Photovoltaic power generation electric energy regulation and control method and system for building

Abstract

The invention discloses a photovoltaic power generation electric energy regulation and control method and system for a building. The method comprises the steps that S1, meteorological data, state data of energy storage equipment and a distributed photovoltaic dynamic model are acquired; s2, obtaining a power grid optimization scheduling model according to the meteorological data and the state data of the energy storage equipment; s3, solving the power grid optimization scheduling model to obtain a distributed photovoltaic optimal active power output reference value; and S4, realizing tracking control of distributed photovoltaic active power output in the distributed photovoltaic dynamic model based on the distributed photovoltaic optimal active power output reference value. By adopting the technical scheme of the invention, more effective energy storage scale and overall power grid asset planning are realized.

Description

Technical Field

[0001] This invention belongs to the field of energy management technology, and in particular relates to a method and system for regulating photovoltaic power generation in buildings. Background Technology

[0002] With the continuous development of photovoltaic power generation technology, photovoltaic power generation systems are playing an increasingly important role in the power system. In order to better integrate photovoltaic power generation systems, it is necessary to rationally allocate energy storage capacity to cope with unpredictable weather changes and the intermittent nature of photovoltaic power generation.

[0003] The combination of photovoltaic (PV) power generation and energy storage systems allows PV systems to be transformed from purely power-based, instantaneous power generation systems into energy-based, dispatchable power generation resources. Therefore, how to achieve more efficient energy storage scale and overall grid asset planning has become a pressing issue. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method and system for regulating photovoltaic power generation in buildings, so as to achieve more effective energy storage scale and overall grid asset planning.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A method for regulating the power output of photovoltaic power generation in buildings, comprising:

[0007] Step S1: Acquire meteorological data, energy storage device status data, and distributed photovoltaic dynamic model;

[0008] Step S2: Based on meteorological data and energy storage device status data, obtain the power grid optimization scheduling model;

[0009] Step S3: Solve the power grid optimization scheduling model to obtain the optimal active power output reference value of distributed photovoltaic power.

[0010] Step S4: Based on the optimal active power output reference value of distributed photovoltaic, realize the tracking control of the active power output of distributed photovoltaic in the dynamic model of distributed photovoltaic.

[0011] Preferably, the status data of the energy storage device includes: real-time photovoltaic power generation, historical photovoltaic power generation, real-time load data, historical load curves, and device status data; meteorological data includes: temperature, humidity, aerosol concentration, cloud cover, light intensity, wind speed, and weather.

[0012] Preferably, in step S3, a Chebyshev graph convolutional neural network is used to solve the power grid optimization scheduling model in real time to obtain the optimal active power output reference value of distributed photovoltaic power.

[0013] As a preferred option, in step S2, the meteorological data and the status data of the energy storage device are cleaned, and the ConvTrans neural network is trained based on the cleaned meteorological data and the status data of the energy storage device to obtain the power grid optimization scheduling model.

[0014] The present invention also provides a photovoltaic power generation control system for buildings, comprising:

[0015] Acquisition device, used to acquire meteorological data, status data of energy storage equipment, and dynamic models of distributed photovoltaic systems;

[0016] The processing device is used to obtain a power grid optimization scheduling model based on meteorological data and energy storage device status data;

[0017] The computing device is used to solve the power grid optimization scheduling model and obtain the optimal active power output reference value of distributed photovoltaic power.

[0018] The control device is used to track and control the active power output of distributed photovoltaics in the dynamic model of distributed photovoltaics based on the optimal active power output reference value of distributed photovoltaics.

[0019] Preferably, the status data of the energy storage device includes: real-time photovoltaic power generation, historical photovoltaic power generation, real-time load data, historical load curves, and device status data; meteorological data includes: temperature, humidity, aerosol concentration, cloud cover, light intensity, wind speed, and weather.

[0020] As a preferred option, the computing device uses a Chebyshev graph convolutional neural network to solve the power grid optimization scheduling model in real time, and obtains the optimal active power output reference value of distributed photovoltaic power.

[0021] As a preferred embodiment, the processing device cleans the meteorological data and the status data of the energy storage equipment, and trains the ConvTrans neural network based on the cleaned meteorological data and the status data of the energy storage equipment to obtain the power grid optimization scheduling model.

[0022] This invention acquires meteorological data, energy storage device status data, and a distributed photovoltaic (PV) dynamic model; based on the meteorological data and energy storage device status data, it obtains a power grid optimization scheduling model; it solves the power grid optimization scheduling model to obtain the optimal active power output reference value for distributed PV; and based on the optimal active power output reference value for distributed PV, it achieves tracking and control of the active power output of distributed PV in the distributed PV dynamic model. By adopting the technical solution of this invention, more effective energy storage scale and overall power grid asset planning can be achieved. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0024] Figure 1 This is a schematic flowchart of a photovoltaic power generation regulation method for buildings according to an embodiment of the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0027] Example 1:

[0028] like Figure 1 As shown, this embodiment of the invention provides a method for regulating the power output of photovoltaic power generation in buildings, comprising:

[0029] Step S1: Acquire meteorological data, energy storage device status data, and distributed photovoltaic dynamic model;

[0030] Step S2: Based on meteorological data and energy storage device status data, obtain the power grid optimization scheduling model;

[0031] Step S3: Solve the power grid optimization scheduling model to obtain the optimal active power output reference value of distributed photovoltaic power.

[0032] Step S4: Based on the optimal active power output reference value of distributed photovoltaic, realize the tracking control of the active power output of distributed photovoltaic in the dynamic model of distributed photovoltaic.

[0033] As one embodiment of the present invention, the status data of the energy storage device includes: real-time photovoltaic power generation, historical photovoltaic power generation, real-time load data, historical load curves, and device status data; meteorological data includes: temperature, humidity, aerosol concentration, cloud cover, irradiance, wind speed, and weather conditions. Based on the topological connection relationship of the distributed photovoltaic system, a dynamic model of the distributed photovoltaic system, taking into account communication latency and load disturbances, is established.

[0034] In one embodiment of the present invention, in step S2, meteorological data and energy storage device status data are cleaned, and a ConvTrans neural network is trained based on the cleaned meteorological data and energy storage device status data to obtain a power grid optimization scheduling model.

[0035] Furthermore, data cleaning is performed on meteorological data and energy storage device status data, specifically as follows:

[0036] By filtering out garbled and out-of-place data from meteorological data and energy storage device status data, the filtering index data is obtained;

[0037] The filtering index data is vectorized to obtain a filtering index vector set. A spatiotemporal vector is added to each filtering index vector in the filtering index vector set to obtain a spatiotemporal vector set.

[0038] The spatiotemporal vector set is subjected to feature clustering to obtain a cluster center dataset;

[0039] The null values ​​in the filtering index data are filled in based on the cluster center dataset.

[0040] Furthermore, the process of training the ConvTrans neural network using meteorological data and energy storage device status data is as follows: Training process: The meteorological data and energy storage device status data are divided into a model training set and a model test set in an 8:2 ratio. Then, the ConvTrans neural network is trained using the training samples. The relevant parameters of the ConvTrans neural network are adjusted, and the quality of the trained ConvTrans neural network is evaluated according to the model evaluation index until the power grid optimization dispatch model is obtained.

[0041] In further embodiments of the present invention, the coefficient of determination R² and mean squared error (MSE) can be used to evaluate the training and testing results of the model to obtain the optimal power grid scheduling model.

[0042] In one embodiment of the present invention, step S3 involves using a Chebyshev graphical convolutional neural network to solve the power grid optimization scheduling model in real time, thereby obtaining the optimal active power output reference value for distributed photovoltaic power. Specifically, the Chebyshev graphical convolutional neural network is used to mine the spatial characteristics of the optimal active power flow of distributed photovoltaic power, establish a mapping relationship between the operating characteristics of the distribution network and the optimal active power output of distributed photovoltaic power, and realize the real-time solution of the power grid optimization scheduling model.

[0043] Example 2:

[0044] This invention also provides a photovoltaic power generation control system for buildings, comprising:

[0045] Acquisition device, used to acquire meteorological data, status data of energy storage equipment, and dynamic models of distributed photovoltaic systems;

[0046] The processing device is used to obtain a power grid optimization scheduling model based on meteorological data and energy storage device status data;

[0047] The computing device is used to solve the power grid optimization scheduling model and obtain the optimal active power output reference value of distributed photovoltaic power.

[0048] The control device is used to track and control the active power output of distributed photovoltaics in the dynamic model of distributed photovoltaics based on the optimal active power output reference value of distributed photovoltaics.

[0049] As one embodiment of the present invention, the status data of the energy storage device includes: real-time photovoltaic power generation, historical photovoltaic power generation, real-time load data, historical load curves, and device status data; meteorological data includes: temperature, humidity, aerosol concentration, cloud cover, irradiance, wind speed, and weather conditions. Based on the topological connection relationship of the distributed photovoltaic system, a dynamic model of the distributed photovoltaic system, taking into account communication latency and load disturbances, is established.

[0050] As one embodiment of the present invention, the processing device performs data cleaning on meteorological data and energy storage device status data, and trains a ConvTrans neural network based on the cleaned meteorological data and energy storage device status data to obtain a power grid optimization scheduling model.

[0051] Furthermore, data cleaning is performed on meteorological data and energy storage device status data, specifically as follows:

[0052] By filtering out garbled and out-of-place data from meteorological data and energy storage device status data, the filtering index data is obtained;

[0053] The filtering index data is vectorized to obtain a filtering index vector set. A spatiotemporal vector is added to each filtering index vector in the filtering index vector set to obtain a spatiotemporal vector set.

[0054] The spatiotemporal vector set is subjected to feature clustering to obtain a cluster center dataset;

[0055] The null values ​​in the filtering index data are filled in based on the cluster center dataset.

[0056] Furthermore, the process of training the ConvTrans neural network using meteorological data and energy storage device status data is as follows: Training process: The meteorological data and energy storage device status data are divided into a model training set and a model test set in an 8:2 ratio. Then, the ConvTrans neural network is trained using the training samples. The relevant parameters of the ConvTrans neural network are adjusted, and the quality of the trained ConvTrans neural network is evaluated according to the model evaluation index until the power grid optimization dispatch model is obtained.

[0057] In further embodiments of the present invention, the coefficient of determination R² and mean squared error (MSE) can be used to evaluate the training and testing results of the model to obtain the optimal power grid scheduling model.

[0058] In one embodiment of the present invention, the computing device employs a Chebyshev graphical convolutional neural network to solve the power grid optimization scheduling model in real time, obtaining the optimal active power output reference value for distributed photovoltaic power. Specifically, the Chebyshev graphical convolutional neural network is used to mine the spatial characteristics of the optimal active power flow of distributed photovoltaic power, establish a mapping relationship between the operating characteristics of the distribution network and the optimal active power output of distributed photovoltaic power, and realize the real-time solution of the power grid optimization scheduling model.

[0059] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for regulating photovoltaic electricity generation for buildings, characterized in that, The method comprises the following steps: Step S1, obtaining meteorological data, state data of energy storage equipment and a distributed photovoltaic dynamic model; Step S2, obtaining an optimal scheduling model of a power grid according to the meteorological data and the state data of the energy storage equipment; Step S3, solving the optimal scheduling model of the power grid to obtain an optimal active power output reference value of the distributed photovoltaic; Step S4, realizing tracking control of active power output of the distributed photovoltaic in the distributed photovoltaic dynamic model based on the optimal active power output reference value of the distributed photovoltaic.

2. The method for regulating the electrical energy produced by photovoltaic power for buildings according to claim 1, characterized in that, The state data of the energy storage equipment comprises real-time photovoltaic power generation, historical photovoltaic power generation, real-time load data, historical load curve and equipment state data; and the meteorological data comprises temperature, humidity, aerosol concentration, cloud cover, light intensity, wind speed and weather.

3. The method for regulating the electrical energy produced by photovoltaic power for buildings according to claim 2, characterized in that, In step S3, a Chebyshev graph convolutional neural network is used to solve the optimal scheduling model of the power grid in real time to obtain the optimal active power output reference value of the distributed photovoltaic.

4. The method for regulating the electrical energy produced by photovoltaic power for buildings according to claim 3, characterized in that, In step S2, data cleaning is performed on the meteorological data and the state data of the energy storage equipment, and a ConvTrans neural network is trained according to the cleaned meteorological data and the state data of the energy storage equipment to obtain the optimal scheduling model of the power grid.

5. A photovoltaic power generation electrical energy regulating system for buildings, characterized in that, The method comprises the following steps: An obtaining device is configured to obtain meteorological data, state data of energy storage equipment and a distributed photovoltaic dynamic model; A processing device is configured to obtain an optimal scheduling model of a power grid according to the meteorological data and the state data of the energy storage equipment; A calculation device is configured to solve the optimal scheduling model of the power grid to obtain an optimal active power output reference value of the distributed photovoltaic; A control device is configured to realize tracking control of active power output of the distributed photovoltaic in the distributed photovoltaic dynamic model based on the optimal active power output reference value of the distributed photovoltaic.

6. The photovoltaic power regulating system for buildings according to claim 5, characterized in that, The state data of the energy storage equipment comprises real-time photovoltaic power generation, historical photovoltaic power generation, real-time load data, historical load curve and equipment state data; and the meteorological data comprises temperature, humidity, aerosol concentration, cloud cover, light intensity, wind speed and weather.

7. The photovoltaic power regulating system for buildings according to claim 6, characterized in that, The calculation device uses a Chebyshev graph convolutional neural network to solve the optimal scheduling model of the power grid in real time to obtain the optimal active power output reference value of the distributed photovoltaic.

8. The photovoltaic power regulating system for buildings according to claim 7, characterized in that, The processing device performs data cleaning on the meteorological data and the state data of the energy storage equipment, and trains a ConvTrans neural network according to the cleaned meteorological data and the state data of the energy storage equipment to obtain the optimal scheduling model of the power grid.

Images