CN115906408A - Energy scheduling system and method based on building load prediction - Google Patents
Energy scheduling system and method based on building load prediction Download PDFInfo
- Publication number
- CN115906408A CN115906408A CN202211289125.1A CN202211289125A CN115906408A CN 115906408 A CN115906408 A CN 115906408A CN 202211289125 A CN202211289125 A CN 202211289125A CN 115906408 A CN115906408 A CN 115906408A
- Authority
- CN
- China
- Prior art keywords
- energy
- data
- load
- historical
- 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 belongs to the technical field of building energy conservation, and discloses an energy scheduling system and method based on building load prediction. The method comprises the following steps: according to the historical load data and the electricity price time partition, a load prediction model is established; establishing an energy generation prediction model according to historical energy generation data and historical environment states; inputting the current electricity price time partition into a load prediction model to obtain predicted load data; inputting the current environment state into an energy generation prediction model to obtain predicted energy generation data; according to the predicted load data, the predicted energy generation data and the current electricity price time partition, determining the corresponding relation between the load and the energy supply; obtaining an energy scheduling strategy according to the corresponding relation between the load and the energy supply; and scheduling the energy supply of the target building according to the energy scheduling strategy. Through the mode, the intelligent scheduling is carried out on the supply of different types of energy, the application value of renewable energy is improved, and green and energy-saving effects are achieved.
Description
Technical Field
The invention relates to the technical field of building energy conservation, in particular to an energy scheduling system and method based on building load prediction.
Background
Under the double pressure of development and environment, the application proportion of renewable energy sources such as solar energy, wind energy and the like in building energy supply is gradually increased. The solar energy resource is widely distributed, is available on the spot and is the most abundant clean energy, but is influenced by weather changes such as overcast and rainy days, and has the characteristics of instability and discontinuity, and wind energy is sensitive to weather and climate, and has the problems of instability and discontinuity, so that the problems limit the use of buildings for renewable energy sources, influence the realization of 'green energy conservation', and how to formulate a multi-energy scheduling strategy becomes a real problem which needs to be solved urgently.
The above is only for the purpose of assisting understanding of the technical aspects of the present invention, and does not represent an admission that the above is prior art.
Disclosure of Invention
The invention mainly aims to provide an energy scheduling system and method based on building load prediction, and aims to solve the technical problems that in the prior art, due to instability and discontinuity of renewable energy, the use of renewable energy by a building is limited, and green and energy conservation is influenced.
In order to achieve the above object, the present invention provides an energy scheduling method based on building load prediction, comprising the steps of:
acquiring historical load data of a target building, and establishing a load prediction model according to the historical load data and power price time partitions;
acquiring historical energy generation data and historical environment states of a target building, and establishing an energy generation prediction model according to the historical energy generation data and the historical environment states;
inputting the current electricity price time into the load prediction model in a partition mode to obtain predicted load data;
inputting the current environment state into the energy generation prediction model to obtain predicted energy generation data;
according to the predicted load data, the predicted energy generation data and the current electricity price time partition, determining the corresponding relation between the load and the energy supply;
obtaining an energy scheduling strategy according to the corresponding relation between the load and the energy supply;
and scheduling the energy supply of the target building according to the energy scheduling strategy.
Optionally, the determining, according to the predicted load data, the predicted energy generation data, and the current power rate time partition, a corresponding relationship between the load and the energy supply includes:
determining current electricity price information according to the current electricity price time partition;
determining an initial corresponding relation between the load and the energy supply according to the predicted load data and the predicted energy generation data;
determining energy cost data according to the current electricity price information;
determining a loss function according to the energy cost data;
and optimizing the initial corresponding relation between the load and the energy supply according to the loss function and a preset multi-objective optimization algorithm to obtain the corresponding relation between the load and the energy supply.
Optionally, the obtaining historical load data of the target building, and establishing a load prediction model according to the historical load data and the electricity price time partition includes:
determining the power price time partition according to the power price information;
acquiring historical load data of a target building;
determining an initial load prediction model according to a preset neural network model;
and according to the historical load data and the electricity price time partition, training the initial load prediction model to obtain a load prediction model.
Optionally, the partitioning according to the historical load data and the electricity price time, training the initial load prediction model to obtain a load prediction model, including:
inputting the electricity price time partition into the initial load prediction model to obtain load output data;
determining a first difference function according to the load output data and historical load data;
and when the first difference function meets a preset training target, determining a load prediction model according to the initial load prediction model.
Optionally, after determining the first difference function according to the load output data and the historical load data, the method further includes:
when the first difference function does not meet a preset training target, adjusting the weight of the initial load prediction model according to the first difference function to obtain a new initial load prediction model;
and returning to execute the step of inputting the power price time partitions into the initial load prediction model according to the new initial load prediction model to obtain load output data.
Optionally, the obtaining historical energy generation data and historical environmental state of the target building, and establishing an energy generation prediction model according to the historical energy generation data and the historical environmental state includes:
determining an initial energy generation prediction model according to a preset neural network model;
acquiring historical energy generation data and historical environmental states of a target building, wherein the historical energy generation data comprises historical light energy generation data and historical wind energy generation data;
and training the initial energy generation prediction model according to the historical energy generation data and the historical environment state to obtain an energy generation prediction model.
Optionally, the training an initial energy generation prediction model according to the historical energy generation data and the historical environmental state to obtain an energy generation prediction model includes:
inputting the historical environment state into the initial energy generation prediction model to obtain energy generation output data;
determining a second difference function according to the energy generation output data and historical energy generation data;
and when the second difference function meets a preset training target, determining an energy generation prediction model according to the initial energy generation prediction model.
Optionally, after determining the second difference function according to the energy generation output data and the historical energy generation data, the method further includes:
when the second difference function does not meet a preset training target, adjusting the weight of the initial energy generation prediction model according to the second difference function to obtain a new initial energy generation prediction model;
and according to the new initial energy generation prediction model, returning to execute the step of inputting the historical environment state into the initial energy generation prediction model to obtain energy generation output data.
Optionally, the obtaining an energy scheduling policy according to the correspondence between the load and the energy supply includes:
determining energy supply data according to the corresponding relation between the load and energy supply, wherein the energy supply data comprises power supply data, light energy supply data and wind energy supply data;
determining an optimal weight coefficient according to the energy supply data;
determining energy demand data according to the optimal weight coefficient;
and determining an energy scheduling strategy according to the energy demand data.
In addition, in order to achieve the above object, the present invention further provides an energy scheduling system based on building load prediction, including:
the model building module is used for obtaining historical load data of a target building and building a load prediction model according to the historical load data and the electricity price time partition;
the model establishing module is also used for acquiring historical energy generation data and historical environmental states of a target building and establishing an energy generation prediction model according to the historical energy generation data and the historical environmental states;
the data prediction module is used for inputting the current electricity price time partition into the load prediction model to obtain predicted load data;
the data prediction module is also used for inputting the current environment state into the energy generation prediction model to obtain predicted energy generation data;
the energy scheduling module is used for determining the corresponding relation between the load and the energy supply according to the predicted load data, the predicted energy generation data and the current electricity price time partition;
the energy scheduling module is further used for obtaining an energy scheduling strategy according to the corresponding relation between the load and the energy supply;
and the energy scheduling module is also used for scheduling the energy supply of the target building according to the energy scheduling strategy.
In addition, to achieve the above object, the present invention further provides an energy scheduling apparatus based on building load prediction, including: a memory, a processor, and an energy scheduler based on building load prediction stored on the memory and operable on the processor, the energy scheduler based on building load prediction being configured to implement the steps of the energy scheduling method based on building load prediction as described above.
In addition, to achieve the above object, the present invention further provides a storage medium, wherein the storage medium stores an energy scheduling program based on building load prediction, and the energy scheduling program based on building load prediction realizes the steps of the energy scheduling method based on building load prediction as described above when being executed by a processor.
The method comprises the steps of obtaining historical load data of a target building, establishing a load prediction model according to the historical load data and power price time partitions, obtaining historical energy generation data and historical environment states of the target building, establishing an energy generation prediction model according to the historical energy generation data and the historical environment states, inputting current power price time partitions into the load prediction model to obtain predicted load data, inputting the current environment states into the energy generation prediction model to obtain predicted energy generation data, determining the corresponding relation between load and energy supply according to the predicted load data, the predicted energy generation data and the current power price time partitions, obtaining an energy scheduling strategy according to the corresponding relation between the load and the energy supply, and scheduling the energy supply of the target building according to the energy scheduling strategy. The renewable energy is easily affected by weather, has instability and discontinuity, and limits the application of the renewable energy.
Drawings
Fig. 1 is a schematic structural diagram of an energy scheduling device based on building load prediction for a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a schematic flowchart illustrating a first embodiment of a method for energy scheduling based on building load prediction according to the present invention;
FIG. 3 is a flowchart illustrating a second embodiment of the energy scheduling method based on building load prediction according to the present invention;
fig. 4 is a block diagram of a first embodiment of the energy scheduling system based on building load prediction according to the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of an energy scheduling device based on building load prediction for a hardware operating environment according to an embodiment of the present invention.
As shown in fig. 1, the energy scheduling apparatus based on building load prediction may include: a processor 1001, such as a Central Processing Unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a Wireless interface (e.g., a Wireless-Fidelity (Wi-Fi) interface). The Memory 1005 may be a Random Access Memory (RAM) Memory, or a Non-Volatile Memory (NVM), such as a disk Memory. The memory 1005 may alternatively be a storage device separate from the processor 1001.
Those skilled in the art will appreciate that the configuration shown in fig. 1 does not constitute a limitation of energy scheduling devices based on building load prediction, and may include more or fewer components than those shown, or some components in combination, or a different arrangement of components.
As shown in fig. 1, the memory 1005, which is a storage medium, may include therein an operating system, a network communication module, a user interface module, and an energy scheduler based on building load prediction.
In the energy scheduling apparatus based on building load prediction shown in fig. 1, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 of the energy scheduling apparatus based on building load prediction according to the present invention may be provided in the energy scheduling apparatus based on building load prediction, which calls the energy scheduling program based on building load prediction stored in the memory 1005 through the processor 1001 and performs the energy scheduling method based on building load prediction according to the embodiment of the present invention.
An embodiment of the present invention provides an energy scheduling method based on building load prediction, and referring to fig. 2, fig. 2 is a schematic flow diagram of a first embodiment of an energy scheduling method based on building load prediction according to the present invention.
In this embodiment, the energy scheduling method based on building load prediction includes the following steps:
step S10: obtaining historical load data of a target building, and establishing a load prediction model according to the historical load data and the electricity price time partition.
The implementation subject of the present embodiment is a computer, and may be any computer that can run an energy scheduling program based on the building load prediction, and schedules the supply of different energy sources by the energy scheduling program based on the building load prediction.
Further, the step S10 includes: determining a power price time partition according to power price information, acquiring historical load data of a target building, determining an initial load prediction model according to a preset neural network model, and training the initial load prediction model according to the historical load data and the power price time partition to obtain a load prediction model.
It can be understood that the target building may be any building that needs energy scheduling, and may be selected according to actual needs, which is not limited in this embodiment, and the historical load data is a historical record of the load of the target building. The electricity rate information is electricity rate peak-valley information, and the electricity rate time partition is a time interval divided according to the electricity rate peak-valley, such as: if the peak time is 7:00 to 11: 00. 19:00 to 23:00, usual period 11:00 to 19:00, valley period 23: 00-day 7:00, the electricity price time partition is 0:00 to 7: 00. 7:00 to 11: 00. 11:00 to 19: 00. 19:00 to 23: 00. 23: 00-0: 00. the preset neural network model in this embodiment is a BP (Back Propagation) neural network model, and other types of neural network models may also be used. The initial load prediction model is a preliminarily constructed load prediction model, and the load prediction model is a model for predicting the building load.
In specific implementation, a model training sample is constructed according to historical load data of a target building and electricity price time partitions, and an initial load prediction model is trained correspondingly, so that a load prediction model is established, and prediction of building load is realized.
It should be understood that in the process of training the initial load prediction model according to the historical load data and the power rate time partition to obtain the load prediction model, the power rate time partition is input into the initial load prediction model to obtain load output data, a first difference function is determined according to the load output data and the historical load data, and when the first difference function meets a preset training target, the load prediction model is determined according to the initial load prediction model. And when the first difference function does not meet a preset training target, adjusting the weight of the initial load prediction model according to the first difference function to obtain a new initial load prediction model, and returning to execute the step of inputting the electricity price time into the initial load prediction model in a partitioning manner according to the new initial load prediction model to obtain load output data.
It should be noted that the load output data is an output of an initial load prediction model, the first difference function is a function formed by a difference between the load output data and historical load data, and the preset training target is a function in which a difference between the output and a target is minimized.
It can be understood that the historical load data includes load data of a plurality of different dates, and each load data of different dates can be divided according to the power rate time partitions, so as to obtain load data corresponding to each power rate time partition on different dates.
In the specific implementation, the electricity price time partition is used as an input sample and input into an initial load prediction model constructed by a BP neural network model to obtain output data, historical load data is used as a target sample, a first error function is calculated, the weight of the initial load prediction model is adjusted according to the first error function, the historical load data and the electricity price time partition are continuously used for training the initial load prediction model until the first error function reaches the minimum, and the trained load prediction model is obtained.
Step S20: historical energy generation data and historical environmental states of a target building are obtained, and an energy generation prediction model is established according to the historical energy generation data and the historical environmental states.
It should be understood that the historical energy generation data is a historical record of energy generation amount, and includes historical light energy generation data and historical wind energy generation data, that is, a historical record of light energy generation amount and a historical record of wind energy generation amount in a unit time, and there may also be generation data of other renewable energy sources, which is not limited in this embodiment and can be flexibly selected according to actual situations, and the unit time used in this embodiment is 1 hour. The historical environmental state is the historical condition of the environment outside the building, including the historical meteorological state and the historical temperature state, and the historical meteorological state is the historical condition of the weather, for example: in the case of sunny, cloudy, rainy, snowy, typhoon, etc., the historical temperature status is the historical condition of temperature, and the grade can be used to represent, for example: the temperature is not more than 0 ℃ and is a first-stage temperature, the temperature is more than 0 ℃ and less than or equal to 10 ℃ and is a second-stage temperature, the temperature is more than 10 ℃ and less than or equal to 20 ℃ and is a third-stage temperature, the temperature is more than 20 ℃ and less than or equal to 30 ℃ and is a fourth-stage temperature, the temperature is more than 30 ℃ and less than or equal to 40 ℃ and is a fifth-stage temperature, and the temperature is more than or equal to 40 ℃ and is a sixth-stage temperature. The energy generation prediction model is a model for predicting the generation amount of each energy, and the embodiment mainly predicts the generation amount of light energy and the generation amount of wind energy.
Further, the step S20 includes: determining an initial energy generation prediction model according to a preset neural network model, acquiring historical energy generation data and historical environment states of a target building, and training the initial energy generation prediction model according to the historical energy generation data and the historical environment states to obtain an energy generation prediction model.
It should be noted that the initial energy generation prediction model is a preliminarily constructed energy generation prediction model, and is constructed by using a BP neural network model in this embodiment.
In the specific implementation, a model training sample is constructed according to historical light energy generation data, historical wind energy generation data, historical meteorological states and historical temperature states of a target building, so that an energy generation prediction model is established, and light energy generation and wind energy generation are predicted.
It can be understood that, in the process of training an initial energy generation prediction model according to the historical energy generation data and the historical environment state to obtain an energy generation prediction model, the historical environment state is input into the initial energy generation prediction model to obtain energy generation output data, a second difference function is determined according to the energy generation output data and the historical energy generation data, and when the second difference function meets a preset training target, the energy generation prediction model is determined according to the initial energy generation prediction model. And when the second difference function does not meet a preset training target, adjusting the weight of the initial energy generation prediction model according to the second difference function to obtain a new initial energy generation prediction model, generating the prediction model according to the new initial energy, and returning to the step of inputting the historical environment state into the initial energy generation prediction model to obtain energy generation output data.
It should be understood that the energy generation output data is an output of the initial energy generation prediction model, and the second difference function is a function of a difference between the energy generation output data and the historical energy generation data.
In the specific implementation, the historical environment state is used as an input sample and input into an initial energy generation prediction model constructed by a BP neural network model to obtain output data, the historical energy generation data is used as a target sample, a second error function is calculated, the weight of the initial energy generation prediction model is adjusted according to the second error function, the historical energy generation data and the historical environment state are continuously used for training the initial energy generation prediction model until the second error function is minimum, and the energy generation prediction model after training is obtained.
Step S30: and inputting the current electricity price time partition into the load prediction model to obtain predicted load data.
It should be noted that the current electricity price time partition is an electricity price time partition in which the current time is located, the load prediction model inputs the electricity price time partition, and outputs predicted load data, and the predicted load data is a building load prediction result obtained by using the load prediction model currently.
In specific implementation, the electricity price time corresponding to the current time is input into the established load prediction model in a partition mode, and a current building load prediction result can be output.
Step S40: and inputting the current environment state into the energy generation prediction model to obtain predicted energy generation data.
It is understood that the current environmental status is a current environment condition outside the building, including a current weather status and a current temperature status, i.e. a current weather condition and a current temperature, for example: and acquiring related data according to actual conditions under the sunny and third-level temperatures. The energy generation prediction model is input as an environmental state, and output as predicted energy generation data, where the predicted energy generation data is a prediction result of the building energy generation amount obtained by using the energy generation prediction model currently, and the predicted energy generation data in this embodiment includes predicted light energy generation data and predicted wind energy generation data, that is, predicted light energy generation amount and wind energy generation amount per unit time.
In specific implementation, the current meteorological state and the current temperature state are input into the established energy generation prediction model, so that the current light energy generation amount and wind energy generation amount can be predicted.
Step S50: and determining the corresponding relation between the load and the energy supply according to the predicted load data, the predicted energy generation data and the current electricity price time partition.
It should be understood that the energy supply includes an electric power supply, which refers to a supply of electric power by an electric power department, a light energy supply, which refers to a supply of converting light energy into electric energy, and a wind energy supply, which refers to a supply of converting wind energy into electric energy. The corresponding relation between the load and the energy supply is a calculated relation between the current load and the energy supply when the cost is the lowest.
In a specific implementation, the light energy supply amount and the wind energy supply amount within the current electricity price time are calculated according to the predicted light energy generation amount and the predicted wind energy generation amount within the unit time, and the calculation relation between the load and the energy supply at the time of lowest cost is calculated according to the light energy supply amount, the wind energy supply amount and the power supply amount within the current electricity price time, wherein the weight coefficient of each energy supply is the optimal weight coefficient.
Step S60: and obtaining an energy scheduling strategy according to the corresponding relation between the load and the energy supply.
It should be noted that the energy scheduling policy is a scheme for scheduling various energy sources.
In the concrete implementation, according to the calculation relationship between the current load and the energy supply, the weight coefficient of each energy supply is determined, each energy supply quantity which enables the cost to reach the lowest is calculated, the demand quantity of each energy is obtained, and then a corresponding energy scheduling strategy is formulated.
Step S70: and scheduling the energy supply of the target building according to the energy scheduling strategy.
In the embodiment, historical load data of a target building is obtained, a load prediction model is established according to the historical load data and the power rate time partition, historical energy generation data and a historical environment state of the target building are obtained, an energy generation prediction model is established according to the historical energy generation data and the historical environment state, the current power rate time partition is input into the load prediction model to obtain predicted load data, the current environment state is input into the energy generation prediction model to obtain predicted energy generation data, the corresponding relation between load and energy supply is determined according to the predicted load data, the predicted energy generation data and the current power rate time partition, an energy scheduling strategy is obtained according to the corresponding relation between load and energy supply, and energy supply of the target building is scheduled according to the energy scheduling strategy. The renewable energy is easily affected by weather, has instability and discontinuity, and limits the application of the renewable energy, and the embodiment calculates the supply quantity of each energy source with the lowest cost based on the building load prediction and in combination with the electricity price information, and intelligently schedules the supply of different types of energy sources, so that the application value of the renewable energy is improved, and green and energy saving is realized.
Referring to fig. 3, fig. 3 is a flowchart illustrating a second embodiment of the energy scheduling method based on building load prediction according to the present invention.
Based on the first embodiment, the step S50 includes:
step S501: and determining the current electricity price information according to the current electricity price time partition.
It should be noted that the current electricity price information is the electricity price corresponding to the current electricity price time partition, for example: the current electricity price time partition is 7:00 to 11:00, at peak time, corresponding to a price of 0.568 yuan/kwh.
Step S502: and determining an initial corresponding relation between the load and the energy supply according to the predicted load data and the predicted energy generation data.
It will be appreciated that the initial correspondence between load and energy supply is a calculated relationship between load and energy supply, as follows:
P＝A·(αL t )+B·(βW t )+C·(E t )
where P is the building load during the current electricity price time, E t Is the amount of power supply in the current electricity rate time, alL t For the light energy supply in the current electricity price time, beta W t Alpha is the light energy conversion coefficient, beta is the wind energy conversion coefficient and L is the wind energy supply amount in the current electricity price time t Amount of light energy production, W t For the wind energy generation, A, B, C are weight coefficients, respectively.
In a specific implementation, the light energy supply quantity and the wind energy supply quantity in the current electricity price time are calculated according to the predicted energy generation data, and then the calculation relational expression is substituted according to the power supply quantity of the power department in the current electricity price time.
Step S503: and determining energy cost data according to the current electricity price information.
It will be appreciated that the energy cost data is the cost of obtaining these energy supplies to the building, primarily the cost of the electricity supply.
In the specific implementation, the cost of the energy supply of the building is calculated according to the current electricity price information and the renewable energy generation cost.
Step S504: and determining a loss function according to the energy cost data, and optimizing the initial corresponding relation between the load and the energy supply according to the loss function and a preset multi-objective optimization algorithm to obtain the corresponding relation between the load and the energy supply.
It should be noted that the preset multi-objective optimization Algorithm is a preset multi-objective optimization Algorithm, and the Non-dominant sequencing Genetic Algorithm (NSGA-II) is selected in this embodiment, or may be another multi-objective optimization Algorithm, which is not limited in this embodiment and can be flexibly adjusted according to actual situations.
In the specific implementation, the energy cost data is used as a loss function, and the NSGA-II algorithm is adopted to optimize the initial corresponding relation between the load and the energy supply, so that the corresponding relation between the load and the energy supply which enables the cost to be minimum is obtained.
Further, the step S60 includes:
step S601: and determining an optimal weight coefficient according to the corresponding relation between the load and the energy supply, and obtaining energy supply data according to the optimal weight coefficient.
It is understood that the optimal weight coefficient is the weight coefficient of each energy supply when the cost is the lowest, and the energy supply data is each energy supply amount when the cost is the lowest, and comprises power supply data, light energy supply data and wind energy supplyData, power supply data refers to the amount of supply of electrical energy transmitted by the power department, light energy supply data refers to the amount of supply that converts light energy into electrical energy, and wind energy supply data refers to the amount of supply that converts wind energy into electrical energy. The power supply data is calculated according to the optimal weight coefficient and the power supply amount in the current price time, the light energy source supply data is calculated according to the optimal weight coefficient and the light energy supply amount in the current price time, and the wind energy supply data is calculated according to the optimal weight coefficient and the wind energy supply amount in the current price time, for example: if the optimal weight coefficient of the light energy supply is A ', the optimal weight coefficient of the wind energy supply is B ', the optimal weight coefficient of the power supply is C ', and the power supply amount in the current price time is alpha L t The light energy supply amount in the current electricity price time is beta W t The wind energy supply amount in the current electricity price time is E t Then, the light energy supply data X = a' · (α L) t ) Wind energy supply data Y = B · (β W) t ) Power supply data Z = C · (E) t )。
Step S602: and determining energy demand data according to the energy supply data, and determining an energy scheduling strategy according to the energy demand data.
It should be understood that the energy demand data is demand of a building for each energy source, and includes power demand data, light energy demand data, and wind energy demand data, the power demand data is power demand, and is the same as power supply data, the light energy demand data is light energy demand, and needs to be calculated according to light energy supply data and a light energy conversion coefficient, and the wind energy demand data is wind energy demand, and needs to be calculated according to wind energy supply data and a wind energy conversion coefficient, for example: if the optical energy supply data is X, the wind energy supply data is Y, the power supply data is Z, the optical energy conversion coefficient is α, and the wind energy conversion coefficient is β, the optical energy demand data is X ' = X/α, the wind energy demand data is Y ' = Y/β, and the power demand data is Z ' =. The energy scheduling policy is a scheme for scheduling various energy sources, such as: if the light energy demand data is X ', the wind energy demand data is Y', the power demand data is Z ', the energy scheduling strategy is to supply Z' to the power department, the building light energy X 'is converted into electric energy, and the building wind energy Y' is converted into electric energy.
In specific implementation, according to the calculation relationship between the current load and the energy supply, the weight coefficient of each energy supply is determined, each energy supply amount which enables the cost to be the lowest is calculated, the demand of each energy is obtained, and then a corresponding energy scheduling strategy is formulated.
In this embodiment, current electricity price information is determined according to a current electricity price time partition, data is generated according to predicted load data and predicted energy, an initial corresponding relationship between load and energy supply is determined, energy cost data is determined as a loss function according to the current electricity price information, the initial corresponding relationship between the load and the energy supply is optimized according to the loss function and a preset multi-objective optimization algorithm, the corresponding relationship between the load and the energy supply is obtained, an optimal weight coefficient is determined, energy supply data is obtained according to the optimal weight coefficient, energy demand data is determined according to the energy supply data, and an energy scheduling strategy is determined according to the energy demand data. According to the energy source scheduling method and the energy source scheduling system, the corresponding relation between the load and the energy source supply with the lowest cost is found according to the energy source cost and the multi-objective optimization algorithm, so that the supply quantity of each energy source is determined, the current energy source scheduling strategy is further determined, the traditional energy source scheduling is optimized, the cost is reduced while the intelligent scheduling is carried out on the multiple energy source supplies, the supply and consumption of traditional electric power are reduced, and green and energy-saving effects are achieved.
In addition, an embodiment of the present invention further provides a storage medium, where the storage medium stores an energy scheduling program based on building load prediction, and the energy scheduling program based on building load prediction implements the steps of the energy scheduling method based on building load prediction as described above when being executed by a processor.
Referring to fig. 4, fig. 4 is a block diagram illustrating a first embodiment of an energy scheduling system based on building load prediction according to the present invention.
As shown in fig. 4, the energy scheduling system based on building load prediction according to the embodiment of the present invention includes:
and the model establishing module 10 is used for acquiring historical load data of the target building and establishing a load prediction model according to the historical load data and the electricity price time partition.
The model establishing module 10 is further configured to obtain historical energy generation data and historical environmental conditions of the target building, and establish an energy generation prediction model according to the historical energy generation data and the historical environmental conditions.
And the data prediction module 20 is used for inputting the current electricity price time partition into the load prediction model to obtain predicted load data.
The data prediction module 20 is further configured to input the current environmental state into the energy generation prediction model to obtain predicted energy generation data.
And the energy scheduling module 30 is configured to determine a corresponding relationship between the load and the energy supply according to the predicted load data, the predicted energy generation data, and the current electricity price time partition.
The energy scheduling module 30 is further configured to obtain an energy scheduling policy according to the corresponding relationship between the load and the energy supply.
The energy scheduling module 30 is further configured to schedule the energy supply of the target building according to the energy scheduling policy.
In the embodiment, historical load data of a target building is obtained, a load prediction model is established according to the historical load data and the power rate time partition, historical energy generation data and a historical environment state of the target building are obtained, an energy generation prediction model is established according to the historical energy generation data and the historical environment state, the current power rate time partition is input into the load prediction model to obtain predicted load data, the current environment state is input into the energy generation prediction model to obtain predicted energy generation data, the corresponding relation between load and energy supply is determined according to the predicted load data, the predicted energy generation data and the current power rate time partition, an energy scheduling strategy is obtained according to the corresponding relation between load and energy supply, and energy supply of the target building is scheduled according to the energy scheduling strategy. The renewable energy is easily affected by weather, has instability and discontinuity, and limits the application of the renewable energy, and the embodiment calculates the supply quantity of each energy source with the lowest cost based on the building load prediction and in combination with the electricity price information, and intelligently schedules the supply of different types of energy sources, so that the application value of the renewable energy is improved, and green and energy saving is realized.
In an embodiment, the energy scheduling module 30 is further configured to determine current electricity price information according to the current electricity price time partition;
determining an initial corresponding relation between the load and the energy supply according to the predicted load data and the predicted energy generation data;
determining energy cost data according to the current electricity price information;
determining a loss function according to the energy cost data;
and optimizing the initial corresponding relation between the load and the energy supply according to the loss function and a preset multi-objective optimization algorithm to obtain the corresponding relation between the load and the energy supply.
In an embodiment, the model building module 10 is further configured to determine a power rate time partition according to the power rate information;
acquiring historical load data of a target building;
determining an initial load prediction model according to a preset neural network model;
and partitioning according to the historical load data and the electricity price time, and training the initial load prediction model to obtain a load prediction model.
In an embodiment, the model establishing module 10 is further configured to input the electricity price time partition into the initial load prediction model to obtain load output data;
determining a first difference function according to the load output data and historical load data;
and when the first difference function meets a preset training target, determining a load prediction model according to the initial load prediction model.
In an embodiment, the model establishing module 10 is further configured to, when the first difference function does not meet a preset training target, adjust a weight of the initial load prediction model according to the first difference function to obtain a new initial load prediction model;
and returning to execute the step of inputting the electricity price time partition into the initial load prediction model according to the new initial load prediction model to obtain load output data.
In an embodiment, the model establishing module 10 is further configured to determine an initial energy generation prediction model according to a preset neural network model;
acquiring historical energy generation data and historical environmental states of a target building, wherein the historical energy generation data comprises historical light energy generation data and historical wind energy generation data;
and training the initial energy generation prediction model according to the historical energy generation data and the historical environment state to obtain an energy generation prediction model.
In an embodiment, the model building module 10 is further configured to input the historical environmental state into the initial energy generation prediction model to obtain energy generation output data;
determining a second difference function according to the energy generation output data and historical energy generation data;
and when the second difference function meets a preset training target, determining an energy generation prediction model according to the initial energy generation prediction model.
In an embodiment, the model establishing module 10 is further configured to, when the second difference function does not meet a preset training target, adjust the weight of the initial energy generation prediction model according to the second difference function, so as to obtain a new initial energy generation prediction model;
and according to the new initial energy generation prediction model, returning to execute the step of inputting the historical environment state into the initial energy generation prediction model to obtain energy generation output data.
In an embodiment, the energy scheduling module 30 is further configured to determine energy supply data according to a corresponding relationship between the load and energy supply, where the energy supply data includes power supply data, light energy supply data, and wind energy supply data;
determining an optimal weight coefficient according to the energy supply data;
determining energy demand data according to the optimal weight coefficient;
and determining an energy scheduling strategy according to the energy demand data.
It should be understood that the above is only an example, and the technical solution of the present invention is not limited in any way, and in a specific application, a person skilled in the art may set the technical solution as needed, and the present invention is not limited thereto.
It should be noted that the above-mentioned work flows are only illustrative and do not limit the scope of the present invention, and in practical applications, those skilled in the art may select some or all of them according to actual needs to implement the purpose of the solution of the present embodiment, and the present invention is not limited herein.
In addition, the technical details that are not described in detail in this embodiment may be referred to the energy scheduling method based on building load prediction provided in any embodiment of the present invention, and are not described herein again.
Further, it is to be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention or portions thereof that contribute to the prior art may be embodied in the form of a software product, where the computer software product is stored in a storage medium (e.g. Read Only Memory (ROM)/RAM, magnetic disk, optical disk), and includes several instructions for enabling a terminal device (e.g. a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. The energy scheduling method based on the building load prediction is characterized by comprising the following steps of:
acquiring historical load data of a target building, and establishing a load prediction model according to the historical load data and power price time partitions;
acquiring historical energy generation data and historical environment states of a target building, and establishing an energy generation prediction model according to the historical energy generation data and the historical environment states;
inputting the current electricity price time into the load prediction model in a partition mode to obtain predicted load data;
inputting the current environment state into the energy generation prediction model to obtain predicted energy generation data;
according to the predicted load data, the predicted energy generation data and the current electricity price time partition, determining the corresponding relation between the load and the energy supply;
obtaining an energy scheduling strategy according to the corresponding relation between the load and the energy supply;
and scheduling the energy supply of the target building according to the energy scheduling strategy.
2. The method of claim 1, wherein determining a correspondence between load and energy supply based on the predicted load data, predicted energy generation data, and current power rate time partition comprises:
determining current electricity price information according to the current electricity price time partition;
determining an initial corresponding relation between the load and the energy supply according to the predicted load data and the predicted energy generation data;
determining energy cost data according to the current electricity price information;
determining a loss function according to the energy cost data;
and optimizing the initial corresponding relation between the load and the energy supply according to the loss function and a preset multi-objective optimization algorithm to obtain the corresponding relation between the load and the energy supply.
3. The method of claim 1, wherein the obtaining historical load data of the target building, and the establishing a load prediction model based on the historical load data and the time division of electricity prices comprises:
determining the power price time partition according to the power price information;
acquiring historical load data of a target building;
determining an initial load prediction model according to a preset neural network model;
and partitioning according to the historical load data and the electricity price time, and training the initial load prediction model to obtain a load prediction model.
4. The method of claim 3, wherein said training said initial load prediction model based on said historical load data and power rate time partitions to obtain a load prediction model comprises:
inputting the electricity price time partition into the initial load prediction model to obtain load output data;
determining a first difference function according to the load output data and historical load data;
and when the first difference function meets a preset training target, determining a load prediction model according to the initial load prediction model.
5. The method of claim 4, wherein after determining a first difference function based on the load output data and historical load data, further comprising:
when the first difference function does not meet a preset training target, adjusting the weight of the initial load prediction model according to the first difference function to obtain a new initial load prediction model;
and returning to execute the step of inputting the electricity price time partition into the initial load prediction model according to the new initial load prediction model to obtain load output data.
6. The method of claim 1, wherein the obtaining historical energy generation data and historical environmental conditions of the target building, and establishing an energy generation prediction model based on the historical energy generation data and historical environmental conditions comprises:
determining an initial energy generation prediction model according to a preset neural network model;
acquiring historical energy generation data and historical environmental states of a target building, wherein the historical energy generation data comprises historical light energy generation data and historical wind energy generation data;
and training the initial energy generation prediction model according to the historical energy generation data and the historical environment state to obtain an energy generation prediction model.
7. The method of claim 6, wherein training an initial energy generation prediction model based on the historical energy generation data and historical environmental conditions to obtain an energy generation prediction model comprises:
inputting the historical environment state into the initial energy generation prediction model to obtain energy generation output data;
determining a second difference function according to the energy generation output data and historical energy generation data;
and when the second difference function meets a preset training target, determining an energy generation prediction model according to the initial energy generation prediction model.
8. The method of claim 7, wherein after determining a second difference function based on the energy generation output data and historical energy generation data, further comprising:
when the second difference function does not meet a preset training target, adjusting the weight of the initial energy generation prediction model according to the second difference function to obtain a new initial energy generation prediction model;
and according to the new initial energy generation prediction model, returning to execute the step of inputting the historical environment state into the initial energy generation prediction model to obtain energy generation output data.
9. The method according to any one of claims 1 to 8, wherein the obtaining an energy scheduling policy according to the correspondence between the load and the energy supply comprises:
determining energy supply data according to the corresponding relation between the load and energy supply, wherein the energy supply data comprises power supply data, light energy supply data and wind energy supply data;
determining an optimal weight coefficient according to the energy supply data;
determining energy demand data according to the optimal weight coefficient;
and determining an energy scheduling strategy according to the energy demand data.
10. An energy scheduling system based on building load prediction, the energy scheduling system based on building load prediction comprising:
the model building module is used for obtaining historical load data of a target building and building a load prediction model according to the historical load data and the electricity price time partition;
the model establishing module is also used for acquiring historical energy generation data and historical environmental states of a target building and establishing an energy generation prediction model according to the historical energy generation data and the historical environmental states;
the data prediction module is used for inputting the current electricity price time partition into the load prediction model to obtain predicted load data;
the data prediction module is also used for inputting the current environment state into the energy generation prediction model to obtain predicted energy generation data;
the energy scheduling module is used for determining the corresponding relation between the load and the energy supply according to the predicted load data, the predicted energy generation data and the current electricity price time partition;
the energy scheduling module is further used for obtaining an energy scheduling strategy according to the corresponding relation between the load and the energy supply;
and the energy scheduling module is also used for scheduling the energy supply of the target building according to the energy scheduling strategy.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211289125.1A CN115906408A (en) | 2022-10-20 | 2022-10-20 | Energy scheduling system and method based on building load prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211289125.1A CN115906408A (en) | 2022-10-20 | 2022-10-20 | Energy scheduling system and method based on building load prediction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115906408A true CN115906408A (en) | 2023-04-04 |
Family
ID=86473507
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211289125.1A Pending CN115906408A (en) | 2022-10-20 | 2022-10-20 | Energy scheduling system and method based on building load prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115906408A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116629775A (en) * | 2023-04-20 | 2023-08-22 | 公安县茂业建材有限公司 | Intelligent production control system for assembled building materials based on Internet of things technology |
| CN116719861A (en) * | 2023-06-27 | 2023-09-08 | 哈尔滨源芯智能科技发展有限公司 | Multi-source data interaction management system and method based on big data |
-
2022
- 2022-10-20 CN CN202211289125.1A patent/CN115906408A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116629775A (en) * | 2023-04-20 | 2023-08-22 | 公安县茂业建材有限公司 | Intelligent production control system for assembled building materials based on Internet of things technology |
| CN116629775B (en) * | 2023-04-20 | 2024-04-12 | 铯镨科技有限公司 | Intelligent production control system for assembled building materials based on Internet of things technology |
| CN116719861A (en) * | 2023-06-27 | 2023-09-08 | 哈尔滨源芯智能科技发展有限公司 | Multi-source data interaction management system and method based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Huang et al. | Demand response for industrial micro-grid considering photovoltaic power uncertainty and battery operational cost | |
| Hafiz et al. | Real-time stochastic optimization of energy storage management using deep learning-based forecasts for residential PV applications | |
| CN115906408A (en) | Energy scheduling system and method based on building load prediction | |
| CN111466063B (en) | Energy storage management and control method, system, computer equipment and storage medium | |
| JPWO2018147194A1 (en) | Storage battery control device, charge / discharge control method, and program | |
| Hooshmand et al. | Experimental demonstration of a tiered power management system for economic operation of grid-tied microgrids | |
| US20040117236A1 (en) | Automated optimization tool for electric utility sypply services | |
| CN113780660A (en) | Resident electricity consumption prediction method, system and storage medium | |
| Xiao et al. | Short-term power load interval forecasting based on nonparametric Bootstrap errors sampling | |
| CN114707737B (en) | Method for predicting power consumption based on edge calculation, computer equipment and storage medium | |
| Wen et al. | Stochastic optimization for security-constrained day-ahead operational planning under pv production uncertainties: Reduction analysis of operating economic costs and carbon emissions | |
| JPH09215192A (en) | Method for predicting daily load curve | |
| CN114662751B (en) | Garden multifunctional short-term load forecasting and optimizing method based on LSTM | |
| Ji et al. | Optimal schedule of solid electric thermal storage considering consumer behavior characteristics in combined electricity and heat networks | |
| Ceccon et al. | Intelligent electric power management system for economic maximization in a residential prosumer unit | |
| Li et al. | Deep integration planning of sustainable energies in district energy system and distributed energy station | |
| CN114118532A (en) | Scheduling method and device for island microgrid, computer equipment and storage medium | |
| WO2020103046A1 (en) | Energy storage management and control methods, systems, computer device, and storage medium | |
| Pacaud et al. | Optimization of a domestic microgrid equipped with solar panel and battery: Model predictive control and stochastic dual dynamic programming approaches | |
| Nath et al. | Online battery scheduling for grid-connected photo-voltaic systems | |
| CN103870886A (en) | Prediction method for monthly maximal power consumption of enterprise | |
| CN114498702A (en) | Micro-grid energy management method and system based on energy router | |
| CN111682539A (en) | Partitioning method and system for power grid with distributed power supply based on simplified comprehensive scene | |
| Zhang et al. | Load Forecasting Considering Demand Response Mechanism | |
| CN112348235A (en) | Wind-solar bus load self-adaptive prediction method and device and computer equipment |
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 |