Power distribution network multi-agent active-reactive collaborative optimization method, system and equipment based on non-cooperative game, and medium

Abstract

The invention relates to the technical field of power distribution network optimization regulation and control, and discloses a power distribution network multi-agent active-reactive collaborative optimization method, system and device based on a non-cooperative game, and a medium, and the method comprises the steps: constructing an active-reactive collaborative optimization regulation and control model in which a distributed resource agent participates based on the multi-agent non-cooperative game; the dynamic supply-demand ratio electricity price model adjusts the purchase and sale electricity price in real time according to the supply-demand state of the power distribution network; when the supply exceeds the demand, the electricity purchasing price is reduced, the electricity selling price is improved, and vice versa. The consumer voltage regulation service income evaluation model encourages the consumer to reasonably regulate the voltage through a reward and punishment mechanism, and maintains the voltage stable. And the active-reactive collaborative optimization model determines an optimal output and energy storage scheduling plan by taking the minimization of the total operation cost of the production consumer as a target. Besides, a non-cooperative game autonomous decision framework is constructed, so that the producer and the consumer independently optimize the strategy under the driving of price and income signals, Nash equilibrium is realized through iterative solution, and safe, stable and economic operation of the power distribution network is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of distribution network optimization and control technology, and in particular to a method, system, equipment and medium for multi-agent active and reactive power collaborative optimization of distribution networks based on non-cooperative game theory. Background Technology

[0002] As my country's energy structure undergoes profound optimization and low-carbon transformation, the penetration rate of distributed photovoltaic (PV) power in the distribution network is growing exponentially, driving the power system to evolve from the traditional "source follows load" paradigm to a new operating mode of "source-load interaction" or even "load follows source." In this process, the role of electricity users is undergoing a fundamental transformation: traditional passive consumers, who generate and consume electricity locally through self-built PV systems, are gradually evolving into "prosumers" with both energy production and consumption attributes, achieving integrated energy production and sales. While promoting the consumption of green electricity and the achievement of carbon emission reduction goals, this transformation also poses unprecedented challenges to the safe and stable operation of the distribution network. The high density and disorderly multi-point access of distributed PV can easily cause severe power fluctuations and voltage surges, leading to frequent voltage overruns, reverse power flows, and equipment overloads at local nodes of the distribution network, significantly weakening system voltage stability and power supply reliability.

[0003] To alleviate voltage quality issues arising from high-proportion distributed photovoltaic (PV) grid connections, existing research largely focuses on centralized active / reactive power regulation strategies, aiming to smooth voltage fluctuations through unified optimization and scheduling of distributed resources. However, this single-path control framework exhibits significant limitations in practical applications: First, centralized regulation relies on global information and unified decision-making, making it difficult to adapt to the geographically dispersed and multi-ownership scenarios of distributed resources. Second, mandatory power regulation can easily lead to secondary risks such as equipment overload, increased curtailment rates, and escalating operation and maintenance costs. Third, traditional voltage regulation equipment (such as OLTC and SVG) has high investment and operation and maintenance costs, and its response speed is difficult to match the rapid fluctuations in PV output. More importantly, with the increasing diversification of investment and operation entities for distributed resources, there are fundamental conflicts between the decision-making objectives, risk preferences, and interests of these entities and the overall operational requirements of the system.

[0004] Therefore, how to fully tap the regulatory potential of producers and consumers while ensuring the autonomy of multi-stakeholder decision-making in distributed resources, and how to construct a voltage coordination governance mechanism adapted to high-proportion distributed photovoltaic access, has become a key issue that urgently needs to be addressed in the transformation of new power distribution networks. Non-cooperative game theory provides a new theoretical paradigm for solving the dilemma of multi-stakeholder coordination. By characterizing the interactive game and goal optimization of each stakeholder in the strategy space, it can guide self-interested stakeholders to spontaneously converge to a Nash equilibrium state of system voltage security and economic operation in the process of pursuing the maximization of individual utility without relying on centralized instructions. This method not only fits the "plug-and-play, autonomous decision-making" operating characteristics of distributed resources, but also is compatible with my country's current energy governance structure of "source-grid-load-storage" multi-stakeholder co-governance at the institutional level, becoming a forward-looking direction for solving the voltage stability problem of new power systems. Summary of the Invention

[0005] In view of the aforementioned existing problems, the present invention is proposed.

[0006] Therefore, this invention provides a method, system, device, and medium for multi-agent active and reactive power collaborative optimization of distribution networks based on non-cooperative game theory. It can solve the problems of existing methods that rarely model the participation of distributed resource subjects in active and reactive power collaborative optimization and control from the perspective of multi-agent non-cooperative game theory, lack the design of mechanisms to incentivize multiple subjects to actively participate in voltage regulation tasks, and for voltage regulation in distribution networks, voltage is affected globally, and the decisions of each distributed resource subject are coupled and influenced by each other, making it difficult to model the compensation mechanism for individual distributed resource nodes participating in voltage regulation tasks.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a multi-agent active and reactive power collaborative optimization method for distribution networks based on non-cooperative game theory, comprising: Based on the net power, photovoltaic power generation output, and energy storage charging and discharging status of each producer and consumer within the time slot, the supply-demand ratio parameters of the distribution network are calculated, and the purchase price and sales price signals that change monotonically with the supply and demand status are generated, which are denoted as the dynamic supply-demand ratio electricity price model. The feeder is equivalent to a resistor-inductor-capacitor series structure. Based on historical operating data and power prediction information, the voltage setting value of the common coupling point and the tap position of the on-load tap changer are determined. The difference in on-load tap changer operation cost under the condition of whether or not producers and consumers participate in voltage regulation is calculated. This difference is then allocated to each producer and consumer as the revenue or penalty of voltage regulation service. This is called the producer and consumer voltage regulation service revenue evaluation model. Construct an active-reactive power coordinated optimization model, the objective of which is to minimize the total operating cost of each producer and consumer; Based on the aforementioned dynamic supply-demand ratio electricity price model, voltage regulation service revenue evaluation model, and active-reactive power coordinated optimization model, multi-entity active-reactive power coordinated optimization of the distribution network is carried out.

[0008] As a preferred embodiment of the multi-agent active-reactive power coordinated optimization method for distribution networks based on non-cooperative game theory described in this invention, it further includes: A non-cooperative game-based autonomous decision-making framework is constructed, which takes producers and consumers with photovoltaic power generation and flexible load capabilities as game participants, and uses their respective active power output, reactive power output and energy storage scheduling plans as strategy variables. Using total operating cost as the loss function, each participant independently optimizes and updates their strategy, driven by price and revenue signals generated by the operator based on the dynamic supply-demand ratio electricity price model and voltage regulation service revenue evaluation model, until the strategy converges. It outputs the reactive power setpoints, active power output plans, energy storage charging and discharging scheduling schemes, and control commands for on-load tap changers and capacitor banks for each producer and consumer.

[0009] This preferred solution, by constructing a non-cooperative game-based autonomous decision-making framework, fully mobilizes the autonomy and initiative of prosumers with photovoltaic power generation and flexible load capabilities. As game participants, prosumers can independently optimize their strategies based on their own circumstances and external price and revenue signals. This makes the operation of the entire distribution network more closely aligned with actual needs and improves resource utilization efficiency.

[0010] As a preferred embodiment of the multi-entity active-reactive power coordinated optimization method for distribution networks based on non-cooperative game theory described in this invention, wherein: in the dynamic supply-demand ratio electricity price model, when the supply-demand ratio is less than 1, the purchase price of electricity is increased and the sales price of electricity is decreased to incentivize local power generation. When the supply-demand ratio is greater than 1, reduce the purchase price of electricity and increase the sales price of electricity to promote electricity consumption. The electricity purchase price and electricity sales price are monotonically adjusted in the direction of change of the supply-demand ratio within adjacent time slots; The supply-demand ratio is calculated based on the positive and negative decomposition of the net power of all producers and consumers within the time slot, where the net power deficit is the sum of the positive net power and the net surplus is the sum of the negative net power. The net power of a single producer-consumer is the algebraic difference between its adjusted power consumption and photovoltaic power generation, and it is also included in the equivalent treatment of energy storage charging as power consumption and energy storage discharging as power generation.

[0011] As a preferred embodiment of the multi-agent active-reactive power coordinated optimization method for distribution networks based on non-cooperative game theory described in this invention, the producer-consumer voltage regulation service revenue evaluation model includes: A damped second-order dynamic voltage model is established based on the resistance, reactance and capacitance parameters of the feeder, and adjacent time slots are discretized. The voltage setpoint of the common coupling point and the tap position of the on-load tap changer are estimated using load and photovoltaic power generation forecast data. The total operating cost is calculated based on the number of taps of the on-load tap changer and the cost of a single tap operation, thus obtaining the cost difference when there is no consumer involved in voltage regulation. Producers and consumers receive positive benefits when they participate in voltage regulation to reduce the number of on-load tap changer operations or the adjustment range. If this leads to voltage deterioration, a penalty is imposed. The benefits or penalties are distributed according to the proportion of reactive power output of each producer and consumer.

[0012] As a preferred embodiment of the multi-subject active-reactive power coordinated optimization method for distribution networks based on non-cooperative game theory described in this invention, the total operating cost includes power utilization loss, generation cost, output adjustment cost, inverter reactive power cost, on-load tap changer operation cost, and capacitor bank operating cost. The reactive power cost of the inverter includes three parts: the power loss cost corresponding to the additional active power loss caused by reactive power output, the life loss cost corresponding to the shortened equipment life due to the increase in apparent power, and the opportunity cost corresponding to the reduction in available active power due to reactive power occupying inverter capacity.

[0013] As a preferred embodiment of the multi-agent active-reactive power coordinated optimization method for distribution networks based on non-cooperative game theory described in this invention, the constraints of the active-reactive power coordinated optimization model include: Power flow equations of a distribution network described in polar coordinates; Upper and lower limits of node voltage and upper limit of branch current; Active power output, reactive power output, and power factor boundary limits of the inverter; Limitations on the tap position range and number of operations of on-load tap changers; Capacitor bank's single reactive power output capability and switching frequency limitations; Upper limit of charging and discharging power, upper and lower limits of state of charge, and rated capacity limit of energy storage system; User-side power comfort requirements and upper limit constraints on the amount of electricity sent to the main grid.

[0014] As a preferred embodiment of the multi-agent active-reactive power collaborative optimization method for distribution networks based on non-cooperative game theory described in this invention, wherein the non-cooperative game autonomous decision-making framework is solved by the optimal response iteration method. In each iteration, the operator recalculates the supply-demand ratio and updates the electricity price signal based on the latest net power, and at the same time recalculates the on-load tap changer operation cost and updates the voltage regulation revenue signal based on the latest voltage status. The system is considered to have reached Nash equilibrium when the changes in the strategy variables of all producers and consumers are less than the preset tolerance.

[0015] Secondly, this invention provides a multi-agent active and reactive power cooperative optimization system for distribution networks based on non-cooperative game theory, comprising: The first model building module is used to calculate the supply-demand ratio parameters of the distribution network based on the net power, photovoltaic power generation output and energy storage charging and discharging status of each producer and consumer in the time slot, and to generate the purchase price and sales price signals that change monotonically with the supply and demand status, which is denoted as the dynamic supply-demand ratio electricity price model. The second model building module is used to equate the feeder to a resistor-inductor-capacitor series structure, determine the voltage setting value of the common coupling point and the tap position of the on-load tap changer based on historical operating data and power prediction information, calculate the difference in on-load tap changer operation cost under the condition of whether or not producers and consumers participate in voltage regulation, and allocate the difference as the revenue or penalty of voltage regulation service to each producer and consumer, which is called the producer and consumer voltage regulation service revenue evaluation model. The third model building module is used to construct an active-reactive power coordinated optimization model, which aims to minimize the total operating cost of each producer and consumer. The collaborative optimization module is used to perform multi-entity active and reactive power collaborative optimization of the distribution network based on the dynamic supply-demand ratio electricity price model, voltage regulation service revenue evaluation model, and active-reactive power collaborative optimization model.

[0016] Thirdly, the present invention provides an electronic device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.

[0017] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0018] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention proposes a multi-agent active and reactive power collaborative optimization method for distribution networks based on non-cooperative game theory. This method is based on multi-agent non-cooperative game theory and constructs an active and reactive power collaborative optimization and control model involving distributed resource entities. The dynamic supply-demand ratio electricity price model adjusts the purchase and sale prices of electricity in real time according to the supply and demand status of the distribution network; when supply exceeds demand, the purchase price is reduced and the sale price is increased, and vice versa. The producer-consumer voltage regulation service revenue evaluation model incentivizes producers and consumers to regulate voltage reasonably through a reward and punishment mechanism to maintain voltage stability. The active and reactive power collaborative optimization model aims to minimize the total operating cost of producers and consumers and determines the optimal output and energy storage scheduling plan. In addition, a non-cooperative game autonomous decision-making framework is constructed, allowing producers and consumers to independently optimize strategies under the drive of price and revenue signals, and achieving Nash equilibrium through iterative solution, ensuring the safe, stable and economical operation of the distribution network. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 The present invention provides a flowchart of a multi-agent active and reactive power collaborative optimization method for distribution networks based on non-cooperative game theory, as an embodiment of the present invention.

[0021] Figure 2 This diagram illustrates the total load, photovoltaic power generation, and net load of a distributed resource entity in a multi-entity active-reactive power cooperative optimization method for a distribution network based on non-cooperative game theory, as provided in one embodiment of the present invention.

[0022] Figure 3 This is a 24-hour distributed resource subject electricity price fluctuation diagram of the last iteration of a multi-subject active-reactive power cooperative optimization method for distribution networks based on non-cooperative game theory, provided as an embodiment of the present invention.

[0023] Figure 4 This invention provides a method for coordinating active and reactive power output in a distribution network based on non-cooperative game theory, showing the price deviation between adjacent iterations of a buy-sell price chart in one embodiment of the invention.

[0024] Figure 5 This is a comparison diagram of the distributed resource subject revenues of a multi-subject active-reactive power collaborative optimization method for distribution networks based on non-cooperative game theory, provided as an embodiment of the present invention.

[0025] Figure 6This is a schematic diagram of photovoltaic active and reactive power adjustment quantities in a multi-agent active-reactive power collaborative optimization method for distribution networks based on non-cooperative game theory, provided as an embodiment of the present invention.

[0026] Figure 7 This is a schematic diagram of the active and reactive power adjustment of energy storage in a multi-subject active-reactive power collaborative optimization method for distribution networks based on non-cooperative game theory, provided as an embodiment of the present invention.

[0027] Figure 8 The voltage comparison diagram shows four scenarios of a multi-subject active-reactive power collaborative optimization method for distribution networks based on non-cooperative game theory, as provided in one embodiment of the present invention.

[0028] Figure 9 This is an internal structure diagram of an electronic device for a multi-agent active-reactive power cooperative optimization method for distribution networks based on non-cooperative game theory, as provided in one embodiment of the present invention. Detailed Implementation

[0029] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0030] It should be noted in advance that the system mentioned in the embodiments as the subject of real-time operation refers to any system configured with this method.

[0031] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides a multi-agent active and reactive power collaborative optimization method for distribution networks based on non-cooperative game theory, including: Existing technologies suffer from several problems. Firstly, they rarely model the collaborative optimization and control of distributed resource entities' participation in active and reactive power from a multi-entity non-cooperative game perspective. This makes it difficult to consider the mutual influence and autonomous decision-making of each entity, hindering the effective mobilization of their enthusiasm for participating in voltage regulation tasks. Secondly, the lack of mechanisms to incentivize multiple entities to actively participate in voltage regulation tasks results in insufficient motivation for producers and consumers to participate, failing to fully leverage their role in maintaining voltage stability. Furthermore, for distribution network voltage regulation, due to the global impact of voltage and the inter-entity coupling of decision-making among entities, it is difficult to model a compensation mechanism for individual distributed resource nodes participating in voltage regulation tasks. In practice, it is difficult to fairly and reasonably compensate participating nodes.

[0032] This invention provides a method that can effectively solve the problems mentioned above. The following will describe in detail how to implement this non-cooperative game theory-based multi-agent active and reactive power cooperative optimization method for power distribution networks. Figure 1 A flowchart of a multi-agent active and reactive power collaborative optimization method for distribution networks based on non-cooperative game theory is shown, including: S101, based on the net power, photovoltaic power generation output and energy storage charging and discharging status of each producer and consumer in the time slot, calculates the supply-demand ratio parameters of the distribution network and generates the purchase price and sales price signals that change monotonically with the supply and demand status, which is recorded as the dynamic supply-demand ratio electricity price model. In an embodiment of the present invention, in the dynamic supply-demand ratio electricity price model, when the supply-demand ratio is less than 1, the purchase price of electricity is increased and the sales price of electricity is decreased to incentivize local power generation. When the supply-demand ratio is greater than 1, reduce the purchase price of electricity and increase the sales price of electricity to promote electricity consumption. The purchase price and the sales price of electricity adjust monotonically in the direction of the supply-demand ratio within adjacent time slots. In one optional implementation, the supply-demand ratio parameter can be calculated based on the positive and negative decomposition of the net power of all prosumers within a time slot. When the supply-demand ratio is less than 1, it indicates that the system's net power deficit is greater than its net power surplus. In this case, the purchase price of electricity from the main grid is increased and the sales price of electricity to the main grid is decreased. This serves as an economic signal to incentivize prosumers to reduce external power purchases and increase local photovoltaic power generation or energy storage discharge. For example, during the evening peak load period and the period when photovoltaic output declines, the supply-demand ratio drops to 0.8. The purchase price is increased from 0.6 yuan / kWh to 0.75 yuan / kWh, and the sales price is decreased from 0.35 yuan / kWh to 0.25 yuan / kWh, prompting prosumers with energy storage to prioritize discharging to meet their own load. When the supply-demand ratio is greater than 1, it indicates that the system's net power surplus exceeds its net power deficit. In this case, the purchase price is decreased and the sales price is increased to promote local consumption or power back to the main grid.

[0033] The supply-demand ratio is defined as the ratio of net surplus electricity to net shortage electricity. Net surplus electricity is the sum of the absolute values ​​of the negative net power of all producers and consumers, and net shortage electricity is the sum of the positive net power. The net power of a single producer and consumer is the algebraic difference between its adjusted power consumption and photovoltaic power generation. Energy storage charging is equivalent to positive power consumption, and energy storage discharging is regarded as negative power consumption, i.e., power generation.

[0034] In this embodiment of the invention, the supply-demand ratio is calculated based on the positive and negative decomposition of the net power of all producers and consumers within the time slot, wherein the net power deficit is the sum of the positive net power and the net surplus is the sum of the negative net power. The net power of a single producer-consumer is the algebraic difference between its adjusted power consumption and photovoltaic power generation, and it is also included in the equivalent treatment of energy storage charging as power consumption and energy storage discharging as power generation.

[0035] In an alternative implementation, the net power of all producers and consumers in each time slot can be decomposed according to positive and negative signs, and the absolute values ​​of the positive and negative parts can be added together to obtain the net power deficit and net power surplus.

[0036] In one alternative implementation, the net power of a single producer-consumer is calculated by subtracting the photovoltaic power generation from its actual adjusted power consumption, and the energy storage charging power is included in the power consumption side and the energy storage discharging power is included in the power generation side for equivalent processing.

[0037] This equivalent treatment ensures that energy storage behavior is accurately reflected in supply and demand balance calculations, avoiding distortion of the supply-demand ratio due to neglecting energy storage status.

[0038] Among them, a positive net power value indicates that the producer-consumer as a whole absorbs power from the system, while a negative value indicates that power is injected into the system. Net power deficit reflects the total power demand that the system needs to supplement from the outside, while net power surplus reflects the total surplus power that the system can transmit to the outside or consume locally.

[0039] It should be noted that this step provides real-time, quantifiable economic incentive signals for subsequent optimization. By dynamically adjusting the purchase and sale of electricity, the degree of active power supply and demand tension in local areas can be accurately reflected, guiding producers and consumers to actively adjust their electricity consumption behavior or photovoltaic output, thereby smoothing net load fluctuations, reducing feeder power flow pressure, and laying the foundation for active power balance for subsequent voltage regulation and collaborative optimization.

[0040] S102, the feeder is equivalent to a resistor-inductor-capacitor series structure. Based on historical operating data and power prediction information, the voltage setting value of the common coupling point and the tap position of the on-load tap changer are determined. The difference in on-load tap changer operation cost under the condition of whether or not producers and consumers participate in voltage regulation is calculated. This difference is allocated to each producer and consumer as the revenue or penalty of voltage regulation service. This is called the producer and consumer voltage regulation service revenue evaluation model. In an alternative implementation, the feeder can be equivalent to a resistor-inductor-capacitor series structure. A damped second-order dynamic voltage model is established based on historical operating data and power prediction information, and adjacent time slots are discretized. For example, the voltage setpoint of the common coupling point is estimated using the total power of producers and consumers at time t−1 and the load and photovoltaic power generation prediction values ​​at time t. In one optional implementation, the tap position and number of operations of the on-load tap changer under voltage regulation without the participation of producers and consumers are calculated based on the deviation between the voltage setpoint and the voltage limit. For example, when the voltage deviation exceeds 0.3125% of the single tap regulation capacity, multiple tap operations are required until the voltage returns to the specified range. Then, the common coupling point voltage is recalculated based on the net power after the producers and consumers participate in voltage regulation, thereby determining the actual tap position and number of operations of the on-load tap changer under the participation of producers and consumers. For example, if the producers and consumers reduce the voltage deviation through reactive power output, the number of tap operations is reduced. In one alternative implementation, the difference in on-load tap changer operation costs between the two scenarios is treated as either a voltage regulation service revenue or a penalty, and allocated according to the proportion of reactive power output of each producer / consumer to the total reactive power output.

[0041] The on-load tap changer operation cost is determined by the product of the cost of a single tap changer operation and the total number of operations. A positive value for the voltage regulation service revenue indicates that the producer-consumer receives a reward for reducing the system voltage regulation cost, while a negative value indicates that the producer-consumer is penalized for deteriorating the voltage.

[0042] In this embodiment of the invention, the revenue includes revenue from the sale of excess electricity and revenue from voltage regulation services. In this embodiment of the invention, the consumer pressure regulation service revenue assessment model includes: A damped second-order dynamic voltage model is established based on the resistance, reactance and capacitance parameters of the feeder, and adjacent time slots are discretized. The voltage setpoint of the common coupling point and the tap position of the on-load tap changer are estimated using load and photovoltaic power generation forecast data. The total operating cost is calculated based on the number of taps of the on-load tap changer and the cost of a single tap operation, thus obtaining the cost difference when there is no consumer involved in voltage regulation. Producers and consumers receive positive benefits when they participate in voltage regulation to reduce the number of on-load tap changer operations or the adjustment range. If this leads to voltage deterioration, penalties are imposed. The benefits or penalties are distributed according to the proportion of reactive power output of each producer and consumer.

[0043] In an optional implementation, a damped second-order dynamic voltage model can be constructed based on the resistance, reactance, and capacitance parameters of the feeder, and adjacent time slots can be discretized. For example, the total injected power at time t−1 and time t can be substituted into the discretized voltage dynamic equation to calculate the voltage change at the common coupling point. The voltage level at the end of the feeder at time t can be derived using the predicted data of load and photovoltaic power generation, and then the common coupling point voltage setting value and the corresponding on-load tap changer tap position required to maintain the voltage within the limit can be estimated. For example, when the predicted voltage is 1.08 pu and the upper limit is 1.05 pu, the tap needs to be lowered to reduce the voltage. In one optional implementation, the baseline total operating cost is calculated based on the number of taps and the cost of a single tap operation when the on-load tap changer is regulating voltage without the participation of producers and consumers. Then, the required number of taps and the corresponding operating cost are recalculated based on the actual net power after the participation of producers and consumers. In one alternative implementation, the difference in action costs between the two scenarios is used as the benefit or penalty of voltage regulation service. When the producer-consumer reduces the number of actions or the adjustment range, a positive benefit is obtained. If its reactive power output causes the voltage to exceed the limit, a negative benefit is generated. The benefit or penalty is strictly allocated according to the proportion of reactive power output provided by each producer-consumer at time t to the total reactive power output of all producers-consumers. Among them, the cost of a single tap changer operation is a fixed maintenance cost parameter, the revenue from voltage regulation service is the algebraic difference of the OLTC operation cost with or without producer-consumer participation, and the allocation ratio is calculated in real time based on the actual reactive power output of each producer-consumer.

[0044] It should be noted that this step transforms the voltage regulation effect into quantifiable and distributable economic benefits or penalties, introducing a clear reactive power value signal for subsequent collaborative optimization models. Through a cost differential mechanism, producers and consumers are incentivized to proactively provide reactive power support in their decision-making, reducing the operation of traditional voltage regulating equipment, improving system voltage stability, and ensuring that benefit distribution matches individual contributions, thereby enhancing participation enthusiasm.

[0045] S103, Construct an active-reactive power co-optimization model, with the goal of minimizing the total operating cost of each producer and consumer; In this embodiment of the invention, the total operating cost includes power utility loss, power generation cost, output adjustment cost, inverter reactive power cost, on-load tap changer operation cost, and capacitor bank operating cost. The reactive power cost of an inverter consists of three parts: the power loss cost corresponding to the additional active power loss caused by reactive power output, the life loss cost corresponding to the shortened equipment life due to the increase in apparent power, and the opportunity cost corresponding to the reduction in available active power due to reactive power occupying inverter capacity.

[0046] In this embodiment of the invention, the constraints of the active-reactive power coordinated optimization model include: Power flow equations of a distribution network described in polar coordinates; Upper and lower limits of node voltage and upper limit of branch current; Active power output, reactive power output, and power factor boundary limits of the inverter; Limitations on the tap position range and number of operations of on-load tap changers; Capacitor bank's single reactive power output capability and switching frequency limitations; Upper limit of charging and discharging power, upper and lower limits of state of charge, and rated capacity limit of energy storage system; User-side power comfort requirements and upper limit constraints on the amount of electricity sent to the main grid.

[0047] It should be noted that this step integrates the dynamic electricity price signal and voltage regulation revenue signal generated in the previous two steps to construct a unified individual optimization objective function. This allows producers and consumers to naturally consider both active power balance and reactive power support needs while pursuing their own cost minimization. This model provides a mathematical foundation and decision-making basis for subsequent multi-agent collaborative optimization, ensuring that individual behavior is consistent with the system objective.

[0048] S104, based on the dynamic supply-demand ratio electricity price model, voltage regulation service revenue evaluation model and active-reactive power coordinated optimization model, performs multi-entity active-reactive power coordinated optimization of distribution network.

[0049] In an embodiment of the present invention, a non-cooperative game autonomous decision-making framework can also be constructed, in which producers and consumers with photovoltaic power generation and flexible load capabilities are taken as game participants, and their respective active power output, reactive power output and energy storage scheduling plans are taken as strategy variables. Using total operating cost as the loss function, each participant independently optimizes and updates their strategy, driven by price and revenue signals generated by the operator based on the dynamic supply-demand ratio electricity price model and the voltage regulation service revenue assessment model, until the strategy converges. It outputs the reactive power setpoints, active power output plans, energy storage charging and discharging scheduling schemes, and control commands for on-load tap changers and capacitor banks for each producer and consumer.

[0050] In an optional implementation, a non-cooperative game-based autonomous decision-making framework can be constructed, with producers and consumers capable of generating photovoltaic power and flexible loads as game participants. Their respective active power output, reactive power output, and energy storage charging and discharging power are used as strategy variables. For example, the strategy of producer and consumer i in time period t includes the active power reduction of the photovoltaic inverter, the reactive power injection, and the charging and discharging status of the energy storage unit. The total operating cost, including electricity purchase cost, equipment loss cost, and voltage regulation service revenue, is used as the loss function. For example, the loss function consists of electricity expenditure under dynamic electricity price minus voltage regulation service revenue plus energy storage cycle loss cost. In an optional implementation, driven by the real-time electricity price signal and voltage regulation revenue signal generated by the operator based on the dynamic supply-demand ratio electricity price model and the voltage regulation service revenue assessment model, each producer-consumer independently solves its own optimization problem and updates its strategy based on local information. For example, it uses gradient response or optimal response algorithms to iteratively update reactive power output and energy storage scheduling plans. Each participant alternately updates its strategy until Nash equilibrium converges. For example, convergence is determined when the strategy changes of all producers-consumers in two consecutive iterations are less than a preset threshold 1e−4. In one optional implementation, the final output includes the inverter reactive power setpoint, active power output plan, energy storage charging and discharging scheduling scheme, and coordinated control commands for the on-load tap changer and capacitor bank on the feeder side. For example, the inverter reactive power setpoint is used for real-time voltage support, and the capacitor bank switching command and the on-load tap changer position together realize feeder-level voltage regulation.

[0051] Among them, the dynamic supply-demand ratio electricity price model refers to the time-of-use electricity price mechanism based on the real-time adjustment of the ratio of regional net load to photovoltaic output; the voltage regulation service revenue assessment model refers to the aforementioned revenue calculation method based on the allocation of the difference in operating costs of on-load tap changers; and the Nash equilibrium refers to the state in which all producers and consumers cannot reduce their own loss function by unilaterally changing their own strategies given the strategies of other participants.

[0052] In this embodiment of the invention, the non-cooperative game autonomous decision-making framework is solved using an optimal response iteration method; In each iteration, the operator recalculates the supply-demand ratio and updates the electricity price signal based on the latest net power, and at the same time recalculates the on-load tap changer operation cost and updates the voltage regulation revenue signal based on the latest voltage status. The system is considered to have reached Nash equilibrium when the changes in the strategy variables of all producers and consumers are less than the preset tolerance.

[0053] In this embodiment of the invention, the feeder voltage calculation in the producer-consumer voltage regulation service revenue assessment model adopts a damped second-order dynamic model, which discretizes the historical power and predicted power of adjacent time slots, estimates the common coupling point voltage setpoint and its change, and is used to predict the OLTC tap position and number of operations and generate a voltage regulation price signal.

[0054] It should be noted that this step achieves closed-loop linkage of the first three steps of the model through a non-cooperative game framework, enabling each producer and consumer to independently optimize and iteratively converge under the drive of price and revenue signals, and finally output coordinated and consistent active power output, reactive power setting, energy storage scheduling and main grid equipment control commands, so as to realize the safe, economical and autonomous collaborative operation of the distribution network under the high proportion of distributed photovoltaic access.

[0055] Example 2: Based on the above examples, the specific implementation of the multi-agent active-reactive power coordinated optimization method for distribution networks based on non-cooperative game theory can be as follows: The dynamic supply-demand ratio (SDR) quantifies the active power supply and demand relationship of the distribution network in real time, and generates dynamic price signals based on the SDR value to guide photovoltaic producers to adjust their active power output. The dynamic supply-demand ratio electricity price model is expressed as follows: (1) in, The net surplus photovoltaic power of all producers and consumers within time t (i.e., the sum when the net power is negative): . Represents the net power deficit of all producers and consumers within time slot h (i.e., the sum of net power when it is positive): .

[0056] The net power of producer i at time t: (2) in The adjusted power consumption, This refers to the photovoltaic power generation capacity. t represents the charging and discharging power of the energy storage unit at time t.

[0057] It should be noted that SDR reflects the supply and demand balance in an energy-sharing region: SDR < 1: demand exceeds supply, requiring electricity to be purchased from the grid. SDR > 1: supply exceeds demand, requiring electricity to be sold to the grid. Prices reflect the supply and demand relationship in real time, incentivizing producers and consumers to adjust their loads to optimize SDR. This mechanism uses mathematical optimization and distributed algorithms to achieve a high-price incentive for demand-side response: when SDR is low (supply falls short of demand), the electricity purchase price (grid selling price) is increased, incentivizing users to reduce electricity consumption or increase local generation; when SDR is high (supply exceeds demand), the electricity selling price is decreased, encouraging users to increase electricity consumption.

[0058] when hour (3) (4) in, The internal electricity price for producers and consumers at time t is... The internal purchase price of electricity for producers and consumers at time t. These are the buying and selling prices of electricity from the power grid.

[0059] when hour (5) Furthermore, the income generated by prosumers through pressure regulating services The evaluation process is as follows: First, the distribution line is treated as an RLC series circuit. The feeder voltage is estimated using the predicted values ​​of photovoltaic power generation and load. The role of the on-load tap changer (OLTC) on the feeder is evaluated, and the corresponding OLTC operating cost is determined. Then, the OLTC operating cost is the regulation price sent to the prosumers. Finally, the benefits that prosumers gain from improving the system voltage condition depend on the amount of voltage violations eliminated. If the system voltage deteriorates, they will also be penalized by this price.

[0060] It should be noted that electricity is supplied to producers and consumers through the point of common coupling (PCC), and the total actual power is... This invention further defines the equivalent resistance, equivalent reactance, and equivalent capacitance of the line between the OLTC and PCC as follows: An OLTC is an autotransformer with automatically adjusting taps. It can perform multiple tap changes until the voltage is brought within specified limits. The taps typically provide a range of ±5% of the rated voltage, with a total of 32 levels, adjusting the rated voltage each time. 0.3125%. The total power of producers and consumers at time t−1 is defined as... The PCC voltage is fixed by operating the OLTC. The voltage at the other end of the OLTC is Its dynamic behavior is described by a second-order differential equation: (6) Where R is the equivalent resistance, L is the equivalent inductance, and C is the equivalent capacitance to ground. The damping coefficient is... It is the resonant angular frequency. For impedance matching factor, This is the resistance attenuation factor.

[0061] The result obtained through Laplace transform is: (7) in At time t Point voltage, It is a dynamic adjustment function , Where K is the damped oscillation frequency, and K is the oscillation amplitude coefficient. , Phase shift .

[0062] Discretize the time intervals t-1 and t: (8) Time step At time t, as the load and photovoltaic power change, the OLTC performs tap switching to maintain a constant PCC voltage. Using the predicted data for time t, the total power injected into the PCC is... The estimated OLTC voltage setpoint at time t is: (9) Discretization parameters , ; It should be noted that the cost of OLTC operation is determined by the number of tap changes and the cost of performing one tap change. Furthermore, the number of tap changes performed by the OLTC is calculated based on the regulated voltage and the regulating capacity of one OLTC tap change, and is 0.3125% of the OLTC's nominal value. Therefore, the estimated OLTC operation cost at time t is: (10) in, It is the cost of performing a tap changer. and It refers to the tap position of the OLTC at times t-1 and t.

[0063] Now assume that the voltage regulation service is provided by the producer-consumer. After power regulation by the producer-consumer, the net power flowing through the PCC at time t becomes... .

[0064] To keep the PCC voltage at its nominal value, the OLTC voltage should be set as follows: (11) The cost of OLTC actions involving producer-consumer intervention becomes: (12) Therefore, compared to the case where prosumers do not participate, in time t, when prosumers participate in voltage regulation, the cost savings from OLTC operation are paid to prosumers as voltage regulation revenue. This revenue can be positive or negative, and is expressed as: (13) in, To estimate the adjustment cost of OLTC at time t, This represents the actual regulation cost of the OLTC after time t following producer-consumer regulation. Cost savings from OLTC operation can be positive or negative, depending on whether the voltage condition improves or worsens. Producers are rewarded for reducing OLTC switching operations and penalized for otherwise negative actions.

[0065] (14) Of these, only and It is the producer-consumer optimization variable, the net power of PCC at time t. and It is the historical data of total producer-consumer power in the previous time slot t-1. and The predicted total active and reactive power injection at time t is also a known parameter. This invention further defines the predicted values ​​of the actual power load and actual photovoltaic power generation of producer-consumer i at time t as follows: and The reactive power generated by producer-consumer i at time t is .

[0066] (15) in: The actual reactive power demand of producer-consumer i in time period t-1; The actual photovoltaic output of producer-consumer i in time period t-1; The reactive power demand of producer-consumer i during time period t-1; The actual reactive power injected by producer-consumer i during time period t-1; The predicted active power of producer-consumer i in time period t-1; The photovoltaic output of producer-consumer i in time period t-1 is the predicted value. The reactive power demand of producer-consumer i during time period t-1; The actual active power demand of producer-consumer i during time period t; The actual photovoltaic output of producer-consumer i during time period t; The reactive power demand of producer-consumer i during time period t; The actual reactive power injected by producer-consumer i during time period t.

[0067] To highlight the differences in individual producer-consumer output, improve the economic efficiency and fairness of the incentive mechanism, and avoid the unfair phenomenon of those who contribute more but receive less, the revenue distribution method is set as proportional to the amount of output from each producer-consumer. This incentivizes producers-consumers to actively provide more reactive power output, thereby improving system voltage stability. The output ratio for an individual producer-consumer is calculated as follows: (16) The complete formula for the output of a single producer-consumer is: (17) The objective function of the active-reactive power coordination optimization model in this invention is: (18) in, The total cost for producer-consumer i during time period t; To adjust costs, For the cost of generating electricity.

[0068] (19) (20) in, The total cost of reactive power at the inverter interface; The cost of adjusting the capacitor bank once; The number of times the capacitor bank is adjusted; The cost of adjusting OLTC once; Adjust the number of times for OLTC; The total electricity consumption efficiency of consumer i during time period t; For the surplus electricity generated by producer-consumer i during time period t; Revenue is generated by providing voltage regulation services to consumer i during time period t.

[0069] (twenty one) in, Let αi be the original power consumption / original electricity usage of consumer i at time t. αi is the user's sensitivity coefficient to adjustments. This represents the adjusted power consumption. Producers and consumers adjust their power consumption plans based on the current electricity price, report their net power to the active and reactive power co-optimization model, recalculate the SDR and update the electricity price, and iterate until convergence is achieved, forming a dynamic equilibrium.

[0070] The reactive power cost of the inverter interface DER consists of three components: power loss cost, shortened lifespan cost, and lost opportunity cost, which can be expressed by the following formula.

[0071] (twenty two) (twenty three) in, It is the unit reactive power cost of the inverter interface; It is the reactive power output of the inverter interface; These are power loss cost, shortened lifespan cost, and lost opportunity cost, respectively.

[0072] The cost of power loss is expressed as: (twenty four) (25) in, The additional power loss of the inverter is caused by reactive power supply. It is the active power of the distribution network. Indicates active and apparent power flow. These are the fitting coefficients determined through experiments.

[0073] The cost of shortened lifespan is expressed as: (26) in, These are the fitting coefficients.

[0074] The opportunity cost of loss can be expressed as follows: (27) (28) in, The reduction in active power is caused by reactive power supply. It refers to the capacity of the inverter.

[0075] (29) in, The net power of producer i at time t; This is the internal buying and selling price, if , ;like , .

[0076] Furthermore, the specific constraints of the active-reactive power coordination optimization model are as follows: (1) Conservation of total electricity consumption: (30) in, The productive activity consumed by producer-consumer i at time t.

[0077] (2) Electricity restrictions: (31) (3) Grid feed-in limitations: (32) in, For photovoltaic power generated by consumer i in time slot t, This represents the upper limit for PV energy feed-in to the public power grid.

[0078] (4) Power flow equation constraints: For each node , The node voltage at time t is a relaxable variable. and This represents the active and reactive power injected into the node at time t. This represents the active power of the distributed resources injected by the node; for each line li-j, and These represent the line conductance and susceptance, respectively. This represents the active power flowing through the line at time t. This represents the phase difference between the two ends of the line at time t. Let be the phase difference between the two ends of the line at time t. Let i be the set of all nodes connected to node i. This paper uses a polar coordinate form for the power flow calculation model: (33) (34) (35) in, This represents the photovoltaic output value of photovoltaic access node i at time t; This represents the user load at node i at time t.

[0079] (5) Operational safety constraints: (36) in, , Representing distribution network nodes The upper and lower limits of the voltage; , Representing distribution network nodes The upper and lower limits of the current; , Representing branch roads The current and the maximum allowable value.

[0080] (6) Inverter power constraints: (37) (38) (39) (40) in, and These are the rated power of the PV inverter for each individual prosumer, and the tangent value corresponding to the inverter power factor adjustment limit; and These are the minimum and maximum values ​​of the inverter's power generation, respectively. and These represent the minimum and maximum load values ​​at time t, respectively.

[0081] (7) OLTC Operating Constraints (41) (42) in, This represents the maximum adjustable gear of the OLTC at time t. This indicates that the reactive power value of node i is adjusted by OLTC. This is the rated power of the OLTC. The reactive power emitted by producer i at time t.

[0082] (8) Capacitor Bank (CB) Operating Constraints (43) in, This represents the maximum number of actions of the v-th capacitor. This indicates the reactive power output during one capacitor switching operation. This indicates that the reactive power value of node i is adjusted by the capacitor bank.

[0083] (9) Energy storage constraints (44) (45) (46) in, Let be the charging and discharging power of the energy storage unit at time t. and These represent the upper limits of the charging and discharging power of the energy storage unit at time t, respectively. and These represent the active and reactive power emitted or absorbed by the energy storage at time t, respectively. The rated capacity of the inverter. Indicates the maximum capacity of the energy storage battery; and and represent the maximum and minimum states of charge of the energy storage battery, respectively. Considering the case where the voltage exceeds the upper limit, the case of the energy storage node discharging is taken.

[0084] Furthermore, regarding the autonomous decision-making framework driven by non-cooperative game theory, game theory is used to model producers and consumers as players who selfishly pursue their own profits through competition. The strategy form of the game is defined as: (47) It consists of the following three elements.

[0085] 1) Player set: Prosumers with flexible loads and PV assets act as participants in the game. Take a finite set N = {1, 2, ..., N}.

[0086] 2) The strategy space of each participant: the strategy space of each participant i It is the set of strategies that a participant can choose. Therefore, the strategy space for all participants is defined as: (48) That is, the Cartesian product of each participant's strategy set, which means that the problem is a standard Nash game, where each player's strategy set is independent.

[0087] 3) Player's loss function: The loss function for each player i is given in the objective function as the total cost for each producer and consumer. .

[0088] As mentioned before, each prosumer's goal is to minimize total cost. In this regard, a suitable solution is NE (Non- ...

[0089] The strategy profile of all participants in a game defined by Definition 1 A solution to a problem called NE, or Nash equilibrium (NEP), is given if and only if the following inequality holds for any participant. Established.

[0090] (49) The definition of Nash equilibrium states that in the defined game, no player i can achieve a better outcome (lower cost) without changing any of the strategies of the other players.

[0091] Example 3, referring to Figures 2-8 In a preferred embodiment, it is proposed that 10 photovoltaic distributed resource entities be connected to a 10kV distribution system feeder. To demonstrate the PCC voltage curves in different scenarios, it is assumed that the OLTC is unavailable throughout the simulation.

[0092] Scenario 1: No adjustment measures are taken. In this scenario, no regulation is involved, thus frequent voltage fluctuations can be observed. The total load, photovoltaic power generation, and net load of all distributed resource entities are as follows: Figure 2 As shown in the figure, photovoltaic output increases sharply between 7:00 and 10:00, then reaches its peak power generation between 10:30 and 14:00. During this period, the output power fluctuates significantly. Finally, the photovoltaic system experiences a sloping production period, typically occurring between 14:00 and 16:30.

[0093] Scenario 2: Single active power regulation The entire process is implemented in a closed-loop manner. Based on the received internal price, and considering predicted photovoltaic power, load consumption, and user willingness, optimal scheduling is performed for each photovoltaic distributed resource entity. For each iteration, the net power of photovoltaic producers may change, generating a new internal price. After a sufficient number of iterations, all photovoltaic distributed resource entities can no longer reduce active power, thus determining the final internal price. All photovoltaic producers and users can benefit from this feasible solution. Otherwise, the energy sharing zone diverges, the energy sharing model has no solution, and photovoltaic producers and users can still trade directly with the grid.

[0094] The 24-hour distributed resource-based main electricity price iteration fluctuation chart is shown below. Figure 3As shown, with the increase of the SDR (Sales and Demands Ratio), both the selling and buying prices exhibit the same trend, decreasing. If the original SDR is large, the internal prices for both selling and buying will be very low, leading them to want to sell less electricity at a low price, while the buying side may want to buy more electricity. Therefore, both sides want to adjust power consumption to minimize costs or maximize profits. Ultimately, adjusting net power consumption will also decrease, which will increase the internal electricity price. Conversely, if the original SDR is small, the dynamic process will be reversed, and the internal price for each time period should eventually be acceptable to all photovoltaic consumers. When the internal price reaches a fixed point, neither side will be willing to adjust power consumption further.

[0095] Using this model and algorithm, the differences between the sell and buy prices between the two iterations are as follows: Figure 4 As shown, the difference between the two iterations converges after dozens of iterations, and the load consumption of photovoltaic power generation users will not change because they have already obtained the minimum cost.

[0096] Scenario 3: Distributed optimization is available, OLTC is unavailable, and traditional DSO is designed with a unified compensation price for the entire network, with all photovoltaic distributed resource entities sharing the same compensation price.

[0097] Scenario 4: Distributed optimization is available, but the OLTC is unavailable. The distributed resource entity's revenue is generated using the method proposed in this paper: the cost required by the OLTC to regulate the voltage to a stable level is sent to the distributed resource entity as its revenue from voltage regulation. Figure 5 Comparison of the benefits of distributed resource entities in Scheme Scenario 3 and Scheme 4.

[0098] Depend on Figure 5 Analysis reveals that under centralized pricing, some nodes experience negative returns due to active power reduction without corresponding compensation. This paper's voltage regulation compensation mechanism considers the autonomy of distributed resource entities in voltage regulation, optimizing their respective reactive power regulation strategies with their own revenue as the objective function. Results show that this scheme delivers high returns for most of the day, with the most significant additional returns occurring around midday during peak photovoltaic power generation. This is because potential overvoltage issues are most severe during this period, resulting in higher profits from voltage regulation compared to other times. In conclusion, this method brings higher collective and individual profits to distributed resource entities.

[0099] Photovoltaic (PV) and energy storage systems play a crucial role in power systems, particularly in voltage regulation and supply-demand balancing. During periods of high PV power generation, output needs to be limited to prevent voltage overshoot, while reactive power regulation is essential to maintain voltage stability during other times of the day. Energy storage systems operate on the principle of "low power generation, high energy storage," increasing voltage at night by discharging excess PV power and absorbing it during the day to prevent voltage overshoot. When actual PV output falls short of forecasts, both active and reactive power regulation are affected, especially at night when the reactive power regulation of PV inverters reaches its maximum capacity. Energy storage systems also need to adjust to PV fluctuations to maintain voltage stability, particularly during extreme PV peak shaving, when energy storage needs to release power to maintain voltage, potentially leading to increased active power regulation and decreased reactive power regulation. While the uncertainties of PV can cause fluctuations in the energy storage regulation curve, overall, energy storage regulation is relatively stable during the day, especially under conditions such as cloudy weather, where active power regulation may increase significantly.

[0100] like Figure 6 , Figure 7 , Figure 8 As shown, the voltage curves of the feeder terminal are compared under four different scenarios where the OLTC is consistently unavailable. Throughout the simulation, the SVR was disabled to demonstrate the voltage curves without conventional regulation measures. Without appropriate incentives, in Scenario 1, the PCC voltage gradually rises in the morning due to increased photovoltaic power generation. Conversely, the PCC voltage amplitude in Scenario 1 decreases significantly from 14:00 to 18:00. At 00:00, as peak load approaches and photovoltaic power output decreases, the proposed model can suppress voltage fluctuations within safe limits, demonstrating that the proposed model can mitigate voltage violations for most of the day without any SVR operation. In summary, the proposed scheme offers more economic benefits to distributed resource providers while maintaining a similar capacity to provide voltage regulation services.

[0101] Example 4, refer to Figure 9 This embodiment also provides a multi-agent active-reactive power cooperative optimization system for distribution networks based on non-cooperative game theory, including: The first model building module is used to calculate the supply-demand ratio parameters of the distribution network based on the net power, photovoltaic power generation output and energy storage charging and discharging status of each producer and consumer in the time slot, and to generate the purchase price and sales price signals that change monotonically with the supply and demand status, which is denoted as the dynamic supply-demand ratio electricity price model. The second model building module is used to equate the feeder to a resistor-inductor-capacitor series structure, determine the voltage setting value of the common coupling point and the tap position of the on-load tap changer based on historical operating data and power prediction information, calculate the difference in on-load tap changer operation cost under the condition of whether or not producers and consumers participate in voltage regulation, and allocate the difference as the revenue or penalty of voltage regulation service to each producer and consumer, which is called the producer and consumer voltage regulation service revenue evaluation model. The third model building module is used to construct an active-reactive power co-optimization model, which aims to minimize the total operating cost of each producer and consumer. The collaborative optimization module is used to perform multi-entity active and reactive power collaborative optimization of the distribution network based on the dynamic supply and demand ratio electricity price model, voltage regulation service revenue evaluation model, and active and reactive power collaborative optimization model.

[0102] The above-mentioned unit modules can be embedded in the processor of the electronic device in hardware form or independent of it, or they can be stored in the memory of the electronic device in software form, so that the processor can call and execute the corresponding operations of the above modules.

[0103] This embodiment also provides an electronic device, which can be a terminal, and its internal structure diagram can be as follows: Figure 9 As shown, the electronic device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a multi-agent active and reactive power collaborative optimization method for power distribution networks based on non-cooperative game theory. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the device's casing, or an external keyboard, touchpad, or mouse.

[0104] This embodiment also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, it performs the following steps: Based on the net power, photovoltaic power generation output, and energy storage charging and discharging status of each producer and consumer within the time slot, the supply-demand ratio parameters of the distribution network are calculated, and the purchase price and sales price signals that change monotonically with the supply and demand status are generated, which are denoted as the dynamic supply-demand ratio electricity price model. The feeder is equivalent to a resistor-inductor-capacitor series structure. Based on historical operating data and power prediction information, the voltage setting value of the common coupling point and the tap position of the on-load tap changer are determined. The difference in on-load tap changer operation cost under the condition of whether or not producers and consumers participate in voltage regulation is calculated. This difference is then allocated to each producer and consumer as the revenue or penalty of voltage regulation service. This is called the producer and consumer voltage regulation service revenue evaluation model. Construct an active-reactive power co-optimization model, with the goal of minimizing the total operating cost of each producer and consumer. Active and reactive power coordinated optimization of distribution networks by multiple stakeholders is carried out based on dynamic supply and demand ratio electricity price model, voltage regulation service revenue evaluation model and active and reactive power coordinated optimization model.

[0105] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0106] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the invention.

[0107] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A multi-agent active and reactive power collaborative optimization method for distribution networks based on non-cooperative game theory, characterized in that, include: Based on the net power, photovoltaic power generation output, and energy storage charging and discharging status of each producer and consumer within the time slot, the supply-demand ratio parameters of the distribution network are calculated, and the purchase price and sales price signals that change monotonically with the supply and demand status are generated, which are denoted as the dynamic supply-demand ratio electricity price model. The feeder is equivalent to a resistor-inductor-capacitor series structure. Based on historical operating data and power prediction information, the voltage setting value of the common coupling point and the tap position of the on-load tap changer are determined. The difference in on-load tap changer operation cost under the condition of whether or not producers and consumers participate in voltage regulation is calculated. This difference is then allocated to each producer and consumer as the revenue or penalty of voltage regulation service. This is called the producer and consumer voltage regulation service revenue evaluation model. Construct an active-reactive power coordinated optimization model, the objective of which is to minimize the total operating cost of each producer and consumer; Based on the aforementioned dynamic supply-demand ratio electricity price model, voltage regulation service revenue evaluation model, and active-reactive power coordinated optimization model, multi-entity active-reactive power coordinated optimization of the distribution network is carried out.

2. The method for multi-agent active and reactive power collaborative optimization of distribution networks based on non-cooperative game theory as described in claim 1, characterized in that, Also includes: A non-cooperative game-based autonomous decision-making framework is constructed, which takes producers and consumers with photovoltaic power generation and flexible load capabilities as game participants, and uses their respective active power output, reactive power output and energy storage scheduling plans as strategy variables. Using total operating cost as the loss function, each participant independently optimizes and updates their strategy, driven by price and revenue signals generated by the operator based on the dynamic supply-demand ratio electricity price model and voltage regulation service revenue evaluation model, until the strategy converges. It outputs the reactive power setpoints, active power output plans, energy storage charging and discharging scheduling schemes, and control commands for on-load tap changers and capacitor banks for each producer and consumer.

3. The method for multi-agent active and reactive power collaborative optimization of distribution networks based on non-cooperative game theory as described in claim 2, characterized in that, In the dynamic supply-demand ratio electricity price model, when the supply-demand ratio is less than 1, the purchase price of electricity is increased and the sales price of electricity is decreased to incentivize local power generation. When the supply-demand ratio is greater than 1, reduce the purchase price of electricity and increase the sales price of electricity to promote electricity consumption. The electricity purchase price and electricity sales price are monotonically adjusted in the direction of change of the supply-demand ratio within adjacent time slots; The supply-demand ratio is calculated based on the positive and negative decomposition of the net power of all producers and consumers within the time slot, where the net power deficit is the sum of the positive net power and the net surplus is the sum of the negative net power. The net power of a single producer-consumer is the algebraic difference between its adjusted power consumption and photovoltaic power generation, and it is also included in the equivalent treatment of energy storage charging as power consumption and energy storage discharging as power generation.

4. The method for multi-agent active and reactive power collaborative optimization of distribution networks based on non-cooperative game theory as described in claim 3, characterized in that, The consumer-driven blood pressure regulation service revenue assessment model includes: A damped second-order dynamic voltage model is established based on the resistance, reactance and capacitance parameters of the feeder, and adjacent time slots are discretized. The voltage setpoint of the common coupling point and the tap position of the on-load tap changer are estimated using load and photovoltaic power generation forecast data. The total operating cost is calculated based on the number of taps of the on-load tap changer and the cost of a single tap operation, thus obtaining the cost difference when there is no consumer involved in voltage regulation. Producers and consumers receive positive benefits when they participate in voltage regulation to reduce the number of on-load tap changer operations or the adjustment range. If this leads to voltage deterioration, a penalty is imposed. The benefits or penalties are distributed according to the proportion of reactive power output of each producer and consumer.

5. The method for multi-agent active and reactive power collaborative optimization of distribution networks based on non-cooperative game theory as described in claim 4, characterized in that, The total operating cost includes power utility loss, power generation cost, output adjustment cost, inverter reactive power cost, on-load tap changer operation cost, and capacitor bank operating cost. The reactive power cost of the inverter includes three parts: the power loss cost corresponding to the additional active power loss caused by reactive power output, the life loss cost corresponding to the shortened equipment life due to the increase in apparent power, and the opportunity cost corresponding to the reduction in available active power due to reactive power occupying inverter capacity.

6. The method for multi-agent active and reactive power collaborative optimization of distribution networks based on non-cooperative game theory as described in claim 5, characterized in that, The constraints of the active-reactive power coordinated optimization model include: Power flow equations of a distribution network described in polar coordinates; Upper and lower limits of node voltage and upper limit of branch current; Active power output, reactive power output, and power factor boundary limits of the inverter; Limitations on the tap position range and number of operations of on-load tap changers; Capacitor bank's single reactive power output capability and switching frequency limitations; Upper limit of charging and discharging power, upper and lower limits of state of charge, and rated capacity limit of energy storage system; User-side power comfort requirements and upper limit constraints on the amount of electricity sent to the main grid.

7. The method for multi-agent active and reactive power collaborative optimization of distribution networks based on non-cooperative game theory as described in claim 6, characterized in that, The non-cooperative game autonomous decision-making framework is solved using an optimal response iteration method. In each iteration, the operator recalculates the supply-demand ratio and updates the electricity price signal based on the latest net power, and at the same time recalculates the on-load tap changer operation cost and updates the voltage regulation revenue signal based on the latest voltage status. The system is considered to have reached Nash equilibrium when the changes in the strategy variables of all producers and consumers are less than the preset tolerance.

8. A multi-agent active-reactive power cooperative optimization system for distribution networks based on non-cooperative game theory, employing the method described in any one of claims 1 to 7, characterized in that, include: The first model building module is used to calculate the supply-demand ratio parameters of the distribution network based on the net power, photovoltaic power generation output and energy storage charging and discharging status of each producer and consumer in the time slot, and to generate the purchase price and sales price signals that change monotonically with the supply and demand status, which is denoted as the dynamic supply-demand ratio electricity price model. The second model building module is used to equate the feeder to a resistor-inductor-capacitor series structure, determine the voltage setting value of the common coupling point and the tap position of the on-load tap changer based on historical operating data and power prediction information, calculate the difference in on-load tap changer operation cost under the condition of whether or not producers and consumers participate in voltage regulation, and allocate the difference as the revenue or penalty of voltage regulation service to each producer and consumer, which is called the producer and consumer voltage regulation service revenue evaluation model. The third model building module is used to construct an active-reactive power coordinated optimization model, which aims to minimize the total operating cost of each producer and consumer. The collaborative optimization module is used to perform multi-entity active and reactive power collaborative optimization of the distribution network based on the dynamic supply-demand ratio electricity price model, voltage regulation service revenue evaluation model, and active-reactive power collaborative optimization model.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of any one of claims 1 to 7 for a multi-subject active-reactive power cooperative optimization method for distribution networks based on non-cooperative game theory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the multi-subject active and reactive power collaborative optimization method for distribution networks based on non-cooperative game theory as described in any one of claims 1 to 7.

Images