CN113361832A - Electric power data center station and working method - Google Patents
Electric power data center station and working method Download PDFInfo
- Publication number
- CN113361832A CN113361832A CN202110911506.8A CN202110911506A CN113361832A CN 113361832 A CN113361832 A CN 113361832A CN 202110911506 A CN202110911506 A CN 202110911506A CN 113361832 A CN113361832 A CN 113361832A
- Authority
- CN
- China
- Prior art keywords
- model
- combined
- predictive model
- data
- prediction model
- 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.)
- Granted
Links
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
- G06Q10/067—Enterprise or organisation modelling
-
- 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
Abstract
The invention relates to a power data center station and a working method thereof, wherein the working method of the power data center station comprises the following steps: reading the stored historical data in response to a trigger command; inputting the read historical data into a pre-constructed combined prediction model for analysis to obtain an analysis result; and outputting an analysis result. By adopting the technical scheme, the combined prediction model constructed according to the related data provides effective service data, algorithm model and calculation capability support for the electric power data middling station.
Description
Technical Field
The invention relates to the field of power systems, in particular to electric quantity of a power data center and a working method thereof.
Background
The station area in the power system refers to the power supply range or area of one or more transformers, and the station in the power data controls and allocates the transformers in the power supply range or area. In power system control, configuration and management of stations in power data is an important aspect of accomplishing power system operations.
Wherein accurate short-term power load prediction is one of the important workflows in station control in power data. Most of the existing load prediction methods are that the power data center station directly predicts the load of power equipment such as a bus and a transformer by analyzing historical load data of the bus and the transformer and adding meteorological factor influence, and the working method of the power data center station has the problem of single input data due to the fact that the power equipment in a station area is limited. Therefore, a new power data center is urgently needed to improve the working efficiency and accuracy of the power data center.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a power data center and a working method thereof, a combined prediction model is pre-established, and the model is corrected by using normalized and large amount of historical service data accumulated by the power data center, so that the prediction precision of the model is improved, and the working process of the normalized power data center is formed.
In order to achieve the purpose, the technical scheme of the invention is as follows:
a power data center station electric quantity working method comprises the following steps:
reading the stored historical data in response to a trigger command;
inputting the read historical data into a pre-constructed combined prediction model for analysis to obtain an analysis result;
and outputting an analysis result.
The combined prediction model is composed of a prediction model and a time model, and the prediction model and the time model are combined to obtain the combined prediction model.
Wherein the prediction model is constructed using the following formula:
wherein, t is the predicted time,
as an initial time of the sample data,
to correspond to
The amount of electricity of (a) is raw,
is composed of
Approaching the theoretical extreme value of ∞ time,
and e is a constant.
Wherein the time model is constructed using the following formula:
wherein D is a monthly electricity quantity sequence; d _ Q, D _ X, D _ S represents a trend component sequence, a seasonal period component sequence, and a random component sequence of the monthly power amount, respectively.
And correcting the combined prediction model by using historical data to obtain an optimized prediction model.
The application also claims a power data center station, which comprises the following components:
the receiver receives the trigger command and forwards the trigger command to the processor;
a memory storing historical data;
a processor that performs the steps of:
reading historical data stored by a memory in response to a trigger command;
inputting the read historical data into a pre-constructed combined prediction model for analysis to obtain an analysis result;
and the output device outputs the analysis result.
The combined prediction model is composed of a prediction model and a time model, and the prediction model and the time model are combined to obtain the combined prediction model.
Wherein the prediction model is constructed using the following formula:
wherein, t is the predicted time,
as an initial time of the sample data,
to correspond to
The amount of electricity of (a) is raw,
is composed of
Approaching the theoretical extreme value of ∞ time,
and e is a constant.
Wherein the time model is constructed using the following formula:
wherein D is a monthly electricity quantity sequence; d _ Q, D _ X, D _ S represents a trend component sequence, a seasonal period component sequence, and a random component sequence of the monthly power amount, respectively.
And correcting the combined prediction model by using historical data to obtain an optimized prediction model.
By adopting the technical scheme, the combined prediction model constructed according to the related data provides effective service data, algorithm model and calculation capability support for the electric power data middling station. Compared with two single prediction models, the combined prediction model established by combining the prediction model and the time model not only considers the internal influence factors related to the electric quantity, but also considers the obvious seasonal difference and the obvious temporal difference of the electric quantity data, minimizes the prediction error of the combined prediction model through the fusion and complementation of the two models, avoids the defects of the two single prediction models, and improves the prediction accuracy; on the other hand, the combined prediction model is corrected by using a large amount of real and effective monthly electric quantity data acquired from the electric power data, the deviation of the prediction result is continuously reduced, accurate prediction is realized, and the prediction result is more reliable compared with two single prediction models.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present invention, and those skilled in the art can also obtain other drawings according to the drawings.
FIG. 1 is a system block diagram of a power data center of the present application;
fig. 2 is a flowchart illustrating the operation of the power data center station according to the present application.
Detailed Description
The technical solutions in the embodiments of the present invention are clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. 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.
As shown in fig. 1, the power data center station 100 of the present application includes the following components: receiver 110, memory 120, processor 130, output 140, wherein:
the receiver receives the trigger command and forwards the trigger command to the processor;
a memory storing historical data;
and a processor for executing the method of operating the station in the power data as described below.
And the output device outputs the analysis result.
Example one
As shown in fig. 2, a method for operating a power data center station includes the following steps:
step S110: reading the stored historical data in response to a trigger command;
the electric power data center station is also called a power grid data center station and is a unified platform for company-level electric power big data resources, data product display, sharing, cooperation, communication, transaction and capacity sharing services, and the electric power data center station stores a standardized and large amount of historical business data, algorithm models and operational capacity and provides effective support for electric quantity prediction models. The method extracts historical data and related information of a required area and forms a data source table through a standard SQL statement (SQL statement, namely structured query language, which is a database query and programming language and is used for accessing data and querying, updating and managing a relational database system) in the power data.
Because the SQL is very simple in commands for inquiring and modifying data and objects, the operation of acquiring the relevant data of the monthly electricity from the power data is simpler, and the formed data source table records the historical monthly electricity data and the relevant information of the required area in detail, so that the construction of a subsequent prediction model is facilitated.
Step S120: and inputting the read historical data into a pre-constructed combined prediction model for analysis to obtain an analysis result.
Step S130: and outputting an analysis result.
The combined prediction model is composed of a prediction model and a time model, and the prediction model and the time model are combined to obtain the combined prediction model.
The read historical data comprises sample data initial time and electric quantity original data corresponding to the sample data initial time, and a formula of a prediction model which is constructed in advance according to relevant data of monthly electric quantity is expressed as follows:
wherein, t is the predicted time,
is sample dataAt the initial time of the day, the user may,
to correspond to
The amount of electricity of (a) is raw,
is composed of
Approaching a theoretical extreme value of ∞ time (estimated by simulation from historical values),
for the undetermined coefficients, p (t) is the predicted electric quantity at the predicted time t.
In order to improve the accuracy of predicting the monthly electricity, because the monthly electricity is predicted by the method, when the prediction model is used for prediction, a corresponding prediction formula needs to be established for each month, and undetermined coefficients in the prediction model formulas of different months
In contrast, the coefficient to be determined is determined on a per-month basis
。
The historical data further comprises a trend component sequence, a seasonal period component sequence and a random component sequence of the monthly electricity quantity, and a specific expression of a time model constructed according to the relevant data of the monthly electricity quantity is as follows:
wherein D is a monthly electricity quantity sequence; d _ Q, D _ X, D _ S represents a trend component sequence, a seasonal period component sequence, and a random component sequence of the monthly power amount, respectively.
And combining the prediction model and the time model to obtain a combined prediction model.
The combined prediction model is obtained by combining the prediction model and the time model, so that the advantages of the prediction model and the time model are complemented, and the defect of a single prediction model in predicting monthly electric quantity is overcome. The specific method for obtaining the combined prediction model by combining the prediction model and the time model is as follows:
s121: respectively carrying out verification prediction on the monthly degrees of the existing sample data by using a prediction model and a time model, and calculating a first deviation matrix { alpha ] between the predicted value and the true value of the prediction model1,……,αnAnd a second deviation moment { beta } between the predicted value and the true value of the temporal model1,……,βn}；
S122: setting a first contribution rate vector { Ɛ for different months of the predictive model according to the first bias matrixiWhere i is a natural number from 1 to n, and setting a second contribution rate vector { Ɯ for different months of the time model according to a second deviation matrixjJ is a natural number from 1 to n;
Ɛi=αi-Di；
Ɯj=βi-Di
wherein Di is a set standard value;
s123: calculating to obtain a first final contribution rate alpha of the prediction model to the combined prediction model according to the first contribution rate vector, and calculating to obtain a second final contribution rate beta of the time model to the combined prediction model according to the second contribution rate vector;
wherein
Is ƐiIs given by the system in advance. Wherein
Is ƜjIs given by the system in advance. S124: and finally fusing the prediction model and the time model by using the first final contribution rate and the second final contribution rate to obtain a combined prediction model with the minimum prediction error.
And weighting the prediction results in the prediction model and the event model respectively by using the first final contribution rate and the second final contribution rate, and taking the weighted results as the output results of the final combined prediction model.
Under the condition that the obtained first final contribution rate and the second final contribution rate are harmonized with the prediction model and the time model, the two models are fused and complemented to minimize the prediction error of the combined prediction model.
Example two
The first embodiment describes a working method of a power data center station, and further, the method may further include the following steps:
and correcting the combined prediction model by using the historical data to obtain the optimized prediction model.
The prediction result of the combined prediction model is automatically compared with the real data of the monthly electric quantity of the electric power data, and the parameters of the combined prediction model are continuously adjusted until the deviation of the prediction result of the combined prediction model is within a preset range.
The electric power data center platform stores a large amount of real and effective relevant data of the monthly electric quantity, and the prediction result of the combined prediction model is compared with the historical data of the real monthly electric quantity to obtain the deviation of the prediction result. And analyzing errors, and performing model feedback and verification optimization by continuously adjusting relevant parameters of the combined prediction model until the deviation between the prediction result of the combined prediction model and the real historical data is smaller than a set threshold value, namely the deviation is within a preset range, namely the correction of the combined prediction model is completed, and the combined prediction model can be put into use.
The application also includes:
and performing later-stage optimization and flow processing on the combined prediction model.
And performing later-stage optimization and flow processing on the combined prediction model, namely storing the combined prediction model in a model algorithm library of the power data center, and adding an algorithm description document for business personnel to use.
The built and corrected combined prediction model is stored in a model algorithm library of the electric power data center for related business personnel to directly use, and an algorithm description document is added to the combined prediction model, so that the business personnel can use the model to predict monthly electric quantity conveniently.
The post-optimization and the process of the combined prediction model further comprise the steps of performing synchronous improvement and version updating on the combined prediction model by using the related data added in the post-power data.
Because the actual monthly electric quantity data can change along with the development of the society, the relevant data of the combined prediction model in the electric power data at the later stage of the use can be increased step by step, the newly increased data is utilized to further perfect and update the version of the combined prediction model, the condition that the combined prediction model which is already used deviates from the actual monthly electric quantity data due to the fact that the parameters of the combined prediction model are corrected only by the historical monthly electric quantity data is avoided, and therefore the prediction result of the combined prediction model is more accurate.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention. It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. A working method of a power data center station is characterized by comprising the following steps:
reading the stored historical data in response to a trigger command;
inputting the read historical data into a pre-constructed combined prediction model for analysis to obtain an analysis result;
and outputting an analysis result.
2. The method of operation of claim 1 wherein the combined predictive model is comprised of a predictive model and a temporal model, the predictive model and the temporal model being combined to obtain the combined predictive model.
3. The method of operation of claim 2, wherein the predictive model is constructed using the formula:
wherein, t is the predicted time,
as an initial time of the sample data,
to correspond to
The amount of electricity of (a) is raw,
is composed of
Approaching the theoretical extreme value of ∞ time,
and e is a constant.
4. The method of operation of claim 2, wherein the time model is constructed using the formula:
wherein D is a monthly electricity quantity sequence; d _ Q, D _ X, D _ S represents a trend component sequence, a seasonal period component sequence, and a random component sequence of the monthly power amount, respectively.
5. The method of operation of claim 2 wherein the combined predictive model is modified using historical data to produce an optimized predictive model.
6. A power data center station comprising the following components:
the receiver receives the trigger command and forwards the trigger command to the processor;
a memory storing historical data;
a processor that performs the steps of:
reading historical data stored by a memory in response to a trigger command;
inputting the read historical data into a pre-constructed combined prediction model for analysis to obtain an analysis result;
and the output device outputs the analysis result.
7. The power data center of claim 6, wherein the combined predictive model is comprised of a predictive model and a temporal model, the predictive model and the temporal model being combined to obtain the combined predictive model.
8. The power data middlebox of claim 7, wherein the predictive model is constructed using the formula:
wherein, t is the predicted time,
as an initial time of the sample data,
to correspond to
The amount of electricity of (a) is raw,
is composed of
Approaching the theoretical extreme value of ∞ time,
and e is a constant.
9. The power data middlebox of claim 7, wherein the time model is constructed using the following equation:
wherein D is a monthly electricity quantity sequence; d _ Q, D _ X, D _ S represents a trend component sequence, a seasonal period component sequence, and a random component sequence of the monthly power amount, respectively.
10. The power data center of claim 7, wherein the combined predictive model is modified using historical data to obtain an optimized predictive model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110911506.8A CN113361832B (en) | 2021-08-10 | 2021-08-10 | Electric power data center station and working method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110911506.8A CN113361832B (en) | 2021-08-10 | 2021-08-10 | Electric power data center station and working method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113361832A true CN113361832A (en) | 2021-09-07 |
| CN113361832B CN113361832B (en) | 2021-11-30 |
Family
ID=77540750
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110911506.8A Active CN113361832B (en) | 2021-08-10 | 2021-08-10 | Electric power data center station and working method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113361832B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109697527A (en) * | 2018-12-19 | 2019-04-30 | 浙江大学 | A kind of power predicating method of the various dimensions based on time series decomposition for trend |
| US20190251484A1 (en) * | 2018-02-13 | 2019-08-15 | The Trustees Of Indiana University | Forecasting and managing daily electrical maximum demands |
| CN110991700A (en) * | 2019-11-08 | 2020-04-10 | 北京博望华科科技有限公司 | Weather and electricity utilization correlation prediction method and device based on deep learning improvement |
| CN112163047A (en) * | 2020-09-21 | 2021-01-01 | 国家电网有限公司大数据中心 | Data center and computing equipment |
| CN112418921A (en) * | 2020-11-11 | 2021-02-26 | 深圳力维智联技术有限公司 | Power demand prediction method, device, system and computer storage medium |
| CN112988718A (en) * | 2021-05-21 | 2021-06-18 | 睿至科技集团有限公司 | Method and system for automatically monitoring power |
-
2021
- 2021-08-10 CN CN202110911506.8A patent/CN113361832B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190251484A1 (en) * | 2018-02-13 | 2019-08-15 | The Trustees Of Indiana University | Forecasting and managing daily electrical maximum demands |
| CN109697527A (en) * | 2018-12-19 | 2019-04-30 | 浙江大学 | A kind of power predicating method of the various dimensions based on time series decomposition for trend |
| CN110991700A (en) * | 2019-11-08 | 2020-04-10 | 北京博望华科科技有限公司 | Weather and electricity utilization correlation prediction method and device based on deep learning improvement |
| CN112163047A (en) * | 2020-09-21 | 2021-01-01 | 国家电网有限公司大数据中心 | Data center and computing equipment |
| CN112418921A (en) * | 2020-11-11 | 2021-02-26 | 深圳力维智联技术有限公司 | Power demand prediction method, device, system and computer storage medium |
| CN112988718A (en) * | 2021-05-21 | 2021-06-18 | 睿至科技集团有限公司 | Method and system for automatically monitoring power |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113361832B (en) | 2021-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2015282362B2 (en) | Power distribution transformer load prediction analysis system | |
| US10580095B2 (en) | Method and system for water production and distribution control | |
| US20120166616A1 (en) | System and method for energy performance management | |
| US10996943B2 (en) | Systems and methods for providing predictive quality analysis | |
| KR20150043338A (en) | Updating cached database query results | |
| CN107168977B (en) | Data query optimization method and device | |
| Ji et al. | CBR revision model for improving cost prediction accuracy in multifamily housing projects | |
| US20190086234A1 (en) | Systems and methods for displaying resource savings | |
| US20190066235A1 (en) | Systems and methods for energy management | |
| CN112150214A (en) | Data prediction method and device and computer readable storage medium | |
| CN113361832B (en) | Electric power data center station and working method | |
| CN112836370B (en) | Heating system scheduling method, apparatus, device, storage medium and program product | |
| CN109886737A (en) | Needing forecasting method, device, electronic equipment and readable storage medium storing program for executing | |
| CN112862189A (en) | Method, device, apparatus, storage medium, and program product for predicting heat source load | |
| US20200320632A1 (en) | Method and system for time series data quality management | |
| JP2023017784A (en) | Computer system and method for updating data aggregation configuration and controlling data aggregation | |
| CN116882767B (en) | Risk prediction method and device based on imperfect heterogeneous relation network diagram | |
| Casarin et al. | Monte carlo within simulated annealing for integral constrained optimizations | |
| CN109426887A (en) | A kind of prediction technique and device of hotel inventory | |
| CN103678636A (en) | System and method for improving reliability of component software system | |
| US20230305548A1 (en) | Generating forecasted emissions value modifications and monitoring for physical emissions sources utilizing machine-learning models | |
| Zeng et al. | Research on the Revision Method of Statistical Data | |
| CN110555746B (en) | Car insurance quotation method | |
| Namerikawa et al. | Filter‐Based Short‐Term Electric Load Prediction Considering Characteristics of Load Curve | |
| CN114706711A (en) | Parameter adjusting 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |