A distribution network expansion scheme evaluation method based on residential load characteristics

Through the evaluation method based on the characteristics of residents' load, the problem of unconsidered load characteristics in the distribution network industry expansion and installation plan in the existing technology was solved, and the accurate assessment of the impact on residents' load and the improvement of the matching degree of distribution capacity was achieved. Equipment utilization and economics.

CN114066253BActive Publication Date: 2024-11-22GUANGZHOU POWER SUPPLY BUREAU GUANGDONG POWER GRID CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111365004.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-11-17
Publication Date
2024-11-22
Estimated Expiration
2041-11-17

AI Technical Summary

Technical Problem

The existing distribution network industry expansion and installation plan lacks the characteristics of residents' load, which leads to an increase in the peak-to-valley difference in the power supply point load, which may touch the maximum transmission capacity limit of the power supply line, and the distribution capacity matching is inappropriate, reducing equipment utilization and economics. .

Method used

The evaluation method based on the resident load characteristics is adopted to determine the load characteristic indicators, build a cloud model, calculate the membership value, determine the evaluation level, quantify the evaluation value, adjust the weight, build a comprehensive evaluation cloud model, and obtain the final comprehensive evaluation level.

Benefits of technology

It has achieved an objective and accurate assessment of the impact on residents' load, improved the matching degree of distribution capacity between the industry expansion and installation plan and load, and improved the utilization rate of equipment and the operational economy of the distribution network.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114066253B_ABST
    Figure CN114066253B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for evaluating a distribution network expansion and installation scheme based on the characteristics of residential loads, including: obtaining the index data of the residential load characteristic index of the distribution network expansion and installation scheme to be evaluated, determining the evaluation level set according to the value range of each characteristic index, constructing a cloud model of each evaluation level of each characteristic index, calculating the membership value of each characteristic index under each evaluation level according to the membership function and index data of the cloud model; determining the evaluation level corresponding to each characteristic index according to the obtained membership value, determining the final evaluation level of each characteristic index according to the evaluation level and the preset reference level; quantifying the final evaluation level of each characteristic index to obtain the index evaluation value; constructing a comprehensive evaluation cloud model according to the adjustment weight of each characteristic index and the cloud model corresponding to its evaluation level to obtain the final comprehensive evaluation level. The present invention can accurately evaluate the impact of residential load on the distribution network expansion and installation scheme, which is conducive to improving the operation efficiency of the distribution network.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network, and in particular to a method for evaluating distribution network expansion and installation plans based on resident load characteristics. Background Art

[0002] With the development of social economy, the level of electrification of user equipment is constantly improving, and the demand for electricity for residents is also increasing. The electricity load has brought new challenges to the online management of power companies. At present, the focus of power companies is mainly on the safe operation of the power system, paying attention to protecting the power supply safety of high-level power users, and not paying enough attention to the power quality issues of low-voltage and low-level power users. However, power products need to be processed simultaneously in the production, supply and sales links. Delays and waste in any link are not conducive to the circulation of the entire power production and operation system. In order to meet the needs of national economic development for power supply, improving the power quality of low-level power users such as residential power loads is an important direction for the transformation and development of power companies.

[0003] Business expansion and installation service is an important business between power companies and residential users. It includes all aspects of residential users from applying for electricity to establishing normal power supply relations, and plays an important role in ensuring the safety of residential electricity use. At present, in the process of formulating business expansion and installation plans, due to the lack of user power load information, power companies generally only formulate access plans based on the maximum load reported by users, and lack detailed consideration of user load characteristics. This may result in a large number of homogeneous power loads at a certain access point. If they are accessed at the same time, the daily peak-to-valley difference of the power supply point load will increase, which may reach the maximum transmission capacity limit of the power supply line.

[0004] In addition, the evaluation indicators of the distribution network expansion plan were not selected reasonably, resulting in inappropriate matching of distribution capacity during expansion, reducing equipment utilization and the economy of the distribution network; the expansion plan ignored the development characteristics of the load and did not take into account the impact of load development on user load characteristics. Summary of the invention

[0005] The purpose of the present invention is to provide a distribution network expansion plan evaluation method based on residential load characteristics, so as to solve the technical problem that the existing methods lack consideration of the impact of residential load on the distribution network expansion plan.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A method for evaluating distribution network expansion plans based on residential load characteristics, comprising:

[0008] Determine the residential load characteristic indicators of the distribution network expansion application plan to be evaluated, and obtain the indicator data of each characteristic indicator;

[0009] Determine the evaluation level set according to the value range of each characteristic indicator, construct a cloud model of each evaluation level of each characteristic indicator, and calculate the membership value of each characteristic indicator at each evaluation level according to the membership function of the cloud model and the indicator data of each characteristic indicator;

[0010] Determine the evaluation level corresponding to each characteristic indicator according to the membership value of each characteristic indicator at each evaluation level, and determine the final evaluation level of each characteristic indicator according to the evaluation level corresponding to each characteristic indicator and the preset reference level;

[0011] Quantify the final evaluation level of each characteristic indicator to obtain the indicator evaluation value of each characteristic indicator;

[0012] According to the initial weight and indicator data of each characteristic indicator, the adjustment weight of each characteristic indicator is obtained, and according to the cloud model corresponding to the adjustment weight and the final evaluation level of each characteristic indicator, a comprehensive evaluation cloud model of the distribution network expansion application plan is constructed;

[0013] The final comprehensive evaluation level of the distribution network expansion application plan is obtained according to the comprehensive evaluation cloud model.

[0014] Optionally, the residential load characteristic indicators for the distribution network expansion application plan to be evaluated specifically include:

[0015] Determine the load index and time domain characteristic index of the residential load of the distribution network expansion application plan to be evaluated;

[0016] Among them, the load indicators include equipment indicators and regional indicators; the time domain characteristic indicators include all-day indicators, peak-valley indicators and flat-time indicators.

[0017] Optionally, determining the evaluation level set according to the value range of each characteristic indicator specifically includes:

[0018] The value range of each characteristic indicator is divided into regions according to the set limit value. Each data interval corresponds to an evaluation level, and all evaluation levels constitute an evaluation level set.

[0019] Optionally, constructing a cloud model for each evaluation level of each characteristic indicator specifically includes:

[0020] According to the limit values ​​of each evaluation level of each characteristic indicator, the standard deviation of the limit values ​​is calculated;

[0021] Calculate the numerical characteristics of the cloud model at each evaluation level of each characteristic indicator according to the standard deviation: expected value E x , Entropy value E n and the super entropy value H e ;

[0022] According to the digital features of the cloud model of each evaluation level of each characteristic indicator, the cloud model of each evaluation level of each characteristic indicator is constructed by using the cloud generator.

[0023] Optionally, using the cloud generator to construct cloud models of each evaluation level of each characteristic indicator specifically includes:

[0024] The cloud generator is used to perform the following steps for each evaluation level of each characteristic indicator:

[0025] S1: Generate E n For expectation, H e Normal random number with standard deviation

[0026] S2: Generate E x For the expectation, A normal random number with standard deviation x i ;

[0027] Repeat steps S1 and S2 until N random numbers of the cloud model are generated, where N is the preset number of random numbers, 1≤i≤N.

[0028] Optionally, calculating the membership value of each characteristic indicator at each evaluation level according to the membership function of the cloud model and the indicator data of each characteristic indicator specifically includes:

[0029] According to the membership function of the cloud model Calculate the membership value of each characteristic indicator at each evaluation level;

[0030] In the formula, k is the index data of each characteristic index, r is the expected value of k at E x , Entropy value E n and the super entropy value H e The membership value of the cloud model corresponding to the evaluation level.

[0031] Optionally, determining the evaluation level corresponding to each characteristic indicator according to the membership value of each characteristic indicator at each evaluation level specifically includes:

[0032] The maximum membership value is selected from the membership values ​​of each characteristic indicator at each evaluation level, and the evaluation level corresponding to the maximum membership value is used as the evaluation level corresponding to each characteristic indicator.

[0033] Optionally, the final evaluation level of each characteristic indicator is quantified to obtain the indicator evaluation value of each characteristic indicator, which specifically includes:

[0034] According to the formula Calculate the correction value Δc i ;

[0035] According to formula C i =C0+Δci Get the cloud drop drop(C0,y i ), C0 is the expected value of the final evaluation level of each characteristic indicator corresponding to the cloud model, E n is the entropy value of the cloud model for the final evaluation level of each characteristic index, y i is a random number x i The membership value at the final evaluation level of each characteristic indicator;

[0036] The precise values ​​of the index evaluation values ​​of N cloud droplets are averaged to obtain the index evaluation value of each characteristic index.

[0037] Optionally, obtaining the adjustment weight of each characteristic indicator according to the initial weight of each characteristic indicator and the indicator data specifically includes:

[0038] Using the balance function For the initial weight

[0039] And the indicator data of each characteristic indicator are processed to obtain the adjustment weight of each characteristic indicator:

[0040]

[0041] In the formula, α is the balance factor, 0<α≤1, x j is the normalized value of the j-th feature index, j = 1, 2, …, n, n is the number of feature indexes.

[0042] Optionally, the comprehensive evaluation cloud model of the distribution network expansion application plan is:

[0043]

[0044]

[0045]

[0046] In the formula, ω′ j is the adjustment weight of the jth feature index, satisfying are the expected value, entropy value and super entropy value of the cloud model corresponding to the final evaluation level of the j-th characteristic indicator.

[0047] The present invention provides a distribution network business expansion and installation scheme evaluation method based on resident load characteristics, comprising: determining resident load characteristic indicators of the distribution network business expansion and installation scheme to be evaluated, and obtaining indicator data of each characteristic indicator; determining an evaluation level set according to the value range of each characteristic indicator, constructing a cloud model of each evaluation level of each characteristic indicator, and calculating the membership value of each characteristic indicator at each evaluation level according to the membership function of the cloud model and the indicator data of each characteristic indicator; determining the evaluation level corresponding to each characteristic indicator according to the membership value of each characteristic indicator at each evaluation level, and determining the final evaluation level of each characteristic indicator according to the evaluation level corresponding to each characteristic indicator and a preset reference level; quantifying the final evaluation level of each characteristic indicator to obtain the indicator evaluation value of each characteristic indicator; obtaining the adjustment weight of each characteristic indicator according to the initial weight and indicator data of each characteristic indicator, and constructing a comprehensive evaluation cloud model of the distribution network business expansion and installation scheme according to the cloud model corresponding to the adjustment weight and the final evaluation level of each characteristic indicator; and obtaining the final comprehensive evaluation level of the distribution network business expansion and installation scheme according to the comprehensive evaluation cloud model.

[0048] In view of this, the beneficial effects brought by the present invention are:

[0049] The present invention obtains the index data of the residential load characteristic index of the distribution network business expansion application plan, divides the evaluation results of each characteristic index into multiple evaluation levels, and establishes a cloud model of each evaluation level of each characteristic index. The digital characteristics of each cloud model include an expected value, an entropy value and a super entropy value. According to the membership function of the cloud model and the index data of each characteristic index, the membership value of each characteristic index at each evaluation level can be obtained, so as to determine the evaluation level corresponding to each characteristic index; the final evaluation level of each characteristic index is determined according to the evaluation level corresponding to each characteristic index and the preset reference level, and the final evaluation level of each characteristic index is converted from a qualitative expression to a quantitative expression, so as to obtain the precise amount of the index evaluation value of each characteristic index; the adjustment weight of each characteristic index is obtained according to the initial weight and index data of each characteristic index, and finally, the comprehensive evaluation cloud model of the distribution network business expansion application plan is constructed according to the cloud model corresponding to the adjustment weight and the final evaluation level of each characteristic index, and the final comprehensive evaluation level is obtained. The present invention can objectively and accurately evaluate the impact of residential loads on the distribution network expansion plan, which is conducive to the targeted formulation of the distribution network expansion plan, improving the matching degree of the distribution capacity between the expansion plan and the residential load, and is conducive to improving the utilization rate of the distribution network equipment and the operating economy of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 Schematic diagram of the process of the evaluation method of the present invention.

[0051] Figure 2 Schematic diagram of the flow chart of the evaluation method embodiment of the present invention. DETAILED DESCRIPTION

[0052] Terminology explanation:

[0053] Cloud model: Cloud model is the specific implementation method of cloud and the basis of cloud-based computing, reasoning and control. It can represent the process from qualitative concept to quantitative representation (forward cloud generator) or the process from quantitative representation to qualitative concept (reverse cloud generator).

[0054] The digital features of the cloud model are expressed as expected E x (Expected Value), Entropy E n (Entropy) and super entropy H e (hyperEntropy), which reflect the quantitative characteristics of qualitative concepts.

[0055] Expected E x :The expectation of the distribution of cloud droplets in the domain space is the point that best represents the qualitative concept and is the most typical sample for the quantification of this concept.

[0056] Entropy n : The concept of "entropy" was originally used as a state parameter to describe thermodynamics. It was later introduced into statistical physics, information theory, complex systems, etc. to measure the degree of uncertainty. In the cloud model, entropy represents the measurable granularity of qualitative concepts. The larger the entropy, the more macroscopic the concept is. It is also a measure of the uncertainty of qualitative concepts, which is determined by the randomness and fuzziness of the concept. On the one hand, En is a measure of the randomness of qualitative concepts, reflecting the degree of discreteness of cloud droplets that can represent this qualitative concept; on the other hand, it is a measure of the both-and-the-other nature of qualitative concepts, reflecting the range of cloud droplets that can be accepted by the concept in the domain space. Using the same numerical feature to reflect randomness and fuzziness will inevitably reflect the correlation between them.

[0057] Super Entropy H e : The uncertainty measure of entropy, that is, the entropy of entropy, is determined by the randomness and fuzziness of entropy. It reflects the cohesion of the degree to which each value belongs to this language value, that is, the cohesion of cloud droplets. The larger the super entropy, the greater the discreteness of the cloud, the greater the randomness of the membership, and the greater the thickness of the cloud.

[0058] The embodiment of the present invention provides a distribution network expansion and installation scheme evaluation method based on residential load characteristics to solve the technical problem that the existing methods lack consideration of the impact of residential loads on the distribution network expansion and installation scheme.

[0059] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. The preferred embodiments of the present invention are shown in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0061] The expansion application of the distribution network needs to consider the changes in the power composition, power nature, capacity, load distribution and load density of the user load, and specify a suitable expansion plan based on the power distribution point, network structure and substation connection of the power grid. Reasonable evaluation of power load characteristics and targeted expansion plans are conducive to improving the utilization rate of power grid equipment, improving the economic interests of power companies and users, and promoting the healthy development of the power industry.

[0062] See also Figure 1 The following is an embodiment of a method for evaluating a distribution network expansion plan based on residential load characteristics of the present invention, including:

[0063] S100: Determine the residential load characteristic index of the distribution network expansion application plan to be evaluated, and obtain index data of each characteristic index;

[0064] S200: determining an evaluation level set according to the value range of each characteristic indicator, constructing a cloud model of each evaluation level of each characteristic indicator, and calculating the membership value of each characteristic indicator at each evaluation level according to the membership function of the cloud model and the indicator data of each characteristic indicator;

[0065] S300: determining the evaluation level corresponding to each characteristic indicator according to the membership value of each characteristic indicator at each evaluation level, and determining the final evaluation level of each characteristic indicator according to the evaluation level corresponding to each characteristic indicator and a preset reference level;

[0066] S400: quantifying the final evaluation level of each characteristic indicator to obtain an indicator evaluation value of each characteristic indicator;

[0067] S500: obtaining the adjustment weight of each characteristic indicator according to the initial weight and indicator data of each characteristic indicator, and constructing a comprehensive evaluation cloud model for the distribution network expansion application plan according to the cloud model corresponding to the adjustment weight and the final evaluation level of each characteristic indicator;

[0068] S600: Obtaining a final comprehensive evaluation grade of the distribution network expansion application plan according to the comprehensive evaluation cloud model.

[0069] In this embodiment, in step S100, the residential load characteristic index of the distribution network expansion application plan to be evaluated is determined, and the index data of each characteristic index is obtained.

[0070] The process of determining the residential load characteristic indicators for the distribution network expansion plan to be evaluated is as follows:

[0071] Considering the completeness of the resident load information, the characteristic index needs to cover multiple aspects of the load. The characteristic index of the resident load selected in this embodiment includes the load index and the time domain characteristic index.

[0072] Specifically, load indicators are used to measure the residential load power consumption and guide the selection of transformer capacity, mainly including equipment indicators and regional indicators. Among them, equipment indicators mainly include: residential user daily power peak P max 、Daily electricity basic load P min The regional indicators mainly include: regional load density A1, regional annual electricity consumption growth rate A2 and distribution transformer commissioning time A3.

[0073] Among them, regional load density refers to the load size of users in a certain area per unit area, which is used to measure power consumption and guide the selection of transformer capacity. The calculation method of regional load density A1 is as follows:

[0074]

[0075] Where P is the total electricity consumption of electricity users in the region, and S is the area of ​​the region.

[0076] Regional load density is used to measure the overall electricity consumption level of a region. Under the same area, the greater the load density, the greater the electricity consumption of users in the region, and the larger the capacity of the required transformer. If the load rate of the transformer at the access point is already at a high level, it is not suitable for new registered users to access.

[0077] The regional annual electricity consumption growth rate indicates the growth rate of residential electricity consumption in the region compared with the previous year. It is used to estimate the growth rate of residential load in the region. The faster the growth rate of residential load, the greater the margin required for the access point transformer. The calculation formula for the regional annual electricity consumption growth rate A2 is as follows:

[0078]

[0079] Where P1 is the electricity consumption of residential users in the area this year, and P0 is the electricity consumption of residential users in the area last year.

[0080] The distribution transformer commissioning time A3 indicates the time when the access point transformer is put into operation. The longer the distribution transformer commissioning time is, the more careful consideration is needed when connecting new loads.

[0081] Specifically, the time domain characteristic index is used to describe the detailed characteristics of different types of residential loads. The more indicator dimensions are selected, the more detailed the description of load characteristics is. According to the time period, the time domain characteristic index is divided into: all-day index, peak and valley index, and flat index.

[0082] Among them, the full-day indicators mainly include: daily load rate B1, daily minimum load rate B2 and daily peak-to-valley difference rate B3.

[0083] The calculation formula of daily load factor B1 is as follows:

[0084]

[0085] Where P av is the daily average load, P max The daily peak load.

[0086] The calculation formula for the daily minimum load factor B2 is as follows:

[0087]

[0088] Where P min It is the daily trough load.

[0089] The calculation formula of daily peak-to-valley difference rate B3 is as follows:

[0090]

[0091] Among them, the peak and valley period indicators mainly include: peak load rate C1 and valley load rate C2.

[0092] The calculation formula of peak load factor C1 is as follows:

[0093]

[0094] Where U peak is the average peak voltage, U av is the average voltage.

[0095] The calculation formula of valley load factor C2 is as follows:

[0096]

[0097] Where U val is the average valley voltage.

[0098] Among them, the flat index mainly includes the flat load rate D, and its calculation formula is as follows:

[0099]

[0100] Where U sh is the average steady-state voltage.

[0101] In step S200, the evaluation level set is determined according to the value range of each characteristic indicator, a cloud model of each evaluation level of each characteristic indicator is constructed, and the membership value of each characteristic indicator at each evaluation level is calculated according to the membership function of the cloud model and the indicator data of each characteristic indicator.

[0102] The collected evaluation index data is input into the information matrix, and each dimension of the information matrix is ​​processed respectively, and each dimension of information corresponds to each characteristic index. The value range of each characteristic index is divided into regions according to the pre-set limit value.

[0103] The evaluation results expressed in natural language are divided into multiple evaluation levels, and all evaluation levels constitute an evaluation level set; the value range of each characteristic indicator is divided into regions according to a pre-set limit value, and each data interval corresponds to an evaluation level. It can be understood that the value range of each characteristic indicator has a corresponding evaluation level set.

[0104] For example, the evaluation results expressed in natural language are divided into five evaluation levels, namely "extremely poor", "poor", "average", "good" and "excellent", and the value range of the indicator data of a certain characteristic indicator (0,10] is also divided into five data intervals accordingly, namely (0,a], (a,b], (b,c], (c,d] and (e,10], and the evaluation levels "extremely poor", "poor", "average", "good" and "excellent" correspond to the data intervals (0,a], (a,b], (b,c], (c,d] and (e,10] respectively, that is, there is a one-to-one correspondence between the evaluation levels and the data intervals.

[0105] It can be understood that each evaluation level has its corresponding limit value, and the limit values ​​corresponding to the evaluation levels of "extremely poor", "poor", "average", "good" and "excellent" are a, b, c, d and e respectively.

[0106] It is worth noting that, since the value ranges of the indicator data of each characteristic indicator are different, the limit values ​​corresponding to each evaluation level of each characteristic indicator are also different.

[0107] After obtaining the limit values ​​of each evaluation level of each characteristic indicator, the cloud model of each evaluation level of each characteristic indicator can be constructed, which specifically includes: calculating the standard deviation of the limit values ​​according to the limit values ​​of each evaluation level of each characteristic indicator; calculating the digital features of the cloud model of each evaluation level according to the standard deviation: expected value E x , Entropy value E n and the super entropy value H eAccording to the digital characteristics of the cloud model at each evaluation level, the cloud generator is used to construct the cloud model of each evaluation level for each characteristic indicator.

[0108] It should be noted that the digital characteristics of the cloud model can be expressed as (E x , E n , H e ), representing the expected value, entropy value and super entropy value respectively.

[0109] Each evaluation level corresponds to a cloud model. For example, the digital features of the cloud models of the five evaluation levels of a certain characteristic indicator are: (-6, 0.1, 0.01), (-3, 0.1, 0.01), (0, 0.1, 0.01), (3, 0.1, 0.01), (6, 0.1, 0.01). It can be understood that the five evaluation levels correspond to five cloud models, and the expected values ​​E of these five cloud models are x The entropy values ​​E of these five cloud models are -6, -3, 0, 3, and 6 respectively. n are all 0.1. The super entropy values ​​H of these five cloud models e The five evaluation levels are all 0.01. The five evaluation levels are associated with the digital features of the five cloud models, and the results are: very poor (-6, 0.1, 0.01), poor (-3, 0.1, 0.01), average (0, 0.1, 0.01), good (3, 0.1, 0.01), and excellent (6, 0.1, 0.01).

[0110] It is worth noting that the digital characteristics of the cloud model are related to the indicator data size of the characteristic indicators, and the digital characteristics of the cloud model corresponding to each evaluation level of each characteristic indicator may be different.

[0111] The cloud models of each evaluation level for each characteristic index are constructed using the cloud generator, including:

[0112] The cloud generator is used to perform the following steps for each evaluation level of each characteristic indicator:

[0113] S1: Generate E n For expectation, H e Normal random number with standard deviation

[0114] S2: Generate E x For the expectation, A normal random number with standard deviation x i ;

[0115] Repeat steps S1 and S2 until N random numbers of the cloud model are generated, where N is the preset number of random numbers, 1≤i≤N.

[0116] It is worth noting that the normal random number is obtained according to the function normrnd for obtaining the normal distribution random number in the matlab software. Under the condition that the computing resources allow, the larger the value of N, the better. In this embodiment, the value of N is 3000.

[0117] The membership value of each characteristic indicator at each evaluation level is calculated based on the membership function of the cloud model and the indicator data of each characteristic indicator, including:

[0118] According to the membership function of the cloud model Calculate the membership value of each characteristic indicator at each evaluation level;

[0119] In the formula, k is the index data of each characteristic index, r is the expected value of k at E x , Entropy value E n and the super entropy value H e The membership value of the cloud model corresponding to the evaluation level.

[0120] It can be understood that, using the same method, according to the membership function of the cloud model and the indicator data of each characteristic indicator, the membership value of each characteristic indicator at each evaluation level can be obtained.

[0121] After obtaining the membership value of each characteristic indicator at each evaluation level, it is possible to determine to which evaluation level the indicator data of each characteristic indicator belongs according to the size of the membership value at each evaluation level. In a preferred implementation, for a certain characteristic indicator, the maximum membership value is selected from the membership values ​​of the characteristic indicator at each evaluation level, and the evaluation level corresponding to the maximum membership value is used as the evaluation level corresponding to the characteristic indicator.

[0122] It can be understood that the same method can be used to determine the evaluation levels corresponding to all characteristic indicators.

[0123] In step S300, the evaluation level corresponding to each characteristic indicator is determined according to the membership value of each characteristic indicator at each evaluation level, and the final evaluation level of each characteristic indicator is determined according to the evaluation level corresponding to each characteristic indicator and the preset reference level.

[0124] According to the designer's experience, conditional statements are compiled one by one to form a rule table for inference. The actual input values ​​of indicators of different dimensions are E j (1≤j≤n), the reference value E set in advance c The state words are converted into the corresponding evaluation levels, which are used as the rows and columns of the rule table, and the corresponding output variable state words are found. It is worth noting that the state words corresponding to the evaluation levels of extremely poor, poor, general, good, and excellent are NB, NS, ZE, PS, and PB respectively.

[0125] The steps to build a cloud inference rule base are as follows:

[0126] For example, the evaluation level set of a certain characteristic indicator is divided into five evaluation levels, and the correlation between the five evaluation levels and the digital characteristics of the five clouds is: extreme (-6, 0.1, 0.01), poor (-3, 0.1, 0.01), general (0, 0.1, 0.01), good (3, 0.1, 0.01), and excellent (6, 0.1, 0.01). Taking the equipment indicator E1 in the load indicator as an example, E1 is the actual input value of the equipment indicator, that is, E1 is the indicator data of the characteristic indicator equipment indicator, and the actual input value E1 and the reference value E c In the rule table, an evaluation level corresponding to the evaluation value Y1 represents E1×E c A mapping to Y1. Do the same mapping for the remaining dimensions, integrate them into corresponding control rules and store them in the rule base.

[0127] It is worth noting that the reference value E c It is a pre-set empirical value or reference value of the evaluation index, and the evaluation value Y1 is the evaluation grade result inferred from the E1 control rule table. The evaluation value Y1 is a rough range, and the index evaluation value in step S400 refers to the calculated specific and accurate index evaluation value.

[0128] According to experience, conditional statements are compiled and assembled into a rule table for reasoning. Input quantities of different dimensions are E1, E2, ... E n The state words are transformed into the corresponding evaluation levels, which are used as the rows and columns of the rule table, and the corresponding output variable state words are found. As shown in Table 1.

[0129] Table 1 E1 control rules table

[0130]

[0131] For example, when the actual input value E1 of the equipment indicator corresponds to the evaluation level of general (ZE), the reference value E c When the corresponding reference level is good (PS), according to the control rule table shown in Table 1, the final evaluation level of the characteristic indicator equipment category indicator is good (PS).

[0132] In this embodiment, it is necessary to establish a corresponding cloud inference rule table for each dimension of information in the information matrix. Since the information matrix has n dimensions, that is, there are n characteristic indicators, it is necessary to establish n cloud inference rule tables, and all cloud inference rule tables constitute a cloud inference rule library.

[0133] In step S400, the final evaluation level of each characteristic indicator is quantified to obtain the indicator evaluation value of each characteristic indicator.

[0134] It is worth noting that, from the above discussion, when the cloud generator is used to construct the cloud model of each evaluation level of each characteristic index, N random numbers x are generated. i , there are N random numbers in each cloud model; using the membership function of the cloud model, the membership values ​​of the N random numbers in the cloud model at the corresponding evaluation level can be obtained respectively. Therefore, using the membership function of the cloud model, the membership values ​​of the N random numbers at the final evaluation level of each characteristic indicator can be obtained.

[0135] Specifically, after determining the final evaluation level of each characteristic indicator, the cloud model corresponding to the final evaluation level of each characteristic indicator can be determined, and the membership function of the cloud model can be used to calculate the cloud model. You can get the random number x of the cloud model i The membership value at the final evaluation level of each characteristic index is repeated N times, and the membership values ​​of N random numbers at the final evaluation level of each characteristic index can be obtained.

[0136] It can be understood that the cloud model corresponding to the final evaluation level of each characteristic indicator includes N random numbers, where each random number x i Under the final evaluation level, there is a corresponding membership value y i , the final evaluation level corresponds to the expected value of the cloud model and each random number x i The corresponding membership value y i Combine them to get cloud droplets.

[0137] The expected value of the corresponding adjustment parameter in the cloud inference rule table is C0. Let drop(C0,y i ) is a cloud drop, cloud drop drop(C0,y i ) is the expected value C0 and the membership degree y i The obtained N cloud droplets are sent to the inverse cloudification processor, which performs inverse cloudification operations on the cloud droplets and calculates the average value to obtain the adjustment amount of the index evaluation value. The inverse cloudification processor converts the fuzzy quantity of the cloudification concept into the precise quantity of control.

[0138] It is worth noting that C0 is the expected value of the cloud model corresponding to the final evaluation level of the characteristic indicator. For example, when the correlation between the five evaluation levels and the digital characteristics of the five cloud models is: extremely poor (-6, 0.1, 0.01), poor (-3, 0.1, 0.01), average (0, 0.1, 0.01), good (3, 0.1, 0.01), and excellent (6, 0.1, 0.01), if the final evaluation level of the equipment type indicator in the characteristic indicator load indicator is good (PS), then the final evaluation level of the equipment type indicator in the characteristic indicator load indicator corresponds to the expected value of the cloud model is 3, that is, the value of C0 is 3.

[0139] The final evaluation level of each characteristic indicator is quantified to obtain the indicator evaluation value of each characteristic indicator. The specific process is as follows:

[0140] According to the formula Calculate the correction value Δc i ;

[0141] According to formula C i =C0+Δc i Get the cloud drop drop(C0,y i ) is the exact value of the indicator evaluation value,

[0142] C0 is the expected value of the cloud model corresponding to the final evaluation level of each characteristic indicator, E n The entropy value of the cloud model for the final evaluation level of each characteristic indicator;

[0143] The precise quantities of the index evaluation values ​​of N cloud droplets are integrated and averaged, and the average value obtained is the index evaluation value of each characteristic index.

[0144] In step S500, the adjustment weight of each characteristic indicator is obtained according to the initial weight and indicator data of each characteristic indicator, and a comprehensive evaluation cloud model for the distribution network expansion application plan is constructed according to the cloud model corresponding to the adjustment weight and the final evaluation level of each characteristic indicator.

[0145] Specifically, the variable weight method is introduced to calculate the adjustment weights of various characteristic indicators under the distribution network expansion and installation plan evaluation index system.

[0146] The indicator weight calculated based on the characteristic index represents the amount of information of each characteristic index, which can provide a reference value for the evaluation and analysis of the distribution network expansion plan. In order to make the overall evaluation results of the distribution network expansion truly reflect the actual situation, the improved variable weight method is introduced to process the initial weights to obtain the final adjustment weights, so as to avoid the impact on the evaluation results when the sample data of the residential load characteristic index is abnormal.

[0147] The process of calculating the adjustment weights of each indicator in the evaluation of distribution network expansion application is as follows:

[0148] The initial weights of the characteristic indicators for distribution network expansion are:

[0149]

[0150] Using the balance function The initial weights and the indicator data of each characteristic indicator are processed to obtain the adjustment weights of each characteristic indicator:

[0151]

[0152] In the formula, α is the balance factor, 0<α≤1, xj is the normalized value of the jth feature index, j = 1, 2, …, n, n is the number of feature indexes.

[0153] When the evaluator is more conservative, that is, when α is less than 1 / 2, more consideration will be given to the balance of many factors; when the evaluator is more open-minded, when α is greater than 1 / 2, the defects or deficiencies of a certain factor can be tolerated more easily. In this embodiment, a compromise is taken as α=1 / 2.

[0154] Based on the characteristics of residential loads, a comprehensive evaluation cloud model for the distribution network expansion application plan is determined, and the comprehensive evaluation cloud model can represent the final comprehensive evaluation level of the distribution network expansion application plan.

[0155] The distribution network expansion plan based on the characteristics of residential loads takes into account the mutual influence between the dimensions of the characteristic indicators of the distribution network expansion plan. By using the inverse cloud processor and combining the adjustment weights of the characteristic indicators, the digital characteristics of the comprehensive evaluation cloud model of the distribution network expansion plan can be calculated as follows:

[0156]

[0157]

[0158]

[0159] In the formula, ω′ j is the adjustment weight of the jth feature index, satisfying are the expected value, entropy value and super entropy value of the cloud model corresponding to the final evaluation level of the j-th characteristic indicator.

[0160] Using the above formula, the actual cloud model parameters of the multi-dimensional comprehensive evaluation index system for distribution network expansion can be calculated, which represents the final comprehensive evaluation level of the distribution network expansion plan based on the characteristics of residential loads. After obtaining the comprehensive evaluation cloud model of the distribution network expansion plan, the final comprehensive evaluation level of the distribution network expansion plan can be determined based on the digital characteristics of the comprehensive evaluation cloud model (i.e., expected value, entropy value, and super entropy value) and the corresponding relationship between the digital characteristics of the cloud model and the evaluation level.

[0161] The embodiment of the present invention obtains the index data of the residential load characteristic index of the distribution network business expansion application plan, divides the evaluation results of each characteristic index into multiple evaluation levels, and establishes a cloud model of each evaluation level of each characteristic index. The digital characteristics of each cloud model include an expected value, an entropy value and a super entropy value. According to the membership function of the cloud model and the index data of each characteristic index, the membership value of each characteristic index at each evaluation level can be obtained, thereby determining the evaluation level corresponding to each characteristic indicator; according to the evaluation level corresponding to each characteristic indicator and the preset reference level, the final evaluation level of each characteristic indicator is determined, and the final evaluation level of each characteristic indicator is converted from a qualitative expression to a quantitative expression, so as to obtain the precise amount of the index evaluation value of each characteristic indicator; according to the initial weight and index data of each characteristic indicator, the adjustment weight of each characteristic indicator is obtained, and finally, according to the adjustment weight and the cloud model corresponding to the final evaluation level of each characteristic indicator, a comprehensive evaluation cloud model of the distribution network business expansion application plan is constructed, and the final comprehensive evaluation level is calculated. The present invention can objectively and accurately evaluate the impact of residential loads on the distribution network expansion plan, which is conducive to the targeted formulation of the distribution network expansion plan, improving the matching degree of the distribution capacity between the expansion plan and the residential load, and is conducive to improving the utilization rate of the power grid equipment and the operating performance of the distribution network.

[0162] The embodiment of the present invention is based on the load characteristics of residential users, approximately simulates the electricity consumption behavior of users, and establishes a mapping from residential users' electricity consumption data to characteristic indicators. The cloud model is used to process the residential load data, calculate the corresponding membership of each dimension of the input data, and distribute the input quantity to each evaluation interval according to the membership. By defining the cloud processor rule base, the mapping from initial data to evaluation value can be achieved. The weight of each evaluation indicator is determined by introducing the variable weight method, taking into account the subjective and objective factors when calculating the weight, so that the indicator weight changes with the state value, reducing the unfair impact of individual cognitive differences on the evaluation results, and improving the scientificity and credibility of the distribution network expansion evaluation results. The present invention can reasonably divide the residential electricity load, improve the matching degree between the decision-making plan and the reported load, and can efficiently process a variety of information. It is more efficient than traditional methods, making the final decision plan more reliable.

[0163] The embodiment of the present invention takes into account the integrity of residential electricity load information and formulates reasonable characteristic indicators of user loads to describe the load characteristics of residential loads. Fuzzy clustering is performed based on the characteristic indicators of the load to obtain a fuzzy evaluation value of the indicator, which is input into the cloud model to obtain the evaluation value. The evaluation value is standardized, and the variable weight method is introduced to calculate the variable weight of each indicator under the distribution network business expansion and installation evaluation indicator system. A multi-dimensional comprehensive evaluation cloud model for distribution network business expansion and installation based on the variable weight method is established, and the actual final result of the comprehensive evaluation of the distribution network business expansion and installation access plan based on the characteristics of the residential load can be obtained. According to the final result of the comprehensive evaluation, the advantages and disadvantages of the distribution network business expansion and installation access plan can be intuitively judged.

[0164] The embodiment of the present invention obtains relevant indicator data required for evaluation from aspects such as load indicators and time domain performance indicators of residential loads, summarizes and fills them into an n-dimensional information matrix, performs cloudization on the n-dimensional information, represents it with three characteristics of the cloud model, establishes cloud inference rules, respectively establishes cloud mappings from initial values ​​to inference values ​​for each dimension of the input information, performs cloud inference and output inverse cloudization, obtains the actual value of the evaluation quantity, constructs a comprehensive evaluation cloud model for the distribution network expansion application plan by adjusting the cloud model corresponding to the final evaluation level of each characteristic indicator, and obtains the final evaluation level based on the comprehensive evaluation cloud model.

[0165] See also Figure 2 Another embodiment of the method for evaluating a distribution network expansion application plan based on residential load characteristics of the present invention determines the residential load characteristic indicators of the distribution network expansion application plan to be evaluated by collecting the information of the registered users, obtains the indicator data of each characteristic indicator, and constructs a residential user load characteristic library; determines the evaluation level set according to the value range of each characteristic indicator, and constructs a cloud model of each evaluation level of each characteristic indicator; establishes a cloud reasoning processing rule library, performs cloud reasoning and outputs inverse cloudization, obtains the expected value corresponding to the final evaluation level of each characteristic indicator, obtains the adjustment weight of each characteristic indicator according to the initial weight and indicator data of each characteristic indicator, and finally obtains the final comprehensive evaluation level result of the distribution network expansion application plan and outputs it.

[0166] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0167] In the embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0168] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0169] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0170] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0171] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating distribution network expansion plans based on residential load characteristics, characterized in that: include: Determine the residential load characteristic indicators of the distribution network expansion application plan to be evaluated, and obtain the indicator data of each characteristic indicator; Determine the evaluation level set according to the value range of each characteristic indicator, divide the value range of each characteristic indicator into regions according to the set limit value, each data interval corresponds to an evaluation level, and all evaluation levels constitute the evaluation level set; construct a cloud model of each evaluation level of each characteristic indicator, and calculate the standard deviation of the limit value according to the limit value of each evaluation level of each characteristic indicator; calculate the digital characteristics of the cloud model of each evaluation level of each characteristic indicator according to the standard deviation: expected value E x , Entropy value E n and the super entropy value H e ; According to the digital features of the cloud model of each evaluation level of each characteristic indicator, the cloud model of each evaluation level of each characteristic indicator is constructed by using the cloud generator respectively; the membership value of each characteristic indicator at each evaluation level is calculated according to the membership function of the cloud model and the indicator data of each characteristic indicator; Determine the evaluation level corresponding to each characteristic indicator according to the membership value of each characteristic indicator at each evaluation level, and determine the final evaluation level of each characteristic indicator according to the evaluation level corresponding to each characteristic indicator and the preset reference level; Quantify the final evaluation level of each characteristic indicator to obtain the indicator evaluation value of each characteristic indicator; According to the initial weight and indicator data of each characteristic indicator, the adjustment weight of each characteristic indicator is obtained, and according to the cloud model corresponding to the adjustment weight and the final evaluation level of each characteristic indicator, a comprehensive evaluation cloud model of the distribution network expansion application plan is constructed; Obtaining the final comprehensive evaluation level of the distribution network expansion application plan according to the comprehensive evaluation cloud model; Among them, the cloud models of each evaluation level of each characteristic indicator are constructed by using the cloud generator, specifically including: The cloud generator is used to perform the following steps for each evaluation level of each characteristic indicator: S1: Generate E n For expectation, H e Normal random number with standard deviation ; S2: Generate E x For expectation, Normal random number with standard deviation ; Repeat steps S1 and S2 until N random numbers of the cloud model are generated, where N is the preset number of random numbers, 1≤i≤N; The calculation of the membership value of each characteristic indicator at each evaluation level according to the membership function of the cloud model and the indicator data of each characteristic indicator specifically includes: According to the membership function of the cloud model Calculate the membership value of each characteristic indicator at each evaluation level; In the formula, is the indicator data of each characteristic indicator, for In the expected value E x , Entropy value E n and the super entropy value H e The membership value of the cloud model corresponding to the evaluation level; Among them, the evaluation level corresponding to each characteristic indicator is determined according to the membership value of each characteristic indicator at each evaluation level, specifically including: Select the maximum membership value from the membership values ​​of each characteristic indicator at each evaluation level, and use the evaluation level corresponding to the maximum membership value as the evaluation level corresponding to each characteristic indicator; Among them, the final evaluation level of each characteristic indicator is quantified, and the indicator evaluation value of each characteristic indicator is obtained, which specifically includes: According to the formula Obtain the correction amount ; According to formula C i =C0+ Get the cloud drop (C0, ), C0 is the expected value of the final evaluation level of each characteristic indicator corresponding to the cloud model, E n is the entropy value of the cloud model for the final evaluation level of each characteristic index, is a random number The membership value at the final evaluation level of each characteristic indicator; The precise values ​​of the index evaluation values ​​of N cloud droplets are averaged to obtain the index evaluation value of each characteristic index; Among them, the adjustment weight of each characteristic indicator obtained according to the initial weight of each characteristic indicator and the indicator data specifically includes: Using the balance function For the initial weight And the indicator data of each characteristic indicator are processed to obtain the adjustment weight of each characteristic indicator: ; In the formula, is the equilibrium factor, , is the normalized value of the jth feature index, , n is the number of characteristic indicators; Among them, the comprehensive evaluation cloud model of the distribution network expansion application plan is: ; ; ; In the formula, is the adjustment weight of the jth feature index, satisfying ; , , are the expected value, entropy value and super entropy value of the cloud model corresponding to the final evaluation level of the j-th characteristic indicator.

2. The method for evaluating distribution network expansion and installation plans based on residential load characteristics according to claim 1 is characterized in that: The specific indicators for determining the residential load characteristics of the distribution network expansion plan to be evaluated include: Determine the load index and time domain characteristic index of the residential load of the distribution network expansion application plan to be evaluated; Among them, the load indicators include equipment indicators and regional indicators; the time domain characteristic indicators include all-day indicators, peak-valley indicators and flat-time indicators.