Method and device for correcting earthquake disaster casualty assessment model and storage medium

Abstract

The application discloses a correction method and device of a seismic disaster personnel casualty assessment model and a storage medium, and relates to the technical field of disaster emergency management. By constructing an emergency preparedness index and an emergency response index as a nonlinear attenuation correction factor, the traditional assessment model's defects of relying on expert experience for manual correction and lacking in automation and intelligence are overcome. The coupling correction formula takes the nonlinear attenuation correction factor as the core, and through the multiplication coupling of independent double indexes, the synergistic attenuation effect of emergency preparedness and emergency response is accurately reflected. Through a closed-loop optimization mechanism, highly reliable data support is provided for emergency command. The correction process not only outputs personnel casualty prediction values, but also quantitatively diagnoses regional resilience by generating an emergency preparedness index and an emergency response index, so that emergency plans can be optimized, rescue force deployment can be adjusted or popular science drills can be strengthened, thereby directly serving the active prevention and control and rapid recovery goals of a resilient city.

Description

Technical Field

[0001] This application relates to the field of disaster emergency management technology, specifically to a method, device, and storage medium for correcting an earthquake disaster casualty assessment model. Background Technology

[0002] Earthquakes are a type of natural disaster characterized by their suddenness, destructive power, and wide-ranging impact. They can easily cause a large number of casualties and huge property losses. Accurate casualty assessment is a core prerequisite for earthquake emergency command, allocation of rescue resources, and disaster loss accounting. It is of great significance for improving the efficiency of earthquake emergency response and reducing casualties.

[0003] Traditional earthquake casualty assessment models primarily rely on building vulnerability. Their core logic involves analyzing the destructive effects of seismic parameters on different types of buildings to determine the degree of damage, and then quantifying casualties by considering the population density of the area. While this method can provide a preliminary estimate of casualties by considering only building damage and population density as key influencing factors, it fails to account for the impact of regional emergency response capabilities in real-world scenarios, leading to significant discrepancies between the assessment results and actual casualties. Furthermore, the methods lack automation and intelligence, requiring manual verification and correction by experts, making them highly dependent on expert experience and failing to meet the demands of emergency command for timely, accurate, and reliable data.

[0004] With the promotion of the concept of resilient cities, the core of urban disaster prevention and control has shifted from "passive response" to "proactive prevention and control + rapid response". Traditional earthquake disaster casualty assessment models and methods do not consider the impact of the actual emergency environment (emergency preparedness, emergency response) on casualty results, resulting in insufficient rationality of earthquake disaster casualty assessment models, limited practicality of implementation methods, and large deviation between assessment results and actual casualty situation. They are difficult to support the optimization of emergency decision-making in resilient cities and therefore cannot meet the needs of resilient cities for accuracy and comprehensiveness in casualty assessment. Summary of the Invention

[0005] The purpose of this application is to provide a method, device, and storage medium for correcting an earthquake disaster casualty assessment model, in order to solve the problem that the traditional earthquake disaster assessment results deviate significantly from the actual casualty situation, making it difficult to meet the emergency decision-making needs of resilient cities.

[0006] To achieve the above objectives, the first aspect of this application provides a method for revising an earthquake disaster casualty assessment model, comprising:

[0007] Obtain emergency preparedness capability indicators and emergency response status indicators for the correction area;

[0008] Based on the emergency preparedness capability index data and the emergency response status index data, an emergency preparedness capability index and an emergency response status index are constructed, and a coupled correction formula is constructed with the emergency preparedness capability index and the emergency response status index as nonlinear attenuation correction factors.

[0009] Based on the initial assessment value output by the earthquake disaster casualty assessment model based on building vulnerability, the corrected target assessment value is calculated by combining the emergency preparedness capability index, the emergency response status index, and the coupling correction formula.

[0010] Based on historical earthquake casualty data of the correction area or adjacent similar areas of the correction area, the corrected target evaluation value is checked for deviation.

[0011] If the deviation between the target evaluation value and the reference data is greater than or equal to a preset deviation value, then the index weights of the emergency preparedness capability index and the emergency response status index are adjusted, the coupling correction formula is reconstructed, and the target evaluation value is calculated again until the deviation between the target evaluation value and the reference data is less than the preset deviation value.

[0012] A second aspect of this application provides a correction device for an earthquake disaster casualty assessment model, comprising:

[0013] The acquisition module is used to acquire emergency preparedness capability index data and emergency response status index data for the correction area;

[0014] The construction module is used to construct an emergency preparedness index and an emergency response index based on the emergency preparedness capability index data and the emergency response status index data, and to construct a coupled correction formula with the emergency preparedness capability index and the emergency response status index as nonlinear attenuation correction factors.

[0015] The calculation module is used to calculate the corrected target assessment value based on the initial assessment value output by the earthquake disaster casualty assessment model based on building vulnerability, combined with the emergency preparedness capability index, the emergency response status index and the coupling correction formula.

[0016] The verification module is used to verify the deviation of the corrected target evaluation value based on historical earthquake casualty data of the corrected area or adjacent similar areas of the corrected area.

[0017] The adjustment module is used to adjust the index weights of the emergency preparedness capability index and the emergency response status index if the deviation between the target evaluation value and the reference data is greater than or equal to a preset deviation value, reconstruct the coupling correction formula, and recalculate the target evaluation value until the deviation between the target evaluation value and the reference data is less than the preset deviation value.

[0018] A third aspect of this application provides a computer-readable storage medium storing a program that can be loaded by a processor and executed by the above-described method for correcting an earthquake disaster casualty assessment model.

[0019] The beneficial effects of this application are:

[0020] This application constructs an emergency preparedness capability index and an emergency response status index as nonlinear attenuation correction factors to correct the initial assessment value based on building vulnerability. It systematically embeds non-engineering emergency factors into the traditional earthquake disaster casualty assessment model, effectively overcoming the shortcomings of traditional models that rely on expert experience for manual correction and lack automation and intelligence. The coupled correction formula, with the nonlinear attenuation correction factor at its core, conforms to the objective law that stronger emergency response capabilities lead to lower casualties. Furthermore, through the multiplicative coupling of independent dual exponents, it accurately reflects the synergistic attenuation effect of emergency preparedness and emergency response, resolving the contradiction in the physical meaning of traditional linear corrections. This significantly improves the rationality of the earthquake disaster casualty assessment model.

[0021] The revised target assessment value is validated using historical earthquake casualty data or data from adjacent similar areas. If the deviation is greater than or equal to the preset deviation value, the weights of the emergency preparedness index and the emergency response index are adjusted, the coupled correction formula is reconstructed, and iterative calculations are performed. A closed-loop optimization mechanism provides highly reliable data support for emergency command. The correction process not only outputs predicted casualty values ​​but also provides a quantitative diagnosis of regional resilience by generating emergency preparedness and emergency response indices. This allows for targeted optimization of emergency plans, adjustment of rescue force deployment, or strengthening of public awareness drills, directly serving the proactive prevention and rapid recovery goals of resilient cities.

[0022] Other features and advantages of this application will be described in detail in the following detailed description section. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating a method for correcting an earthquake disaster casualty assessment model provided in an embodiment of this application.

[0024] Figure 2 This is a schematic diagram of the structure of a correction device for an earthquake disaster casualty assessment model provided in an embodiment of this application. Detailed Implementation

[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0026] In the description of this application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "a plurality of" means two or more, unless otherwise explicitly specified. Details are set forth in the following description for illustrative purposes. It should be understood that those skilled in the art will recognize that this application can be implemented without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid unnecessarily obscuring the description of this application. Therefore, this application is not intended to be limited to the embodiments shown, but rather to be consistent with the broadest scope of the principles and features disclosed herein.

[0027] Figure 1 This is a flowchart illustrating a method for correcting an earthquake disaster casualty assessment model provided in an embodiment of this application. Figure 1 As shown, the correction method may include steps 101-105, which will be described in detail below.

[0028] Step 101: Obtain emergency preparedness capability index data and emergency response status index data for the correction area.

[0029] The revised area refers to a specific geographical region (such as a city, district, or township) where earthquake casualties assessment is required. Its boundaries can be determined based on administrative divisions, geological risk zoning, or emergency management needs. Emergency preparedness capability indicators reflect the foundation of earthquake prevention and control in the revised area and may include quantitative data on personnel emergency drills, emergency plans, early warning terminal configuration, and earthquake science education. Emergency response indicators reflect the effectiveness of post-earthquake relief efforts in the revised area and may include quantitative or qualitative data on emergency response systems, rescue forces, and relief supplies.

[0030] In one example, the modified area to be evaluated can be identified (e.g., a municipal administrative district or a specific grid unit). Subsequently, two types of core data are acquired through multi-source data collection channels: emergency preparedness indicator data and emergency response status indicator data. For example, data on the static defense and preparedness status of the modified area before the earthquake is collected. This includes: per capita emergency shelter area, density of emergency material reserves, coverage rate of community emergency plans, residents' awareness of earthquake prevention and disaster reduction knowledge (obtained through questionnaires or records of popular science activities), and the proportion of professional rescue teams deployed. Data on the dynamic response of the modified area after the earthquake is also collected (if it is a predicted scenario, estimates based on historical drills or simulations are used). This includes: average rescue arrival time, speed of medical treatment point activation, communication restoration time, traffic management efficiency, and the proportion of volunteers mobilized. After data cleaning, the above qualitative or quantitative data are standardized into a numerical sequence with uniform dimensions.

[0031] This step breaks away from the limitations of traditional earthquake disaster casualty assessment models that rely solely on population and building data, quantifying soft power (management and response) as input parameters. By collecting multi-dimensional data, it comprehensively reflects a region's true resilience in the face of earthquake disasters, laying a solid data foundation for the subsequent construction of a refined and corrected earthquake disaster casualty assessment model.

[0032] Step 102: Based on emergency preparedness capability index data and emergency response status index data, construct the emergency preparedness capability index and the emergency response status index, and construct a coupled correction formula with the emergency preparedness capability index and the emergency response status index as nonlinear attenuation correction factors.

[0033] The emergency preparedness capability index is a comprehensive index (range [0,1]) obtained by standardizing and quantifying emergency preparedness capability indicator data and then summing it using dynamic weights. A higher index indicates stronger pre-disaster prevention and control capabilities. The emergency response status index is a comprehensive index (range [0,1]) obtained by standardizing and quantifying emergency response status indicator data and then summing it using dynamic weights. A higher index indicates stronger post-disaster relief effectiveness.

[0034] The coupled correction formula is the core formula for the initial assessment value of the dual emergency index and the traditional earthquake disaster casualty assessment model, enabling an organic combination of traditional physical assessment and emergency capability correction. The coupled correction formula includes a nonlinear attenuation correction factor. This nonlinear attenuation correction factor is an independent attenuation factor constructed based on an exponential function, including the emergency preparedness capability index and the emergency response status index. Each nonlinear attenuation correction factor includes a corresponding weighting coefficient, used to characterize the attenuation effect of emergency capability on casualties, conforming to the objective law that stronger emergency capability leads to lower casualties.

[0035] By transforming abstract emergency response capabilities into quantifiable dual indices, this approach addresses the industry pain point of integrating emergency factors into earthquake disaster casualty assessment models, enabling standardized assessment of non-engineering factors. Dynamic weight allocation adapts to different regional characteristics and disaster levels, making index construction more targeted. The coupled correction formula aligns with the actual operational patterns of emergency response capabilities, reducing the physical inconsistencies inherent in traditional linear corrections.

[0036] Step 103: Based on the initial assessment value output by the earthquake disaster casualty assessment model based on building vulnerability, the target assessment value is calculated by combining the emergency preparedness capability index, the emergency response status index, and the coupled correction formula.

[0037] The initial assessment value is a predicted casualty figure calculated based on a traditional earthquake disaster casualty assessment model, considering only the degree of building damage and population density. In one example, using a traditional earthquake disaster casualty assessment model based on building vulnerability, the initial casualty assessment value can be calculated by inputting the building structure type, construction year, seismic resistance level (building vulnerability parameter), and population density data of the correction area. However, the initial assessment value requires manual correction by experts and has a low level of automation.

[0038] The target assessment value is the final casualty assessment value calculated using the coupled correction formula and verified for deviation, taking into account both engineering and non-engineering factors. In one example, this embodiment of the application incorporates the concept of resilient cities, substituting the weight coefficients of the constructed emergency preparedness index, emergency response status index, and calibrated nonlinear attenuation correction factor into the coupled correction formula, and obtaining the target assessment value through multiplication.

[0039] In this way, the traditional manual correction relying on expert experience is transformed into automatic model correction, improving the authenticity and practicality of the assessment results and directly supporting emergency command without human intervention. Furthermore, the calculation process is simple and repeatable, and can be quickly implemented by those skilled in the art through standardized procedures, requiring no additional creative effort and facilitating practical application.

[0040] Step 104: Based on historical earthquake casualty data of the correction area or adjacent similar areas of the correction area, perform deviation verification on the corrected target assessment value.

[0041] In this embodiment, historical earthquake casualty data of the correction area (such as the actual number of casualties from similar magnitude earthquakes that have occurred in the area) can be preferentially selected, and the data must be complete and accurately recorded. If there is no historical data of the same type in the correction area, historical earthquake casualty data of adjacent similar areas (such as areas with similar geographical environment, population size, economic level, and building characteristics) can be selected. Next, the deviation rate is calculated using the relative deviation formula, and the calculated deviation rate is compared with the preset deviation value for deviation verification. For example, if the deviation rate is less than the set percentage, i.e., the preset deviation value, the target evaluation value meets the accuracy requirements and is directly output. If the deviation rate is greater than or equal to the set percentage, i.e., the preset deviation value, it indicates that the current indicator weight allocation is unreasonable, and the correction method of the earthquake disaster casualty assessment model fails to accurately reflect the actual emergency response capability of the correction area. In this case, step 105 is entered for optimization and adjustment. By verifying historical data, the reliability and rationality of the evaluation results are ensured, reducing the problem of the calculation of the earthquake disaster casualty assessment model correction method being out of touch with the actual scenario, and providing accurate data support for emergency decision-making. The reference data selection rules are flexible and can be adapted to the data source conditions of different correction areas, expanding the applicability of the earthquake disaster casualty assessment model correction method.

[0042] Step 105: If the deviation between the target assessment value and the reference data is greater than or equal to the preset deviation value, adjust the index weights of the emergency preparedness capability index and the emergency response status index, reconstruct the coupling correction formula, and recalculate the target assessment value until the deviation between the target assessment value and the reference data is less than the preset deviation value.

[0043] The preset deviation value is an assessment accuracy threshold determined based on industry standards and historical data statistics, used to judge the reasonableness of the target assessment value. The weights of the two indices can be adjusted according to the direction and magnitude of the deviation. For example, if the target assessment value is significantly higher than the actual value, it indicates that the mitigating effect of emergency response capabilities on casualties is underestimated, and the weight of the corresponding emergency response index can be increased. Conversely, if the target assessment value is significantly lower than the actual value, it indicates that the mitigating effect of emergency response capabilities on casualties is overestimated, and the weight of the corresponding emergency response index can be decreased.

[0044] Next, based on the adjusted weights, the emergency preparedness capability index and the emergency response status index are recalculated. The coupling correction formula is updated synchronously (the formula form remains unchanged after the weight adjustment, only the values ​​of the two indices are updated), and the initial assessment values ​​are substituted back into the formula to calculate the new target assessment values. The deviation check in step 104 is repeated. If the deviation rate is still greater than or equal to the preset deviation value, the weights are adjusted again until the deviation rate is less than the preset deviation value.

[0045] A closed-loop mechanism of construction, calculation, verification, and optimization is established to enable the revised earthquake disaster casualty assessment model to adaptively adjust and adapt to the assessment needs of different regions and disaster scenarios, thereby improving the model's versatility and robustness. By precisely identifying the shortcomings in emergency response capability assessment through weight adjustments, not only are the assessment results optimized, but targeted directions are also provided for subsequent emergency response capability enhancement.

[0046] This application's embodiments systematically embed non-engineering emergency factors into the traditional earthquake disaster casualty assessment model, effectively overcoming the shortcomings of traditional earthquake disaster casualty assessment models that rely on expert experience for manual correction and lack automation and intelligence. The coupled correction formula uses a nonlinear attenuation correction factor as its core, conforming to the objective law that the stronger the emergency response capability, the lower the casualty rate. Furthermore, through independent double-exponential multiplicative coupling, it accurately reflects the synergistic attenuation effect of emergency preparedness and emergency response, resolving the defect of the contradictory physical meaning of traditional linear correction. The rationality of the corrected earthquake disaster casualty assessment model is significantly improved. A closed-loop optimization mechanism provides highly reliable data support for emergency command. The correction process not only outputs the predicted casualty value but also generates an emergency preparedness capability index and an emergency response status index to quantitatively diagnose regional resilience. This allows for targeted optimization of emergency plans, adjustment of rescue force deployment, or strengthening of science popularization drills, directly serving the proactive prevention and rapid recovery goals of resilient cities.

[0047] In step 101, emergency preparedness indicator data can be constructed based on the personnel drill indicator parameters, emergency plan indicator parameters, early warning terminal indicator parameters, and earthquake science popularization indicator parameters of the modified area.

[0048] Personnel drill indicators are quantitative data reflecting the improvement of self-rescue and mutual rescue skills and coordination capabilities in the modified area through simulated combat before an earthquake. They are elements for measuring the level of civil defense and can include the annual frequency of earthquake emergency drills and the percentage of participants (e.g., number of participants / total population of the area × 100%) to reflect the level of training of the region's personnel's emergency response capabilities.

[0049] Emergency response plan indicators are parameters that characterize the quality and practicality of earthquake emergency response plans in the revised area. These include the completeness of plan modules (such as whether core modules such as risk assessment and evacuation routes are covered), update frequency (such as the number of updates per unit time), and operability score (such as the quantitative evaluation of the difficulty of implementing the plan by experts).

[0050] The early warning terminal index parameters are parameters that characterize the deployment and operation status of earthquake early warning terminals in the correction area. These include the number of early warning terminals deployed, the total number of terminals suitable for deployment in the area, and the number of terminals operating normally. Derivative indicators include early warning terminal coverage rate (e.g., number of deployed terminals / total number of suitable deployments × 100%) and normal operation rate (e.g., number of terminals operating normally / number of deployed terminals × 100%).

[0051] Earthquake science popularization indicators are parameters that characterize the coverage and frequency of earthquake science popularization within the correction area. They include the number of science popularization activities per year and the percentage of people covered (e.g., the number of people covered / the total population of the area × 100%), reflecting the degree of public awareness of earthquake protection knowledge in the area.

[0052] By refining the parameters across four dimensions, the abstract emergency preparedness capability is concretized into measurable and quantifiable data indicators. For example, parameters related to personnel drills and public awareness campaigns are directly linked to the probability of residents' self-rescue and mutual aid, effectively reducing secondary casualties caused by panic. Emergency plan parameters ensure the orderly organization during a disaster. Early warning terminal parameters are directly related to the success rate of pre-earthquake emergency evacuation. This multi-dimensional data collection enables the revised earthquake disaster casualty assessment model to accurately identify areas with strong hardware but weak software (e.g., abundant equipment but lack of user skills) or high awareness but lack of contingency plans, providing fine-grained input for the subsequent construction of high-precision correction factors.

[0053] At the same time, emergency response status indicator data can be constructed based on the emergency response system indicator parameters, rescue force indicator parameters, and rescue material indicator parameters of the corrected area.

[0054] Emergency response system indicators are parameters that characterize the operational effectiveness of the emergency response platform in the correction area. These include response start time (e.g., the time from the occurrence of a disaster to system startup), data transmission latency (e.g., the time from the issuance of instructions / data to their receipt), and the success rate of multi-department collaborative scheduling (e.g., the number of times collaborative scheduling was completed / the total number of scheduling times × 100%).

[0055] The rescue force index parameters are parameters that characterize the configuration and capability of emergency rescue teams in the modified area. These include the number of professional rescue teams, the ratio of rescue personnel per 10,000 people (e.g., total number of rescue personnel / total population of the area × 10,000), the certification rate of rescue personnel (e.g., number of certified personnel / total number of rescue personnel × 100%), and the completeness of rescue equipment (e.g., number of complete core equipment / total number of core equipment that should be equipped × 100%).

[0056] The parameters of relief supplies indicators are parameters that characterize the relief supplies reserve and allocation capacity of the modified area, including the reserve of supplies per 10,000 people (e.g., total quantity of supplies / total population of the area × 10,000), the coverage rate of core material types (e.g., the quantity of core categories already reserved / the total number of core categories that should be reserved × 100%), and the response time for material allocation (e.g., the time from issuing an allocation order to the arrival of materials).

[0057] The detailed breakdown of emergency response indicators accurately captures key dynamic variables affecting post-earthquake casualties. The high efficiency of the emergency response system significantly shortens the command transmission chain, reducing rescue delays caused by command confusion. The quantification of rescue forces (especially arrival time and equipment level) directly corresponds to the success rate of search and rescue within the critical 72 hours, significantly reducing the mortality rate of those buried. The rapid turnover of relief supplies effectively prevents additional casualties caused by post-disaster epidemics and secondary survival crises. Incorporating these dynamic response parameters into the revised earthquake disaster casualty assessment model transforms the assessment result from a static theoretical value into a dynamic prediction that fully considers the speed of rescue and the efficiency of material preservation, making the revised earthquake disaster casualty assessment model more applicable and accurate in real disaster scenarios.

[0058] In step 102, the frequency and participation rate of annual earthquake emergency drills in the modified region can be used as personnel drill indicator parameters. The completeness, update frequency, and operability score of the earthquake emergency plan in the modified region can be used as emergency plan indicator parameters. The number of earthquake early warning terminals deployed, coverage density, and operational normalization rate in the modified region can be used as early warning terminal indicator parameters. The frequency, coverage, and scope of earthquake science popularization activities in the modified region can be used as earthquake science popularization indicator parameters. Then, the personnel drill indicator parameters, emergency plan indicator parameters, early warning terminal indicator parameters, and earthquake science popularization indicator parameters are weighted and summed to obtain the emergency preparedness capability index. The emergency preparedness capability index is used to quantify the basic level of emergency preparedness in the modified region before an earthquake occurs, reflecting the region's proactive ability to resist earthquake disasters and reduce casualties. It is highly consistent with the core objective of resilient city risk prevention and control and is one of the core input factors of the modified earthquake disaster casualty assessment model.

[0059] In one example, the specific formula for calculating the emergency preparedness index P is:

[0060] P=α1×A+α2×B+α3×C+α4×D.

[0061] A represents the quantified value of the personnel drill indicator parameter, ranging from [0,1]. It is calculated based on the frequency of annual earthquake emergency drills and the percentage of participants relative to the on-duty employees of participating units within the correction area. For example, if the frequency of annual earthquake emergency drills is ≥2 and the percentage of participants is ≥60%, A=1. If the frequency of annual earthquake emergency drills is 1 and the percentage of participants is between 30% and 60%, A=0.6. If the frequency of annual earthquake emergency drills is 0 or the percentage of participants is <30%, A=0.2. All other cases are calculated using linear interpolation.

[0062] B represents the quantified value of the emergency response plan's indicator parameters, ranging from [0,1]. It is calculated based on the completeness, update frequency, and operability scores of the earthquake emergency response plan within the modified area. For example, if the emergency response plan includes complete modules such as risk assessment, evacuation routes, and rescue procedures, and is updated at least once a year and can be directly implemented, B=1. If the modules are complete, updated every two years, and have moderate operability, B=0.6. If the modules are missing, not updated, or cannot be implemented, B=0.2. All other cases are calculated using linear interpolation.

[0063] C represents the quantified value of the early warning terminal indicator parameters, ranging from [0,1]. It is calculated based on the early warning terminal coverage rate and normal operation rate within the correction area. For example, when the early warning terminal coverage rate is ≥90% and the normal operation rate is ≥95%, C=1. When the coverage rate is between 70% and 90% and the normal operation rate is between 85% and 95%, C=0.6. When the coverage rate is <70% or the normal operation rate is <85%, C=0.2. Other cases are calculated using linear interpolation. Wherein, early warning terminal coverage rate = number of early warning terminals deployed / total suitable deployment area, and normal operation rate = number of normally operating terminals / number of early warning terminals deployed.

[0064] D is the quantified value of the earthquake science popularization indicator parameter, ranging from [0,1]. It is calculated based on the number of earthquake emergency science popularization events and the number of participants relative to the resident population within the correction area each year. For example, if the number of science popularization events is ≥4 and the participation rate is ≥50%, D=1. If the number of science popularization events is 2 and the participation rate is between 25% and 50%, D=0.6. If the number of science popularization events is <2 or the participation rate is <25%, D=0.2. All other cases are calculated using linear interpolation.

[0065] α1, α2, α3, and α4 are the weights of each indicator, and α1 + α2 + α3 + α4 = 1. The weight values ​​can be dynamically adjusted according to the population size, economic development level, and geological environment characteristics of the correction area. For example, for densely populated, economically developed, and geologically hazardous areas, α3 (the weight of the early warning terminal indicator parameter) is set to 0.3-0.4, and α1, α2, and α4 are set to 0.2-0.3 respectively. For sparsely populated, economically underdeveloped, and low-risk areas, α1 (the weight of the personnel drill indicator parameter) is set to 0.4, and α2, α3, and α4 are set to 0.2 respectively.

[0066] In step 102, the response speed and data transmission latency of the emergency response system in the modified region can also be used as indicators of the emergency response system. The number of professional rescue teams, the certification rate of rescue personnel, the availability of rescue equipment, and the cross-regional rescue coordination capabilities in the modified region can be used as indicators of rescue strength. The quantity, distribution, allocation efficiency, and replenishment capacity of rescue supplies in the modified region can be used as indicators of rescue supplies. The emergency response system indicators, rescue strength indicators, and rescue supplies indicators are weighted and summed to obtain the emergency response status index. The emergency response status index is used to quantify the emergency response effectiveness of the modified region after the earthquake, reflecting the region's ability to quickly carry out rescue operations and reduce casualties. It is highly consistent with the core objective of rapid recovery in resilient cities and is another core input factor in the modified earthquake disaster casualty assessment model. The approach combines quantitative and qualitative methods. Core indicators (response speed, quantity of materials, etc.) are scored numerically, while auxiliary indicators (coordination efficiency, command and dispatch capabilities, etc.) are quantified after being graded qualitatively (excellent / good / average / poor). The results are then integrated through weight allocation to obtain an emergency response index (range 0-1, the higher the index, the stronger the emergency response effectiveness).

[0067] In one example, the specific formula for calculating the emergency response index R is:

[0068] R = β1 × E + β2 × F + β3 × G.

[0069] E is a quantified value of the emergency response system indicator, ranging from [0,1]. It is calculated based on the emergency response system's response speed, data transmission efficiency, and collaborative scheduling capability. For example, when data transmission delay is ≤1 minute and multi-department collaborative scheduling is possible, E=1. When the response time is 5-10 minutes, transmission delay is 1-3 minutes, and collaborative scheduling capability is average, E=0.6. When the response time is >10 minutes, transmission delay is >3 minutes, or collaborative scheduling is not possible, E=0.2. Other cases are calculated using linear interpolation.

[0070] F is the quantified value of the rescue force indicator parameter, ranging from [0,1]. It is quantified based on the number of professional rescue teams in the correction area, the certification rate of rescue personnel (i.e., the qualifications of rescue personnel), and the availability of emergency rescue equipment. When there are ≥5 professional rescue personnel per 10,000 people, a certification rate of ≥90%, and complete equipment for demolition, search and rescue, and emergency medical care, F=1. When there are 2-5 professional rescue personnel per 10,000 people, a certification rate of 70%-90%, and basic equipment availability, F=0.6. When there are <2 professional rescue personnel per 10,000 people, a certification rate <70%, or severe equipment deficiencies, F=0.2. Other cases are calculated using linear interpolation.

[0071] G is the quantified value of the relief supplies indicator parameter, ranging from [0,1]. It is calculated based on the reserve quantity, reserve types, allocation efficiency, and replenishment capacity (such as coverage area) of the relief supplies. For example, when the relief supplies reserve quantity meets the following conditions: ≥50 supplies per 10,000 people, reserve types cover core categories such as medical supplies, food, and tents, allocation response time ≤30 minutes, and coverage of all key areas, G=1. When the relief supplies reserve quantity is <50 supplies per 10,000 people, types are basically complete, allocation response time is 30-60 minutes, and coverage of major areas, G=0.6. When the relief supplies reserve quantity is <10 supplies per 10,000 people, types are missing, allocation response time is >60 minutes, or coverage area is limited, G=0.2. Other cases are calculated using linear interpolation.

[0072] β1, β2, and β3 are the weights of each indicator, and β1 + β2 + β3 = 1. The weight values ​​are dynamically adjusted according to the earthquake disaster level and the scope of the correction area. For example, for major earthquakes and large-scale correction areas, β2 (weight of rescue force indicator) and β3 (weight of rescue material indicator) are respectively set to 0.35-0.4, and β1 is set to 0.2-0.3. For minor earthquakes and small-scale correction areas, β1 (weight of emergency response system indicator) is set to 0.4, and β2 and β3 are respectively set to 0.3.

[0073] In step 102, four main construction principles (constraints) are first determined. These principles include ensuring that the modified target assessment value is no greater than the initial assessment value, that the rate of casualty decline is faster when emergency response capability improves from a low to a medium level, that the rate of casualty decline slows down when emergency response capability approaches saturation, and that the formula is mathematically calibrable. These constraints are used to construct the coupled correction formula. The constraints of the coupled correction formula adhere to four technical principles. First, there is a non-incremental constraint, ensuring that the modified target assessment value (predicted modified assessment of casualty numbers C) is never greater than the initial assessment value based on building vulnerability (theoretical maximum number of casualties C0) under any circumstances, i.e., C ≤ C0. Second, there is a diminishing marginal effect constraint, simulating real-world patterns. When emergency response capability (P or R) improves from a low to a medium level, the rate of casualty decline should be faster (high sensitivity). When capability approaches saturation (high level), the rate of casualty decline should slow down to avoid infinite decline. Third, there is an independence constraint, acknowledging that emergency preparedness and emergency response are two relatively independent dimensions of action, requiring separate attenuation mechanisms. Fourth, there is the constraint of mathematical calibrability, which requires that the formula structure must support parameter inversion and fitting using historical data.

[0074] Then, define the non-linear attenuation factor. The non-linear attenuation factor is a mathematical term whose function value decreases exponentially and asymptotically approaches zero as the independent variable increases, accurately describing the law of "diminishing marginal benefit" of emergency investment, that is, the initial investment has a significant effect, and the effect of later investment slows down. Define the first attenuation factor in exponential form for the emergency preparedness capacity index P, and define the first attenuation factor as the exponential function form e^(-αP). Define the second attenuation factor e^(-βR) in exponential form for the emergency response situation index R. The first attenuation factor and the second attenuation factor are independent non-linear attenuation factors. Based on the initial evaluation value of the earthquake disaster casualty assessment model, multiply the first attenuation factor and the second attenuation factor to obtain the initial coupling correction formula. Multiplicative coupling means that multiple correction factors act on the base value in the form of a product. Compared with additive coupling, multiplicative coupling can better reflect the "barrel effect", that is, if either the preparedness capacity or the response capacity is zero (or extremely low), the overall disaster reduction effect will be greatly limited, which conforms to the collaborative logic of disaster rescue.

[0075] Among them, the initial coupling correction formula satisfies:

[0076] C = C0 × e^(-αP) × e^(-βR).

[0077] Among them, C is the corrected evaluation casualty number, C0 is the initial evaluation value, P is the emergency preparedness capacity index, R is the emergency response situation index, e^(-αP) is the first attenuation factor, e^(-βR) is the second attenuation factor, α is the first weight coefficient of the first attenuation factor, β is the second weight coefficient of the second attenuation factor, and α > 0 and β > 0.

[0078] In the process of selecting the correction formula of the earthquake disaster casualty assessment model in the embodiments of this application, the core derivation basis is carried out around four dimensions: physical meaning self-consistency, historical data fitting suitability, mathematical rigor, and practical application feasibility. Through combining and fitting and verifying multiple groups of historical earthquake case data, it is finally determined to use the double exponential function to correct the earthquake disaster casualty assessment model. The reasons are as follows.

[0079] First of all, the double exponential function e^(-x) is a monotonically decreasing function, and 0 < e^(-x) ≤ 1, which can naturally ensure that the corrected evaluation casualty number C ≤ C0, conforming to the law that emergency intervention can only reduce casualties and cannot increase casualties, without the need to add additional constraint conditions. Secondly, the double exponential form can accurately reflect the law of diminishing marginal effect, which is completely matched with the attenuation characteristics of emergency capacity on casualties. Then, the double exponential function can be transformed into a linear form by taking the logarithm, which is convenient for calibrating the weight coefficient by the least squares method, and the calculation is simple and the operability is strong.

[0080] The derivation process of the initial coupling correction formula includes the following steps.

[0081] Step 1: Define the casualty attenuation coefficient corresponding to emergency preparedness capability as e^(-αP) and the casualty attenuation coefficient corresponding to emergency response capability as e^(-βR), where α and β are weighting coefficients, representing the degree of contribution of the two emergency capabilities to casualty attenuation, α>0 and β>0.

[0082] Step 2: Since the attenuation effects of the two emergency response capabilities are independent and can be superimposed, the total attenuation coefficient is the product of the two individual attenuation coefficients, i.e., e^(-αP)×e^(-βR)=e^(-(αP+βR)).

[0083] Step 3: Correct the assessment of casualties C = the initial assessment value C0 of the traditional earthquake disaster casualty assessment model × the total attenuation coefficient, which gives the initial coupling correction formula of the corrected earthquake disaster casualty assessment model: C = C0 × e^(-αP) × e^(-βR).

[0084] The double exponential function-corrected earthquake casualty assessment model aligns with the actual impact of emergency preparedness capabilities. The influence of the emergency preparedness index P and the emergency response index R on earthquake casualties is essentially a non-linear attenuation effect—casualties decrease rapidly when emergency preparedness improves from 0 to a moderate level. As emergency preparedness approaches saturation, the rate of decrease gradually slows, exhibiting a clear diminishing marginal effect, which is an objective reality of earthquake emergency scenarios. The initial coupled correction formula uses e^(-αP)×e^(-βR) as an independent attenuation factor. The inherent characteristics of the exponential function accurately match this non-linear attenuation law. Furthermore, because the larger the emergency preparedness coefficients P and R, the smaller the attenuation factor, and the smaller the corrected casualty assessment number C, it perfectly matches the core logic that stronger emergency preparedness leads to lower casualties.

[0085] In one example, the process may also include a step of fitting and calibrating the first and second weighting coefficients using the least squares method. For instance, the least squares method is used to fit and calibrate the weighting coefficients α and β using four sets of historical earthquake case data (three sets of valid cases are selected), ensuring that the calibration process is repeatable and verifiable. The specific steps are as follows.

[0086] Specifically, multiple sets of historical earthquake casualty data can be collected first. Each set of historical earthquake casualty data can include the initial assessment value C0 of the earthquake disaster casualty assessment model, the emergency preparedness index P, the emergency response status index R, and the actual number of casualties Cactual.

[0087] For example, collect data from 4 sets of historical earthquake cases. Each set of cases should include C0 (initial assessment value of the traditional earthquake disaster casualty assessment model), P (emergency preparedness capability coefficient), R (emergency response capability coefficient), and Cactual (actual number of casualties). Specific data are shown in Table 1, which is an example of the collection of historical case data.

[0088] Table 1

[0089]

[0090] Since the double exponential function requires subsequent logarithmic operations, and the first case C i=0, ln(0) is meaningless, the first case is removed, and the second, third and fourth cases are selected as valid fitting cases.

[0091] To facilitate the solution, the initial coupling correction formula can be linearized by taking the natural logarithm to obtain a linear fitting model. For example, taking the natural logarithm of both sides of the initial coupling correction formula C = C0 × e^(-αP) × e^(-βR) transforms the nonlinear model into a linear model, making it easier to fit using the least squares method: ln(C) = ln(C0) - αP - βR. Rearranging, we get: ln(C / C0) = -αP - βR. Let Y = ln(C / C0), then the linear model can be expressed as: Y = -αP - βR, where Y is the dependent variable, P and R are the independent variables, and α and β are the weighting coefficients to be calibrated.

[0092] The goal is to minimize the sum of squared residuals between the fitted values ​​and the actual values ​​of the linear fitting model. A least-squares objective function is then constructed, incorporating historical earthquake casualty data into this function. For example, to minimize the sum of squared residuals between the actual value Y_i and the fitted value (-αP_i-βR_i) in valid cases, the least-squares objective function is: minS(α,β)=Σ[Y_i-(-αP_i-βR_i)] 2 =Σ[Y_i+αP_i+βR_i] 2 (i=2,3,4, i.e., valid case numbers). Substitute the valid case numbers Y_i, P_i, and R_i into the equation. The objective function is: S(α,β)=[-0.470004+0.765α+0.92β] 2 +[-0.916291+0.90α+0.93β] 2 +[-0.282330+0.90α+0.94β] 2 .

[0093] Next, the partial derivatives of the first and second weight coefficients of the least squares objective function are calculated and set to zero. A system of objective equations is then established, and the optimal estimates of the first and second weight coefficients are obtained by solving the system.

[0094] For example, by taking the partial derivatives of α and β with respect to the objective function S(α,β) and setting the partial derivatives to 0, we obtain a system of normal equations. Solving the system of equations will give us the optimal estimates of α and β.

[0095] (1) Find the partial derivative and set it equal to 0, then find the partial derivative with respect to α:

[0096] ;

[0097] Take the partial derivative with respect to β:

[0098] .

[0099] Since 2≠0, it can be simplified to:

[0100] Σ[Y_iP_i+αP_i 2 +βP_iR_i]=0→αΣP_i 2 +βΣP_iR_i=-ΣY_iP_i;

[0101] Σ[Y_iR_i+αP_iR_i+βR_i 2 ]=0→αΣP_iR_i+βΣR_i 2 =-ΣY_iR_i.

[0102] (2) Calculate the sum of each term, and substitute the values ​​of Y_i, P_i, and R_i from the valid cases to accurately calculate the sum of each term:

[0103] ΣP_i 2 =0.765 2 +0.90 2 +0.90 2 =0.585225+0.81+0.81=2.20522;

[0104] ΣP_iR_i=0.765×0.92+0.90×0.93+0.90×0.94=0.7038+0.837+0.846=2.3868;

[0105] ΣR_i 2 =0.92 2 +0.93 2 +0.94 2 =0.8464+0.8649+0.8836=2.5949;

[0106] ΣY_iP_i=(-0.470004)×0.765+(-0.916291)×0.90+(-0.282330)×0.90≈-0.359553+(-0.824662)+(-0.254097)=-1.438312;

[0107] ΣY_iR_i=(-0.470004)×0.92+(-0.916291)×0.93+(-0.282330)×0.94≈-0.432404+(-0.852151)+(-0.265390)=-1.549945.

[0108] (3) Establish and solve the normal equation system. Substitute the above summation value into the simplified normal equation system to obtain: 2.205225α+2.3868β=1.438312; 2.3868α+2.5949β=1.549945.

[0109] Solve this system of two linear equations using the substitution method:

[0110] Solving the first equation, we get: α = (1.438312 - 2.3868β) / 2.20522 ≈ 0.6522 - 1.0823β;

[0111] Substituting α into the second equation, we get: β≈0.41.

[0112] Substituting β≈0.41 into the expression for α, we get α≈0.62.

[0113] Therefore, the optimal estimated values ​​of the weighting coefficients are: α=0.62, β=0.41, both of which are positive values, which is consistent with the physical meaning that the stronger the emergency response capability, the more obvious the casualty reduction.

[0114] The calibrated first and second weighting coefficients are then substituted into the initial coupling correction formula, and the fitting accuracy is verified using historical earthquake casualty data. If the fitting accuracy meets the preset accuracy requirements, and both the first and second weighting coefficients are greater than zero, the coupling correction formula is determined based on the optimal estimate. If the fitting accuracy does not meet the preset accuracy requirements, historical earthquake casualty data is reacquired or the fitting method is adjusted until the fitting accuracy of the initial coupling correction formula meets the accuracy requirements.

[0115] For example, substituting the calibrated values ​​α=0.62 and β=0.41 into the initial coupling correction formula, the fitted value Cmodi for each valid case is calculated and compared with the actual number of casualties Cactuali to verify the fitting accuracy. The specific verification process is as follows:

[0116] The formula for calculating the fitted value is: C_model_i = C_0i × e^(-0.62 × P_i) × e^(-0.41 × R_i).

[0117] (1) Case group 2 (i=2):

[0118] C_model2 = 40 × e^(-0.62 × 0.765) × e^(-0.41 × 0.92) ≈ 40 × e^(-0.4743) × e^(-0.3772) ≈ 40 × 0.6225 × 0.685 ≈ 24.8 (people);

[0119] Deviation rate = |24.8-25| / 25×100%=0.8%.

[0120] (2) Case group 3 (i=3):

[0121] C_model 3 = 10 × e^(-0.62 × 0.90) × e^(-0.41 × 0.93) ≈ 10 × e^(-0.558) × e^(-0.3813) ≈ 10 × 0.572 × 0.683 ≈ 3.9 (people).

[0122] Deviation rate = |3.9-4| / 4×100% = 2.5%.

[0123] (3) Case group 4 (i=4):

[0124] C_model 4 = 260 × e^(-0.62 × 0.90) × e^(-0.41 × 0.94) ≈ 260 × e^(-0.558) × e^(-0.3854) ≈ 260 × 0.572 × 0.680 ≈ 197.2 (people);

[0125] Deviation rate = |197.2-196| / 196×100%≈0.6%.

[0126] The overall fitting accuracy calculation includes the following steps.

[0127] (1) Goodness of fit R 2 The goodness-of-fit index is used to evaluate the explanatory power of the modified earthquake disaster casualty assessment model. The calculation formula is as follows:

[0128] R 2 =1-[Σ(CrealiCmodulari)] 2 ] / [Σ(CrealiCrealaverage) 2 ];

[0129] Among them, the average value of C is (25+4+196) / 3 = 225 / 3 = 75 (people).

[0130] Calculate the values ​​for each item:

[0131] Σ(CrealiCmodi) 2 =(25-24.8) 2 +(4-3.9) 2 +(196-197.2) 2=0.04+0.01+1.44=1.49;

[0132] Σ(CrealiCrealaverage) 2 =(25-75) 2 +(4-75) 2 +(196-75) 2 =2500+5041+14641=22182;

[0133] Therefore, R 2 =1-1.49 / 22182≈1-0.000067≈0.999933, that is, R 2 The value is approximately 0.9999, close to 1, indicating that the revised earthquake disaster casualty assessment model has extremely strong explanatory power.

[0134] (2) Mean Absolute Deviation Rate (MAPE): The mean absolute deviation rate is used to evaluate the prediction accuracy of the corrected earthquake disaster casualty assessment model. The calculation formula is as follows:

[0135] MAPE = (1 / 3) × Σ[|(CrealiCmodi) / Creali| × 100%] = (1 / 3) × (0.8% + 2.5% + 0.6%) ≈ 1.2%.

[0136] Verification conclusion: Goodness of fit R 2 The mean absolute deviation (MAPE) is approximately 0.9999, and the fitting accuracy is extremely high. Since both α and β are positive values ​​with reasonable physical meaning, α=0.62 and β=0.41 are determined as the final weighting coefficients of the double exponential function-corrected earthquake disaster casualty assessment model.

[0137] Therefore, the double exponential function modified earthquake disaster casualty assessment model selected in this application embodiment conforms to the objective laws of earthquake disaster emergency response, fits the actual data of historical earthquake cases, and has both physical self-consistency and mathematical rigor. It can not only accurately reflect the nonlinear attenuation effect of emergency preparedness and emergency response capabilities on casualties, and solve the technical defects of linear modification model fitting failure and logical contradiction, but also achieve high-precision casualty assessment through weight coefficients (α=0.62, β=0.41) calibrated by historical cases. It provides reliable data support for casualty assessment, emergency rescue dispatch and post-disaster recovery planning in resilient cities, and is fully adapted to the invention purpose and practical application needs of this application embodiment.

[0138] In step 103, the constructed emergency preparedness capability index, emergency response status index, and calibrated coupling correction formula can be substituted into the initial evaluation value, and the corrected target evaluation value can be obtained through multiplication coupling operation. During the operation, the data format is kept consistent and the calculation accuracy is maintained.

[0139] In one example, the initial assessment value is first read from the output interface of the earthquake disaster casualty assessment model prior to step 103. The system automatically detects the data type (e.g., floating-point number). If it is an integer, it is converted to double-precision floating-point (64-bit) to ensure the accuracy of subsequent decimal operations and avoid truncation errors caused by integer division. Then, the index data (P and R) are loaded, and the emergency preparedness index P and emergency response index R generated in step 102 are called. A dimension consistency check is performed to verify whether P and R have been normalized (i.e., whether the numerical range is within the preset intervals such as [0,1] or [0,10]). If a dimension inconsistency is found (e.g., one is a percentage and the other is an absolute quantity), the system immediately calls the standardization function to map them to the unified dimension interval required by the formula. The calibration parameters (α and β) are loaded, and the optimal first weight coefficient α and second weight coefficient β determined in step 102 are read to ensure that the latest parameter values ​​verified by historical data are used.

[0140] Then, the attenuation factor is calculated item by item. Using a high-precision mathematical library, exponential operations are performed to calculate the reduction factor of emergency preparedness capability, i.e., the first attenuation factor. Similarly, using the same high-precision mathematical library, exponential operations are performed to calculate the reduction factor of emergency response status, i.e., the second attenuation factor. A logical check is added during the calculation process: if the calculated first or second attenuation factor approaches 0 due to floating-point underflow, it is forcibly set to a very small positive number to avoid subsequent multiplication results being absolutely zero (which physically implies zero casualties, but is usually not realistic).

[0141] Next, the multiplication coupling operation is performed. First, the two independent attenuation factors are multiplied and coupled to obtain the comprehensive correction coefficient. The initial evaluation value is multiplied by the comprehensive correction coefficient to obtain the final corrected target evaluation value. The entire multiplication chain adopts the double-precision floating-point arithmetic standard. No intermediate truncation is performed before the final output; rounding is only performed in the last step according to business requirements (such as personnel casualties usually being integers or retaining one decimal place).

[0142] Finally, the system outputs the results and performs anomaly checks. It automatically compares the calculated corrected estimate of casualties C with the initial estimate C0. If an anomaly occurs where C > C0, the system immediately issues an alarm and terminates the calculation, backtracking to check the validity of the parameters. Non-negativity checks ensure C ≥ 0. The final calculated C value is encapsulated into a standard data object, containing the numerical value, confidence level (based on fitting accuracy), calculation timestamp, and parameter version information, and then passed to step 104 for deviation verification.

[0143] Through the refined calculation process described above, a lossless conversion from theoretical formulas to actual numerical values ​​is ensured. Standardized format eliminates calculation errors caused by heterogeneous multi-source data. High-precision floating-point arithmetic reduces numerical drift during multiple exponentiation and multiplication operations. Boundary value protection and double verification mechanisms guarantee that the output always conforms to physical laws (i.e., the correction value must be less than or equal to the initial value and non-negative). This makes the final target evaluation value not only mathematically rigorous but also highly reliable and stable in engineering applications.

[0144] Steps 104 and 105 constitute a "self-calibrating closed loop" for revising the earthquake disaster casualty assessment model. By introducing historical or analogous data for verification and dynamically adjusting the indicator weights based on deviation feedback, the assessment results maintain high accuracy even in the absence of local historical data or in the event of sudden environmental changes.

[0145] In step 104, it is determined whether the correction area contains historical earthquake casualty data that matches the current assessment environment. If the correction area contains historical earthquake casualty data that matches the current assessment environment, then the historical earthquake casualty data of the correction area is selected as reference data. For example, the system first queries the disaster history database of the correction area to search for historical earthquake casualty data whose magnitude (±0.5), focal depth (±10km), occurrence time (day / night), and seasonal characteristics highly match the current assessment environment. If at least one historical record with a matching score higher than a preset threshold (e.g., 0.85) is found, then the actual disaster casualty data of that record is directly extracted as reference data.

[0146] If the correction area does not contain historical earthquake casualty data matching the current assessment environment, then historical earthquake casualty data from neighboring similar areas with similar geographical environment, population size, and economic level are selected. "Similar" means that the quantitative differences in geographical environment, population size, and economic level between the correction area and the current correction area are within a set range. For example, if the correction area has no matching historical data (such as newly developed urban areas or long-term earthquake-free zones), the system initiates a spatial analogy mechanism. The feature vector Vtarget={geographical features, population density, GDP per capita, building age distribution, urbanization rate} of the correction area is extracted. In the surrounding administrative region database, the cosine similarity or Euclidean distance between the feature vectors Vneighbor and Vtarget of each adjacent area is calculated. The top three adjacent similar areas with the highest similarity are selected, and their historical earthquake casualty data are weighted and averaged (weighted by the reciprocal of the similarity) to generate synthetic reference data.

[0147] Finally, based on the verification standards for this earthquake disaster level adjustment, the relative deviation formula was used to calculate the degree of deviation between the target assessment value and the reference data. The relative deviation formula is the absolute value of the difference between the target assessment value and the reference data divided by the reference data. The relative deviation formula is a dimensionless index used to measure the degree to which the predicted value deviates from the actual value. It eliminates the influence of differences in magnitude on error assessment, making the assessment results of cities of different sizes comparable.

[0148] This solves the "data silo" problem commonly encountered in earthquake assessment. For areas without historical earthquake records, a reliable reference benchmark is generated through a scientific analogy algorithm, preventing the revised earthquake disaster casualty assessment model from running blindly due to a lack of validation data. The method of calculating relative deviation unifies the error measurement standards for disasters of different scales, providing a precise quantitative basis for subsequent automated adjustments.

[0149] In step 105, if the deviation between the target assessment value and the reference data is greater than or equal to a preset deviation value, a weight adjustment strategy is determined by combining the deviation direction, deviation magnitude, and characteristics of the correction area. The weight adjustment strategy is a heuristic rule based on the idea of ​​error backpropagation. It corrects the cognitive bias of the earthquake disaster casualty assessment model regarding the emergency response capability of the correction area by dynamically changing the proportion of each evaluation indicator in the comprehensive index.

[0150] For example, the calculated deviation δ is compared with a preset deviation value (e.g., 20% for intensity VII and below, and 15% for intensity VIII and above). If δ < preset deviation value: the accuracy of the corrected earthquake disaster casualty assessment model is deemed acceptable, the current target assessment value is output, and the process ends. If δ ≥ preset deviation value: a weight adjustment strategy is triggered. The system generates adjustment instructions based on the deviation direction and the characteristics of the correction area.

[0151] Scenario A (Overestimation) indicates that the revised earthquake casualty assessment model underestimates the emergency response capabilities of the revised area. The strategy is to increase the weight of key indicators in the emergency preparedness index P or the emergency response situation index R (e.g., increasing the weight coefficient of drill frequency or response speed) to make the attenuation factor larger, thereby reducing predicted casualties. Scenario B (Underestimation) indicates that the revised earthquake casualty assessment model overestimates the region's emergency response capabilities or that there are unconsidered vulnerabilities. The strategy is to reduce the weight of relevant capability indicators or increase the influence of basic vulnerability factors to make the attenuation factor smaller, thereby improving predicted casualties. If the revised area is an old urban area, the contribution of the early warning terminal weight is automatically reduced, and the influence of implicit factors related to building seismic performance is increased (indirectly achieved by adjusting α and β).

[0152] Then, based on the adjusted indicator weights, the emergency preparedness capability index and emergency response status index are recalculated according to the preset quantitative weighting rules, and the coupling correction formula is updated simultaneously. The quantitative weighting rules are mathematical rules that define how various sub-indicators (such as drill frequency and material reserves) are aggregated into a total index (P or R), usually expressed as a linear weighted sum or non-linear combination. The reconstructed emergency preparedness capability index, emergency response status index, and updated coupling correction formula are substituted into the initial evaluation values ​​to recalculate the target evaluation values. Deviation checks are repeatedly performed on the newly generated target evaluation values ​​until the deviation between the target evaluation value and the reference data is less than the preset deviation value.

[0153] This step addresses different urban characteristics (such as aging communities vs. high-tech industrial parks) or sudden special circumstances. The revised earthquake disaster casualty assessment model no longer rigidly applies fixed weights, but automatically "calibrates" its sensitivity to various emergency factors based on deviation feedback. Through multiple rounds of iterative approximation, it ensures that the final output casualty assessment value is always controlled within an acceptable error range. The revised earthquake disaster casualty assessment model has better robustness in complex real-world scenarios. The weight adjustment process itself reveals the shortcomings of the regional emergency system (for example, if the weight of rescue forces needs to be significantly reduced to fit the actual casualties, it indicates that the actual rescue efficiency in the region is far lower than the reported data), providing in-depth insights for the government to formulate targeted improvement measures.

[0154] The following example illustrates the correction of casualty assessment in a certain region due to earthquake disaster.

[0155] 1. Implementation scenario: Taking the correction of casualties in a 7.0 magnitude earthquake disaster as an example (corresponding to the 4th group of historical cases).

[0156] 2. The implementation steps are as follows.

[0157] Step 1: Data collection. Collect various types of data for this area as follows.

[0158] (1) Assessment of personnel casualty data: The system assesses personnel casualties as 260;

[0159] (2) Data related to emergency preparedness: 294 earthquake emergency drills were conducted annually, covering 61,200 people; a special earthquake emergency plan is available and updated regularly every year; there are 61 earthquake early warning terminals, covering 8 townships; 31 popular science publicity activities were carried out, covering 46,500 people.

[0160] (3) Emergency response capability related data: emergency response activation time: 5 minutes; number of rescue personnel: 31.54 people / 10,000 people; current quantity of rescue supplies: 49 sets / 10,000 people;

[0161] Step 2: Calculation of emergency response capability coefficient.

[0162] (1) Calculate the emergency preparedness capability coefficient P:

[0163] P1=0.8;

[0164] P2 = Emergency Response Plan Score = 1.0 (Having a specific emergency response plan and updating it annually);

[0165] P3 = Early warning terminal coverage rate = 1;

[0166] P4 = Popular science coverage rate = 0.8

[0167] The values ​​α1=0.3, α2=0.2, α3=0.3, and α4=0.2 are determined by the dynamic hierarchical assignment method. Therefore, P=0.3×0.8+0.2×1.0+0.3×1+α4×0.8=0.9.

[0168] (2) Calculate the emergency response capability coefficient R:

[0169] R1 = Quantitative value of emergency response start time = 0.9 (start time 5 minutes ≤ 10 minutes);

[0170] R2 = Number of rescuers per 10,000 people = 31.54 people / 10,000 people = 1;

[0171] R3 = Amount of relief supplies per 10,000 people = 49 sets / 10,000 people = 0.9;

[0172] The values ​​β1=0.25, β2=0.4, and β3=0.35 were determined using the dynamic hierarchical assignment method.

[0173] Therefore, R = 0.25 × 0.9 + 0.4 × 1 + 0.35 × 0.9 = 0.94.

[0174] Step 3: Weight coefficient calibration.

[0175] Using the historical case data from groups 2, 3, and 4, the least squares method was used to fit the data, resulting in α=0.62 and β=0.41.

[0176] Step 4: Model correction and output.

[0177] Substituting into the coupling correction formula, we obtain:

[0178] C = 260 × e^(-0.62 × 0.9) × e^(-0.41 × 0.94) ≈ 40 × 0.6225 × 0.685 ≈ 260 × 0.572 × 0.680 ≈ 197.2 (people).

[0179] Step 5: Result verification and optimization.

[0180] The actual number of casualties in the area, Cactual, is 196. The deviation rate is |197.2 - 196| / 196 × 100% = 0.6% ≤ 10%, which meets the accuracy requirements. The final assessment value is 197.2 (in practical applications, it can be rounded to 197).

[0181] Compared to the initial assessment value of 260 people in the traditional earthquake disaster casualty assessment model (with a deviation rate of 32.7% from the actual value), the assessment accuracy has been significantly improved.

[0182] This application's embodiments overcome the shortcomings of traditional technologies and improve assessment accuracy. Non-engineering disaster mitigation factors (management, organization, behavior) are systematically embedded in quantifiable indices into the traditional earthquake disaster casualty assessment model, which primarily focuses on engineering factors (buildings, intensity). This achieves a paradigm shift from static physical assessment to dynamic comprehensive assessment. This application's embodiments introduce an emergency preparedness capability index and an emergency response status index as dual correction factors, incorporating real-world environmental factors such as emergency preparedness and emergency response into the earthquake disaster casualty assessment system. This solves the problem of traditional earthquake disaster casualty assessment models only considering building damage and population density, resulting in large assessment biases, and makes the assessment results more closely reflect actual earthquake disaster scenarios.

[0183] Secondly, the embodiments of this application provide clear quantitative definitions for the constituent indicators of the two correction factors. Each indicator has a clear value standard and calculation method, and the weight can be dynamically adjusted according to the population, economy, geological characteristics and earthquake disaster level of the correction area, reducing the influence of subjective factors, adapting to various earthquake scenarios, and facilitating practical application.

[0184] Furthermore, the correction method of the earthquake disaster casualty assessment model in this application embodiment can more accurately predict the number of casualties. At the same time, through the calculation of two correction factors, it clearly reflects the shortcomings of emergency preparedness and emergency response in the correction area. It not only provides accurate data support for the allocation of rescue resources and emergency command, but also provides targeted suggestions for subsequent emergency capability improvement and emergency plan optimization.

[0185] Finally, the embodiments of this application are based on the traditional earthquake disaster casualty assessment model. There is no need to completely reconstruct the existing assessment system. It can be directly connected to the existing earthquake disaster assessment system. The emergency preparedness and emergency response related data collection and calculation modules are embedded in the existing system to achieve rapid upgrades and reduce promotion costs.

[0186] Figure 2 This is a schematic diagram of the structure of a correction device for an earthquake disaster casualty assessment model provided in an embodiment of this application. Figure 2As shown, the correction device 200 for the earthquake disaster casualty model may include an acquisition module 201, a construction module 202, a calculation module 203, a verification module 204, and an adjustment module 205.

[0187] The acquisition module 201 is used to acquire emergency preparedness capability index data and emergency response status index data of the correction area.

[0188] The construction module 202 is used to construct an emergency preparedness index and an emergency response index based on emergency preparedness capability index data and emergency response status index data, and to construct a coupled correction formula with the emergency preparedness capability index and the emergency response status index as nonlinear attenuation correction factors.

[0189] The calculation module 203 is used to calculate the corrected target assessment value based on the initial assessment value output by the earthquake disaster casualty assessment model based on building vulnerability, combined with the emergency preparedness capability index, the emergency response status index and the coupled correction formula.

[0190] The verification module 204 is used to verify the deviation of the corrected target assessment value based on historical earthquake casualty data of the corrected area or adjacent similar areas of the corrected area.

[0191] The adjustment module 205 is used to adjust the index weights of the emergency preparedness capability index and the emergency response status index if the deviation between the target assessment value and the reference data is greater than or equal to the preset deviation value, reconstruct the coupled correction formula, and recalculate the target assessment value until the deviation between the target assessment value and the reference data is less than the preset deviation value.

[0192] The acquisition module 201, construction module 202, calculation module 203, verification module 204 and adjustment module 205 can be used to execute steps 101-105 in the embodiment of the above-mentioned correction method for the earthquake disaster casualty assessment model. For the specific implementation of these modules and more details, please refer to the corresponding method section, which will not be elaborated here.

[0193] This application also provides a computer-readable storage medium storing a program that can be loaded by a processor and executed by any of the earthquake disaster casualty assessment model correction methods in this application.

[0194] Those skilled in the art will understand that all or part of the functions of the various methods in the above embodiments can be implemented by hardware or by computer programs. When all or part of the functions in the above embodiments are implemented by computer programs, the program can be stored in a computer-readable storage medium, which may include: read-only memory, random access memory, disk, optical disk, hard disk, etc., and the program is executed by a computer to achieve the above functions. For example, the program can be stored in the memory of a device, and when the program in the memory is executed by the processor, all or part of the above functions can be achieved. In addition, when all or part of the functions in the above embodiments are implemented by computer programs, the program can also be stored in a server, another computer, disk, optical disk, flash drive, or external hard drive, etc., and can be downloaded or copied to the memory of a local device, or the system of the local device can be updated. When the program in the memory is executed by the processor, all or part of the functions in the above embodiments can be achieved.

[0195] The above examples illustrate this application only to aid understanding and are not intended to limit its scope. Those skilled in the art to which this application pertains can make various simple deductions, modifications, or substitutions based on the ideas presented.

Claims

1. A method for correcting an earthquake disaster casualty assessment model, characterized in that, include: Based on the personnel drill index parameters, emergency plan index parameters, early warning terminal index parameters, and earthquake science popularization index parameters of the correction area, emergency preparedness capability index data is constructed. Based on the emergency response system index parameters, rescue force index parameters, and rescue material index parameters of the modified area, emergency response status index data is constructed. The frequency of annual earthquake emergency drills and the percentage of participants in the modified area are used as the personnel drill indicator parameters. The completeness, update frequency, and operability score of the earthquake emergency plan for the modified area are used as the indicator parameters of the emergency plan. The number of earthquake early warning terminals deployed, coverage density, and normal operation rate in the modified area are used as the indicator parameters of the early warning terminals. The frequency, coverage, and scope of earthquake science popularization activities in the modified area will be used as the earthquake science popularization indicator parameters. The emergency preparedness index is obtained by weighted summing of the personnel drill index parameters, emergency plan index parameters, early warning terminal index parameters, and earthquake science popularization index parameters. The response speed and data transmission delay of the emergency response system in the correction area are used as the indicator parameters of the emergency response system. The number of professional rescue teams in the modified area, the rate of certified rescue personnel, the availability of rescue equipment, and the cross-regional rescue coordination capability are used as the rescue force indicator parameters. The quantity, layout, allocation efficiency, and replenishment capacity of the relief supplies in the modified area are used as the indicator parameters of the relief supplies. The emergency response status index is obtained by weighted summing of the emergency response system index parameters, rescue force index parameters, and rescue material index parameters. The constraints for constructing the coupled correction formula are: the corrected target assessment value is not greater than the initial assessment value; the rate of decrease in casualties is faster when the emergency response capability is improved from a low level to a medium level; the rate of decrease in casualties tends to slow down when the emergency response capability is close to saturation; and the formula is mathematically calibrable. A first attenuation factor in exponential form is defined for the emergency preparedness capability index, and a second attenuation factor in exponential form is defined for the emergency response status index. The first attenuation factor and the second attenuation factor are independent nonlinear attenuation factors. Based on the initial assessment value of the earthquake disaster casualty assessment model, the first attenuation factor and the second attenuation factor are multiplied and coupled to obtain the initial coupling correction formula; The initial coupling correction formula satisfies: ； Where C represents the revised assessment of casualties, C0 represents the initial assessment value, P represents the emergency preparedness index, and R represents the emergency response status index. As the first attenuation factor, Let α be the first weighting coefficient of the first attenuation factor, and β be the second weighting coefficient of the second attenuation factor, where α > 0 and β > 0. Multiple sets of historical earthquake casualty data were collected. Each set of historical earthquake casualty data included the initial assessment value of the earthquake disaster casualty assessment model, the emergency preparedness capability index, the emergency response status index, and the corrected assessment number of casualties. By linearizing the initial coupling correction formula using the natural logarithm, a linear fitting model is obtained; A least squares objective function is constructed with the goal of minimizing the sum of squared residuals between the fitted values ​​and the actual values ​​of the linear fitting model, and the historical earthquake casualty data is substituted into the least squares objective function. The partial derivatives of the first weight coefficient and the second weight coefficient are calculated for the least squares objective function, and the partial derivatives are set to zero. A system of objective equations is established, and the optimal estimates of the first weight coefficient and the second weight coefficient are obtained by solving the system. Substitute the calibrated first and second weighting coefficients into the initial coupling correction formula, and verify the fitting accuracy using historical earthquake casualty data. If the fitting accuracy meets the preset accuracy requirement, and both the first weight coefficient and the second weight coefficient are greater than zero, then the coupling correction formula is determined based on the optimal estimate. If the fitting accuracy does not meet the preset accuracy requirements, return to re-collect the historical earthquake casualty data or adjust the fitting method until the fitting accuracy of the initial coupling correction formula meets the accuracy requirements; Based on the initial assessment value output by the earthquake disaster casualty assessment model based on building vulnerability, the corrected target assessment value is calculated by combining the emergency preparedness capability index, the emergency response status index, and the coupling correction formula. Based on historical earthquake casualty data of the correction area or adjacent similar areas of the correction area, the corrected target evaluation value is checked for deviation. If the deviation between the target evaluation value and the reference data is greater than or equal to a preset deviation value, then the index weights of the emergency preparedness capability index and the emergency response status index are adjusted, the coupling correction formula is reconstructed, and the target evaluation value is calculated again until the deviation between the target evaluation value and the reference data is less than the preset deviation value.

2. The method for correcting the earthquake disaster casualty assessment model according to claim 1, characterized in that, The calculation, which combines the emergency preparedness capability index, the emergency response status index, and the coupled correction formula, yields the corrected target evaluation value, including: Substitute the constructed emergency preparedness capability index, the emergency response status index, and the calibrated coupling correction formula into the initial evaluation value, and obtain the corrected target evaluation value through multiplicative coupling operation, while maintaining data format consistency and calculation accuracy during the operation.

3. The method for correcting the earthquake disaster casualty assessment model according to claim 1, characterized in that, The deviation verification of the corrected target assessment value based on historical earthquake casualty data of the corrected area or adjacent similar areas of the corrected area includes: Determine whether the modified area contains historical earthquake casualty data that matches the current assessment environment; If the correction area contains historical earthquake casualty data that matches the current assessment environment, then the historical earthquake casualty data of the correction area is selected as reference data. If the correction area does not contain historical earthquake casualty data that matches the current assessment environment, then historical earthquake casualty data from adjacent similar areas with similar geographical environment, population size, and economic level to the correction area are selected. Based on the verification standards for this earthquake disaster level adjustment, the degree of deviation between the target assessment value and the reference data is calculated using the relative deviation formula, which is the absolute value of the difference between the target assessment value and the reference data divided by the reference data.

4. The method for correcting the earthquake disaster casualty assessment model according to claim 1, characterized in that, If the deviation between the target evaluation value and the reference data is greater than or equal to a preset deviation value, then the index weights of the emergency preparedness capability index and the emergency response status index are adjusted, the coupling correction formula is reconstructed, and the target evaluation value is calculated again until the deviation between the target evaluation value and the reference data is less than the preset deviation value, including: If the deviation between the target evaluation value and the reference data is greater than or equal to a preset deviation value, a weight adjustment strategy is determined by combining the deviation direction, deviation magnitude and correction area characteristics. Based on the adjusted indicator weights, the emergency preparedness capability index and the emergency response status index are recalculated according to the preset quantitative weighting rules, and the coupling correction formula is updated synchronously. Substitute the reconstructed emergency preparedness index, the emergency response status index, and the updated coupling correction formula into the initial evaluation value to recalculate the target evaluation value. Repeat the deviation check on the newly generated target evaluation value until the deviation between the target evaluation value and the reference data is less than the preset deviation value.

5. A correction device for an earthquake disaster casualty assessment model, characterized in that, include: The acquisition module is used to construct emergency preparedness capability indicator data based on personnel drill indicator parameters, emergency plan indicator parameters, early warning terminal indicator parameters, and earthquake science popularization indicator parameters of the modified area; and to construct emergency response status indicator data based on emergency response system indicator parameters, rescue force indicator parameters, and rescue material indicator parameters of the modified area. The module is used to: use the frequency and participation rate of annual earthquake emergency drills in the modified area as personnel drill indicator parameters; use the completeness, update frequency, and operability score of the earthquake emergency plan in the modified area as emergency plan indicator parameters; use the number of earthquake early warning terminals deployed, coverage density, and operational normalization rate in the modified area as early warning terminal indicator parameters; use the frequency, coverage, and scope of earthquake science popularization in the modified area as earthquake science popularization indicator parameters; perform a weighted summation of the personnel drill indicator parameters, emergency plan indicator parameters, early warning terminal indicator parameters, and earthquake science popularization indicator parameters to obtain an emergency preparedness capability index; use the response speed and data transmission latency of the emergency response system in the modified area as the emergency response system indicator parameters; and use the number of professional rescue teams and the certification of rescue personnel in the modified area as indicators. The emergency response system index, including its rate of response, equipment availability, and cross-regional emergency response coordination capabilities, is used as the emergency response force index. The quantity, layout, allocation efficiency, and replenishment capacity of emergency supplies in the corrected region are used as the emergency supplies index. The emergency response status index is obtained by weighted summation of the emergency response system index, emergency response force index, and emergency supplies index. The following constraints are used to construct the coupled correction formula: the corrected target assessment value is not greater than the initial assessment value; the rate of casualty reduction is faster when emergency response capabilities improve from low to medium levels; the rate of casualty reduction slows down when emergency response capabilities approach saturation; and mathematical calibrability is required. A first attenuation factor in exponential form is defined for the emergency preparedness capability index, and a second attenuation factor in exponential form is defined for the emergency response status index. The first and second attenuation factors are independent nonlinear attenuation factors. Based on the initial assessment values ​​of the earthquake disaster casualty assessment model, the first attenuation factor and the second attenuation factor are multiplied and coupled to obtain the initial coupling correction formula; the initial coupling correction formula satisfies: ； Where C represents the revised assessment of casualties, C0 represents the initial assessment value, P represents the emergency preparedness index, and R represents the emergency response status index. As the first attenuation factor, Let α be the first weighting coefficient of the first attenuation factor, and β be the second weighting coefficient of the second attenuation factor, where α > 0 and β > 0. Multiple sets of historical earthquake casualty data were collected. Each set of historical earthquake casualty data included the initial assessment value, emergency preparedness index, emergency response status index, and corrected assessment number of casualties from the earthquake disaster casualty assessment model. The initial coupling correction formula was linearized by taking the natural logarithm to obtain a linear fitting model. A least squares objective function was constructed with the goal of minimizing the sum of squared residuals between the fitted values ​​and actual values ​​of the linear fitting model. The historical earthquake casualty data were substituted into the least squares objective function. The partial derivatives of the first weight coefficient and the second weight coefficient were calculated for the least squares objective function, and the partial derivatives were set to... The derivative is set to zero, and a system of objective equations is established. The optimal estimates of the first and second weighting coefficients are obtained by solving the system. The calibrated first and second weighting coefficients are substituted into the initial coupling correction formula, and the fitting accuracy is verified using historical earthquake casualty data. If the fitting accuracy meets the preset accuracy requirements, and both the first and second weighting coefficients are greater than zero, the coupling correction formula is determined based on the optimal estimates. If the fitting accuracy does not meet the preset accuracy requirements, the historical earthquake casualty data is re-acquired or the fitting method is adjusted until the fitting accuracy of the initial coupling correction formula meets the accuracy requirements. The calculation module is used to calculate the corrected target assessment value based on the initial assessment value output by the earthquake disaster casualty assessment model based on building vulnerability, combined with the emergency preparedness capability index, the emergency response status index and the coupling correction formula. The verification module is used to verify the deviation of the corrected target evaluation value based on historical earthquake casualty data of the corrected area or adjacent similar areas of the corrected area. The adjustment module is used to adjust the index weights of the emergency preparedness capability index and the emergency response status index if the deviation between the target evaluation value and the reference data is greater than or equal to a preset deviation value, reconstruct the coupling correction formula, and recalculate the target evaluation value until the deviation between the target evaluation value and the reference data is less than the preset deviation value.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that can be loaded by a processor and executed as a correction method for the earthquake disaster casualty assessment model as described in any one of claims 1 to 4.

Images