Artificial intelligence-based regional power distribution network energy optimization scheduling method and system

Abstract

The application discloses a regional power distribution network energy optimization scheduling method and system based on artificial intelligence. The application models each distributed resource, aggregates the distributed resources to build a virtual power plant, and solves the feasible region of the virtual power plant by using a vertex enumeration method to obtain the external output power characteristics of the distributed resources. On the basis, a regional power distribution network optimization scheduling model is established based on the feasible region of the virtual power plant, and a regional power distribution network optimization scheduling is realized by using a cuckoo optimization algorithm, so that the regional power distribution network energy autonomous aggregation optimization is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of distribution networks, and in particular relates to an artificial intelligence-based regional distribution network energy optimization scheduling method and system. Background Art

[0002] With the large number of distributed resources connected to the distribution network, ensuring the matching of supply and demand and flexible operation of the distribution network has become a research focus. To solve this problem, it is necessary to aggregate distributed resources to form a virtual power plant (VPP). Through coordinated control of the VPP, the efficient utilization of distributed resources can be achieved. Currently, it is known that the probabilistic characteristics of the virtual power plant dispatch boundary (VFR) under the condition that new energy satisfies the probability distribution require the Monte Carlo method to randomly extract the operating state of new energy, calculate the virtual power plant threshold exchange power under different distributed resource operating states, and then obtain the probability distribution of the dispatch boundary. However, this method is extremely computationally intensive and requires solving a large number of optimization problems. However, related optimization scheduling methods have weak regulation capabilities when facing multi-object optimization scheduling tasks, and usually only consider active power or reactive power, without considering both active power and reactive power simultaneously, resulting in imperfect regulation capabilities. Summary of the Invention

[0003] The purpose of the present invention is to overcome the defects of the prior art. In response to the problem of distributed resource aggregation, the present invention provides a regional distribution network energy optimization scheduling method and system based on artificial intelligence. The present invention models each distributed resource, aggregates the distributed resources to construct a virtual power plant, and solves the feasible domain of the virtual power plant through the vertex enumeration method to obtain the external output power characteristics of the distributed resources. On this basis, the regional distribution network is optimized and scheduled based on the feasible domain of the virtual power plant. The two are combined to achieve regional distribution network energy optimization scheduling.

[0004] In order to achieve the expected effect, the present invention adopts the following technical solutions:

[0005] The present invention discloses an artificial intelligence-based regional distribution network energy optimization scheduling method, comprising:

[0006] Determine the objective function and constraints based on the distributed resource parameters of the regional distribution network and the virtual power plant network topology model;

[0007] Determine the optimization problem according to the objective function and the constraint conditions, and solve it by using the vertex enumeration method to obtain the feasible region of the virtual power plant;

[0008] A regional distribution network optimization scheduling model is established based on the feasible region of the virtual power plant, and the cuckoo optimization algorithm is used to solve the model to achieve optimal energy scheduling of the regional distribution network.

[0009] Furthermore, the distributed resources include energy storage resources and power supply resources. The energy storage resources include distributed energy storage and electric vehicles. The power supply resources include fully controlled power supplies and semi-controlled power supplies. The constraints of the fully controlled power supplies include power constraints, capacity constraints and ramp constraints. The semi-controlled power supplies first construct a power uncertainty model and then convert the power uncertainty model into a power deterministic constraint. The energy storage resources adopt power constraints and energy constraints.

[0010] Furthermore, the network topology model includes node and line power constraints, and node and line voltage and current safety constraints.

[0011] Furthermore, the constraints include: constraints on fully controlled power supplies; constraints on semi-controlled power supplies; constraints on energy storage resources; constraints on network topology models; and constraints on power factor angle conditional capabilities.

[0012] Furthermore, the constraint of the power factor angle on the conditional capability is expressed as:

[0013] Among them, P is active power, Q is reactive power, is the power factor angle.

[0014] Furthermore, the objective function is:

[0015] max z μ T z;

[0016] in, is the unit direction vector of the new vertex obtained by enumeration, z=[P,Q] T is the point in the feasible region; P and Q are the active power and reactive power output by the virtual power plant at the common coupling point of the distribution network, respectively; is the power factor angle.

[0017] Furthermore, the use of the virtual power plant feasible region to optimize the energy scheduling of the regional distribution network specifically includes:

[0018] Constructing an optimization scheduling model through the feasible region of the virtual power plant;

[0019] The cuckoo optimization algorithm is used to solve the optimization scheduling model to achieve regional distribution network energy optimization scheduling based on the feasible region of the virtual power plant.

[0020] Furthermore, the optimization scheduling model includes a first objective function, branch flow constraints, node power balance constraints, variable upper and lower limit constraints, and virtual power plant scheduling boundary constraints.

[0021] Furthermore, the use of the cuckoo optimization algorithm to solve the optimization scheduling model to achieve regional distribution network energy optimization scheduling based on the feasible region of the virtual power plant specifically includes:

[0022] Initialize the population;

[0023] Calculate the fitness of each sample and iterate based on the cuckoo optimization algorithm to find the optimal solution when the stopping condition is met;

[0024] Determine whether the optimal solution meets the various constraints of the optimization scheduling model. If so, output the final regional distribution network scheduling plan.

[0025] The present invention also discloses an artificial intelligence-based regional distribution network energy optimization scheduling system, comprising:

[0026] Acquisition module, used to collect distributed resource parameters and network topology models of regional distribution networks;

[0027] The optimization scheduling module is used to perform regional distribution network energy optimization scheduling according to any of the methods described above.

[0028] Compared with the prior art, the beneficial effects of the present invention are: the present invention provides a method and system for optimizing and dispatching energy of a regional distribution network based on artificial intelligence. The present invention models each distributed resource, aggregates the distributed resources to construct a virtual power plant, and solves the feasible domain of the virtual power plant through the vertex enumeration method to obtain the external output power characteristics of the distributed resources. On this basis, the regional distribution network is optimized and dispatched based on the feasible domain of the virtual power plant. The combination of the two realizes the autonomous aggregation optimization of energy of the regional distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0030] Figure 1 This is a flowchart of an artificial intelligence-based regional distribution network energy optimization scheduling method provided by an embodiment of the present invention.

[0031] Figure 2 This is a flowchart of a regional distribution network optimization scheduling model using a cuckoo optimization algorithm provided by an embodiment of the present invention.

[0032] Figure 3 This is a schematic diagram of solving the feasible domain of a virtual power plant using a vertex enumeration method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0034] See also Figures 1 to 3 The present invention discloses an artificial intelligence-based regional distribution network energy optimization scheduling method, comprising:

[0035] Step 1: Determine the objective function and constraints based on the distributed resource parameters of the regional distribution network and the virtual power plant network topology model;

[0036] In one embodiment, the distributed resources include energy storage resources and power supply resources. The energy storage resources include distributed energy storage and electric vehicles. The power supply resources include fully controlled power supply and semi-controlled power supply.

[0037] The constraints of the fully controlled power supply class include power constraints, capacity constraints and ramp constraints;

[0038] Specifically, the constraints of the fully controlled power supply class are expressed as:

[0039]

[0040] in, and are the active power and reactive power at time t, are the minimum active power and the maximum active power respectively, is the rated capacity, is the upper limit of climbing, The lower limit of climbing.

[0041] The semi-controlled power supply class first builds a power uncertainty model,

[0042] Specifically, the power uncertainty model is expressed as:

[0043]

[0044] Where, and are active power and reactive power respectively, is the maximum active power, and They are the minimum power factor angle and the maximum power factor angle respectively.

[0045] The power uncertainty model is then converted into a power deterministic constraint. Specifically, the following formula is used:

[0046]

[0047] By piecewise polynomial fitting to obtain a specific form.

[0048] The energy storage resources adopt power constraints and energy constraints.

[0049]

[0050] Where, and are the discharge power and charging power of the resource at time t, Is the charge and discharge status, 1 is discharging, 0 is charging; is the maximum charge and discharge power; E i,t is the state of charge of the resource at time t, is the charge and discharge efficiency.

[0051] In another embodiment, the network topology model includes node and line power constraints, and node and line voltage and current safety constraints.

[0052] Specifically, the formula is as follows:

[0053]

[0054]

[0055] Where, are the active power and reactive power of node i at time t respectively; are the active power and reactive power of line ij at time t; ri j 、xi j are the resistance and reactance of line ij respectively; I ij,t is the square value of the current in line ij; v i,t is the square value of the voltage at node i at time t; and v i are the upper and lower limits of the voltage squared respectively; P t and Q t are the active power and reactive power output of the virtual power plant (VPP) at PCC at time t, respectively.

[0056] Furthermore, the constraints include: constraints on fully controlled power supplies; constraints on semi-controlled power supplies; constraints on energy storage resources; constraints on network topology models; and constraints on power factor angle conditional capabilities.

[0057] In one embodiment, the power factor angle constraint on conditional capability is expressed as:

[0058] Among them, P is active power, Q is reactive power, is the power factor angle.

[0059] In another embodiment, the objective function is:

[0060]

[0061] in, is the unit direction vector of the new vertex obtained by enumeration, z=[P,Q] T is the point in the feasible region; P and Q are the active power and reactive power output by the virtual power plant at the point of common coupling (PCC) of the distribution network, respectively; is the power factor angle.

[0062] Step 2: Determine the optimization problem based on the objective function and the constraints, and solve it using the vertex enumeration method to obtain the feasible region of the virtual power plant;

[0063] Specifically, if Figure 3 The figure shows a schematic diagram of solving the feasible region of virtual power plant using vertex enumeration method.

[0064] This is the basis for subsequent energy optimization scheduling, which is only to solve the external power operation characteristics of the aggregate to further obtain the process of virtual power plant scheduling boundary.

[0065] Step three: establish a regional distribution network optimization scheduling model based on the feasible region of the virtual power plant, adopt the cuckoo optimization algorithm to solve the model, and realize the energy optimization scheduling of the regional distribution network.

[0066] In one embodiment, the optimizing scheduling of regional distribution network energy using the virtual power plant feasible region specifically includes:

[0067] Step 1: constructing an optimization scheduling model through the feasible region of the virtual power plant;

[0068] Furthermore, the optimization scheduling model includes a first objective function, branch flow constraints, node power balance constraints, variable upper and lower limit constraints, and virtual power plant scheduling boundary constraints.

[0069] Specifically, the first objective function is to minimize the power generation cost of the unit and the virtual power plant as the optimization scheduling objective function, which is expressed as:

[0070]

[0071] Where, is the active power of conventional units in the distribution network, is the active power of the virtual power plant, and the operating cost is a quadratic function, where are the quadratic, linear and constant coefficients of the operating cost function. Similarly, the operating cost of the virtual power plant is also defined as a quadratic function. are the secondary, primary and constant coefficients of the virtual power plant operating cost; are the dimensions of conventional units in the distribution network and the number of virtual power plants participating in the dispatch, which are row vectors with all values ​​1.

[0072] It is worth noting that the present invention selects branch flow constraints, node power balance constraints, variable upper and lower limit constraints, and virtual power plant scheduling boundary constraints as constraint conditions for regional distribution network optimization scheduling based on the feasible domain of virtual power plants.

[0073] Specifically, the branch power flow constraint sets a given value for the power flow of the main distribution network so that it cannot exceed the given value. The formula is as follows:

[0074]

[0075] Where, is the square sum of the voltage amplitudes at the distribution network nodes; θ m is the voltage phase angle of the distribution network node; are the active power and reactive power of the distribution network branch respectively; They are the lower limit and upper limit of the active power of each branch of the distribution network respectively.

[0076] The constraint condition for node power balance is that the sum of the output power and the sum of the input power of each node in the distribution network are the same, as shown below.

[0077]

[0078] in, are the input active power and reactive power of the main network node of the distribution network respectively; They are respectively the active power demand and reactive power demand of the main distribution network. is the reactive power of conventional units in the distribution network; They are the correlation matrices of the conventional units of the main distribution network, load demand, virtual power plant and distribution network interface nodes; is the association matrix between the main nodes and branches of the distribution network; is the ground conductivity matrix of the main distribution network; is the ground susceptance matrix of the distribution network main grid.

[0079] For the upper and lower line constraints of the variables, let the square of the voltage amplitude of each node of the distribution network be The node voltage phase angle of the main distribution network is θm , Active power of conventional units in distribution network and reactive power Set the upper and lower limits as follows:

[0080]

[0081] Finally, based on Figure 1 The feasible region of the virtual power plant obtained by the process shown in the figure determines the scheduling boundary of the virtual power plant. As a random variable, on the basis of finding its probability, the constraint modeling of the virtual power plant scheduling boundary is realized by utilizing chance constraints to deal with the uncertainty of the virtual power plant, as shown below.

[0082]

[0083] Among them, α max , α min are the confidence levels for satisfying the upper and lower bounds of virtual power plant dispatch, respectively.

[0084] In order to ensure the speed of solving the optimal scheduling, due to the formula The constraints in are nonlinear constraints, which can be converted into linear constraints using the quantile transformation method as shown below.

[0085]

[0086] in, They are obtained through the feasible region and confidence level of the virtual power plant respectively.

[0087] Step 2: Use the cuckoo optimization algorithm to solve the optimization scheduling model to achieve regional distribution network energy optimization scheduling based on the feasible region of the virtual power plant.

[0088] In one embodiment, solving the optimization scheduling model using the cuckoo optimization algorithm to achieve regional distribution network energy optimization scheduling based on the virtual power plant feasible region specifically includes:

[0089] Initialize the population;

[0090] Calculate the fitness of each sample and iterate based on the cuckoo optimization algorithm to find the optimal solution. When the stopping condition is met, output the optimal solution at this time.

[0091] Determine whether the optimal solution meets the various constraints of the optimization scheduling model. If so, output the final regional distribution network scheduling plan.

[0092] like Figure 2 As shown in Figure 2, the process of using the cuckoo optimization algorithm to solve the regional distribution network optimization scheduling model is as follows:

[0093] S1, initialize a group consisting of N nests;

[0094] S2. Create a chick for each cuckoo;

[0095] S3, calculating the fitness function of the cuckoo and its chicks;

[0096] S4. If the sample is the best, replace the cuckoo with a chick;

[0097] S5. Select the best sample;

[0098] S6. Update the cuckoo clock;

[0099] S7. Find the best nesting group;

[0100] S8, determine whether the stopping condition is met; if so, return to the best solution, otherwise return to S2;

[0101] S9. Determine whether the optimal solution satisfies the constraint conditions. If so, the process ends; otherwise, the process returns to S2.

[0102] Specifically, a comparison of the present invention's AI-based regional distribution network energy optimization and scheduling method with traditional centralized optimization is shown in Table 1. As can be seen from the table, compared with centralized optimization, the present invention fully utilizes the distributed resources' ability to regulate active and reactive power, saving costs and increasing the flexibility of distributed resources in participating in grid operations after being combined into a virtual power plant.

[0103] Table 1

[0104]

[0105] The present invention also discloses an artificial intelligence-based regional distribution network energy optimization scheduling system, comprising:

[0106] Acquisition module, used to collect distributed resource parameters and network topology models of regional distribution networks;

[0107] The optimization scheduling module is used to perform regional distribution network energy optimization scheduling according to any of the methods described above.

[0108] The system embodiment can be implemented in a one-to-one correspondence with the aforementioned method embodiment, and will not be described in detail here.

[0109] Based on the same inventive concept, the present invention also discloses an electronic device, which may include: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus. The processor can call logic instructions in the memory to execute an artificial intelligence-based regional distribution network energy optimization scheduling method, including:

[0110] Determine the objective function and constraints based on the distributed resource parameters of the regional distribution network and the virtual power plant network topology model;

[0111] Determine the optimization problem according to the objective function and the constraint conditions, and solve it by using the vertex enumeration method to obtain the feasible region of the virtual power plant;

[0112] A regional distribution network optimization scheduling model is established based on the feasible region of the virtual power plant, and the cuckoo optimization algorithm is used to solve the model to achieve optimal energy scheduling of the regional distribution network.

[0113] In addition, the logical instructions in the above-mentioned memory can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0114] On the other hand, an embodiment of the present invention further provides a computer program product, comprising a computer program stored on a non-transitory computer-readable storage medium, wherein the computer program comprises program instructions. When the program instructions are executed by a computer, the computer is capable of executing an artificial intelligence-based regional distribution network energy optimization scheduling method provided in the above-mentioned method embodiments, comprising:

[0115] Determine the objective function and constraints based on the distributed resource parameters of the regional distribution network and the virtual power plant network topology model;

[0116] Determine the optimization problem according to the objective function and the constraint conditions, and solve it by using the vertex enumeration method to obtain the feasible region of the virtual power plant;

[0117] A regional distribution network optimization scheduling model is established based on the feasible region of the virtual power plant, and the cuckoo optimization algorithm is used to solve the model to achieve optimal energy scheduling of the regional distribution network.

[0118] In another aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for optimizing and dispatching energy in a regional distribution network based on artificial intelligence provided in the above embodiments is implemented, including:

[0119] Determine the objective function and constraints based on the distributed resource parameters of the regional distribution network and the virtual power plant network topology model;

[0120] Determine the optimization problem according to the objective function and the constraint conditions, and solve it by using the vertex enumeration method to obtain the feasible region of the virtual power plant;

[0121] A regional distribution network optimization scheduling model is established based on the feasible region of the virtual power plant, and the cuckoo optimization algorithm is used to solve the model to achieve optimal energy scheduling of the regional distribution network.

[0122] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0123] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A regional distribution network energy optimization scheduling method based on artificial intelligence, characterized in that: include: Determine the objective function and constraints based on the distributed resource parameters of the regional distribution network and the virtual power plant network topology model; Determine the optimization problem according to the objective function and the constraint conditions, and solve it by using the vertex enumeration method to obtain the feasible region of the virtual power plant; Establishing a regional distribution network optimization scheduling model based on the feasible region of the virtual power plant, and using the cuckoo optimization algorithm to solve the model to achieve optimal energy scheduling of the regional distribution network; The optimization scheduling model includes a first objective function, branch flow constraints, node power balance constraints, variable upper and lower limit constraints, and virtual power plant scheduling boundary constraints. The first objective function is to minimize the power generation cost of the unit and the virtual power plant as the optimization scheduling objective function, which is expressed as: ； Where, is the active power of conventional units in the distribution network, is the active power of the virtual power plant, and the operating cost is a quadratic function, where are the quadratic, linear and constant coefficients of the operating cost function. Similarly, the operating cost of the virtual power plant is also defined as a quadratic function. are the secondary, primary and constant coefficients of the virtual power plant operating cost; are the dimensions of conventional units in the distribution network and the number of virtual power plants participating in the dispatch, which are row vectors with all values ​​1.

2. The method according to claim 1, wherein: The distributed resources include energy storage resources and power supply resources. The energy storage resources include distributed energy storage and electric vehicles. The power supply resources include fully controlled power supplies and semi-controlled power supplies. The constraints of the fully controlled power supplies include power constraints, capacity constraints and ramp constraints. The semi-controlled power supplies first construct a power uncertainty model and then convert the power uncertainty model into power deterministic constraints. The energy storage resources adopt power constraints and energy constraints.

3. The method according to claim 2, wherein: The network topology model includes node and line power constraints, and node and line voltage and current safety constraints.

4. The method according to claim 3, wherein: The constraints include: constraints on fully controlled power supplies; constraints on semi-controlled power supplies; constraints on energy storage resources; constraints on network topology models; and constraints on power factor angle conditional capabilities.

5. The method according to claim 4, wherein: The power factor angle constraint on conditional capability is expressed as: ; Among them, P is active power, Q is reactive power, is the power factor angle.

6. The method according to claim 1, wherein: The objective function is: ； in, is the unit direction vector of the new vertex obtained by enumeration, is the point in the feasible region, P and Q are the active power and reactive power output by the virtual power plant at the common coupling point of the distribution network, respectively; is the power factor angle.

7. The method according to claim 1, wherein: The use of the virtual power plant feasible region to optimize the energy scheduling of the regional distribution network specifically includes: Constructing an optimization scheduling model through the feasible region of the virtual power plant; The cuckoo optimization algorithm is used to solve the optimization scheduling model to achieve regional distribution network energy optimization scheduling based on the feasible region of the virtual power plant.

8. The method according to claim 1, wherein: The method of solving the optimization scheduling model by using the cuckoo optimization algorithm to realize the regional distribution network energy optimization scheduling based on the feasible region of the virtual power plant specifically includes: Initialize the population; Calculate the fitness of each sample and iterate based on the cuckoo optimization algorithm to find the optimal solution. When the stopping condition is met, output the optimal solution at this time. Determine whether the optimal solution meets the various constraints of the optimization scheduling model. If so, output the final regional distribution network scheduling plan.

9. An artificial intelligence-based regional distribution network energy optimization and dispatching system, characterized in that: include: Acquisition module, used to collect distributed resource parameters and network topology models of regional distribution networks; An optimization scheduling module is used to perform regional distribution network energy optimization scheduling according to the method according to any one of claims 1-8.

Images