CN114862057A - Agent electric quantity prediction method, device, equipment and medium - Google Patents
Agent electric quantity prediction method, device, equipment and medium Download PDFInfo
- Publication number
- CN114862057A CN114862057A CN202210661857.2A CN202210661857A CN114862057A CN 114862057 A CN114862057 A CN 114862057A CN 202210661857 A CN202210661857 A CN 202210661857A CN 114862057 A CN114862057 A CN 114862057A
- Authority
- CN
- China
- Prior art keywords
- target
- electric quantity
- month
- component
- monthly
- 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
- 238000000034 method Methods 0.000 title claims abstract description 81
- 230000001932 seasonal effect Effects 0.000 claims description 36
- 230000006870 function Effects 0.000 claims description 32
- 238000004590 computer program Methods 0.000 claims description 17
- 230000007423 decrease Effects 0.000 claims description 8
- 230000035784 germination Effects 0.000 claims description 8
- 238000006467 substitution reaction Methods 0.000 claims description 8
- 230000015572 biosynthetic process Effects 0.000 claims description 7
- 238000000354 decomposition reaction Methods 0.000 claims description 7
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 3
- 230000005611 electricity Effects 0.000 description 48
- 238000011161 development Methods 0.000 description 9
- 230000018109 developmental process Effects 0.000 description 9
- 238000012545 processing Methods 0.000 description 9
- 230000007246 mechanism Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 238000012887 quadratic function Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Images
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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas 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
Abstract
The application discloses a method, a device, equipment and a medium for predicting agent electric quantity, which relate to the technical field of electric power systems and comprise the following steps: determining a target monthly proxy electric quantity corresponding to the first historical month and not considering user loss based on the first loss rate of the first historical month and the historical monthly proxy electric quantity; decomposing the target monthly agent electric quantity to obtain a plurality of load components, predicting each load component according to a preset rule to obtain a predicted load component of a target month, and obtaining a reconstructed target monthly agent electric quantity based on all predicted load components; and determining a predicted monthly agent electric quantity corresponding to the target month and considering user loss based on the second loss rate and the reconstructed target monthly agent electric quantity. According to the technical scheme, the prediction precision of the power grid enterprise agent electric quantity can be effectively improved.
Description
Technical Field
The present invention relates to the field of power system technologies, and in particular, to a method, an apparatus, a device, and a medium for predicting a proxy power amount.
Background
The power grid enterprise agent electricity purchasing mechanism is a clear requirement for further deepening the market reformation of the coal-fired electricity generation online electricity price in China, and has important significance for orderly and stably realizing that all industrial and commercial users enter the electric power market and promoting the electric power market to accelerate the construction and development. The power grid enterprise needs to make a clear of the range of the agent electricity purchasing users, predict the electricity utilization scale of the agent industry and business users, and form the electricity price of the agent electricity purchasing users through a marketized electricity purchasing mode. Because the power grid enterprise agent electricity purchase price comprises the deviation electricity quantity generated by agent electricity purchase besides the market electricity purchase price, the power transmission and distribution price, the government fund and the addition, and the deviation electricity quantity is measured according to the month and is published in advance, the power grid enterprise needs to accurately predict the monthly electricity consumption of the agent user so as to reduce the deviation cost, reasonably establish the agent electricity purchase price and ensure the healthy development of a power grid enterprise agent electricity purchase mechanism.
The electric quantity prediction can be roughly divided into short-term electric quantity prediction and medium-and-long-term electric quantity prediction, and the latter is mainly oriented to the planning of a power grid, safe and economic operation and the marketing work of power grid enterprises. The method for predicting the medium and long term load comprises an exponential smoothing method, a time series method, a gray model and the like. In addition, artificial intelligence methods such as artificial neural networks and support vector machines are theoretically used for medium-term load prediction, in which a nonlinear relationship between input and output is established by training a large number of samples. Although medium and long term load data sequences generally have good tendency, the sample size in prediction is often small, so that the method is difficult to be applied practically. In addition, as the power market encourages industrial and commercial users to directly participate in market trading, and the power purchasing range of the power grid enterprise agent is continuously reduced, the influence of market environment and policy factors on the power grid enterprise agent power quantity is large, the trend of a historical load data sequence is damaged, and the power grid enterprise agent power quantity is difficult to be directly used for medium-term load prediction.
In summary, how to improve the prediction accuracy of the power grid enterprise agent power quantity is a problem to be solved at present.
Disclosure of Invention
In view of this, the present invention provides a method, an apparatus, a device and a medium for predicting agent power, which can improve the prediction accuracy of the agent power of the power grid enterprise. The specific scheme is as follows:
in a first aspect, the present application discloses a method for predicting a proxy power, including:
determining a target monthly proxy electric quantity corresponding to a first historical month and not considering user loss based on a first loss rate of the first historical month and the historical monthly proxy electric quantity;
decomposing the target monthly agent electric quantity to obtain a plurality of load components, predicting each load component according to a preset rule to obtain a predicted load component of a target month, and obtaining a reconstructed target monthly agent electric quantity based on all the predicted load components;
and establishing a target user loss rate model, predicting a second loss rate of the target month by using the target user loss rate model, and determining predicted monthly agent electric quantity corresponding to the target month and considering user loss based on the second loss rate and the reconstructed target monthly agent electric quantity.
Optionally, the decomposing the target monthly agent electric quantity to obtain a plurality of load components includes:
and decomposing the target monthly agent electric quantity by using a time series decomposition algorithm to obtain a trend component, a seasonal component and a random component.
Optionally, the predicting each load component according to a preset rule to obtain a predicted load component of the target month includes:
predicting the trend component by utilizing a polynomial curve fitting method to obtain a predicted trend component of the target month;
predicting the seasonal component by using a preset historical contemporaneous data substitution method to obtain a predicted seasonal component of the target month;
and predicting the random component by using an average value estimation method to obtain a predicted random component of the target month.
Optionally, the predicting the trend component by using a polynomial curve fitting method to obtain a predicted trend component of the target month includes:
determining a polynomial function of the trend component, and performing fitting evaluation on the polynomial function by using a mean square error function comprising a regular term function to determine a target order of the polynomial function;
a target polynomial is constructed based on the target order and determined as a predictive trend component for the target month.
Optionally, the predicting the seasonal component by using a preset historical contemporaneous data replacement method to obtain the predicted seasonal component of the target month includes:
and presetting a historical contemporaneous discrimination coefficient, and constructing a predicted seasonal component of the target month based on the historical contemporaneous discrimination coefficient and the seasonal component.
Optionally, the predicting the random component by using an average estimation method to obtain a predicted random component of the target month includes:
determining an average random component of a preset number of the random components, and determining the average random component as a predicted random component of the target month.
Optionally, the constructing a target user attrition rate model further includes:
constructing an index function for representing the germination stage, the formation stage and the decline stage of the user attrition rate based on a life cycle theory, and constructing an initial user attrition rate model based on the index function;
and determining a parameter value of a preset target parameter in the initial user attrition rate model based on a third attrition rate of a second historical month to obtain a target user attrition rate model.
In a second aspect, the present application discloses an agent power prediction apparatus, including:
the electric quantity determining module is used for determining a target monthly proxy electric quantity corresponding to a first historical month, which does not account for user loss, based on a first loss rate of the first historical month and the historical monthly proxy electric quantity;
the component prediction module is used for decomposing the target monthly agent electric quantity to obtain a plurality of load components, predicting each load component according to a preset rule to obtain a predicted load component of a target month, and obtaining a reconstructed target monthly agent electric quantity based on all the predicted load components;
and the electric quantity prediction module is used for constructing a target user loss rate model, predicting a second loss rate of the target month by using the target user loss rate model, and determining predicted monthly agent electric quantity corresponding to the target month and considering user loss based on the second loss rate and the reconstructed target monthly agent electric quantity.
In a third aspect, the present application discloses an electronic device, comprising:
a memory for storing a computer program;
a processor for executing the computer program to implement the steps of the aforementioned disclosed proxy power prediction method.
In a fourth aspect, the present application discloses a computer readable storage medium for storing a computer program; wherein the computer program when executed by a processor implements the steps of the aforementioned disclosed proxy power prediction method.
Therefore, the target monthly proxy electric quantity corresponding to the first historical month and not considering user loss is determined based on the first loss rate of the first historical month and the historical monthly proxy electric quantity; decomposing the target monthly agent electric quantity to obtain a plurality of load components, predicting each load component according to a preset rule to obtain a predicted load component of a target month, and obtaining a reconstructed target monthly agent electric quantity based on all the predicted load components; and establishing a target user loss rate model, predicting a second loss rate of the target month by using the target user loss rate model, and determining predicted monthly agent electric quantity corresponding to the target month and considering user loss based on the second loss rate and the reconstructed target monthly agent electric quantity. Therefore, the target monthly proxy electric quantity corresponding to the first historical month and not considering the user loss is determined by utilizing the first loss rate of the first historical month and the historical monthly proxy electric quantity, then the target monthly proxy electric quantity is decomposed to obtain a plurality of load components, each load component is predicted, and the target monthly proxy electric quantity is reconstructed based on the predicted load components; and finally, predicting the monthly agent electric quantity of the target month under the condition of considering user loss by using the second loss rate and the reconstructed target monthly agent electric quantity. By the technical scheme, the prediction precision of the power grid enterprise agent electricity quantity is effectively improved, and the analysis of the development condition of the regional power grid enterprise agent electricity purchasing mechanism is facilitated.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a flowchart of an agent power prediction method disclosed in the present application;
fig. 2 is a flowchart of a specific proxy power prediction method disclosed in the present application;
FIG. 3 is a flow chart illustrating a specific proxy power prediction process disclosed herein;
fig. 4 is a schematic structural diagram of an agent power prediction apparatus disclosed in the present application;
fig. 5 is a block diagram of an electronic device disclosed in the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The power grid enterprise needs to make a clear proxy electricity purchasing user range, predict the electricity utilization scale of the proxy industrial and commercial users and form the electricity price of the proxy electricity purchasing users in a marketized electricity purchasing mode. Because the power grid enterprise agent electricity purchase price comprises the deviation electricity quantity generated by agent electricity purchase besides the market electricity purchase price, the power transmission and distribution price, the government fund and the addition, and the deviation electricity quantity is measured according to the month and is published in advance, the power grid enterprise needs to accurately predict the monthly electricity consumption of the agent user so as to reduce the deviation cost, reasonably establish the agent electricity purchase price and ensure the healthy development of a power grid enterprise agent electricity purchase mechanism. At present, the electric power market encourages industrial and commercial users to directly participate in market trading, the power grid enterprise agent electricity purchasing range is continuously reduced, so that the power grid enterprise agent electricity quantity is greatly influenced by market environment and policy factors, and the trend of a historical load data sequence is damaged. Therefore, the embodiment of the application discloses a method, a device, equipment and a medium for predicting the proxy electric quantity, which can improve the prediction precision of the proxy electric quantity of a power grid enterprise.
Step S11: determining a target monthly agent electric quantity corresponding to a first historical month and not considering user loss based on a first loss rate of the first historical month and the historical monthly agent electric quantity.
In this embodiment, a first attrition rate and a historical monthly proxy electric quantity of a first historical month are first obtained, and a target monthly proxy electric quantity corresponding to the first historical month and not accounting for user attrition is determined according to the first attrition rate and the historical monthly proxy electric quantity. It should be noted that the loss rate refers to a reduction rate of the number of users of the power grid enterprise agent caused by the fact that the users directly participate in the power market transaction from the power grid enterprise agent electricity purchase or the power selling company agent participates in the market transaction. Because the number of the power grid enterprise agent users is large and is mostly small and medium-sized industrial and commercial users, the number of the power grid enterprise agent users is in direct proportion to the agent electric quantity. That is, by considering the loss rate of the power grid enterprise agent users, the monthly agent power amount of the power grid enterprise in the first historical month is recovered under the condition that the user loss is not considered by the user, and assuming that the first historical month is tth month, the corresponding calculation formula is as follows:
Q t,re ＝Q t,0 /(1-L t )；
wherein Q is t,re Representing monthly agent power of the power grid enterprise in the t month under the condition of not considering user loss; q t,0 Representing the historical monthly agent electric quantity of the power grid enterprise in the tth month; l is t Representing a first attrition rate of the grid enterprise proxy user in month t.
Step S12: decomposing the target monthly agent electric quantity to obtain a plurality of load components, predicting each load component according to a preset rule to obtain a predicted load component of a target month, and obtaining a reconstructed target monthly agent electric quantity based on all the predicted load components.
In this embodiment, the restored target monthly proxy electric quantity not considering the user loss is decomposed to obtain a plurality of decomposed load components, each load component is predicted according to a preset rule to obtain a predicted load component of the target month, and then the reconstructed target monthly proxy electric quantity is obtained based on all the predicted load components. The reconstruction method may be an addition operation.
Step S13: and establishing a target user loss rate model, predicting a second loss rate of the target month by using the target user loss rate model, and determining predicted monthly agent electric quantity corresponding to the target month and considering user loss based on the second loss rate and the reconstructed target monthly agent electric quantity.
In this embodiment, a target user churn rate model is constructed, a second churn rate of the target month is predicted by using the target user churn rate model, and a predicted monthly agent electric quantity corresponding to the target month and accounting for user churn is determined based on the second churn rate and the reconstructed target monthly agent electric quantity. Namely, the second loss rate of the power grid enterprise proxy users in the target month is predicted by constructing a target user loss rate model so as to correct the prediction result of the monthly proxy power quantity of the proxy power grid enterprise.
Therefore, the target monthly proxy electric quantity corresponding to the first historical month and not considering user loss is determined based on the first loss rate of the first historical month and the historical monthly proxy electric quantity; decomposing the target monthly agent electric quantity to obtain a plurality of load components, predicting each load component according to a preset rule to obtain a predicted load component of a target month, and obtaining a reconstructed target monthly agent electric quantity based on all the predicted load components; and establishing a target user loss rate model, predicting a second loss rate of the target month by using the target user loss rate model, and determining predicted monthly agent electric quantity corresponding to the target month and considering user loss based on the second loss rate and the reconstructed target monthly agent electric quantity. Therefore, the target monthly proxy electric quantity corresponding to the first historical month and not considering the user loss is determined by utilizing the first loss rate of the first historical month and the historical monthly proxy electric quantity, then the target monthly proxy electric quantity is decomposed to obtain a plurality of load components, each load component is predicted, and the target monthly proxy electric quantity is reconstructed based on the predicted load components; and finally, predicting the monthly agent electric quantity of the target month under the condition of considering the user loss by using the second loss rate and the reconstructed target monthly agent electric quantity. By the technical scheme, the prediction precision of the power grid enterprise agent electricity quantity is effectively improved, and the analysis of the development condition of the regional power grid enterprise agent electricity purchasing mechanism is facilitated.
Referring to fig. 2, the embodiment of the present application discloses a specific method for predicting the proxy power amount, and compared with the previous embodiment, the present embodiment further describes and optimizes the technical solution. The method specifically comprises the following steps:
step S21: determining a target monthly agent electric quantity corresponding to a first historical month and not considering user loss based on a first loss rate of the first historical month and the historical monthly agent electric quantity.
Step S22: and decomposing the target monthly agent electric quantity by using a time series decomposition algorithm to obtain a trend component, a seasonal component and a random component.
In this embodiment, the target monthly agent electric quantity of the power grid enterprise is decomposed into a trend component Q based on a time series decomposition method t,tr Seasonal component Q t,se Random component Q t,ra Specifically, it can be expressed as:
Q t,re ＝Q t,tr +Q t,se +Q t,ra ；
wherein the trend component reflects a relatively stable user monthly electricity usage development direction resulting from regional long term economic growth; the seasonal component reflects the periodic fluctuation of the electricity consumption of the regional users which is alternately influenced along with seasons; the random component reflects a random small perturbation of the user's monthly power usage.
Step S23: predicting the trend component by utilizing a polynomial curve fitting method to obtain a predicted trend component of the target month; predicting the seasonal component by using a preset historical contemporaneous data substitution method to obtain a predicted seasonal component of the target month; and predicting the random component by using an average value estimation method to obtain a predicted random component of the target month.
In this embodiment, a polynomial curve fitting method, a historical contemporaneous data substitution method, and an average value estimation method are respectively used to predict the trend component, the seasonal component, and the random component to obtain a corresponding predicted trend component, a predicted seasonal component, and a predicted random component.
The predicting the trend component by using a polynomial curve fitting method to obtain the predicted trend component of the target month includes: determining a polynomial function of the trend component, and performing fitting evaluation on the polynomial function by using a mean square error function comprising a regular term function to determine a target order of the polynomial function; a target polynomial is constructed based on the target order and determined as a predictive trend component for the target month. It will be appreciated that the polynomial function corresponding to the trend component is first determined, i.e. the trend component Q is represented by constructing a polynomial function t,tr The functional relation with the month t is as follows:
wherein W is { W ═ W 1 ,w 2 ,…,w N Is a polynomial coefficient matrix and N is a polynomial order. Q tr (t, W) represents a trend component Q of the t-th month under a coefficient W t,tr The result of polynomial fitting.
The fitting effect of the polynomial function is then evaluated using a mean square error function comprising a regular term function, expressed as:
in the formula, e (w) represents a mean square error of a polynomial curve fitting based on the lunar degree trend component, M is the number of months of the known historical lunar degree proxy electric quantity, and λ is a regular term coefficient.
In the fitting process, the minimization E (W) is taken as a target, the polynomial order N is adjusted, namely the target order of the polynomial function is determined, and the polynomial coefficient matrix is further solved by adopting a gradient descent method. And finally, constructing a target polynomial based on the target order and the corresponding polynomial coefficient matrix, and determining the target polynomial as a prediction trend component of the target month, wherein the prediction trend component can be expressed as:
in the formula, Q T,tr Representing the predicted trend component for the target month, i.e., month T.
Further, the predicting the seasonal component by using a preset historical contemporaneous data substitution method to obtain the predicted seasonal component of the target month includes: and presetting a historical contemporaneous discrimination coefficient, and constructing a predicted seasonal component of the target month based on the historical contemporaneous discrimination coefficient and the seasonal component. It can be understood that, in this embodiment, the prediction seasonal component of the target month is obtained by predicting the seasonal component by using a historical contemporaneous data substitution method, and the specific relation is as follows:
in the formula, Q T,se Representing the predicted seasonal component, Q, of the target month, i.e. month Tth t,se Is the seasonal component of month t, h t,T Is a historical contemporaneous discrimination coefficient, h if and only if the historical agent electric quantity month is the same as the target month t,T 1, otherwise h t,T M is a preset number of months for which historical monthly proxy electricity quantities are known. Namely, the method for constructing the target month based on the historical contemporaneous discriminative coefficient and the seasonal componentAnd measuring seasonal components.
In addition, the predicting the random component by using the average value estimation method to obtain the predicted random component of the target month includes: acquiring random components of a preset number of historical months, and determining the average value of all the random components as the prediction random component of the target month. It can be understood that, in this embodiment, the random component is predicted by using an average value estimation method to obtain the predicted random component of the target month, that is, the specific relation formula of the predicted random component, in which the preset number, that is, the average value of the random components of M historical months is determined as the target month, is:
wherein Q T,ra A predictive random component, Q, representing the target month, i.e. month Tth t,ra For the random component of month t, M is a preset number of months for which historical monthly agent capacities are known.
Step S24: and obtaining the reconstructed target monthly agent electric quantity based on the predicted trend component, the predicted seasonal component and the predicted random component.
In this embodiment, the reconstructed target monthly proxy electric quantity is obtained based on the predicted trend component, the predicted seasonal component, and the predicted random component, and the specific relation is as follows:
Q T,re ＝Q T,tr +Q T,se +Q T,ra ；
wherein Q T,re Under the condition that user loss is not considered, the predicted target month of the power grid enterprise, namely the target monthly agent electric quantity of the Tth month.
Step S25: and constructing an index function for representing the germination stage, the formation stage and the decline stage of the user attrition rate based on a life cycle theory, and constructing an initial user attrition rate model based on the index function.
In this embodiment, an index function used for representing the germination stage, the formation stage and the decline stage of the user attrition rate is constructed based on the life cycle theory, and an initial user attrition rate model is constructed based on the index function. It can be understood that the loss rate of the power grid enterprise agent user can be divided into three stages of a germination stage, a formation stage and a decline stage based on a life cycle theory. The germination period refers to a guarantee period that the electric charge of the industrial and commercial users is converted from the catalog electricity price settlement into the electric charge of the power grid enterprise agency participating in the electric power market transaction, and the industrial and commercial users generally choose the power grid enterprise agency to participate in the electric power market transaction because the industrial and commercial users do not adapt to the electric power market competition environment, so that the loss rate of the power grid enterprise agency users is low. The forming period refers to a period that industrial and commercial users gradually adapt to the competition environment of the power market and select the rapid development period of the power market which directly or by the agency of the power selling company to participate in the power market transaction, and the industrial and commercial users can participate in the power market transaction and obtain more flexible power selling service through the direct participation period or the agency of the power selling company, so that the loss rate of the agency users of the power grid enterprise is higher. The decline period refers to the mature period of the electricity selling market that most industrial and commercial users choose to directly participate in the electricity market transaction or transfer to the electricity selling company agents, and a few industrial and commercial users continue to participate in the electricity market transaction by the power grid enterprise agents. And then representing three stages of a germination stage, a formation stage and a decline stage of the loss rate of the power grid enterprise agent users by using an exponential function, wherein the three stages are represented as follows:
L(t)＝L max exp(-k(t-b) 2 )；
in the formula, L max K, b are function unknown parameters, L max And the maximum value of the loss rate of the power grid enterprise agent users is represented.
Step S26: and determining a parameter value of a preset target parameter in the initial user attrition rate model based on a third attrition rate of a second historical month to obtain a target user attrition rate model.
In this embodiment, a third attrition rate of the second historical month is obtained, and then a parameter value of a preset target parameter in the initial user attrition rate model is determined by using the third attrition rate, so as to obtain a target user attrition rate model. The specific steps in the process are as follows:
firstly, logarithm is solved on an exponential function of the loss rate of the power grid enterprise agent users, and a quadratic function is obtained and expressed as:
L ln (t)＝lnL(t)＝-k(t-b) 2 +lnL max ＝-αt 2 +βt+γ；
in the formula, alpha, beta and gamma are quadratic functions L ln (t) unknown parameters.
Then, logarithm is obtained on the obtained third loss rate of the second historical month, namely logarithm is obtained on historical monthly loss rate data of the power grid enterprise agent users, and a quadratic function L is fitted by minimizing mean square error ln (t) determining the parameter value alpha of the preset target parameter in a curve mode opt 、β opt 、γ opt And then, performing exponential operation on the quadratic function to obtain a final target user loss rate model, which is expressed as:
L T ＝exp(L ln (T))＝exp(-α opt T 2 +β opt T+γ opt )；
in the formula, alpha opt 、β opt 、γ opt Is a quadratic function L ln (T), namely the parameter value of the preset target parameter.
Step S27: predicting a second loss rate of the target month by using the target user loss rate model, and determining a predicted monthly agent electric quantity corresponding to the target month and considering user loss based on the second loss rate and the reconstructed target monthly agent electric quantity.
In this embodiment, the second attrition rate of the target month is predicted by using the obtained target user attrition rate model, that is, a specific T value is input to obtain L T . And then determining a predicted monthly proxy electric quantity corresponding to the target month and accounting for user loss based on the second loss rate and the reconstructed target monthly proxy electric quantity, namely correcting a predicted result of the monthly proxy electric quantity by using the second loss rate, wherein the specific relation is as follows:
Q T,pr ＝(1-L T )Q T,re ；
in the formula, Q T,pr Representing predicted monthly proxy power, L, corresponding to target month T, accounting for user churn T Indicating the predicted second loss rate, Q T,re And representing the reconstructed target monthly agent electric quantity.
For a more specific processing procedure of the step S21, reference may be made to corresponding contents disclosed in the foregoing embodiments, and details are not repeated here.
Therefore, when the target monthly agent electric quantity is decomposed, the time series decomposition algorithm is specifically used for decomposing the target monthly agent electric quantity to obtain a trend component, a seasonal component and a random component, then a polynomial curve fitting method, a historical contemporaneous data substitution method and an average value estimation method are respectively adopted for predicting the trend component, the seasonal component and the random component, and the target monthly agent electric quantity is reconstructed. In addition, a user loss rate model used for representing the germination period, the formation period and the decline period of the user loss rate is built on the basis of a life cycle theory, the initially built user loss rate model is trained by using the third loss rate of the second historical month to obtain a target user loss rate model, so that the second loss rate is predicted by using the target user loss rate model, and the predicted monthly agent electric quantity corresponding to the target month and accounting for the user loss is determined by using the second loss rate and the reconstructed target monthly agent electric quantity. By the technical scheme, the prediction precision of the power grid enterprise agent electricity quantity is effectively improved, and the analysis of the development condition of the regional power grid enterprise agent electricity purchasing mechanism is facilitated.
Referring to fig. 3, fig. 3 discloses a specific flow chart of agent power prediction, which includes: calculating the loss rate of power grid enterprise agent users, restoring monthly agent electric quantity data of the power grid enterprise, decomposing the restored historical monthly agent electric quantity data of the power grid enterprise into a trend component, a seasonal component and a random component based on a time sequence decomposition method (a time sequence decomposition method), predicting corresponding load components by respectively adopting a polynomial curve fitting method, a historical contemporaneous data substitution method and an average value estimation method, and reconstructing to obtain a user monthly load prediction result; and constructing a user loss rate model based on a life cycle theory, fitting a user loss rate curve, predicting the power grid enterprise agent user loss rate of the target month by using the user loss rate model, and correcting the user monthly load prediction result.
Referring to fig. 4, an embodiment of the present application discloses an agent electric quantity prediction apparatus, including:
the electric quantity determining module 11 is configured to determine, based on a first attrition rate of a first historical month and historical monthly proxy electric quantity, a target monthly proxy electric quantity corresponding to the first historical month, which does not account for user attrition;
the component prediction module 12 is configured to decompose the target monthly agent electric quantity to obtain a plurality of load components, predict each load component according to a preset rule to obtain a predicted load component of the target month, and obtain a reconstructed target monthly agent electric quantity based on all the predicted load components;
and the electric quantity prediction module 13 is configured to construct a user loss rate model, predict a second loss rate of the target month by using the user loss rate model, and determine a predicted monthly agent electric quantity corresponding to the target month and accounting for user loss based on the second loss rate and the reconstructed target monthly agent electric quantity.
Therefore, the target monthly proxy electric quantity corresponding to the first historical month and not considering user loss is determined based on the first loss rate of the first historical month and the historical monthly proxy electric quantity; decomposing the target monthly agent electric quantity to obtain a plurality of load components, predicting each load component according to a preset rule to obtain a predicted load component of a target month, and obtaining a reconstructed target monthly agent electric quantity based on all the predicted load components; and establishing a target user loss rate model, predicting a second loss rate of the target month by using the target user loss rate model, and determining predicted monthly agent electric quantity corresponding to the target month and considering user loss based on the second loss rate and the reconstructed target monthly agent electric quantity. Therefore, the target monthly proxy electric quantity corresponding to the first historical month and not considering the user loss is determined by utilizing the first loss rate of the first historical month and the historical monthly proxy electric quantity, then the target monthly proxy electric quantity is decomposed to obtain a plurality of load components, each load component is predicted, and the target monthly proxy electric quantity is reconstructed based on the predicted load components; and finally, predicting the monthly agent electric quantity of the target month under the condition of considering user loss by using the second loss rate and the reconstructed target monthly agent electric quantity. By the technical scheme, the prediction precision of the power grid enterprise agent electricity quantity is effectively improved, and the analysis of the development condition of the regional power grid enterprise agent electricity purchasing mechanism is facilitated.
Fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application. The method specifically comprises the following steps: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input output interface 25, and a communication bus 26. The memory 22 is used for storing a computer program, and the computer program is loaded and executed by the processor 21 to implement relevant steps in the agent power prediction method executed by an electronic device disclosed in any of the foregoing embodiments.
In this embodiment, the power supply 23 is configured to provide a working voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and an external device, and a communication protocol followed by the communication interface is any communication protocol applicable to the technical solution of the present application, and is not specifically limited herein; the input/output interface 25 is configured to obtain external input data or output data to the outside, and a specific interface type thereof may be selected according to specific application requirements, which is not specifically limited herein.
The processor 21 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and the like. The processor 21 may be implemented in at least one hardware form of a DSP (Digital Signal Processing), an FPGA (Field-Programmable Gate Array), and a PLA (Programmable Logic Array). The processor 21 may also include a main processor and a coprocessor, where the main processor is a processor for Processing data in an awake state, and is also called a Central Processing Unit (CPU); a coprocessor is a low power processor for processing data in a standby state. In some embodiments, the processor 21 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content required to be displayed on the display screen. In some embodiments, the processor 21 may further include an AI (Artificial Intelligence) processor for processing a calculation operation related to machine learning.
In addition, the storage 22 is used as a carrier for storing resources, and may be a read-only memory, a random access memory, a magnetic disk or an optical disk, etc., the resources stored thereon include an operating system 221, a computer program 222, data 223, etc., and the storage may be a transient storage or a permanent storage.
The operating system 221 is used for managing and controlling each hardware device on the electronic device 20 and the computer program 222, so as to implement the operation and processing of the mass data 223 in the memory 22 by the processor 21, which may be Windows, Unix, Linux, or the like. The computer program 222 may further include a computer program that can be used to perform other specific tasks in addition to the computer program that can be used to perform the agent power prediction method performed by the electronic device 20 disclosed in any of the foregoing embodiments. The data 223 may include data received by the electronic device and transmitted from an external device, or may include data collected by the input/output interface 25 itself.
Further, an embodiment of the present application further discloses a computer-readable storage medium, where a computer program is stored in the storage medium, and when the computer program is loaded and executed by a processor, the method steps executed in the agent power prediction process disclosed in any of the foregoing embodiments are implemented.
The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The method, the device, the equipment and the storage medium for predicting the agent electric quantity provided by the invention are described in detail, a specific example is applied in the text to explain the principle and the implementation mode of the invention, and the description of the embodiment is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. A method for predicting a proxy power level, comprising:
determining a target monthly proxy electric quantity corresponding to a first historical month and not considering user loss based on a first loss rate of the first historical month and the historical monthly proxy electric quantity;
decomposing the target monthly agent electric quantity to obtain a plurality of load components, predicting each load component according to a preset rule to obtain a predicted load component of a target month, and obtaining a reconstructed target monthly agent electric quantity based on all the predicted load components;
and establishing a target user loss rate model, predicting a second loss rate of the target month by using the target user loss rate model, and determining predicted monthly agent electric quantity corresponding to the target month and considering user loss based on the second loss rate and the reconstructed target monthly agent electric quantity.
2. The method of predicting the proxy power consumption as claimed in claim 1, wherein the decomposing the target monthly proxy power consumption to obtain a plurality of load components comprises:
and decomposing the target monthly agent electric quantity by using a time series decomposition algorithm to obtain a trend component, a seasonal component and a random component.
3. The agent electric quantity prediction method according to claim 2, wherein the predicting each load component according to a preset rule to obtain a predicted load component of a target month comprises:
predicting the trend component by utilizing a polynomial curve fitting method to obtain a predicted trend component of the target month;
predicting the seasonal component by using a preset historical contemporaneous data substitution method to obtain a predicted seasonal component of the target month;
and predicting the random component by using an average value estimation method to obtain a predicted random component of the target month.
4. The method for predicting the agent electric quantity according to claim 3, wherein the predicting the trend component by using a polynomial curve fitting method to obtain the predicted trend component of the target month comprises:
determining a polynomial function of the trend component, and performing fitting evaluation on the polynomial function by using a mean square error function comprising a regular polynomial function to determine a target order of the polynomial function;
a target polynomial is constructed based on the target order and determined as a predictive trend component for the target month.
5. The agent power prediction method of claim 3, wherein the predicting the seasonal component by using a preset historical contemporaneous data replacement method to obtain the predicted seasonal component of the target month comprises:
and presetting a historical contemporaneous discrimination coefficient, and constructing a predicted seasonal component of the target month based on the historical contemporaneous discrimination coefficient and the seasonal component.
6. The agent power prediction method of claim 3, wherein the predicting the random component by using an average estimation method to obtain the predicted random component of the target month comprises:
determining an average random component of a preset number of the random components, and determining the average random component as a predicted random component of the target month.
7. The agent power prediction method according to any one of claims 1 to 6, wherein the constructing a target user attrition rate model further comprises:
constructing an index function for representing the germination stage, the formation stage and the decline stage of the user attrition rate based on a life cycle theory, and constructing an initial user attrition rate model based on the index function;
and determining a parameter value of a preset target parameter in the initial user attrition rate model based on a third attrition rate of a second historical month to obtain a target user attrition rate model.
8. A proxy power prediction apparatus, comprising:
the electric quantity determining module is used for determining a target monthly proxy electric quantity corresponding to a first historical month, which does not account for user loss, based on a first loss rate of the first historical month and the historical monthly proxy electric quantity;
the component prediction module is used for decomposing the target monthly agent electric quantity to obtain a plurality of load components, predicting each load component according to a preset rule to obtain a predicted load component of a target month, and obtaining a reconstructed target monthly agent electric quantity based on all the predicted load components;
and the electric quantity prediction module is used for constructing a target user loss rate model, predicting a second loss rate of the target month by using the target user loss rate model, and determining predicted monthly agent electric quantity corresponding to the target month and considering user loss based on the second loss rate and the reconstructed target monthly agent electric quantity.
9. An electronic device, comprising:
a memory for storing a computer program;
a processor for executing said computer program to implement the steps of the proxy power prediction method as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium for storing a computer program; wherein the computer program when executed by a processor implements the steps of the proxy power prediction method according to any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210661857.2A CN114862057A (en) | 2022-06-13 | 2022-06-13 | Agent electric quantity prediction method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210661857.2A CN114862057A (en) | 2022-06-13 | 2022-06-13 | Agent electric quantity prediction method, device, equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114862057A true CN114862057A (en) | 2022-08-05 |
Family
ID=82625560
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210661857.2A Pending CN114862057A (en) | 2022-06-13 | 2022-06-13 | Agent electric quantity prediction method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114862057A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109727066A (en) * | 2018-12-27 | 2019-05-07 | 浙江华云信息科技有限公司 | A kind of big industrial electricity consumers load forecasting method based on XGBoost algorithm |
| US20190265768A1 (en) * | 2018-02-24 | 2019-08-29 | Hefei University Of Technology | Method, system and storage medium for predicting power load probability density based on deep learning |
| US20200143208A1 (en) * | 2018-11-05 | 2020-05-07 | Royal Bank Of Canada | Opponent modeling with asynchronous methods in deep rl |
| CN113516378A (en) * | 2021-06-24 | 2021-10-19 | 东北电力大学 | Retail electricity price package design method considering electric energy quality difference |
| WO2021208342A1 (en) * | 2020-04-14 | 2021-10-21 | 广东卓维网络有限公司 | Power system based on cooperative interaction between diverse users and power grid |
| CN113962491A (en) * | 2021-12-01 | 2022-01-21 | 国网河北省电力有限公司雄安新区供电公司 | Short-term combined prediction method and device for cooling, heating and power loads based on seasonal decomposition |
| CN114119098A (en) * | 2021-11-30 | 2022-03-01 | 广东电网有限责任公司东莞供电局 | Decision tree-based customer loss analysis method and system |
-
2022
- 2022-06-13 CN CN202210661857.2A patent/CN114862057A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190265768A1 (en) * | 2018-02-24 | 2019-08-29 | Hefei University Of Technology | Method, system and storage medium for predicting power load probability density based on deep learning |
| US20200143208A1 (en) * | 2018-11-05 | 2020-05-07 | Royal Bank Of Canada | Opponent modeling with asynchronous methods in deep rl |
| CN109727066A (en) * | 2018-12-27 | 2019-05-07 | 浙江华云信息科技有限公司 | A kind of big industrial electricity consumers load forecasting method based on XGBoost algorithm |
| WO2021208342A1 (en) * | 2020-04-14 | 2021-10-21 | 广东卓维网络有限公司 | Power system based on cooperative interaction between diverse users and power grid |
| CN113516378A (en) * | 2021-06-24 | 2021-10-19 | 东北电力大学 | Retail electricity price package design method considering electric energy quality difference |
| CN114119098A (en) * | 2021-11-30 | 2022-03-01 | 广东电网有限责任公司东莞供电局 | Decision tree-based customer loss analysis method and system |
| CN113962491A (en) * | 2021-12-01 | 2022-01-21 | 国网河北省电力有限公司雄安新区供电公司 | Short-term combined prediction method and device for cooling, heating and power loads based on seasonal decomposition |
Non-Patent Citations (2)
| Title |
|---|
| 叶利 等: "基于战略地图的电网企业市场竞争力评价研究", 湖北电力, vol. 43, no. 3, 30 June 2019 (2019-06-30), pages 36 - 41 * |
| 钱仲文;陈浩;纪德良;: "一种基于XGBoost算法的月度负荷预测方法", 浙江电力, no. 05, 22 May 2019 (2019-05-22), pages 80 - 85 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| McNerney et al. | How production networks amplify economic growth | |
| US7765123B2 (en) | Indicating which of forecasting models at different aggregation levels has a better forecast quality | |
| Sornette et al. | Clarifications to questions and criticisms on the Johansen–Ledoit–Sornette financial bubble model | |
| Safari et al. | Very short-term wind power prediction interval framework via bi-level optimization and novel convex cost function | |
| CN112200489B (en) | Integrated optimization system, method and device for nonferrous metal smelting production and supply marketing | |
| Klein et al. | Deep distributional time series models and the probabilistic forecasting of intraday electricity prices | |
| Lawson et al. | Bayesian framework for power network planning under uncertainty | |
| CN112785057B (en) | Method, device, equipment and storage medium for predicting quantity of parts based on exponential smoothing | |
| Jiménez et al. | A closed-loop data-driven optimization framework for the unit commitment problem: A Q-learning approach under real-time operation | |
| EP3961507A1 (en) | Optimal policy learning and recommendation for distribution task using deep reinforcement learning model | |
| Ma et al. | Distribution-free continuous review inventory model with controllable lead time and setup cost in the presence of a service level constraint | |
| CN111695967A (en) | Method, device, equipment and storage medium for determining quotation | |
| CN114862057A (en) | Agent electric quantity prediction method, device, equipment and medium | |
| CN117114776A (en) | Price reporting method for provincial day-ahead spot transaction | |
| CN114971736A (en) | Power metering material demand prediction method and device, electronic equipment and storage medium | |
| CN113222300B (en) | Method, device, readable medium and equipment for processing product modification data | |
| Arifovic et al. | Are sunspots learnable? An experimental investigation in a simple macroeconomic model | |
| CN114282881A (en) | Depreciation measuring and calculating method and device, storage medium and computer equipment | |
| CN114066642A (en) | Electric power retail risk prediction method, terminal and storage medium | |
| CN114298847A (en) | Prediction method, device, equipment and storage medium for employee medical insurance fund balance | |
| CN113592263A (en) | Resource return increment prediction method and device based on dynamic resource return increase ratio | |
| Manco et al. | Risker: Platform implementation of complex system model for financial risk management in energy markets | |
| Radet et al. | Dynamic aware aging design of a simple distributed energy system: A comparative approach with single stage design strategies | |
| Sanchez Guadarrama | Power system adequacy: on two-area models and the capacity procurement decision process | |
| CN112351105B (en) | Cloud service evaluation method and device |
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 |