Multi-objective decision-making method based on security, economy and availability

Abstract

The invention provides a multi-objective decision-making method based on security, economy and availability. The method comprises the following steps: generating a plurality of corresponding implementation schemes according to an obtained project file; screening the plurality of implementation schemes in each project file based on a preset condition to obtain corresponding sequence data; each piece of sequence data is processed, and corresponding key parameters are generated; averaging each type of key parameters of all the sequence data to generate corresponding target parameters; calculating the similarity between the key parameter and the target parameter of each piece of sequence data, and generating the total similarity of each piece of sequence data; and screening the total similarity of all the sequence data to obtain the sequence data corresponding to the optimal total similarity, and executing the corresponding implementation scheme based on the sorting of the implementation schemes in the sequence data. According to the multi-objective decision-making method based on safety, economy and availability provided by the invention, complex projects can be managed.

Description

Technical Field

[0001] The present invention relates to the field of nuclear power, and particularly to a multi-objective decision-making method based on safety, economy, and availability. Background Art

[0002] In modern industrial and infrastructure construction, the management of complex projects faces many challenges. These projects usually include various activities, such as equipment maintenance, system testing, and facility renovation. Each activity involves multiple interdependent factors, such as cost, resource allocation, time arrangement, and risk control. Traditional project management methods often can only optimize for a single objective and are difficult to achieve effective balance and comprehensive decision-making among multiple objectives.

[0003] Especially in high-complexity fields such as nuclear power plants, large bridges, and tunnels, the limitations of these methods are more obvious. Traditional optimization techniques usually do not consider the impact of dynamic data, such as the change in reliability after equipment maintenance or upgrade, the update of the probabilistic safety assessment (PSA) model, and the power generation risk changing over time. Ignoring these dynamic factors may lead to inaccurate decision-making and increase the risk of project implementation. Therefore, there is room for improvement. Summary of the Invention

[0004] The purpose of the present invention is to provide a multi-objective decision-making method based on safety, economy, and availability, which can manage complex projects.

[0005] To solve the above technical problems, the present invention is realized through the following technical solutions:

[0006] The present invention provides a multi-objective decision-making method based on safety, economy, and availability, including:

[0007] Generating multiple corresponding implementation plans according to the obtained project documents;

[0008] Performing screening processing on the multiple implementation plans in each of the project documents based on preset conditions to obtain corresponding sequence data; wherein, the sequence data includes multiple sorted implementation plans;

[0009] Processing each of the sequence data to generate corresponding key parameters;

[0010] Performing average processing on each type of key parameter of all the sequence data respectively to generate corresponding target parameters;

[0011] Calculating the similarity between the key parameters and the target parameters of each of the sequence data to generate the total similarity of each of the sequence data;

[0012] Screen the total similarity of all the sequence data, obtain the sequence data corresponding to the optimal total similarity, and execute the corresponding implementation plan based on the sorting of the implementation plans in the sequence data.

[0013] In an embodiment of the present invention, the step of generating a corresponding plurality of implementation plans according to the parameters of the obtained project files includes:

[0014] Obtain a plurality of project files;

[0015] Perform data extraction processing on the plurality of project files to generate corresponding initial parameters;

[0016] Perform format conversion processing on the initial parameters to generate corresponding intermediate parameters;

[0017] Generate a corresponding plurality of implementation plans based on the intermediate parameters corresponding to each project file.

[0018] In an embodiment of the present invention, the step of screening and processing a plurality of implementation plans in each project file based on preset conditions to obtain corresponding sequence data includes:

[0019] Screen a plurality of implementation plans in each project file respectively based on different preset screening conditions to screen out the implementation plans of the corresponding project files under different preset screening conditions;

[0020] Under each preset screening condition, sort the screened implementation plans based on preset sorting conditions to obtain different types of sequence data under each preset screening condition;

[0021] Judge whether the quantity of the sequence data reaches the preset capacity: if not, adjust the preset screening conditions and / or the preset sorting conditions, and repeat the processing until the quantity of the sequence data reaches the preset capacity.

[0022] In an embodiment of the present invention, the sequence data includes cost data, revenue data, equipment information, failure probability, and total personnel irradiation dose of each implementation plan; the step of processing each sequence data to calculate the corresponding key parameters includes:

[0023] Process the equipment information and failure probability in each sequence data based on the probabilistic safety average model to generate the corresponding core damage frequency and early large-scale radioactive release frequency;

[0024] Process the cost data, revenue data, and power generation risk assessment data in each sequence data based on the power generation risk assessment model and the financial model to generate the corresponding power generation risk assessment data and net present value;

[0025] Average the total irradiation dose of personnel for all embodiments in the sequence data to generate the corresponding average irradiation dose;

[0026] Summarize the corresponding core damage frequency, the early large radioactive release frequency, the net present value, the power generation risk assessment data, and the average irradiation dose to generate the key parameters for each sequence data.

[0027] In an embodiment of the present invention, the step of respectively averaging each type of key parameter of all the sequence data to generate the corresponding target parameter includes:

[0028] Average the core damage frequency, the early large radioactive release frequency, the net present value, the power generation risk assessment data, and the total irradiation dose of personnel for all embodiments in all the sequence data to generate the corresponding target damage frequency, target radioactive release frequency, target net present value, target risk assessment data, and target irradiation dose;

[0029] Summarize the corresponding target damage frequency, target radioactive release frequency, target net present value, target risk assessment data, and target irradiation dose to generate the corresponding target parameter.

[0030] In an embodiment of the present invention, the step of calculating the similarity between the key parameter and the target parameter of each sequence data to generate the total similarity of each sequence data includes:

[0031] Calculate the similarity of each type of key parameter in each sequence data according to the data types of each type of key parameter and the corresponding target parameter in each sequence data;

[0032] Calculate the sum of the similarities of all types of key parameters of the sequence data to generate the total similarity of the sequence data.

[0033] In an embodiment of the present invention, the step of calculating the similarity of each type of key parameter in each sequence data according to the data types of each type of key parameter and the corresponding target parameter in each sequence data includes:

[0034] Judge the types of each type of key parameter and the corresponding target parameter in each sequence data:

[0035] When the key parameter and the corresponding target parameter are numerical types, calculate the similarity of the key parameter according to the key parameter, the corresponding target parameter, the maximum and minimum values of the key parameter in all the sequence data;

[0036] When the key parameter is numerical and the corresponding target parameter is interval - type, calculate the similarity of the key parameter according to the key parameter, the minimum value and the maximum value of the interval range of the corresponding target parameter.

[0037] When the key parameter is interval - type and the corresponding target parameter is numerical, calculate the similarity of the key parameter according to the minimum value, the maximum value of the interval range of the key parameter and the corresponding target parameter.

[0038] When the key parameter and the corresponding target parameter are both interval - type, calculate the similarity of the key parameter according to the minimum value, the maximum value of the interval range of the key parameter and the minimum value, the maximum value of the interval range of the corresponding target parameter.

[0039] In an embodiment of the present invention, when the key parameter and the corresponding target parameter are numerical, the similarity Sim(x i ,a i ) of the key parameter is expressed as: where xi represents the i - th key parameter, ai represents the corresponding i - th target parameter, a min and a max represent the maximum value and the minimum value of the key parameter respectively.

[0040] In an embodiment of the present invention, when the key parameter and the corresponding target parameter are interval - type, the similarity Sim([x i ,y i ,[a i ,b i ) of the key parameter is expressed as: where [x i ,y i represents the interval range of the i - th key parameter, [a i ,b i represents the interval range of the i - th target parameter, u represents the middle value of the interval range of the i - th key parameter, and v represents the middle value of the interval range of the i - th target parameter.

[0041] In an embodiment of the present invention, when the key parameter is interval - type and the corresponding target parameter is numerical, the similarity Sim([x i ,y i ,a i ) of the key parameter is expressed as: where [x i ,y i represents the interval range of the i - th key parameter, and ai represents the corresponding i - th target parameter.

[0042] As described above, the present invention provides a multi-objective decision-making method based on security, economy, and usability. By integrating a variety of advanced analysis methods and intelligent algorithms, it comprehensively considers multi-dimensional factors in the project implementation process, significantly improving the efficiency, economic benefits, and security of complex project management. It is not only applicable to high-complexity and high-risk projects such as nuclear power plants and large-scale infrastructure, but also has broad applicability and flexibility, and can promote the sustainable development of enterprises and society.

[0043] Of course, it is not necessary for any product implementing the present invention to simultaneously achieve all the above-mentioned advantages. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for describing the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0045] Figure 1 It is a flowchart of the multi-objective decision-making method based on security, economy, and usability in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0047] Please refer to Figure 1 , the present invention discloses a multi-objective decision-making method based on security, economy, and usability. The decision-making method can optimize a complex project so that the optimized complex project can be successfully executed. The decision-making method may include the following steps:

[0048] Step S10: Generate multiple corresponding implementation plans according to the obtained project files;

[0049] Step S20: Screen and process the multiple implementation plans in each project file based on preset conditions to obtain corresponding sequence data; wherein, the sequence data includes multiple sorted implementation plans;

[0050] Step S30: Process each sequence data to generate corresponding key parameters;

[0051] Step S40: Perform averaging processing on each type of key parameter of all sequence data respectively to generate corresponding target parameters;

[0052] Step S50: Calculate the similarity between the key parameters and the target parameters of each sequence data, and generate the total similarity of each sequence data.

[0053] Step S60: Screen the total similarities of all sequence data, obtain the sequence data corresponding to the optimal total similarity, and execute the corresponding implementation plan based on the sorting of the implementation plans in this sequence data.

[0054] In some embodiments, when step S10 is executed, specifically, step S10 may include the following steps:

[0055] Step S11: Obtain a plurality of project files.

[0056] Step S12: Perform data extraction processing on the plurality of project files to generate corresponding initial parameters.

[0057] Step S13: Perform format conversion processing on the initial parameters to generate corresponding intermediate parameters.

[0058] Step S14: Generate a plurality of corresponding implementation plans based on the intermediate parameters corresponding to each project file.

[0059] In some embodiments, when step S11 is executed, specifically, various project files can be imported from multiple data sources. These project files may contain different formats, such as PDF, Word, Excel, etc. These project files contain detailed information about the implementation plans, including various initial parameters. Among them, the data sources can include databases, file servers, cloud storage, etc. The project files can also include implementation projects, such as maintenance, tests, renovations, etc. Each project file can contain detailed descriptions and relevant data of these activities.

[0060] In some embodiments, when step S12 is executed, specifically, natural language processing (NLP) technology can be used to parse the unstructured data in the project files, automatically standardize and organize the recognition of initial parameters. For example, when the project files are in PDF and Word formats, OCR (Optical Character Recognition) technology can be used to extract the text content from the PDF files, or NLP technology can be used to parse the text content in the Word documents. Another example is that when the project files are in Excel format, a data processing library (such as Pandas) can be used to read the data in the Excel tables.

[0061] In some embodiments, after extraction processing through natural language processing technology, the initial parameters of each project file can be obtained. The initial parameters may include the cost data of the project, the revenue data after implementation, the duration (repair time), equipment information (specific unavailable equipment), failure probability (how many times the equipment fails in a year, obtained through calculation), probabilistic safety assessment data (impact of the PSA model), power generation risk assessment data (impact of the GRA model), and the total personnel irradiation dose.

[0062] In some embodiments, the cost data refers to the total cost of project implementation, including material costs, labor costs, equipment rental fees, etc. During the extraction process, unit conversion can be performed on cost data in different units (for example, converting ten thousand yuan to yuan). The revenue data refers to the average annual revenue after project implementation. During the extraction process, unit conversion can be performed on the revenue data to ensure a consistent currency unit. The duration refers to the implementation duration of the project, including the repair time. During the extraction process, the time unit can be converted (for example, converting days to hours). The equipment information refers to specific unavailable equipment information, including the equipment name, model, and unavailable time period. The failure probability refers to the failure probability data of the equipment, such as the number of times the equipment fails within a year. For example, specific failure probability values can be obtained through calculation methods. The probabilistic safety assessment data (impact of the PSA model) refers to the key parameters in the PSA model, such as the fault tree analysis results, event tree analysis results, etc. The power generation risk assessment data (impact of the GRA model) refers to the key parameters in the GRA model, such as the power generation risk value, risk distribution, etc. The total personnel irradiation dose refers to the total personnel irradiation dose data during project implementation. During the extraction process, unit conversion can be performed on the data (for example, converting mSv to Sv).

[0063] In some embodiments, when step S13 is executed, specifically, during the extraction processing, operations such as unit conversion, missing value filling, and outlier identification can be performed on the extracted initial parameters to ensure that the generated intermediate parameters can be effectively used by subsequent analysis models. Among them, unit conversion refers to converting all cost data to the same currency unit (for example, yuan), converting all time data to the same time unit (for example, hours), and converting all irradiation dose data to the same unit (for example, Sv). Missing value filling refers to reasonably filling in the missing data, which can use the average value, median, or interpolation method based on other data. Outlier identification refers to identifying and processing outliers, which can be identified and processed through statistical methods (such as standard deviation, quantile) or machine learning methods (such as isolation forest). The extracted and processed initial parameters can be standardized and integrated to generate a unified data table for subsequent model calculation and analysis.

[0064] In some embodiments, when performing step S14, specifically, in order to generate multiple different implementation schemes, a multi-objective optimization model can be adopted. The multi-objective optimization model aims to optimize multiple objectives simultaneously, such as minimizing cost, maximizing revenue, minimizing risk, etc. For example, multiple objective functions can be defined, and the objective functions can include a cost minimization function, a revenue maximization function, a risk minimization function, etc. At the same time, the constraint conditions of the multi-objective optimization model can be defined to ensure that the generated schemes are feasible. The constraint conditions can include time constraint conditions, resource constraint conditions, safety constraint conditions, equipment reliability constraint conditions, etc. Among them, the time constraint means that the duration of the project must be within the allowed range. The resource constraint means that the resources required for the project (such as personnel and equipment) cannot exceed the available resources. The safety constraint means that the total irradiation dose of personnel cannot exceed the safety standard. The equipment reliability constraint means that the unavailable time of the equipment must be within the plan.

[0065] In some embodiments, a multi-objective optimization algorithm (such as a genetic algorithm, a particle swarm optimization algorithm, etc.) can be used to generate multiple different implementation schemes. For example, first, a set of initial schemes can be randomly generated, and each scheme includes parameters such as the project schedule, resource allocation, cost, revenue, etc. Subsequently, the objective function values and the satisfaction degrees of the constraint conditions can be calculated for each scheme. Subsequently, the schemes with higher fitness can be selected to enter the next generation. Subsequently, new schemes can be generated through crossover operations, combining the characteristics of two or more existing schemes. Subsequently, new characteristics can be introduced through mutation operations to increase the diversity of the population. Finally, when the predetermined number of iterations is reached or the objective function value no longer improves significantly, the optimization process is terminated, and multiple corresponding implementation schemes are generated.

[0066] Please refer to Table 1. In some embodiments, taking the example of using a genetic algorithm to generate multiple implementation plans based on the intermediate parameters of a project file for illustration. The intermediate parameters of the project file may include the cost data of the project (1,000,000 yuan), the revenue data after implementation (150,000 yuan / year), the duration (240 hours), the equipment information (equipment A is unavailable from 08:00 on January 1, 2024 to 16:00 on January 2, 2024), the failure probability (0.05 times / year), the probabilistic safety assessment data (0.001), the power generation risk assessment data (0.002), and the total personnel irradiation dose (50 μSv). By initializing the population, 100 random plans are generated, and each plan contains parameters such as time arrangement and resource allocation. By evaluating the plans, calculate the cost, revenue, and risk index of each plan, and at the same time check whether each plan meets the time, resource, safety, and equipment reliability constraints. Through the selection operation, select the top 50 plans with higher fitness. Through the crossover operation, randomly select two from these 50 plans for crossover to generate 50 new plans. Through the mutation operation, randomly mutate some parameters in the newly generated 50 plans. By evaluating the new plans, recalculate the cost, revenue, and risk index of the new plans, and check whether the new plans meet the constraint conditions. Through repeated iteration, repeat the selection, crossover, mutation, and evaluation processes until a predetermined number of iterations (e.g., 100 times) is reached or the objective function value no longer improves significantly.

[0067] Table 1: Five implementation plans generated by the genetic algorithm.

[0068]

[0069] In some embodiments, when step S20 is executed, specifically, step S20 may include the following steps:

[0070] Step S21: Screen the multiple implementation plans in each project file based on different preset screening conditions to screen out the implementation plans of the corresponding project file under different preset screening conditions;

[0071] Step S22: Under each preset screening condition, sort the screened implementation plans based on the preset sorting condition to obtain different types of sequence data under each preset screening condition;

[0072] Step S23: Determine whether the number of sequence data reaches the preset capacity: If not, adjust the preset screening condition and / or the preset sorting condition, and repeat the process until the number of sequence data reaches the preset capacity.

[0073] In some embodiments, when performing step S21, specifically, screening conditions can be preset. The screening conditions can include cost limit, revenue requirement, risk tolerance, time limit, equipment availability, personnel irradiation dose, etc. Among them, the cost limit means that the cost of the implementation plan cannot exceed a certain threshold. The revenue requirement means that the implementation plan must reach a certain revenue level. The risk tolerance means that the risk index of the implementation plan cannot exceed a certain acceptance level. The time limit means that the duration of the implementation plan must be within a specific range. The equipment availability means that the implementation plan must consider the unavailable time of the equipment. The personnel irradiation dose means that the personnel irradiation dose in the implementation plan must meet the safety standards.

[0074] In some embodiments, an intelligent sampling algorithm can be used to extract one implementation plan from multiple implementation plans of each project file, and diversity and representativeness are ensured. Specifically, according to the preset screening conditions, each group of implementation plans can be evaluated and screened. At the same time, an optimal implementation plan is selected from each group, while ensuring overall diversity and representativeness.

[0075] In some embodiments, taking the screening by the intelligent sampling algorithm as an example for illustration. The preset screening conditions can be set as follows: the cost does not exceed 1,000,000 yuan, the revenue is at least 150,000 yuan / year, the risk index does not exceed 0.0015, the duration does not exceed 240 hours, and the personnel irradiation dose does not exceed 55 μSv. For example, a certain project file has 5 implementation plans. Implementation plan a: the cost is 950,000 yuan, the revenue is 140,000 yuan / year, and the revenue does not meet the standard. Implementation plan b: the cost is 980,000 yuan, the revenue is 155,000 yuan / year, the risk is 0.0011, the time is 230 hours, and the dose is 50 μSv, which meets the conditions. Implementation plan c: the cost is 960,000 yuan, the revenue is 148,000 yuan / year, and the revenue does not meet the standard. Implementation plan d: the cost is 970,000 yuan, the revenue is 152,000 yuan / year, the risk is 0.0012, the time is 228 hours, and the dose is 47 μSv, which meets the conditions. Implementation plan e: the cost is 955,000 yuan, the revenue is 145,000 yuan / year, and the revenue does not meet the standard. The implementation plans that meet the preset screening conditions include b and d. Among them, when the number of implementation plans that meet the preset screening conditions is multiple, the implementation plan with the highest revenue can be given priority.

[0076] In some embodiments, when step S22 is executed, specifically, in the screened implementation solutions, sorting can be performed based on another set of preset sorting conditions. Through the sorting operation, different types of sequence data are generated (such as ascending order by cost, descending order by revenue, ascending order by risk, etc.). The preset sorting conditions can include cost from low to high (ascending order) or from high to low (descending order), revenue from high to low (descending order) or from low to high (ascending order), risk index from low to high (ascending order), duration from short to long (ascending order) or from long to short (descending order), personnel irradiation dose from low to high (ascending order), etc.

[0077] In some embodiments, for the screened implementation solutions, sorting is performed according to the preset sorting conditions. For example, sorting in ascending order by cost: the solution with the lowest cost is ranked at the front; sorting in descending order by revenue: the solution with the highest revenue is ranked at the front; sorting in ascending order by risk: the solution with the lowest risk is ranked at the front; sorting in ascending order by duration: the solution with the shortest duration is ranked at the front; sorting in ascending order by personnel irradiation dose: the solution with the lowest personnel irradiation dose is ranked at the front.

[0078] In some embodiments, through the sorting operation, different types of sequence data can be generated. For example, a sequence in ascending order by cost: list all eligible implementation solutions and arrange them in ascending order by cost; a sequence in descending order by revenue: list all eligible implementation solutions and arrange them in descending order by revenue; a sequence in ascending order by risk: list all eligible implementation solutions and arrange them in ascending order by risk; a sequence in ascending order by duration: list all eligible implementation solutions and arrange them in ascending order by duration; a sequence in ascending order by personnel irradiation dose: list all eligible implementation solutions and arrange them in ascending order by personnel irradiation dose.

[0079] In some embodiments, when step S23 is executed, specifically, after the sequence data is generated, it can be checked whether the number of the generated sequence data reaches a preset capacity (such as 20,000). If the number of column data is insufficient, the screening conditions and / or sorting conditions need to be adjusted, and then the sequence data is regenerated. The above steps can be repeated until the number of the generated sequence data reaches the preset capacity.

[0080] In some embodiments, when step S30 is executed, specifically, step S30 may include the following steps:

[0081] Step S31: Process the equipment information and failure probability in each sequence data based on the probabilistic safety average model to generate the corresponding core damage frequency and early large-scale radioactive release frequency;

[0082] Step S32: Process the cost data and revenue data in each sequence of data based on the power generation risk assessment model and the financial model to generate corresponding power generation risk assessment data and net present value;

[0083] Step S33: Averagely process the total irradiation dose of personnel in all implementation schemes in the sequence of data to generate the corresponding average irradiation dose;

[0084] Step S34: Summarize and process the corresponding core damage frequency, early large-scale radioactive release frequency, net present value, power generation risk assessment data, and average irradiation dose to generate the key parameters of each sequence of data.

[0085] In some embodiments, when performing step S31, specifically, the sequence of data may include the equipment information and failure probability of each implementation scheme. The Probabilistic Safety Assessment (PSA) model can be used to comprehensively analyze multiple parameters of each implementation scheme to generate the core damage frequency and early large-scale radioactive release frequency. PSA is a method for evaluating the safety risk of nuclear facilities. By quantifying the probabilities and consequences of various failure events, it evaluates the safety.

[0086] In some embodiments, first, the equipment information and failure probability of each implementation scheme can be obtained. Subsequently, the PSA model can be used for modeling. The modeling content may include Event Tree Analysis (ETA), Fault Tree Analysis (FTA), and risk assessment. Among them, ETA refers to defining various possible event sequences and their consequences. FTA refers to analyzing the failure paths and probabilities leading to various events. Risk assessment refers to combining the results of ETA and FTA to evaluate the risks of each event. Finally, each sequence of data can be processed to generate the core damage frequency (CDF) and early large-scale radioactive release frequency (LERF).

[0087] In some embodiments, when performing step S32, specifically, by applying the power generation risk assessment model and the financial model to process a series of data (including costs and revenues), the net present value (NPV, Net Present Value) and power generation risk assessment data of each sequence of data are finally generated. The net present value is an important indicator for evaluating the profitability of investment projects. It takes into account the time value of money and determines the economic feasibility of the project by discounting future cash flows to the current point in time and subtracting the initial investment cost.

[0088] In some embodiments, first, cost data and revenue data for each implementation plan can be obtained. Subsequently, the above data can be processed using a power generation risk assessment model and a financial model. Among them, the power generation risk assessment model can be used to evaluate power generation risk assessment data during the power generation process, which may include safety risks, environmental impacts, health risks, etc. The financial model can be used to calculate the net present value based on the cost data and revenue data, considering the time value of cash flows.

[0089] In some embodiments, the input data of the power generation risk assessment model can be cost data and revenue data, and the output data can be power generation risk assessment data. The power generation risk assessment model can be used to quantify the risk level, which may be converted into economic costs (such as insurance costs, compensation costs, etc.).

[0090] In some embodiments, the input data of the financial model can be cost data and revenue data. The financial model can be used to predict future cash flows based on revenue and cost data; the financial model can also be used to apply a risk adjustment factor to the cash flows to reflect the impact of risk on the cash flows; the financial model can also be used to select an appropriate discount rate to reflect the time value of funds and project risks; the financial model can also be used to discount future cash flows to the current point in time, subtract the initial investment, and obtain the net present value. The output data of the financial model can be the net present value of each sequence of data.

[0091] In some embodiments, when step S33 is executed, specifically, statistical averaging processing is performed on the data of multiple implementation plans to generate representative indicators. For example, the total personnel irradiation dose for each implementation plan can be aggregated and calculated to obtain the average value of these indicators for subsequent analysis or decision-making.

[0092] In some embodiments, when step S34 is executed, specifically, the core damage frequency, early large-scale radioactive release frequency, net present value, power generation risk assessment data, and average irradiation dose of each sequence of data can be comprehensively processed to generate one or more key parameters for subsequent analysis or decision-making. Among them, the core damage frequency refers to the cumulative probability of describing the occurrence of a certain event (such as equipment failure, risk event). The early large-scale radioactive release frequency refers to the frequency of evaluating the occurrence of a large-scale radioactive material release within a specific time. The net present value refers to an indicator for evaluating the economic benefits of a project, reflecting the present value of future cash flows minus the initial investment. The average irradiation dose refers to the average value of the total personnel irradiation dose in all implementation plans, reflecting the radiation risk.

[0093] In some embodiments, when step S40 is executed, specifically, step S40 may include the following steps:

[0094] Step S41: Average the core damage frequency, early large radioactive release frequency, net present value, power generation risk assessment data, and total personnel irradiation dose for all implementation schemes in all sequence data to generate corresponding target damage frequency, target radioactive release frequency, target net present value, target risk assessment data, and target irradiation dose.

[0095] Step S42: Aggregate the corresponding target damage frequency, target radioactive release frequency, target net present value, target failure probability, target risk assessment data, and target irradiation dose to generate corresponding target parameters.

[0096] In some embodiments, when performing Step S41, specifically, for each indicator, the corresponding data of the implementation schemes in all sequence data can be averaged to generate the target damage frequency, target radioactive release frequency, target net present value, target risk assessment data, and target irradiation dose.

[0097] In some embodiments, the average value of the core damage frequency of all implementation schemes in all sequence data can be taken to generate the target damage frequency. The average value of the early large radioactive release frequency of all implementation schemes in all sequence data can be taken to generate the target radioactive release frequency. The average value of the net present value of all implementation schemes in all sequence data can be taken to generate the target net present value. The average value of the power generation risk assessment data of all implementation schemes in all sequence data can be taken to generate the target risk assessment data. The average value of the total personnel irradiation dose of all implementation schemes in all sequence data can be taken to generate the target irradiation dose.

[0098] In some embodiments, the target damage frequency can be used to evaluate the overall core damage frequency level. The target radioactive release frequency can be used to evaluate the overall early large radioactive release frequency. The target net present value can be used to evaluate the overall economic benefit level. The target irradiation dose can be used to evaluate the overall personnel radiation risk.

[0099] In some embodiments, when performing Step S42, specifically, the calculated target damage frequency, target radioactive release frequency, target net present value, target risk assessment data, and target irradiation dose can be further aggregated to generate comprehensive target parameters.

[0100] In some embodiments, when performing Step S50, specifically, Step S50 may include the following steps:

[0101] Step S51: Calculate the similarity of each type of key parameter in each sequence data according to the data types of each type of key parameter and the corresponding target parameters in each sequence data.

[0102] Step S52: Calculate the sum of the similarities of all types of key parameters of the sequence data to generate the total similarity of the sequence data.

[0103] In some embodiments, when performing step S51, specifically, step S51 may include the following steps:

[0104] Step S511: Determine the types of each type of key parameter in each sequence data and the corresponding target parameter:

[0105] Step S512: When the key parameter and the corresponding target parameter are numerical types, calculate the similarity of the key parameter according to the key parameter, the corresponding target parameter, the maximum value and the minimum value of the key parameter in all sequence data.

[0106] Step S513: When the key parameter is a numerical type and the corresponding target parameter is an interval type, calculate the similarity of the key parameter according to the key parameter, the minimum value and the maximum value of the interval range of the corresponding target parameter.

[0107] Step S514: When the key parameter is an interval type and the corresponding target parameter is a numerical type, calculate the similarity of the key parameter according to the minimum value, the maximum value of the interval range of the key parameter and the corresponding target parameter.

[0108] Step S515: When the key parameter and the corresponding target parameter are numerical types, calculate the similarity of the key parameter according to the minimum value, the maximum value of the interval range of the key parameter and the minimum value, the maximum value of the interval range of the corresponding target parameter.

[0109] In some embodiments, when performing step S511, specifically, since the type of each type of key parameter in each sequence data may be one of the numerical type and the interval type, and the type of the corresponding target parameter may also be one of the numerical type and the interval type. Therefore, it is necessary to determine the types of each type of key parameter and the corresponding target parameter in each sequence data to perform different operations according to the determination results.

[0110] In some embodiments, when performing step S512, specifically, when the key parameter and the corresponding target parameter are numerical types, calculate the similarity Sim(x i , a i ) of the key parameter according to the key parameter, the corresponding target parameter, the maximum value and the minimum value of the key parameter in all sequence data; where, xi represents the i-th key parameter, ai represents the corresponding i-th target parameter, a min and a max respectively represent the maximum value and the minimum value of the key parameter.

[0111] In some embodiments, when step S513 is executed, specifically, when the key parameter is numerical and the corresponding target parameter is interval-type, according to the key parameter, the minimum value and the maximum value of the interval range of the corresponding target parameter, calculate the similarity Sim(x i ,[a i-min ,a i-max ) of the key parameter;

[0112]

[0113] wherein, [a i-min ,a i-max represents the interval range of the i-th target parameter, and x i represents the corresponding i-th key parameter.

[0114] In some embodiments, when step S514 is executed, specifically, when the key parameter is interval-type and the corresponding target parameter is numerical, according to the minimum value, the maximum value of the interval range of the key parameter and the corresponding target parameter, calculate the similarity Sim([x i ,y i ,a i ) of the key parameter; wherein, [x i ,y i represents the interval range of the i-th key parameter.

[0115] In some embodiments, when step S515 is executed, specifically, when the key parameter and the corresponding target parameter are interval-type, according to the minimum value, the maximum value of the interval range of the key parameter and the minimum value, the maximum value of the interval range of the corresponding target parameter, calculate the similarity Sim([x i ,y i ,[a i ,b i ) of the key parameter; wherein, [x i ,y i represents the interval range of the i-th key parameter, [a i ,b i represents the interval range of the i-th target parameter, u represents the middle value of the interval range of the i-th key parameter, and v represents the middle value of the interval range of the i-th target parameter.

[0116] In some embodiments, when step S52 is executed, specifically, by calculating the sum of the similarities of the key parameters of the sequence data, generate the total similarity SimX of the sequence data, wherein, X represents the X-th sequence data, xi represents the i-th key parameter, n represents the total number of key parameters, w iRepresents the weight value of the i-th key parameter,

[0117] In some embodiments, when step S60 is executed, specifically, the optimal sequence data can be selected through similarity screening, and the corresponding implementation plans can be executed based on the sorting of the implementation plans in the sequence data, thereby improving the scientificity and effectiveness of decision-making.

[0118] It can be seen that in the above solution, by integrating a variety of advanced analysis methods and intelligent algorithms, comprehensively considering multi-dimensional factors in the project implementation process, the efficiency, economic benefits and safety of complex project management have been significantly improved. It is not only applicable to high-complexity and high-risk projects such as nuclear power plants and large infrastructure, but also has wide applicability and flexibility, and can promote the sustainable development of enterprises and society. Through scientific and intelligent decision support, specifically, through intelligent sampling technology, key data can be efficiently screened, combined with multi-objective optimization algorithms, and the risks, costs and benefits of different plans can be comprehensively evaluated to ensure the scientificity and systematicness of decision-making. By combining methods such as PSA (Probabilistic Safety Assessment), GRA (Generation Risk Assessment), financial analysis and multi-objective optimization, various multi-dimensional factors such as risks, costs, benefits, etc. involved in the project implementation process are comprehensively evaluated. Through comprehensive risk assessment and economic benefit assessment, the optimal balance between risks and benefits is achieved. Through similarity screening and implementation plan sorting, the optimal implementation plan can be quickly determined, reducing decision-making time and improving the efficiency of project management. This solution is particularly applicable to project scenarios with high complexity and high risk such as nuclear power plants and large infrastructure construction, and can effectively address decision-making challenges under complex conditions.

[0119] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0120] The embodiments of the present invention disclosed above are only used to help explain the present invention. The embodiments do not describe all the details in detail, nor limit the invention to the specific embodiments described. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification to better explain the principles and practical applications of the present invention, so that those skilled in the art in the technical field can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A multi-objective decision-making method based on security, economy, and usability, characterized in that Including: Generating a plurality of corresponding implementation plans according to the obtained project files; Performing screening processing on the plurality of implementation plans in each of the project files based on preset conditions to obtain corresponding sequence data; wherein, the sequence data includes a plurality of sorted implementation plans; Processing each of the sequence data to generate corresponding key parameters; Respectively performing averaging processing on each type of key parameters of all the sequence data to generate corresponding target parameters; Calculating the similarity between the key parameters and the target parameters of each of the sequence data to generate the total similarity of each of the sequence data; Screening the total similarities of all the sequence data, obtaining the sequence data corresponding to the optimal total similarity, and executing the corresponding implementation plan based on the sorting of the implementation plans in the sequence data.

2. The multi-objective decision-making method based on security, economy, and usability according to claim 1, wherein The step of generating corresponding multiple implementation plans according to the parameters of the obtained project files includes: Obtaining a plurality of project files; Performing data extraction processing on the plurality of project files to generate corresponding initial parameters; Performing format conversion processing on the initial parameters to generate corresponding intermediate parameters; Generating a plurality of corresponding implementation plans based on the intermediate parameters corresponding to each of the project files.

3. The multi-objective decision-making method based on security, economy, and usability according to claim 1, characterized in that, The step of performing screening processing on the plurality of implementation plans in each of the project files based on preset conditions to obtain corresponding sequence data includes: Screening the plurality of implementation plans in each of the project files respectively based on different preset screening conditions to screen out the implementation plans of the corresponding project files under different preset screening conditions; Under each of the preset screening conditions, sorting the screened implementation plans based on preset sorting conditions to obtain different types of sequence data under each of the preset screening conditions; Judging whether the quantity of the sequence data reaches a preset capacity: if not, adjusting the preset screening conditions and / or the preset sorting conditions, and repeating the processing until the quantity of the sequence data reaches the preset capacity.

4. The multi-objective decision-making method based on security, economy, and usability according to claim 1, wherein, The sequence data includes cost data, revenue data, equipment information, failure probability, and total personnel irradiation dose of each implementation plan; The step of processing each of the sequence data to calculate the corresponding key parameters includes: Processing the equipment information and failure probability in each of the sequence data based on a probabilistic safety average model to generate corresponding core damage frequency and early large-scale radioactive release frequency; Processing the cost data and revenue data in each of the sequence data based on a power generation risk assessment model and a financial model to generate corresponding power generation risk assessment data and net present value; Performing averaging processing on the total personnel irradiation dose of all the implementation plans in the sequence data to generate corresponding average irradiation dose; Performing summarization processing on the corresponding core damage frequency, the early large-scale radioactive release frequency, the net present value, the power generation risk assessment data, and the average irradiation dose to generate the key parameters of each of the sequence data.

5. The multi-objective decision-making method based on security, economy, and usability according to claim 1, characterized in that, The step of respectively performing averaging processing on each type of key parameters of all the sequence data to generate corresponding target parameters includes: Average the core damage frequency, early large radioactive release frequency, net present value, power generation risk assessment data, and total personnel irradiation dose for all embodiments in all the sequence data to generate corresponding target damage frequency, target radioactive release frequency, target net present value, target risk assessment data, and target irradiation dose; Summarize the corresponding target damage frequency, target radioactive release frequency, target net present value, target risk assessment data, and target irradiation dose to generate corresponding target parameters.

6. The multi-objective decision-making method based on security, economy, and usability according to claim 1, characterized in that, The step of calculating the similarity between the key parameters of each sequence data and the target parameters to generate the total similarity of each sequence data includes: Calculate the similarity of each type of key parameter in each sequence data according to the data types of each type of key parameter and the corresponding target parameter in each sequence data; Calculate the sum of the similarities of all types of key parameters of the sequence data to generate the total similarity of the sequence data.

7. The multi-objective decision-making method based on security, economy, and usability according to claim 6, wherein The step of calculating the similarity of each type of key parameter in each sequence data according to the data types of each type of key parameter and the corresponding target parameter in each sequence data includes: Judge the types of each type of key parameter and the corresponding target parameter in each sequence data: When the key parameter and the corresponding target parameter are numerical types, calculate the similarity of the key parameter according to the key parameter, the corresponding target parameter, the maximum and minimum values of the key parameter in all sequence data; When the key parameter is a numerical type and the corresponding target parameter is an interval type, calculate the similarity of the key parameter according to the key parameter, the minimum value and the maximum value of the interval range of the corresponding target parameter; When the key parameter is an interval type and the corresponding target parameter is a numerical type, calculate the similarity of the key parameter according to the minimum value and the maximum value of the interval range of the key parameter and the corresponding target parameter; When the key parameter and the corresponding target parameter are interval types, calculate the similarity of the key parameter according to the minimum value and the maximum value of the interval range of the key parameter and the minimum value and the maximum value of the interval range of the corresponding target parameter.

8. The multi-objective decision-making method based on security, economy, and usability according to claim 7, characterized in that When the key parameter and the corresponding target parameter are numerical, the similarity Sim(x i , a i ) of the key parameter is expressed as: where x i represents the i-th key parameter, ai represents the corresponding i-th target parameter, a min and a max respectively represent the maximum and minimum values of the key parameter.

9. The multi-objective decision-making method based on security, economy, and usability according to claim 7, wherein When the key parameter and the corresponding target parameter are interval types, the similarity Sim([x i ,y i ,[a i ,b i ) of the key parameter is expressed as: where [x i ,y i represents the interval range of the i-th key parameter, [a i ,b i represents the interval range of the i-th target parameter, u represents the intermediate value of the interval range of the i-th key parameter, and v represents the intermediate value of the interval range of the i-th target parameter.

10. The multi-objective decision-making method based on security, economy, and usability according to claim 7, characterized in that, When the key parameter is interval - type and the corresponding target parameter is numerical - type, the similarity Sim([x i ,y i ,a i ) of the key parameter is expressed as: where [x i ,y i represents the interval range of the i - th key parameter, and ai represents the corresponding i - th target parameter.

Images