Distributed photovoltaic power station day-ahead output prediction method based on space-time correlation model
A space-time correlation, distributed photovoltaic technology, applied in the direction of electrical digital data processing, data processing applications, instruments, etc., can solve the problems of limited prediction accuracy, poor robustness, and high dependence on mathematical models of urban weather forecast information, and achieve the elimination of Dependency, guaranteed efficiency and privacy, and the effect of improving prediction accuracy
Pending Publication Date: 2022-06-21
HUNAN UNIV
View PDF1 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] The characteristics of distributed photovoltaic power plants lead to three problems in day-ahead forecasting tasks: 1) Due to the spatial resolution, the prediction accuracy that urban weather forecast information can improve is very limited
This scheme has a high dependence on the mathematical model, resulting in poor robustness
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0073] The present invention will be further described below in conjunction with the accompanying drawings and specific preferred embodiments, but the protection scope of the present invention is not limited thereby.
[0074] like figure 1 As shown, the idea of this scheme is: the server side generates linear trend scenarios for each distributed photovoltaic power plant based on the spatiotemporal correlation model and the discrete empirical cumulative distribution function (ECDF), and integrates these scenarios containing spatiotemporal correlation knowledge After receiving these linear trend scenarios, each photovoltaic power station will input the Gated Recurrent Unit (GRU) together with the local historical data to train the prediction model. After the prediction model training is completed, the The historical data to be predicted and the linear trend scenario are input into the forecasting model to obtain the output forecasting results of each photovoltaic power station...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a distributed photovoltaic power station day-ahead output prediction method based on a spatio-temporal correlation model, and the method comprises the steps: obtaining the historical output data of all photovoltaic power stations at a server side, and constructing the spatio-temporal correlation model according to a linear trend segment and a linear trend segment mode; according to the space-time correlation model, generating a linear trend segment mode sequence scene corresponding to the photovoltaic power station by using a discrete experience cumulative distribution function; and sending the linear trend segment mode sequence scene to the target photovoltaic power station. The photovoltaic power station comprises the following steps: acquiring a linear trend segment mode sequence scene and an initial mode, matching local historical data by using the initial mode, and reconstructing linear trend time sequence data from the linear trend segment mode sequence scene by using an ARIMA model; training a prediction model by using the linear trend time sequence data and corresponding historical data; and calculating an output prediction result by using the trained prediction model. According to the method, the dependency on a mathematical model is canceled, and meanwhile, the prediction efficiency and privacy are ensured.
Description
technical field [0001] The invention relates to the field of photovoltaic power generation, in particular to a method for predicting the day-ahead output of a distributed photovoltaic power station based on a time-space correlation model. Background technique [0002] Distributed photovoltaic power plants are usually connected to medium and low voltage power distribution systems. The popularity of distributed photovoltaic power plants has turned the distribution network into an active system. With this kind of environmentally friendly renewable energy, the flexibility and reliability of the power distribution system can be greatly improved. Therefore, the installed capacity of distributed photovoltaics has increased significantly in recent years. Compared with large-scale centralized access methods, distributed photovoltaic power plants mainly have the following characteristics: 1) There are many access points, and the single-point access capacity is small, mostly below 50...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06F30/27G06F17/16G06Q50/06G06F119/02
CPCG06F30/27G06F17/16G06Q50/06G06F2119/02
Inventor 屈尹鹏黄晟张冀沈非凡颜畅
Owner HUNAN UNIV