, this invention categorizes emergency resources according to their functional attributes into fault repair resources and power supply support resources, formulating differentiated allocation strategies for each, further improving the precision of resource allocation.

[0112] Engineering Cases This embodiment takes nine regions (R1-R9) of a provincial power grid as the research object to simulate the process of sudden risk classification and emergency resource coordination and allocation under an extreme typhoon.

[0113] Step 1: Synchronous access and preprocessing of multi-source data.

[0114] First, the power grid operation data and meteorological forecast data for each region are synchronized and preprocessed. The meteorological data server provides gridded meteorological forecast data covering all regions of the power grid for future periods, including parameters such as wind speed, rainfall intensity, and temperature. The power grid dispatching system provides the real-time topology, line parameters, generator parameters, and current power flow sections of the power grid in each region. After the two types of data are aligned in both time and space dimensions, they together form the basic input for subsequent risk classification and emergency resource allocation.

[0115] Step 2: Sudden risk classification based on multidimensional electrical quantity trajectory extrapolation.

[0116] During the typhoon, power grids in various regions suffered varying degrees of wind and rain damage. The system first accessed real-time monitoring data from the power grids in each region, and then performed extreme operating condition evolution simulations based on time-domain simulations to obtain voltage, frequency, and power angle trajectories for each region. Subsequently, a parallel evaluation process across four dimensions was initiated to determine the comprehensive risk level of each region according to the principle of the most severe level.

[0117] Heat maps of initial risk levels for each region are shown below. Figure 1 As shown. The initial risk levels for each region are: R3 and R9 are considered moderate risk; R1, R2, R4, R5, R6, R7, and R8 are considered risk-free.

[0118] As the typhoon's path evolved and power grid conditions continued to deteriorate, the risk levels in various regions dynamically changed. The resulting heatmap is shown below. Figure 2As shown: R1, R3, R5, R7, and R9 represent no risk; R2 and R6 represent moderate risk; and R4 and R8 represent severe risk.

[0119] Step 3: Quantitative assessment of the matching degree between emergency resources and risk levels based on cloud models.

[0120] like Figure 3 The diagram shows the flowchart for quantitative evaluation of matching degree based on cloud model.

[0121] The maximum allowable response time is determined based on the risk level of each region: 2 hours for high-risk regions (R4, R8); 4 hours for moderate-risk regions (R2, R6); and 6 hours for no-risk regions (R1, R3, R5, R7, R9).

[0122] According to the method described in this invention, the emergency response capability of each region's resources is quantified from five dimensions: resource type, action response time, repair operation speed, available capacity, and current status. An evaluation standard cloud model is established for each matching indicator under each risk level, with the weight coefficients for each dimension corresponding to resource type, action response time, repair operation speed, available capacity, and current status, respectively. Based on actual data for each region, the certainty of each indicator for achieving the "standard cloud" under the risk level is calculated, and then the comprehensive risk matching degree of each region is calculated.

[0123] Risk levels and overall matching degree of each region before allocation are as follows: Figure 4 As shown.

[0124] Based on the matching data before allocation, it can be seen that: the matching degree of severely risky areas R4 and R8 is extremely low, indicating a severe undermatch, and they urgently need emergency resources to be allocated; the matching degree of general risk areas R2 and R6 is also severely undermatched, and they also need resource supplementation; the matching degree of risk-free areas R1, R3, R5, R7, and R9 is good, and they can be used as resource providers. Therefore, the set of undermatched areas is identified as {R2, R4, R6, R8}.

[0125] Step 4: Coordinate and optimize the allocation of emergency resources across regions.

[0126] like Figure 5 The diagram shows the process flow for cross-regional emergency resource coordination and optimization.

[0127] For the undermatched region set {R2, R4, R6, R8}, a cross-regional resource coordination and optimization allocation process is initiated. An optimization model is constructed with the dual objectives of maximizing matching degree and minimizing movement path cost. Risk level weight coefficients are set as follows: severe risk = 1, moderate risk = 0, and no risk = -1. The path cost penalty coefficient is set to 0.5 million yuan / (km·unit resource).

[0128] In terms of constraints, the system synchronously verifies six types of constraints: adjusted resource capacity constraints, adjusted resource matching degree constraints, high-risk area matching compliance constraints, risk-free area non-mismatch constraints, allocation time constraints, and allocation variable linkage constraints.

[0129] A genetic algorithm was used to solve the problem, with a population size of 100, a maximum number of iterations of 200, a crossover probability of 0.80, and a mutation probability of 0.05. After iterative optimization, the optimal allocation scheme was obtained, as shown in Table 1.

[0130] Table 1 Optimal Resource Allocation Scheme Source region target area Resource types quantity Total cost R1 R4 RC (Repair Team) 1 15.0 R1 R4 MEG 1 24.75 R3 R2 MEG 1 24.75 R3 R6 MEG 1 24.75 R5 R4 RC (Repair Team) 1 15.0 R5 R8 RC (Repair Team) 2 30.0 R7 R4 RC (Repair Team) 1 15.0 R7 R4 MESS 1 21.45 R7 R8 RC (Repair Team) 1 15.0 R7 R8 MESS 1 21.45 R9 R8 RC (Repair Team) 2 30.0 R9 R8 MEG 1 24.75 R9 R8 MESS 1 21.45 The allocation plan shows that: risk-free area R1 allocates 1 repair team and 1 MEG to high-risk area R4; risk-free area R3 allocates 1 MEG each to general risk areas R2 and R6; risk-free area R5 allocates 1 repair team to high-risk area R4 and 2 repair teams to high-risk area R8; risk-free area R7 allocates 1 repair team and 1 MESS to high-risk area R4 and 1 repair team and 1 MESS to high-risk area R8; risk-free area R9 allocates 2 repair teams, 1 MEG, and 1 MESS to high-risk area R8. All allocation routes meet the 2-hour emergency response time constraint, and the amount of resources allocated from each area does not exceed the maximum available allocation limit.

[0131] Step 5: Differentiated allocation of emergency resources.

[0132] This invention categorizes emergency resources into two main types based on their functional attributes and formulates allocation strategies for each category.

[0133] Emergency repair resources are primarily provided by repair teams. The allocation plan includes 10 RC (Repair, Response, and Control) units: 1 unit for R1→R4, 1 unit for R5→R4, 2 units for R5→R8, 1 unit for R7→R4, 1 unit for R7→R8, and 2 units for R9→R8. Allocation must meet traffic constraints, fleet constraints, and power grid operation constraints. Repair teams will prioritize support for the high-risk areas R4 and R8.

[0134] Power supply support resources mainly consist of mobile emergency generator sets (MEGs) and mobile energy storage systems (MESSs). The allocation plan includes four MEGs and three MESSs. Allocation is differentiated based on risk level: severely risky areas R4 and R8 receive priority support from MEGs and MESSs; moderately risky areas R2 and R6 receive MEG support; and in risk-free areas, redundant resources are appropriately redistributed while ensuring their own matching degree does not decrease.

[0135] Step Six: Verify the mixing effect.

[0136] After the resource allocation is completed, the overall matching degree of each region is recalculated. The resource matching degree of each region after the allocation is as follows: Figure 6 As shown.

[0137] Before the allocation, the matching degree of R4 in the high-risk area was only 0.01 and that of R8 was only 0.00, both of which were in a state of severe undermatch; the matching degree of R2 and R6 in the general-risk area was 0.27, which were also in a state of severe undermatch; the matching degree of R1, R3, R5, R7 and R9 in the risk-free area was 1.00, which was a good match.

[0138] After cross-regional resource coordination and optimization, the matching degree of each region has significantly improved. The matching degree of region R4 improved from 0.01 to 0.89, jumping from severely under-matched to well-matched, meeting the requirements for a severe risk level. The matching degree of region R8 improved from 0.00 to 0.87, also jumping from severely under-matched to well-matched, meeting the requirements for a severe risk level. The matching degree of region R2 improved from 0.27 to 0.93, jumping from severely under-matched to well-matched; the matching degree of region R6 improved from 0.27 to 0.93, jumping from severely under-matched to well-matched. After redundant resources were removed from risk-free regions R1, R3, R5, R7, and R9, the matching degree remained at 1.00, with no mismatch occurring.

[0139] After the resource allocation, the matching degree of all regions reached the threshold corresponding to their risk levels. The number of well-matched regions increased from 5 to 9, while the number of severely mismatched regions was reduced to zero. There were no large-scale mismatches, eliminating the need to activate the external emergency resource mobilization mechanism, thus verifying the effectiveness and feasibility of the method of this invention. Comparison of resource matching degrees of each region before and after allocation is shown below. Figure 7 As shown.

[0140] Example 2 The purpose of this embodiment is to provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the above-described method.

[0141] Example 3 The purpose of this embodiment is to provide a computer-readable storage medium.

[0142] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the steps of the above-described method.

[0143] Example 4 The purpose of this embodiment is to provide a system for power grid risk assessment and collaborative optimization of emergency resource allocation under extreme operating conditions, including: The module for determining the final risk level of each region is configured to: acquire power grid operation data and weather forecast data for each region, obtain the time-domain trajectory of electrical quantities for each region, conduct a comprehensive assessment of each region based on the acquired time-domain trajectory of electrical quantities, and determine the final risk level of each region; The matching degree quantitative assessment module is configured to: establish an evaluation cloud model, quantitatively assess the matching degree between emergency resources and risk levels based on the cloud model, calculate the regional risk comprehensive matching degree index, and identify severely mismatched areas; The resource allocation module is configured to: for areas where emergency resources and risk levels are severely mismatched, construct a multi-objective optimization model with the objectives of maximizing the matching degree and minimizing the movement distance of emergency resources, optimize the allocation of emergency resources in different areas under the premise of satisfying traffic constraints and ensuring no mismatch in normal areas, and propose a hierarchical solution method to determine the type, capacity and path of resource allocation.

[0144] Example 5 The purpose of this embodiment is to provide a computer program product containing instructions that, when run on a computer, cause the computer to perform the methods and functions involved in any of the above embodiments.

[0145] The steps and methods involved in the apparatus of the above embodiments correspond to those in Embodiment 1. For specific implementation details, please refer to the relevant description section of Embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood as including any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and enabling the processor to perform any of the methods in this invention.

[0146] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0147] 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 method for collaborative optimization and allocation of power grid risk assessment and emergency resources under extreme operating conditions, characterized by: include: By acquiring power grid operation data and meteorological forecast data for each region, the time-domain trajectory of electrical quantities in each region is obtained. Based on the acquired time-domain trajectory of electrical quantities, a comprehensive assessment of each region is conducted to determine the final risk level of each region. Establish an evaluation cloud model, quantitatively assess the matching degree between emergency resources and risk levels based on the cloud model, calculate the regional risk comprehensive matching degree index, and identify areas with serious under-match; For regions where emergency resources and risk levels are severely mismatched, a multi-objective optimization model is constructed with the goals of maximizing the matching degree and minimizing the movement distance of emergency resources. Under the premise of satisfying traffic constraints and ensuring no mismatch in normal areas, the model optimizes the allocation of emergency resources in different regions and proposes a hierarchical solution method to determine the type, capacity, and path of resource allocation.

2. The method for power grid risk assessment and collaborative optimization of emergency resource allocation under extreme operating conditions as described in claim 1, characterized in that, Based on the acquisition of power grid operation data and meteorological forecast data in various regions, when a sudden risk warning signal for the power grid is detected, extreme operating condition evolution is extrapolated based on time-domain simulation for different regional power grids, and the time-domain trajectory of electrical quantities in each region is obtained, including voltage trajectory, frequency trajectory and power angle trajectory. Based on the acquired electrical quantity time-domain trajectory and in conjunction with national standards, a comprehensive assessment of each region is conducted from four dimensions: line overload safety margin, power angle stability level, frequency stability level, and voltage stability level.

3. The method for power grid risk assessment and collaborative optimization of emergency resource allocation under extreme operating conditions as described in claim 1, characterized in that, During the comprehensive assessment, if any dimension is rated as urgent, the overall risk level is severe risk, corresponding to system disconnection, generator loss of synchronism, system instability, large-scale load loss, severe frequency or voltage deviation, and severe overload of multiple lines. If there is a dimension that is rated as alarm but there is no emergency dimension, the overall risk level is general risk, corresponding to consequences such as severe frequency or voltage deviation, severe overload of multiple lines, etc. If all dimensions are rated as normal, the overall risk level is no risk, corresponding to voltage, frequency, and line load all being within the specified range.

4. The method for power grid risk assessment and collaborative optimization of emergency resource allocation under extreme operating conditions as described in claim 1, characterized in that, When establishing the evaluation cloud model, the emergency response capability of each region is quantified from five dimensions: resource type, action response time, repair operation speed, available capacity, and current status, and various risk levels are established. g The following matching indicators q Evaluation criteria for cloud models: In the formula, g The risk level is represented by a value of severe risk, moderate risk, or no risk. q To match the metrics, five dimensions are used: resource type, action response time, emergency repair speed, available capacity, and current status. Risk level g Lowering the target q The expected value of the cloud model reflects the ideal central value of the indicator at this level; The corresponding entropy reflects the degree of fuzziness in the index value; This corresponds to hyperentropy, reflecting the randomness of the index values. Risk level g Matching Indicators q The evaluation criteria for cloud models.

5. The method for power grid risk assessment and collaborative optimization of emergency resource allocation under extreme operating conditions as described in claim 1, characterized in that, Based on actual emergency resource data for a specific region, the degree of matching and the certainty of the risk level for that region are calculated, reflecting the region's [risk level]. q The measured values ​​of each matching indicator belong to the risk level. G r The degree to which the target cloud is met; the larger the value, the closer the indicator is to meeting the target level. Then, the overall risk matching degree of the region is calculated. The higher the value, the higher the degree of matching between the emergency resources and the risk level of the region. To determine the matching level, four matching cloud levels are defined: severely undermatched, undermatched, basic match, and good match. Calculate the degree of certainty of the overall risk matching degree for each matching level in this region; The matching level determination result for this region is one of the following: severely undermatched, undermatched, basically matched, or well matched. That is, the matching level with the highest certainty is taken as the final matching level for this region.

6. The method for power grid risk assessment and collaborative optimization of emergency resource allocation under extreme operating conditions as described in claim 1, characterized in that, In the multi-objective optimization model, the first optimization objective is to maximize the matching degree between emergency resources and risk levels in each region after adjustment. Optimization objective two: Minimize the movement path cost of cross-regional emergency resource allocation; The constraints include: adjusted resource capacity constraints, adjusted resource matching degree constraints, matching compliance constraints in high-risk areas, no mismatch constraints in normal areas, allocation time constraints, and linkage constraints of allocation variables.

7. A power grid risk assessment and emergency resource collaborative optimization and allocation system under extreme operating conditions, characterized by: include: The module for determining the final risk level of each region is configured to: acquire power grid operation data and weather forecast data for each region, obtain the time-domain trajectory of electrical quantities for each region, conduct a comprehensive assessment of each region based on the acquired time-domain trajectory of electrical quantities, and determine the final risk level of each region; The matching degree quantitative assessment module is configured to: establish an evaluation cloud model, quantitatively assess the matching degree between emergency resources and risk levels based on the cloud model, calculate the regional risk comprehensive matching degree index, and identify severely mismatched areas; The resource allocation module is configured to: for areas where emergency resources and risk levels are severely mismatched, construct a multi-objective optimization model with the objectives of maximizing the matching degree and minimizing the movement distance of emergency resources, optimize the allocation of emergency resources in different areas under the premise of satisfying traffic constraints and ensuring no mismatch in normal areas, and propose a hierarchical solution method to determine the type, capacity and path of resource allocation.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 6.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method described in any one of claims 1-6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it performs the steps of the method described in any one of claims 1-6 above.