CN114336603A - Power grid side coal-electricity fuel inventory prediction system and method - Google Patents
Power grid side coal-electricity fuel inventory prediction system and method Download PDFInfo
- Publication number
- CN114336603A CN114336603A CN202111623103.XA CN202111623103A CN114336603A CN 114336603 A CN114336603 A CN 114336603A CN 202111623103 A CN202111623103 A CN 202111623103A CN 114336603 A CN114336603 A CN 114336603A
- Authority
- CN
- China
- Prior art keywords
- coal
- electricity
- fuel
- power grid
- 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.)
- Pending
Links
Images
Classifications
-
- 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 invention discloses a system and a method for predicting a power grid side coal-electricity fuel inventory in the technical field of power grid side coal-electricity fuel information management, and the method for predicting the power grid side coal-electricity fuel inventory comprises the following steps: acquiring original data of a coal-electricity fuel stock; decomposing original data of a coal-electricity fuel stock to obtain a plurality of modal components; recombining each modal component to obtain a plurality of recombined components; establishing a respective depth belief network prediction model of each weight group; recombining all the deep belief network prediction models to obtain a final prediction model; and performing fuel inventory prediction based on the final prediction model. The invention can provide a basis for the coal-electricity enterprises to keep proper fuel stock by predicting the load and the fuel consumption of the power grid, is beneficial to the safe operation of the power grid, and simultaneously reduces the operation cost of the coal-electricity enterprises.
Description
Technical Field
The invention relates to a power grid side coal-electricity fuel inventory prediction system and method, and belongs to the technical field of power grid side coal-electricity fuel information management.
Background
The current carbon peak reaching and carbon neutral climate commitment is formally brought into the economic and social development planning and ecological civilization prospect construction overall layout of China. In order to realize the double-carbon target, a novel power system taking new energy as a main body is built, meanwhile, efficient and clean coal and electricity are guaranteed to be required to be developed, and stable operation of the coal and electricity provides a foundation for development of the new energy. By the end of 2020, the installed capacity of coal electricity in China accounts for 49.1 percent of the total installed capacity and reaches about 10.8 hundred million kilowatts, the generated energy accounts for about 70 percent of the whole power grid, the coal electricity is still the main body of power supply, and the safe and stable operation of the coal electricity is important for the power grid. The fuel is the core problem of coal electricity, is influenced by global epidemic situation development and economic situation, coal price fluctuates greatly, and certain randomness and uncertainty exist in coal supply, so that the phenomena of recent tense electricity utilization situation, limited electricity and limited production and the like are caused, and the national economic development is influenced. Meanwhile, under the condition of a double-carbon policy, new energy such as wind power, photovoltaic and the like is further built, the proportion of the new energy in a power grid is gradually increased, coal power is gradually developed into peak load regulation, and the uncertainty of fuel consumption is further brought by deep peak regulation of the coal power. The specific gravity of installed capacity of water and electricity is large by taking a southern power grid as an example (hereinafter referred to as a southern power grid), the water and electricity have obvious seasonal distribution, and new energy and water and electricity loads are overlapped, so that the power generation load in the southern power grid range shows more obvious climatic and seasonal laws. The balance between loads on the power utilization side and the power generation side is important for the safety of a power grid, the real-time prediction and research on the loads on the power utilization side are quite sufficient at present, and the changes of short-term and long-term power utilization loads can be basically and accurately predicted. In the south grid range, the load margin of a generator set at the power generation side is large, the problem of load scheduling at the power generation side does not exist generally, but under extreme conditions, the loads of new energy and water and electricity are limited, and the coal and electricity are required to bear the top load of a power grid. At the moment, the coal-electricity fuel stock is the most important factor influencing the coal-electricity output, and if the stock is insufficient, large-scale electricity limiting and power failure accidents of the power grid similar to the disaster of ice in the south in 08 years can be caused, so that great economic loss is caused. Therefore, the safety of the power grid is ensured, and the fuel stock condition of the coal-electricity enterprise is also a factor to be considered in important consideration for load scheduling on the power generation side of the power grid.
The fuel inventory is the core problem of safe operation of coal and electricity, and each power generation group and coal and electricity enterprise are provided with fuel informatization systems for enhancing the fuel inventory management. The information systems realize better control on the fuel, but after the plant network is separated, for the power grid, the coal-electricity fuel management system is an information isolated island, and the fuel information of the thermal power enterprise is in a scheduling blind area, so that the lack of unified management and prediction analysis of the coal-electricity fuel inventory in the south network range is caused. The coal-electricity fuel management system of the south network is constructed by utilizing the internet + technology, the coal-electricity fuel information in the south network range is unified to a comprehensive management platform, the coal-electricity information in the whole network range can be managed and analyzed, meanwhile, the fuel stock is predicted to a certain extent, and the fuel stock accidents are prevented. For coal-electricity enterprises, too high fuel stock causes overlarge coal-electricity capital pressure, which brings difficulty to operation, and too small stock is not beneficial to the safety of a power grid.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a system and a method for predicting the coal-electricity fuel inventory on the power grid side.
In order to achieve the purpose, the invention is realized by adopting the following technical scheme:
in a first aspect, the present invention provides a power grid side coal-electricity fuel inventory prediction method, including:
acquiring original data of a coal-electricity fuel stock;
decomposing original data of a coal-electricity fuel stock to obtain a plurality of modal components;
recombining each modal component to obtain a plurality of recombined components;
establishing a respective depth belief network prediction model of each weight group;
recombining all the deep belief network prediction models to obtain a final prediction model;
and performing fuel inventory prediction based on the final prediction model.
Further, decomposing the original data of the coal-electricity fuel stock to obtain a plurality of modal components, wherein the expression is as follows:
wherein X (t) is the original data of the coal-electricity fuel stock, r (t) is the final residual signal of decomposition,
for the Kth modal component, the expression is:
wherein r isk-1(t) is the k-1 th residual signal, εk-1Is the k-1 th Gaussian white noise amplitude constant, Ek-1The k-1 EMD decomposition operator.
Further, recombining each modal component to obtain a plurality of recombined components, including:
performing complexity evaluation on each modal component through sample entropy;
and recombining the mode components according to the complexity to obtain a plurality of recombined components.
Further, the sample entropy is:
wherein SE (m, r, N) is the sample entropy value; m is the number of dimensions; b ism(r) is the probability that a sequence matches m points with a similarity tolerance r.
Further, the probability B that the sequence matches m points with a similarity tolerance rm(r) is obtained by calculating the number of absolute values of the maximum value of the difference between the corresponding elements in the m-dimensional vector group, which is smaller than a threshold, and the expression is as follows:
d[Xm(i),Xm(j)]＝max[|x(i+k)-x(j+k)|]
Xm(i)＝{x(i),x(i+1),|,x(i+m-1)}
Xm(j)＝{x(j),x(j+1),…,x(j+m-1)}
wherein N is the number of repeated decompositions and m is the number of dimensions,
is the probability that a sequence matches i of m points under a similar tolerance (i.e., threshold r), Xm(i) And Xm(j) Is a vector group with m dimensions, and k is the number of vector elements.
Further, the deep belief network model includes a limited boltzmann machine, and an energy function, a visible layer probability distribution function and a likelihood function of the limited boltzmann machine are respectively:
L＝lnP(V)
wherein E (v, h) is an energy function, p (V) is a visible layer probability distribution function, L is a likelihood function, wijFor connecting between hidden layer and display layerValue viTo show layer, aiFor display layer bias, hi、hjAre all hidden layers, biFor hidden layer biasing, m, n are network dimensions.
Further, the gradient of the likelihood function is obtained by using a contrast divergence algorithm with one-step gibbs sampling, and the expression is as follows:
In a second aspect, the present invention provides a grid-side coal-to-electricity fuel inventory prediction system, comprising:
a data acquisition module: the system is used for acquiring the raw data of the coal-electricity fuel stock;
a decomposition module: the system comprises a data acquisition module, a data processing module and a data processing module, wherein the data acquisition module is used for acquiring data of a coal-electricity fuel stock;
a recombination module: the system is used for recombining each modal component to obtain a plurality of recombined components;
a modeling module: the deep belief network prediction model is used for establishing a respective deep belief network prediction model of each weight group;
a superposition module: the prediction model is used for recombining all the deep belief network prediction models to obtain a final prediction model;
a prediction module: for fuel inventory prediction based on the final prediction model.
In a third aspect, the invention provides a power grid side coal-electricity fuel inventory predicting device, which comprises 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 the above.
In a fourth 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 any of the methods described above.
Compared with the prior art, the invention has the following beneficial effects:
according to the method, the basis can be provided for the coal-electricity enterprises to keep proper fuel stocks through forecasting the load and the fuel consumption of the power grid, the safe operation of the power grid is facilitated, the operation cost of the coal-electricity enterprises is reduced, the CEEMDAN method is adopted to decompose the original data of the coal-electricity fuel stocks into a plurality of Intrinsic Mode Functions (IMFs), and each IMF represents different oscillation components. The complexity of each IMF is calculated by utilizing the sample entropy, the IMFs are recombined according to the sample entropy, the model complexity and the calculated amount are reduced, the calculation efficiency is improved, the DBN prediction models of the decomposed and recombined recombinant components are respectively established, the final prediction model is formed by overlapping the prediction models of the recombinant components, and compared with a single prediction model, the established final prediction model has higher prediction accuracy.
Drawings
Fig. 1 is a schematic diagram of a power grid side fuel management integrated management platform according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a CEEMDAN-SE and DBN-based coal-electricity fuel inventory model according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating a decomposition of a CEEMDAN modal component of an inventory signal according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of modal components and residual sample entropy provided by an embodiment of the present invention;
FIG. 5 is a schematic diagram of modal component reorganization according to an embodiment of the present invention;
fig. 6 is a schematic view of an RBM structure according to an embodiment of the invention;
FIG. 7 is a diagram illustrating a comparison between a DBN model fuel inventory and actual values 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.
The first embodiment is as follows:
the embodiment provides a CEEMDAN-SE (adaptive noise-based complete set empirical mode decomposition sample entropy) fused Deep Belief Network (DBN) south network power generation side fuel combination prediction model. Referring to fig. 1, the collection terminal and the collection network in fig. 1 are disposed in each coal-electricity enterprise, and are data interfaces between the integrated management platform and the coal-electricity enterprise, and are mainly responsible for collecting fuel inventory data from the coal-electricity fuel management system and uploading the data to the integrated platform. In order to ensure the information security of the platform, the server is set in a layered and partitioned manner, and is provided with two servers which are respectively a main server and a DMZ server, wherein the main server is arranged in a security three-zone, the DMZ server is arranged in an isolation zone, and a firewall is arranged between the DMZ server and the main server to ensure the data security of the main server. The DMZ server is a main server for safe backup and provides intranet and extranet data access services.
Referring to FIG. 2, FIG. 2 is a model for coal-to-electricity fuel inventory prediction in the south grid area based on CEEMDAN-SE and DBN. The comprehensive fuel management system on the south grid power generation side established in fig. 1 collects the original inventory data of coal-electricity fuels in the past year and other auxiliary information data, such as meteorological data and the like. The CEEMDAN method is used to decompose raw data of a coal-electricity fuel stock into a plurality of Intrinsic Mode Functions (IMFs), wherein each IMF represents different oscillation components. And calculating the complexity of each IMF by using the sample entropy, recombining the IMFs according to the sample entropy, reducing the model complexity and the calculated amount, and improving the calculation efficiency. And respectively establishing a DBN prediction model for each weight, wherein the final prediction model is formed by overlapping the prediction models of the weight, and compared with a single prediction model, the established final prediction model has higher prediction precision.
Constructing an original curve X (t) +. epsilon of an original coal-electricity fuel stock0ni(t) wherein ε0Is an initial constant of Gaussian white noise amplitude, X (t) is raw data of a coal-electric fuel stock, ni(t) is white gaussian noise satisfying a standard normal distribution;
performing N times of repeated decomposition on the original data of the coal-electricity fuel stock by adopting an EMD method, and obtaining a first modal component by calculating an average value
Comprises the following steps:
wherein N is the number of repeated decompositions,
is EMD single decomposition modal component;
obtaining a first residual signal r1(t) is:
construction of the sequence r1(t)+ε1E1(ni(t)), continuing the EMD iterative decomposition of the sequence N times to obtain a second modal component
Comprises the following steps:
wherein E is1For the first EMD decomposition operator, ∈1Is a first gaussian white noise amplitude constant;
obtaining a second residual signal r2(t) is:
repeating the above steps, and obtaining the Kth modal component when the residual signal meets the termination decomposition condition (namely the residual signal is a monotone function)
Wherein r isk-1(t) is the k-1 th residual signal, εk-1Is the k-1 th Gaussian white noise amplitude constant, Ek-1The k-1 EMD decomposition operator.
Decompose the final residual signal r (t) into:
through the above method, the raw data x (t) of the coal-electricity fuel stock can be decomposed into:
according to the characteristic, original fuel stock data X (t) of the coal-electricity fuel stock are decomposed by CEEMDAN to obtain a plurality of modal components, namely 365 continuous data points are decomposed. The white noise group number NR is 100 and the standard deviation Nstd is 0.2, the stock signal sequence is decomposed into 7 modal components with different characteristics and 1 residual signal by the CEEMDAN algorithm, and the decomposition result is shown in fig. 3.
The sample entropy calculation method is as follows:
sequentially reconstructing a group of m-dimensional vectors, namely X, from raw data X (t) of the coal-electricity fuel stockm(i) And Xm(j)：
Xm(i)＝{x(i),x(i+1),…,x(i+m-1)}
Xm(j)＝{x(j),x(j+1),…,x(j+m-1)} (8)
Definition vector Xm(i) And Xm(j) The absolute value of the maximum value of the difference between the corresponding elements is the distance, wherein k is the number of the vector elements:
d[Xm(i),Xm(j)]＝max[|x(i+k)-x(j+k)|] (9)
setting a threshold r, calculating d [ X ]m(i),Xm(j)]The number of r is calculated to obtain Bm(r)，Bm(r) is the probability that a sequence matches m points with a similarity tolerance (i.e. threshold r),
the probability that a sequence matches i of m points under a similar tolerance (i.e., threshold r);
wherein N is the repeated decomposition times, and m is the dimension;
repeating the steps, and calculating to obtain a theoretical sample entropy value SE (m, r) as follows:
for finite N, the sample entropy approximation SE (m, r, N) is:
the complexity (i.e., sample entropy) of each IMF component in fig. 3 is evaluated by using sample entropy, the embedding dimension (m) is taken as 2 in the calculation process, the similarity tolerance is taken as 0.25, and the obtained IMF entropy curve is as shown in fig. 4.
And recombining the components with similar complexity, wherein IMF1, IMF2 and IMF3 are recombined into components F1, IMF6 and IMF7 are recombined into components F4, and as a result, as shown in FIG. 5, F2, F3 and F5 are all modal components.
The DBN model is built for each of the reconstruction components of fig. 5, and then the models are reconstructed. It should be noted that the Deep Belief Network (DBN) is essentially a probabilistic deep network combining unsupervised learning and supervised learning, and is formed by stacking Restricted Boltzmann Machines (RBMs). The training of the DBN model is mainly divided into two steps of unsupervised training and BP fine adjustment. Obtaining the initial values of the RBM pre-training network by adopting an unsupervised greed layer-by-layer algorithm; the BP network pair is then used to reverse fine tune network parameters with supervision. As shown in FIG. 6, the RBM, which is an important component of the DBN, is composed of a visible layer and a hidden layer. In RBM, the energy function E (v, h) is defined as:
in the formula: w is aijThe connection weight between the hidden layer and the display layer; v. ofiTo show layer, aiBiasing for the display layer; biFor hidden layer bias, hi、hjAre hidden layers, and m and n are network dimensions. The visible layer probability distribution function p (V) is:
the RBM weight parameter w, the visible layer bias parameter a, and the hidden layer bias parameter b set θ ═ w, a, b, the likelihood function given by the equation needs to be maximized, and the likelihood function form is defined as:
L＝lnP(V) (15)
the gradient of the likelihood function is obtained using a Contrast Divergence (CD) algorithm with one step Gibbs sampling (Gibbs):
The recombined model is used for predicting the fuel stock of the power generation side of the south China network 2020, and the prediction result is shown in FIG. 7.
According to the method, the load data of the south power grid is used as the basis, the prediction result shows that the model has high prediction precision, the prediction model reduces the inventory cost of coal-electricity fuel, improves the fuel safety, and further improves the safety of the power grid.
Example two:
a power grid side coal-electricity fuel inventory prediction system can realize a power grid side coal-electricity fuel inventory prediction method in the first embodiment, and comprises the following steps:
a data acquisition module: the system is used for acquiring the raw data of the coal-electricity fuel stock;
a decomposition module: the system comprises a data acquisition module, a data processing module and a data processing module, wherein the data acquisition module is used for acquiring data of a coal-electricity fuel stock;
a recombination module: the system is used for recombining each modal component to obtain a plurality of recombined components;
a modeling module: the deep belief network prediction model is used for establishing a respective deep belief network prediction model of each weight group;
a superposition module: the prediction model is used for recombining all the deep belief network prediction models to obtain a final prediction model;
a prediction module: for fuel inventory prediction based on the final prediction model.
Example three:
the embodiment of the invention also provides a device for predicting the coal-electricity fuel stock on the power grid side, which can realize the method for predicting the coal-electricity fuel stock on the power grid side in the embodiment one and comprises 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 of:
acquiring original data of a coal-electricity fuel stock;
decomposing original data of a coal-electricity fuel stock to obtain a plurality of modal components;
recombining each modal component to obtain a plurality of recombined components;
establishing a respective depth belief network prediction model of each weight group;
recombining all the deep belief network prediction models to obtain a final prediction model;
and performing fuel inventory prediction based on the final prediction model.
Example four:
the embodiment of the present invention further provides a computer-readable storage medium, which can implement the method for predicting a grid-side coal-electric fuel inventory in the first embodiment, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the computer program implements the following steps of the method:
acquiring original data of a coal-electricity fuel stock;
decomposing original data of a coal-electricity fuel stock to obtain a plurality of modal components;
recombining each modal component to obtain a plurality of recombined components;
establishing a respective depth belief network prediction model of each weight group;
recombining all the deep belief network prediction models to obtain a final prediction model;
and performing fuel inventory prediction based on the final prediction model.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
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. A power grid side coal-electricity fuel inventory prediction method is characterized by comprising the following steps:
acquiring original data of a coal-electricity fuel stock;
decomposing original data of a coal-electricity fuel stock to obtain a plurality of modal components;
recombining each modal component to obtain a plurality of recombined components;
establishing a respective depth belief network prediction model of each weight group;
recombining all the deep belief network prediction models to obtain a final prediction model;
and performing fuel inventory prediction based on the final prediction model.
2. The grid-side coal-to-electricity fuel inventory prediction method according to claim 1,
decomposing original data of the coal-electricity fuel stock to obtain a plurality of modal components, wherein the expression is as follows:
wherein X (t) is the original data of the coal-electricity fuel stock, r (t) is the final residual signal of decomposition,
for the Kth modal component, the expression is:
wherein r isk-1(t) is the k-1 th residual signal, εk-1Is the k-1 th Gaussian white noise amplitude constant, Ek-1The k-1 EMD decomposition operator.
3. The grid-side coal-to-electricity fuel inventory prediction method according to claim 1,
recombining each modal component to obtain a plurality of recombined components, including:
performing complexity evaluation on each modal component through sample entropy;
and recombining the mode components according to the complexity to obtain a plurality of recombined components.
4. The grid-side coal-to-electricity fuel inventory prediction method according to claim 3,
the sample entropy is:
wherein SE (m, r, N) is the sample entropy value; m is the number of dimensions; b ism(r) is the probability that a sequence matches m points with a similarity tolerance r.
5. The grid-side coal-electric fuel inventory prediction method as claimed in claim 4, wherein the sequence has a probability B of matching m points under a similarity tolerance rm(r) is obtained by calculating the number of absolute values of the maximum value of the difference between the corresponding elements in the m-dimensional vector group, which is smaller than a threshold, and the expression is as follows:
d[Xm(i),Xm(j)]＝max[|x(i+k)-x(j+k)|]
Xm(i)＝{x(i),x(i+1),…,x(i+m-1)}
Xm(j)＝{x(j),x(j+1),…,x(j+m-1)}
wherein N is the number of repeated decompositions and m is the number of dimensions,
is the probability that a sequence matches i of m points under a similar tolerance (i.e., threshold r), Xm(i) And Xm(j) Is a vector group with m dimensions, and k is the number of vector elements.
6. The grid-side coal-electric fuel inventory prediction method according to claim 1, wherein the deep belief network model comprises a constrained boltzmann machine whose energy function, visible layer probability distribution function, and likelihood function are respectively:
L＝lnP(V)
wherein E (v, h) is an energy function, p (V) is a visible layer probability distribution function, L is a likelihood function, wijIs the connection weight between the hidden layer and the display layer, viTo show layer, aiFor display layer bias, hi、hjAre all hidden layers, biFor hidden layer biasing, m, n are network dimensions.
7. The power grid side coal-electric fuel inventory prediction method according to claim 6, characterized in that the gradient of the likelihood function is obtained by a contrast divergence algorithm with one-step Gibbs sampling, and the expression is as follows:
8. A power grid side coal-electricity fuel inventory prediction system is characterized by comprising:
a data acquisition module: the system is used for acquiring the raw data of the coal-electricity fuel stock;
a decomposition module: the system comprises a data acquisition module, a data processing module and a data processing module, wherein the data acquisition module is used for acquiring data of a coal-electricity fuel stock;
a recombination module: the system is used for recombining each modal component to obtain a plurality of recombined components;
a modeling module: the deep belief network prediction model is used for establishing a respective deep belief network prediction model of each weight group;
a superposition module: the prediction model is used for recombining all the deep belief network prediction models to obtain a final prediction model;
a prediction module: for fuel inventory prediction based on the final prediction model.
9. A power grid side coal-electricity fuel inventory prediction device is characterized by 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 one of claims 1 to 7.
10. Computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111623103.XA CN114336603A (en) | 2021-12-28 | 2021-12-28 | Power grid side coal-electricity fuel inventory prediction system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111623103.XA CN114336603A (en) | 2021-12-28 | 2021-12-28 | Power grid side coal-electricity fuel inventory prediction system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114336603A true CN114336603A (en) | 2022-04-12 |
Family
ID=81015162
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111623103.XA Pending CN114336603A (en) | 2021-12-28 | 2021-12-28 | Power grid side coal-electricity fuel inventory prediction system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114336603A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292453A (en) * | 2017-07-24 | 2017-10-24 | 国网江苏省电力公司电力科学研究院 | A kind of short-term wind power prediction method based on integrated empirical mode decomposition Yu depth belief network |
| CN112734333A (en) * | 2020-12-31 | 2021-04-30 | 华润电力技术研究院有限公司 | Method, device and equipment for scheduling coal fuel inventory |
| CN112785345A (en) * | 2021-02-03 | 2021-05-11 | 贵州乌江水电开发有限责任公司 | All-round power generation fuel management system |
-
2021
- 2021-12-28 CN CN202111623103.XA patent/CN114336603A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292453A (en) * | 2017-07-24 | 2017-10-24 | 国网江苏省电力公司电力科学研究院 | A kind of short-term wind power prediction method based on integrated empirical mode decomposition Yu depth belief network |
| CN112734333A (en) * | 2020-12-31 | 2021-04-30 | 华润电力技术研究院有限公司 | Method, device and equipment for scheduling coal fuel inventory |
| CN112785345A (en) * | 2021-02-03 | 2021-05-11 | 贵州乌江水电开发有限责任公司 | All-round power generation fuel management system |
Non-Patent Citations (1)
| Title |
|---|
| 岳有军;刘英翰;赵辉;王红君: "基于CEEMDAN-SE和DBN的短期电力负荷预测", 电测与仪表 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chang | An RBF neural network combined with OLS algorithm and genetic algorithm for short-term wind power forecasting | |
| Musbah et al. | Energy management using multi-criteria decision making and machine learning classification algorithms for intelligent system | |
| CN109412161B (en) | Power system probability load flow calculation method and system | |
| Xiong et al. | A short-term wind power forecast method via XGBoost hyper-parameters optimization | |
| CN114048896B (en) | Method, system, equipment and medium for predicting photovoltaic power generation data | |
| CN110717610A (en) | Wind power prediction method based on data mining | |
| Li et al. | Short-term power generation forecasting of a photovoltaic plant based on PSO-BP and GA-BP neural networks | |
| CN112418526A (en) | Comprehensive energy load control method and device based on improved deep belief network | |
| Chen et al. | Research on wind power prediction method based on convolutional neural network and genetic algorithm | |
| CN113011680A (en) | Power load prediction method and system | |
| Jiang et al. | Research and application of a new hybrid wind speed forecasting model on BSO algorithm | |
| Zhou et al. | Coal consumption prediction in thermal power units: A feature construction and selection method | |
| Liu et al. | Super-resolution perception for wind power forecasting by enhancing historical data | |
| CN117200352A (en) | Photovoltaic power generation regulation and control method and system based on cloud edge fusion | |
| Wang et al. | An experimental investigation of FNN model for wind speed forecasting using EEMD and CS | |
| CN114336603A (en) | Power grid side coal-electricity fuel inventory prediction system and method | |
| Hiba et al. | Stacked Denoising Autoencoder network for short-term prediction of electrical Algerian load | |
| Xiaosheng et al. | A deep learning approach for wind power prediction based on stacked denoising auto encoders optimized by bat algorithm | |
| Kazmi et al. | A Concurrent CNN-RNN Approach for Multi-Step Wind Power Forecasting | |
| Yuan et al. | A novel hybrid short-term wind power prediction framework based on singular spectrum analysis and deep belief network utilized improved adaptive genetic algorithm | |
| Li et al. | EMLP: short-term gas load forecasting based on ensemble multilayer perceptron with adaptive weight correction | |
| CN113536508A (en) | Method and system for classifying manufacturing network nodes | |
| Lu et al. | Enhanced grey theory system model with weight table for now-casting PV power generation forecast | |
| Almohri et al. | Data Analysis and Prediction of Power Generated by Photovoltaic Systems | |
| Tsianikas et al. | Comparison of neural network based approaches for short-term residential energy load forecasting |
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 |