Water price reference index determination method and device based on multi-element mutual feedback optimization

Abstract

The invention relates to a water price reference index determination method and device based on multi-element mutual feedback optimization, and the method comprises the following steps: obtaining data in a research region, and constructing an index system affecting the water price; selecting different indexes for combination to obtain a plurality of first index combinations; based on the constructed first index combinations, adjusting to obtain a plurality of second index combinations; calculating a comprehensive index of each second index combination by using an entropy weight method; and calculating the relative deviation of the comprehensive index of each second index combination, evaluating and screening out one or more second index combinations with the minimum relative deviation, and determining the indexes as the water price reference indexes. According to the method, the problems of low efficiency, unstable analysis, dependence on a static model or subjective weight and the like when an analyzed reference index is determined in an existing water price analysis or adjustment method are solved, and the scientificity, the accuracy and the objectivity of data arrangement and analysis and a result obtained are remarkably improved.

Description

Technical Field

[0001] This invention relates to a method and apparatus for determining water price reference indicators based on multi-factor feedback optimization, belonging to the field of data processing technology. Background Technology

[0002] Water pricing, as a crucial tool for urban water resource management, directly impacts water resource allocation efficiency, supply-demand balance, and sustainable development. Existing water price analysis and adjustment methods require analysis based on relevant indicators to determine the adjustment cycle and magnitude. However, for more accurate and scientific water price determination and adjustment, analysis cannot rely solely on a single indicator. Instead, a comprehensive analysis of multi-dimensional indicators (such as GDP, population, industrial structure, and water use efficiency) is necessary. When dealing with massive amounts of data involving multi-dimensional indicators, manual analysis is extremely inefficient, and the selection of indicator data is subjective, lacking objectivity and supported by advanced data processing technologies.

[0003] The invention patent application with application number 202411038552.1 provides a method, device, system, and storage medium for determining a water supply price reference index based on a large language model. It achieves the processing, analysis, and result generation of massive amounts of collected data through cleaning and normalization, solving the problems of slow processing speed and high processing costs associated with traditional manual analysis methods, and also removing useless information mixed in with the data. However, the reference index determined by this method relies on public feedback, and the analysis results are biased towards public subjectivity, easily ignoring objective indicators such as cost and water resources. It still does not solve the problem of lack of objectivity in the determination of indicator data in existing analysis methods. Furthermore, this method for determining the water supply price reference index uses a large language model to construct a parse tree, and uses the parse tree volume and branch number weighting (preset N1+N2=1) to correct the weights. While AI-driven, it relies on model accuracy, which carries the risk of unstable parsing.

[0004] Therefore, it is necessary to seek a more objective, scientific, and accurate technical solution for determining water price reference indicators. Summary of the Invention

[0005] To address the shortcomings of the existing technologies, the present invention aims to propose a method and apparatus for determining water price reference indicators based on multi-factor feedback optimization, which comprehensively processes multi-dimensional indicators to determine water price reference indicators more efficiently, objectively, scientifically, and accurately.

[0006] This invention provides a first technical solution: a method for determining water price reference indicators based on multi-factor feedback optimization, comprising the following steps: S1. Obtain data within the study area and construct an indicator system affecting water prices; S2. Based on the constructed indicator system, different indicators are selected and combined to obtain multiple indicator combinations. The adjustment period and adjustment range are added to each indicator combination to obtain multiple first indicator combinations. S3. Based on the constructed first indicator combination, calculate the average annual compound growth rate of the corresponding indicator in each first indicator combination, and calculate the adjustment value of the corresponding indicator in the planning level year according to the following formula to obtain multiple adjusted first indicator combinations, and define the adjusted first indicator combination as the second indicator combination. , in, This is the adjustment value for the indicator. is the initial value of the indicator, CAGR is the average annual compound growth rate of the indicator, and a is the number of adjustment years; S4. Obtain the indicator data of the planning level year of the study area to form the third indicator combination; based on multiple second and third indicator combinations, use the entropy weight method to calculate the entropy value and weight of each indicator, and calculate the comprehensive index of each second indicator combination and the comprehensive index of the third indicator combination. S5. Based on the comprehensive index of the multiple second indicator combinations obtained, and using the comprehensive index of the third indicator combination as a comparison benchmark, calculate the relative deviation of the comprehensive index of each second indicator combination, evaluate and select the second indicator combination with the smallest relative deviation, and determine the indicator among them as the water price reference indicator.

[0007] Furthermore, in step S2, the adjustment range in each first indicator combination is selected from any one of 5%, 10%, 15%, and 20%; the adjustment period in each first indicator combination is selected from any one of two years, three years, and five years.

[0008] Furthermore, in step S3, the average annual compound growth rate of the corresponding indicator is calculated according to the following formula two. , in, These are the initial values ​​for the corresponding indicators. N represents the final value of the corresponding indicator, and N is the year difference between the year of the final value and the year of the initial value.

[0009] Furthermore, in step S3, the number of years of adjustment is an integer multiple of the adjustment period in the first indicator combination.

[0010] Further, in step S4, the entropy value and weight of each indicator are calculated using the entropy weight method, and the comprehensive index of each combination of second indicators and the comprehensive index of the combination of third indicators are calculated. The method includes: S4.1 If the indicator is a contrarian indicator, take the reciprocal of its adjustment value to obtain the corresponding processing value. And obtain the processed index combination. , n represents the number of indicators in a single indicator combination. ;in, The second set of indicators is the processed combination, and m is the number of combinations of the second indicator. This is the third combination of indicators after processing; S4.2 Constructing a matrix based on the processed index combination , among them , , If the corresponding indicator is a contrarian indicator, then its processed value is used. If the corresponding indicator is a positive indicator, then its adjusted value is used. ;right Min-Max normalization is performed to obtain the corresponding and matrix ; S4.3 Constructing the Proportion Matrix ,matrix The following formula (3) is used for calculation: ; S4.4 Calculate the entropy value of the index Entropy The following formula (four) is used for calculation: ; S4.5 Calculate the weight of the indicator Weight The following formula (five) is used for calculation: ; S4.6 Calculate the composite index of each indicator combination. The comprehensive index The following formula (6) is used for calculation: ; in, This is the composite index of the combination of each second indicator. This is the composite index of the third indicator combination.

[0011] Further, in step S5, the relative deviation of the comprehensive index of each second indicator combination is calculated, and the second indicator combination with the smallest relative deviation is evaluated and selected. The indicator in this combination is then determined as the water price reference indicator. The method includes: S5.1 Calculate the relative deviation of the composite index for each combination of second indicators. relative deviation The following formula (7) is used for calculation: , in, , The benchmark composite index is the composite index of the third indicator combination obtained in step S4.6. ; S5.2 The relative deviation is determined according to... , and The standards are divided into positive deviation group, negative deviation group and near-zero deviation group; S5.3 Construct a sensitivity matrix using positive deviation, negative deviation, and near-zero deviation groups; calculate the sensitivity of each input variable to the output variable; determine the trend of the output variable with respect to a single variable through regression analysis; quantify the relative importance of each input variable to the output variable through correlation analysis; identify the factors with the greatest impact on the results; and obtain the dynamic feedback effect among multiple factors. Here, the single variable is the adjustment period or adjustment magnitude, and the input variable is the one obtained in step S4.2. ,in The output variable is the composite index of each combination of second indicators or the relative deviation between the composite index of the combination of second indicators and the benchmark composite index. S5.4 Comparison Scheme: Construct a decision matrix and list the second indicator combination and features with the smallest relative deviation; S5.5 Determine the threshold: Based on the schemes selected in step S5.4, draw a three-dimensional graph of deviation-period-amplitude, identify the period and amplitude corresponding to the combination of scenarios with the minimum deviation, select the period that makes the relative deviation stable and close to zero in the long term as the period threshold, and select the minimum amplitude that can effectively reduce the relative deviation as the amplitude threshold.

[0012] The present invention also provides a second technical solution: a water price reference index determination system based on multi-factor feedback optimization, comprising: The data acquisition module is used to acquire data within the study area, construct an indicator system affecting water prices, acquire indicator data for the planning level year of the study area, and obtain the adjustment cycle and adjustment range. The grouping module is used to select different indicators based on the constructed indicator system to obtain multiple indicator combinations, and to add the adjustment period and adjustment range to each indicator combination to obtain and output multiple first indicator combinations; and to construct and output third indicator combinations based on the indicator data of the acquired planning level year. The first analysis module is used to obtain the average annual compound growth rate of the corresponding indicators in each first indicator combination, and to obtain the adjustment value of the corresponding indicator in the planning level year, thereby obtaining and outputting multiple second indicator combinations. The second analysis module is used to analyze and process multiple combinations of second indicators and third indicators based on the entropy weight method, and to obtain and output the comprehensive index of each combination of second indicators and the comprehensive index of the third indicator combination. The third analysis module is used to obtain the relative deviation of the comprehensive index of each second indicator combination based on the comprehensive index of the third indicator combination, evaluate and screen out one or more second indicator combinations with the smallest relative deviation, determine the indicators among them as water price reference indicators and output them.

[0013] The present invention also provides a third technical solution: a water price reference index determination device based on multi-factor mutual feedback optimization, comprising: Memory for storing non-transitory computer-readable instructions; and A processor is used to run non-transitory computer-readable instructions, such that when the non-transitory computer-readable instructions are executed by the processor, the above-mentioned method for determining water price reference indicators based on multi-factor mutual feedback optimization is implemented.

[0014] The present invention also provides a fourth technical solution: a computer-readable storage medium for storing non-transitory computer-readable instructions, which, when executed by a computer, cause the computer to execute the above-mentioned method for determining water price reference indicators based on multi-factor feedback optimization.

[0015] Due to the application of the above technical solution, the present invention has the following advantages compared with the prior art: This invention generates all index combinations, adjusts index values ​​based on the compound growth rate, and objectively determines weights using the entropy weight method. It comprehensively considers and efficiently processes the multi-dimensional factors affecting water prices, overcoming the problems of low efficiency, unstable analysis, reliance on static models or subjective weights in existing water price analysis or adjustment methods. It significantly improves the scientificity, accuracy, and objectivity of data processing, analysis, and the results obtained. Detailed Implementation

[0016] The technical solution of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0018] Example 1: The technical solution provided in this embodiment is: a method for determining water price reference indicators based on multi-factor feedback optimization, which includes the following steps: S1. Obtain data within the study area, identify the factors influencing water prices, and construct an indicator system affecting water prices. Specifically, the data within the study area includes the region's Gross Domestic Product (GDP), the ratio of secondary industry, the number of permanent residents, per capita disposable income, water consumption, water consumption per 10,000 yuan of GDP, per capita water consumption, and water prices.

[0019] S2. Based on the constructed indicator system, different indicators are selected and combined to obtain multiple indicator combinations. The adjustment period and adjustment magnitude are added to each indicator combination to obtain multiple first indicator combinations within the study area. Furthermore, the adjustment magnitude in each first indicator combination is selected from any one of 5%, 10%, 15%, and 20%; the adjustment period in each first indicator combination is selected from any one of two years, three years, and five years.

[0020] S3. Based on the constructed combinations of multiple first indicators, adjust the indicators within each combination under the corresponding adjustment period. Specifically, this includes: S3.1 Assuming that each indicator grows at a fixed average annual growth rate, based on the obtained historical data, preferably data from the past five years, calculate the average annual compound growth rate of the corresponding indicator in each first indicator combination, as follows: , Wherein, CAGR is the compound annual growth rate of this indicator. These are the initial values ​​for the corresponding indicators. is the final value of the corresponding indicator, and N is the year difference between the year of the final value and the year of the initial value.

[0021] S3.2 Calculate the adjustment value of the corresponding indicator for the planning level year according to the following formula: , in, This is the adjustment value for the indicator. is the initial value of the indicator, CAGR is the average annual compound growth rate of the indicator, and 'a' is the number of adjustment years. Specifically, the number of adjustment years is an integer multiple of the adjustment period in the first indicator combination in which the indicator belongs.

[0022] After calculating the adjusted values ​​of the indicators, multiple combinations of the first indicators are obtained after adjustment. These adjusted combinations of the first indicators are defined as the second indicator combinations.

[0023] S4. Obtain indicator data for the planning level year of the study area to construct a third indicator combination; based on multiple second and third indicator combinations, calculate the entropy value and weight of each indicator using the entropy weight method, and calculate the comprehensive index of each second indicator combination and the comprehensive index of the third indicator combination. Specifically, this includes: S4.1 If the indicator is a contrarian indicator, then adjust its value. Take the reciprocal to get the corresponding processing value. And obtain the processed index combination. , n represents the number of indicators in a single indicator combination. ,in, The second set of indicators is the processed combination, and m is the number of combinations of the second indicator. This is the third indicator combination after processing.

[0024] It should be further noted that in this field, indicators with larger values ​​and worse evaluation results are defined as inverse indicators, such as per capita water consumption and water consumption per 10,000 yuan of GDP. The larger the value, the worse the water resource utilization efficiency. Positive indicators are the remaining indicators in the indicator system other than inverse indicators, including GDP, the ratio of secondary industry, the number of permanent residents, per capita disposable income of residents, water consumption, water price, etc.

[0025] S4.2 Based on the index combination processed in step S4.1 Construct matrix X (m+1)×n， Among them ( , If the corresponding indicator is a contrarian indicator, then its processed value is used. If the corresponding indicator is a positive indicator, then its adjusted value is used. ;right Min-Max normalization is performed to obtain the corresponding and matrix ; , in, For indicator combination The minimum value in, For indicator combination The maximum value in, here It is processed according to the value selection principles of reverse and positive indicators when constructing the matrix.

[0026] S4.3 Matrix-based Construct a proportional matrix ,matrix The following formula (3) is used for calculation: ; in, , m is the number of second indicator combinations, and n is the number of indicators in a single second indicator combination.

[0027] S4.4 Calculate the entropy value of the index Entropy The following formula (four) is used for calculation: ; in, , m is the number of second indicator combinations, and n is the number of indicators in a single second indicator combination.

[0028] S4.5 Calculate the weight of the indicator Weight The following formula (five) is used for calculation: ; in, , where n is the number of indicators in a single second indicator combination.

[0029] S4.6 Calculate the composite index of each indicator combination. Composite Index The following formula (6) is used for calculation: , in, Let i be the composite index of the i-th indicator combination. , , where m is the number of second indicator combinations, and n is the number of indicators in a single second indicator combination. Further, when hour, This is the composite index of the combination of each second indicator. hour, This is the composite index of the third indicator combination.

[0030] S5. Based on the comprehensive index of multiple combinations of second indicators obtained, and using the comprehensive index of the third indicator combination as a benchmark, calculate the relative deviation of the comprehensive index of each combination of second indicators, evaluate and select the second indicator combination with the smallest relative deviation, and determine the indicators among them as water price reference indicators. Specifically, this includes: S5.1 Calculate the relative deviation of the composite index for each combination of second indicators. relative deviation The following formula (7) is used for calculation: , in, Let be the relative deviation of the composite index of the i-th combination of second indicators. m is the number of combinations of the second indicator. The benchmark composite index is the composite index of the third indicator combination obtained in step S4.6. .

[0031] S5.2 will determine the relative deviation They are divided into three categories according to the following criteria: positive deviation group, negative deviation group, and near-zero deviation group. , The value of ε is set according to the actual situation. The number and characteristics of the second indicator combination corresponding to each category are counted. The deviation distribution is displayed using box plots or scatter plots. Outliers are identified and the causes of outliers are analyzed. The value of ε is optimized and determined, and the final classification is determined.

[0032] The relative deviations in the near-zero deviation group from the final classification are filtered out, and their corresponding second indicator combinations are given priority consideration. The positive deviation group and the negative deviation group are also considered together for comparison, in order to analyze outliers and reveal unsuitable parameter indicators.

[0033] S5.3 Construct a sensitivity matrix using positive deviation, negative deviation, and near-zero deviation groups; calculate the sensitivity of each input variable to the output variable; determine the trend of the output variable with respect to a single variable through regression analysis; quantify the relative importance of each input variable to the output variable through correlation analysis; identify the factors with the greatest impact on the results; and obtain the dynamic feedback effect among multiple factors. Here, the single variable is the adjustment period or adjustment magnitude, and the input variable is the one obtained in step S4.2. ,in The output variable is the composite index of each combination of second indicators or the relative deviation of the composite index of each combination of second indicators from the benchmark composite index. The sensitivity matrix is ​​constructed as follows, taking relative deviation as an example: , Where S is the sensitivity matrix, The value of the j-th input variable in the i-th second index combination is the value obtained in step S4.2. ( ), Let m be the relative deviation of the i-th second indicator combination, m be the number of second indicator combinations, and n be the number of input variables.

[0034] S5.4 Comparison Scheme: Construct a decision matrix and list one or more combinations of second indicators with the smallest relative deviation.

[0035] Specifically, Pareto optimality analysis can be used to select one or more combinations of second indicators with the smallest relative deviation. This embodiment provides one implementation method for reference: the non-dominated set, i.e., the Pareto front, obtained through algorithmic calculation, can be used to traverse all points using the NSGA-II algorithm, removing dominated points to determine the minimum deviation. On the Pareto front, the points or subsets with the lowest deviation can be preferentially selected, for example, those with deviations less than a defined threshold. This is done by sorting the front points according to... The top k items (k≥1, depending on actual conditions and needs) are selected in ascending order, and their corresponding second indicator combinations are then determined, thus obtaining one or more second indicator combinations with the smallest relative deviation. Since Pareto optimality analysis is a method already in use, the specific calculation formulas involved will not be elaborated here. Preferably, when constructing the decision matrix and selecting schemes, the regional characteristics of the study area should be fully considered. This facilitates qualitative comparison and scheme filtering, helps explain the heterogeneity of the deviation distribution, and thus explains the scheme priority ranking in Pareto analysis. Specific details can be determined based on the actual implementation situation.

[0036] If the determined second indicator combination is one, then one of the indicators will be designated as the water price reference indicator; if the determined second indicator combination is multiple, then if there are overlapping indicators among all the indicators, the extra ones will be removed, that is, only one of the same indicators needs to be kept, and the final indicator will be designated as the water price reference indicator.

[0037] It should be noted that the water price reference indicators determined in this application are only used as analytical objects in the process of water price adjustment and determination, and do not directly determine water prices. This application provides more scientific and reasonable reference elements and analytical objects for existing water price adjustment and determination methods.

[0038] Example 2: The technical solution provided in this embodiment is: a water price reference index determination system based on multi-factor feedback optimization, including a data acquisition module, a grouping module, a first analysis module, a second analysis module, and a third analysis module. Wherein: The data acquisition module is used to acquire data within the study area, construct an indicator system affecting water prices, and obtain indicator data for the planning level year of the study area, as well as the adjustment cycle and adjustment range. Data is collected from existing databases and includes the study area's GDP, secondary industry ratio, resident population, per capita disposable income, water consumption, water consumption per 10,000 yuan of GDP, per capita water consumption, and water prices, with adjustment ranges of 5%, 10%, 15%, and 20%, and adjustment cycles selected from two, three, and five years.

[0039] The input of the grouping module is connected to the output of the data acquisition module. The grouping module is used to select different indicators based on the constructed indicator system, combine them to obtain multiple indicator combinations, and add the adjustment period and adjustment magnitude to each indicator combination to obtain and output multiple first indicator combinations. It is also used to construct and output a third indicator combination based on the acquired indicator data for the planning level year. Specifically, when adding the adjustment period and adjustment magnitude to each of the indicator combinations, the adjustment magnitude is selected from any one of 5%, 10%, 15%, and 20%, and the adjustment period is selected from any one of two, three, and five years.

[0040] The input of the first analysis module is connected to the output of the grouping module. The first analysis module is used to obtain the average annual compound growth rate of the corresponding indicator in each first indicator combination, and to obtain the adjustment value of the corresponding indicator in the planning level year, thereby obtaining and outputting multiple second indicator combinations.

[0041] Furthermore, the first analysis module includes a first processing unit for obtaining the average annual compound growth rate of the corresponding indicator in each first indicator combination. The average annual compound growth rate of the indicator is obtained according to the following formula: , Wherein, CAGR is the compound annual growth rate of this indicator. These are the initial values ​​for the corresponding indicators. is the final value of the corresponding indicator, and N is the year difference between the year of the final value and the year of the initial value.

[0042] Furthermore, the first analysis module also includes a second processing unit for obtaining the adjusted values ​​of the corresponding indicators for the planning level year and for obtaining multiple combinations of second indicators. The adjusted values ​​of the indicators for the planning level year are obtained according to the following formula: , in, This is the adjustment value for the indicator. is the initial value of the indicator, CAGR is the average annual compound growth rate of the indicator, and 'a' is the number of adjustment years. Specifically, the number of adjustment years is an integer multiple of the adjustment period in the first indicator combination in which the indicator belongs.

[0043] The first analysis module analyzes and processes multiple combinations of first indicators output by the grouping module to obtain multiple combinations of second indicators.

[0044] The input terminals of the second analysis module are connected to the output terminals of the first analysis module and the grouping module, respectively. The second analysis module is used to analyze and process multiple combinations of second and third indicators based on the entropy weight method, and to obtain and output the comprehensive index of each combination of second indicators and the comprehensive index of each combination of third indicators.

[0045] Furthermore, the second analysis module includes a third processing unit, which is used to mark all indicators in the second and third indicator combinations as positive or negative indicators, respectively, and to adjust the values ​​of the indicators marked as negative indicators. Perform the reciprocal operation to obtain its processed value. Output the combination of indicators after the above processing. , n represents the number of indicators in a single indicator combination. ,in, The second set of indicators is the processed combination, and m is the number of combinations of the second indicator. This is the third indicator combination after processing.

[0046] Furthermore, the second analysis module also includes a fourth processing unit, which is used to process the combination of indicators output by the third processing unit. Construct matrix X (m+1)×n， And conduct on Perform Min-Max normalization to obtain and output the corresponding and matrix .

[0047] Among them ( , If the corresponding indicator is a contrarian indicator, then its processed value is used. If the corresponding indicator is a positive indicator, then its adjusted value is used. ; , in, For indicator combination The minimum value in, For indicator combination The maximum value in, here It is processed according to the value selection principles of reverse and positive indicators when constructing the matrix.

[0048] Furthermore, the second analysis module also includes methods for matrix-based analysis. Construct and output the scaling matrix The fifth processing unit. Among them, It is obtained from the following formula: ; in, , m is the number of second indicator combinations, and n is the number of indicators in a single second indicator combination.

[0049] Furthermore, the second analysis module also includes a sixth processing unit for obtaining and outputting the entropy value of the indicator. The entropy value... It is obtained from the following formula: ; in, , m is the number of second indicator combinations, and n is the number of indicators in a single second indicator combination.

[0050] Furthermore, the second analysis module also includes a seventh processing unit for obtaining and outputting the weights of the indicators. The weights... It is obtained from the following formula: ; in, , where n is the number of indicators in a single second indicator combination.

[0051] Furthermore, the second analysis module also includes an eighth processing unit for obtaining and outputting a composite index of the combinations of indicators. The composite index... It is obtained from the following formula: , in, Let i be the composite index of the i-th indicator combination. , , where m is the number of second indicator combinations, and n is the number of indicators in a single second indicator combination. Furthermore, when hour, This is the composite index of the combination of each second indicator. hour, This is the composite index of the third indicator combination.

[0052] The input of the third analysis module is connected to the output of the second analysis module. The third analysis module is used to obtain the relative deviation of the comprehensive index of each second indicator combination using the comprehensive index of the third indicator combination as a comparison benchmark, evaluate and screen the second indicator combination with the smallest relative deviation, determine the indicators in it as water price reference indicators and output them.

[0053] Furthermore, the third analysis module includes a ninth processing unit for obtaining and outputting the relative deviation of the composite index of each combination of second indicators. The relative deviation... It is obtained from the following formula: , in, Let be the relative deviation of the composite index of the i-th combination of second indicators. m is the number of combinations of the second indicator. The benchmark composite index, i.e., the composite index of the third indicator combination. .

[0054] Furthermore, the third analysis module also includes methods for analyzing relative deviations. The tenth processing unit performs classification. Specifically, it classifies the relative deviations. They are divided into three categories according to the following criteria: positive deviation group, negative deviation group, and near-zero deviation group. , The value of ε is set according to the actual situation. The number and characteristics of the second indicator combination corresponding to each category are counted. The deviation distribution is displayed using box plots or scatter plots. Outliers are identified and the causes of outliers are analyzed. The value of ε is optimized and determined, and the final classification is determined and output.

[0055] Furthermore, the third analysis module also includes an eleventh processing unit, used to construct a sensitivity matrix using positive deviation groups, negative deviation groups, and near-zero deviation groups. This unit analyzes the sensitivity of each input variable to the output variable, determines the trend of the output variable with respect to a single variable based on regression analysis, quantifies the relative importance of each input variable to the output variable based on correlation analysis, identifies the factors with the greatest impact on the results, and obtains the dynamic feedback effect among multiple factors. Here, the single variable is the adjustment period or adjustment magnitude, and the input variables are obtained from the second analysis module. ,in The preferred output variable is the relative deviation between the composite index of each combination of second indicators and the benchmark composite index; The sensitivity matrix is ​​constructed as follows, taking relative deviation as an example: , Where S is the sensitivity matrix, The value of the j-th input variable in the i-th second indicator combination, i.e. ( ), Let m be the relative deviation of the i-th second indicator combination, m be the number of second indicator combinations, and n be the number of input variables.

[0056] Furthermore, the third analysis module also includes a twelfth processing unit, used to construct a decision matrix, list one or more combinations of second indicators with the smallest relative deviation, determine the indicators among them as water price reference indicators, and output them. Preferably, Pareto optimality analysis is used to select one or more combinations of second indicators with the smallest relative deviation. This embodiment provides one implementation method for reference: the non-dominated set, i.e., the Pareto front, obtained through algorithmic calculation, can be used to traverse all points using the NSGA-II algorithm, removing dominated points to determine the minimum deviation. On the Pareto front, the points or subsets with the lowest deviation can be preferentially selected, for example, those with deviations less than a defined threshold. This is done by sorting the front points according to... The first k indicators (k≥1, depending on actual conditions and needs) are selected in ascending order to determine their corresponding second indicator combinations, resulting in one or more second indicator combinations with small relative deviations. If a single second indicator combination is determined, the indicator within it is designated as the water price reference indicator. If multiple second indicator combinations are determined, any overlapping indicators are removed, meaning only one identical indicator is retained. The resulting indicators are then used as the water price reference indicators and output.

[0057] The present invention also provides a third technical solution: a water price reference index determination device based on multi-factor mutual feedback optimization, comprising: Memory for storing non-transitory computer-readable instructions; and A processor is used to run non-transitory computer-readable instructions, such that when the non-transitory computer-readable instructions are executed by the processor, the above-mentioned method for determining water price reference indicators based on multi-factor mutual feedback optimization is implemented.

[0058] The present invention also provides a fourth technical solution: a computer-readable storage medium for storing non-transitory computer-readable instructions, which, when executed by a computer, cause the computer to execute the above-mentioned method for determining water price reference indicators based on multi-factor feedback optimization.

[0059] Therefore, due to the application of the above technical solution, the present invention has the following advantages compared with the prior art: This invention comprehensively addresses and efficiently processes the multi-dimensional factors affecting water prices by generating all indicator combinations, adjusting indicator values ​​based on the compound growth rate, and objectively determining weights using the entropy weight method. It overcomes the problems of low efficiency, unstable analysis, reliance on static models or subjective weights in existing water price analysis or adjustment methods, and significantly improves the scientificity, accuracy, and objectivity of data processing, analysis, and results.

[0060] The above embodiments are only for illustrating the technical concept and features of the present invention. Their purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be used to limit the scope of protection of the present invention. All equivalent changes or modifications made in accordance with the spirit and essence of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A method for determining water price reference indicators based on multi-factor feedback optimization, characterized in that: Includes the following steps: S1. Obtain data within the study area and construct an indicator system affecting water prices; S2. Based on the constructed indicator system, different indicators are selected and combined to obtain multiple indicator combinations. The adjustment period and adjustment range are added to each of the indicator combinations to obtain multiple first indicator combinations. S3. Based on the constructed multiple first indicator combinations, calculate the average annual compound growth rate of the corresponding indicator in each first indicator combination, and calculate the adjustment value of the corresponding indicator in the planning level year according to the following formula to obtain the adjusted multiple first indicator combinations, and define the adjusted first indicator combination as the second indicator combination. ， in, This is the adjustment value for the indicator. is the initial value of the indicator, CAGR is the average annual compound growth rate of the indicator, and a is the number of adjustment years; S4. Obtain the indicator data of the planning level year of the study area to form a third indicator combination; based on multiple second indicator combinations and the third indicator combination, use the entropy weight method to calculate the entropy value and weight of each indicator, and calculate the comprehensive index of each second indicator combination and the comprehensive index of the third indicator combination. S5. Based on the comprehensive index of the multiple second indicator combinations obtained, and using the comprehensive index of the third indicator combination as a comparison benchmark, calculate the relative deviation of the comprehensive index of each second indicator combination, evaluate and select one or more second indicator combinations with the smallest relative deviation, and determine the indicators among them as water price reference indicators.

2. The method for determining water price reference indicators based on multi-factor feedback optimization according to claim 1, characterized in that: In step S2, the adjustment range in each of the first indicator combinations is selected from any one of 5%, 10%, 15%, and 20%; the adjustment period in each of the first indicator combinations is selected from any one of two years, three years, and five years.

3. The method for determining water price reference indicators based on multi-factor feedback optimization according to claim 1, characterized in that: In step S3, the average annual compound growth rate of the corresponding indicator is calculated according to the following formula two. ， in, These are the initial values ​​for the corresponding indicators. N represents the final value of the corresponding indicator, and N is the year difference between the year of the final value and the year of the initial value.

4. The method for determining water price reference indicators based on multi-factor feedback optimization according to claim 1, characterized in that: In step S3, the number of adjustment years is an integer multiple of the adjustment period in the first indicator combination.

5. The method for determining water price reference indicators based on multi-factor feedback optimization according to claim 1, characterized in that: In step S4, the method for calculating the entropy value and weight of each indicator using the entropy weight method, and calculating the comprehensive index of each combination of second indicators and the comprehensive index of the combination of third indicators, includes: S4.1 If the indicator is a contrarian indicator, take the reciprocal of its adjustment value to obtain the corresponding processing value. And obtain the processed index combination. , n represents the number of indicators in a single indicator combination. ;in, The second set of indicators is the processed combination, and m is the number of combinations of the second indicator. This is the third combination of indicators after processing; S4.2 Constructing a matrix based on the processed index combination , among them , , If the corresponding indicator is a contrarian indicator, then its processed value is used. If the corresponding indicator is a positive indicator, then its adjusted value is used. ;right Min-Max normalization is performed to obtain the corresponding and matrix ; S4.3 Constructing the Proportion Matrix ,matrix The following formula (3) is used for calculation: ； S4.4 Calculate the entropy value of the index The entropy value The following formula (four) is used for calculation: ； S4.5 Calculate the weight of the indicator The weight The following formula (five) is used for calculation: ； S4.6 Calculate the composite index of each indicator combination. The comprehensive index The following formula (6) is used for calculation: ； in, This is the composite index of the combination of each second indicator. This is the composite index of the third indicator combination.

6. The method for determining water price reference indicators based on multi-factor feedback optimization according to claim 5, characterized in that: In step S5, the method for calculating the relative deviation of the comprehensive index of each combination of second indicators, evaluating and selecting one or more combinations of second indicators with the smallest relative deviation, and determining the indicators therein as water price reference indicators includes: S5.1 Calculate the relative deviation of the composite index for each combination of second indicators. The relative deviation The following formula (7) is used for calculation: ， in, , The benchmark composite index, namely the composite index of the third indicator combination. ; S5.2 The relative deviation is determined according to... , and The standards are divided into positive deviation group, negative deviation group and near-zero deviation group; S5.3 A sensitivity matrix is ​​constructed using the positive deviation group, negative deviation group, and near-zero deviation group. The sensitivity of each input variable to the output variable is calculated. Regression analysis is used to determine the trend of the output variable with respect to a single variable. Correlation analysis is used to quantify the relative importance of each input variable to the output variable, identifying the factors that have the greatest impact on the results and obtaining the dynamic feedback effect among multiple factors. Here, the single variable is the adjustment period or the adjustment magnitude, and the input variable is the... ,in The output variable is the composite index of each combination of second indicators or the relative deviation of the composite index of each combination of second indicators from the benchmark composite index. S5.4 Comparison Scheme: Construct a decision matrix and list one or more combinations of second indicators with the smallest relative deviation.

7. A water price reference index determination system based on multi-factor feedback optimization, characterized in that: include The data acquisition module is used to acquire data within the study area, construct an indicator system affecting water prices, acquire indicator data for the planning level year of the study area, and obtain the adjustment cycle and adjustment range. The grouping module is used to select different indicators based on the constructed indicator system to obtain multiple indicator combinations, and to add the adjustment period and adjustment range to each indicator combination to obtain and output multiple first indicator combinations. It is also used to construct and output a third indicator combination based on the indicator data of the acquired planning level year. The first analysis module is used to obtain the average annual compound growth rate of the corresponding indicator in each first indicator combination, and to obtain the adjustment value of the corresponding indicator in the planning level year, thereby obtaining and outputting multiple second indicator combinations. The second analysis module is used to analyze and process multiple combinations of the second indicators and the third indicators based on the entropy weight method, and to obtain and output the comprehensive index of each combination of the second indicators and the comprehensive index of the third indicator. The third analysis module is used to obtain the relative deviation of the comprehensive index of each second indicator combination based on the comprehensive index of the third indicator combination, evaluate and screen out one or more second indicator combinations with the smallest relative deviation, determine the indicators among them as water price reference indicators and output them.

8. A device for determining water price reference indicators based on multi-factor feedback optimization, characterized in that: include Memory is used to store non-transitory computer-readable instructions; as well as A processor is configured to execute the non-transitory computer-readable instructions such that when the non-transitory computer-readable instructions are executed by the processor, the method for determining water price reference indicators based on multi-factor mutual feedback optimization as described in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium for storing non-transitory computer-readable instructions, which, when executed by a computer, cause the computer to perform the water price reference index determination method based on multi-factor feedback optimization as described in any one of claims 1 to 6.

Images