CN115713360A - Power market operation risk prediction method and device and storage medium - Google Patents
Power market operation risk prediction method and device and storage medium Download PDFInfo
- Publication number
- CN115713360A CN115713360A CN202211261437.1A CN202211261437A CN115713360A CN 115713360 A CN115713360 A CN 115713360A CN 202211261437 A CN202211261437 A CN 202211261437A CN 115713360 A CN115713360 A CN 115713360A
- Authority
- CN
- China
- Prior art keywords
- electric power
- data
- module
- power market
- electric
- 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
Images
Abstract
The invention discloses a method, a device and a storage medium for predicting the operation risk of an electric power market, wherein the method comprises the steps of acquiring electric power market data; screening electric power market data through linkage parameters input in advance; executing an Action module to obtain an initial evaluation value of the Jing electric index; after the Action module is executed, index calculation is carried out on the Beijing electric index initial evaluation value through the Action mapping module, a linkage result is output, and meanwhile, action configuration parameters are further read and input to the parameter prediction adjusting module for adjustment, so that the adjusted linkage requirement is obtained; the method has the advantages that the operation effect and risk assessment of the electric power market are completed, the analysis model is complete, the analysis speed is high, and accurate prediction can be performed on the results.
Description
Technical Field
The invention relates to a method and a device for predicting power market operation risk and a storage medium, and belongs to the technical field of power.
Background
The electric power market is just an electric power trading market as the name implies, with the deepening promotion of the electric power market construction in China, the market trading varieties are continuously increased, trading subjects are diversified day by day, trading periods are more flexible and changeable, and the market operation effect and the risk condition thereof are more difficult to master. The method is used for grasping and analyzing the effect and possible risks brought by market operation in time and is of great importance to the stable and ordered development of the market. The electric power market operation effect and risk assessment system needs huge data from different sources, analysis modeling needs to be carried out and result prediction needs to be carried out on the basis of the data from different sources, but the existing electric power market operation effect and risk assessment system has the problems that an analysis model is incomplete, analysis results are long in time consumption, results cannot be predicted and the like, and an electric power market operation risk prediction method needs to be researched to guarantee the correctness and the effectiveness of the market effect and risk assessment index system.
Disclosure of Invention
The present invention is directed to a method, an apparatus, and a storage medium for predicting an operation risk in an electric power market, so as to solve the problems set forth in the foregoing background.
In order to achieve the purpose, the invention adopts the following technical scheme:
in a first aspect, the present invention provides a method for predicting an operation risk of an electric power market, including:
acquiring electric power market data;
a data screening step, comprising: performing first data screening on the electric power market data through a linkage parameter input in advance;
calling a pre-constructed analysis library ActionForm module to analyze and verify the power market data after the first screening, and using the successfully verified data as an index to accord with the power market data imported into the Action module;
importing the electric power market data with indexes conforming to the indexes imported into the Action module into a pre-constructed Action module and operating the Action module to obtain an initial evaluation value of the Beijing power index;
index calculation is carried out on the Beijing electric index initial evaluation value through a pre-constructed ActionMapping module, and a Beijing electric index final linkage result is output;
reading configuration parameters of the Action module, inputting the configuration parameters to a pre-constructed parameter prediction adjusting module to adjust the electric quantity and the electricity price of each transaction variety type, and acquiring adjusted linkage parameters;
returning the adjusted linkage parameters as input to the data screening step, performing secondary data screening, and constructing a recursion process, wherein the recursion result of each time is the prediction result of the next round;
predicting the future electric power market operation risk through multiple rounds of prediction results obtained through multiple recursions;
the parameter prediction adjusting module further predicts and adjusts parameters by adopting a pre-constructed time sequence prediction algorithm model, and predicts the parameters by combining Action configuration parameters and parameters of the first data screening through a time sequence prediction algorithm.
Further, the linkage parameters comprise any one or more of electric power market delivery electric quantity, electric power market delivery price, large user direct transaction electric quantity, large user direct transaction electric price, power generation right electric quantity, power generation right electric price, spot transaction electric quantity, spot transaction electric price, pumping bidding electric quantity and pumping bidding electric price.
Further, the first data screening is performed on the electric power market data, and the first data screening specifically includes: and screening and removing garbage data and abnormal data in the electric power market data, and reserving accurate electric power market data.
Further, before the first data screening of the electric power market data, big data cleaning is carried out on the electric power market data, key data and information are extracted, correlated and stored as data which can be directly used for mining analysis.
Further, the analysis library ActionForm module which is constructed in advance is called to analyze and verify the electric power market data which are screened for the first time, and the data which are successfully verified are used as indexes to accord with the electric power market data which are led into the Action module; the method comprises the following steps:
calling a pre-constructed analysis library ActionForm module;
inputting the power market data after the first screening into an analysis library ActionForm module, and verifying whether the indexes of the power market data meet the verification indexes or not based on the analysis library ActionForm module;
if the verification result is not consistent with the verification result, the verification fails, an error prompt is given, and if the verification result is consistent with the verification result, the verification is successful;
and taking the successfully verified data as an index to accord with the electric power market data imported into the Action module.
Further, a standard value and a threshold value range of the requirements of an electric power market risk index algorithm, outgoing electric power charge, large user direct transaction electric charge, power generation right electric charge, spot transaction electric charge and pumping and bidding electric charge are stored in a pre-constructed Action module; importing the electric power market data with indexes conforming to the indexes imported into the Action module into a pre-constructed Action module and operating the Action module to obtain the Beijing electric power index initial evaluation value, wherein the method comprises the following steps:
importing the electric power market data with indexes conforming to the indexes imported into the Action module into the pre-constructed Action module;
and operating the Action module, carrying out comprehensive evaluation on the electric power market data with indexes conforming to the imported Action module through the Action module by adopting the electric power market risk index algorithm and combining the standard value and the threshold range, and obtaining the Beijing electric power index initial evaluation value.
Further, an index calculation algorithm is stored in the ActionMapping module in advance, and the index calculation of the kyoto electric index initial evaluation value through the pre-constructed ActionMapping module includes:
inputting the initial evaluation value of the Jing electric index into the index calculation algorithm to perform index calculation, and obtaining a final linkage result of the Jing electric index.
Further, the method for constructing the time sequence prediction algorithm model comprises the following steps:
establishing an exponential smoothing prediction method model, wherein a recursion formula of the exponential smoothing prediction method is as follows:
Y t+1 ＝αX t +α(1-α)X t-1 +α(1-α) 2 X t-2 +…+α(1-α) n X t-n (1.1)
in formula (1.1): y is t+1 Predicting the value by an exponential smoothing method; alpha is a weight coefficient, n is a smoothing index, t is a period number, X t 、X t-1 、X t-2 、…、X t-n Is the observed value of each phase;
and smoothing again on the basis of the first exponential smoothing to obtain a second exponential smoothing method, wherein the smoothing formula is as follows:
in formulae (1.2) and (1.3)
And
respectively the first and second exponential smoothing values of the t-th period;
X t the t =1,2,3, …, n, n is the number of original data; by
And
two quadratic exponential smoothing model parameters, at and bt, can be obtained:
the time sequence prediction algorithm model obtained by sorting is as follows: y is t+m ＝α t +b t X m, wherein: y is t+m Is the predicted value, and m is the predicted lead period number.
In a second aspect, the present invention provides an electric power market operation risk prediction device, including:
the data acquisition module is used for acquiring electric power market data;
the data screening module is used for carrying out primary data screening on the electric power market data through the linkage parameters input in advance;
the analysis and verification module is used for calling a pre-constructed analysis library ActionForm module to analyze and verify the electric power market data after the first screening, and the data which is successfully verified is used as an index to accord with the electric power market data led into the Action module;
the processing module is used for importing the electric power market data with indexes conforming to the indexes imported into the Action module into a pre-constructed Action module and operating the Action module to obtain an initial evaluation value of the Jing electric power index;
the calculation module is used for carrying out index calculation on the Jing electric index initial evaluation value through a pre-constructed ActionMapping module and outputting a Jing electric index final linkage result;
the adjusting module is used for reading the configuration parameters of the Action module and inputting the configuration parameters to the pre-constructed parameter prediction adjusting module to adjust the electric quantity and the electricity price of each transaction variety type, so as to obtain the adjusted linkage parameters; the parameter prediction adjusting module is used for further predicting and adjusting parameters by adopting a pre-constructed time sequence prediction algorithm model, and predicting through a time sequence prediction algorithm by combining Action configuration parameters and parameters of first data screening;
the recursion module is used for returning the adjusted linkage parameters as input to the data screening module, performing secondary data screening and constructing a recursion process, wherein each recursion result is a prediction result of the next round;
and the risk prediction module is used for predicting the future electric power market operation risk through multiple rounds of prediction results obtained by multiple recursions.
In a third aspect, the invention provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method of any one of the preceding claims.
Compared with the prior art, the invention has the following beneficial effects:
the invention provides a method, a device and a storage medium for predicting the operation risk of an electric power market, wherein an Action module is called to obtain an initial evaluation value of a Jing electric index, an Action mapping module is used for carrying out index calculation on the initial evaluation value of the Jing electric index, a final linkage result of the Jing electric index is output, a parameter adjusting module is used for adjusting, and finally the Action module and the parameter predicting and adjusting module are executed to carry out secondary data screening on the adjusted linkage requirement, so that a cycle is constructed, the operation effect and the risk evaluation of the electric power market are completed, an analysis model is complete, the analysis speed is high, and the result can be accurately predicted.
Drawings
Fig. 1 is a flowchart of a method for predicting an operation risk of an electric power market according to an embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
Example 1
The embodiment introduces an electric power market operation risk prediction method, which includes:
acquiring electric power market data;
a data screening step, comprising: performing first data screening on the electric power market data through a linkage parameter input in advance;
calling a pre-constructed analysis library ActionForm module to analyze and verify the power market data after the first screening, and using the successfully verified data as an index to accord with the power market data imported into the Action module;
importing the electric power market data with indexes conforming to the indexes imported into the Action module into a pre-constructed Action module and operating the Action module to obtain an initial evaluation value of the Jing electric index;
index calculation is carried out on the Beijing electric index initial evaluation value through a pre-constructed ActionMapping module, and a Beijing electric index final linkage result is output;
reading configuration parameters of the Action module, inputting the configuration parameters to a pre-constructed parameter prediction adjusting module to adjust the electric quantity and the electricity price of each transaction variety type, and acquiring adjusted linkage parameters;
returning the adjusted linkage parameters as input to the data screening step, performing secondary data screening, and constructing a recursion process, wherein the recursion result of each time is the prediction result of the next round;
predicting the future electric power market operation risk through multiple rounds of prediction results obtained through multiple recursions;
the parameter prediction adjusting module further predicts and adjusts parameters by adopting a pre-constructed time sequence prediction algorithm model, and predicts the parameters by combining Action configuration parameters and parameters of the first data screening through a time sequence prediction algorithm.
As shown in fig. 1, an application process of the power market operation risk prediction method provided by this embodiment specifically involves the following steps:
1) Carrying out prediction on Beijing power index theme indexes of power market operation risks, and taking the collected power market data as input;
2) Big data cleaning is carried out on the data; due to the existence of multi-source heterogeneous (such as structured, unstructured and semi-structured data) data sources, after data are acquired, one or more times of preprocessing are carried out, key data and information are extracted, correlated and stored as data which can be directly used for mining analysis. For example, performing text analysis on the operation log information of a trading platform or a trader, and extracting data; after the data storage is finished, the data which is already arranged and associated is further processed according to the requirement of data mining analysis, such as partial automatic calculation. Pre-processing may occur both before and after storage. The model and the structure of data storage also meet the requirement of mining analysis, and continuous optimization is carried out by taking the convenient and efficient search of required data as a target;
3) Preparing linkage parameters required by a Beijing electricity index in advance, wherein the linkage parameters comprise electricity delivery quantity of an electricity market, electricity delivery price of the electricity market, direct transaction quantity of a large user, direct transaction electricity price of the large user, electricity generation right quantity of electricity, electricity generation right price, spot transaction quantity of electricity, spot transaction electricity price, water pumping and bidding electricity quantity and water pumping and bidding electricity price, performing first-time data screening on detailed data of the electricity market through the pre-input linkage parameters, and reserving accurate electricity transaction data by screening and removing garbage data and abnormal data such as empty values and repeated data;
4) Then calling an analysis library ActionForm module for processing
A. The Actionform module mainly analyzes and verifies the data, and whether the indexes of the data meet the import Action is verified to be considered for processing.
B. If the verification is not in conformity with the Action module, the verification fails, an error prompt is given, and the Action module is further executed if the verification is successful.
The analysis library ActionForm module gives an error prompt for failure in processing the verification result, and further executes an Action module after the verification is successful, wherein the Action module is mainly used for comprehensively evaluating the power market risk; the Action module stores an electric power market risk index algorithm in advance, wherein the electric power charge risk evaluation algorithm of the Jing electric power index is electric power charge = electric power price (market average) electric quantity, electric quantity and electric power price, the Action module stores standard values and threshold values of the requirements of outgoing electric power charge, direct electric power charge transaction of a large user, power generation right (taking a positive value) electric power charge, spot transaction electric power charge and pumping and bidding electric power charge in advance, the standard value is 30000, the unit ten thousand yuan, the threshold value range is more than or equal to 0 and less than 1000 and is a reasonable range value, more than or equal to 1000 is a risk range value, and the Action module carries out comprehensive evaluation on data after primary screening by combining the standard value and the threshold value range to obtain the Jing electric power index initial evaluation value.
5) After the Action module is executed, skipping is carried out after processing is carried out through Action mapping, the Action mapping carries out index calculation on the Jing electric index evaluation value, the specific algorithm is (outgoing electric charge + large user direct transaction electric charge + power generation right (taking a positive value), electric charge + spot transaction electric charge + pumping and bidding electric charge)/19330953799 1000, a final linkage result of the Jing electric index is output through calculation, for example, the Jing electric index is 4000, the linkage result is output according to the numerical value requirement of the electric market on the Jing electric index subject index, and meanwhile, action configuration is further read and input to the parameter prediction adjusting module to adjust the electric quantity and the electric price of each transaction type;
6) The parameter prediction adjusting module is mainly used for further predicting and adjusting parameters by adopting a time sequence prediction algorithm model, and predicting by combining action configuration parameters of the previous round and parameters of data screening through a time sequence prediction algorithm.
7) And then screening data, calling an analysis library ActionForm module, executing an Action and a parameter prediction adjusting module, and further providing the adjusted linkage requirement for data screening to construct a cycle. Each cycle is a prediction for the next round. For example, after the first round of historical results comes out, the second round is to predict the results of the next year, and so on.
Wherein 7) the time sequence prediction algorithm of the parameter prediction adjustment module is specifically
By setting prediction parameters such as a prediction algorithm, an analysis graph, sample data and a fitting method, a conjecture period number, a confidence level and the like, a dimension value of the next period and a prediction index result are obtained according to a time sequence sample period rule, and the data change trend is observed. And in the first round, self-defined setting parameters are used for trend prediction to obtain a trend prediction change curve, so that the service development change trend is predicted.
The algorithm model is as follows:
firstly, an exponential smoothing prediction method model is established, which is a prediction method most suitable for predicting the trend of medium-short term time series data, and the unequal weight processing of the data at different times is more consistent with the actual situation. The exponential smoothing method can be generally classified into n-order exponential smoothing methods such as one-order, two-order, three-order, and the like, according to the number of smoothing times.
The recursion formula for the exponential smoothing method is given below:
Y t+1 ＝αX t +α(1-α)X t-1 +α(1-α) 2 X t-2 +…+α(1-α) n X t-n , (1.1)
in formula (1.1): y is t+1 Predicting a value by an exponential smoothing method; alpha is a weight coefficient, n is a smoothing exponent, t is a period number, X t 、X t-1 、X t-2 、…、X t-n Is the observed value of each period.
Parameter values corresponding to several rounds; the parameter is increased by one round for each round.
The recursion formula shows that the capability of adjusting the predicted value is strong; the information quantity contained in the predicted value is all historical data, and the weighting is characterized in that the weight closer to the prediction period is larger, and the weight farther away is smaller. The sum of the weights is 1; this changing trend can be quickly reflected in the exponential moving average.
And smoothing again on the basis of the first exponential smoothing to obtain a second exponential smoothing method, wherein the smoothing formula is as follows:
in formulae (1.2) and (1.3)
And
respectively the first and second exponential smoothing values of the t-th period;
X t the t =1,2,3, …, n, n is the number of original data; by
And
two quadratic exponential smoothing model parameters, at and bt, can be obtained:
and finally, sorting to obtain a time sequence prediction algorithm model as follows: y is t+m ＝α t +b t X m, wherein: y is t+m To predict the value, m is the predicted number of lead periods.
For example
The number of conjecture periods: the default is to predict 1 phase later, the minimum prediction phase is 1, and the maximum prediction phase is limited to 10. And (4) judging the period rule of the time series dimension value, and calculating the tolerance of the sample data by using the arithmetic progression to judge the dimension value of the next stage.
1) If the data has a periodic rule, displaying a next-stage dimension value according to the rule, namely 'the current maximum dimension value + tolerance', if the time sequence sample data is 2012, 2014, 2016 and 2018, then the next stage is 2020;
2) If the data does not have a periodic rule, the next period is "the current maximum dimension value +1", and if the time series sample data is 2011, 2014, 2015 and 2018, the next period is 2019 at this time.
Example 2
The present embodiment provides an electric power market operation risk prediction device, including:
the data acquisition module is used for acquiring electric power market data;
the data screening module is used for carrying out primary data screening on the electric power market data through the linkage parameters input in advance;
the analysis and verification module is used for calling a pre-constructed analysis library ActionForm module to perform analysis and verification on the power market data after the first screening, and taking the data which is successfully verified as an index to accord with the power market data which is imported into the Action module;
the processing module is used for importing the electric power market data with indexes conforming to the indexes imported into the Action module into a pre-constructed Action module and operating the Action module to obtain an initial evaluation value of the Jing electric power index;
the calculation module is used for carrying out index calculation on the Jing electric index initial evaluation value through a pre-constructed ActionMapping module and outputting a Jing electric index final linkage result;
the adjusting module is used for reading the configuration parameters of the Action module and inputting the configuration parameters to the pre-constructed parameter prediction adjusting module to adjust the electric quantity and the electricity price of each transaction variety type, so as to obtain the adjusted linkage parameters; the parameter prediction adjusting module is used for further predicting and adjusting parameters by adopting a pre-constructed time sequence prediction algorithm model, and predicting by combining Action configuration parameters and parameters of first data screening through a time sequence prediction algorithm;
the recursion module is used for returning the adjusted linkage parameters to the data screening module as input, performing secondary data screening and constructing a recursion process, wherein the recursion result of each time is the prediction result of the next round;
and the risk prediction module is used for predicting the future electric power market operation risk through multiple rounds of prediction results obtained by multiple recursions.
Example 3
The present embodiment provides a computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the steps of the method of any of the embodiment 1.
Example 4
An electronic device comprising a processor and a storage medium;
the storage medium is used for storing instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method according to any of embodiment 1.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.
Claims (10)
1. An electric power market operation risk prediction method is characterized by comprising the following steps:
acquiring electric power market data;
a data screening step, comprising: performing first data screening on the electric power market data through a linkage parameter input in advance;
calling a pre-constructed analysis library ActionForm module to analyze and verify the power market data after the first screening, and using the successfully verified data as an index to accord with the power market data imported into the Action module;
importing the electric power market data with indexes conforming to the indexes imported into the Action module into a pre-constructed Action module and operating the Action module to obtain an initial evaluation value of the Jing electric index;
index calculation is carried out on the Beijing electric index initial evaluation value through a pre-constructed ActionMapping module, and a Beijing electric index final linkage result is output;
reading configuration parameters of an Action module, inputting the configuration parameters to a pre-constructed parameter prediction adjusting module to adjust the electric quantity and the electricity price of each transaction variety type, and acquiring adjusted linkage parameters;
returning the adjusted linkage parameters as input to the data screening step, performing secondary data screening, and constructing a recursion process, wherein the recursion result of each time is the prediction result of the next round;
predicting the future electric power market operation risk through multiple rounds of prediction results obtained through multiple recursions;
the parameter prediction adjusting module further predicts and adjusts parameters by adopting a pre-constructed time sequence prediction algorithm model, and predicts through a time sequence prediction algorithm by combining Action configuration parameters and parameters of first data screening.
2. The electric power market operation risk prediction method according to claim 1, wherein the linkage parameters include any one or more of electric power market delivery electric quantity, electric power market delivery electric price, large user direct trading electric quantity, large user direct trading electric price, power generation right electric quantity, power generation right electric price, spot trading electric quantity, spot trading electric price, pumping bidding electric quantity, and pumping bidding electric price.
3. The electric power market operation risk prediction method according to claim 1, wherein the first data screening is performed on the electric power market data, specifically: and screening and removing garbage data and abnormal data in the electric power market data, and reserving accurate electric power market data.
4. The electric power market operation risk prediction method according to claim 1, wherein before the electric power market data is subjected to the first data screening, the electric power market data is subjected to big data cleaning, key data and information are extracted, and are associated and stored as data which can be directly used for mining analysis.
5. The electric power market operation risk prediction method according to claim 1, wherein the analysis library ActionForm module which is constructed in advance is called to analyze and verify the electric power market data which is screened for the first time, and the data which is successfully verified is used as an index to accord with the electric power market data which is imported into the Action module; the method comprises the following steps:
calling a pre-constructed analysis library ActionForm module;
inputting the electric power market data after the first screening into an analysis library ActionForm module, and verifying whether the indexes of the electric power market data meet the verification indexes or not based on the analysis library ActionForm module;
if the verification result is not consistent with the verification result, the verification fails, an error prompt is given, and if the verification result is consistent with the verification result, the verification is successful;
and taking the successfully verified data as an index to accord with the electric power market data imported into the Action module.
6. The electric power market operation risk prediction method according to claim 1, wherein a standard value and a threshold range of electric power market risk index algorithm, outgoing electric power charge, large user direct transaction electric charge, power generation right electric charge, spot transaction electric charge, pumping and bidding electric charge requirements are stored in a pre-constructed Action module; importing the electric power market data with indexes conforming to the indexes imported into the Action module into a pre-constructed Action module and operating the Action module to obtain the Beijing electric power index initial evaluation value, wherein the method comprises the following steps:
importing the electric power market data with indexes conforming to the indexes imported into the Action module into the pre-constructed Action module;
and operating the Action module, carrying out comprehensive evaluation on the electric power market data with indexes conforming to the imported Action module through the Action module by adopting the electric power market risk index algorithm and combining the standard value and the threshold range, and obtaining the Beijing electric power index initial evaluation value.
7. The electric power market operation risk prediction method according to claim 1, wherein an ActionMapping module stores an index calculation algorithm in advance, and performing index calculation on the kyoto electric power index initial evaluation value through the pre-constructed ActionMapping module includes:
inputting the initial evaluation value of the Jing electric index into the index calculation algorithm to perform index calculation, and obtaining a final linkage result of the Jing electric index.
8. The electric power market operation risk prediction method according to claim 1, wherein the time sequence prediction algorithm model is constructed by a method comprising:
establishing an exponential smoothing prediction method model, wherein a recursion formula of the exponential smoothing prediction method is as follows:
Y t+1 ＝αX t +α(1-α)X t-1 +α(1-α) 2 X t-2 +…+α(1-α) n X t-n (1.1)
in formula (1.1): y is t+1 Predicting the value by an exponential smoothing method; alpha is a weight coefficient, n is a smoothing exponent, t is a period number, X t 、X t-1 、X t-2 、…、X t-n Is the observed value of each phase;
and smoothing again on the basis of the first exponential smoothing to obtain a second exponential smoothing method, wherein the smoothing formula is as follows:
in formulae (1.2) and (1.3)
And
respectively the first and second exponential smoothing values of the t-th period;
X t the t-stage observed value is t =1,2,3, …, n, n is the number of the original data; by
And
two quadratic exponential smoothing model parameters, at and bt, can be obtained:
the time sequence prediction algorithm model obtained by sorting is as follows: y is t+m ＝α t +b t X m, wherein: y is t+m To predict the value, m is the predicted number of lead periods.
9. An electric power market operation risk prediction device, comprising:
the data acquisition module is used for acquiring power market data;
the data screening module is used for carrying out primary data screening on the electric power market data through the linkage parameters input in advance;
the analysis and verification module is used for calling a pre-constructed analysis library ActionForm module to perform analysis and verification on the power market data after the first screening, and taking the data which is successfully verified as an index to accord with the power market data which is imported into the Action module;
the processing module is used for importing the electric power market data with indexes conforming to the indexes imported into the Action module into a pre-constructed Action module and operating the Action module to obtain an initial evaluation value of the Jing electric index;
the calculation module is used for carrying out index calculation on the Jing electric index initial evaluation value through a pre-constructed ActionMapping module and outputting a Jing electric index final linkage result;
the adjusting module is used for reading the configuration parameters of the Action module and inputting the configuration parameters to the pre-constructed parameter prediction adjusting module to adjust the electric quantity and the electricity price of each transaction variety type, so as to obtain the adjusted linkage parameters; the parameter prediction adjusting module is used for further predicting and adjusting parameters by adopting a pre-constructed time sequence prediction algorithm model, and predicting by combining Action configuration parameters and parameters of first data screening through a time sequence prediction algorithm;
the recursion module is used for returning the adjusted linkage parameters as input to the data screening module, performing secondary data screening and constructing a recursion process, wherein each recursion result is a prediction result of the next round;
and the risk prediction module is used for predicting the future electric power market operation risk through multiple rounds of prediction results obtained by multiple recursions.
10. A computer-readable storage medium having stored thereon a computer program, characterized in that: the program when executed by a processor implements the steps of the method of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211261437.1A CN115713360A (en) | 2022-10-14 | 2022-10-14 | Power market operation risk prediction method and device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211261437.1A CN115713360A (en) | 2022-10-14 | 2022-10-14 | Power market operation risk prediction method and device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115713360A true CN115713360A (en) | 2023-02-24 |
Family
ID=85231241
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211261437.1A Pending CN115713360A (en) | 2022-10-14 | 2022-10-14 | Power market operation risk prediction method and device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115713360A (en) |
-
2022
- 2022-10-14 CN CN202211261437.1A patent/CN115713360A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112910690A (en) | Network traffic prediction method, device and equipment based on neural network model | |
| CN115600824B (en) | Carbon emission early warning method and device, storage medium and electronic equipment | |
| CN110647995A (en) | Rule training method, device, equipment and storage medium | |
| CN111199469A (en) | User payment model generation method and device and electronic equipment | |
| CN116307215A (en) | Load prediction method, device, equipment and storage medium of power system | |
| CN111210332A (en) | Method and device for generating post-loan management strategy and electronic equipment | |
| CN114548494A (en) | Visual cost data prediction intelligent analysis system | |
| CN113256036A (en) | Power supply cost analysis and prediction method based on Prophet-LSTNet combined model | |
| CN117596191A (en) | Power Internet of things abnormality detection method, device and storage medium | |
| CN113033898A (en) | Electrical load prediction method and system based on K-means clustering and BI-LSTM neural network | |
| CN116706884A (en) | Photovoltaic power generation amount prediction method, device, terminal and storage medium | |
| CN112595918A (en) | Low-voltage meter reading fault detection method and device | |
| CN111475548A (en) | Power utilization abnormity analysis decision system based on big data mining technology | |
| CN116561569A (en) | Industrial power load identification method based on EO feature selection and AdaBoost algorithm | |
| CN111080037A (en) | Short-term power load prediction method and device based on deep neural network | |
| CN115713360A (en) | Power market operation risk prediction method and device and storage medium | |
| CN113704407B (en) | Complaint volume analysis method, device, equipment and storage medium based on category analysis | |
| CN115616333A (en) | Power distribution network line loss prediction method and system | |
| CN115528684A (en) | Ultra-short-term load prediction method and device and electronic equipment | |
| CN114169923A (en) | Photovoltaic power station power prediction method, system, terminal and storage medium | |
| CN114399076A (en) | Method for screening electricity stealing suspicion users based on big data analysis | |
| CN117435870B (en) | Load data real-time filling method, system, equipment and medium | |
| CN109753497A (en) | Data processing empty value method, apparatus and terminal device | |
| CN116226687B (en) | Reservoir daily initial water level estimation method and device based on data mining technology | |
| CN117332893A (en) | Method and system for generating electric power declaration data |
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 |