CN112651651A - Investment benefit evaluation method based on feature extraction and lasso regression - Google Patents
Investment benefit evaluation method based on feature extraction and lasso regression Download PDFInfo
- Publication number
- CN112651651A CN112651651A CN202011624938.2A CN202011624938A CN112651651A CN 112651651 A CN112651651 A CN 112651651A CN 202011624938 A CN202011624938 A CN 202011624938A CN 112651651 A CN112651651 A CN 112651651A
- Authority
- CN
- China
- Prior art keywords
- index
- investment
- incremental
- reliability
- formula
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000008901 benefit Effects 0.000 title claims abstract description 28
- 238000011156 evaluation Methods 0.000 title claims abstract description 20
- 238000000605 extraction Methods 0.000 title claims abstract description 12
- 238000010276 construction Methods 0.000 claims abstract description 32
- 238000000034 method Methods 0.000 claims abstract description 22
- 238000012216 screening Methods 0.000 claims abstract description 18
- 150000001875 compounds Chemical class 0.000 claims description 12
- 230000000694 effects Effects 0.000 claims description 5
- 238000012417 linear regression Methods 0.000 claims description 5
- 238000012935 Averaging Methods 0.000 claims description 3
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 claims description 2
- 125000004432 carbon atom Chemical group C* 0.000 description 6
- 238000013210 evaluation model Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008092 positive effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/06—Asset management; Financial planning or analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Technology Law (AREA)
- Quality & Reliability (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention provides an investment benefit evaluation method based on feature extraction and lasso regression. The method comprises the steps of constructing an investment index and power supply reliability index feature pool, screening investment indexes associated with each reliability index, constructing a lasso regression association model based on power supply reliability indexes and investment index combinations obtained through screening, and evaluating investment benefits based on the lasso regression association model. The method can realize accurate and objective evaluation of benefits of power distribution network investment projects, and guarantee high efficiency and accuracy of power grid investment construction.
Description
Technical Field
The invention relates to the field of electric power, in particular to an investment benefit evaluation method based on feature extraction and lasso regression.
Background
The investment construction of the power distribution network is an important measure for guaranteeing the power consumption requirements of industry and residents. With the continuous increase of the investment amount of each level of project, the benefit generated by each investment project is objectively and reasonably evaluated, and the method is an important working link for promoting the development of power grid construction towards a scientific, environment-friendly and sustainable direction. By evaluating the reliability benefits of the investment construction of the power distribution network, the improvement degree of different investment projects on the reliability of the power distribution network can be quantitatively found, core investment projects for effectively improving the reliability of the power distribution network are found, the effect of the existing investment construction projects of the power distribution network is retrospectively summarized, a specific decision reference basis is provided for future investment construction planning, the effectiveness of investment can be further optimized, the reliability of power supply is improved, the investment construction cost of the power distribution network is controlled, and meanwhile the safety performance and the economic benefits of the power distribution network are improved.
The research in the aspect of evaluation of investment and construction benefits of the power distribution network at the present stage focuses on an evaluation method mainly qualitative and quantitatively assisted, and relates to multiple dimensions such as economic benefits, environmental benefits and social benefits, the main methods include an analytic hierarchy process, an expert evaluation method, a gray level model and the like, but some methods involve excessively numerous and complicated index systems and excessively large evaluation dimensions, so that the evaluation method has excessively large volatility and the evaluation result may have distortion or unstable conditions; most methods lack sufficient objectivity and have more subjective evaluation standards, and even though expert opinions are adopted, the methods are still influenced by the knowledge, experience and preference of experts and have greater contingency; in general, the existing method has insufficient utilization degree of objective data information, and a specific evaluation model with pertinence is difficult to obtain. The few methods for evaluation based on historical investment data are limited by insufficient effective data volume, excessive indexes to be evaluated and the like, and a comprehensive evaluation model is difficult to form.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides an investment benefit evaluation method based on feature extraction and lasso regression. In order to achieve the purpose of the invention, the technical scheme of the invention is as follows.
An investment benefit assessment method based on feature extraction and lasso regression comprises the following steps:
constructing an investment index for power distribution network construction and a reliability index characteristic pool for power supply of the power distribution network;
screening investment indexes associated with each reliability index;
constructing a lasso regression correlation model based on the reliability index and investment index combination obtained by screening;
and evaluating the investment benefit based on the lasso regression correlation model.
Preferably, the constructing of the investment index and power supply reliability index feature pool includes:
an arithmetic incremental investment index delta x is constructed by adopting a formula (1) and a formula (2)iProportional incremental investment index
In the formula (I), the compound is shown in the specification,
for the data value of the standardized investment index in the ith year,
is a data value i-1 year of the standardized investment index>1；
An arithmetic incremental reliability index delta y is constructed by adopting a formula (3) and a formula (4)iSum-ratio type incremental reliability index
In the formula (I), the compound is shown in the specification,
for the data value of the normalized reliability index of the i-th year,
the data value of the standardized reliability index in the i-1 th year is obtained;
and (3) carrying out intra-group standardization treatment on the arithmetic incremental investment index by adopting a formula (5):
in the formula, xiTo select the data value for the index year i,
is the mean value, σXIs the standard deviation.
Preferably, the screening the investment index associated with each reliability index comprises:
calculating the Pearson correlation coefficient between the non-incremental reliability index and the non-incremental investment index by adopting a formula (6):
in the formula (I), the compound is shown in the specification,
a data value representing a normalized non-incremental investment index,
a data value representing a normalized non-incremental reliability indicator;
calculating the Pearson correlation coefficient between the arithmetic incremental reliability index and the arithmetic incremental investment index by adopting a formula (7):
in the formula,. DELTA.xjData value, Δ y, representing an arithmetic incremental investment indexiA data value representing an arithmetic type incremental reliability index,
represents the average of the arithmetic incremental reliability indicators,
means representing an arithmetic incremental investment index;
calculating the Pearson correlation coefficient between the arithmetic incremental reliability index and the proportional incremental investment index by adopting a formula (8):
in the formula (I), the compound is shown in the specification,
data value, Δ y, representing a proportional incremental investment indexiA data value representing an arithmetic type incremental reliability index,
represents the average of the arithmetic incremental reliability indicators,
representing an average of the proportional incremental investment indicators;
calculating the Pearson correlation coefficient between the proportional incremental reliability index and the arithmetic incremental investment index by adopting a formula (9):
in the formula,. DELTA.xjA data value representing an arithmetic incremental investment index,
a data value representing a proportional-type incremental reliability indicator,
represents the average of the proportional-type incremental reliability indicators,
means representing an arithmetic incremental investment index;
calculating the Pearson correlation coefficient between the proportional incremental reliability index and the proportional incremental investment index by adopting a formula (10), wherein the formula is as follows:
in the formula (I), the compound is shown in the specification,
a data value representing a proportional incremental investment index,
a data value representing a proportional-type incremental reliability indicator,
represents the average of the proportional-type incremental reliability indicators,
representing an average of the proportional incremental investment indicators;
according to the Pearson correlation coefficient, selecting a reliability index and an investment index combination of which R is more than or equal to R, wherein,
and R is a preset threshold value.
Preferably, the constructing of the lasso regression correlation model based on the combination of the reliability index and the investment index obtained by screening includes:
constructing a first model (11):
Y＝α0+α1X1+α2X2+…+αpXp+ε (11)
wherein Y is a reliability index, XiFor the ith investment index, alpha, corresponding to the reliability indexiFor the weight coefficient of the influence of the investment index on the reliability index, i is 1, …, p, α0The power distribution network reflecting the data source has reliability index weight, and epsilon is other index model not included but corresponding to reliability indexMarking Y as an index and an error with potential influence;
averaging the linear regression models (11), and eliminating the unobservable term epsilon to obtain a second model (12):
E(Y)＝α0+α1X1+α2X2+…+αpXp (12)
wherein E (Y) is an average value of the reliability index;
forming sample data array by n data values of p investment indexes
In the formula, xijJ-th data value representing an investment index i;
the n data values of the reliability index corresponding to the p investment indexes form a reliability index vector:
in the formula, yiAn ith data value representing the selected reliability indicator.
Lasso estimation value for calculating investment index weight coefficient
Controlling the estimation of the weighting coefficients of the investment indicators by means of a band linearity constraint (15)
Wherein t is more than or equal to 0 as an adjusting parameter;
calculating goodness of fit R of the lasso regression correlation model by adopting formula (16)e；
In the formula (I), the compound is shown in the specification,
is the i-th term of the reliability index model value based on the lasso regression correlation model.
Preferably, if the goodness of fit is greater than a preset value, the following evaluation is performed: if the lasso estimation value of the influence weight coefficient of the investment index on the reliability index is 0, the corresponding investment index has no influence on the considered reliability index; if the lasso estimation value of the influence weight coefficient of the investment index on the reliability index is not 0, the corresponding investment index has a reverse or forward effect on the size of the considered reliability index.
Compared with the prior art, the invention has the beneficial technical effects that:
1. according to the method, an investment index system with potential influence on the reliability index is selected based on three factors of relevance between the investment index and the reliability index, continuity of actual project construction planning, specialty of expert opinions and the like, uncertainty of a relevance model caused by less historical data can be avoided, and objectivity and accuracy of power distribution network investment construction project evaluation are improved.
2. The evaluation model disclosed by the invention can be used for establishing an increment index system comprising arithmetic type increment and proportional type increment by determining the investment index of key construction projects of the power distribution network and selecting the key reliability index, so that the characteristics of the existing index system can be effectively extracted, and the evaluation process and the model significance degree are improved.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application are clearly and completely described below, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all embodiments.
The investment benefit evaluation method based on feature extraction and lasso regression comprises the following steps:
determining investment indexes of key investment projects of the power distribution network;
selecting a power supply reliability index of the power distribution network;
collecting historical data of construction project investment and power supply reliability indexes of a power distribution network of a power enterprise; constructing an incremental investment and reliability index feature pool based on historical data;
according to historical data, aiming at each reliability index, screening possible investment indexes related to the reliability index by calculating a Pearson correlation coefficient;
and constructing a lasso regression correlation model based on the reliability index and investment index combination obtained by screening.
The key construction investment indexes of the power distribution network cover 10 indexes which can represent the investment project requirements of the power distribution network most, and the power distribution network provides power for newly outgoing lines of the transformer substation to meet newly added loads; solving the heavy overload of the medium-voltage line; the capability of coping with natural disasters is improved; industrial expansion engineering; perfecting a medium-voltage net rack; network distribution automation; replacing the old equipment or modifying the line; the problems of heavy overload of distribution transformer and low voltage of transformer area are solved; the newly-built platform area meets the requirement of load increase; and (5) total investment.
The selected power distribution network power supply reliability indexes cover 4 power supply reliability indexes which are obviously influenced by power distribution network construction projects directly or indirectly: average power off time of a client; the average prearranged power failure time of the customers; average fault outage time of customers; medium voltage line failure rate.
The standardization processing and the incremental index construction are realized by the following methods:
collecting historical data of construction project investment and power supply reliability indexes of a power distribution network of a power enterprise;
screening investment project indexes with delayed effectiveness of investment benefits, and constructing an increase index, wherein the formula is as follows:
in the above formula, the first and second carbon atoms are,
for the data value of the standardized investment index in the ith year,
for the data value of the standardized investment index from the i-1 year, the incremental index needs to start from the 2 nd year due to no index value of the 0 th year, and delta xiAn arithmetic type incremental investment index is represented,
and (4) expressing a proportional incremental investment index.
Selecting the power failure time of a user as a reliability index of the power distribution network, and constructing an incremental reliability index, wherein the formula is as follows:
in the above formula, the first and second carbon atoms are,
for the data value of the normalized reliability index of the i-th year,
for the normalized data value of the reliability index of the i-1 year, the incremental index needs to start from the 2 nd year, delta y, because of no index value of the 0 th yeariAn arithmetic type incremental reliability index is represented,
a proportional incremental reliability indicator is indicated.
The method comprises the following steps of respectively carrying out intragroup standardization treatment on each original investment index and each arithmetic incremental investment index, wherein the standardization formula is as follows:
in the above formula, xiTo select the data value for the index year i,
is the average of several years of data of the index, sigmaXThe standard deviation of the data of a plurality of years of the index is calculated by the following formula:
meanwhile, the screening method of the investment index having potential correlation with the reliability index in the step 4 is carried out by calculating the Pearson correlation coefficient, and the specific implementation method is divided into the following steps:
calculating the Pearson correlation coefficient between the non-incremental reliability index and the non-incremental investment index, wherein the formula is as follows:
in the above formula, the first and second carbon atoms are,
a data value representing a normalized non-incremental investment index j,
a data value representing the normalized non-incremental reliability index i.
Calculating the Pearson correlation coefficient between the arithmetic incremental reliability index and the arithmetic incremental investment index, wherein the formula is as follows:
in the above formula,. DELTA.xjData value, Δ y, representing an arithmetic incremental investment index jiA data value representing an arithmetic type incremental reliability index i,
represents the average of an arithmetic incremental reliability index i,
the mean of the arithmetic incremental investment index j is shown.
Calculating the Pearson correlation coefficient between the arithmetic incremental reliability index and the proportional incremental investment index, wherein the formula is as follows:
in the above formula, the first and second carbon atoms are,
data value, Δ y, representing a proportional incremental investment index jiA data value representing an arithmetic type incremental reliability index i,
represents the average of an arithmetic incremental reliability index i,
the average value of the proportional incremental investment index j is shown.
Calculating the Pearson correlation coefficient between the proportional incremental reliability index and the arithmetic incremental investment index, wherein the formula is as follows:
in the above formula,. DELTA.xjA data value representing an arithmetic incremental investment index j,
a data value representing a proportional-type incremental reliability index i,
represents the average value of the proportional-type incremental reliability index i,
the mean of the arithmetic incremental investment index j is shown.
Calculating the Pearson correlation coefficient between the proportional incremental reliability index and the proportional incremental investment index, wherein the formula is as follows:
in the above formula, the first and second carbon atoms are,
a data value representing a proportional incremental investment index j,
a data value representing a proportional-type incremental reliability index i,
represents the average value of the proportional-type incremental reliability index i,
the average value of the proportional incremental investment index j is shown.
Selecting a reliability index and an investment index combination, wherein r is more than or equal to 0.4 according to the five groups of Pearson correlation coefficients obtained by calculation and according to actual requirements, and the threshold can be adjusted according to the total planning item number of the construction of the power distribution network, wherein
If the selected index combination does not contain some investment indexes which are already included in the construction plan and have long benefit generation period, or other investment indexes which are recommended by experts but are not selected in the selected index combination exist, the additional selection of the index combination is carried out.
Constructing a lasso regression correlation model between the reliability indexes and the investment index combinations, wherein a specific model algorithm is as follows:
construction of the following Linear regression model
Y＝α0+α1X1+α2X2+…+αpXp+ε
In the above formula, Y is the selected reliability index (non-incremental or incremental), XiFor the ith investment index, alpha, corresponding to the reliability indexiFor the weight coefficient of the influence of the investment index on the reliability index, i is 1, …, p, α0The power distribution network reflecting the data source has reliability index weight, and epsilon is other indexes and errors which are not included in the index model but have potential influence on the reliability index Y.
Averaging the linear regression models, eliminating the unobservable term epsilon to obtain the following model
E(Y)＝α0+α1X1+α2X2+…+αpXp
In the above formula, e (y) is an average value of the reliability indexes, and reflects an average reliability state of the operation of the power distribution network.
The following sample data array is formed by n data values of p investment indexes
In the above formula, xijA j-th data value representing an investment index i.
(4) The n data values of the reliability index corresponding to the p investment indexes in (1) constitute a reliability index vector,
in the above formula, yiAn ith data value representing the selected reliability indicator.
Calculating lasso estimation values of each investment index weight coefficient of linear regression equation
Quadratic programming problem with linear constraints by solving
In the above formula, t is not less than 0 as an adjustment parameter, and the estimation of the weight coefficient of the investment index is controlled. The above optimization problem can be solved quickly by a conventional quadratic programming method or a minimum angle regression algorithm.
And (3) evaluating the goodness of fit of the correlation model, wherein the formula is as follows:
in the above formula, the first and second carbon atoms are,
representing a reliability index model value based on the correlation model,
is composed of
The ith component of (a). ReWhen the value is more than or equal to 0.1, the threshold value of 0.1 can be combined with the expert meaning according to the actual data volume and the investment index volumeAnd if the correlation model is adjusted, the correlation model can be regarded as describing to reach the required precision. If the lasso estimation value of the influence weight coefficient of the investment index on the reliability index is 0, the corresponding investment index has no influence on the considered reliability index; if the lasso estimation value of the influence weight coefficient of the investment index on the reliability index is negative (positive), the corresponding investment index has a reverse (positive) effect on the size of the considered reliability index.
Constructing a correlation model between investment construction and reliability indexes of the power distribution network, and depending on construction and screening of key core indexes; the influence degree of the investment index on the reliability index is reflected and is estimated by the weight coefficient of the correlation model. The invention provides a power distribution network construction investment benefit evaluation model based on feature extraction and lasso regression, which is characterized in that arithmetic type increment indexes and proportional type increment indexes are used for construction, then a Pearson coefficient is used for screening a reliability index and an investment index combination, and then the lasso regression is used for training a correlation model, so that the accurate and objective evaluation of the benefits of power distribution network investment projects is realized, and the high efficiency and accuracy of power distribution network investment construction are ensured.
Illustratively, the newly outgoing line of the transformer substation is selected as the object to be evaluated for the construction investment benefit of the power distribution network in the embodiment of the invention to meet the requirement of newly added load power supply; solving the heavy overload of the medium-voltage line; industrial expansion engineering; perfecting a medium-voltage net rack; network distribution automation; replacing the old equipment or modifying the line; the problems of heavy overload of distribution transformer and low voltage of transformer area are solved; the new platform building area satisfies 8 construction investments such as load increase, that is, 8 objects to be evaluated in this embodiment, X respectively1，…,X8。
And establishing a power distribution network construction investment index system. The newly outgoing line of the transformer substation meets the requirement of newly increased load power supply; solving the heavy overload of the medium-voltage line; industrial expansion engineering; perfecting a medium-voltage net rack; network distribution automation; replacing the old equipment or modifying the line; the problems of heavy overload of distribution transformer and low voltage of transformer area are solved; the newly-built platform area meets 8 investments such as load increase and the like, and then an investment index system is formed.
And establishing a power distribution network power supply reliability index system. And selecting the average fault power failure time of the client as a reliability index.
Collecting historical data of construction project investment and power supply reliability indexes of a power distribution network of a power enterprise; the historical data of 8 subjects to be evaluated is shown in table 1.
TABLE 1
| Object of the project | 2014 | 2015 years | 2016 (year) | 2017 |
| New outgoing line of transformer substation meets new load power supply | 7921.47 | 6679.86 | 11289.2 | 18759.68 |
| Solving heavy overload of medium-voltage line | 141.32 | 363.89 | 685.49 | 955 |
| Business expansion project | 1085 | 3815 | 14094.8 | 30842.6 |
| Perfecting medium-pressure net rack | 3773.74 | 9379.5 | 17048.09 | 21790.97 |
| Distribution network automation | 1088.65 | 1775.96 | 1608.69 | 7983.57 |
| Replacement of worn equipment or line modifications | 851.81 | 988.78 | 1204.74 | 4466.16 |
| Solves the problems of heavy overload and low voltage in distribution transformer area | 2166.55 | 4026.32 | 5681.14 | 3896.48 |
| Newly-built platform area satisfying load increase | 4971.14 | 5344.79 | 5406.7 | 8204.9 |
The historical data of the average customer outage time Y is shown in table 2.
TABLE 2
| Index name | 2014 | 2015 years | 2016 (year) | 2017 |
| Mean time to failure of customer | 0.43 | 0.3 | 0.82 | 0.83 |
And respectively carrying out standardization processing on each index, and constructing an arithmetic type increment index and a proportional type increment index. And 4, screening possible related investment indexes by calculating a Pearson correlation coefficient to obtain the relation that 8 investment indexes and the average power failure time of a client meet the requirement of a model.
Taking the non-incremental reliability index and the non-incremental investment index, and the proportional incremental reliability index and the non-incremental investment index as examples, a lasso regression correlation model is constructed, and the weight coefficient estimation values of the objects to be evaluated in each model are shown in table 3.
TABLE 3
It can be found that the goodness of fit of the correlation model established by lasso regression meets the requirements.
The coefficients of the new outgoing line of the transformer substation meeting the requirements of newly increased load power supply and the industrial expansion project are both larger than 0, which indicates that the current two investments have negative influence on the non-incremental reliability index; the coefficients of 'distribution network automation' and 'solving the problems of distribution variable weight overload and transformer area low voltage' are all less than 0, which indicates that the current two investments have positive effects on non-incremental reliability indexes. The investment of 'the new outgoing line of the transformer substation meets the requirement of newly added load power supply' has positive effect on the proportional increment reliability index. The investment of "distribution network automation" has a negative impact on the comparative incremental reliability indicators. The investment of 'solving the problems of heavy distribution and overload and low voltage of transformer area' has negative influence on the scale increment reliability index.
The model for evaluating the investment benefit of power distribution network construction based on feature extraction and lasso regression in the embodiment determines the core indexes of key construction investment of the power distribution network and selects the key indexes of power supply reliability, standardizes the investment indexes and the reliability indexes based on historical investment and power failure data, further, an arithmetic type increment index and a proportional type increment index are constructed, the matching screening of the reliability index and the investment index combination is carried out based on the Pearson correlation coefficient, and combines with expert experience, actual project overall planning arrangement and the like to determine a final reliability index and an investment index pool, constructs a correlation model of the investment index and the power failure reliability index through lasso regression, and finally obtaining a core investment index combination which obviously influences the reliability index and a degree weight coefficient of each investment project for improving the reliability index by means of the compression coefficient characteristic of lasso regression.
The above-mentioned embodiments are only specific embodiments of the present application, and are used for illustrating the technical solutions of the present application, but not limiting the same, and the scope of the present application is not limited thereto, and although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope disclosed in the present application; such modifications, changes or substitutions do not depart from the spirit and scope of the present disclosure, which should be construed in light of the above teachings. Are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (5)
1. An investment benefit assessment method based on feature extraction and lasso regression is characterized by comprising the following steps:
constructing an investment index for power distribution network construction and a reliability index characteristic pool for power supply of the power distribution network;
screening investment indexes associated with each reliability index;
constructing a lasso regression correlation model based on the reliability index and investment index combination obtained by screening;
and evaluating the investment benefit based on the lasso regression correlation model.
2. The method of claim 1, wherein the constructing a feature pool of investment index and power supply reliability index comprises:
an arithmetic incremental investment index delta x is constructed by adopting a formula (1) and a formula (2)iProportional incremental investment index
In the formula (I), the compound is shown in the specification,
for the data value of the standardized investment index in the ith year,
is a data value i-1 year of the standardized investment index>1；
An arithmetic incremental reliability index delta y is constructed by adopting a formula (3) and a formula (4)iSum-ratio type incremental reliability index
In the formula (I), the compound is shown in the specification,
for the data value of the normalized reliability index of the i-th year,
the data value of the standardized reliability index in the i-1 th year is obtained;
and (3) carrying out intra-group standardization treatment on the arithmetic incremental investment index by adopting a formula (5):
in the formula, xiTo select the data value for the index year i,
is the mean value, σXIs the standard deviation.
3. The method of claim 2, wherein the screening the investment indicators associated with each reliability indicator comprises:
calculating the Pearson correlation coefficient between the non-incremental reliability index and the non-incremental investment index by adopting a formula (6):
in the formula (I), the compound is shown in the specification,
a data value representing a normalized non-incremental investment index,
a data value representing a normalized non-incremental reliability indicator;
calculating the Pearson correlation coefficient between the arithmetic incremental reliability index and the arithmetic incremental investment index by adopting a formula (7):
in the formula,. DELTA.xjData value, Δ y, representing an arithmetic incremental investment indexiA data value representing an arithmetic type incremental reliability index,
represents the average of the arithmetic incremental reliability indicators,
means representing an arithmetic incremental investment index;
calculating the Pearson correlation coefficient between the arithmetic incremental reliability index and the proportional incremental investment index by adopting a formula (8):
in the formula (I), the compound is shown in the specification,
data value, Δ y, representing a proportional incremental investment indexiA data value representing an arithmetic type incremental reliability index,
represents the average of the arithmetic incremental reliability indicators,
representing an average of the proportional incremental investment indicators;
calculating the Pearson correlation coefficient between the proportional incremental reliability index and the arithmetic incremental investment index by adopting a formula (9):
in the formula,. DELTA.xjA data value representing an arithmetic incremental investment index,
a data value representing a proportional-type incremental reliability indicator,
represents the average of the proportional-type incremental reliability indicators,
means representing an arithmetic incremental investment index;
calculating the Pearson correlation coefficient between the proportional incremental reliability index and the proportional incremental investment index by adopting a formula (10), wherein the formula is as follows:
in the formula (I), the compound is shown in the specification,
a data value representing a proportional incremental investment index,
a data value representing a proportional-type incremental reliability indicator,
represents the average of the proportional-type incremental reliability indicators,
representing an average of the proportional incremental investment indicators;
according to the Pearson correlation coefficient, selecting a reliability index and an investment index combination of which R is more than or equal to R, wherein,
and R is a preset threshold value.
4. The method for evaluating investment benefits based on feature extraction and lasso regression as claimed in claim 1, wherein said constructing a lasso regression correlation model based on the combination of the reliability indicators and the investment indicators obtained by screening comprises:
constructing a first model (11):
Y＝α0+α1X1+α2X2+…+αpXp+ε (11)
wherein Y is a reliability index, XiFor the ith investment index, alpha, corresponding to the reliability indexiFor the weight coefficient of the influence of the investment index on the reliability index, i is 1, …, p, α0The reliability index weight is provided for reflecting the power distribution network of the data source, and epsilon is other indexes and errors which are not included in the index model but have potential influence on the reliability index Y;
averaging the linear regression models (11), and eliminating the unobservable term epsilon to obtain a second model (12):
E(y)＝α0+α1X1+α2X2+…+αpXp (12)
wherein E (Y) is an average value of the reliability index;
forming sample data array by n data values of p investment indexes
In the formula, xijJ-th data value representing an investment index i;
the n data values of the reliability index corresponding to the p investment indexes form a reliability index vector:
in the formula, yiAn ith data value representing the selected reliability indicator.
Lasso estimation value for calculating investment index weight coefficient
Controlling the estimation of the weighting coefficients of the investment indicators by means of a band linearity constraint (15)
Wherein t is more than or equal to 0 as an adjusting parameter;
calculating goodness of fit R of the lasso regression correlation model by adopting formula (16)e；
In the formula (I), the compound is shown in the specification,
is the i-th term of the reliability index model value based on the computationally lasso regression correlation model.
5. The method for investment benefit evaluation based on feature extraction and lasso regression according to any of claims 1-4, wherein if the goodness-of-fit is greater than a preset value, the following evaluation is performed: if the lasso estimation value of the influence weight coefficient of the investment index on the reliability index is 0, the corresponding investment index has no influence on the considered reliability index; if the lasso estimation value of the influence weight coefficient of the investment index on the reliability index is not 0, the corresponding investment index has a reverse or forward effect on the size of the considered reliability index.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011624938.2A CN112651651A (en) | 2020-12-30 | 2020-12-30 | Investment benefit evaluation method based on feature extraction and lasso regression |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011624938.2A CN112651651A (en) | 2020-12-30 | 2020-12-30 | Investment benefit evaluation method based on feature extraction and lasso regression |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112651651A true CN112651651A (en) | 2021-04-13 |
Family
ID=75366745
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011624938.2A Pending CN112651651A (en) | 2020-12-30 | 2020-12-30 | Investment benefit evaluation method based on feature extraction and lasso regression |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112651651A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113793077A (en) * | 2021-11-17 | 2021-12-14 | 广东电网有限责任公司佛山供电局 | Method and system for analyzing power failure influence of power distribution network user fault |
-
2020
- 2020-12-30 CN CN202011624938.2A patent/CN112651651A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113793077A (en) * | 2021-11-17 | 2021-12-14 | 广东电网有限责任公司佛山供电局 | Method and system for analyzing power failure influence of power distribution network user fault |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5248372B2 (en) | Power generation plan creation method, device, program, and storage device | |
| CN103985000B (en) | Medium-and-long term typical daily load curve prediction method based on function type nonparametric regression | |
| CN112712203B (en) | Day-highest load prediction method and system for power distribution network | |
| CN106779280B (en) | Decision-making determination method and system for secondary equipment major repair and technical modification | |
| CN112330121B (en) | Power distribution network automation construction investment benefit evaluation method | |
| CN109165809B (en) | Power grid planning project investment sequencing assessment method under new electricity-to-electricity environment | |
| CN108694470A (en) | A kind of data predication method and device based on artificial intelligence | |
| CN111539844B (en) | Moon electricity quantity prediction method based on K-L information quantity and ARIMA error correction | |
| CN110866658A (en) | Method for predicting medium and long term load of urban power grid | |
| CN110533291B (en) | Medium voltage distribution network weak link identification method based on risk assessment | |
| CN114862229A (en) | Power quality evaluation method and device, computer equipment and storage medium | |
| CN112651651A (en) | Investment benefit evaluation method based on feature extraction and lasso regression | |
| CN112330030B (en) | System and method for predicting requirements of expansion materials | |
| CN109711621A (en) | The industrial park load forecasting method combined based on discriminant analysis and support vector machines | |
| CN112836871A (en) | Power distribution network reliability prediction method, device, equipment and medium | |
| CN110991748A (en) | Short-term load prediction method for urban power grid | |
| CN113988557B (en) | Construction method and device of investment performance evaluation index system of power grid enterprise | |
| CN113743994A (en) | Provider's season-busy prediction method, system, equipment and storage medium | |
| Fathin et al. | Model for Predicting Electrical Energy Consumption Using ARIMA Method | |
| Jeihoonian et al. | Modeling and comparing energy consumption in basic metal industries by neural networks and ARIMA | |
| CN118365373B (en) | Prediction method, device and storage medium based on business expansion data | |
| Koksal et al. | Determination of optimal component maintenance process for RCAM of power transmission system using TOPSIS method | |
| Rakhmawati et al. | Linear and Quadratic Regression Model to Predict the Number of Electricity Distributed in Indonesia | |
| CN117575337A (en) | Prophet model and LSTM model-based power distribution network line peak load prediction method | |
| CN117767278A (en) | Unified load prediction method suitable for short-term price early warning in electric power market |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210413 |
|
| RJ01 | Rejection of invention patent application after publication |