Power distribution network multi-power supply partition collaborative optimization method, system and equipment and storage medium

Abstract

The invention belongs to the technical field of power distribution networks, and discloses a power distribution network multi-power-supply-zone collaborative optimization method, system and device and a storage medium, and the method comprises the steps: firstly obtaining power supply zone combination information based on a power supply zone tie line rule agent according to a superior main transformer fault state, and then carrying out the collaborative optimization of multiple power supply zones according to the power supply zone combination information; and updating the operation state of each power supply subarea in the power distribution network, and finally obtaining the control decision of each power supply subarea through the pre-trained power supply subarea control decision optimization agent of each power supply subarea according to the operation state of each power supply subarea, thereby realizing the collaborative optimization control of each power supply subarea, and ensuring the safe operation of the power distribution network. According to the method, advantages of scheduling experience knowledge and a data driving method can be utilized at the same time, and intelligent decisions are provided for rapid optimization of a large number of source network load storage resources in power supply partitions while power supply partition tie line on-off artificial experience and regulation and control rules are reserved; and local autonomy of the power supply subarea and mutual aid optimization of the whole power distribution network system are effectively ensured.

Description

Technical Field

[0001] The present invention belongs to the technical field of distribution network, and relates to a method, system, device and storage medium for collaborative optimization of multiple power supply partitions in a distribution network. Background Art

[0002] With the construction of new power systems, the distribution network has gradually transformed from a network that simply receives and distributes electric energy to a system that integrates and interacts with sources, grids, loads and storage. Traditional mathematical optimization algorithms face difficulties in solving complex large-scale mixed integer optimization problems. It is difficult to solve large-scale examples within an effective time and it is difficult to find the best solution. Currently, we are exploring the application of reinforcement learning in the coordinated optimization of sources, grids, loads and storage in distribution networks. Multi-agent reinforcement learning algorithms are usually used to divide the power grid and reduce the dimensions of the action / state space. However, in the optimization process of sources, grids, loads and storage, there are both continuous decision variables with finer control time granularity and discrete decision variables with coarser control time granularity. Decision-making relies entirely on data-driven reinforcement learning methods, which makes modeling and optimization difficult.

[0003] In the face of the control decision optimization of distribution networks, the two commonly used implementation methods are the mixed integer programming method for distribution network operation optimization and the source-grid-load-storage collaborative optimization method based on multi-agent reinforcement learning. Among them, the mixed integer programming method for distribution network operation optimization models the physical models of all equipment and devices in the distribution network, and uses mixed integer programming methods such as interior point method, robust optimization, probability optimization, and distributed optimization to solve continuous discrete decision variables. It is a pure mechanism knowledge-driven method. The source-grid-load-storage collaborative optimization method based on multi-agent reinforcement learning models the state-action space of all equipment and devices in the distribution network, and then uses multi-agent reinforcement learning methods to train the decision network. It is a pure data-driven method.

[0004] However, the mixed integer programming method for distribution network operation optimization relies heavily on accurate physical mechanism modeling, and has high requirements for the accuracy of equipment and line parameters in the power supply partition. It is difficult to achieve real-time solution for the high-dimensional mixed integer optimization problem brought about by the large-scale access of source, grid, load and storage resources to the distribution network, and the online decision-making speed is slow. The source, grid, load and storage collaborative optimization method based on multi-agent reinforcement learning is difficult to ensure the effectiveness and safety of decisions in scenarios with multiple decision time scales (for example, the state of the power supply partition interconnection switch in the distribution network is always open and changes slowly, and the action time scale of the interconnection switch is much larger than the action time scale of other equipment), continuous and discrete decision variables coexist, and decision safety requirements are high (for example, the state change of the interconnection switch in the power supply partition often means planned maintenance or power transfer), which in turn affects the safe operation of the power system. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and to provide a method, system, device and storage medium for collaborative optimization of multiple power supply partitions in a distribution network.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] According to a first aspect of the present invention, a method for collaborative optimization of multiple power supply partitions in a distribution network is provided, comprising: obtaining the fault status of an upper-level main transformer of each power supply partition in the distribution network, and obtaining power supply partition merging information according to the fault status of the upper-level main transformer through a power supply partition interconnection line rule intelligent agent; updating the operating status of each power supply partition in the distribution network according to the power supply partition merging information; and obtaining control decisions of each power supply partition according to the operating status of each power supply partition through a power supply partition control decision optimization intelligent agent pre-trained in each power supply partition.

[0008] Optionally, the power supply partition merging information obtained through the power supply partition interconnection line rule intelligent body according to the fault status of the upper main transformer includes: the power supply partition merging information of the power supply partition where the fault status of the upper main transformer is a fault alarm is merged with each other, and the power supply partition merging information of the power supply partition where the fault status of the upper main transformer is no fault alarm is kept independent.

[0009] Optionally, updating the operating status of each power supply partition in the distribution network according to the power supply partition merging information includes: obtaining the operating status of the distribution network in the previous period; calculating the distribution network power flow distribution according to the operating status of the distribution network in the previous period and the power supply partition merging information; and updating the operating status of each power supply partition in the distribution network according to the distribution network power flow distribution.

[0010] Optionally, the power supply partition control decision optimization agent includes: physical model constraints, agent state space, agent action space and reward function; wherein the physical model constraints include: gas turbine output constraints, energy storage system constraints, medium voltage flexible control device constraints, renewable energy constraints, distribution network flow balance constraints, distribution network node voltage constraints and distribution network current carrying capacity constraints.

[0011] Optionally, the gas turbine output constraint is:

[0012]

[0013] in, is the minimum active power of the jth gas turbine; is the active power of the jth gas turbine in period t; is the maximum active power of the jth gas turbine; Gas turbine assembly for power supply zone i; is the maximum downward climbing power of the jth gas turbine; is the active power of the jth gas turbine in period t-1; is the maximum upward climbing power of the jth gas turbine; is the reactive power of the jth gas turbine in period t; is the maximum apparent power of the jth gas turbine.

[0014] The energy storage system constraints are:

[0015]

[0016] soc j,min ≤soc j,t ≤soc j,max

[0017] in, is the minimum value of the charging and discharging active power of the jth energy storage system; is the charging and discharging active power of the j-th energy storage system in period t; is the maximum value of the charging and discharging active power of the jth energy storage system; For power supply partition i energy storage system collection; soc j,t is the charge of the j-th energy storage system in period t; soc j,t-1 is the charge of the j-th energy storage system in period t-1; is the charging efficiency of the jth energy storage system; is the charging power of the jth energy storage system in period t-1; is the maximum storage capacity of the jth energy storage system; Δt is the time difference between period t and period t-1; is the discharge power of the jth energy storage system in period t-1; is the discharge efficiency of the jth energy storage system; soc j,min is the minimum charge value of the j-th energy storage system; soc j,t is the charge of the j-th energy storage system in period t; soc j,max is the maximum charge value of the jth energy storage system.

[0018] The constraints of the medium voltage soft control device are:

[0019]

[0020] in, is the active power transmitted by the port on the node i of the medium voltage flexible control device during period t; is the active power transmitted by the port on the node j of the medium voltage flexible control device during period t; is the reactive power provided by the port on the node i of the medium voltage flexible control device during period t, is the reactive power provided by the port at the node j of the medium voltage flexible control device during period t, and the positive directions of the power flow are respectively emitted from the node i of the medium voltage flexible control device and emitted from the node j of the medium voltage flexible control device; S i,max is the maximum capacity of the converter at the node i side of the medium voltage flexible control device; S j,max is the maximum capacity of the converter on the node j side of the medium voltage flexible control device.

[0021] The renewable energy constraint is:

[0022]

[0023] in, is the active power of the jth renewable energy source in period t; is the active power of the jth renewable energy source in period t in the maximum power point tracking mode; is the reactive power of the jth renewable energy source in period t; To supply electricity to the district i renewable energy collection; is the maximum apparent power of the jth renewable energy source.

[0024] The power flow balance constraint of the distribution network is:

[0025]

[0026] Among them, P k,t is the net active power injected at node k in the distribution network during period t; Ω(k) is the set of nodes connected to node k; V k,t is the voltage amplitude at node k during period t; V l,t is the voltage amplitude at node l during period t; G kl is the conductance of the branch between node k and node l; θ kl,t is the phase angle difference between the branch between node k and node l; B kl is the susceptance of the branch between node k and node l; is the active power of the kth renewable energy source in period t; is the active power of the kth gas turbine in period t; is the active power discharged by the kth energy storage system during period t; is the active power of the load at node k in the distribution network during period t; Q is the total active power flowing out of the connecting line at node k in the distribution network during period t; k,t is the net injected reactive power at node k in the distribution network during period t; is the reactive power of the kth renewable energy source in period t; is the reactive power of the kth gas turbine in period t; is the reactive power of the kth static VAR compensation device in period t; is the reactive power of the kth capacitor in period t; is the reactive power of the load at node k in the distribution network during period t; It is the total reactive power flowing out of the connecting line at node k in the distribution network during period t.

[0027] The voltage constraint of the distribution network node is:

[0028] V min ≤V k,t ≤V max

[0029] Among them, V min is the lower limit of the node voltage constraint; V max is the upper limit of node voltage constraint.

[0030] The current carrying capacity constraint of the distribution network is:

[0031] |I kl,t |≤I kl,max

[0032] Among them, I kl,t I is the current flowing through the branch between node k and node l during period t; kl,max is the upper limit of the branch current between node k and node l.

[0033] Agent state space for:

[0034]

[0035] in, is the power of the jth renewable energy source in period t; is the power of the jth load in period t; is the observable space of the jth gas turbine in period t, is the reactive power of the jth gas turbine in period t-1; is the observable space of the j-th energy storage system in period t, is the charging and discharging active power of the jth energy storage system in period t-1; is the observable space of the j-th medium voltage flexible control device in period t,

[0036] Agent action space for:

[0037]

[0038] in, is the reactive power value of renewable energy in power supply zone i during period t; is the active power output of the gas turbine in power supply zone i during period t; is the reactive power output of the gas turbine in power supply zone i during period t; is the charging and discharging power of the energy storage system in power supply partition i during period t.

[0039] The reward function is:

[0040]

[0041] Among them, R i,T is the discount reward return of power supply partition i within the scheduling period T; γ is the discount coefficient, 0<γ<1; r i,t is the single-step reward value of power supply partition i during period t; N is the number of gas turbines in power supply partition i; is the operating cost of the j-th gas turbine in power supply zone i during period t; M is the number of energy storage systems in power supply zone i; is the operating cost of the j-th energy storage system in power supply partition i during period t; F t penalty For penalty items; is the voltage constraint violation penalty function of the distribution network during period t; ψ 1 is the first weight; is the penalty function for the branch power flow constraint exceeding the limit in the distribution network during period t; W is the total number of nodes in the distribution network; V i,t is the voltage amplitude at node i during period t; V is the total number of distribution network branches; ψ 2 is the second weight; I j,t is the j-th branch current in period t; I j,max is the upper limit of the j-th branch current.

[0042] According to a second aspect of the present invention, a collaborative optimization system for multiple power supply partitions in a distribution network is provided, comprising: a partition merging module, which is used to obtain the fault status of the upper-level main transformer of each power supply partition in the distribution network, and obtain power supply partition merging information according to the fault status of the upper-level main transformer through the power supply partition interconnection line rule intelligent agent; an operation update module, which is used to update the operating status of each power supply partition in the distribution network according to the power supply partition merging information; and a decision optimization module, which is used to obtain the control decision of each power supply partition according to the operating status of each power supply partition through the power supply partition control decision optimization intelligent agent pre-trained in each power supply partition.

[0043] Optionally, the power supply partition merging information obtained through the power supply partition interconnection line rule intelligent body according to the fault status of the upper main transformer includes: the power supply partition merging information of the power supply partition where the fault status of the upper main transformer is a fault alarm is merged with each other, and the power supply partition merging information of the power supply partition where the fault status of the upper main transformer is no fault alarm is kept independent.

[0044] Optionally, updating the operating status of each power supply partition in the distribution network according to the power supply partition merging information includes: obtaining the operating status of the distribution network in the previous period; calculating the distribution network power flow distribution according to the operating status of the distribution network in the previous period and the power supply partition merging information; and updating the operating status of each power supply partition in the distribution network according to the distribution network power flow distribution.

[0045] Optionally, the power supply partition control decision optimization agent includes: physical model constraints, agent state space, agent action space and reward function; wherein the physical model constraints include: gas turbine output constraints, energy storage system constraints, medium voltage flexible control device constraints, renewable energy constraints, distribution network flow balance constraints, distribution network node voltage constraints and distribution network current carrying capacity constraints.

[0046] Optionally, the gas turbine output constraint is:

[0047]

[0048]

[0049] in, is the minimum active power of the jth gas turbine; is the active power of the jth gas turbine in period t; is the maximum active power of the jth gas turbine; Gas turbine assembly for power supply zone i; is the maximum downward climbing power of the jth gas turbine; is the active power of the jth gas turbine in period t-1; is the maximum upward climbing power of the jth gas turbine; is the reactive power of the jth gas turbine in period t; is the maximum apparent power of the jth gas turbine.

[0050] The energy storage system constraints are:

[0051]

[0052] soc j,min ≤soc j,t ≤soc j,max

[0053] in, is the minimum value of the charging and discharging active power of the jth energy storage system; is the charging and discharging active power of the j-th energy storage system in period t; is the maximum value of the charging and discharging active power of the jth energy storage system; For power supply partition i energy storage system collection; soc j,tis the charge of the j-th energy storage system in period t; soc j,t-1 is the charge of the j-th energy storage system in period t-1; is the charging efficiency of the jth energy storage system; is the charging power of the jth energy storage system in period t-1; is the maximum storage capacity of the jth energy storage system; Δt is the time difference between period t and period t-1; is the discharge power of the jth energy storage system in period t-1; is the discharge efficiency of the jth energy storage system; soc j,min is the minimum charge value of the j-th energy storage system; soc j,t is the charge of the j-th energy storage system in period t; soc j,max is the maximum charge value of the jth energy storage system.

[0054] The constraints of the medium voltage soft control device are:

[0055]

[0056] in, is the active power transmitted by the port on the node i of the medium voltage flexible control device during period t; is the active power transmitted by the port on the node j of the medium voltage flexible control device during period t; is the reactive power provided by the port on the node i of the medium voltage flexible control device during period t, is the reactive power provided by the port at the node j of the medium voltage flexible control device during period t, and the positive directions of the power flow are respectively emitted from the node i of the medium voltage flexible control device and emitted from the node j of the medium voltage flexible control device; S i,max is the maximum capacity of the converter at the node i side of the medium voltage flexible control device; S j,max is the maximum capacity of the converter on the node j side of the medium voltage flexible control device.

[0057] The renewable energy constraint is:

[0058]

[0059] in, is the active power of the jth renewable energy source in period t; is the active power of the jth renewable energy source in period t in the maximum power point tracking mode; is the reactive power of the jth renewable energy source in period t; To supply electricity to the district i renewable energy collection; is the maximum apparent power of the jth renewable energy source.

[0060] The power flow balance constraint of the distribution network is:

[0061]

[0062] Among them, P k,t is the net active power injected at node k in the distribution network during period t; Ω(k) is the set of nodes connected to node k; V k,t is the voltage amplitude at node k during period t; V l,t is the voltage amplitude at node l during period t; G kl is the conductance of the branch between node k and node l; θ kl,t is the phase angle difference between the branch between node k and node l; B kl is the susceptance of the branch between node k and node l; is the active power of the kth renewable energy source in period t; is the active power of the kth gas turbine in period t; is the active power discharged by the kth energy storage system during period t; is the active power of the load at node k in the distribution network during period t; Q is the total active power flowing out of the connecting line at node k in the distribution network during period t; k,t is the net injected reactive power at node k in the distribution network during period t; is the reactive power of the kth renewable energy source in period t; is the reactive power of the kth gas turbine in period t; is the reactive power of the kth static VAR compensation device in period t; is the reactive power of the kth capacitor in period t; is the reactive power of the load at node k in the distribution network during period t; It is the total reactive power flowing out of the connecting line at node k in the distribution network during period t.

[0063] The voltage constraint of the distribution network node is:

[0064] V min ≤V k,t ≤V max

[0065] Among them, V min is the lower limit of the node voltage constraint; V max is the upper limit of node voltage constraint.

[0066] The current carrying capacity constraint of the distribution network is:

[0067] |I kl,t |≤I kl,max

[0068] Among them, I kl,t is the current flowing through the branch between node k and node l during period t; I kl,max is the upper limit of the branch current between node k and node l.

[0069] Agent state space for:

[0070]

[0071] in, is the power of the jth renewable energy source in period t; is the power of the jth load in period t; is the observable space of the jth gas turbine in period t, is the reactive power of the jth gas turbine in period t-1; is the observable space of the j-th energy storage system in period t, is the charging and discharging active power of the jth energy storage system in period t-1; is the observable space of the j-th medium voltage flexible control device in period t,

[0072] Agent action space for:

[0073]

[0074] in, is the reactive power value of renewable energy in power supply zone i during period t; is the active output of the gas turbine in power supply zone i during period t; is the reactive power output of the gas turbine in power supply zone i during period t; is the charging and discharging power of the energy storage system in power supply partition i during period t.

[0075] The reward function is:

[0076]

[0077] Among them, R i,T is the discount reward return of power supply partition i within the scheduling period T; γ is the discount coefficient, 0<γ<1; ri,t is the single-step reward value of power supply partition i in period t; N is the number of gas turbines in power supply partition i; is the operating cost of the j-th gas turbine in power supply zone i during period t; M is the number of energy storage systems in power supply zone i; is the operating cost of the j-th energy storage system in power supply partition i during period t; F t penalty For penalty items; is the voltage constraint violation penalty function of the distribution network during period t; ψ 1 is the first penalty weight; is the penalty function for the branch power flow constraint exceeding the limit in the distribution network during period t; W is the total number of nodes in the distribution network; V i,tis the voltage amplitude at node i during period t; V is the total number of distribution network branches; ψ 2 is the second penalty weight; I j,t is the j-th branch current in period t; I j,max is the upper limit of the j-th branch current.

[0078] According to a third aspect of the present invention, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned method for collaborative optimization of multiple power supply zones in a distribution network when executing the computer program.

[0079] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned method for collaborative optimization of multiple power supply zones in a distribution network are implemented.

[0080] Compared with the prior art, the present invention has the following beneficial effects:

[0081] The collaborative optimization method for multiple power supply partitions of the distribution network of the present invention first obtains the power supply partition merging information according to the fault status of the upper main transformer based on the power supply partition interconnection line rule agent, and then updates the operating status of each power supply partition in the distribution network based on the power supply partition merging information. Finally, according to the operating status of each power supply partition, the control decision of each power supply partition is obtained through the power supply partition control decision optimization agent pre-trained in each power supply partition, so as to realize the collaborative optimization control of each power supply partition to ensure the safe operation of the distribution network. The present invention generates decisions for slow time scale discrete control variables through the power supply partition interconnection line rule agent, and generates decisions for fast time scale continuous control variables through the power supply partition control decision optimization agent. It can simultaneously utilize the advantages of scheduling experience knowledge and data-driven methods, and while retaining the manual experience and regulation rules of the power supply partition interconnection line, it provides intelligent decisions for the rapid optimization of a large number of source, network, load and storage resources within the power supply partition, effectively ensuring the local autonomy of the power supply partition and the mutual optimization of the entire distribution network system. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 This is a flow chart of a method for collaborative optimization of multiple power supply zones in a distribution network according to an embodiment of the present invention.

[0083] Figure 2 This is a flowchart of the power supply partition control decision optimization agent pre-training according to an embodiment of the present invention.

[0084] Figure 3 This is a dispatching cycle flow chart of a method for collaborative optimization of multiple power supply zones in a distribution network according to an embodiment of the present invention.

[0085] Figure 4It is a structural block diagram of a distribution network multi-power supply partition collaborative optimization system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0086] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. 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 creative work should fall within the scope of protection of the present invention.

[0087] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0088] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0089] See also Figure 1 In one embodiment of the present invention, a method for collaborative optimization of multiple power supply partitions in a distribution network is provided, specifically a multi-agent power grid optimization decision-making method based on a hybrid drive of data and knowledge. The method controls the power supply partition control decision optimization agent through a decision network, controls the power supply partition interconnection line rule agent through rules, and uses a multi-agent decision model based on the fusion of artificial knowledge experience and reinforcement learning algorithm to solve the high-dimensional mixed integer optimization problem faced by the collaborative optimization of source, grid, load and storage, and effectively improves the efficiency of intelligent scheduling decision-making.

[0090] Specifically, the method for collaborative optimization of multiple power supply zones in a distribution network of the present invention comprises the following steps:

[0091] S1: Obtain the fault status of the upper-level main transformer of each power supply partition in the distribution network, and obtain the power supply partition merging information through the power supply partition interconnection line rule intelligent agent according to the fault status of the upper-level main transformer.

[0092] S2: Update the operating status of each power supply partition in the distribution network according to the power supply partition merging information.

[0093] S3: According to the operating status of each power supply partition, the control decision of each power supply partition is obtained through the power supply partition control decision optimization intelligent agent pre-trained in each power supply partition.

[0094] The collaborative optimization method for multiple power supply partitions of the distribution network of the present invention first obtains the power supply partition merging information according to the fault status of the upper main transformer based on the power supply partition interconnection line rule agent, and then updates the operating status of each power supply partition in the distribution network based on the power supply partition merging information. Finally, according to the operating status of each power supply partition, the control decision of each power supply partition is obtained through the power supply partition control decision optimization agent pre-trained in each power supply partition, so as to realize the collaborative optimization control of each power supply partition to ensure the safe operation of the distribution network. The present invention generates decisions for slow time scale discrete control variables through the power supply partition interconnection line rule agent, and generates decisions for fast time scale continuous control variables through the power supply partition control decision optimization agent. It can simultaneously utilize the advantages of scheduling experience knowledge and data-driven methods, and while retaining the manual experience and regulation rules of the power supply partition interconnection line, it provides intelligent decisions for the rapid optimization of a large number of source, network, load and storage resources within the power supply partition, effectively ensuring the local autonomy of the power supply partition and the mutual optimization of the entire distribution network system.

[0095] Explanatory, under normal circumstances, the power supply partition tie lines are disconnected, and only when the failure of important equipment affects the reliability of power supply will they be closed. Therefore, the action frequency of tie lines is much lower than that of other equipment, and it is not suitable to be constructed by data-driven modeling. Among them, tie lines refer to power lines used to connect different power sources or feeders in the distribution network, which are usually used to improve the flexibility and reliability of the power grid. Through tie lines, different power sources or lines can achieve mutual support and load transfer, ensuring that power supply can be maintained in the event of a failure or maintenance, and optimizing the stability and resource utilization efficiency of the distribution network.

[0096] In this implementation manner, the power supply partition interconnection line rule agent adopts a rule agent construction method, and the action of the power supply partition interconnection line rule agent does not consider the purpose of power transfer and load reversal, but only considers the fault situation of the upper main transformer.

[0097] In a possible implementation manner, the power supply partition merging information obtained through the power supply partition interconnection line rule intelligent body according to the fault status of the upper main transformer includes: the power supply partition merging information of the power supply partition where the fault status of the upper main transformer is a fault alarm is merged with each other, and the power supply partition merging information of the power supply partition where the fault status of the upper main transformer is no fault alarm is kept independent.

[0098] Explanatory, it can be divided into two specific situations for analysis: 1. Compared with period t-1, time period t has no fault alarm information of the upper main transformers of power supply partition i and power supply partition j, and the interconnection line does not operate and maintains the disconnected state; 2. Compared with time period t-1, time period t has fault alarm information of the upper main transformers of power supply partition i and power supply partition j, and the interconnection line operates and turns to a closed state, so that the two partitions are combined for power supply, thereby improving power supply reliability.

[0099] Explanatory, the power supply partition tie line rule agent receives the current upper main transformer fault status, outputs the tie line on-off decision, and judges the tie line on-off decision of time period t and the tie line on-off decision of time period t-1.

[0100] In a possible implementation, updating the operating status of each power supply partition in the distribution network according to the power supply partition merging information includes: obtaining the operating status of the distribution network in the previous period; calculating the distribution network flow distribution according to the operating status of the distribution network in the previous period and the power supply partition merging information; and updating the operating status of each power supply partition in the distribution network according to the distribution network flow distribution.

[0101] Explanatory: 1. If the tie-line on-off decision in time period t is the same as the tie-line on-off decision in time period t-1, the operating status of each power supply partition in the current distribution network is extracted and input into the power supply partition control decision optimization intelligent agent, which outputs the control decision of the power supply partition.

[0102] 2. If the tie-line on-off decision in time period t is different from the tie-line on-off decision in time period t-1, the simulation interactive environment calculates the new distribution network flow distribution according to the tie-line on-off, and then extracts the operating status of each power supply partition in the distribution network from the new distribution network flow distribution, and inputs it into the power supply partition control decision optimization intelligent body, which then outputs the control decision of the power supply partition.

[0103] Explanatory, Reinforcement Learning (RL), also known as Reinforcement Learning, Evaluation Learning or Enhanced Learning, is one of the paradigms and methodologies of machine learning. It is used to describe and solve the problem of how an agent learns strategies to maximize rewards or achieve specific goals during its interaction with the environment. Reinforcement learning is a process in which an agent learns by trial and error, and the rewards obtained by interacting with the environment guide the behavior. The goal is to maximize the reward for the agent. In reinforcement learning, the environment evaluates the quality of the agent's actions (usually a scalar signal, rather than telling the agent how to take the correct action). Since the external environment provides little information, the agent must learn through its own experience. In this way, the agent gains knowledge in an action-evaluation environment and improves its action plan to adapt to the environment.

[0104] Reinforcement learning regards learning as a trial-and-error evaluation process. The agent selects an action for the environment. After the environment accepts the action, its state changes and a reinforcement signal (reward or punishment) is generated and fed back to the agent. The agent selects the next action based on the reinforcement signal and the current state of the environment. The principle of selection is to increase the probability of receiving positive reinforcement (reward). The selected action not only affects the immediate reward value, but also affects the state of the environment at the next moment and the final reward value.

[0105] The agent action space is the set of actions output by the agent in reinforcement learning, the agent state space is the set of environment operating states input by the agent in reinforcement learning, and the reward value is an evaluation of the quality of the actions generated by the agent by the environment in reinforcement learning.

[0106] Collaborative optimization is to optimize each sub-objective function while comprehensively considering the results of other sub-objective functions, so that the optimization results between sub-objective functions can be consistent. In the power system, general optimization goals include reducing system operating costs, improving system energy efficiency, and reducing system environmental costs.

[0107] In the present invention, when constructing the power supply partition control decision optimization intelligent agent, the distribution network intelligent agent is divided according to the distribution network power supply partition, the gas turbine, energy storage system and medium voltage flexible control device are modeled according to the power supply partition composition, and the power supply partition control decision optimization intelligent agent training is carried out using a multi-agent reinforcement learning algorithm.

[0108] Explanatory, Medium Voltage Flexible Control Device is a device used in 1kV to 35kV power systems, using power electronics technology (such as IGBT or GTO) to achieve fast and flexible power regulation and control. Compared with traditional mechanical switches, flexible control has higher response speed and accuracy, and is widely used to regulate, protect and optimize the performance of medium voltage power transmission systems.

[0109] In one possible implementation, the power supply partition control decision optimization agent includes: physical model constraints, agent state space, agent action space and reward function; wherein the physical model constraints include: gas turbine output constraints, energy storage system constraints, medium voltage flexible control device constraints, renewable energy constraints, distribution network flow balance constraints, distribution network node voltage constraints and distribution network current carrying capacity constraints.

[0110] Exemplarily, the gas turbine output constraint is:

[0111]

[0112] in, is the minimum active power of the jth gas turbine; is the active power of the jth gas turbine in period t; is the maximum active power of the jth gas turbine; Gas turbine assembly for power supply zone i; is the maximum downward climbing power of the jth gas turbine; is the active power of the jth gas turbine in period t-1; is the maximum upward climbing power of the jth gas turbine; is the reactive power of the jth gas turbine in period t; is the maximum apparent power of the jth gas turbine.

[0113] Exemplarily, the energy storage system constraints are:

[0114]

[0115] soc j,min ≤soc j,t ≤soc j,max

[0116] in, is the minimum value of the charging and discharging active power of the jth energy storage system; is the charging and discharging active power of the j-th energy storage system in period t; is the maximum value of the charging and discharging active power of the jth energy storage system; For power supply partition i energy storage system collection; soc j,t is the charge of the j-th energy storage system in period t; soc j,t-1 is the charge of the j-th energy storage system in period t-1; is the charging efficiency of the jth energy storage system; is the charging power of the jth energy storage system in period t-1; is the maximum storage capacity of the jth energy storage system; Δt is the time difference between period t and period t-1; is the discharge power of the jth energy storage system in period t-1; is the discharge efficiency of the jth energy storage system; soc j,min is the minimum charge value of the j-th energy storage system; soc j,t is the charge of the j-th energy storage system in period t; soc j,max is the maximum charge value of the jth energy storage system.

[0117] Exemplarily, the medium voltage soft control device is constrained as follows:

[0118]

[0119] Among them, the medium voltage soft control device is connected between computing node i and node j. is the active power transmitted by the port on the node i of the medium voltage flexible control device during period t; is the active power transmitted by the port on the node j of the medium voltage flexible control device during period t; is the reactive power provided by the port on the node i of the medium voltage flexible control device during period t, is the reactive power provided by the port at the node j of the medium voltage flexible control device during period t, and the positive directions of the power flow are respectively emitted from the node i of the medium voltage flexible control device and emitted from the node j of the medium voltage flexible control device; S i,max is the maximum capacity of the converter at the node i side of the medium voltage flexible control device; S j,max is the maximum capacity of the converter on the node j side of the medium voltage flexible control device.

[0120] For example, to improve economic efficiency, renewable energy is generated in maximum power point tracking mode, and only reactive capacity is used for regulation. The renewable energy constraint is:

[0121]

[0122] in, is the active power of the jth renewable energy source in period t; is the active power of the jth renewable energy source in period t in the maximum power point tracking mode; is the reactive power of the jth renewable energy source in period t; To supply electricity to the district i renewable energy collection; is the maximum apparent power of the jth renewable energy source.

[0123] Exemplarily, the power flow balance constraint of the distribution network is:

[0124]

[0125] Among them, P k,t is the net active power injected at node k in the distribution network during period t; Ω(k) is the set of nodes connected to node k; V k,t is the voltage amplitude at node k during period t; V l,t is the voltage amplitude at node l during period t; G kl is the conductance of the branch between node k and node l; θ kl,t is the phase angle difference between the branch between node k and node l; B kl is the susceptance of the branch between node k and node l; is the active power of the kth renewable energy source in period t; is the active power of the kth gas turbine in period t; is the active power discharged by the kth energy storage system during period t; is the active power of the load at node k in the distribution network during period t; Q is the total active power flowing out of the connecting line at node k in the distribution network during period t; k,t is the net injected reactive power at node k in the distribution network during period t; is the reactive power of the kth renewable energy source in period t; is the reactive power of the kth gas turbine in period t; is the reactive power of the kth static VAR compensation device in period t; is the reactive power of the kth capacitor in period t; is the reactive power of the load at node k in the distribution network during period t; It is the total reactive power flowing out of the connecting line at node k in the distribution network during period t.

[0126] Exemplarily, the distribution network node voltage constraint is:

[0127] V min ≤V k,t ≤V max

[0128] Among them, V min is the lower limit of the node voltage constraint; V max is the upper limit of node voltage constraint.

[0129] Exemplarily, the current carrying capacity constraint of the distribution network is:

[0130] |I kl,t |≤I kl,max

[0131] Among them, I kl,t is the current flowing through the branch between node k and node l during period t; I kl,max is the upper limit of the branch current between node k and node l.

[0132] Exemplarily, the agent state space for:

[0133]

[0134] in, is the power of the jth renewable energy source in period t; is the power of the jth load in period t; is the observable space of the jth gas turbine in period t, is the reactive power of the jth gas turbine in period t-1; is the observable space of the j-th energy storage system in period t, is the charging and discharging active power of the jth energy storage system in period t-1; is the observable space of the j-th medium voltage flexible control device in period t,

[0135] Exemplarily, the agent action space for:

[0136]

[0137] in, is the reactive power value of renewable energy in power supply zone i during period t; is the active power output of the gas turbine in power supply zone i during period t; is the reactive power output of the gas turbine in power supply zone i during period t; is the charging and discharging power of the energy storage system in power supply partition i during period t. Among them, renewable energy generates electricity in maximum power point tracking mode, and only reactive capacity is used for regulation.

[0138] Exemplarily, the reward function is:

[0139]

[0140] Among them, R i,T is the discount reward return of power supply partition i within the scheduling period T; γ is the discount coefficient, 0<γ<1; r i,t is the single-step reward value of power supply partition i during period t; N is the number of gas turbines in power supply partition i; is the operating cost of the jth gas turbine in power supply zone i during period t; M is the number of energy storage systems in power supply zone i; is the operating cost of the j-th energy storage system in power supply partition i during period t; F t penalty For penalty items; is the voltage constraint violation penalty function of the distribution network during period t; ψ 1 is the first weight; is the penalty function for the branch power flow constraint exceeding the limit in the distribution network during period t; W is the total number of nodes in the distribution network; V i,t is the voltage amplitude at node i during period t; V is the total number of distribution network branches; ψ 2 is the second weight; I j,t is the j-th branch current in period t; I j,max is the upper limit of the j-th branch current.

[0141] For example, see Figure 2 ,The pre-training process of the power supply partition control decision optimization agent is as follows:

[0142] 1. Initialize the training environment and determine the training iteration rounds.

[0143] 2. Input the current agent state space vector of the power supply partition control decision optimization agent.

[0144] 3. Output the agent action space vector and the agent state space vector at the next moment.

[0145] 4. If the upper limit of training iterations is reached and the constraints are met, go to step 5; otherwise, go to step 6.

[0146] 5. Calculate the final reward value.

[0147] 6. Calculate the current reward value.

[0148] 7. Output power supply partition control decision optimization intelligent agent network parameters, end.

[0149] For example, see Figure 3 ,The coordinated optimization process of multiple power supply partitions in the distribution network within the dispatching period is as follows:

[0150] 1. Start by setting the scheduling period T.

[0151] 2. The power supply partition interconnection line rule agent reads the fault status of the upper-level main transformer in time period t.

[0152] 3. Output the on-off decision of the tie line in period t.

[0153] 4. If the power supply partition merging information remains unchanged, go to step 5; otherwise, go to step 6.

[0154] 5. The power supply partition control decision optimization agent reads the power supply partition status during period t.

[0155] 6. The simulation environment inputs the power supply partition merging information and calculates the distribution network status.

[0156] 7. The power supply partition control decision optimization agent reads the power supply partition status during time period t.

[0157] 8. Control decision of output power supply partition, time period t=t+1.

[0158] 9. If the current period t < scheduling period T, go to step 10; otherwise, go to step 2.

[0159] 10. End the scheduling and output the control decision sequence within the scheduling period T.

[0160] The following are device embodiments of the present invention, which can be used to implement the method embodiments of the present invention. For details not disclosed in the device embodiments, please refer to the method embodiments of the present invention.

[0161] See also Figure 4In another embodiment of the present invention, a distribution network multi-power supply partition collaborative optimization system is provided, which can be used to implement the above-mentioned distribution network multi-power supply partition collaborative optimization method. Specifically, the distribution network multi-power supply partition collaborative optimization system includes a partition merging module, an operation update module and a decision optimization module.

[0162] Among them, the partition merging module is used to obtain the fault status of the upper-level main transformer of each power supply partition in the distribution network, and obtain the power supply partition merging information according to the fault status of the upper-level main transformer through the power supply partition interconnection line rule agent; the operation update module is used to update the operation status of each power supply partition in the distribution network according to the power supply partition merging information; the decision optimization module is used to obtain the control decision of each power supply partition according to the operation status of each power supply partition through the power supply partition control decision optimization agent pre-trained in each power supply partition.

[0163] In a possible implementation manner, the power supply partition merging information obtained through the power supply partition interconnection line rule intelligent body according to the fault status of the upper main transformer includes: the power supply partition merging information of the power supply partition where the fault status of the upper main transformer is a fault alarm is merged with each other, and the power supply partition merging information of the power supply partition where the fault status of the upper main transformer is no fault alarm is kept independent.

[0164] In a possible implementation, updating the operating status of each power supply partition in the distribution network according to the power supply partition merging information includes: obtaining the operating status of the distribution network in the previous period; calculating the distribution network flow distribution according to the operating status of the distribution network in the previous period and the power supply partition merging information; and updating the operating status of each power supply partition in the distribution network according to the distribution network flow distribution.

[0165] In one possible implementation, the power supply partition control decision optimization agent includes: physical model constraints, agent state space, agent action space and reward function; wherein the physical model constraints include: gas turbine output constraints, energy storage system constraints, medium voltage flexible control device constraints, renewable energy constraints, distribution network flow balance constraints, distribution network node voltage constraints and distribution network current carrying capacity constraints.

[0166] In a possible implementation manner, the gas turbine output constraint is:

[0167]

[0168] in, is the minimum active power of the jth gas turbine; is the active power of the jth gas turbine in period t; is the maximum active power of the jth gas turbine; Gas turbine assembly for power supply zone i; is the maximum downward climbing power of the jth gas turbine; is the active power of the jth gas turbine in period t-1; is the maximum upward climbing power of the jth gas turbine; is the reactive power of the jth gas turbine in period t; is the maximum apparent power of the jth gas turbine.

[0169] The energy storage system constraints are:

[0170]

[0171] soc j,min ≤soc j,t ≤soc j,max

[0172] in, is the minimum value of the charging and discharging active power of the jth energy storage system; is the charging and discharging active power of the j-th energy storage system in period t; is the maximum value of the charging and discharging active power of the jth energy storage system; For power supply partition i energy storage system collection; soc j,t is the charge of the j-th energy storage system in period t; soc j,t-1 is the charge of the j-th energy storage system in period t-1; is the charging efficiency of the jth energy storage system; is the charging power of the jth energy storage system in period t-1; is the maximum storage capacity of the jth energy storage system; Δt is the time difference between period t and period t-1; is the discharge power of the jth energy storage system in period t-1; is the discharge efficiency of the jth energy storage system; soc j,min is the minimum charge value of the j-th energy storage system; soc j,t is the charge of the j-th energy storage system in period t; soc j,max is the maximum charge value of the jth energy storage system.

[0173] The constraints of the medium voltage soft control device are:

[0174]

[0175] in, is the active power transmitted by the port on the node i of the medium voltage flexible control device during period t; is the active power transmitted by the port on the node j of the medium voltage flexible control device during period t; is the reactive power provided by the port on the node i of the medium voltage flexible control device during period t, is the reactive power provided by the port at the node j of the medium voltage flexible control device during period t, and the positive directions of the power flow are respectively emitted from the node i of the medium voltage flexible control device and emitted from the node j of the medium voltage flexible control device; S i,max is the maximum capacity of the converter at the node i side of the medium voltage flexible control device; S j,max is the maximum capacity of the converter on the node j side of the medium voltage flexible control device.

[0176] The renewable energy constraint is:

[0177]

[0178] in, is the active power of the jth renewable energy source in period t; is the active power of the jth renewable energy source in period t in the maximum power point tracking mode; is the reactive power of the jth renewable energy source in period t; To supply electricity to the district i renewable energy collection; is the maximum apparent power of the jth renewable energy source.

[0179] The power flow balance constraint of the distribution network is:

[0180]

[0181] Among them, P k,t is the net active power injected at node k in the distribution network during period t; Ω(k) is the set of nodes connected to node k; V k,t is the voltage amplitude at node k during period t; V l,t is the voltage amplitude at node l during period t; G kl is the conductance of the branch between node k and node l; θ kl,t is the phase angle difference between the branch between node k and node l; B kl is the susceptance of the branch between node k and node l; is the active power of the kth renewable energy source in period t; is the active power of the kth gas turbine in period t; is the active power discharged by the kth energy storage system during period t; is the active power of the load at node k in the distribution network during period t; Q is the total active power flowing out of the connecting line at node k in the distribution network during period t; k,t is the net injected reactive power at node k in the distribution network during period t; is the reactive power of the kth renewable energy source in period t; is the reactive power of the kth gas turbine in period t; is the reactive power of the kth static VAR compensation device in period t; is the reactive power of the kth capacitor in period t; is the reactive power of the load at node k in the distribution network during period t; It is the total reactive power flowing out of the connecting line at node k in the distribution network during period t.

[0182] The voltage constraint of the distribution network node is:

[0183] V min ≤V k,t ≤V max

[0184] Among them, V min is the lower limit of the node voltage constraint; V max is the upper limit of node voltage constraint.

[0185] The current carrying capacity constraint of the distribution network is:

[0186] |I kl,t |≤I kl,max

[0187] Among them, I kl,t I is the current flowing through the branch between node k and node l during period t; kl,max is the upper limit of the branch current between node k and node l.

[0188] Agent state space for:

[0189]

[0190] in, is the power of the jth renewable energy source in period t; is the power of the jth load in period t; is the observable space of the jth gas turbine in period t, is the reactive power of the jth gas turbine in period t-1; is the observable space of the j-th energy storage system in period t, is the charging and discharging active power of the jth energy storage system in period t-1; is the observable space of the j-th medium voltage flexible control device in period t,

[0191] Agent action space for:

[0192]

[0193] in, is the reactive power value of renewable energy in power supply zone i during period t; is the active power output of the gas turbine in power supply zone i during period t; is the reactive power output of the gas turbine in power supply zone i during period t; is the charging and discharging power of the energy storage system in power supply partition i during period t.

[0194] The reward function is:

[0195]

[0196]

[0197] Among them, R i,T is the discount reward return of power supply partition i within the scheduling period T; γ is the discount coefficient, 0<γ<1; r i,t is the single-step reward value of power supply partition i during period t; N is the number of gas turbines in power supply partition i; is the operating cost of the j-th gas turbine in power supply zone i during period t; M is the number of energy storage systems in power supply zone i; is the operating cost of the j-th energy storage system in power supply partition i during period t; F t penalty For penalty items; is the voltage constraint violation penalty function of the distribution network during period t; ψ 1 is the first weight; is the penalty function for the branch power flow constraint exceeding the limit in the distribution network during period t; W is the total number of nodes in the distribution network; V i,t is the voltage amplitude at node i during period t; V is the total number of distribution network branches; ψ 2 is the second weight; I j,t is the j-th branch current in period t; I j,max is the upper limit of the j-th branch current.

[0198] All relevant contents of each step involved in the embodiment of the aforementioned distribution network multi-power supply partition collaborative optimization method can be referred to the functional description of the functional modules corresponding to the distribution network multi-power supply partition collaborative optimization system in the embodiment of the present invention, and will not be repeated here.

[0199] The division of modules in the embodiments of the present invention is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional module in each embodiment of the present invention may be integrated into one processor, or may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of software functional modules.

[0200] In another embodiment of the present invention, a computer device is provided, the computer device including a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, which are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions in a computer storage medium to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the collaborative optimization method of multiple power supply partitions in the distribution network.

[0201] In another embodiment of the present invention, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It can be understood that the computer-readable storage medium here can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the method for collaborative optimization of multiple power supply partitions in the distribution network in the above embodiment.

[0202] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0203] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0204] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0205] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0206] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for collaborative optimization of multiple power supply partitions in a distribution network, characterized in that: include: Obtain the fault status of the upper-level main transformer of each power supply partition in the distribution network, and obtain the power supply partition merging information through the power supply partition tie line rule agent according to the fault status of the upper-level main transformer; Update the operation status of each power supply partition in the distribution network according to the power supply partition merging information; According to the operating status of each power supply partition, the control decision of each power supply partition is obtained through the power supply partition control decision optimization intelligent agent pre-trained in each power supply partition.

2. The method for collaborative optimization of multiple power supply partitions in a distribution network according to claim 1, characterized in that: The obtaining of the power supply partition merging information through the power supply partition tie line rule agent according to the fault state of the upper main transformer includes: The power supply partition merging information of the power supply partition where the fault status of the upper main transformer is a fault alarm is merged with each other, and the power supply partition merging information of the power supply partition where the fault status of the upper main transformer is no fault alarm is kept independent.

3. The method for collaborative optimization of multiple power supply partitions in a distribution network according to claim 1, characterized in that: The updating of the operation status of each power supply partition in the distribution network according to the power supply partition merging information includes: Get the operating status of the distribution network in the previous period; Calculate the power flow distribution of the distribution network based on the operation status of the distribution network in the previous period and the combined information of the power supply partitions; According to the power flow distribution of the distribution network, the operating status of each power supply partition in the distribution network is updated.

4. The method for collaborative optimization of multiple power supply partitions in a distribution network according to claim 1, characterized in that: The power supply partition control decision optimization agent includes: physical model constraints, agent state space, agent action space and reward function; wherein the physical model constraints include: gas turbine output constraints, energy storage system constraints, medium voltage flexible control device constraints, renewable energy constraints, distribution network flow balance constraints, distribution network node voltage constraints and distribution network current carrying capacity constraints.

5. The method for collaborative optimization of multiple power supply partitions in a distribution network according to claim 4, characterized in that: The gas turbine output constraint is: in, is the minimum active power of the jth gas turbine; is the active power of the jth gas turbine in period t; is the maximum active power of the jth gas turbine; Gas turbine assembly for power supply zone i; is the maximum downward climbing power of the jth gas turbine; is the active power of the jth gas turbine in period t-1; is the maximum upward climbing power of the jth gas turbine; is the reactive power of the jth gas turbine in period t; is the maximum apparent power of the jth gas turbine; The energy storage system constraints are: in, is the minimum value of the charging and discharging active power of the jth energy storage system; is the charging and discharging active power of the j-th energy storage system in period t; is the maximum value of the charging and discharging active power of the jth energy storage system; For power supply partition i energy storage system collection; soc j,t is the charge of the j-th energy storage system in period t; soc j,t-1 is the charge of the j-th energy storage system in period t-1; is the charging efficiency of the jth energy storage system; is the charging power of the jth energy storage system in period t-1; is the maximum storage capacity of the jth energy storage system; Δt is the time difference between period t and period t-1; is the discharge power of the jth energy storage system in period t-1; is the discharge efficiency of the jth energy storage system; soc j,min is the minimum charge value of the j-th energy storage system; soc j,t is the charge of the j-th energy storage system in period t; soc j,max is the maximum charge value of the jth energy storage system; The constraints of the medium voltage soft control device are: in, is the active power transmitted by the port on the node i of the medium voltage flexible control device during period t; is the active power transmitted by the port on the node j of the medium voltage flexible control device during period t; is the reactive power provided by the port on the node i of the medium voltage flexible control device during period t, is the reactive power provided by the port at the node j of the medium voltage flexible control device during period t, and the positive directions of the power flow are respectively emitted from the node i of the medium voltage flexible control device and emitted from the node j of the medium voltage flexible control device; S i,max is the maximum capacity of the converter at the node i side of the medium voltage flexible control device; S j,max is the maximum capacity of the converter at the node j side of the medium voltage flexible control device; The renewable energy constraint is: in, is the active power of the jth renewable energy source in period t; is the active power of the jth renewable energy source in period t in the maximum power point tracking mode; is the reactive power of the jth renewable energy source in period t; To supply electricity to the district i renewable energy collection; is the maximum apparent power of the jth renewable energy source; The power flow balance constraint of the distribution network is: Among them, P k,t is the net active power injected at node k in the distribution network during period t; Ω(k) is the set of nodes connected to node k; V k,t is the voltage amplitude at node k during period t; V l,t is the voltage amplitude at node l during period t; G kl is the conductance of the branch between node k and node l; θ kl,t is the phase angle difference between the branch between node k and node l; B kl is the susceptance of the branch between node k and node l; is the active power of the kth renewable energy source in period t; is the active power of the kth gas turbine in period t; is the active power discharged by the kth energy storage system during period t; is the active power of the load at node k in the distribution network during period t; Q is the total active power flowing out of the connecting line at node k in the distribution network during period t; k,t is the net injected reactive power at node k in the distribution network during period t; is the reactive power of the kth renewable energy source in period t; is the reactive power of the kth gas turbine in period t; is the reactive power of the kth static VAR compensation device in period t; is the reactive power of the kth capacitor in period t; is the reactive power of the load at node k in the distribution network during period t; is the total reactive power flowing out of the connecting line at node k in the distribution network during period t; The voltage constraint of the distribution network node is: In min ≤V k,t ≤V max Among them, V min is the lower limit of the node voltage constraint; V max is the upper limit of node voltage constraint; The current carrying capacity constraint of the distribution network is: |I kl,t |≤I kl,max Among them, I kl,t is the current flowing through the branch between node k and node l during period t; I kl,max is the upper limit of the branch current between node k and node l; Agent state space for: in, is the power of the jth renewable energy source in period t; is the power of the jth load in period t; is the observable space of the jth gas turbine in period t, is the reactive power of the jth gas turbine in period t-1; is the observable space of the j-th energy storage system in period t, is the charging and discharging active power of the jth energy storage system in period t-1; is the observable space of the j-th medium voltage flexible control device in period t, Agent action space for: in, is the reactive power value of renewable energy in power supply zone i during period t; is the active output of the gas turbine in power supply zone i during period t; is the reactive power output of the gas turbine in power supply zone i during period t; is the charging and discharging power of the energy storage system in power supply partition i during period t; The reward function is: Among them, R i,T is the discount reward return of power supply partition i within the scheduling period T; γ is the discount coefficient, 0<γ<1; r i,t is the single-step reward value of power supply partition i during period t; N is the number of gas turbines in power supply partition i; is the operating cost of the jth gas turbine in power supply zone i during period t; M is the number of energy storage systems in power supply zone i; is the operating cost of the j-th energy storage system in power supply partition i during period t; F t penalty For penalty items; is the voltage constraint violation penalty function of the distribution network in period t; ψ1 is the first weight; is the penalty function for the branch power flow constraint exceeding the limit in the distribution network during period t; W is the total number of nodes in the distribution network; V i,t is the voltage amplitude at node i during period t; V is the total number of distribution network branches; ψ2 is the second weight; I j,t is the j-th branch current in period t; I j,max is the upper limit of the j-th branch current.

6. A distribution network multi-power supply partition collaborative optimization system, characterized in that: include: The partition merging module is used to obtain the fault status of the upper main transformer of each power supply partition in the distribution network, and obtain the power supply partition merging information through the power supply partition tie line rule intelligent body according to the fault status of the upper main transformer; An operation update module is used to update the operation status of each power supply partition in the distribution network according to the power supply partition merging information; The decision optimization module is used to obtain the control decision of each power supply partition according to the operating status of each power supply partition through the power supply partition control decision optimization intelligent agent pre-trained in each power supply partition.

7. The distribution network multi-power supply partition collaborative optimization system according to claim 6 is characterized in that: The obtaining of the power supply partition merging information through the power supply partition tie line rule agent according to the fault state of the upper main transformer includes: The power supply partition merging information of the power supply partition where the fault status of the upper main transformer is a fault alarm is merged with each other, and the power supply partition merging information of the power supply partition where the fault status of the upper main transformer is no fault alarm is kept independent.

8. The distribution network multi-power supply partition collaborative optimization system according to claim 6 is characterized in that: The updating of the operation status of each power supply partition in the distribution network according to the power supply partition merging information includes: Get the operating status of the distribution network in the previous period; Calculate the power flow distribution of the distribution network based on the operation status of the distribution network in the previous period and the combined information of the power supply partitions; According to the power flow distribution of the distribution network, the operating status of each power supply partition in the distribution network is updated.

9. The distribution network multi-power supply partition collaborative optimization system according to claim 6, characterized in that: The power supply partition control decision optimization agent includes: physical model constraints, agent state space, agent action space and reward function; wherein the physical model constraints include: gas turbine output constraints, energy storage system constraints, medium voltage flexible control device constraints, renewable energy constraints, distribution network flow balance constraints, distribution network node voltage constraints and distribution network current carrying capacity constraints.

10. The distribution network multi-power supply partition collaborative optimization system according to claim 6, characterized in that: The gas turbine output constraint is: in, is the minimum active power of the jth gas turbine; is the active power of the jth gas turbine in period t; is the maximum active power of the jth gas turbine; Gas turbine assembly for power supply zone i; is the maximum downward climbing power of the jth gas turbine; is the active power of the jth gas turbine in period t-1; is the maximum upward climbing power of the jth gas turbine; is the reactive power of the jth gas turbine in period t; is the maximum apparent power of the jth gas turbine; The energy storage system constraints are: soc j,min ≤soc j,t ≤soc j,max in, is the minimum value of the charging and discharging active power of the jth energy storage system; is the charging and discharging active power of the j-th energy storage system in period t; is the maximum value of the charging and discharging active power of the jth energy storage system; For power supply partition i energy storage system collection; soc j,t is the charge of the j-th energy storage system in period t; soc j,t-1 is the charge of the j-th energy storage system in period t-1; is the charging efficiency of the jth energy storage system; is the charging power of the jth energy storage system in period t-1; is the maximum storage capacity of the jth energy storage system; Δt is the time difference between period t and period t-1; is the discharge power of the jth energy storage system in period t-1; is the discharge efficiency of the jth energy storage system; soc j,min is the minimum charge value of the j-th energy storage system; soc j,t is the charge of the j-th energy storage system in period t; soc j,max is the maximum charge value of the jth energy storage system; The constraints of the medium voltage soft control device are: in, is the active power transmitted by the port on the node i of the medium voltage flexible control device during period t; is the active power transmitted by the port on the node j of the medium voltage flexible control device during period t; is the reactive power provided by the port on the node i of the medium voltage flexible control device during period t, is the reactive power provided by the port at the node j of the medium voltage flexible control device during period t, and the positive directions of the power flow are respectively emitted from the node i of the medium voltage flexible control device and emitted from the node j of the medium voltage flexible control device; S i,max is the maximum capacity of the converter at the node i side of the medium voltage flexible control device; S j,max is the maximum capacity of the converter at the node j side of the medium voltage flexible control device; The renewable energy constraint is: in, is the active power of the jth renewable energy source in period t; is the active power of the jth renewable energy source in period t in the maximum power point tracking mode; is the reactive power of the jth renewable energy source in period t; To supply electricity to the district i renewable energy collection; is the maximum apparent power of the jth renewable energy source; The power flow balance constraint of the distribution network is: Among them, P k,t is the net active power injected at node k in the distribution network during period t; Ω(k) is the set of nodes connected to node k; V k,t is the voltage amplitude at node k during period t; V l,t is the voltage amplitude at node l during period t; G kl is the conductance of the branch between node k and node l; θ kl,t is the phase angle difference between the branch between node k and node l; B kl is the susceptance of the branch between node k and node l; is the active power of the kth renewable energy source in period t; is the active power of the kth gas turbine in period t; is the active power discharged by the kth energy storage system during period t; is the active power of the load at node k in the distribution network during period t; Q is the total active power flowing out of the connecting line at node k in the distribution network during period t; k,t is the net injected reactive power at node k in the distribution network during period t; is the reactive power of the kth renewable energy source in period t; is the reactive power of the kth gas turbine in period t; is the reactive power of the kth static VAR compensation device in period t; is the reactive power of the kth capacitor in period t; is the reactive power of the load at node k in the distribution network during period t; is the total reactive power flowing out of the connecting line at node k in the distribution network during period t; The voltage constraint of the distribution network node is: In min ≤V k,t ≤V max Among them, V min is the lower limit of the node voltage constraint; V max is the upper limit of node voltage constraint; The current carrying capacity constraint of the distribution network is: |I kl,t |≤I kl,max Among them, I kl,t is the current flowing through the branch between node k and node l during period t; I kl,max is the upper limit of the branch current between node k and node l; Agent state space for: in, is the power of the jth renewable energy source in period t; is the power of the jth load in period t; is the observable space of the jth gas turbine in period t, is the reactive power of the jth gas turbine in period t-1; is the observable space of the j-th energy storage system in period t, is the charging and discharging active power of the jth energy storage system in period t-1; is the observable space of the j-th medium voltage flexible control device in period t, Agent action space for: in, is the reactive power value of renewable energy in power supply zone i during period t; is the active output of the gas turbine in power supply zone i during period t; is the reactive power output of the gas turbine in power supply zone i during period t; is the charging and discharging power of the energy storage system in power supply partition i during period t; The reward function is: Among them, R i,T is the discount reward return of power supply partition i within the scheduling period T; γ is the discount coefficient, 0<γ<1; r i,t is the single-step reward value of power supply partition i during period t; N is the number of gas turbines in power supply partition i; is the operating cost of the jth gas turbine in power supply zone i during period t; M is the number of energy storage systems in power supply zone i; is the operating cost of the j-th energy storage system in power supply partition i during period t; F t penalty For penalty items; is the voltage constraint violation penalty function of the distribution network in period t; ψ1 is the first penalty weight; is the penalty function for the branch power flow constraint exceeding the limit in the distribution network during period t; W is the total number of nodes in the distribution network; V i,t is the voltage amplitude at node i during period t; V is the total number of distribution network branches; ψ2 is the second penalty weight; I j,t is the j-th branch current in period t; I j,max is the upper limit of the j-th branch current.

11. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for collaborative optimization of multiple power supply partitions in a distribution network as claimed in any one of claims 1 to 5 are implemented.

12. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for collaborative optimization of multiple power supply zones in a distribution network as claimed in any one of claims 1 to 5 are implemented.

Images