Load refined grading method and system for power system containing high-proportion new energy

Abstract

The invention discloses a fine load grading method and system for a power system containing high-proportion new energy, and relates to the related technical field of a power load classification system.The method comprises the steps that standardization processing is conducted on the basis of social safety indexes and functional indexes considering power loads, and after index weights are obtained through a data-driven weighting method, the power loads are classified to be classified according to the index weights; forming a power load refined grading table; in consideration of changes of load demand characteristics along with time and geographic positions in different scenes, load grades in different scenes are divided according to a power load refined grading table. The load priority can be dynamically adjusted in different scenes, the complex power demand of the modern society is met, the load management precision and flexibility in long-term planning of a power grid are improved, and the method has remarkable technical progress and application value.

Description

Technical Field

[0001] The present invention relates to the technical field of power load classification systems. Specifically, it relates to a method and system for refined classification of loads in a power system with a high proportion of new energy. Background Art

[0002] Currently, the existing power load classification system adopts the traditional three-level load classification method. This method classifies loads according to the requirements for power supply reliability and the degree of losses or impacts caused by power supply interruption, and divides power loads into first-level loads, second-level loads, and third-level loads: First-level loads: including important government agencies, financial institutions, hospitals, and life support systems, etc., and uninterrupted power supply must be ensured. Second-level loads: including commercial buildings, industrial production, etc., power supply interruption will affect the economy but will not immediately endanger life. Third-level loads: including residential and non-important industrial loads, and power supply interruption can be accepted for a short time.

[0003] Although the traditional three-level load classification method has played an important role in the power system, in the modern context of large-scale grid connection of new energy, increasing difficulty in supply-demand balance, and gradually diversified power load characteristics, it is difficult to meet the load power supply requirements in different supply-demand balance scenarios of complex and changing power systems, mainly reflected in the following aspects: (1) Coarse classification: The three-level load classification is too general, and does not reflect the production value of different types of electrical loads and their importance and functional contributions to social safety, and it is difficult to reflect the actual diversity and complexity of loads. In the evolution of the power grid form, the response requirements of different loads vary greatly, and the simple three-level classification cannot support refined power grid management and control. (2) Ignoring the spatio-temporal characteristics of loads: The three-level load classification does not fully consider the differences in load time and space. With the increase in the types of loads, the load characteristics in different regions vary significantly at different times, and detailed analysis and dynamic management of regional load levels become particularly important to effectively respond to changes in supply-demand balance in various scenarios. Summary of the Invention

[0004] In order to solve the above problems, the purpose of the present invention is to provide a method for refined classification of loads in a power system with a high proportion of new energy, aiming to achieve efficient allocation of power resources.

[0005] In order to achieve the above technical objectives, the present application provides a method for refined classification of loads in a power system with a high proportion of new energy, including the following steps: Based on considering the social safety index and functional index of the power load, perform standardization processing, and adopt a data-driven weighting method. After obtaining the index weights, form a refined classification table of power loads; Consider the variation of load demand characteristics with time and geographical location under different scenarios, and classify the load levels under different scenarios according to the refined power load classification table.

[0006] Preferably, when obtaining the social security index, generate the social security index according to the life support sub-index, public security sub-index, emergency demand sub-index, and social basic function sub-index. Among them, the life support sub-index is used to measure the role of power load in life support; the public security sub-index is used to evaluate the role of power load in maintaining public security; the emergency demand sub-index is used to reflect the power demand and response speed of power load in extreme situations; the social basic function sub-index is used to measure the basic support ability of power load in daily social operation.

[0007] Preferably, when obtaining the functional index, generate the functional index according to the unit load function contribution sub-index and the function interruption impact sub-index. Among them, the unit load function contribution sub-index is used to evaluate the economic value contribution of the load to society or the system during normal operation; the function interruption impact sub-index is used to measure the economic loss or system impact caused by power interruption to the load.

[0008] Preferably, when performing standardization processing, perform standardization processing through the Z-score standardization method to eliminate the influence of different sub-index dimensions and value ranges.

[0009] Preferably, when obtaining the index weight, obtain the index weight according to the amount of index data. Among them, when the amount of index data is limited, use the entropy method to obtain the index weight; when the amount of index data is sufficient, use the principal component analysis method or the decision tree model to obtain the index weight.

[0010] Preferably, when classifying the load levels under different scenarios, analyze the load demand in the current situation, including the social security and functional changes of different loads during the current time period, and the social security and functional attributes of the regions where they are located; According to the demand analysis results, classify various loads according to the current situation, give priority to retaining loads with high social security, and refine the classification according to the functional index among loads with the same social security level; According to the classification results, when the power supply is insufficient, sort the various loads according to their priorities, and gradually adjust the loads from low to high in turn to ensure that the power loads with significant impacts on social security and high functionality are preferentially guaranteed.

[0011] The present invention discloses a refined load classification system for a power system with a high proportion of new energy, including: A load classification module, which is used to perform standardization processing based on considering the social security index and functional index of the power load, and adopt a data-driven weight assignment method. After obtaining the index weights, a refined power load classification table is formed. A load division module, which is used to consider the variation of load demand characteristics with time and geographical location under different scenarios, and divide the load levels under different scenarios according to the refined power load classification table.

[0012] Preferably, the load classification module is further used to generate a social security index according to the life support sub-index, public safety sub-index, emergency demand sub-index and social basic function sub-index, wherein the life support sub-index is used to measure the role of the power load in life support; the public safety sub-index is used to evaluate the role of the power load in maintaining public safety; the emergency demand sub-index is used to reflect the power demand and response speed of the power load in extreme situations; the social basic function sub-index is used to measure the basic support ability of the power load in daily social operation.

[0013] Preferably, the load classification module is further used to generate a functional index according to the unit load function contribution sub-index and function interruption impact sub-index, wherein the unit load function contribution sub-index is used to evaluate the economic value contribution of the load to society or the system during normal operation; the function interruption impact sub-index is used to measure the economic loss or system impact caused by power interruption to the load.

[0014] Preferably, the load classification module is further used to perform standardization processing by the Z-score standardization method to eliminate the influence of different sub-index dimensions and value ranges; and obtain the index weights according to the amount of index data. Among them, when the amount of index data is limited, the entropy method is used to obtain the index weights; when the amount of index data is sufficient, the principal component analysis method or the decision tree model is used to obtain the index weights.

[0015] The present invention discloses the following technical effects: (1) Refined load classification: The load classification method proposed by the present invention is based on two types of indicators, social security and functionality, and performs more refined load classification, avoiding the deficiency of rough classification in the existing solutions.

[0016] (2) Objective weight assignment: Based on the data-driven weight assignment method, the scientificity and reliability of the classification are improved.

[0017] (3) Flexibility: The solution can dynamically adjust the weight assignment based on the load characteristics under specific scenarios, and has a wider applicability. Description of the Drawings

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0019] Figure 1 It is a schematic flow diagram of the method described in the present invention. Detailed implementation manners

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some, rather than all, of the embodiments of the present application. Usually, the components of the embodiments of the present application described and illustrated in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the present application claimed, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0021] As Figure 1 shown, the present invention provides a method for refined classification of loads in a power system with a high proportion of new energy, which belongs to a refined classification technology of loads considering the social safety and functionality of power loads. This method combines the technical and physical characteristics of various loads, refinedly measures the social safety and functionality of various loads, takes into account indicators of different dimensions and units, can accurately identify the cost of power loss of various loads and the impact on social safety when the power supply capacity of a power system with a high proportion of new energy is lacking, improves the scientificity of formulating the load classification of a new power system with a high proportion of new energy and a low proportion of thermal power, and realizes the efficient allocation of power resources. The method for refined classification of loads specifically includes the following contents: (1) Calculation of social safety indicators of power loads; (2) Calculation of functional indicators of power loads; (3) Standardization of each indicator; (4) Selection of the weights of each indicator; (5) Formation of a refined classification table of power loads; (6) Generation of a process for classifying power load levels in a specific scenario.

[0022] In step (1), the calculation method of the social safety indicators of power loads is as follows: The indicator formula for considering the social safety of power loads is: , wherein, represents the importance of measuring the load in terms of life support; represents the importance of measuring the load in maintaining public safety; represents the importance of the load in extreme situations such as disasters and wars; represents the supporting role of the load in basic social functions such as transportation and communication; , , , represent the weights of each sub - index of social safety and satisfy .

[0023] (1) Life support sub - index: The life support sub - index measures the role of the power load in life support, especially applicable to places such as hospitals and emergency centers. This index considers the frequency and duration of the load requirements of emergency medical assistance facilities and equipment, reflecting the key role of the power load in ensuring people's life safety. The formula for the life support sub - index is: , wherein, represents the occurrence frequency of life - safety events, such as the number of times of sudden medical assistance needs; represents the duration of the system to handle life - safety events.

[0024] (2) Public safety sub - index: The public safety sub - index is used to evaluate the role of the power load in maintaining public safety, such as the power load requirements in places like police stations and fire stations. This sub - index considers the frequency of sudden public safety events and the degree of impact on society to assess the importance of the power load in ensuring public safety. The formula for the public safety sub - index is: , wherein, represents the occurrence frequency of public safety events, such as the frequency of public order emergencies; represents the degree of impact of the event on society.

[0025] (3) Emergency demand sub - index: The emergency demand sub - index reflects the power demand and response speed of the power load in extreme situations such as disasters and wars. This item especially focuses on the power demand of key loads in emergency scenarios to ensure the reliable power supply of key loads in emergency situations. The formula for the emergency demand sub - index is: , wherein, Indicates the load demand under emergency conditions, representing the power demand of this load in the emergency scenario; Indicates the response speed of the system to emergency events.

[0026] (4)Social basic function sub - index: The social basic function sub - index focuses on the basic support functions of the power load in the daily social operation, such as the power demand of basic facilities like transportation and communication. This index includes the service coverage area and the daily duration of the service, measuring the degree of support of the load for daily life and social operation. The formula for the social basic function sub - index is: , Among them, Indicates the service coverage area, representing the user or geographical scope supported by this load; Indicates the daily duration of the service, characterizing the continuous demand of this load in the normal state.

[0027] In step (2), the calculation method of the power load functional index is as follows: The functional index combines the unit function of the load and the impact of load interruption to measure the overall importance of the load functionality. Its calculation formula is: , Among them: Indicates the functional contribution of the unit load; Indicates the impact of power interruption on the load function; and respectively represent the weights of the unit load function contribution index and the function interruption impact index, and satisfy .

[0028] (1)Unit load function contribution sub - index: The unit load function contribution sub - index is used to evaluate the economic value contribution of the load to society or the system during normal operation. This index is applicable to analyzing economic outputs such as the annual output value of a factory or the sales volume of commercial facilities, reflecting the economic importance of the load in the normal state. The formula for the unit load function contribution sub - index is: , Among them, Indicates the gross production value of this load per unit time; Indicates the power consumption of the load per unit time.

[0029] (2)Function interruption impact sub - index: The function interruption impact sub - index is used to measure the economic losses or system impacts caused by power interruptions on the load. It mainly examines the duration and impact depth of power interruptions to evaluate the potential losses of power outages on the economy or system operation. The formula for the load function interruption impact sub - index is as follows: , where, represents the critical function duration of the load, that is, the duration of the impact of this load on the system or society; represents the interruption impact depth, that is, the potential economic loss caused by the power interruption.

[0030] The standardization methods for each index in step (3) are as follows: In the refined classification method of power loads of the present invention, the Z - score standardization method is used to standardize each sub - item of the social safety and functionality indexes, so that each index is comparable on the same scale. Since each sub - item of the social safety and functionality indexes has different dimensions and data ranges, directly performing weighted summation will lead to inaccurate evaluation results. Therefore, the Z - score method is used to dimensionless process each sub - index.

[0031] (1) The principle of the Z - score standardization method: Z - score standardization is a common dimensionless method. By adjusting the mean and standard deviation of the original data, the data is transformed into a standard normal distribution with zero mean and unit variance. The calculation formula of Z - score is as follows: , where, Z represents the value after standardization; X represents the original data value; μ represents the mean of this index; σ represents the standard deviation of this index.

[0032] After Z - score standardization, the data will fall into a symmetric standard normal distribution, and most of the data is concentrated between - 3 and + 3, so that they can be compared and weighted on the same scale.

[0033] (2) The specific application of the Z - score method in the present invention: In the refined load classification method proposed in the present invention, the social safety and functionality indexes include multiple sub - items. To evaluate these sub - items on a unified scale, the present invention performs Z - score standardization processing on each sub - index respectively to eliminate the influence of different dimensions and numerical ranges of different sub - indexes. For example, for the life - guarantee sub - index , after calculating its mean and standard deviation in different load objects, the life - guarantee values of each load object are converted into Z - score standardized values. This process is applicable to all sub - indexes.

[0034] The method for selecting the weights of each index in step (4) is as follows: To ensure the objectivity of the sub-index weights, this scheme considers using a data-driven weight assignment method. Different methods for selecting index weights can be adopted according to the amount of index data.

[0035] (1) When the amount of data is limited: When the amount of data for an index is limited, the entropy method is used. The entropy method objectively assigns weights by calculating the dispersion degree of each sub-index. The smaller the entropy value of an index, the greater the dispersion degree, the more information, and thus it is given a higher weight, so as to accurately reflect its importance to the overall load classification.

[0036] (2) When the amount of data is sufficient: When the amount of data for a sub-index is sufficient, the principal component analysis (PCA) or decision tree model is used. PCA extracts the principal components based on the covariance matrix between data and assigns weights according to the contribution rate of the principal components. The decision tree model automatically analyzes the relative importance of each feature to objectively determine the weights.

[0037] After determining the weights of each sub-index, the standardized sub-index values are weighted and summed to obtain the comprehensive evaluation value of the power load in terms of social security and functionality.

[0038] In step (5), a refined power load classification table is formed, as follows: After standardization and weighting, the values of the social security index and functionality index of the load are both in the range of 0 - 1.

[0039] To illustrate the refined load classification, the social security index and functionality index of the load can be divided into three categories: high, medium, and low. According to the different combinations of high, medium, and low weights of the social security and functionality of the load, all loads can be divided into nine categories. The classification of the nine categories of loads is shown in the following table: Nine-level load classification High functionality Medium functionality Low functionality High social security A1 A2 A3 Medium social security B1 B2 B3 Low social security C1 C2 C3

[0040] The characteristic descriptions of the nine-level loads are shown in the following table: Characteristic descriptions of nine-level loads Classification Social security Functionality Typical applications Main features A1 High High National financial system, medical equipment in core hospitals, government emergency command centers, large data centers, etc. It has extremely high functional contribution and is closely related to social security. Power interruption will seriously affect the country's core economy and security system, and may endanger life and national security. A2 High Medium High-end manufacturing industries (such as chips, precision machinery), logistics transportation networks, financial transaction back-end systems, etc. The functional contribution is large and it has medium importance for social security. Power interruption mainly affects economic operations but does not directly threaten public safety. A3 High Low Remote medical facilities, key community infrastructure, etc. High social security. Although the functionality is low, it is crucial in specific regions or groups. Power interruption may seriously affect social stability. B1 Medium High Traffic command systems, water supply systems, some medical service facilities (such as emergency departments), etc. Medium social security, high functional contribution. Power interruption affects public safety and also has a greater impact on economic activities. B2 Medium Medium Medium-sized industrial parks, schools, general government office buildings, etc. It takes into account both functionality and social security. Power interruption will not immediately cause serious consequences, but will cause certain economic losses and social inconveniences. B3 Medium Low General service industries (such as catering, hotels), ordinary commercial office buildings, etc. Medium social security, low functionality. The impact of power interruption is mainly reflected in commercial losses and has little impact on social security. C1 Low High Large commercial office areas, large manufacturing plants, etc. Low social security but high functional contribution. Power interruption mainly affects production and operation efficiency and has little impact on public safety. C2 Low Medium Ordinary commercial facilities, residential communities, street lights, etc. Medium functional contribution, but low social security. Power interruption affects the quality of life and has little impact on social stability. C3 Low Low Entertainment facilities (such as cinemas, advertising screens), etc. Both social security and functionality are low. Power interruption will not cause serious consequences and is applicable to non-essential service areas. In step (6), a power load level classification process for a specific scenario is generated, as follows: In the actual operation of power systems, the load demand characteristics under different scenarios can vary significantly over time and geographical locations. Therefore, it is necessary to dynamically adjust the load levels in specific situations to more accurately reflect the importance of the loads and ensure a reasonable allocation of power resources. When classifying load levels in specific situations, social safety and functional indicators need to be combined, and time and geographical characteristics should be considered to determine the priorities of various loads. The process of classifying power load levels in specific scenarios is as follows Figure 1 as shown.

[0041] (1)Load characteristic analysis: In specific situations, by analyzing the time and geographical characteristics of the loads, the load levels can be adjusted more flexibly to ensure a reasonable allocation of power resources.

[0042] 1)Time characteristics: The social safety and functionality of loads may vary greatly over time. For example: Some industrial loads are mainly used for production during the day, with a high functional contribution, while at night they are mainly for self-use such as lighting, with a relatively low functional contribution. When adjusting the loads, the functional weights can be dynamically adjusted according to the load characteristics during the day and at night; Medical and emergency loads have a higher social safety at night and during emergencies, and need to be given priority to ensure the normal operation of first aid and important medical services.

[0043] 2)Geographical characteristics: The importance of loads is affected by geographical locations. For example: Key loads in economic center areas, such as the financial system and large commercial office areas, have a high functional contribution and may have a higher functional weight; Public safety facilities (such as medical and emergency services) in remote areas have a higher social safety and need to be given priority under specific conditions.

[0044] (2)Dynamic adjustment process for classifying power loads In specific situations, the adjustment of loads by the power system needs to be based on the assessment of social safety and functional indicators, and combined with time and geographical characteristics. The specific process is as follows: 1)Analyze load characteristics: Analyze the load demand in the current situation, including the changes in social safety and functionality of different loads during the current time period, as well as the social safety and functional attributes of their locations.

[0045] 2)Determine load levels: According to the refined load classification method, classify various loads according to the current situation and dynamically adjust if necessary. Loads with high social safety are given priority to be retained, and within the same social safety level, they are further classified according to functional indicators.

[0046] 3) Sorting and priority scheduling: When the power supply is insufficient, sort according to the priorities of various types of loads, and gradually adjust the loads from low to high in sequence to ensure that the power with a significant impact on social safety and high functionality is preferentially guaranteed.

[0047] The present invention can dynamically adjust the priorities of loads in different scenarios, meet the complex power demands of modern society, improve the accuracy and flexibility of load management in the medium- and long-term planning of the power grid, and has significant technological progress and application value.

[0048] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, and the combination of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to generate a machine, so that the instructions executed by the processors of the computer or other programmable data processing devices generate means for realizing the functions specified in one Figure 1 one process or multiple processes and / or blocks Figure 1 or multiple blocks.

[0049] In the description of the present invention, it should be understood that the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, "a plurality" means two or more unless otherwise specifically defined.

[0050] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A method for refined classification of loads in a power system with a high proportion of new energy, characterized in that, It includes the following steps: Based on considering the social safety indicators and functional indicators of the power load, conduct standardization processing, and adopt a data-driven weighting method. After obtaining the indicator weights, form a refined classification table of power loads; Consider the variation of load demand characteristics with time and geographical location under different scenarios, and according to the refined classification table of power loads, divide the load levels under different scenarios.

2. The refined load classification method for a power system with a high proportion of new energy according to claim 1, characterized in that: When obtaining the social safety indicators, generate the social safety indicators according to the life support sub-indicator, public safety sub-indicator, emergency demand sub-indicator, and social basic function sub-indicator. Among them, the life support sub-indicator is used to measure the role of the power load in life support; the public safety sub-indicator is used to evaluate the role of the power load in maintaining public safety; the emergency demand sub-indicator is used to reflect the power demand and response speed of the power load in extreme situations; the social basic function sub-indicator is used to measure the basic support ability of the power load in the daily social operation.

3. The refined load classification method for a power system with a high proportion of new energy according to claim 2, characterized in that: When obtaining the functional indicators, generate the functional indicators according to the unit load function contribution sub-indicator and the function interruption impact sub-indicator. Among them, the unit load function contribution sub-indicator is used to evaluate the economic value contribution of the load to society or the system during normal operation; the function interruption impact sub-indicator is used to measure the economic loss or system impact caused by power interruption to the load.

4. The refined load classification method for a power system with a high proportion of new energy according to claim 3, characterized in that: When conducting standardization processing, perform standardization processing through the Z-score standardization method to eliminate the influence of different sub-indicator dimensions and value ranges.

5. The refined load classification method for a power system with a high proportion of new energy according to claim 4, characterized in that: When obtaining the indicator weights, obtain the indicator weights according to the amount of indicator data. Among them, when the amount of indicator data is limited, use the entropy method to obtain the indicator weights; when the amount of indicator data is sufficient, use the principal component analysis method or the decision tree model to obtain the indicator weights.

6. The refined load classification method for a power system with a high proportion of new energy according to claim 5, characterized in that: When dividing the load levels under different scenarios, analyze the load demand in the current situation, including the social safety and functional changes of different loads during the current time period, and the social safety and functional attributes of the area where they are located; According to the demand analysis results, classify various loads according to the current situation, give priority to retaining the loads with high social safety, and then refine the classification according to the functional indicators among the same social safety levels; According to the classification results, when the power supply is insufficient, sort the various loads according to their priorities, and gradually adjust the loads from low to high in turn to ensure that the power loads with a significant impact on social safety and higher functionality are preferentially guaranteed.

7. A load fine - grading system for a power system with a high proportion of new - energy power, characterized in that, Including: A load classification module, which is used to perform standardized processing based on considering the social security index and functional index of the power load, and adopt a data-driven weighting method. After obtaining the index weights, a refined power load classification table is formed. A load division module, which is used to consider the variation of load demand characteristics with time and geographical location under different scenarios, and divide the load levels under different scenarios according to the refined power load classification table.

8. The refined load classification system for a power system with a high proportion of new energy according to claim 7, wherein: The load classification module is further used to generate the social security index according to the life support sub-index, public safety sub-index, emergency demand sub-index and social basic function sub-index. Among them, the life support sub-index is used to measure the role of the power load in life support; the public safety sub-index is used to evaluate the role of the power load in maintaining public safety; the emergency demand sub-index is used to reflect the power demand and response speed of the power load in extreme situations; the social basic function sub-index is used to measure the basic support ability of the power load in the daily social operation.

9. The refined load classification system for a power system with a high proportion of new energy according to claim 8, wherein: The load classification module is further used to generate the functional index according to the unit load function contribution sub-index and function interruption impact sub-index. Among them, the unit load function contribution sub-index is used to evaluate the economic value contribution of the load to society or the system during normal operation; the function interruption impact sub-index is used to measure the economic loss or system impact caused by power interruption to the load.

10. The refined load classification system for a power system with a high proportion of new energy according to claim 9, wherein: The load classification module is further used to perform standardized processing through the Z-score standardization method to eliminate the influence of different sub-index dimensions and value ranges; and obtain the index weights according to the amount of index data. Among them, when the amount of index data is limited, the entropy method is used to obtain the index weights; when the amount of index data is sufficient, the principal component analysis method or decision tree model is used to obtain the index weights.

Images