Optimization method and system for power distribution voltage regulation by considering excitation of micro-grid

Abstract

The invention provides an optimization method and system for power distribution voltage regulation by considering excitation of a micro-grid, and belongs to the technical field of power distribution network voltage control. Inputting the acquired data into the established microgrid optimal operation model and reinforcement learning agent before considering the voltage regulation excitation mechanism; based on the microgrid optimal operation model before the voltage regulation excitation mechanism is considered, calculating the operation voltage of the power distribution network when the power purchasing and selling power is obtained through microgrid optimization before the voltage regulation excitation mechanism is considered, and taking the operation voltage at the moment as a reference voltage; the reinforcement learning agent decides the current operation state according to the input data, outputs the power of each device, and then obtains the actual operation voltage through the power flow calculation of the power distribution network; and the intelligent agent strategy is updated based on the voltage regulation compensation, the reinforcement learning intelligent agent continuously updates the self strategy so as to make a decision on the current operation state, and the optimized operation power of each device is output.

Description

Technical Field

[0001] The present invention belongs to the technical field of power distribution network voltage control, and in particular, relates to an optimization method and system for power distribution voltage regulation considering incentive microgrids. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] The global energy and environmental crisis has accelerated the large-scale development of renewable energy and distributed energy, so microgrid technology has received widespread attention from the power industry. A microgrid is a small power system that includes distributed energy and loads, which can operate in grid-connected mode or island mode. The operation of a microgrid is relatively flexible, and the advantages of incorporating a microgrid into the distribution system are multifaceted: first, the distributed energy units in the microgrid can meet local energy needs, reduce dependence on the upper power grid, and improve the reliability of power supply; second, microgrids can promote the full utilization of new energy through renewable energy such as wind turbines and photovoltaics; and through microgrids, energy can be produced and sold locally, which can reduce long-distance transmission losses and investment in large transformers and transmission lines.

[0004] Domestic and foreign scholars have conducted a lot of research on the optimal operation of microgrids, but they mainly focus on the optimal scheduling of single microgrids, multiple microgrids and microgrid-distribution networks, and focus on solving the problem of how to improve the benefits of microgrids or reduce the operating costs of microgrids. However, with the continuous increase in the penetration rate of microgrids, the peak load and peak reverse energy flow during non-peak / peak hours caused by the mismatch between the characteristics of new energy generation and the power consumption characteristics of loads will violate the voltage limitations of the distribution network. Most studies have ignored the consideration of the voltage constraints of the distribution network.

[0005] For example, in the prior art, CN118157148A_An intelligent multi-agent collaborative voltage control method for microgrid groups. This method first establishes a master-slave game model between the distribution network and multiple microgrids, and uses a reasonable economic incentive mechanism to encourage multiple microgrids to autonomously participate in the distribution network voltage control; in order to protect the privacy of multiple subjects, multiple subjects make dynamic game decisions based on incomplete information and deep reinforcement learning algorithms, so that they can maximize their own interests while providing voltage control auxiliary services for the distribution network; the proposed method can enable the distribution network to encourage multiple stakeholders to autonomously participate in the optimization of distribution network operation under a reasonable economic incentive mechanism, so as to reduce the additional configuration of voltage regulation equipment in the distribution network, thereby improving the economy and energy efficiency of the distribution network.

[0006] The incentive problem raised in the above patent is mainly the voltage regulation subsidy for the active change of the microgrid, and the power transaction price is determined by the master-slave game bidding between the distribution network and the microgrid. However, the voltage regulation subsidy in the above patent is only measured by the active change of the microgrid, and the impact of the actual voltage regulation effect of the distribution network node as an incentive mechanism is not considered. In fact, when the distribution network is not considered to set voltage regulation subsidies or voltage regulation incentives, the microgrid often maximizes its own benefits. The operation plan under this state may often lead to a series of problems such as voltage exceeding the limit. By measuring the actual voltage regulation effect, the contribution of the microgrid in the voltage regulation process can be better reflected. By converting the actual voltage regulation effect into voltage regulation compensation, and expressing it together with the microgrid operating cost through the economic objective function, the enthusiasm of the microgrid for voltage regulation can be fully stimulated.

[0007] Traditional equipment involved in the voltage regulation of distribution networks mainly includes on-load voltage regulating transformers, capacitor banks, photovoltaics, energy storage systems, etc., while microgrids, as small systems including sources, loads, and storage, can flexibly adjust active and reactive power to contribute to solving the problem of voltage regulation in distribution networks. However, in actual operation, microgrids and microgrids belong to different operators. Voltage regulation of microgrids may increase operating costs or reduce profits. There are many practical problems in direct control and scheduling of distribution networks. The main problems solved by the present invention are: how to fully mobilize the enthusiasm of microgrid voltage regulation, compensate for the cost increase caused by voltage regulation of microgrids, and mobilize different microgrids to minimize the cost of voltage regulation. Summary of the invention

[0008] In order to overcome the above-mentioned deficiencies of the prior art, the present invention provides an optimization method for power distribution voltage regulation taking into account incentives for microgrids, thereby mobilizing the enthusiasm of microgrids to participate in power distribution network voltage regulation, which is conducive to building a safer and more reliable power distribution system.

[0009] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: In a first aspect, a method for optimizing distribution voltage regulation by considering incentive microgrids is provided, comprising: Obtain the distribution network load data predicted by the day before, the photovoltaic power output data predicted, the wind turbine power output data predicted, the energy storage system capacity, the micro gas turbine capacity, the electric boiler capacity and the time-of-use electricity price; The acquired data is input into the established microgrid optimal operation model and reinforcement learning agent before considering the voltage regulation incentive mechanism; The optimal operation model of the microgrid before considering the voltage regulation incentive mechanism calculates the operating voltage of the distribution network when the microgrid is optimized to purchase and sell power before considering the voltage regulation incentive based on the input data. The operating voltage at this time is used as the reference voltage. The reinforcement learning agent makes decisions on the current operating status based on the input data, outputs the power of each device, and then calculates the actual operating voltage through the distribution network flow; The voltage regulation compensation is calculated according to the reference voltage, the actual operating voltage and various set voltage regulation incentives to obtain the voltage regulation compensation. The agent strategy is updated based on the voltage regulation compensation. The reinforcement learning agent continuously updates its own strategy to make decisions on the current operating status and outputs the optimized operating power of each device.

[0010] As a further technical solution, the optimal operation model of the microgrid before the voltage regulation incentive mechanism is considered to minimize the microgrid operation cost, which mainly includes equipment operation cost, fuel cost and electricity purchase and sales cost; The operation constraints of energy storage system, micro gas turbine, electric boiler and microgrid power balance constraints are considered.

[0011] As a further technical solution, various voltage regulation incentives are provided including: Reduced voltage regulation costs in distribution networks; Reduced DG disconnection risk cost in distribution network; Reduced electricity costs for voltage-dependent loads in microgrids.

[0012] As a further technical solution, the voltage regulation compensation is calculated according to the reference voltage, the actual operating voltage and various set voltage regulation incentives to obtain the voltage regulation compensation, which specifically includes: The compensation calculated for the three types of incentives includes the following three parts: ; ; ; Therefore, microgrids are The total voltage regulation compensation obtained at all times is: .

[0013] As a further technical solution, regarding reinforcement learning agents, each microgrid is represented by an agent, which learns the optimal strategy by repeatedly exploring the sequential decision-making process.

[0014] As a further technical solution, the elements of the intelligent agent include ,in: is the set of states that the agent can perceive, is the strategy of the agent; The action taken by the agent according to the strategy in the current state; is the state transition function, that is, the probability of transitioning to the next state under the current state and action; It is the set of rewards that the agent can obtain after performing actions in the environment; State Space: Agents Instead of making learning decisions for the microgrid, the current environment is observed at each decision moment to obtain the current state , including photovoltaic power, wind turbine power, electrical load power, thermal load power, energy storage charging and discharging power, micro gas turbine power, electric boiler power, electricity purchase and sales price and current decision time; Action space: Based on the current state and strategy, the agent Make decisions and obtain the actual actions of each device after linear mapping , including the charging and discharging power of energy storage, the change in power of micro gas turbine and the power of electric boiler; Reward space: Apply the actual actions of each device in the simulation environment to get the reward at that moment , including voltage regulation compensation, income from purchasing and selling electricity to the superior power grid, charging and discharging costs of energy storage, fuel costs of micro gas turbines and reactive power regulation costs.

[0015] In a second aspect, an optimization system for power distribution voltage regulation considering incentive microgrid is provided, including: The data acquisition module is configured to: obtain the distribution network load data predicted a day ago, the photovoltaic predicted output data, the wind turbine predicted output data, the energy storage system capacity, the micro gas turbine capacity, the electric boiler capacity and the time-of-use electricity price; The acquired data is input into the established microgrid optimal operation model and reinforcement learning agent before considering the voltage regulation incentive mechanism; The reference voltage determination module is configured to: the optimal operation model of the microgrid before considering the voltage regulation incentive mechanism calculates the operating voltage of the distribution network when the microgrid is optimized to purchase and sell power before considering the voltage regulation incentive based on the input data, and the operating voltage at this time is used as the reference voltage; The actual operating voltage determination module is configured as follows: the reinforcement learning agent makes a decision on the current operating state according to the input data, outputs the power of each device, and then obtains the actual operating voltage through the distribution network flow calculation; The voltage regulation compensation module is configured to calculate the voltage regulation compensation according to the reference voltage, the actual operating voltage and various set voltage regulation incentives to obtain the voltage regulation compensation, update the intelligent agent strategy based on the voltage regulation compensation, and the reinforcement learning intelligent agent continuously updates its own strategy to make decisions on the current operating status and output the optimized operating power of each device.

[0016] One or more of the above technical solutions have the following beneficial effects: The technical solution of the present invention fully considers the incentives for microgrids in the voltage control problem, measures the voltage regulation effect of microgrids through the voltage regulation incentive mechanism, and effectively improves their enthusiasm for participating in voltage regulation. The voltage regulation incentive includes direct incentives for the distribution network to microgrids to participate in voltage regulation: the voltage regulation cost of the distribution network is reduced, and the DG disconnection risk cost of the distribution network due to overvoltage problem is reduced; indirect incentives obtained by reducing the cost of load power demand in the microgrid: the electricity cost of voltage-related loads in the microgrid is reduced. Through the above-mentioned voltage regulation incentive, a reasonable and effective voltage regulation incentive mechanism is constructed to measure the voltage regulation contribution of the microgrid. And by introducing the SQDDPG algorithm of Shaply Q value, the voltage regulation rewards of different microgrids are allocated, and fair credit distribution among intelligent entities is achieved. In the optimization strategy of considering the incentive microgrid for distribution voltage regulation, the microgrid can obtain benefits by providing voltage regulation support for the distribution network on the basis of ensuring its own operating economy, and the overall operating economy is improved, and the voltage over-limit problem of the distribution network is solved, which is conducive to building a safe and reliable distribution system.

[0017] The voltage regulation incentive proposed in the technical solution of the present invention takes into account that the microgrid can regulate reactive power. Reactive power has a lower regulation cost and a more significant voltage regulation effect. It is more reasonable to consider active power and reactive power together.

[0018] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0020] Figure 1 This is a schematic diagram of the distribution network structure; Figure 2 It is a schematic diagram of the voltage regulation incentive mechanism; Figure 3 Schematic diagram of the optimization strategy for distribution voltage regulation considering incentive microgrid; Figure 4 The voltage distribution diagram of the distribution network before the voltage regulation incentive mechanism is considered; Figure 5 This is a schematic diagram of voltage distribution in the distribution network after considering the voltage regulation incentive mechanism; Figure 6 The figure is a flow chart of a sub-method of an embodiment of the present invention. DETAILED DESCRIPTION

[0021] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0022] It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.

[0023] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0024] When voltage regulation compensation is not set, the microgrid always makes full use of its flexibility resources, optimizes the purchase and sale of electricity according to the time-of-use electricity price on the basis of meeting the local load, and optimizes its operation plan with the goal of minimizing costs, without considering the operating status of the distribution network. However, if the distribution network is actually operated according to this operation plan, the problem of node voltage exceeding the limit will inevitably occur, resulting in increased difficulty in operation and control. The sub-technical scheme of the embodiment of the present invention proposes a strategy to incentivize microgrids to regulate the voltage of the distribution network. The proposed invention fully considers the incentive factors for the voltage regulation of the microgrid, reasonably measures the contribution to the voltage regulation of the distribution network and fairly distributes the rewards. The microgrid can improve its own operating economy while providing voltage support for the distribution network, which is of great significance to the construction of a safe and reliable distribution network.

[0025] Embodiment 1 See attached Figure 6 As shown, this embodiment discloses an optimization method for power distribution voltage regulation considering incentive microgrid, including: Step 1: Input the distribution network load data predicted a day ago, the photovoltaic power output data predicted, the wind turbine power output data predicted, the energy storage system capacity, the micro gas turbine capacity, the electric boiler capacity, and the time-of-use electricity price; Step 2: Introduce the optimal operation model of the microgrid before the voltage regulation incentive mechanism to minimize the microgrid operation cost, mainly including equipment operation cost, fuel cost and electricity purchase and sales cost; consider the operation constraints of the energy storage system, the operation constraints of the micro gas turbine, the operation constraints of the electric boiler and the power balance constraints of the microgrid; Step 3: The voltage regulation incentive mechanism considers reducing the voltage regulation cost of the distribution network, reducing the risk cost of DG disconnection in the distribution network due to overvoltage problem, and reducing the electricity cost of voltage-related loads in the microgrid as incentives for microgrid voltage regulation. The voltage regulation compensation can be obtained by comparing the voltage operating status of the nodes before and after the voltage regulation.

[0026] The voltage regulation incentive mechanism is attached Figure 2 The entire content is essentially a calculation formula used to calculate the attached Figure 2The microgrid voltage regulation compensation in the formula requires three parts: (1) Various voltage regulation incentives include three types: reduction of voltage regulation costs of distribution networks, reduction of DG disconnection risk costs of distribution networks due to overvoltage problems, and reduction of electricity costs of voltage-related loads in microgrids; (2) The voltage operation state of the distribution network is obtained by the optimal operation plan of the microgrid without considering the voltage regulation incentive and used as the reference voltage; (3) The calculation formula for voltage regulation compensation is used to obtain the voltage regulation compensation of the microgrid through the reference voltage, the actual operating voltage and the calculation formula; The voltage regulation incentive mechanism is all of the above, and the voltage regulation incentive is the three voltage regulation compensation types set in (1).

[0027] Step 4: Establish a reinforcement learning agent based on the SQDDPG algorithm, construct a microgrid optimization problem based on the voltage regulation incentive mechanism as an interactive environment, construct the state space, action space, and reward space of the agent, and calculate the reward for the agent through the node voltage of the system after the agent makes a decision and the reference voltage before considering the voltage regulation incentive mechanism. The optimal strategy of the agent is obtained through continuous learning and iteration.

[0028] The data and parameters in step one are the input data of the "microgrid optimal operation model before the introduction of the voltage regulation incentive mechanism" in step two and the "microgrid optimization problem based on the voltage regulation incentive mechanism" in step four.

[0029] The model constructed in step 2 is optimized based on the data and parameters in step 1, that is, the data and parameters in step 1 are known data, and the optimization variable is the active power of each device.

[0030] Step 4 constructs the interactive environment required by the reinforcement learning agent based on the data and parameters in step 1. Similarly, the optimization variable controlled by the agent is the power of each device, and the data and parameters in step 1 are known data used to construct the state space of the agent.

[0031] The operation plan of the microgrid, i.e., the power purchase and sale, can be obtained through the optimization model in step 2. The distribution network can obtain the node voltage by substituting the power purchase and sale power into the power flow calculation. Because this voltage is obtained through the operation plan of the microgrid before considering the voltage regulation incentive, this voltage is used as the reference voltage.

[0032] The intelligent agent in step 4 makes decisions based on the current operating status and controls the power of each device. Because the voltage regulation incentive is taken into account in the objective function in step 4, in addition to the income from purchasing and selling electricity, voltage regulation compensation can also be obtained. The voltage regulation compensation is calculated through the actual operating voltage of the distribution network, the reference voltage and the voltage regulation incentive mechanism set in step 3. Through the calculated voltage regulation compensation, the intelligent agent in step 3 updates its parameters to obtain higher operating benefits.

[0033] In step 2, the "microgrid optimal operation model before the voltage regulation incentive mechanism" is used to calculate the operating voltage of the distribution network when the microgrid is optimized to purchase and sell electricity before the voltage regulation incentive is considered. The operating voltage at this time is used as a reference voltage and compared with the actual operating voltage after considering the voltage regulation incentive mechanism. The contribution of the microgrid to voltage regulation is measured, and the voltage regulation compensation of the microgrid is calculated in combination with the voltage regulation incentive mechanism to encourage its enthusiasm for participating in voltage regulation.

[0034] The voltage regulation incentive mechanism in step three calculates the voltage regulation compensation based on the reference voltage in step two, the actual operating voltage in step four and the various voltage regulation incentives set. After obtaining the voltage regulation compensation, the microgrid can not only change the power to obtain revenue through the purchase and sale of electricity, but also obtain revenue through voltage regulation by changing the power. The reinforcement learning agent obtains higher overall operating benefits by constantly updating its own strategies.

[0035] In step 4, known data such as operating data and equipment parameters are obtained through step 1 to build an interactive environment for reinforcement learning; voltage regulation compensation can be obtained through steps 2 and 3 to update its own strategy to obtain higher returns.

[0036] Through the continuous updating of the intelligent agents, a stable control strategy for each intelligent agent can be obtained, which can control the power of each device in the microgrid, output the optimized operating power of each device and the obtained electricity purchase and sales income and voltage regulation compensation.

[0037] The intelligent body adjusts the interactive power between the microgrid and the distribution network by controlling the active power and reactive power of the equipment. The interactive power includes not only the income from purchasing and selling electricity with the distribution network, but also the voltage regulation compensation obtained by adjusting the node voltage. Through the optimized strategy, the operating power of the microgrid can be obtained. This operating power can reduce the operating voltage deviation of the distribution network while ensuring the improvement of the overall operating income of the microgrid, thereby ensuring the safety of operation.

[0038] like Figure 2As shown in the figure, in order to construct a voltage regulation incentive mechanism, first, without considering the voltage regulation incentive, the optimal operation plan of the microgrid is obtained and the power flow calculation is performed to obtain the operating status of each node. The node voltage operating status of the system at this time is used as the reference voltage, and compared with the actual operating voltage after the introduction of the incentive, the voltage regulation compensation of the microgrid is obtained. The microgrid can adjust the node voltage of the distribution network by changing the operation plan or adjusting the interactive power. In this process, the microgrid is motivated to participate in the voltage regulation of the distribution network through the voltage regulation incentive. The voltage regulation incentive consists of the following three parts: the voltage regulation cost of the distribution network is reduced, the DG disconnection risk cost of the distribution network is reduced due to the solution of the overvoltage problem, and the electricity cost of the voltage-related load in the microgrid is reduced. Among them, the first two items are direct incentives for the distribution network to the microgrid to participate in voltage regulation, and the third item is an indirect incentive obtained by reducing the electricity demand cost of the load in the microgrid.

[0039] The voltage regulation incentive mechanism can calculate the voltage regulation compensation of the microgrid through the reference voltage, actual voltage and voltage regulation incentive. As shown in formula 19-22, the reference voltage and actual voltage are the corresponding voltage values ​​in the formula. Each voltage regulation incentive corresponds to a calculation formula. Substituting the reference voltage and actual voltage into the corresponding voltage regulation incentive calculation formula can obtain the corresponding voltage regulation compensation.

[0040] Consider the optimization strategy of incentive microgrid for distribution voltage regulation. Figure 3 As shown. The optimal operation strategy of the microgrid considering voltage regulation incentives is obtained through SQDDPG algorithm optimization. In the algorithm, each agent represents a microgrid to control the power of internal equipment, and the global reward distribution is carried out by introducing Shapley Q value to measure the contribution of different microgrids. The algorithm adopts a centralized training distributed execution framework. After continuous exploration and learning, the agent gradually converges to the optimal operation strategy. When the strategy is running intraday, it can be quickly controlled only through local information and trained local strategies, effectively reducing data transmission and calculation.

[0041] In this implementation case, the distribution network structure is as follows: Figure 1As shown. The voltage regulation incentive mechanism of the microgrid includes voltage regulation incentive setting, the optimal operation model of the microgrid without considering the voltage regulation incentive, and the voltage regulation compensation calculation. The optimal operation model of the microgrid without considering the voltage regulation incentive takes into account the operation constraints of the energy storage system, the operation constraints of the micro gas turbine, the operation constraints of the electric boiler, and the power balance constraints. The Gurobi solver is called in Python to solve the problem, and the optimal operation plan of the microgrid without considering the voltage regulation incentive is obtained, that is, the output power of each device; the voltage regulation strategy of the microgrid for the distribution network is stimulated by using the power flow calculation and voltage regulation incentive mechanism of the distribution network as the interactive environment to train the set reinforcement learning agent through deep reinforcement learning, and the operation strategy of the microgrid is obtained through iterative optimization of the agent and the environment, including the charging and discharging power of the energy storage, the output power of the gas turbine, the power of the electric boiler, the reactive power of the fan, and the reactive power of the photovoltaic.

[0042] The voltage regulation incentive mechanism is attached Figure 2 All contents in the Figure 2 The microgrid voltage regulation compensation in the formula requires three parts: (1) Various voltage regulation incentive settings, including three types: reducing the voltage regulation cost of the distribution network, reducing the risk cost of DG disconnection in the distribution network due to overvoltage problems, and reducing the electricity cost of voltage-related loads in the microgrid; (2) The voltage operation state of the distribution network is obtained by using the optimal operation model of the microgrid without considering the voltage regulation incentive and used as the reference voltage; (3) Voltage regulation compensation calculation: the voltage regulation compensation of the microgrid is obtained through the reference voltage, actual operating voltage and voltage regulation incentive calculation formula.

[0043] Voltage regulation incentive setting: voltage regulation cost in distribution network is reduced, DG disconnection risk cost in distribution network is reduced, and electricity cost of voltage-related loads in microgrid is reduced.

[0044] Regarding the reduction of voltage regulation costs in distribution networks: Distribution networks usually regulate voltage by installing on-load voltage-regulating transformers, capacitor banks, static VAR compensators and other devices. The configuration, operation and maintenance costs of these devices can be considered as the voltage regulation costs of distribution networks. When the node voltage is closer to the rated value, the voltage regulation cost required by the distribution network is smaller, which can be expressed by the following formula: (1); In the formula, for The cost of voltage regulation of distribution network at all times; is the total number of distribution network nodes; The voltage regulation cost coefficient of the distribution network representing the unit node voltage deviation; for Time Node The voltage amplitude of is the rated voltage, and formula (1) is used in the subsequent formula (19) to calculate the voltage regulation compensation of the first type of voltage regulation excitation.

[0045] Regarding the reduction of DG disconnection risk cost in the distribution network: Failure to take voltage regulation measures or improper voltage regulation will lead to overvoltage in the distribution network, causing unnecessary economic losses to the distribution network. For example, DG connected through an inverter needs to be temporarily disconnected from the grid to ensure operational safety, and the disconnection of DG will increase the cost of purchasing electricity from the upper grid. The following formula is used to express the DG disconnection risk cost caused by overvoltage in the distribution network: (2); In the formula, for The risk cost of DG disconnection at the moment; represents the DG disconnection risk cost coefficient of unit node overvoltage; is the upper limit of the voltage at the distribution network node. Formula (2) is used in the subsequent formula (20) to calculate the voltage regulation compensation of the second type of voltage regulation incentive.

[0046] Regarding the reduction of electricity costs for voltage-related loads in microgrids: From the static voltage characteristics of the load, it can be seen that the active power of the load will change with the slow change of the node voltage. Under overvoltage conditions, the load often consumes more electrical energy. Therefore, if the node voltage can be effectively regulated, for the microgrid, the load power is reduced while allowing more electricity to be connected to the grid to increase revenue, effectively stimulating its enthusiasm for participating in voltage regulation. The ZIP model is used to model the load in the microgrid, which consists of three parts: constant impedance (Z), constant current (I) and constant power (P).

[0047] (3); In the formula, for Always access the node The voltage is The active power of the load in the microgrid at this time; Indicates rated voltage Hourly load active power; , and They respectively represent the proportion of constant impedance, constant current and constant power in the total node load. Formula (3) is used in the subsequent formula (21) to calculate the voltage regulation compensation of the third type of voltage regulation excitation.

[0048] In step 2, the optimal operation model of the microgrid without considering the voltage regulation incentive: the microgrid is mainly composed of distributed power sources, energy storage systems, electric boilers and user loads (electric load, thermal load), among which the distributed power sources can be divided into uncontrollable units and controllable units. The uncontrollable units include wind turbines and photovoltaics, while the controllable units include micro gas turbines. When the voltage regulation incentive is not considered, the operating cost of the microgrid mainly includes the operating cost of the above equipment, fuel cost and electricity purchase and sales cost. The optimal operation model includes the objective function and constraints.

[0049] Objective function: The microgrid is optimized with the goal of minimizing operating costs. The objective function is as follows: (4); In the formula, To optimize the time period collection; Assemble for microgrids; for Microgrid The charging and discharging costs of the energy storage system; for Microgrid The operating cost of micro gas turbines; and for Microgrid Power purchase and sale; and for The electricity purchase and sales price at the moment.

[0050] The constraints include: energy storage system operation constraints, micro gas turbine operation constraints, electric boiler operation constraints, and microgrid power balance constraints.

[0051] Regarding the operation constraints of the energy storage system: Considering the service life of the energy storage system, energy efficiency, discharge depth and capacity attenuation of the electrochemical energy storage technology, the charging and discharging cost of the energy storage system can be obtained by leveling the cost and power generation over the entire life cycle of the energy storage system, also known as the cost per kilowatt-hour. The charging and discharging cost of the energy storage system can be expressed by the following formula, where is the charging and discharging cost coefficient of the energy storage system: (5); (6); (7); (8); (9); In the formula, for Microgrid The charging power of the energy storage system, for Microgrid The discharge power of the energy storage system; Formula (6) is used to calculate the state of charge (SOC) of the energy storage system. and for Moment and Microgrid The SOC status of the energy storage system; and is a 0-1 variable, indicating Microgrid The charging and discharging status of the energy storage system; and for Microgrid The charging and discharging efficiency of the energy storage system; For microgrid Rated capacity of the energy storage system; is the optimization time interval. Formula (7) represents the nuclear power state constraint of the energy storage system, and are the minimum and maximum SOC states allowed by the energy storage system respectively; Formula (8) is the charge and discharge power constraint of the energy storage system, and are the maximum charging and discharging powers of the energy storage system respectively; Formula (9) indicates that the energy storage system can only operate in one state: charging or discharging.

[0052] Regarding the operating constraints of micro gas turbines: Micro gas turbines are generally used together with bromide refrigeration machines to form a cogeneration device. The high-grade thermal energy generated by natural gas combustion drives the micro gas turbine to generate electricity, and the high-temperature waste heat flue gas discharged is used for heating and supplying domestic hot water after passing through the bromide refrigeration machine. The mathematical model of its operating cost and thermal power relationship is as follows: (10); (11); In the formula, is the unit price of natural gas; The efficiency of micro gas turbine power generation; It is the low calorific value of natural gas; for Microgrid The heating power of the medium MT; is the heat dissipation loss; is the flue gas recovery rate of the cold bromine machine; is the heating coefficient of the cold bromine machine.

[0053] The operating constraints of the microturbine are as follows: (12); (13); In the formula, is the power limit of the micro gas turbine; and is the climbing power limit of the micro gas turbine.

[0054] Regarding the operating constraints of electric boilers: Electric boilers convert electrical energy into thermal energy to meet the thermal load, and the operating constraints are: (14); (15); In the formula, for Microgrid The electrical power consumed by EB; for Microgrid Thermal power converted from EB; is the electrothermal conversion efficiency of EB; It is the power limit of electric boiler.

[0055] About the power balance constraints of microgrid: The power balance constraints of electricity and heat in microgrid are as follows: (16); (17); (18); In the formula, and for Microgrid Active power of photovoltaic and wind turbines; for Microgrid The load power of the power grid; Formula (17) indicates that the microgrid can only operate in one state of purchasing or selling electricity. and is a 0-1 variable, indicating Microgrid The status of electricity purchase and sale; for Microgrid Medium heat load power.

[0056] Voltage regulation compensation calculation in step 3: After the optimal operation plan of each microgrid is known, the distribution network obtains the voltage state of the system through power flow calculation As a reference voltage, the actual voltage after considering the voltage regulation excitation is The microgrid voltage regulation compensation is obtained by comparison. The compensation calculated by three types of incentives includes the following three parts: (19); (20); (twenty one); Therefore, microgrids are The total voltage regulation compensation obtained at all times is: (twenty two).

[0057] In step 4, consider the optimization strategy of incentivizing microgrids for distribution voltage regulation: each microgrid in the SQDDPG algorithm is represented by an agent, and the agent learns the optimal strategy by repeatedly exploring the sequential decision-making process. Usually, multi-agent reinforcement learning needs to be described by a Markov game process (MGP).

[0058] The elements of an agent include ,in: is the set of states that the agent can perceive, is the strategy of the agent; The action taken by the agent according to the strategy in the current state; is the state transition function, that is, the probability of transitioning to the next state under the current state and action; is the reward set that the agent can obtain after performing actions in the environment. For each microgrid, the corresponding relationship between each element and the actual problem is as follows.

[0059] State Space: Agents Instead of making learning decisions for the microgrid, the current environment is observed at each decision moment to obtain the current state , including photovoltaic power, wind turbine power, electrical load power, thermal load power, energy storage charging and discharging power, micro gas turbine power, electric boiler power, electricity purchase and sales price and current decision time: (27); In the formula, and for Microgrid Active power of photovoltaic and wind turbines; for Microgrid China power load power; for Always access the node The voltage is The active power of the load in the microgrid at this time; for Microgrid The charging power of the energy storage system, for Microgrid The discharge power of the energy storage system; for Microgrid The power of the micro gas turbine; for Microgrid The power of the electric boiler; and for The electricity purchase and sales price at the time; For the current moment.

[0060] Action space: Based on the current state and strategy, the agent Make decisions and obtain the actual actions of each device after linear mapping , including the charging and discharging power of energy storage, the change in power of micro gas turbine and the power of electric boiler.

[0061] (28); In the formula, for Microgrid The charging power of the energy storage system, for Microgrid The discharge power of the energy storage system; for Microgrid The power change of the micro gas turbine; for Microgrid The power of the electric boiler; and for Microgrid Reactive power of photovoltaic and wind turbines.

[0062] Since the idle reactive capacity of photovoltaic and wind turbine inverters has strong voltage regulation capability and low regulation cost, they are high-quality voltage regulation resources, so they are used as part of the decision variables in the strategy considering voltage regulation incentives. The reactive output constraints and reactive costs of photovoltaic and wind turbine converters are as follows: (29); (30); In the formula, for Microgrid Reactive power of photovoltaic and wind turbines; and is the minimum and maximum power factor angle of the inverter; It is the utilization rate of the capacity occupied by the converter reactive power regulation. Under the condition of occupying the same capacity, the higher the utilization rate, the higher the reactive power regulation cost, which is generally 5%~10%; is the capacity of the grid-connected inverter for photovoltaic and wind turbines.

[0063] Reward space: Apply the actual actions of each device in the simulation environment to get the reward at that moment , including voltage regulation compensation, income from purchasing and selling electricity to the superior power grid, charging and discharging costs of energy storage, fuel costs of micro gas turbines and reactive power regulation costs: (31); In the formula, for The reward obtained by the agent at this moment; for The voltage regulation compensation obtained by the agent at each moment; and for The electricity purchase and sales price at the time; and for The power purchase and sale of the microgrid at any given moment; for Microgrid The operating costs of medium energy storage; for Microgrid The operating cost of micro gas turbines; for Microgrid The operating cost of the reactive power output of photovoltaic and wind turbines.

[0064] Because the agent always optimizes the strategy to obtain the maximum reward, the benefit of the microgrid is taken as the reward instead of the operating cost.

[0065] The technical solution of the present invention sets up a voltage regulation incentive mechanism. The microgrid obtains corresponding compensation by actively participating in voltage regulation. The incentive includes three parts: the reduction of voltage regulation cost in the distribution network, the reduction of DG disconnection risk cost in the distribution network, and the reduction of electricity cost of voltage-related loads in the microgrid. Without considering the voltage regulation incentive, the microgrid total plan maximizes the profit optimization operation plan, and is verified by substituting it into the distribution network flow calculation to obtain the voltage of each node as the reference value in the incentive mechanism. The voltage regulation compensation of the microgrid can be obtained by comparing the reference voltage and the actual voltage after the introduction of the voltage regulation incentive. Secondly, the microgrid optimization strategy considering the voltage regulation incentive is learned through the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm, and in order to measure the contribution of different microgrids in the voltage regulation process, the Shapley Q value is introduced in the traditional MADDPG algorithm for global reward distribution, which solves the problem of fair credit distribution among microgrids.

[0066] As attached Figure 3 As shown in , during the update process of the reinforcement learning algorithm, after the Shapley Q value is calculated, the parameters of the agent are updated using this value. This value is used in the update process of the algorithm to ensure that the agent can obtain higher rewards and improve the algorithm's optimization performance.

[0067] The Shapley Q-value Deep Deterministic Policy Gradient (SQDDPG) algorithm based on the voltage regulation incentive mechanism and the introduction of Shapley Q value mobilizes the enthusiasm of microgrids to participate in the voltage regulation of the distribution network, which is conducive to building a safer and more reliable distribution system. Considering the voltage distribution of the distribution network before and after the voltage regulation incentive mechanism, as shown in Figure 4 and Figure 5 As shown, the distribution network operates in a safer state.

[0068] Embodiment 2 The purpose of this embodiment is to provide a computer device, including 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 method when executing the program.

[0069] Embodiment 3 The purpose of this embodiment is to provide a computer-readable storage medium.

[0070] A computer-readable storage medium stores a computer program, which executes the steps of the above method when executed by a processor.

[0071] Embodiment 4 The purpose of this embodiment is to provide an optimization system for power distribution voltage regulation considering incentive microgrids, including: The data acquisition module is configured to: obtain the distribution network load data predicted a day ago, the photovoltaic predicted output data, the wind turbine predicted output data, the energy storage system capacity, the micro gas turbine capacity, the electric boiler capacity and the time-of-use electricity price; The acquired data is input into the established microgrid optimal operation model and reinforcement learning agent before considering the voltage regulation incentive mechanism; The reference voltage determination module is configured to: the optimal operation model of the microgrid before considering the voltage regulation incentive mechanism calculates the operating voltage of the distribution network when the microgrid is optimized to purchase and sell power before considering the voltage regulation incentive based on the input data, and the operating voltage at this time is used as the reference voltage; The actual operating voltage determination module is configured as follows: the reinforcement learning agent makes a decision on the current operating state according to the input data, outputs the power of each device, and then obtains the actual operating voltage through the distribution network flow calculation; The voltage regulation compensation module is configured to calculate the voltage regulation compensation according to the reference voltage, the actual operating voltage and various set voltage regulation incentives to obtain the voltage regulation compensation, update the intelligent agent strategy based on the voltage regulation compensation, and the reinforcement learning intelligent agent continuously updates its own strategy to make decisions on the current operating status and output the optimized operating power of each device.

[0072] Embodiment 5 The purpose of this embodiment is to provide a computer program product containing instructions, which, when running on a computer, enables the computer to execute the methods and functions involved in any of the above embodiments. The steps involved in the apparatus of the above embodiment correspond to the method embodiment 1, and the specific implementation method can refer to the relevant description part of embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode or carry an instruction set for execution by a processor and enable the processor to execute any method in the present invention.

[0073] Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0074] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A method for optimizing distribution voltage regulation by considering incentive microgrids, characterized by comprising: Obtain the distribution network load data predicted by the day before, the photovoltaic power output data predicted, the wind turbine power output data predicted, the energy storage system capacity, the micro gas turbine capacity, the electric boiler capacity and the time-of-use electricity price; The acquired data is input into the established microgrid optimal operation model and reinforcement learning agent before considering the voltage regulation incentive mechanism; The optimal operation model of the microgrid before considering the voltage regulation incentive mechanism calculates the operating voltage of the distribution network when the microgrid is optimized to purchase and sell power before considering the voltage regulation incentive based on the input data. The operating voltage at this time is used as the reference voltage. The reinforcement learning agent makes decisions on the current operating status based on the input data, outputs the power of each device, and then calculates the actual operating voltage through the distribution network flow; The voltage regulation compensation is calculated according to the reference voltage, the actual operating voltage and various set voltage regulation incentives to obtain the voltage regulation compensation. The intelligent agent strategy is updated based on the voltage regulation compensation. The reinforcement learning agent continuously updates its own strategy to make decisions on the current operating status and outputs the optimized operating power of each device.

2. The optimization method for power distribution voltage regulation considering incentive microgrid as claimed in claim 1, characterized in that: The optimal operation model of the microgrid before considering the voltage regulation incentive mechanism aims to minimize the microgrid operation cost, which mainly includes equipment operation cost, fuel cost and electricity purchase and sales cost; The operation constraints of energy storage system, micro gas turbine, electric boiler and microgrid power balance constraints are considered.

3. The optimization method for power distribution voltage regulation considering incentive microgrid as claimed in claim 1, characterized in that: The various voltage regulation incentives that can be set include: Reduced voltage regulation costs in distribution networks; Reduced DG disconnection risk cost in distribution network; Reduced electricity costs for voltage-dependent loads in microgrids.

4. The optimization method for power distribution voltage regulation considering incentive microgrid as claimed in claim 1, characterized in that: The objective function: The microgrid is optimized with the goal of minimizing the operating cost. The objective function is as follows: (4)； In the formula, To optimize the time period collection; Assemble for microgrids; for Microgrid The charging and discharging costs of the energy storage system; for Microgrid The operating cost of micro gas turbines; and for Microgrid Power purchased and sold; and for The electricity purchase and sales price at the moment.

5. The optimization method for power distribution voltage regulation considering incentive microgrid as claimed in claim 1, characterized in that: Regarding reinforcement learning agents, each microgrid is represented by an agent, which learns the optimal strategy by repeatedly exploring the sequential decision-making process.

6. The optimization method for power distribution voltage regulation considering incentive microgrid as claimed in claim 1, characterized in that: The elements of an agent include ,in: is the set of states that the agent can perceive, is the strategy of the agent; The action taken by the agent according to the strategy in the current state; is the state transition function, that is, the probability of transitioning to the next state under the current state and action; It is the set of rewards that the agent can obtain after performing actions in the environment; State Space: Agents Instead of making learning decisions for the microgrid, the current environment is observed at each decision moment to obtain the current state , including photovoltaic power, wind turbine power, electrical load power, thermal load power, energy storage charging and discharging power, micro gas turbine power, electric boiler power, electricity purchase and sales price and current decision time; Action space: Based on the current state and strategy, the agent Make decisions and obtain the actual actions of each device after linear mapping , including the charging and discharging power of energy storage, the change in power of micro gas turbine and the power of electric boiler; Reward space: Apply the actual actions of each device in the simulation environment to get the reward at that moment , including voltage regulation compensation, income from purchasing and selling electricity to the superior power grid, charging and discharging costs of energy storage, fuel costs of micro gas turbines and reactive power regulation costs.

7. An optimization system for power distribution voltage regulation considering incentive microgrids, characterized by comprising: The data acquisition module is configured to: obtain the distribution network load data predicted a day ago, the photovoltaic predicted output data, the wind turbine predicted output data, the energy storage system capacity, the micro gas turbine capacity, the electric boiler capacity and the time-of-use electricity price; The acquired data is input into the established microgrid optimal operation model and reinforcement learning agent before considering the voltage regulation incentive mechanism; The reference voltage determination module is configured to: the optimal operation model of the microgrid before considering the voltage regulation incentive mechanism calculates the operating voltage of the distribution network when the microgrid is optimized to purchase and sell power before considering the voltage regulation incentive based on the input data, and the operating voltage at this time is used as the reference voltage; The actual operating voltage determination module is configured as follows: the reinforcement learning agent makes a decision on the current operating state according to the input data, outputs the power of each device, and then obtains the actual operating voltage through the distribution network flow calculation; The voltage regulation compensation module is configured to calculate the voltage regulation compensation according to the reference voltage, the actual operating voltage and various set voltage regulation incentives to obtain the voltage regulation compensation, update the intelligent agent strategy based on the voltage regulation compensation, and the reinforcement learning intelligent agent continuously updates its own strategy to make decisions on the current operating status and output the optimized operating power of each device.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

9. 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 program, the steps of the method described in any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 6 are performed.

Images