Method, device and system for selecting power energy of inland vessel, and storage medium

Abstract

The application discloses a kind of inland river ships power energy selection method, device, system, storage medium and computer program product, belong to inland river ships engineering technical field, this method includes: determining evaluation index set and alternative power energy type, qualitative evaluation information of each index is converted into fuzzy relation matrix based on fuzzy mathematics theory, entropy weight method is used to calculate the weight value of each index, combined with weight value and fuzzy relation matrix carries out fuzzy synthesis operation and the result is normalized, finally by solving evaluation total score, the relative closeness of each alternative type and optimal solution is calculated, the target power energy type of adaptation is determined.The scheme realizes the quantification, overall and precision of inland river ships power energy selection.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of inland river ships, and particularly relates to a power energy selection method, device, system, storage medium and computer program product for an inland river ship. BACKGROUND

[0002] Under the driving of the overall green transformation of the economic society, the process of applying clean energy to inland river ships in China has accelerated significantly. At present, progress has been made in the construction and application of small and medium-sized inland river ships using various clean energy powers, and the number of related ship orders is still increasing. At present, the clean energy power technology of inland river ships presents a diversified development trend. The types of power energy that can be selected include liquefied natural gas (LNG), methanol, pure lithium batteries, hydrogen fuel cells, etc. The type of power technology has a significant impact on the economy and emission reduction effect of ship operation, which makes the selection of power energy a focus problem for all parties in the industry chain when making investment, production and application decisions.

[0003] In the actual selection process of clean energy power technology for small and medium-sized inland river ships, economic factors such as construction cost, operation cost and maintenance cost, support factors such as the perfection of construction facilities, the perfection of energy supplement facilities and the perfection of maintenance facilities, and policy factors such as subsidy support and preferential navigation support, as well as some individual factors that cannot be determined, all affect the selection result. The existing selection method usually relies on fuzzy qualitative language such as "excellent, good, medium and poor" to describe and analyze the above factors, which is difficult to convert the selection decision-making process into accurate and quantifiable analysis results, resulting in a lack of accurate technical selection basis for the upstream, midstream and downstream related parties in the shipping, equipment, energy and port industry chains when making investment, production and application decisions for clean energy ships, and unable to effectively cope with the difficulties brought by the uncertainty of the technical route.

[0004] The above content is only used to assist in understanding the technical solutions of the present application and does not represent the acknowledgement of the above content as prior art. SUMMARY

[0005] The purpose of the present application is to provide a power energy selection method, device, system, storage medium and computer program product for an inland river ship, to solve the problem that the selection of clean energy power technology for inland river ships in related solutions is difficult to form accurate and quantifiable results, resulting in a lack of accurate selection basis for related parties in the industry chain and an inability to cope with the uncertainty of the technical route, achieving the effects of quantification, comprehensiveness and precision in the selection of power energy for inland river ships.

[0006] The application provides a power energy selection method for an inland waterway ship, comprising: determining an evaluation index set for power energy selection of the inland waterway ship, and specifying a type of a candidate power energy to be selected; obtaining qualitative evaluation information of each evaluation index in the evaluation index set, and converting the qualitative evaluation information into a fuzzy relation matrix based on fuzzy mathematics theory; the fuzzy relation matrix is used to represent membership degrees of each evaluation index under different evaluation levels; based on multi-index decision logic, an entropy weight method is used to perform hierarchical calculation and processing on the evaluation index set, to obtain weight values of each evaluation index in selection decision; the weight values are used to reflect selection priorities of each evaluation index under different dimension classifications in the evaluation index set; a fuzzy weight vector is determined based on the weight values, fuzzy synthesis operation is performed on the fuzzy weight vector and the fuzzy relation matrix, a fuzzy comprehensive evaluation vector is obtained, and normalization processing is performed on the fuzzy comprehensive evaluation vector; a corresponding evaluation total score is solved based on the fuzzy comprehensive evaluation vector, relative proximities of each candidate power energy type to an optimal solution are calculated based on the evaluation total score, and a target power energy type that is adapted is determined from the candidate power energy types based on the relative proximities.

[0007] In some embodiments, based on multi-index decision logic, hierarchical calculation and processing on the evaluation index set by using an entropy weight method comprises: dividing each evaluation index into positive feedback type indexes and negative feedback type indexes; for the positive feedback type indexes and the negative feedback type indexes, corresponding standardization formulas are used for standardization deformation respectively; normalized processing is performed on index values after standardization deformation, to obtain intermediate forms; entropy values of each evaluation index are calculated based on the intermediate forms, and a deviation coefficient is solved through the entropy values, and finally, weight values of each evaluation index are obtained based on the deviation coefficient.

[0008] In some embodiments, when the normalization processing is performed on the fuzzy comprehensive evaluation vector, a preset normalization formula is used to convert the fuzzy comprehensive evaluation vector into a numerical value of a uniform dimension.

[0009] In some embodiments, the relative proximities of each candidate power energy type to the optimal solution are calculated based on the evaluation total score, comprising: constructing a score vector, obtaining evaluation total scores of each candidate power energy type through vector operation of the score vector and the fuzzy comprehensive evaluation vector; based on the evaluation total scores, a first distance between each candidate power energy type and an optimal score, and a second distance between each candidate power energy type and a worst score are calculated respectively; the relative proximities are calculated based on the first distance and the second distance.

[0010] In some implementations, determining the suitable target power energy type from the candidate power energy types based on the relative proximity includes: sorting the relative proximity corresponding to each candidate power energy type in descending order of value; and selecting the candidate power energy type that is ranked first and whose relative proximity is greater than a preset threshold as the suitable target power energy type.

[0011] In some implementations, when the qualitative evaluation information is transformed into a fuzzy relation matrix based on fuzzy mathematics theory, the factor domain and the level domain of the evaluation index set are also simultaneously associated, so that the construction dimension of the fuzzy relation matrix corresponds one-to-one with the hierarchy and evaluation level of the evaluation index.

[0012] In conjunction with the above method, another aspect of the present invention provides a power energy selection device for inland waterway vessels, comprising: a determining unit configured to determine a set of evaluation indicators for power energy selection of inland waterway vessels and to identify candidate power energy types; a conversion unit configured to acquire qualitative evaluation information of each evaluation indicator in the set of evaluation indicators, and to convert the qualitative evaluation information into a fuzzy relation matrix based on fuzzy mathematics theory; the fuzzy relation matrix is ​​used to characterize the membership degree of each evaluation indicator under different evaluation levels; and a calculation unit configured to perform hierarchical calculation processing on the set of evaluation indicators based on multi-indicator decision logic and using the entropy weight method to obtain the evaluation indicators in the selection decision. The weight values ​​in the evaluation index set are used to reflect the selection priority of each evaluation index under different dimension classifications. The calculation unit is also configured to determine a fuzzy weight vector based on the weight values, perform fuzzy synthesis operation with the fuzzy weight vector and the fuzzy relation matrix to obtain a fuzzy comprehensive evaluation vector, and perform normalization processing on the fuzzy comprehensive evaluation vector. The selection unit is configured to solve the corresponding total evaluation score based on the fuzzy comprehensive evaluation vector, calculate the relative closeness between each of the candidate power energy types and the optimal solution based on the total evaluation score, and determine the suitable target power energy type from the candidate power energy types based on the relative closeness.

[0013] In some implementations, the calculation unit, based on multi-index decision-making logic, uses the entropy weight method to perform hierarchical calculation processing on the set of evaluation indicators, including: dividing each evaluation indicator into positive feedback indicators and negative feedback indicators; standardizing and transforming the positive feedback indicators and negative feedback indicators respectively using corresponding standardization formulas; normalizing the standardized indicator values ​​to obtain intermediate formulas; calculating the entropy value of each evaluation indicator based on the intermediate formulas; then solving for the deviation coefficient through the entropy value; and finally obtaining the weight value of each evaluation indicator based on the deviation coefficient.

[0014] In some embodiments, the computing unit, when normalizing the fuzzy comprehensive evaluation vector, converts the fuzzy comprehensive evaluation vector into a numerical value of a uniform dimension by using a preset normalization formula.

[0015] In some embodiments, the selecting unit calculates the relative closeness of each of the candidate power energy types to the optimal solution in combination with the evaluation total score, including: constructing a score vector, and obtaining the evaluation total score of each of the candidate power energy types through vector operation of the score vector and the fuzzy comprehensive evaluation vector; based on the evaluation total score, calculating a first distance of each of the candidate power energy types to the optimal score and a second distance to the worst score, respectively; and calculating the relative closeness based on the first distance and the second distance.

[0016] In some embodiments, the selecting unit determines the target power energy type from the candidate power energy types based on the relative closeness, including: sorting the relative closeness corresponding to each of the candidate power energy types in descending order of numerical value; and taking the candidate power energy type with the highest relative closeness and a relative closeness greater than a preset threshold as the target power energy type.

[0017] In some embodiments, the converting unit, when converting the qualitative evaluation information into a fuzzy relation matrix based on fuzzy mathematics theory, simultaneously associates a factor domain and a level domain of the evaluation index set, so that the construction dimension of the fuzzy relation matrix corresponds to the level and evaluation level of the evaluation index one by one.

[0018] To match the above device, the present application further provides a power energy selection system for an inland ship, including: the power energy selection device for an inland ship as described above.

[0019] To match the above method, the present application further provides a storage medium including a stored program, wherein the program controls the device where the storage medium is located to execute the power energy selection method for an inland ship as described above when the program is running.

[0020] To match the above method, the present application further provides a computer program product including a computer program, which realizes the steps of the power energy selection method for an inland ship when the computer program product is processed and executed.

[0021] The scheme of the present application determines the evaluation index set and the alternative power energy type, converts qualitative evaluation information of each index into a fuzzy relation matrix based on fuzzy mathematics theory, calculates the weight value of each index by using entropy weight method, performs fuzzy synthesis operation combining the weight value and the fuzzy relation matrix and normalizes the result, and finally determines the adapted target power energy type by solving the total score of evaluation and calculating the relative closeness of each alternative type to the optimal solution. The quantitative, comprehensive and accurate selection of the power energy type of the inland ship is realized, and the reliability of the selection result is improved.

[0022] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application.

[0023] The technical solutions of the present application will be further described in detail below with the help of the accompanying drawings and embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 The figure is a flowchart of an embodiment of the power energy selection method of the inland ship of the present application.

[0025] Figure 2 The figure is a structural schematic diagram of an embodiment of the power energy selection device of the inland ship of the present application.

[0026] Figure 3 The figure is a structural schematic diagram of the power energy selection system of the inland ship.

[0027] In combination with the accompanying drawings, the reference signs in the embodiments of the present application are as follows:

[0028] 101-determining unit; 102-converting unit; 103-calculating unit; 104-selection unit. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described clearly and completely below in combination with specific embodiments of the present application and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the present application.

[0030] According to the embodiments of the present application, a power energy selection method for an inland ship is provided, as shown in Figure 1 The figure is a flowchart of an embodiment of the method of the present application. The power energy selection method for the inland ship can include steps S110 to S150.

[0031] At step S110, the evaluation index set of the selection of the power energy of the inland river ship is determined, and the type of the alternative power energy to be selected is specified.

[0032] The evaluation index set is a combination of indexes built for accurately judging the adaptability of the power energy of the inland river ship, including endogenous power indexes and external support indexes. The endogenous power indexes are comprehensive indexes that measure the inherent competitiveness of the power energy of the ship in terms of cost, efficiency, performance, etc. from the technical and operational characteristics of the power energy of the ship, including economic indexes. The external support indexes are comprehensive indexes that measure the external resource and condition support obtained by the power energy of the ship in terms of facility support, policy support, and external environment adaptation, including security indexes and policy indexes.

[0033] For the endogenous power indexes, the shipbuilding and operation are market-oriented, and profits need to be obtained by controlling costs. The construction (initial investment), operation (fuel / supplies), and maintenance (equipment update) are the three most critical stages that affect costs, so the endogenous power indexes include construction cost, operation cost, and maintenance cost, which directly determine the economic feasibility of technology selection.

[0034] For the external support indexes, whether new energy ships can be put into operation depends on the support of construction hardware (such as specialized shipyards), energy supplement networks (such as LNG filling stations and charging piles), and maintenance networks. If these facilities are lacking, even if the technology is economically feasible, it cannot be practically applied. Therefore, the external support indexes include construction facility perfection, energy supplement facility perfection, and maintenance facility perfection. Moreover, new energy ships have high initial investment (such as hydrogen fuel cell ships, which have significantly higher costs than traditional ships), subsidies can alleviate cost pressure, and preferential navigation can improve operational efficiency. These two types of policies directly affect the market competitiveness of the technology and are key factors that cannot be ignored in selection, so the external support indexes also include subsidy support intensity and preferential navigation support intensity. The final evaluation index set includes two first-level indexes (endogenous power indexes and external support indexes) and eight second-level indexes (construction cost, operation cost, maintenance cost, construction facility perfection, energy supplement facility perfection, maintenance facility perfection, subsidy support intensity, and preferential navigation support intensity).

[0035] The alternative power energy types include liquefied natural gas (LNG) power, methanol power, hydrogen fuel cell power, and pure lithium battery power, which are clean energy with relatively mature industrial chains. These clean energy technologies can avoid the problem of missing industrial chain support due to being too advanced.

[0036] Specifically, in the index screening, the core influencing factors of the application of power energy of inland river vessels are combed, and a dual framework of endogenous power index and external support index is constructed. The endogenous power index screens the cost, efficiency and other self-characteristics index; the external support index screens the external condition index such as facility supporting and policy support, forming a non-redundant and non-missing evaluation index set. In determining the type of the selected power energy, the clean energy power technologies that have appeared in the current inland river vessel field are collected, and based on the standards such as technology maturity, pilot verification, and inland river environment adaptability, the types that have not been piloted, unstable technology or missing industrial chain supporting are excluded, and the selected power energy types with actual application feasibility are determined, such as the final selection of LNG, methanol, hydrogen fuel cell, and pure lithium battery power as the selected type.

[0037] At step S120, qualitative evaluation information of each evaluation index in the evaluation index set is obtained, and the qualitative evaluation information is converted into a fuzzy relation matrix based on fuzzy mathematics theory; the fuzzy relation matrix is used to represent the membership degree of each evaluation index under different evaluation levels.

[0038] The qualitative evaluation information is obtained through industry expert evaluation, field research, data statistics and other methods, and is a non-quantitative description of the importance or performance of each evaluation index (such as “important”, “general”, “perfect”, “insufficient”, etc.), which is used to reflect the index characteristics that are difficult to be quantified directly. The fuzzy mathematics theory is a theoretical method that relies on fuzzy mathematics tools to convert elements with fuzzy boundaries and difficult to quantify into quantitative indexes. The core is to realize the quantitative processing of qualitative information through membership function and fuzzy relation matrix. The fuzzy relation matrix is a structured matrix constructed based on fuzzy mathematics theory, which is used to represent the membership degree of each evaluation index under different evaluation levels (i.e. the degree to which a certain index belongs to a certain evaluation level), and is a key carrier connecting qualitative evaluation and quantitative calculation.

[0039] Some evaluation indexes (such as facility perfection and policy support) are difficult to be quantified directly, and need to be reflected by qualitative evaluation; the fuzzy mathematics theory can solve the problem that qualitative information cannot be directly involved in calculation, and converts fuzzy description into structured quantitative data, laying a foundation for subsequent calculation.

[0040] Specifically, by inviting experts in the fields of ship engineering, energy application, and shipping operation, and combining field research data, the importance or performance of each evaluation index is qualitatively evaluated to form qualitative evaluation information such as “important”, “better”, “general”, and “insufficient”. The factor domain (all evaluation index set) and evaluation level domain (such as “good, better, general, worse, poor”) of the evaluation index are determined, the membership degree of each evaluation index under different evaluation levels is solved, the fuzzy relation matrix is constructed based on the membership degree, and the matrix elements accurately represent the belonging degree of each index under the corresponding evaluation level.

[0041] In some embodiments, when the qualitative evaluation information is converted into a fuzzy relation matrix based on fuzzy mathematics theory, the factor domain and the grade domain of the evaluation index set are also associated synchronously, so that the construction dimension of the fuzzy relation matrix corresponds to the hierarchy and evaluation grade of the evaluation index.

[0042] The factor domain of the evaluation index set refers to a set composed of all the indexes participating in the evaluation of the selection of the power energy source, which is a clear definition of the evaluation dimension, and specifically corresponds to all the subordinate specific indexes of the endogenous power index and the external support index. The grade domain refers to a set composed of all the grade entries used to evaluate the performance of the index (such as “good, better, general, worse, and poor”), which is a standardized division of the performance degree of the index.

[0043] The factor domain clearly defines the specific index range of the evaluation, and the grade domain clearly defines the standard grade of the evaluation. If the construction of the fuzzy relation matrix is not associated with the two, it will lead to confusion in the matrix dimension (such as the number of indexes not matching the number of rows of the matrix, and the number of evaluation grades not matching the number of columns of the matrix), and the corresponding relationship between the index and the evaluation grade cannot be accurately represented. The evaluation index has a clear hierarchical structure, and if the fuzzy relation matrix does not conform to this hierarchy, the evaluation data of different hierarchical indexes will be confused, affecting the accuracy of subsequent weight calculation and comprehensive evaluation. The synchronous association of the factor domain, the grade domain, and the index hierarchy can ensure that the construction logic of the fuzzy relation matrix is completely consistent with the evaluation system, so that the matrix can accurately carry the membership degree information of each hierarchical index under different evaluation grades.

[0044] Specifically, first, all the specific indicators participating in the selection evaluation are combed to form a clear factor domain (such as {F1: construction cost, F2: operation cost, F3: construction facility perfection, F4: complementary facility perfection, F5: subsidy support intensity, F6: priority navigation support intensity}); then, according to the selection accuracy requirement, a standardized grade domain (such as {G1: good, G2: better, G3: general, G4: worse, G5: poor}) is determined to ensure that the domain range is not redundant and not missed. Then, the core level to which each indicator belongs is determined to form a hierarchical relationship between the core level and the specific indicators (such as "endogenous power indicators-F1, F2; external support indicators-F3, F4, F5, F6"), ensuring that the hierarchical division is clear and there is no cross-level indicator confusion. Then, according to the number of indicators in the factor domain (such as 6) and the number of evaluation grades in the grade domain (such as 5), the dimension of the fuzzy relationship matrix is determined to be 6 rows and 5 columns (the number of rows corresponds to the number of indicators, and the number of columns corresponds to the number of evaluation grades); each row of the matrix corresponds to a specific indicator in the factor domain, and the core level to which the row belongs is consistent with the level of the indicator itself (such as the rows corresponding to F1, F2 belong to the "endogenous power indicator" related row group, and the rows corresponding to F3-F6 belong to the "external support indicator" related row group); each column of the matrix corresponds to an evaluation grade in the grade domain, and the matrix elements are the membership degrees of the corresponding indicators under the evaluation grade, and finally a fuzzy relationship matrix is formed which is completely matched with the structure and domain, and the level. Finally, check whether the number of rows of the fuzzy relationship matrix is consistent with the number of indicators in the factor domain, whether the number of columns is consistent with the number of evaluation grades in the grade domain, whether the level of each indicator row is accurate, and ensure that there is no dimension error or level confusion in the matrix construction.

[0045] At step S130, based on the multi-index decision logic, the entropy weight method is used to perform hierarchical calculation processing on the evaluation index set to obtain the weight value of each evaluation index in the selection decision; the weight value is used to reflect the selection priority of each evaluation index under different dimension classifications in the evaluation index set.

[0046] The multi-index decision logic is a logic system that comprehensively considers multiple interrelated or conflicting evaluation indicators, integrates the influence of each indicator through a specific calculation method, and realizes optimal decision, which is used to coordinate the effect of different dimension indicators on the selection result. The entropy weight method is a method for objectively calculating the weight based on the dispersion degree of index data, which determines the importance of the index by measuring the entropy value, and the smaller the entropy value (the higher the dispersion degree of the index), the greater the weight, which can avoid the deviation caused by subjective judgment. The weight value can represent the relative importance of each evaluation index in the selection decision, and the higher the weight value, the greater the influence of the index on the final selection result.

[0047] The influence degree of different evaluation indexes on the selection result is different, and the weight value can quantify the difference; the entropy weight method objectively assigns weights based on the dispersion degree of index data, avoids the deviation caused by expert subjective judgment, and ensures the rationality and fairness of weight distribution; the multi-index decision logic can coordinate the correlation or conflict relationship between indexes, and realize the comprehensive optimization.

[0048] Specifically, first, the evaluation indexes are divided into positive feedback type indexes (the higher the index value, the better, such as facility perfection) and negative feedback type indexes (the lower the index value, the better, such as cost). The corresponding standardization formula is used for standardization transformation for the two types of indexes respectively, and the dimensional difference is eliminated; the normalized value after standardization is normalized to obtain an intermediate form; the entropy value of each index is calculated based on the intermediate form, the deviation coefficient is solved by entropy value (deviation coefficient = 1-entropy value), and finally the deviation coefficient is normalized to obtain the weight value of each evaluation index.

[0049] In some embodiments, in step S130, based on the multi-index decision logic, the entropy weight method is used for hierarchical calculation and processing of the evaluation index set, including steps S210 to S240.

[0050] Step S210, dividing each evaluation index into positive feedback type index and negative feedback type index.

[0051] The evaluation index set contains two types of indexes with opposite action directions. If not classified and directly converted by the same formula, the index advantage and disadvantage trend will not be consistent with the numerical value (for example, the cost index value is large, which represents the disadvantage, and if the same formula is used as the positive index, it will be misjudged as an advantage). After classification, the appropriate standardization formula can be used for targeted processing to ensure that the larger the value of all indexes, the better.

[0052] Specifically, the characteristics and action direction of each index in the evaluation index set are analyzed one by one, and the correlation between the index value and the power energy adaptability is determined. The index with the characteristic of “the larger the value, the better the power energy adaptability” is classified as a positive feedback type index, such as construction facility perfection, energy supplement facility perfection, subsidy support intensity, and priority navigation support intensity. The index with the characteristic of “the smaller the value, the better the power energy adaptability” is classified as a negative feedback type index, such as construction cost and operation cost.

[0053] Step S220, for the positive feedback type index and the negative feedback type index, the corresponding standardization formula is used for standardization transformation.

[0054] The raw data for different evaluation indicators have different dimensions (e.g., construction cost is measured in "ten thousand yuan", facility completeness in "grades", and subsidy support in "percentages"). Directly using the raw data for calculations would result in incomparable numerical values ​​due to these different dimensions, thus affecting the objectivity of weight calculations. Standardization can map all indicator data to the [0,1] interval, eliminating dimensional differences and making data from different types of indicators horizontally comparable. Standardization formulas are used to transform raw indicator data with different dimensions and numerical ranges into dimensionless data within a unified range (usually [0,1]), eliminating the impact of dimensional differences on the calculation results.

[0055] Specifically, for positive feedback indicators, a standardized formula is used. ( Let j be the original value of the i-th alternative power energy type under the j-th index. The minimum value among all candidate types under the j-th indicator. (where the maximum value of all candidate types under the j-th indicator is used), the original data is converted into dimensionless data in the interval [0,1]; for negative feedback indicators, a standardized formula is used. Similarly, the original data is converted into dimensionless data in the range [0,1] to ensure that both types of indicators achieve the unified logic that "the larger the value, the better the adaptability".

[0056] Step S230: Normalize the standardized and deformed index values ​​to obtain the intermediate formula.

[0057] Although the standardized indicator data has eliminated dimensional differences, the data distribution still does not meet the requirements for entropy calculation (entropy calculation needs to be based on the logic of "data proportion"). Normalization can convert the standardized data of each indicator into the proportion of all alternative data types under that indicator, so that the data presents the characteristics of "relative contribution", providing reasonable input for accurately calculating the entropy value (measuring the degree of data dispersion).

[0058] Specifically, for each evaluation indicator, standardized and modified data of all candidate power energy types under that indicator are collected, and a normalization formula is used. ( Given the standardized data of the i-th candidate type under the j-th indicator (the denominator is the sum of the standardized data of all candidate types under all indicators), we transform each standardized data point to obtain the corresponding intermediate formula. This ensures that the sum of the intermediate expressions for all candidate types under each indicator and the sum of the intermediate expressions for all indicators satisfy the logical requirements for entropy calculation.

[0059] Step S240, entropy values of each evaluation index are calculated based on the intermediate formula, and then a deviation coefficient is solved by using the entropy values, and finally the weight value of each evaluation index is obtained based on the deviation coefficient.

[0060] The entropy value is a core index for measuring the discrete degree of index data. The higher the discrete degree of data, the more effectively the index can distinguish the adaptability difference of different alternative power energy types, and the greater the reference value for selection decision-making. The deviation coefficient is calculated by "1-entropy value", which can directly quantify the importance potential of the index.

[0061] Specifically, based on the intermediate formula , the entropy value calculation formula is adopted (where k is the number of alternative power energy types, n is the number of evaluation indexes, if , then is processed as 0), and the entropy value of each evaluation index is calculated one by one ; then the entropy value of each index is converted into the corresponding deviation coefficient by using the deviation coefficient formula . The greater the deviation coefficient, the stronger the importance potential of the index.

[0062] The deviation coefficient can only reflect the relative size of the importance potential of the index, and does not form a unified weight distribution system (the sum of the deviation coefficients of all indexes is not necessarily 1). By normalizing the deviation coefficient, it can be converted into a weight value, ensuring that the sum of the weight values of all indexes is 1, forming an intuitive and comparable importance ratio.

[0063] Specifically, the deviation coefficients of all evaluation indexes are collected , the sum of all deviation coefficients is calculated (n is the number of evaluation indexes), the weight value calculation formula is adopted, and the weight value of each evaluation index is calculated one by one . The final weight value set needs to satisfy that the sum of all weight values is 1, ensuring the rationality and logic of weight distribution.

[0064] At step S140, a fuzzy weight vector is determined based on the weight value, a fuzzy synthesis operation is performed combining the fuzzy weight vector and the fuzzy relation matrix to obtain a fuzzy comprehensive evaluation vector, and the fuzzy comprehensive evaluation vector is normalized.

[0065] The fuzzy weight vector is a vector constructed based on the weight values of each evaluation index, and is used to reflect the importance proportion of different indexes in the fuzzy comprehensive evaluation. The fuzzy synthesis operation is a process of operating the fuzzy weight vector and the fuzzy relation matrix through a specific fuzzy operator, and is used to integrate the evaluation results of each index to obtain a comprehensive evaluation vector. The fuzzy comprehensive evaluation vector can represent the comprehensive performance of all evaluation indexes under different evaluation grades, and reflect the overall adaptability trend of the candidate power energy type. The normalization processing is to convert the fuzzy comprehensive evaluation vector into a value with unified dimension and range, eliminate the dimensional differences brought by different indexes or evaluation grades, and ensure the fairness and accuracy of subsequent calculations.

[0066] The fuzzy weight vector can reflect the importance proportion of each index, ensuring that important indexes play a leading role in the comprehensive evaluation. The fuzzy synthesis operation can integrate the evaluation results of each index to form an overall evaluation trend. The normalization processing can eliminate the numerical differences brought by different evaluation grades or indexes, making the evaluation results comparable.

[0067] Specifically, the weight values of each evaluation index are arranged in the corresponding order to construct a fuzzy weight vector, ensuring that the vector dimension is consistent with the number of evaluation indexes. A fuzzy operator (such as the max-min operator, the product-addition operator, etc.) is used to perform synthesis operation on the fuzzy weight vector and the fuzzy relation matrix to obtain a fuzzy comprehensive evaluation vector, which reflects the comprehensive membership of all indexes under each evaluation grade. A pre-set normalization formula (such as linear normalization, normalization to the [0, 1] interval, etc.) is used to convert the fuzzy comprehensive evaluation vector into a value with unified dimension and range, completing the normalization processing.

[0068] In some embodiments, when the fuzzy comprehensive evaluation vector is normalized, a pre-set normalization formula is used to convert the fuzzy comprehensive evaluation vector into a value with unified dimension.

[0069] The fuzzy comprehensive evaluation vector is the original result obtained through fuzzy synthesis operation, and its value may differ due to the number of evaluation grades and the distribution of membership. If it is directly used for subsequent calculations, the comparison will be distorted due to the inconsistent value scale. The fuzzy comprehensive evaluation vectors corresponding to different candidate power energy types lack a unified measurement benchmark, and it is difficult to objectively judge their adaptability without normalization processing. Normalization processing can map all vector values to a unified interval, eliminate dimensional and scale differences, and make the comprehensive evaluation results of different candidate types comparable.

[0070] Specifically, the original data characteristics of the fuzzy comprehensive evaluation vector are determined, including vector dimension (consistent with the number of evaluation levels), numerical range, membership degree distribution law, etc., to ensure that the normalization formula is adapted to the data characteristics; a preset normalization formula is selected, common formulas include linear normalization formula, interval normalization formula, etc., and the specific formula selection needs to be combined with the data resolution requirements of the selection decision; each value in the fuzzy comprehensive evaluation vector is substituted into the selected normalization formula, and each value is calculated to obtain the corresponding standardized value; the normalized values are checked to ensure that all values are in the preset unified interval (such as [0, 1]), and the value distribution can truly reflect the membership degree trend of the original vector.

[0071] At step S150, the corresponding evaluation total score is solved based on the fuzzy comprehensive evaluation vector, the relative closeness of each of the candidate power energy types to the optimal solution is calculated combined with the evaluation total score, and the adapted target power energy type is determined from the candidate power energy types based on the relative closeness.

[0072] The evaluation total score is a quantitative value obtained by vector operation based on the fuzzy comprehensive evaluation vector and the preset score vector, which directly reflects the comprehensive adaptation level of the candidate power energy type. The relative closeness is an index that quantifies the closeness to the theoretical optimal solution by calculating the distance between the candidate power energy type and the optimal score and the worst score, and the larger the value, the better the adaptability.

[0073] The evaluation total score converts the fuzzy comprehensive evaluation vector into a direct quantitative value, which is convenient for horizontal comparison of each candidate type; the relative closeness quantifies the adaptability by the distance from the theoretical optimal solution and the worst solution, avoiding the problem that a single value is difficult to judge the adaptation degree; the selection based on the relative closeness can ensure the objectivity and accuracy of the results, avoiding subjective decision bias.

[0074] Specifically, a score vector corresponding to the evaluation level domain is constructed (such as "good = 90 points, better = 80 points, general = 70 points, worse = 60 points, poor = 50 points"), and the fuzzy comprehensive evaluation vector is converted into the evaluation total score of each candidate power energy type through vector operation. The evaluation total scores of all candidate types are counted to determine the optimal score (the maximum value among all scores) and the worst score (the minimum value among all scores); the first distance of each candidate type from the optimal score and the second distance from the worst score are calculated, and the relative closeness of each candidate type is calculated through the formula (relative closeness = second distance / (first distance + second distance)). The relative closeness of each candidate type is sorted in descending order, and the type with the highest ranking and a relative closeness greater than a preset threshold (set according to actual operation requirements or industry standards) is selected as the adapted target power energy type.

[0075] In some embodiments, in step S150, in combination with the calculation of the evaluation total score, the relative closeness of each of the candidate power energy types to the optimal solution is calculated, including steps S310 to S330.

[0076] In step S310, a score vector is constructed, and the evaluation total score of each of the candidate power energy types is obtained through vector operation of the score vector and the fuzzy comprehensive evaluation vector.

[0077] The score vector refers to a vector containing corresponding quantitative scores preset according to the evaluation level domain, and the dimension is consistent with the fuzzy comprehensive evaluation vector, which is used to convert the fuzzy membership distribution into an intuitive total score. The fuzzy comprehensive evaluation vector can only reflect the membership distribution of each evaluation level, and cannot directly reflect the comprehensive adaptation level of the candidate type; the score vector can give each evaluation level a clear quantitative score, and through vector operation, the fuzzy distribution is converted into a single total score, realizing the horizontal comparison of different candidate types.

[0078] Specifically, according to the setting of the evaluation level domain, the score vector is constructed in combination with the accuracy requirement of the selection decision and the industry conventional scoring standard. For example, if the evaluation level is 5 levels, the score vector can be set as [90, 80, 70, 60, 50] (the higher the score, the better the evaluation level), ensuring that the dimension of the score vector is completely consistent with the fuzzy comprehensive evaluation vector. For the fuzzy comprehensive evaluation vector corresponding to each candidate power energy type, the vector dot product operation rule is used for calculation. Assuming that the fuzzy comprehensive evaluation vector of a certain candidate type is [S1, S2, S3, S4, S5], and the score vector is [F1, F2, F3, F4, F5], then the evaluation total score Z = S1xF1 + S2xF2 + S3xF3 + S4xF4 + S5xF5, and the membership distribution is converted into a single quantitative score through the operation. Repeat the above operation to obtain the evaluation total score of all candidate power energy types one by one, and form a comparable total score set.

[0079] In step S320, based on the evaluation total score, the first distance of each of the candidate power energy types from the optimal score and the second distance from the worst score are calculated respectively.

[0080] The optimal score refers to the maximum value of the evaluation total score of all candidate power energy types, reflecting the highest level of power energy adaptability in the current selection range. The worst score refers to the minimum value of the evaluation total score of all candidate power energy types, which is a quantitative embodiment of the lowest level of adaptability in the selection range. The first distance and the second distance can quantify the relative position of the candidate type from the optimal and the worst.

[0081] Specifically, from the evaluation total score of all candidate power energy types, the score with the maximum value is selected as the optimal score Z max, the minimum value of the scores is selected as the worst score Z min ; the evaluation total score Z for each alternative power energy type i , the first distance D1 is calculated using the absolute value distance formula i = |Z i - Z max |, which directly reflects the gap between the alternative type and the optimal adaptation state, and the smaller the value, the closer to the optimal; similarly, the evaluation total score Z for each alternative type i , the second distance D2 is calculated using the absolute value distance formula i = |Z i - Z min |, which reflects the gap between the alternative type and the worst adaptation state, and the larger the value, the better the adaptation than the worst level.

[0082] Step S330, the relative closeness is calculated based on the first distance and the second distance.

[0083] Specifically, for each alternative power energy type, the relative closeness calculation formula C i = D2 i / (D1 i + D2 i ) is used to calculate (C i is the relative closeness of the i-th alternative type). After the calculation is completed, the calculation result is checked to ensure that the relative closeness value is in the interval [0, 1]: if the evaluation total score of a certain alternative type is equal to the optimal score (Z i = Z max ), then D1 i = 0, C i = 1, which means that the type completely meets the optimal adaptation state; if the evaluation total score of a certain alternative type is equal to the worst score (Z i = Z min ), then D2 i = 0, C i = 0, which means that the type is in the worst adaptation state; in other cases, the relative closeness value is between 0 and 1, and the larger the value, the better the adaptation. Finally, the relative closeness of all alternative power energy types is calculated to form a complete set of relative closeness.

[0084] In some embodiments, in step S150, the adapted target power energy type is determined from the alternative power energy types based on the relative closeness, including: sorting the relative closeness corresponding to each of the alternative power energy types in descending order of value; and selecting the alternative power energy type with the highest ranking and a relative closeness greater than a preset threshold as the adapted target power energy type.

[0085] The preset threshold is used to determine whether the adaptability of the candidate power energy type meets the minimum requirement. Through the double rules of sorting priority and threshold screening, it is ensured that the selected type is relatively optimal and its adaptability meets the actual demand, avoiding the selection of unqualified types due to single sorting.

[0086] Specifically, the relative closeness values corresponding to all candidate power energy types are collected to form a complete relative closeness dataset (such as LNG power 0.92, methanol power 0.75, hydrogen fuel cell power 0.31, and pure lithium battery power 0.28). According to the rule that "the larger the value, the higher the priority", the dataset is arranged in descending order to obtain a clear adaptability sorting result (such as LNG power > methanol power > hydrogen fuel cell power > pure lithium battery power), which ensures that the sorting result can directly reflect the relative adaptability level of each candidate type. The top candidate power energy type in the descending order is determined, and it is judged whether the relative closeness value is greater than the preset threshold. If the relative closeness of the type is greater than the preset threshold (such as the relative closeness of the first LNG power 0.92 > 0.6), it is directly determined as the target power energy type; if the relative closeness of the type is less than or equal to the preset threshold (such as the relative closeness of the first type 0.55 ≤ 0.6), the previous evaluation process (such as index selection, weight calculation, and relative closeness calculation) is rechecked for deviation, or the preset threshold is adjusted, the candidate type is supplemented, and the selection process is reperformed until the target type that meets the requirements is screened out.

[0087] In some embodiments, the specific process of evaluating each index by using the fuzzy comprehensive evaluation method includes steps 1 to 7.

[0088] Step 1, based on the distribution range of the evaluation index, determine the relevant factor domain F. F = {f1, f2, …, fm}, which represents the set formed after integrating m evaluation indexes. m}, indicating the set of p level terms possessed by each evaluation index. For example, the level set is taken as R = {good, better, average, worse, poor}.

[0089] Step 2, based on the degree level of the evaluation index, determine the relevant level domain G. G = {g1, g2, …, gp}, which represents the set composed of p level terms of each evaluation index. For example, the level set is taken as R = {good, better, average, worse, poor}. p}, indicating the set of p level terms possessed by each evaluation index. For example, the level set is taken as R = {good, better, average, worse, poor}.

[0090] Step 3, based on each evaluation index, construct the algebraic form of the fuzzy relation matrix R. Specifically, after determining the factor domain and the level domain, the factor of each evaluation index f i (i = 1, 2, …, m) is quantified, that is, the membership degree N~(f i ) of the fuzzy subset of each evaluation index under different levels is solved, and thus the fuzzy relation matrix N is obtained.

[0091]

[0092] In the matrix, the element eij in the ith row and jth column means the evaluation index qj ij , and its meaning is the membership degree of the fuzzy subset corresponding to the evaluation grade r i This single dimension is used to obtain the evaluated object belonging to the evaluation grade r j .

[0093] Step 4, an expert team participating in the evaluation is organized, and a scoring method is used to quantitatively evaluate according to the relevant grade standards, so as to concretize the abstract form of the matrix N. First, the value range [0, 10] of the score value is divided into 5 continuous subintervals, which are (8, 10], (6.5, 8], (4.5, 6.5], (3, 4.5], and [0, 3] in turn, and the 5 subintervals correspond to the v1 (good), v2 (better), v3 (general), v4 (worse), and v5 (poor) grades in the evaluation set. Subsequently, statistical analysis is performed on the score results of all the experts, and the membership degree values of the evaluation factors corresponding to different evaluation grades are calculated.

[0094] Step 5, a relevant fuzzy weight vector J (j1, j2, …, j m ) is determined. In an actual evaluation scene, the importance of the m evaluation factors to the evaluated object is often inconsistent, and the performance of each single factor also has different effects on the overall evaluation result, so the relevant fuzzy weight vector needs to be determined before the fuzzy relationship synthesis operation is performed.

[0095] Step 6, the fuzzy operator is used to perform a synthesis operation on J and N, and the fuzzy comprehensive evaluation vector S of each evaluated object can be obtained. The vector is the result of degree description of the comprehensive condition of each evaluation index according to different grades, and the calculation formula is S = J × N.

[0096] The fuzzy comprehensive evaluation result vector obtained through the above operation is:

[0097] The calculation result is: JN = (0.292, 0.333, 0.178, 0.019, 0.010) and the normalized processing result is (0.383, 0.362, 0.206, 0.026, 0.013).

[0098] Step 7, the evaluation total score is solved through vector operation. The score vector H is constructed, H = (90, 80, 70, 60, 50), and the evaluation total score Z can be calculated by the following formula: Z = JN × F.

[0099] In some embodiments, the specific process of solving the index weight by using the entropy weight method includes steps 11 to 15.

[0100] Step 11, index classification, dividing the evaluation index into positive feedback type index and negative feedback type index.

[0101] Step 12, judgment matrix standardization deformation, for positive feedback type index and negative feedback type index, respectively, using the corresponding formula to eliminate the dimensional effect, the standardization deformation formula of positive feedback type index is as follows:

[0102]

[0103] The standardization deformation formula of negative feedback type index is as follows:

[0104]

[0105] Step 13, normalization processing, n ij After normalization processing, the intermediate formula m ij (K ij is the standardized index value):

[0106]

[0107] Step 14, calculate the index entropy value, using the intermediate formula m ij The entropy value corresponding to the kth index can be calculated as follows:

[0108]

[0109] From the above formula, the entropy values of each secondary evaluation index under the first-level evaluation index X1 are e1={0.698, 0.8123, 0.7895, 0.8012}. The entropy values of each secondary evaluation index under the second-level evaluation index X2 are e2={0.7002, 0.7312, 0.7256, 0.8166}, and the entropy values of each secondary evaluation index under the third-level evaluation index X3 are e3={0.7765, 0.8165, 0.5712, 0.8022}.

[0110] Step 15, calculate the deviation coefficient and entropy weight value, calculate the deviation coefficient c of each evaluation index, the formula is: j W j =1-t

[0111] Based on the above calculation results, the entropy weight value of each evaluation index is calculated, the formula is: .

[0112] The calculation result shows that the entropy weight values of each secondary evaluation index under the first-level evaluation index are C1=[0.1288, 0.1235, 0.1036 0, 0.1135, 0.0925]; the entropy weight values of each secondary evaluation index under the secondary evaluation index X2 are C2=[0.1128, 0.1365, 0.1653, 0.0928, 0.0903]; the entropy weight values of each secondary evaluation index under the tertiary evaluation index X3 are C3=[0.4912, 0.1689, 0.2723, 0.1662]; and the entropy weight values of the three first-level evaluation indexes are C=[0.3788, 0.2998, 0.3686].

[0113] The scheme realizes the comprehensiveness, quantification and precision of the selection of the power energy of the inland ship. The problems of subjective weight distribution and difficulty in quantifying the qualitative evaluation in the traditional selection are solved, the risk caused by the uncertainty of the technical route is effectively reduced, and reliable support is provided for the investment, production and application decision of the relevant parties in the inland ship industry chain.

[0114] For example, when selecting a power energy type for an inland river cargo ship, the internal power index selects the construction cost and operation cost; the external support index selects the construction facility perfection, energy supplement facility perfection, subsidy support intensity, and preferential navigation support intensity, forming an evaluation index set of six specific indexes; the selected power energy types are LNG power, methanol power, hydrogen fuel cell power, and pure lithium battery power. Five ship industry experts are invited to evaluate each index, such as the energy supplement facility perfection of LNG power being evaluated as 'good' and the construction facility perfection of hydrogen fuel cell power being evaluated as 'general'; the factor domain is determined as the six indexes, the level domain is determined as 'good, better, general, worse, and poor', the membership degree of each index is calculated, and a 6*5 order fuzzy relationship matrix is constructed. The construction cost and operation cost are divided into negative feedback indexes, and the rest are positive feedback indexes; standardization and normalization processing are respectively performed, the construction cost weight is 0.25, the operation cost weight is 0.20, the construction facility perfection weight is 0.15, the energy supplement facility perfection weight is 0.18, the subsidy support intensity weight is 0.12, and the preferential navigation support intensity weight is 0.10. Based on the weight value, a fuzzy weight vector is constructed, and the fuzzy relationship matrix is combined through a product-sum operator to obtain a fuzzy comprehensive evaluation vector [0.35, 0.32, 0.21, 0.08, 0.04], which is normalized to [0.38, 0.35, 0.23, 0.09, 0.05] after normalization processing. A score vector [90, 80, 70, 60, 50] is constructed, the total score of LNG power evaluation is 82.3 points, the total score of methanol power evaluation is 78.5 points, the total score of hydrogen fuel cell power evaluation is 65.2 points, and the total score of pure lithium battery power evaluation is 63.8 points; the optimal score is 82.3 points, the worst score is 63.8 points, the relative closeness of LNG power is 0.92, the relative closeness of methanol power is 0.75, the relative closeness of hydrogen fuel cell power is 0.31, and the relative closeness of pure lithium battery power is 0.28; the preset threshold is 0.6, and finally, LNG power is determined as the target power energy type.

[0115] The technical scheme of the embodiment determines the evaluation index set and the selected power energy type, converts the qualitative evaluation information of each index into a fuzzy relationship matrix based on the fuzzy mathematics theory, calculates the weight value of each index by using the entropy weight method, performs fuzzy synthesis operation on the weight value and the fuzzy relationship matrix, and normalizes the result. Finally, the optimal solution is determined by solving the total score of evaluation and calculating the relative closeness of each selected type, and the target power energy type is determined. The quantification, comprehensiveness, and precision of the selection of the power energy type of the inland river ship are realized, and the reliability of the selection result is improved.

[0116] According to the embodiments of the present application, a power energy selection device for an inland river ship corresponding to the power energy selection method of the inland river ship is also provided. Referring to Figure 2Structure diagram of an embodiment of the device of the application. The power energy selection device of the inland river ship can comprise: a determination unit 101, a conversion unit 102, a calculation unit 103, and a selection unit 104.

[0117] The determination unit 101 is configured to determine a set of evaluation indexes for the selection of the power energy of the inland river ship and to specify the types of the alternative power energy to be selected. For the specific functions and processing of the unit, see step S110.

[0118] The set of evaluation indexes is a combination of indexes constructed for accurately judging the adaptability of the power energy of the inland river ship, including endogenous power indexes and external support indexes. The endogenous power indexes are comprehensive indexes for measuring the inherent competitiveness of the power energy of the ship in terms of cost, efficiency, performance, etc. from the technical and operational characteristics of the ship, specifically including economic indexes. The external support indexes are comprehensive indexes for measuring the external resource and condition support obtained by the power energy of the ship in terms of facility support, policy support, and external environment adaptation, specifically including security indexes and policy indexes.

[0119] For the endogenous power indexes, the shipbuilding and operation are market-oriented, and profits need to be obtained by controlling costs. The construction (initial investment), operation (fuel / supplies), and maintenance (equipment update) are the three most critical stages that affect the cost, so the endogenous power indexes include construction cost, operation cost, and maintenance cost, which directly determine the economic feasibility of technology selection.

[0120] For the external support indexes, whether new energy ships can be put into operation depends on the support of construction hardware (such as specialized shipyards), energy supplement networks (such as LNG filling stations and charging piles), and maintenance networks. If these facilities are lacking, even if the technology is economically feasible, it cannot be practically applied. Therefore, the external support indexes include the perfection of construction facilities, the perfection of energy supplement facilities, and the perfection of maintenance facilities. Moreover, the initial investment of new energy ships is high (such as the cost of hydrogen fuel cell ships being significantly higher than that of traditional ships), subsidies can alleviate the cost pressure, and preferential navigation can improve operational efficiency. These two types of policies directly affect the market competitiveness of the technology and are key factors that cannot be ignored in selection, so the external support indexes also include subsidy support intensity and preferential navigation support intensity. The final set of evaluation indexes includes two first-level indexes (endogenous power indexes and external support indexes) and eight second-level indexes (construction cost, operation cost, maintenance cost, perfection of construction facilities, perfection of energy supplement facilities, perfection of maintenance facilities, subsidy support intensity, and preferential navigation support intensity).

[0121] The alternative power energy types include clean energy such as liquefied natural gas (LNG) power, methanol power, hydrogen fuel cell power, and pure lithium battery power. These clean energy technologies have relatively mature industrial chains and can avoid the problem of missing industrial chain support due to overly advanced technology.

[0122] Specifically, in the index screening, the core influencing factors of the application of power energy of inland river vessels are combed, and a dual framework of endogenous power index and external support index is constructed. The endogenous power index screens the cost, efficiency and other self-characteristics indexes; the external support index screens the external condition indexes such as facility supporting and policy support, forming a non-redundant and non-missing evaluation index set. In determining the types of alternative power energy, the clean energy power technologies that have appeared in the current inland river vessel field are collected, and based on the standards of technology maturity, pilot verification, and inland environment adaptability, the types that have not been piloted, unstable technology or missing industrial chain supporting are excluded, and the types of alternative power energy that have practical application feasibility are determined, such as finally selecting LNG, methanol, hydrogen fuel cell, and pure lithium battery power as alternative types.

[0123] The conversion unit 102 is configured to obtain qualitative evaluation information of each evaluation index in the evaluation index set, and convert the qualitative evaluation information into a fuzzy relation matrix based on fuzzy mathematics theory; the fuzzy relation matrix is used to represent the membership degree of each evaluation index at different evaluation levels. For specific functions and processing of this unit, see step S120.

[0124] The qualitative evaluation information is obtained through industry expert evaluation, field research, data statistics and other methods, and is a non-quantitative description of the importance or performance of each evaluation index (such as “important”, “general”, “perfect”, “insufficient”, etc.), which is used to reflect the characteristics of the index that is difficult to quantify directly. Fuzzy mathematics theory is a theoretical method that relies on fuzzy mathematics tools to convert elements with fuzzy boundaries and difficult to quantify into quantitative indexes, and the core is to realize the quantitative processing of qualitative information through membership function and fuzzy relation matrix. The fuzzy relation matrix is a structured matrix constructed based on fuzzy mathematics theory, which is used to represent the membership degree of each evaluation index at different evaluation levels (i.e. the degree to which a certain index belongs to a certain evaluation level), and is a key carrier connecting qualitative evaluation and quantitative calculation.

[0125] Some evaluation indexes (such as facility perfection and policy support) are difficult to quantify directly, and need to reflect the actual situation through qualitative evaluation; fuzzy mathematics theory can solve the problem that qualitative information cannot be directly involved in calculation, and convert fuzzy description into structured quantitative data, laying a foundation for subsequent calculation.

[0126] Specifically, by inviting experts in the fields of ship engineering, energy application, shipping operation, etc., and combining with field research data, the importance or performance of each evaluation index is qualitatively evaluated to form qualitative evaluation information such as "important", "better", "general", and "insufficient". The factor domain (set of all evaluation indexes) and the evaluation level domain (such as "good, better, general, worse, and poor") of the evaluation indexes are determined, the membership degrees of the evaluation indexes under different evaluation levels are solved, and a fuzzy relationship matrix is constructed based on the membership degrees, and the matrix elements accurately represent the degree of belonging of each index under the corresponding evaluation level.

[0127] In some embodiments, when the conversion unit 102 converts the qualitative evaluation information into a fuzzy relationship matrix based on fuzzy mathematics theory, it also synchronously associates the factor domain of the set of evaluation indexes with the level domain, so that the construction dimension of the fuzzy relationship matrix corresponds to the hierarchy of the evaluation indexes and the evaluation levels one by one.

[0128] The factor domain of the set of evaluation indexes refers to the set composed of all indexes participating in the evaluation of power energy selection, which is a clear definition of the evaluation dimension, and specifically corresponds to all subordinate specific indexes of endogenous power indexes and external support indexes. The level domain refers to the set composed of all level terms used to evaluate the performance of the indexes (such as "good, better, general, worse, and poor"), which is a standardized division of the performance degree of the indexes.

[0129] The factor domain clearly defines the specific index range of the evaluation, and the level domain clearly defines the standard level of the evaluation. If the construction of the fuzzy relationship matrix is not associated with the two, it will cause confusion in the matrix dimension (such as mismatch between the number of indexes and the number of matrix rows, mismatch between the number of evaluation levels and the number of matrix columns), and cannot accurately represent the correspondence between the indexes and the evaluation levels. The evaluation indexes have a clear hierarchical structure, and if the fuzzy relationship matrix does not fit this hierarchy, the evaluation data of different level indexes will be confused, affecting the accuracy of subsequent weight calculation and comprehensive evaluation. Simultaneous association of the factor domain, the level domain, and the index hierarchy can ensure that the construction logic of the fuzzy relationship matrix is completely consistent with the evaluation system, so that the matrix can accurately carry the membership degree information of each level index under different evaluation levels.

[0130] Specifically, first, all the specific indicators participating in the selection evaluation are combed to form a clear factor domain (such as {F1: construction cost, F2: operation cost, F3: construction facility perfection, F4: complementary facility perfection, F5: subsidy support intensity, F6: priority navigation support intensity}); then, according to the selection accuracy requirement, a standardized grade domain (such as {G1: good, G2: better, G3: general, G4: worse, G5: poor}) is determined to ensure that the domain range is not redundant and not missed. Then, the core level to which each indicator belongs is determined to form a hierarchical relationship between the core level and the specific indicators (such as "endogenous power indicators-F1, F2; external support indicators-F3, F4, F5, F6"), ensuring that the hierarchical division is clear and there is no cross-level indicator confusion. Then, according to the number of indicators in the factor domain (such as 6) and the number of evaluation grades in the grade domain (such as 5), the dimension of the fuzzy relationship matrix is determined to be 6 rows and 5 columns (the number of rows corresponds to the number of indicators, and the number of columns corresponds to the number of evaluation grades); each row of the matrix corresponds to a specific indicator in the factor domain, and the core level to which the row belongs is consistent with the level of the indicator itself (such as the rows corresponding to F1, F2 belong to the "endogenous power indicator" related row group, and the rows corresponding to F3-F6 belong to the "external support indicator" related row group); each column of the matrix corresponds to an evaluation grade in the grade domain, and the matrix elements are the membership degrees of the corresponding indicators under the evaluation grade, and finally a fuzzy relationship matrix is formed which is completely matched with the structure and domain, and the level. Finally, check whether the number of rows of the fuzzy relationship matrix is consistent with the number of indicators in the factor domain, whether the number of columns is consistent with the number of evaluation grades in the grade domain, whether the level of each indicator row is accurate, and ensure that there is no dimension error or level confusion in the matrix construction.

[0131] The calculation unit 103 is configured to perform hierarchical calculation and processing on the set of evaluation indicators based on the multi-index decision logic using the entropy weight method to obtain the weight values of each evaluation indicator in the selection decision; the weight values are used to reflect the selection priority of each evaluation indicator in different dimension classifications in the set of evaluation indicators. For specific functions and processing of this unit, see step S130.

[0132] The multi-index decision logic is a logic system that comprehensively considers multiple interrelated or conflicting evaluation indicators, integrates the influence of each indicator through a specific calculation method, and realizes optimal decision, which is used to coordinate the effect of different dimension indicators on the selection result. The entropy weight method is a method for objectively calculating the weight based on the discrete degree of indicator data, which determines the importance of the indicator by measuring the entropy value, and the smaller the entropy value (the higher the discrete degree of the indicator), the greater the weight, which can avoid the deviation caused by subjective judgment. The weight value can represent the relative importance of each evaluation indicator in the selection decision, and the higher the weight value, the greater the influence of the indicator on the final selection result.

[0133] The influence degree of different evaluation indexes on the selection result is different, and the weight value can quantify the difference; the entropy weight method objectively assigns weights based on the dispersion degree of index data, avoids the deviation caused by expert subjective judgment, and ensures the rationality and fairness of weight distribution; the multi-index decision logic can coordinate the correlation or conflict relationship between indexes, and realize the comprehensive optimization.

[0134] Specifically, first, the evaluation indexes are divided into positive feedback type indexes (the higher the index value, the better, such as facility perfection) and negative feedback type indexes (the lower the index value, the better, such as cost). The corresponding standardization formula is used for standardization transformation for the two types of indexes respectively, and the dimensional difference is eliminated; the normalized value after standardization is normalized to obtain an intermediate form; the entropy value of each index is calculated based on the intermediate form, the deviation coefficient (deviation coefficient = 1-entropy value) is solved by entropy value, and finally the deviation coefficient is normalized to obtain the weight value of each evaluation index.

[0135] In some embodiments, the computing unit 103, based on the multi-index decision logic, uses the entropy weight method to perform hierarchical calculation and processing on the evaluation index set, including:

[0136] The computing unit 103 is specifically further configured to divide each evaluation index into a positive feedback type index and a negative feedback type index. For specific functions and processing of this unit, see step S210.

[0137] The evaluation index set contains two types of indexes with opposite action directions. If not classified and directly converted by the same formula, the index advantage and disadvantage trend will not be consistent with the numerical value (for example, the cost type index value is large, which represents the disadvantage, and if the same formula is used as the positive index, it will be misjudged as an advantage). After classification, the appropriate standardization formula can be used to ensure that the larger the value of all indexes, the better.

[0138] Specifically, the characteristics and action directions of each index in the evaluation index set are analyzed one by one to determine the correlation between the index value and the adaptability of the power energy. The index with the characteristic of “the larger the value, the better the adaptability of the power energy” is classified as a positive feedback type index, such as construction facility perfection, energy supplement facility perfection, subsidy support intensity, and priority navigation support intensity. The index with the characteristic of “the smaller the value, the better the adaptability of the power energy” is classified as a negative feedback type index, such as construction cost and operation cost.

[0139] The computing unit 103 is specifically further configured to use the corresponding standardization formula for standardization transformation for the positive feedback type index and the negative feedback type index. For specific functions and processing of this unit, see step S220.

[0140] The raw data for different evaluation indicators have different dimensions (e.g., construction cost is measured in "ten thousand yuan", facility completeness in "grades", and subsidy support in "percentages"). Directly using the raw data for calculations would result in incomparable numerical values ​​due to these different dimensions, thus affecting the objectivity of weight calculations. Standardization can map all indicator data to the [0,1] interval, eliminating dimensional differences and making data from different types of indicators horizontally comparable. Standardization formulas are used to transform raw indicator data with different dimensions and numerical ranges into dimensionless data within a unified range (usually [0,1]), eliminating the impact of dimensional differences on the calculation results.

[0141] Specifically, for positive feedback indicators, a standardized formula is used. ( Let j be the original value of the i-th alternative power energy type under the j-th index. The minimum value among all candidate types under the j-th indicator. (where the maximum value of all candidate types under the j-th indicator is used), the original data is converted into dimensionless data in the interval [0,1]; for negative feedback indicators, a standardized formula is used. Similarly, the original data is converted into dimensionless data in the range [0,1] to ensure that both types of indicators achieve the unified logic that "the larger the value, the better the adaptability".

[0142] The calculation unit 103 is further configured to normalize the standardized and deformed index values ​​to obtain an intermediate formula. See step S230 for the specific functions and processing of this unit.

[0143] Although the standardized indicator data has eliminated dimensional differences, the data distribution still does not meet the requirements for entropy calculation (entropy calculation needs to be based on the logic of "data proportion"). Normalization can convert the standardized data of each indicator into the proportion of all alternative data types under that indicator, so that the data presents the characteristics of "relative contribution", providing reasonable input for accurately calculating the entropy value (measuring the degree of data dispersion).

[0144] Specifically, for each evaluation indicator, standardized and modified data of all candidate power energy types under that indicator are collected, and a normalization formula is used. ( Given the standardized data of the i-th candidate type under the j-th indicator (the denominator is the sum of the standardized data of all candidate types under all indicators), we transform each standardized data point to obtain the corresponding intermediate formula. This ensures that the sum of the intermediate expressions for all candidate types under each indicator and the sum of the intermediate expressions for all indicators satisfy the logical requirements for entropy calculation.

[0145] The computing unit 103 is further configured to calculate the entropy value of each evaluation index based on the intermediate formula, solve the deviation coefficient based on the entropy value, and finally obtain the weight value of each evaluation index based on the deviation coefficient. For specific functions and processing of this unit, see step S240.

[0146] The entropy value is a core index for measuring the degree of data dispersion. The higher the degree of data dispersion, the more effectively the index can distinguish the adaptability differences of different alternative power energy types, and the greater the reference value for selection decision-making. The deviation coefficient is calculated by "1-entropy value", which can directly quantify the importance potential of the index.

[0147] Specifically, based on the intermediate formula , the entropy value calculation formula is used, where k is the number of alternative power energy types, n is the number of evaluation indexes, if , then is processed as 0), the entropy value of each evaluation index is calculated one by one ; then the entropy value of each index is converted into the corresponding deviation coefficient by using the deviation coefficient formula . The larger the deviation coefficient, the stronger the importance potential of the index.

[0148] The deviation coefficient can only reflect the relative size of the importance potential of the index, and does not form a unified weight distribution system (the sum of the deviation coefficients of all indexes is not necessarily 1). By normalizing the deviation coefficient, it can be converted into a weight value, ensuring that the sum of the weight values of all indexes is 1, forming an intuitive and comparable importance ratio.

[0149] Specifically, the deviation coefficients of all evaluation indexes are collected , the sum of all deviation coefficients is calculated (n is the number of evaluation indexes), the weight value calculation formula is used, and the weight value of each evaluation index is calculated one by one . The final weight value set must satisfy the sum of all weight values being 1, ensuring the rationality and logic of weight distribution.

[0150] The computing unit 103 is further configured to determine a fuzzy weight vector based on the weight value, perform fuzzy synthesis operation on the fuzzy weight vector and the fuzzy relation matrix to obtain a fuzzy comprehensive evaluation vector, and normalize the fuzzy comprehensive evaluation vector. For specific functions and processing of this unit, see step S140.

[0151] The fuzzy weight vector is a vector constructed based on the weight values ​​of each evaluation indicator, used to reflect the importance proportion of different indicators in the fuzzy comprehensive evaluation. Fuzzy synthesis is the process of operating the fuzzy weight vector and the fuzzy relation matrix using specific fuzzy operators to integrate the evaluation results of each indicator, obtaining the comprehensive evaluation vector. The fuzzy comprehensive evaluation vector can characterize the comprehensive performance of all evaluation indicators under different evaluation levels, reflecting the overall adaptability trend of candidate power energy types. Normalization converts the fuzzy comprehensive evaluation vector into values ​​of a unified dimension and range, eliminating dimensional differences caused by different indicators or evaluation levels, and ensuring the fairness and accuracy of subsequent calculations.

[0152] Fuzzy weight vectors can reflect the importance proportion of each indicator, ensuring that important indicators play a leading role in the comprehensive evaluation; fuzzy synthesis operations can integrate the evaluation results of each indicator to form an overall evaluation trend; and standardization processing can eliminate numerical differences caused by different evaluation levels or indicators, making the evaluation results comparable.

[0153] Specifically, the weight values ​​of each evaluation indicator are arranged in corresponding order to construct a fuzzy weight vector, ensuring that the vector dimension is consistent with the number of evaluation indicators. Fuzzy operators (such as the maximum-minimum operator, product-addition operator, etc.) are used to synthesize the fuzzy weight vector and the fuzzy relation matrix to obtain a fuzzy comprehensive evaluation vector, which reflects the comprehensive membership degree of all indicators under each evaluation level. A preset normalization formula (such as linear normalization, normalization to the [0,1] interval, etc.) is used to convert the fuzzy comprehensive evaluation vector into values ​​with a unified dimension and a unified range, completing the normalization process.

[0154] In some implementations, when the calculation unit 103 performs normalization processing on the fuzzy comprehensive evaluation vector, it uses a preset normalization formula to convert the fuzzy comprehensive evaluation vector into a value of a unified dimension.

[0155] The fuzzy comprehensive evaluation vector is the raw result obtained through fuzzy synthesis operations. Its numerical value may vary depending on the number of evaluation levels and the distribution of membership degrees. Directly using it for subsequent calculations will lead to distortion in comparison due to inconsistent numerical scales. Furthermore, the fuzzy comprehensive evaluation vectors corresponding to different candidate power energy types lack a unified benchmark, making it difficult to objectively judge their suitability without standardization. Standardization can map all vector values ​​to a unified interval, eliminating differences in dimensions and scales, and making the comprehensive evaluation results of different candidate types comparable.

[0156] Specifically, the original data characteristics of the fuzzy comprehensive evaluation vector are determined, including vector dimension (consistent with the number of evaluation levels), numerical range, membership degree distribution law, etc., to ensure that the normalization formula is adapted to the data characteristics; a preset normalization formula is selected, common formulas include linear normalization formula, interval normalization formula, etc., and the specific formula selection needs to be combined with the data resolution requirements of the selection decision; each value in the fuzzy comprehensive evaluation vector is substituted into the selected normalization formula, and each value is calculated to obtain the corresponding standardized value; the normalized values are checked to ensure that all values are in a preset unified interval (such as [0, 1]), and the value distribution can truly reflect the membership degree trend of the original vector.

[0157] The selection unit 104 is configured to solve a corresponding evaluation total score based on the fuzzy comprehensive evaluation vector, calculate the relative closeness of each of the candidate power energy types to the optimal solution in combination with the evaluation total score, and determine an adapted target power energy type from the candidate power energy types based on the relative closeness. The specific functions and processing of this unit are described in step S150.

[0158] The evaluation total score is a quantitative value obtained by vector operation based on the fuzzy comprehensive evaluation vector and a preset score vector, which directly reflects the comprehensive adaptation level of the candidate power energy type. The relative closeness is an index that quantifies the closeness to the theoretical optimal solution by calculating the distance between the candidate power energy type and the optimal score and the worst score, and the larger the value, the better the adaptability.

[0159] The evaluation total score converts the fuzzy comprehensive evaluation vector into a direct quantitative value, which is convenient for horizontal comparison of each candidate type; the relative closeness quantifies the adaptability by the distance from the theoretical optimal solution and the worst solution, avoiding the problem that a single value is difficult to judge the adaptation degree; the selection based on the relative closeness can ensure the objectivity and accuracy of the result, avoiding subjective decision bias.

[0160] Specifically, a score vector corresponding to the evaluation level domain is constructed (such as "good = 90 points, better = 80 points, general = 70 points, worse = 60 points, poor = 50 points"), and the fuzzy comprehensive evaluation vector is converted into the evaluation total score of each candidate power energy type through vector operation. The evaluation total scores of all candidate types are counted to determine the optimal score (the maximum value among all scores) and the worst score (the minimum value among all scores); the first distance between each candidate type and the optimal score and the second distance between each candidate type and the worst score are calculated, and the relative closeness of each candidate type is calculated through the formula (relative closeness = second distance / (first distance + second distance)). The relative closeness of each candidate type is sorted in descending order, and the type with the highest ranking and a relative closeness greater than a preset threshold (set according to actual operation requirements or industry standards) is selected as the adapted target power energy type.

[0161] In some embodiments, the selection unit 104, in combination with the evaluation total score, calculates the relative closeness of each of the candidate power energy types to the optimal solution, including:

[0162] The selection unit 104 is specifically further configured to construct a score vector, and the evaluation total score of each of the candidate power energy types is obtained through vector operation of the score vector and the fuzzy comprehensive evaluation vector. The specific functions and processes of this unit are described in step S310.

[0163] The score vector refers to a vector containing corresponding quantitative scores preset according to the evaluation level domain, and the dimension is consistent with the fuzzy comprehensive evaluation vector, which is used to convert the fuzzy membership distribution into an intuitive total score. The fuzzy comprehensive evaluation vector can only reflect the membership distribution of each evaluation level, and cannot directly reflect the comprehensive adaptation level of the candidate type; the score vector can give each evaluation level a clear quantitative score, and through vector operation, the fuzzy distribution is converted into a single total score, realizing the horizontal comparison of different candidate types.

[0164] Specifically, according to the setting of the evaluation level domain, the score vector is constructed in combination with the accuracy requirement of the selection decision and the industry conventional scoring standard. For example, if the evaluation level is 5 levels, the score vector can be set as [90, 80, 70, 60, 50] (the higher the score, the better the evaluation level), ensuring that the dimension of the score vector is completely consistent with the fuzzy comprehensive evaluation vector. For the fuzzy comprehensive evaluation vector corresponding to each candidate power energy type, the vector dot product operation rule is used for calculation. Assuming that the fuzzy comprehensive evaluation vector of a certain candidate type is [S1, S2, S3, S4, S5], and the score vector is [F1, F2, F3, F4, F5], the evaluation total score Z=S1xF1+S2xF2+S3xF3+S4xF4+S5xF5, and the membership distribution is converted into a single quantitative score through the operation. Repeat the above operation to obtain the evaluation total score of all candidate power energy types one by one, and form a comparable total score set.

[0165] The selection unit 104 is specifically further configured to calculate the first distance of each of the candidate power energy types from the optimal score and the second distance from the worst score based on the evaluation total score. The specific functions and processes of this unit are described in step S320.

[0166] The optimal score refers to the maximum value of the evaluation total score of all candidate power energy types, reflecting the highest level of power energy adaptability in the current selection range. The worst score refers to the minimum value of the evaluation total score of all candidate power energy types, which is a quantitative embodiment of the lowest level of adaptability in the selection range. The first distance and the second distance can quantify the relative position of the candidate type from the optimal and the worst.

[0167] Specifically, the maximum value of the total score of the evaluation of all alternative power energy types is selected as the optimal score Z max , and the minimum value of the total score of the evaluation of all alternative power energy types is selected as the worst score Z min ; the first distance D1 i =|Z i -Z max | is calculated by using an absolute value distance formula for the total score of the evaluation of each alternative power energy type Z i , which directly reflects the gap between the alternative type and the optimal adaptation state, and the smaller the value, the closer to the optimal; similarly, the second distance D2 i =|Z i -Z min | is calculated by using an absolute value distance formula for the total score of the evaluation of each alternative type Z i , which reflects the gap between the alternative type and the worst adaptation state, and the larger the value, the better the adaptation than the worst level.

[0168] The selection unit 104 is specifically configured to calculate the relative closeness based on the first distance and the second distance. The specific functions and processes of this unit are described in step S330.

[0169] Specifically, for each alternative power energy type, the relative closeness calculation formula C i =D2 i / (D1 i +D2 i ) is used to calculate the relative closeness of the i-th alternative type C i . After the calculation is completed, the calculation result is checked to ensure that the relative closeness value is in the interval [0, 1]: if the total score of the evaluation of a certain alternative type is equal to the optimal score (Z i =Z max ), then D1 i =0, C i =1, which means that the type completely meets the optimal adaptation state; if the total score of the evaluation of a certain alternative type is equal to the worst score (Z i =Z min ), then D2 i =0, C i =0, which means that the type is in the worst adaptation state; in other cases, the relative closeness value is between 0 and 1, and the larger the value, the better the adaptation. Finally, the relative closeness of all alternative power energy types is calculated to form a complete set of relative closeness.

[0170] In some embodiments, the selecting unit 104 determines the adapted target power energy type from the candidate power energy types based on the relative proximities, including: sorting the relative proximities corresponding to each of the candidate power energy types in descending order of values; and taking the candidate power energy type with the highest sorting priority and a relative proximity greater than a preset threshold as the adapted target power energy type.

[0171] The preset threshold is used to determine whether the adaptability of the candidate power energy type meets a minimum requirement. Through the double rules of sorting priority and threshold screening, it is ensured that the selected type is relatively optimal and its adaptability meets the actual demand, avoiding the selection of unqualified types due to single sorting.

[0172] Specifically, the relative proximity values corresponding to all candidate power energy types are collected to form a complete relative proximity dataset (such as LNG power 0.92, methanol power 0.75, hydrogen fuel cell power 0.31, and pure lithium battery power 0.28). According to the rule that "the larger the value, the higher the priority", the dataset is sorted in descending order to obtain a clear adaptability sorting result (such as LNG power > methanol power > hydrogen fuel cell power > pure lithium battery power), which ensures that the sorting result can directly reflect the relative adaptability level of each candidate type. The candidate power energy type with the highest sorting priority after descending sorting is determined, and it is determined whether the relative proximity value of the type is greater than the preset threshold. If the relative proximity of the type is greater than the preset threshold (such as the relative proximity of the first LNG power 0.92 > 0.6), the type is directly determined as the adapted target power energy type. If the relative proximity of the type is less than or equal to the preset threshold (such as the relative proximity of the first type 0.55 ≤ 0.6), it is rechecked whether there is deviation in the previous evaluation process (such as index selection, weight calculation, and relative proximity calculation), or the preset threshold is adjusted, the selection type process is re-performed after supplementing the candidate type, until the target type meeting the requirements is screened out.

[0173] The scheme realizes the comprehensiveness, quantification and precision of the selection of power energy types for inland ships. It solves the problems of qualitative evaluation in traditional selection, subjective weight allocation, and effectively reduces the risk of technical route uncertainty, providing reliable support for investment, production and application decisions of related parties in the inland ship industry chain.

[0174] Since the processing and functions realized by the device of the present embodiment are basically corresponding to the embodiments, principles and examples of the foregoing method, details not described in the description of the present embodiment can be referred to the relevant description in the foregoing embodiments, which will not be repeated here.

[0175] The technical scheme of the present application determines the evaluation index set and the alternative power energy type, converts qualitative evaluation information of each index into a fuzzy relation matrix based on fuzzy mathematics theory, calculates the weight value of each index by using an entropy weight method, performs fuzzy synthesis operation in combination with the weight value and the fuzzy relation matrix, and normalizes the result, finally determines the adapted target power energy type by solving the total evaluation score and calculating the relative closeness of each alternative type to the optimal solution, thereby realizing quantification, comprehensiveness and precision of the power energy type selection of the inland ship, and improving the reliability of the selection result.

[0176] According to the embodiment of the present application, a power energy type selection system of an inland ship corresponding to the power energy type selection device of the inland ship is also provided. The system can include the power energy type selection device of the inland ship described above.

[0177] Figure 3 The figure is a structural schematic diagram of the power energy type selection system of the inland ship, the expert grader is used for expert input scoring, the storage is used for storing scoring and evaluation calculation data, the control circuit is used for controlling the fuzzy and vector processor, the fuzzy and vector processor is used for model calculation, the display is used for displaying the evaluation calculation result, and the transmission interface is used for outputting the evaluation result data.

[0178] Since the processing and functions realized by the system of the present embodiment are basically corresponding to the foregoing embodiments, principles and examples of the device, the description of the present embodiment will not be elaborated on the related descriptions in the foregoing embodiments, which will not be repeated here.

[0179] The technical scheme of the present application determines the evaluation index set and the alternative power energy type, converts qualitative evaluation information of each index into a fuzzy relation matrix based on fuzzy mathematics theory, calculates the weight value of each index by using an entropy weight method, performs fuzzy synthesis operation in combination with the weight value and the fuzzy relation matrix, and normalizes the result, finally determines the adapted target power energy type by solving the total evaluation score and calculating the relative closeness of each alternative type to the optimal solution, thereby realizing quantification, comprehensiveness and precision of the power energy type selection of the inland ship, and improving the reliability of the selection result.

[0180] According to the embodiment of the present application, a storage medium corresponding to the power energy type selection method of the inland ship is also provided. The storage medium includes a stored program, wherein when the program runs, the device where the storage medium is located performs the power energy type selection method of the inland ship described above.

[0181] Since the processing and functions realized by the storage medium of the present embodiment are basically corresponding to the foregoing embodiments, principles and examples of the method, the description of the present embodiment will not be elaborated on the related descriptions in the foregoing embodiments, which will not be repeated here.

[0182] The technical scheme of the present application determines the evaluation index set and the alternative power energy type, converts qualitative evaluation information of each index into a fuzzy relation matrix based on fuzzy mathematics theory, calculates the weight value of each index by using an entropy weight method, performs fuzzy synthesis operation in combination with the weight value and the fuzzy relation matrix, and normalizes the result, finally determines the adapted target power energy type by solving the total evaluation score and calculating the relative closeness of each alternative type to the optimal solution.

[0183] According to the embodiments of the present application, a computer program product corresponding to the power energy type selection method of the inland ship is also provided, and the computer program product comprises a computer program, and the computer program product is processed to implement the steps of the above-mentioned power energy type selection method of the inland ship.

[0184] Since the processing and functions implemented by the computer program product of the present embodiment are basically corresponding to the embodiments, principles and examples of the above-mentioned method, the unexplained parts in the description of the present embodiment can be referred to the related description in the above-mentioned embodiments, which will not be repeated here.

[0185] The technical scheme of the present application determines the evaluation index set and the alternative power energy type, converts qualitative evaluation information of each index into a fuzzy relation matrix based on fuzzy mathematics theory, calculates the weight value of each index by using an entropy weight method, performs fuzzy synthesis operation in combination with the weight value and the fuzzy relation matrix, and normalizes the result, finally determines the adapted target power energy type by solving the total evaluation score and calculating the relative closeness of each alternative type to the optimal solution.

[0186] In summary, those skilled in the art can easily understand that the above-mentioned advantageous modes can be freely combined and superimposed without conflict.

[0187] The above only describes the embodiments of the present application and is not used to limit the present application, and those skilled in the art can make various modifications and changes to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the scope of claims of the present application.

Claims

1. A method for selecting the power source for inland waterway vessels, characterized in that, The method includes: Determine the set of evaluation indicators for the selection of power sources for inland waterway vessels, and identify the alternative power source types to be selected; Qualitative evaluation information for each evaluation indicator in the set of evaluation indicators is obtained, and the qualitative evaluation information is transformed into a fuzzy relation matrix based on fuzzy mathematics theory; the fuzzy relation matrix is ​​used to characterize the membership degree of each evaluation indicator under different evaluation levels; Based on multi-indicator decision-making logic, the entropy weight method is used to perform hierarchical calculation processing on the evaluation index set to obtain the weight value of each evaluation index in the selection decision. The weight value is used to reflect the selection priority of each evaluation index under different dimension classifications in the evaluation index set. The hierarchical calculation processing of the evaluation index set using the entropy weight method includes: dividing each evaluation index into positive feedback indicators and negative feedback indicators; standardizing and transforming the positive feedback indicators and negative feedback indicators respectively using corresponding standardization formulas; normalizing the standardized index values ​​to obtain intermediate formulas; calculating the entropy value of each evaluation index based on the intermediate formulas; then solving the deviation coefficient through the entropy value; and finally obtaining the weight value of each evaluation index based on the deviation coefficient. Based on the weight values, a fuzzy weight vector is determined. The fuzzy weight vector and the fuzzy relation matrix are then combined to perform a fuzzy synthesis operation to obtain a fuzzy comprehensive evaluation vector. The fuzzy comprehensive evaluation vector is then normalized. The corresponding total evaluation score is obtained based on the fuzzy comprehensive evaluation vector. The relative proximity between each of the candidate power energy types and the optimal solution is calculated in combination with the total evaluation score. Based on the relative proximity, the suitable target power energy type is determined from the candidate power energy types. The qualitative evaluation information is used only to obtain fuzzy descriptions of each evaluation indicator to transform them into fuzzy relation matrices. The entropy weight method calculates weight values ​​based on the dispersion of indicator data and does not rely on experts' subjective judgment of the importance of indicators.

2. The method for selecting the power source for inland waterway vessels according to claim 1, characterized in that, When normalizing the fuzzy comprehensive evaluation vector, a preset normalization formula is used to convert the fuzzy comprehensive evaluation vector into a value with a unified dimension.

3. The method for selecting the power source for inland waterway vessels according to claim 1, characterized in that, The relative similarity between each of the candidate power energy types and the optimal solution is calculated based on the total evaluation score, including: Construct a score vector, and obtain the total evaluation score for each of the candidate power energy types through vector operation between the score vector and the fuzzy comprehensive evaluation vector; Based on the total evaluation score, calculate the first distance between the candidate power energy type and the best score, and the second distance between the candidate power energy type and the worst score, respectively. The relative proximity is calculated based on the first distance and the second distance.

4. The method for selecting the power source for inland waterway vessels according to claim 1 or 3, characterized in that, Determining the suitable target power energy type from the candidate power energy types based on the relative proximity includes: The relative proximity of each of the candidate power energy types is sorted in descending order of numerical value. The candidate power energy type that ranks highest and whose relative proximity is greater than a preset threshold is selected as the target power energy type for adaptation.

5. The method for selecting the power source for inland waterway vessels according to claim 1, characterized in that, When transforming the qualitative evaluation information into a fuzzy relation matrix based on fuzzy mathematics theory, the factor domain and the level domain of the evaluation index set are also simultaneously associated, so that the construction dimension of the fuzzy relation matrix corresponds one-to-one with the hierarchy and evaluation level of the evaluation index.

6. A power energy selection device for inland waterway vessels, characterized in that, include: The unit is configured to determine the set of evaluation indicators for selecting the power source of inland waterway vessels and to identify the alternative power source types to be selected. The transformation unit is configured to acquire qualitative evaluation information for each evaluation indicator in the set of evaluation indicators, and to transform the qualitative evaluation information into a fuzzy relation matrix based on fuzzy mathematics theory. The fuzzy relation matrix is ​​used to characterize the membership degree of each evaluation index under different evaluation levels; The calculation unit is configured to perform hierarchical calculation processing on the set of evaluation indicators based on multi-indicator decision-making logic, using the entropy weight method to obtain the weight value of each evaluation indicator in the selection decision; the weight value is used to reflect the selection priority of each evaluation indicator under different dimension classifications in the set of evaluation indicators; wherein, the hierarchical calculation processing of the set of evaluation indicators using the entropy weight method includes: dividing each evaluation indicator into positive feedback indicators and negative feedback indicators; standardizing and transforming the positive feedback indicators and negative feedback indicators respectively using corresponding standardization formulas; normalizing the standardized indicator values ​​to obtain intermediate formulas; calculating the entropy value of each evaluation indicator based on the intermediate formulas, then solving for the deviation coefficient through the entropy value, and finally obtaining the weight value of each evaluation indicator based on the deviation coefficient; The computing unit is further configured to determine a fuzzy weight vector based on the weight value, perform a fuzzy synthesis operation by combining the fuzzy weight vector with the fuzzy relation matrix to obtain a fuzzy comprehensive evaluation vector, and perform normalization processing on the fuzzy comprehensive evaluation vector. The selection unit is configured to solve the corresponding total evaluation score based on the fuzzy comprehensive evaluation vector, calculate the relative proximity between each of the candidate power energy types and the optimal solution based on the total evaluation score, and determine the suitable target power energy type from the candidate power energy types based on the relative proximity. The qualitative evaluation information is used only to obtain fuzzy descriptions of each evaluation indicator to transform them into fuzzy relation matrices. The entropy weight method calculates weight values ​​based on the dispersion of indicator data and does not rely on experts' subjective judgment of the importance of indicators.

7. A power energy selection system for inland waterway vessels, characterized in that, include: The power energy selection device for inland waterway vessels as described in claim 6.

8. A storage medium, characterized in that, The storage medium includes a stored program, wherein, when the program is executed, the device containing the storage medium is controlled to perform the power energy selection method for inland waterway vessels as described in any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the power energy selection method for inland waterway vessels as described in any one of claims 1 to 5.

Images