CN114465264A

CN114465264A - Integrated energy system and optimization control method thereof

Info

Publication number: CN114465264A
Application number: CN202111568613.1A
Authority: CN
Inventors: 朱林; 金阳坤; 黄重月; 牟世忠
Original assignee: Tsinghua University; Huawei Digital Power Technologies Co Ltd
Current assignee: Tsinghua University; Huawei Digital Power Technologies Co Ltd
Priority date: 2021-12-21
Filing date: 2021-12-21
Publication date: 2022-05-10

Abstract

The application provides a comprehensive energy system and an optimization control method thereof, which are used for identifying and eliminating transient data of each subsystem according to real-time data of each subsystem, overcoming the influence of a transient process on the comprehensive energy system and improving the robustness of a control model of the comprehensive energy system. The method and the device can identify and eliminate the transient data of each subsystem in the comprehensive energy system, and obtain the steady-state data of each subsystem. And inputting the steady-state data of each subsystem into the control model of each subsystem to solve, and acquiring the output information of each subsystem. And the output information of each subsystem is input into the control model of the comprehensive energy system to be solved, the control strategy of the comprehensive energy system is obtained, and the optimal control of the comprehensive energy system is realized.

Description

Integrated energy system and optimization control method thereof

Technical Field

The present application relates to the field of energy technology, and more particularly, to an integrated energy system and an optimization control method thereof.

Background

With the development of renewable energy technology, comprehensive energy systems (including renewable energy subsystems, load subsystems and the like) are widely applied. Generally, the safety, economy and reliability of the comprehensive energy system can be improved by adjusting the operation parameters and configuration parameters of equipment in each subsystem in the comprehensive energy system, so that the utilization rate of the renewable energy subsystem is improved, the operation cost of the load subsystem is reduced, the large-scale development of the renewable energy subsystem is promoted, and the energy conservation and emission reduction are realized.

Due to the existence of transient processes (start-up and shut-down processes, disturbance processes and the like) and steady-state processes (i.e., normal operation) in the integrated energy system, the transient processes affect the control of the entire integrated energy system. Moreover, because the operation conditions of the subsystems are different, the fixed single model cannot be adopted to carry out edge cooperative control on the subsystems under different operation conditions. Therefore, a technical scheme capable of overcoming the influence of the transient process and realizing the optimization control of the comprehensive energy system through the edge cloud cooperative control of each subsystem is urgently needed.

Disclosure of Invention

The application provides an integrated energy system and an optimization control method thereof, which are used for identifying and eliminating transient data of each subsystem according to real-time data of each subsystem, overcoming the influence of a transient process on the integrated energy system, and further realizing the optimization control of the integrated energy system through steady-state data of each subsystem, a control model of each subsystem and a control model of the integrated energy system.

In a first aspect, the present application provides a method for optimizing and controlling an integrated energy system, which may include: and identifying and eliminating the transient data of each subsystem according to the real-time data of each subsystem in the comprehensive energy system, and acquiring the steady-state data of each subsystem. And inputting the steady-state data of each subsystem into the control model of each subsystem which is acquired in advance to solve, and acquiring the output information of each subsystem. And inputting the output information of each subsystem into a pre-constructed control model of the comprehensive energy system for solving to obtain a control strategy of the comprehensive energy system.

Optionally, each subsystem comprises at least two of a renewable energy subsystem, an energy storage subsystem and a load subsystem, the renewable energy subsystem comprising a wind electronic system and/or a photovoltaic subsystem.

Thus, the real-time data for each subsystem may include at least two of the real-time data for the renewable energy subsystem, the real-time data for the energy storage subsystem, and the real-time data for the load subsystem.

The real-time data of the renewable energy subsystem can include transient data of the renewable energy subsystem (including power generation data of the wind power subsystem when the actual wind speed is less than a preset wind speed threshold and/or power generation data of the photovoltaic subsystem when the solar irradiance is less than a preset solar irradiance threshold) and steady-state data of the renewable energy subsystem (including power generation data of the wind power subsystem when the actual wind speed is greater than or equal to the preset wind speed threshold and/or power generation data of the photovoltaic subsystem when the solar irradiance is greater than or equal to the preset solar irradiance threshold).

The state of the energy storage subsystem may be a sleep state (the energy storage subsystem is neither charged nor discharged) or an operational state (the energy storage subsystem is charged or discharged). Thus, the real-time data of the energy storage subsystem may include data that the energy storage subsystem is in a sleep state and data that the energy storage subsystem is in an operational state.

The data of the energy storage subsystem in the dormant state may include actual electric quantity of the energy storage subsystem in the dormant state, and the data of the energy storage subsystem in the working state may include charging electric quantity, charging duration, discharging electric quantity and discharging duration of the energy storage subsystem.

The real-time data of the load subsystem may include load data in which the power fluctuation value of the load subsystem is less than a preset power fluctuation threshold and load data in which the power fluctuation value of the load subsystem is greater than or equal to the preset power fluctuation threshold.

Thus, the transient data of each subsystem may include at least two of data of the energy storage subsystem in a sleep state, load data of the load subsystem having a power fluctuation value smaller than a preset power fluctuation threshold, and transient data of the renewable energy subsystem (described above).

Similarly, the steady state data of each subsystem comprises at least two of data of the energy storage subsystem in a working state, load data of the load subsystem with a power fluctuation value larger than or equal to a preset power fluctuation threshold value, and steady state data of the renewable energy subsystem.

Alternatively, the output information of each subsystem may include power usage of the load and the power generation of the renewable subsystem.

Optionally, the control strategy of the integrated energy system may include at least two items of power generation amount of the renewable energy subsystem within a preset time range, charging and discharging information of the energy storage subsystem, and power utilization information of the load subsystem.

Illustratively, the electricity consumption information of the load subsystem includes the electricity consumption duration and the electricity consumption amount of the load subsystem.

The charging and discharging information of the energy storage subsystem may include charging information and/or discharging information of the energy storage subsystem. The charging information may include a charging duration and a charging electric quantity of the energy storage subsystem, and the discharging information may include a discharging duration and a discharging electric quantity of the energy storage subsystem.

The method and the system can identify and eliminate the transient data of each subsystem according to the real-time data of each subsystem, overcome the influence of the transient process on the comprehensive energy system, and further realize the optimal control of the comprehensive energy system through the steady-state data of each subsystem, the control model of each subsystem and the control model of the comprehensive energy system.

In a possible implementation manner, identifying and eliminating transient data of each subsystem according to real-time data of each subsystem in the integrated energy system, and acquiring steady-state data of each subsystem may include: and extracting the frequency spectrum characteristics of the real-time data of each subsystem by adopting a frequency spectrum characteristic analysis method. And identifying the transient data of each subsystem according to the frequency spectrum characteristics to obtain the transient data of each subsystem. And eliminating the transient data of each subsystem from the real-time data of each subsystem to obtain the steady-state data of each subsystem.

The method and the system identify and remove the transient data of each subsystem according to the real-time data of each subsystem, overcome the influence of the transient process on the comprehensive energy system, ensure the robustness of identification of the identifiable model family, and provide a foundation for the optimal control of the comprehensive energy system.

In a possible implementation manner, the control model of each subsystem can be obtained by constructing the recognizable model family of each subsystem and recognizing the recognizable model family of each subsystem by using the interactive multi-model.

Optionally, the process of building the recognizable model family of each subsystem may include: and identifying the operation condition of each subsystem by adopting a Gaussian mixture model. And constructing an identifiable model family of each subsystem according to the operation condition of each subsystem.

The operation conditions of the subsystems may include at least two of the operation conditions of the load subsystem, the operation conditions of the energy storage subsystem, and the operation conditions of the renewable energy subsystem. The operation condition of the renewable energy subsystem may include an operation condition of the wind power subsystem and/or an operation condition of the photovoltaic subsystem.

Further, the operation condition of the load subsystem comprises a user operation condition or an industrial and commercial operation condition.

The operation condition of the energy storage subsystem comprises a household operation condition, an industrial and commercial operation condition or a ground power station operation condition.

The operation working conditions of the wind power subsystem comprise plain operation working conditions, mountain operation working conditions or seaside operation working conditions.

The operation condition of the photovoltaic subsystem comprises a household operation condition, an industrial and commercial operation condition or a ground power station operation condition.

In one possible implementation, the distinguishable family of models may include a distinguishable family of models of the load subsystem and a distinguishable family of models of the renewable energy generation subsystem.

The identifiable model family of the renewable energy power generation subsystem may include an identifiable model of the wind power subsystem and/or an identifiable model of the photovoltaic subsystem.

Further, the identifiable model family of the load subsystem includes at least two of a differential integrated moving average autoregressive model, a recurrent neural network model, and a long-term and short-term memory network model.

The identifiable model of the wind power subsystem may include at least two of a semi-empirical model, a vortex model, and a computational fluid dynamics model.

The distinguishable model of the photovoltaic subsystem may include at least two of a first polynomial nonlinear model, a second polynomial nonlinear model, and a third polynomial nonlinear model.

The first polynomial nonlinear model is constructed by taking the horizontal irradiance and the air temperature as input, according to the corresponding relation between the power generation capacity of the photovoltaic subsystem and the horizontal irradiance and the air temperature, and taking the power generation capacity of the photovoltaic subsystem as output.

Similarly, the second polynomial nonlinear model is constructed by taking the inclined irradiance and the air temperature as input, according to the corresponding relation of the power generation amount of the photovoltaic subsystem and the inclined irradiance and the air temperature, and taking the power generation amount of the photovoltaic subsystem as output.

Similarly, the third polynomial nonlinear model is constructed by taking the horizontal irradiance, the oblique irradiance and the air temperature as input, according to the corresponding relation of the power generation capacity of the photovoltaic subsystem and the horizontal irradiance, the oblique irradiance and the air temperature, and taking the power generation capacity of the photovoltaic subsystem as output.

In one possible implementation, the first objective function of the integrated energy system may be constructed with the goal of minimizing the total cost of the integrated energy system. And constructing the constraint conditions of the comprehensive energy system according to the first objective function. And constructing a control model of the comprehensive energy system according to the first objective function and the constraint condition.

Alternatively, the first objective function may be constructed as follows:

min C_Total＝C_{bill_grid_buy}-E_{income_grid_sale}

in the formula, C_TotalRepresents the total cost of the integrated energy system, C_{bill_grid_buy}Indicating the electricity purchase rate paid by the user to the grid, E_{income_grid_sale}And showing the online electricity selling income of the renewable energy subsystem.

In another possible implementation, the second objective function of the integrated energy system may be constructed with the goal of maximizing the total profit of the integrated energy system. And constructing the constraint conditions of the comprehensive energy system according to the second objective function. And constructing a control model of the comprehensive energy system according to the second objective function and the constraint condition.

Optionally, the second objective function is constructed as follows:

maxE_Total＝E_{income_grid_sale}+E_{bill_self_sufficency}

in the formula, E_TotalRepresenting the total profit of the integrated energy system, E_{income_grid_sale}Representing the net electricity sales revenue of the renewable energy subsystem, E_{bill_self_sufficency}Representing a self-sufficient saving of electricity charges by the user using the renewable energy subsystem.

In one possible implementation, the constraint condition may include a charging and discharging constraint and/or a power grid constraint.

The charging and discharging constraint can be used for indicating that the state of charge of the energy storage subsystem is greater than or equal to a preset state of charge lower limit and less than or equal to a preset state of charge upper limit.

Grid constraints may be used to indicate that the user's power usage is within the range of power usage provided by the grid and does not exceed the power usage agreed upon by the user with the grid.

In a possible implementation manner, inputting the steady-state data of each subsystem into the control model of each subsystem to perform solution, and acquiring the output information of each subsystem may include: the steady state data of each subsystem can be input into the pre-constructed control model of each subsystem according to the preset time interval granularity, and the control model of each subsystem is solved by adopting an unconstrained optimization method to obtain the output information of each subsystem.

Alternatively, the unconstrained optimization method may be any one of a gradient descent method (i.e., steepest descent method), a conjugate direction method (i.e., conjugate gradient method), a newton method, and a quasi-newton method. Of course, other unconstrained optimization methods may be used, and the present application is not limited thereto.

In a possible implementation manner, inputting the output information of each subsystem into a control model of the integrated energy system to solve, and obtaining a control strategy of the integrated energy system may include: and inputting the output information of each subsystem into the control model of the comprehensive energy system, and solving the control model of the comprehensive energy system by adopting a constrained optimization method to obtain a control strategy of the comprehensive energy system.

Alternatively, the constrained optimization method may be any one of a linear programming method, a non-linear programming method, and the like. Of course, other constrained optimization methods may be adopted to solve the control model of the integrated energy system, and the application is not limited in this respect.

The influence of the transient process on the optimization control of the comprehensive energy system is overcome by identifying and eliminating the transient data of each subsystem, and the edge cloud cooperative control of each subsystem is realized through the control model of each subsystem and the control model of the comprehensive energy system, so that the optimization control of the comprehensive energy system is realized.

In a second aspect, the present application further provides an integrated energy system that may include a controller and a plurality of subsystems, the controller being coupled to each of the plurality of subsystems.

The controller may be configured to:

and identifying and eliminating the transient data of each subsystem according to the real-time data of each subsystem in the comprehensive energy system, and acquiring the steady-state data of each subsystem.

And inputting the steady-state data of each subsystem into the control model of each subsystem which is acquired in advance to solve, and acquiring the output information of each subsystem.

And inputting the output information of each subsystem into a pre-constructed control model of the comprehensive energy system for solving to obtain a control strategy of the comprehensive energy system.

Each subsystem may be for: and operating according to the control strategy of the comprehensive energy system.

The power generation amount of the renewable subsystem can comprise the power generation amount of the photovoltaic subsystem and/or the power generation amount of the wind power subsystem. The control strategy of the integrated energy system can comprise at least two items of generated energy of the renewable energy subsystem in a preset time range, charging and discharging information of the energy storage subsystem and power utilization information of the load subsystem.

Further, the charging and discharging information of the energy storage subsystem may include charging information and/or discharging information of the energy storage subsystem. The charging information may include a charging duration and a charging electric quantity of the energy storage subsystem, and the discharging information may include a discharging duration and a discharging electric quantity of the energy storage subsystem.

The power consumption information of the load subsystem can comprise the power consumption duration and the power consumption of the load subsystem.

In one possible implementation, the controller may be configured to: and extracting the frequency spectrum characteristics of the real-time data of each subsystem by adopting a frequency spectrum characteristic analysis method. And identifying the transient data of each subsystem according to the frequency spectrum characteristics to obtain the transient data of each subsystem. And eliminating the transient data of each subsystem from the real-time data of each subsystem to obtain the steady-state data of each subsystem.

In another possible implementation, the controller may be configured to: and identifying the operation condition of each subsystem by adopting a Gaussian mixture model. And constructing an identifiable model family of each subsystem according to the operation condition of each subsystem.

Further, the operation condition of each subsystem may include at least two of the operation condition of the load subsystem, the operation condition of the energy storage subsystem, and the operation condition of the renewable energy subsystem. The operation condition of the renewable energy subsystem may include an operation condition of the wind power subsystem and/or an operation condition of the photovoltaic subsystem.

Further, the controller may construct a first polynomial nonlinear model with the horizontal irradiance and the air temperature as inputs, according to a correspondence of the power generation of the photovoltaic subsystem with the horizontal irradiance and the air temperature, and with the power generation of the photovoltaic subsystem as an output.

Similarly, the controller may construct a second polynomial nonlinear model based on the correspondence of the power generation of the photovoltaic subsystem to the oblique irradiance and the air temperature with the oblique irradiance and the air temperature as inputs and with the power generation of the photovoltaic subsystem as an output.

Similarly, the controller may construct a third polynomial nonlinear model based on the correspondence of the power generation of the photovoltaic subsystem to the horizontal irradiance, the oblique irradiance, and the air temperature with the horizontal irradiance, the oblique irradiance, and the air temperature as inputs, and with the power generation of the photovoltaic subsystem as an output.

According to the embodiment of the application, the control models of the subsystems under different operation conditions are constructed according to the different operation conditions of the subsystems, the reusability of the identifiable model family is solved, and the adaptability of the comprehensive energy system under various operation condition scenes is improved.

In a possible implementation manner, the controller may identify the identifiable model families of the subsystems by using an interactive multi-model to obtain the control models of the subsystems.

In one possible implementation, the controller may be configured to: and constructing a first objective function of the comprehensive energy system by taking the lowest total cost of the comprehensive energy system as an objective. And constructing the constraint conditions of the comprehensive energy system according to the first objective function. And constructing a control model of the comprehensive energy system according to the first objective function and the constraint condition.

Further, the controller may construct the first objective function as follows:

min C_Total＝C_{bill_grid_buy}-E_{income_grid_sale}

In another possible implementation, the controller may be configured to: the second objective function of the integrated energy system may be constructed with the goal of maximizing the total profit of the integrated energy system. And constructing the constraint conditions of the comprehensive energy system according to the second objective function. And constructing a control model of the comprehensive energy system according to the second objective function and the constraint condition.

Further, the controller may construct the second objective function as follows:

maxE_Total＝E_{income_grid_sale}+E_{bill_self_sufficency}

In one possible implementation, the constraints constructed by the controller may include charging and discharging constraints and/or grid constraints.

In one possible implementation, the controller may be configured to: and inputting the steady-state data of each subsystem into a pre-constructed control model of each subsystem according to the preset time interval granularity, and solving the control model of each subsystem by adopting an unconstrained optimization method to obtain the output information of each subsystem.

In one possible implementation, the controller may be configured to: and inputting the output information of each subsystem into the control model of the comprehensive energy system, and solving the control model of the comprehensive energy system by adopting a constrained optimization method to obtain a control strategy of the comprehensive energy system.

Alternatively, the constrained optimization method may be any one of a linear programming method, a non-linear programming method, and the like. Of course, other constrained optimization methods may also be used to solve the control model of the integrated energy system, and the embodiment of the present application is not limited.

It should be understood that the second aspect of the present application is consistent with the technical solution of the first aspect of the present application, and similar advantageous effects are obtained in various aspects and corresponding possible embodiments, and thus, detailed description is omitted.

Drawings

In order to more clearly illustrate the technical solutions in the present application or the prior art, the following briefly introduces the drawings needed to be used in the description of the embodiments or the prior art, and it is obvious that the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to the drawings without inventive labor.

FIG. 1 provides a schematic flow chart diagram of a method for optimizing control of an integrated energy system according to an embodiment of the present application;

FIG. 2 provides a schematic flow chart of building a family of identifiable models for each subsystem according to an embodiment of the present application;

fig. 3 provides a schematic flowchart of an optimization control method using an intelligent light storage system according to an embodiment of the present application;

fig. 4 provides a schematic diagram of the roll optimization control of the user intelligent light storage system according to the embodiment of the present application;

fig. 5 provides a schematic structural diagram of an integrated energy system according to an embodiment of the present application.

Detailed Description

The technical solution in the present application will be described below with reference to the accompanying drawings.

To make the purpose, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be clearly and completely described below with reference to the drawings in the present application, and it is obvious that the described embodiments are some, but not all embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The terms "first," "second," and the like in the description examples and claims of this application and in the drawings are used for descriptive purposes only and are not to be construed as indicating or implying relative importance, nor order. Furthermore, the terms "comprises" and "comprising," as well as any variations thereof, are intended to cover a non-exclusive inclusion, such as a list of steps or elements. A method, system, article, or apparatus is not necessarily limited to those steps or elements explicitly listed, but may include other steps or elements not explicitly listed or inherent to such process, system, article, or apparatus.

It should be understood that in the present application, "at least one" means one or more, "a plurality" means two or more. "and/or" for describing an association relationship of associated objects, indicating that there may be three relationships, e.g., "a and/or B" may indicate: only A, only B and both A and B are present, wherein A and B may be singular or plural. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of single item(s) or plural items. For example, at least one (one) of a, b, or c, may represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", wherein a, b, c may be single or plural.

With the development of renewable energy technology, comprehensive energy systems are widely applied. The integrated energy system may include at least two of a renewable energy subsystem, an energy storage subsystem, and a load subsystem. Since renewable energy sources may typically include wind power and photovoltaic, the renewable energy subsystem may include a wind electronic system and/or a photovoltaic subsystem. Of course, the integrated energy system may further include other subsystems besides the renewable energy subsystem, the energy storage subsystem, and the load subsystem, and the renewable energy subsystem may also include other subsystems besides the wind electronic system and/or the photovoltaic subsystem, which is not limited in the embodiment of the present application.

Generally, the safety, economy and reliability of the comprehensive energy system can be improved by adjusting the operation parameters and configuration parameters of equipment in each subsystem (i.e. the renewable energy subsystem, the energy storage subsystem and the load subsystem) in the comprehensive energy system, so that the utilization rate of the renewable energy subsystem is improved, the operation cost of the load subsystem is reduced, the large-scale development of the renewable energy subsystem is promoted, and energy conservation and emission reduction are realized.

Due to the existence of transient processes (start-up and shut-down processes, disturbance processes and the like) and steady-state processes (i.e., normal operation) in the integrated energy system, the transient processes affect the control of the entire integrated energy system. Moreover, because the operation conditions of the subsystems are different, the fixed single model cannot be adopted to carry out edge cooperative control on the subsystems under different operation conditions.

In order to overcome the influence of the transient process and realize the optimal control of the integrated energy system through the edge cloud cooperative control of each subsystem, an embodiment of the present application provides an optimal control method of the integrated energy system, and as shown in fig. 1, the optimal control method 100 may be implemented according to the following steps:

step S101: and identifying and eliminating the transient data of each subsystem according to the real-time data of each subsystem in the comprehensive energy system, and acquiring the steady-state data of each subsystem.

The real-time data of each subsystem can comprise at least two of the real-time data of the renewable energy subsystem, the real-time data of the energy storage subsystem and the real-time data of the load subsystem.

The state of the energy storage subsystem may be a sleep state (the energy storage subsystem is neither charged nor discharged) or an operational state (the energy storage subsystem is charged or discharged). Therefore, the real-time data of the energy storage subsystem may include data that the energy storage subsystem is in a dormant state (including the actual amount of power that the energy storage subsystem is in the dormant state) and data that the energy storage subsystem is in an operating state (including the amount of charge power, the charging duration, the discharging power and the discharging duration of the energy storage subsystem).

The real-time data of the load subsystem may include load data in which a power fluctuation value of the load subsystem is less than a preset power fluctuation threshold and load data in which a power fluctuation value of the load subsystem is greater than or equal to the preset power fluctuation threshold.

In one possible implementation, step S101 may include: and extracting the frequency spectrum characteristics of the real-time data of each subsystem by adopting a frequency spectrum characteristic analysis method. And identifying the transient data of each subsystem according to the frequency spectrum characteristics to obtain the transient data of each subsystem. And eliminating the transient data of each subsystem from the real-time data of each subsystem to obtain the steady-state data of each subsystem.

The transient data of each subsystem are identified and eliminated according to the real-time data of each subsystem, the influence of the transient process on the comprehensive energy system is overcome, the robustness of identification of the identifiable model family is guaranteed, and a foundation is provided for the optimization control of the comprehensive energy system.

Step S102: and inputting the steady-state data of each subsystem into the pre-acquired control model of each subsystem to solve, and acquiring the output information of each subsystem.

Alternatively, the control model of each subsystem may be obtained by identifying a pre-constructed identifiable model family of each subsystem by using an Interactive Multiple Model (IMM). Of course, the recognizable model families of the subsystems can be recognized in other manners, and the embodiment of the present application is not limited.

Further, as shown in fig. 2, the recognizable model family of each subsystem can be pre-constructed according to the following steps:

step S102a 1: and identifying the operation condition of each subsystem by adopting a Gaussian mixture model.

Optionally, the operation condition of each subsystem can be identified by adopting a gaussian mixture model according to the actual condition of each subsystem.

In a possible implementation manner, the operation conditions of the subsystems may include at least two of the operation conditions of the load subsystem, the operation conditions of the energy storage subsystem, and the operation conditions of the renewable energy subsystem. The operation condition of the renewable energy subsystem may include an operation condition of the wind power subsystem and/or an operation condition of the photovoltaic subsystem.

Further, the operation condition of the load subsystem comprises a user operation condition or an industrial and commercial operation condition;

the operation working conditions of the energy storage subsystem comprise household operation working conditions, industrial and commercial operation working conditions or ground power station operation working conditions;

the operation working conditions of the wind power subsystem comprise plain operation working conditions, mountain operation working conditions or seaside operation working conditions;

Of course, the operation conditions of the subsystems are not limited to the listed operation conditions, and may also be other operation conditions, which is not limited in the embodiment of the present application.

Step S102a 2: and constructing an identifiable model family of each subsystem according to the operation condition of each subsystem.

Since the identifiable model family of each subsystem is closely related to the operating condition, the identifiable model family of each subsystem needs to be constructed according to the operating condition.

In one possible implementation, the recognizable model family may include a recognizable model family of the load subsystem and a recognizable model family of the renewable energy power generation subsystem.

Further, the family of identifiable models of the renewable energy power generation subsystem may include an identifiable model of the wind power subsystem and/or an identifiable model of the photovoltaic subsystem.

Still further, the recognizable model family of the load subsystem may include at least two of a differential integrated moving average autoregressive (ARIMA) model (referred to as ARIMA model), a Recurrent Neural Network (RNN) model (referred to as RNN model), and a long short term memory network (LSTM) model (referred to as LSTM model).

It should be noted that, since the recognizable model of the load subsystem needs to be identified to obtain the control model of the load subsystem, the recognizable model family of the load subsystem includes at least two models.

Similarly, since the recognizable model of the wind power subsystem needs to be recognized to obtain the control model of the wind power subsystem, the recognizable model family of the wind power subsystem includes at least two models.

The distinguishable model of the photovoltaic subsystem includes at least two of a first polynomial nonlinear model, a second polynomial nonlinear model, and a third polynomial nonlinear model. Similarly, since the recognizable model of the photovoltaic subsystem needs to be recognized to obtain the control model of the photovoltaic subsystem, the recognizable model family of the photovoltaic subsystem includes at least two models.

In the photovoltaic subsystem, the power generation amount of the photovoltaic subsystem may be in a corresponding relationship (such as a second order relationship) with a horizontal irradiance (GHI), an oblique irradiance (GTI), and an air temperature, but in actual application, the horizontal irradiance, the oblique irradiance, and the air temperature are easily affected by factors such as an installation angle of the photovoltaic subsystem (which may be a photovoltaic array in the photovoltaic subsystem), and the horizontal irradiance, the oblique irradiance, and the air temperature all have a certain uncertainty.

Therefore, the first polynomial nonlinear model can be constructed by taking the horizontal irradiance and the air temperature as input, according to the corresponding relation (such as a second-order relation) of the power generation amount of the photovoltaic subsystem and the horizontal irradiance and the air temperature, and taking the power generation amount of the photovoltaic subsystem as output. It can be seen that the first polynomial nonlinear model is a binary quadratic polynomial nonlinear model. It can be seen that the first polynomial nonlinear model can be a binary quadratic nonlinear model.

The second polynomial nonlinear model can be constructed by taking the oblique irradiance and the air temperature as input, according to the corresponding relation (such as a second-order relation) of the power generation amount of the photovoltaic subsystem and the oblique irradiance and the air temperature, and taking the power generation amount of the photovoltaic subsystem as output. It can be seen that the second polynomial nonlinear model can also be a binary quadratic nonlinear model.

The third polynomial nonlinear model can be constructed by taking the horizontal irradiance, the oblique irradiance and the air temperature as input, according to the corresponding relation (such as a second-order relation) of the power generation capacity of the photovoltaic subsystem and the horizontal irradiance, the oblique irradiance and the air temperature, and taking the power generation capacity of the photovoltaic subsystem as output. It can be seen that the third polynomial nonlinear model can be a ternary quadratic nonlinear model.

In a possible implementation manner, the step S102 may be implemented according to the following procedures:

according to a preset time interval granularity (for example, 5 minutes, 15 minutes, 30 minutes, 1 hour, 2 hours, 4 hours, 24 hours and the like), the steady-state data of each subsystem can be input into the identified control model of each subsystem, and the control model of each subsystem is solved by using an unconstrained optimization method to obtain the output information of each subsystem.

Alternatively, the unconstrained optimization method may be any one of a gradient descent method (i.e., steepest descent method), a conjugate direction method (i.e., conjugate gradient method), a newton method, and a quasi-newton method. Of course, other unconstrained optimal methods can be adopted, which is not limited in the embodiments of the present application.

For example, the output information of each subsystem may include power consumption of the renewable subsystem and power consumption of the load. The power generation amount of the renewable subsystem can comprise the power generation amount of the photovoltaic subsystem and/or the power generation amount of the wind power subsystem.

Step S103: and inputting the output information of each subsystem into a pre-constructed control model of the comprehensive energy system for solving to obtain a control strategy of the comprehensive energy system.

In a possible implementation manner, the control model of the integrated energy system can be constructed on the basis of the control models of the subsystems according to the following two manners:

the first method is as follows: and constructing a first objective function of the comprehensive energy system by taking the lowest total cost of the comprehensive energy system as an objective. And constructing the constraint conditions of the comprehensive energy system according to the first objective function. And constructing a control model of the comprehensive energy system according to the first objective function and the constraint condition.

Further, the first objective function may be constructed according to the following equation:

min C_Total＝C_{bill_grid_buy}-E_{income_grid_sale}

It should be noted that, the present application is described by taking as an example that the integrated energy system includes at least two of a renewable energy subsystem (including a wind electronic system and/or a photovoltaic subsystem), an energy storage subsystem, and a load subsystem. Of course, the integrated energy system may also include a distributed power source, a gas turbine, and the like, so the first objective function may also be constructed according to the following formula:

min C_Total＝C_DG+C_Gas+C_{bill_grid_buy}-E_{income_grid_sale}

in the formula, C_DGRepresents the operating cost of the distributed power supply, C_GasRepresenting the operating cost of the gas turbine.

The second method comprises the following steps: and constructing a second objective function of the comprehensive energy system by taking the maximum total income of the comprehensive energy system as an objective. And constructing the constraint conditions of the comprehensive energy system according to the second objective function. And constructing a control model of the comprehensive energy system according to the second objective function and the constraint condition.

Further, the second objective function may be constructed according to the following equation:

maxE_Total＝E_{income_grid_sale}+E_{bill_self_sufficency}

Of course, in addition to the above two ways, the control model of the integrated energy system may also be constructed by adopting other ways on the basis of the control models of the subsystems in the embodiment of the present application, which is not limited in the embodiment of the present application.

It should be noted that, since the control model of the integrated energy system is constructed, the constraint condition of the integrated energy system constructed according to the first objective function may be the same as the constraint condition of the integrated energy system constructed according to the second objective function, and the constraint condition of the integrated energy system is described below.

Optionally, the constraint condition may include a charging and discharging constraint and/or a grid constraint. Of course, the constraint condition may also include other constraints, which are not described in detail in the embodiments of the present application.

Where charge-discharge constraints may be used to indicate the state of charge of the energy storage subsystem (state of charge,SOC) is greater than or equal to a preset lower state of charge limit (available SOC)_minRepresented by) and less than or equal to a preset upper state of charge (SOC may be used)_maxExpress), that is, satisfy the SOC_min≤SOC≤SOC_max。

In order to ensure the safe power utilization of the user, the grid constraint may be used to indicate that the power utilization of the user is within the range of the power utilization provided by the grid and does not exceed the power utilization agreed by the user and the grid, and may be formulated as:

-P_D-(t)≤P_Load(t)-P_PV(t)+P_{BES_charge}(t)-P_{BES_discharge}(t)≤P_D+(t)

in the formula, P_Load(t) represents the load power of the user, P_{PV_max}(t) represents the maximum output of the photovoltaic subsystem at time t, P_{BES_charge}(t) represents the charging power of the energy storage subsystem during the period t, P_{BES_discharge}(t) represents the discharge power of the energy storage subsystem during the period t, P_D+(t) represents the maximum forward power capacity of the consumer (maximum power purchased by the consumer from the grid), P_D-(t) represents the maximum reverse power capacity of the user (i.e., the maximum power sold by the user to the grid).

According to the control model of the comprehensive energy system, the control model of the comprehensive energy system is constructed according to the control models of the subsystems, so that the solution of the control model of the comprehensive energy system becomes possible, the optimal solution of the optimization control (namely the optimization control strategy of the comprehensive energy system) can be found, and the control model of the comprehensive energy system has strong practicability.

In a possible implementation manner, the step S103 may be implemented according to the following procedures:

and inputting the output information of each subsystem into the control model of the comprehensive energy system, and solving the control model of the comprehensive energy system by adopting a constrained optimization method to obtain a control strategy of the comprehensive energy system.

For example, the control strategy of the integrated energy system may include at least two of the power generation amount of the renewable energy subsystem within a preset time range, the charging and discharging information of the energy storage subsystem, and the power utilization information of the load subsystem.

Further, the power generation amount of the renewable energy subsystem in the preset time range may include the power generation amount of the photovoltaic subsystem in the preset first time range and/or the power generation amount of the wind power subsystem in the preset second time range. The first time range and the second time range may be the same or different.

Optionally, the charging and discharging information of the energy storage subsystem may include charging information and/or discharging information of the energy storage subsystem. The charging information may include a charging duration and a charging electric quantity of the energy storage subsystem, and the discharging information may include a discharging duration and a discharging electric quantity of the energy storage subsystem.

Optionally, the power consumption information of the load subsystem may include the power consumption duration and the power consumption amount of the load subsystem.

In a possible implementation manner, the controller may send the control strategy of the integrated energy system to each subsystem in a communication manner such as RS485, MBUS, or 5G, and adjust real-time operation information (including charge and discharge information of the energy storage subsystem, power generation amount of the renewable energy subsystem, power consumption duration and power consumption of the load subsystem, and the like) of each subsystem.

Furthermore, the adjusted real-time operation information of each subsystem can be collected through a sensor, and the collected real-time operation information is fed back to the controller.

It should be noted that, if the data volume of the collected real-time operation information is large, the real-time operation information may be compressed (for example, edge feature calculation) and the like, and then fed back to the controller.

The embodiment of the application can identify and eliminate the transient data of each subsystem according to the real-time data of each subsystem, overcomes the influence of the transient process on the comprehensive energy system, and further realizes the optimal control of the comprehensive energy system through the steady-state data of each subsystem, the control model of each subsystem and the control model of the comprehensive energy system.

Moreover, the optimization control method provided by the embodiment of the application has universality, and can perform continuous identification of digital twins and edge cloud cooperative optimization control on the comprehensive energy system in a plurality of different areas.

On the basis of the optimization control method provided by the embodiment of the application, the utilization efficiency of social energy can be further improved, the operating cost of a comprehensive energy system is reduced, the large-scale development of renewable energy is promoted, and energy conservation and emission reduction are realized.

The optimization control method provided by the embodiment of the present application is described below by taking a user intelligent light storage system (i.e., a comprehensive energy system including a photovoltaic subsystem, an energy storage subsystem, and a load subsystem (such as a heat pump air conditioning subsystem, a charging pile subsystem, a lighting subsystem, and an elevator subsystem, etc.), and as shown in fig. 3, the method can be implemented according to the following steps:

step S201: according to the real-time data of each subsystem in the user intelligent optical storage system, transient data of each subsystem is identified and eliminated, and steady-state data of each subsystem is obtained.

The real-time data of each subsystem can comprise real-time data of the photovoltaic subsystem, real-time data of the energy storage subsystem and real-time data of the load subsystem.

Wherein the transient data of the photovoltaic subsystem comprises power generation data of the photovoltaic subsystem when the solar irradiance is less than a preset solar irradiance threshold and power generation data of the photovoltaic subsystem when the solar irradiance is greater than or equal to the preset solar irradiance threshold (such as 100W/m)²Etc.) of the power generation data.

Thus, the transient data of each subsystem may include data that the energy storage subsystem is in a dormant state, load data that the power fluctuation value of the load subsystem is smaller than a preset power fluctuation threshold, and power generation data of the photovoltaic subsystem when the solar irradiance is smaller than the preset solar irradiance threshold.

Similarly, the steady-state data of each subsystem comprises data of the energy storage subsystem in a working state, load data of the load subsystem with a power fluctuation value larger than or equal to a preset power fluctuation threshold value, and power generation data of the photovoltaic subsystem with solar irradiance larger than or equal to a preset solar irradiance threshold value.

In one possible implementation, the spectral feature of the real-time data of each subsystem may be extracted by using a spectral feature analysis method. And identifying the transient data of each subsystem according to the frequency spectrum characteristics to obtain the transient data of each subsystem. And eliminating the transient data of each subsystem from the real-time data of each subsystem to obtain the steady-state data of each subsystem.

Step S202: and identifying the operation condition of each subsystem by adopting a Gaussian mixture model.

It should be noted that the operation conditions of each subsystem in the household intelligent optical storage system are all the operation conditions of the user.

Step S203: and constructing an identifiable model family of each subsystem according to the user operating condition of each subsystem.

Optionally, according to the operation condition of the user, constructing an identifiable model family of the load subsystem and an identifiable model family of the photovoltaic subsystem. It should be noted that there is no need to construct a family of identifiable models for the energy storage subsystem.

Further, due to the uncertainty of the load subsystem over the week (i.e., weekday) and weekend, a family of identifiable models of the load subsystem may be constructed for the week and weekend, respectively.

For example, a recognizable model family for a load subsystem may include an ARIMA model, an RNN model, and an LSTM model) for a week.

Similarly, the recognizable model family of load subsystems may also include an ARIMA model, an RNN model, and an LSTM model for weekends).

Optionally, the recognizable model family of the photovoltaic subsystem includes a first polynomial nonlinear model, a second polynomial nonlinear model, and a third polynomial nonlinear model.

The first polynomial nonlinear model can be constructed by taking the horizontal irradiance and the air temperature as input, according to the corresponding relation (such as a second-order relation) of the power generation amount of the photovoltaic subsystem and the horizontal irradiance and the air temperature, and taking the power generation amount of the photovoltaic subsystem as output. It can be seen that the first polynomial nonlinear model is a binary quadratic polynomial nonlinear model. It can be seen that the first polynomial nonlinear model can be a binary quadratic nonlinear model.

Similarly, the second polynomial nonlinear model may be constructed with the oblique irradiance and the air temperature as inputs, according to a corresponding relationship (e.g., a second order relationship) of the power generation of the photovoltaic subsystem to the oblique irradiance and the air temperature, and with the power generation of the photovoltaic subsystem as an output. It can be seen that the second polynomial nonlinear model can also be a binary quadratic nonlinear model.

Step S204: and identifying the identifiable model family of each subsystem by adopting an interactive multi-model to obtain the control model of each subsystem.

It should be noted that other manners may also be adopted to identify the recognizable model families of each subsystem, and the embodiment of the present application takes an interactive multi-model as an example for description.

Step S205: and inputting the steady-state data of each subsystem into the control model of each subsystem for solving, and acquiring the output information of each subsystem.

Optionally, a time interval granularity (e.g., 1 hour) may be set, steady-state data of each subsystem is input into the control model of each subsystem, and the control model of each subsystem is solved by using an unconstrained optimization method to obtain the power generation amount of the photovoltaic subsystem and the power consumption amount of the load.

It should be noted that, in the embodiment of the present application, a gradient descent method is used to solve the control models of the subsystems, and other unconstrained optimization methods may also be used to solve the control models of the subsystems, which is not limited in the embodiment of the present application.

Step S206: and constructing a control model of the household intelligent light storage system.

Optionally, the main purpose of the customer installation of the photovoltaic subsystem and the energy storage subsystem is to reduce the electricity charges that the customer needs to pay to the grid. Because the user carries out time-of-use (TOU) or real-time electricity price (RTP), the transfer of a load peak value can be effectively realized by utilizing the energy storage subsystem, the generated energy of the photovoltaic subsystem is consumed, and the total cost of the household intelligent light storage system is reduced.

Thus, an objective function can be constructed with the goal of minimizing the total cost of the user intelligent optical storage system, and the objective function can be expressed by the following formula:

min C_Total＝C_{bill_grid_buy}+C_{BES_depreciation}-C_{PV_subsidy}-E_{income_grid_sale}-C_{demand_reward}

wherein, C_TotalRepresenting the total cost of the household intelligent light storage system, C_{bill_grid_buy}Indicating the electricity purchase rate paid by the user to the grid, C_{BES_depreciation}Representing the life-depreciation cost, C, of the energy storage subsystem_{PV_subsidy}Representing a patch obtained by the photovoltaic subsystem, E_{income_grid_sale}Representing the on-line electricity sales revenue of the photovoltaic subsystem, C_{demand_reward}The incentive obtained for the user to participate in the demand response.

Further, the electricity purchasing fee C paid by the user to the power grid_{bill_grid_buy}Depending on the actual amount of electricity purchased by the user, it can be represented by the following formula:

wherein TOU (t) represents the time-of-use electricity price (unit may be yuan/kWh) in time period t, P_Load(t) represents the power of the load subsystem during time t, P_PV(t) represents the generated power of the photovoltaic subsystem in the period of t, P_{BES_charge}(t) represents the charging power of the energy storage subsystem during the period t, P_{BES_discharge}(t) represents the discharge power stored during the period t, and Δ t represents the period length (in min).

Life reduction cost C of energy storage subsystem_{BES_depreciation}The energy storage subsystem can be obtained according to the service life, the charging depth, the discharging depth, the charging multiplying power, the discharging multiplying power, the ambient temperature and the like of the energy storage subsystem.

The household photovoltaic subsystem can obtain subsidy and demand response income, and the subsidy C obtained by the photovoltaic subsystem_{PV_subsidy}The power generation amount of the photovoltaic subsystem can be obtained.

Further, a patch C can be obtained for the photovoltaic subsystem_{PV_subsidy}Can be represented by the following formula:

in the formula, ρ_{PV_subsidy}And the electricity subsidy price of the photovoltaic subsystem is expressed in units of element/kWh.

Photovoltaic subsystem's income E of selling electricity on surfing the net_{income_grid_sale}Can be represented by the following formula:

in the formula, sale (t) represents the electricity selling price of the user in t period, and the unit is yuan/kWh.

Incentives C obtained by user engagement with demand response_{demand_reward}Can be represented by the following formula:

in the formula, ρ_{demand_reward}A coefficient representing user participation in demand response incentives, in units of dollars per kW; Δ P (t) represents the power of the engaged user participating in the demand response.

Further, constraint conditions of the user intelligent light storage system can be constructed according to the objective function.

Optionally, the constraints may include constraints of the energy storage subsystem (i.e., the charging and discharging constraints above), constraints of the photovoltaic subsystem, grid constraints, and demand response constraints.

Because the load subsystem and the photovoltaic power generation subsystem have obvious daily cycle characteristics, the scheduling cycle of the energy storage subsystem can be defined as one day, and the energy storage subsystem ends at the time t within the day_endState of charge (available SOC (t)_end) Expressed) needs to return to an initial state of charge value (which may be expressed as SOC (0), e.g., 0.5), i.e., SOC (t)_end) And the SOC (0) is 0.5 so as to ensure the continuous operation of the energy storage subsystem.

The charging efficiency of the energy storage subsystem can be assumed (eta may be used)_chargeExpressed) and discharge efficiency (which can be expressed as η)_dischargeExpressed) is constant during the operation of the energy storage subsystem, and thus, to ensure safe operation of the energy storage subsystem, the constraints of the energy storage subsystem may be expressed by the following equation:

in the formula, C_{BES_capacity}Representing the energy storage capacity of the energy storage subsystem, SOC (t-1) representing the state of charge of the energy storage subsystem in a period of t-1, SOC_minIndicating presetsLower limit of state of charge, SOC_maxRepresenting a preset upper state of charge limit.

Further, charging power P of the energy storage subsystem in the period of t_{BES_charge}(t) can be greater than or equal to 0 and less than or equal to the rated power (P can be used) of the energy storage inverter in the energy storage subsystem_PCSExpressed), i.e., satisfies 0. ltoreq. P_{BES_charge}(t)≤P_PCS。

Similarly, the discharge power P for storing energy in the period of t_{BES_discharge}(t) can be greater than or equal to 0 and less than or equal to the rated power of the energy storage inverter, namely, P is more than or equal to 0_{BES_discharge}(t)≤P_PCS。

The redundant power output by the photovoltaic subsystem (namely the power of the photovoltaic subsystem except the power consumed by the load subsystem in the generated power) cannot be reversely transmitted to the power grid, and the output power of the photovoltaic subsystem is also considered as a schedulable variable by considering that the photovoltaic inverter in the photovoltaic subsystem has the power regulation capability, so that the output power of the photovoltaic subsystem can be regulated on the basis of the maximum power tracking power. The constraints of the photovoltaic subsystem can then be expressed by:

0≤P_PV(t)≤P_{PV_max}(t)

in the formula, P_{PV_max}(t) represents the maximum output of the photovoltaic subsystem in kW during the period t.

For safe power utilization of the user, the grid constraint may be used to indicate that the power utilization of the user is within the range of the power utilization provided by the grid and does not exceed the power utilization agreed by the user and the grid, and may be represented by the following formula:

-P_D-(t)≤P_Load(t)-P_PV(t)+P_{BES_charge}(t)-P_{BES_discharge}(t)≤P_D+(t)

in the formula, P_Load(t)-P_PV(t)+P_{BES_charge}(t)-P_{BES_discharge}(t) represents the power consumption of the user, P_Load(t) represents the load power of the user, P_{PV_max}(t) represents the maximum output of the photovoltaic subsystem at time t, P_{BES_charge}(t) represents the charging power of the energy storage subsystem during the period t, P_{BES_discharge}(t) represents the discharge power of the energy storage subsystem during the period t, P_D+(t) represents the maximum forward power capacity of the consumer (maximum power purchased by the consumer from the grid), P_D-(t) represents the maximum reverse power capacity of the user (i.e., the maximum power sold by the user to the grid).

Load shedding may be performed downward on a load basis, taking into account user participation in demand response. The load baseline may be a difference between the power usage of the load (related to the power usage of the load) and the power generation of the photovoltaic subsystem (related to the power generation of the photovoltaic subsystem). Then, the demand response constraint may be represented by:

in the formula,. DELTA.P_cut(t) represents the minimum curtailed power of the user,

representing the output information of the load subsystem (i.e. the above electrical power usage of the load) obtained by solving the control model of the load subsystem,

the output information of the photovoltaic subsystem (i.e., the above generated power of the photovoltaic subsystem) obtained by solving the control model of the photovoltaic subsystem is represented.

In a possible implementation manner, the objective function and the constraint condition may be combined to obtain a control model of the user intelligent optical storage system.

Step S207: and inputting the output information of each subsystem into the control model of the household intelligent optical storage system to solve, and acquiring the control strategy of the household intelligent optical storage system.

Optionally, the power generation amount of the photovoltaic subsystem and the power consumption of the load can be input into the control model of the household intelligent light storage system, the control model of the household intelligent light storage system is solved by a constrained optimization method, and a control strategy of the household intelligent light storage system is obtained.

For example, the control strategy of the household intelligent light storage system may include the power generation amount of the photovoltaic subsystem within a preset time range, the charging and discharging information of the energy storage subsystem, and the power utilization information of the load subsystem.

Further, the charging and discharging information of the energy storage subsystem may include charging information and discharging information of the energy storage subsystem. The charging information may include a charging duration and a charging capacity of the energy storage subsystem, and the discharging information may include a discharging duration and a discharging capacity of the energy storage subsystem.

Step S208: the control strategy of the user intelligent light storage system is sent to each subsystem (by the control device of the user intelligent light storage system), and the real-time operation information of each subsystem is adjusted.

Step S209: and acquiring the adjusted real-time operation information of each subsystem, and feeding back the real-time operation information (to a control device of the intelligent light storage system for a user).

In one possible implementation, as shown in fig. 4, the user may have roll optimization control over the smart light store system at a granularity of two hours. Namely, the output information of each subsystem in the day is updated according to the output information of each subsystem in the day. In FIG. 4, day-1 indicates day ahead, day indicates day in, and day +1 indicates day after (i.e., the next day). Meanwhile, the real-time operation information of the energy storage subsystem, the photovoltaic subsystem and the load subsystem can be adjusted according to the control strategy sent by the control device.

The embodiment of the application also provides an integrated energy system, as shown in fig. 5. The integrated energy system 1 may include a controller 11 and a plurality of subsystems 12, and the controller 11 may be connected to each of the plurality of subsystems 12.

Optionally, the plurality of subsystems may include at least two of a renewable energy subsystem, which may include a wind electronic system and/or a photovoltaic subsystem, an energy storage subsystem, and a load subsystem. In the embodiment of the present application, the multiple subsystems 12 include a wind power subsystem 121, a photovoltaic subsystem 122, an energy storage subsystem 123, and a load subsystem 124.

Thus, as shown in fig. 5, the controller 11 may be connected to the wind power subsystem 121, the photovoltaic subsystem 122, the energy storage subsystem 123, and the load subsystem 124, respectively.

Optionally, the controller 11 may be configured to:

And inputting the steady-state data of each subsystem into the control model of each subsystem for solving, and acquiring the output information of each subsystem.

And inputting the output information of each subsystem into a control model of the comprehensive energy system to solve, and acquiring a control strategy of the comprehensive energy system.

It is to be construed that the renewable energy subsystem may include a wind electronic system and/or a photovoltaic subsystem, as the integrated energy system may include at least two of the renewable energy subsystem, the energy storage subsystem, and the load subsystem. The real-time data of each subsystem may then comprise at least two of the real-time data of the renewable energy subsystem, the real-time data of the energy storage subsystem and the real-time data of the load subsystem.

The transient data of each subsystem may then comprise at least two of data of the energy storage subsystem in a dormant state, load data of the load subsystem with a power fluctuation value smaller than a preset power fluctuation threshold, and transient data of the renewable energy subsystem (as described above).

The embodiment of the application takes the control strategy of the comprehensive energy system including the power generation amount of the wind power subsystem in the preset time range, the power generation amount of the photovoltaic subsystem in the preset time range, the charging and discharging information of the energy storage subsystem and the power utilization information of the load subsystem as examples for explanation.

Therefore, the wind power subsystem can operate according to the generated energy of the wind power subsystem in a preset time range, the photovoltaic subsystem can operate according to the generated energy of the photovoltaic subsystem in the preset time range, the energy storage subsystem can operate according to the charging and discharging information of the energy storage subsystem, and the load subsystem can operate according to the power utilization information of the load subsystem.

In one possible implementation, the controller 11 may be configured to: and extracting the frequency spectrum characteristics of the real-time data of each subsystem by adopting a frequency spectrum characteristic analysis method. And identifying the transient data of each subsystem according to the frequency spectrum characteristics to obtain the transient data of each subsystem. And eliminating the transient data of each subsystem from the real-time data of each subsystem to obtain the steady-state data of each subsystem.

The controller in the embodiment of the application identifies and eliminates the transient data of each subsystem according to the real-time data of each subsystem, overcomes the influence of the transient process on the comprehensive energy system, ensures the identification robustness of the identifiable model family, and provides a basis for the optimal control of the comprehensive energy system.

In a possible implementation manner, the controller 11 may identify the identifiable model family of each subsystem by using an interactive multi-model, so as to obtain the control model of each subsystem.

In an example, the controller 11 may identify the operation condition of each subsystem by using a gaussian mixture model, and construct a recognizable model family of each subsystem according to the operation condition of each subsystem.

Alternatively, the controller 11 may identify the operation condition of each subsystem according to the actual condition of each subsystem by using a gaussian mixture model.

Since the recognizable model family of each subsystem is closely related to the operation condition, the controller 11 needs to construct the recognizable model family of each subsystem according to the operation condition.

Still further, the recognizable model family of the load subsystem may include at least two of a differential integrated moving average autoregressive (ARIMA) model (referred to as ARIMA model for short), a Recurrent Neural Network (RNN) model (referred to as RNN model for short), and a long short term memory network (LSTM) model (referred to as LSTM model for short).

Therefore, the controller 11 may construct the first polynomial nonlinear model based on the correspondence (e.g., second order relationship) between the power generation of the photovoltaic subsystem and the horizontal irradiance and the air temperature with the horizontal irradiance and the air temperature as inputs and with the power generation of the photovoltaic subsystem as an output. It can be seen that the first polynomial nonlinear model is a binary quadratic polynomial nonlinear model. It can be seen that the first polynomial nonlinear model can be a binary quadratic nonlinear model.

The controller 11 may construct a second polynomial nonlinear model based on the corresponding relationship (e.g., a second order relationship) between the power generation of the photovoltaic subsystem and the oblique irradiance and the air temperature, with the oblique irradiance and the air temperature as inputs, and with the power generation of the photovoltaic subsystem as an output. It can be seen that the second polynomial nonlinear model can also be a binary quadratic nonlinear model.

The controller 11 may use the horizontal irradiance, the oblique irradiance, and the air temperature as inputs, construct a third polynomial nonlinear model according to a corresponding relationship (e.g., a second order relationship) between the power generation of the photovoltaic subsystem and the horizontal irradiance, the oblique irradiance, and the air temperature, and use the power generation of the photovoltaic subsystem as an output. It can be seen that the third polynomial nonlinear model can be a ternary quadratic nonlinear model.

Further, the controller 11 may input the steady-state data of each subsystem into the identified control model of each subsystem according to a preset time interval granularity (e.g., 5 minutes, 15 minutes, 30 minutes, 1 hour, 2 hours, 4 hours, 24 hours, etc.), and solve the control model of each subsystem by using an unconstrained optimization method to obtain output information of each subsystem.

The controller in the embodiment of the application can construct the control models of the subsystems under different operation conditions according to the different operation conditions of the subsystems, the reusability of the identifiable model family is achieved, and the adaptability of the comprehensive energy system under various operation condition scenes is improved.

In one possible implementation, the controller 11 may construct a control model of the integrated energy system.

In one example, the controller 11 may construct a first objective function of the integrated energy system with a goal of minimizing the total cost of the integrated energy system. The controller 11 may construct constraints of the integrated energy system based on the first objective function. The controller can construct a control model of the integrated energy system according to the first objective function and the constraint condition.

Further, the controller 11 may construct the first objective function according to the following equation:

min C_Total＝C_{bill_grid_buy}-E_{income_grid_sale}

It should be noted that, the present application is described by taking as an example that the integrated energy system includes at least two of a renewable energy subsystem (including a wind electronic system and/or a photovoltaic subsystem), an energy storage subsystem, and a load subsystem. Of course, the integrated energy system may also include a distributed power source, a gas turbine, and the like, so the controller 11 may also construct the first objective function according to the following equation:

min C_Total＝C_DG+C_Gas+C_{bill_grid_buy}-E_{income_grid_sale}

In another example, the controller 11 may construct the second objective function of the integrated energy system with the total profit margin of the integrated energy system as the objective. The controller 11 may construct constraints of the integrated energy system according to the second objective function. The controller 11 may construct a control model of the integrated energy system according to the second objective function and the constraint condition.

Further, the controller 11 may construct the second objective function according to the following equation:

maxE_Total＝E_{income_grid_sale}+E_{bill_self_sufficency}

Of course, in addition to the above two modes, the controller 11 in the embodiment of the present application may also use other modes to construct the control model of the integrated energy system on the basis of the control models of the subsystems, which is not limited in the embodiment of the present application.

It should be noted that, since the control model of the integrated energy system is constructed, the constraint condition of the integrated energy system constructed according to the first objective function by the controller 11 may be the same as the constraint condition of the integrated energy system constructed according to the second objective function, and the constraint condition of the integrated energy system will be described below.

The charge-discharge constraint may be used to indicate that a state of charge (SOC) of the energy storage subsystem is greater than or equal to a preset SOC lower limit (available SOC)_minRepresented by) and less than or equal to a preset upper state of charge (SOC may be used)_maxExpress), that is, satisfy the SOC_min≤SOC≤SOC_max。

-P_D-(t)≤P_Load(t)-P_PV(t)+P_{BES_charge}(t)-P_{BES_discharge}(t)≤P_D+(t)

In a possible implementation manner, the controller 11 may input the output information of each subsystem into the control model of the integrated energy system, and solve the control model of the integrated energy system by using a constrained optimization method to obtain a control strategy of the integrated energy system.

In a possible implementation manner, the controller 11 may send the control strategy of the integrated energy system to each subsystem by using communication manners such as RS485, MBUS, or 5G, and adjust real-time operation information (including charge and discharge information of the energy storage subsystem, power generation amount of the renewable energy subsystem, power consumption duration and power consumption amount of the load subsystem, and the like) of each subsystem.

Further, the adjusted real-time operation information of each subsystem may be collected by a sensor or the like, and the collected real-time operation information is fed back to the controller 11.

It should be noted that, if the data volume of the collected real-time operation information is large, the real-time operation information may be compressed (for example, edge feature calculation) and the like, and then fed back to the controller 11.

The controller in the embodiment of the application can identify and eliminate the transient data of each subsystem according to the real-time data of each subsystem, overcomes the influence of the transient process on the comprehensive energy system, and further realizes the optimal control of the comprehensive energy system through the steady-state data of each subsystem, the control model of each subsystem and the control model of the comprehensive energy system.

Moreover, the controller in the embodiment of the application can perform continuous identification of digital twins and optimized control of edge cloud coordination on the comprehensive energy system in a plurality of different areas.

It should be understood that, in the various embodiments of the present application, the sequence numbers of the above-mentioned processes do not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. An optimization control method of an integrated energy system is characterized by comprising the following steps:

identifying and eliminating transient data of each subsystem according to real-time data of each subsystem in the comprehensive energy system, and acquiring steady-state data of each subsystem;

inputting the steady-state data of each subsystem into a control model of each subsystem, which is acquired in advance, to solve, and acquiring output information of each subsystem;

2. The optimization control method according to claim 1, wherein the identifying and eliminating transient state data of each subsystem according to real-time data of each subsystem in the integrated energy system to obtain steady state data of each subsystem comprises:

extracting the frequency spectrum characteristics of the real-time data of each subsystem by adopting a frequency spectrum characteristic analysis method;

identifying the transient data of each subsystem according to the frequency spectrum characteristics to obtain the transient data of each subsystem;

and removing the transient data of each subsystem from the real-time data of each subsystem to obtain the steady-state data of each subsystem.

3. The optimization control method according to claim 1 or 2, wherein the subsystems comprise at least two of a renewable energy subsystem, an energy storage subsystem and a load subsystem;

the renewable energy subsystem comprises a wind electronic system and/or a photovoltaic subsystem;

the transient data of each subsystem comprises at least two items of data of the energy storage subsystem in a dormant state, load data of the load subsystem with a power fluctuation value smaller than a preset power fluctuation threshold value and transient data of the renewable energy subsystem;

the transient data of the renewable energy subsystem comprises power generation data of the wind power subsystem when the actual wind speed is smaller than a preset wind speed threshold value and/or power generation data of the photovoltaic subsystem when the solar irradiance is smaller than a preset solar irradiance threshold value.

4. The optimization control method according to claim 3, wherein the steady state data of each subsystem comprises at least two of data of the energy storage subsystem in a working state, load data of the load subsystem with a power fluctuation value larger than or equal to a preset power fluctuation threshold value, and steady state data of the renewable energy subsystem;

the steady state data of the renewable energy subsystem comprises power generation data of the wind power subsystem when the actual wind speed is larger than or equal to a preset wind speed threshold value and/or power generation data of the photovoltaic subsystem when the solar irradiance is larger than or equal to a preset solar irradiance threshold value.

5. The optimal control method according to claim 4, wherein the data that the energy storage subsystem is in the sleep state comprises an actual amount of power that the energy storage subsystem is in the sleep state;

the data of the energy storage subsystem in the working state comprise the charging electric quantity, the charging time length, the discharging electric quantity and the discharging time length of the energy storage subsystem.

6. The control method according to any one of claims 3 to 5, characterized in that the method further comprises:

identifying the operation condition of each subsystem by adopting a Gaussian mixture model;

and constructing an identifiable model family of each subsystem according to the operation condition of each subsystem.

7. The optimization control method according to claim 6, wherein the operation condition of each subsystem comprises at least two of the operation condition of the load subsystem, the operation condition of the energy storage subsystem and the operation condition of the renewable energy subsystem;

the operation condition of the renewable energy subsystem comprises the operation condition of the wind power subsystem and/or the operation condition of the photovoltaic subsystem.

8. The optimization control method according to claim 7, wherein the operation condition of the load subsystem comprises a user operation condition or an industrial and commercial operation condition;

the operation working conditions of the photovoltaic subsystem comprise household operation working conditions, industrial and commercial operation working conditions or ground power station operation working conditions.

9. The optimization control method according to any one of claims 6 to 8, wherein the distinguishable family of models comprises a distinguishable family of models of the load subsystem and a distinguishable family of models of the renewable energy power generation subsystem;

the identifiable model family of renewable energy power generation subsystems comprises an identifiable model of the wind power subsystem and/or an identifiable model of the photovoltaic subsystem.

10. The optimization control method according to claim 9, wherein the identifiable family of models of the load subsystem comprises at least two of a differential integrated moving average autoregressive model, a recurrent neural network model, and a long-term and short-term memory network model;

the recognizable model of the wind power subsystem comprises at least two items of a semi-empirical model, a vortex model and a computational fluid dynamics model;

the identifiable model of the photovoltaic subsystem comprises at least two of a first polynomial nonlinear model, a second polynomial nonlinear model, and a third polynomial nonlinear model;

the first polynomial nonlinear model is constructed by taking horizontal irradiance and air temperature as input, according to the corresponding relation between the power generation capacity of the photovoltaic subsystem and the horizontal irradiance and the air temperature, and taking the power generation capacity of the photovoltaic subsystem as output;

the second polynomial nonlinear model is constructed by taking the inclined irradiance and the air temperature as input, according to the corresponding relation of the power generation capacity of the photovoltaic subsystem and the inclined irradiance and the air temperature, and taking the power generation capacity of the photovoltaic subsystem as output;

the third polynomial nonlinear model is constructed by taking the horizontal irradiance, the inclined irradiance and the air temperature as input, according to the corresponding relation between the power generation capacity of the photovoltaic subsystem and the horizontal irradiance, the inclined irradiance and the air temperature, and taking the power generation capacity of the photovoltaic subsystem as output.

11. The optimization control method according to any one of claims 6 to 10, characterized in that the control method further includes:

and identifying a pre-constructed identifiable model family of each subsystem by adopting an interactive multi-model to obtain a control model of each subsystem.

12. The control method according to any one of claims 3 to 11, characterized in that the method further comprises:

constructing a first objective function of the comprehensive energy system by taking the lowest total cost of the comprehensive energy system as a target;

constructing a constraint condition of the comprehensive energy system according to the first objective function;

and constructing a control model of the comprehensive energy system according to the first objective function and the constraint condition.

13. The optimization control method according to any one of claims 3 to 11, characterized in that the method further comprises:

constructing a second objective function of the comprehensive energy system with the maximum total income of the comprehensive energy system as an objective;

constructing a constraint condition of the comprehensive energy system according to the second objective function;

and constructing a control model of the comprehensive energy system according to the second objective function and the constraint condition.

14. The optimization control method according to claim 12 or 13, wherein the constraints comprise the charge-discharge constraints and/or grid constraints;

the charge and discharge constraint is used for indicating that the state of charge of the energy storage subsystem is greater than or equal to a preset state of charge lower limit and less than or equal to a preset state of charge upper limit;

the power grid constraint is used for indicating that the power consumption power of the user is within the power consumption power range provided by the power grid and does not exceed the power consumption power agreed by the user and the power grid.

15. The optimization control method according to any one of claims 3 to 14, wherein the inputting steady-state data of each subsystem into a control model of each subsystem acquired in advance to solve and acquire output information of each subsystem includes:

inputting steady-state data of each subsystem into a pre-constructed control model of each subsystem according to a preset time interval granularity, and solving the control model of each subsystem by adopting an unconstrained optimization method to obtain output information of each subsystem;

the output information of each subsystem comprises the power generation amount of the renewable subsystem and the power consumption amount of the load.

16. The optimal control method according to any one of claims 3 to 15, wherein the inputting the output information of each subsystem into a pre-constructed control model of the integrated energy system for solving to obtain the control strategy of the integrated energy system comprises:

inputting the output information of each subsystem into a control model of the comprehensive energy system, and solving the control model of the comprehensive energy system by adopting a constrained optimization method to obtain a control strategy of the comprehensive energy system;

the control strategy of the comprehensive energy system comprises at least two items of generated energy of the renewable energy subsystem in a preset time range, charging and discharging information of the energy storage subsystem and power utilization information of the load subsystem.

17. The optimization control method according to claim 16, wherein the charge and discharge information of the energy storage subsystem comprises charge information and/or discharge information of the energy storage subsystem, the charge information comprises a charge duration and a charge capacity of the energy storage subsystem, and the discharge information comprises a discharge duration and a discharge capacity of the energy storage subsystem;

the power consumption information of the load subsystem comprises the power consumption duration and the power consumption of the load subsystem.

18. An integrated energy system comprising a controller and a plurality of subsystems, the controller being connected to each of the plurality of subsystems;

the controller is configured to:

inputting the output information of each subsystem into a pre-constructed control model of the comprehensive energy system for solving, and acquiring a control strategy of the comprehensive energy system;

each subsystem is configured to: and operating according to the control strategy of the comprehensive energy system.

19. The integrated energy system of claim 18, wherein the controller is configured to:

transient data of each subsystem is removed from the real-time data of each subsystem, and steady-state data of each subsystem is obtained;

each subsystem comprises at least two of a renewable energy subsystem, an energy storage subsystem and a load subsystem;

the transient data of each subsystem comprises at least two items of data of the energy storage subsystem in a dormant state, load data of the load subsystem with a power fluctuation value smaller than a preset power fluctuation threshold value and transient data of the renewable energy subsystem; the data of the energy storage subsystem in the dormant state comprises the actual electric quantity of the energy storage subsystem in the dormant state, and the transient data of the renewable energy subsystem comprises the power generation data of the wind power subsystem when the actual wind speed is smaller than a preset wind speed threshold and/or the power generation data of the photovoltaic subsystem when the solar irradiance is smaller than a preset solar irradiance threshold;

the steady state data of each subsystem comprises at least two of data of the energy storage subsystem in a working state, load data of which the power fluctuation value of the load subsystem is greater than or equal to a preset power fluctuation threshold value and steady state data of the renewable energy subsystem; the data of the energy storage subsystem in the working state comprise the charging electric quantity, the charging time length, the discharging electric quantity and the discharging time length of the energy storage subsystem, and the steady-state data of the renewable energy subsystem comprise the power generation data of the wind power subsystem when the actual wind speed is larger than or equal to the preset wind speed threshold value and/or the power generation data of the photovoltaic subsystem when the solar irradiance is larger than or equal to the preset solar irradiance threshold value.