Power system frequency distributed game optimization control method considering unit dynamics

Abstract

This application discloses a power system frequency distributed game-theoretic optimization control method that takes into account generator unit dynamics. The method includes: collecting regional frequency deviations and tie-line power from adjacent generation areas using a measurement unit, and calculating regional control deviations based on these deviations and power; obtaining the nodal price of the generation area; inputting the regional control deviation and nodal price into a preset market game model to generate a power reference value, thereby coupling the power reference value and the regional control deviation in real time; and generating control commands corresponding to the generator units based on the power reference value using a preset generator unit generation response model, and then adjusting the frequency of the generator units based on these control commands. In this embodiment, the generator units can respond in real time to control commands from economic dispatch.

Description

Technical Field

[0001] This application relates to, but is not limited to, the field of power grid frequency control technology, and in particular to a power system frequency distributed game optimization control method that takes into account the dynamics of generating units. Background Technology

[0002] The power system is a fundamental infrastructure of modern society, and its safe and stable operation is crucial to the national economy and people's lives. Within the power system, frequency is a vital indicator reflecting the balance between active power supply and demand. When the load suddenly increases, the generating capacity is insufficient to meet the load demand, and the system frequency will decrease; conversely, when the load suddenly decreases, the system frequency will increase. Excessive deviation of the frequency from the rated value can lead to equipment damage, system disconnection, and even large-scale power outages.

[0003] Existing power system frequency control is often achieved through Automatic Generation Control (AGC). Traditional AGC is based on the Area Control Deviation (ACE) signal, which comprehensively considers frequency deviation and tie-line power deviation. It responds to load changes by adjusting the power reference value of generator sets, bringing the ACE to zero, thereby restoring the frequency and maintaining inter-area power exchange at the planned value.

[0004] However, with the large-scale integration of renewable energy into inverter interfaces at a high proportion, the system's rotational inertia has decreased significantly, exacerbating the amplitude and frequency of frequency fluctuations. In the aforementioned technical scenario, simply reducing the ACE (Average Frequency Interval) to zero through frequency control is no longer sufficient to meet the operational needs of the power system.

[0005] Existing power systems must not only ensure physical safety and stability but also maximize economic benefits, i.e., implement economic dispatch to minimize total generation costs. Current frequency control frameworks separate economic dispatch and AGC (Automatic Generation Control) into different time scales: economic dispatch is calculated offline in minutes and then the base point is issued, while AGC handles real-time frequency control in seconds. This separation of time scales is feasible in traditional power grids; however, in the aforementioned high-proportion renewable energy power grid technology scenario, continuous source-load imbalance forces generating units to frequently deviate from the economically optimal dispatch point, resulting in severe economic penalties. This leads to frequent fluctuations in the frequency control of the units involved in the power system, preventing them from responding to economic dispatch in real time. Summary of the Invention

[0006] This application provides a power system frequency distributed game optimization control method that takes into account the dynamics of generating units, enabling generating units to respond in real time to control commands from economic dispatch.

[0007] In a first aspect, embodiments of this application provide a power system frequency distributed game optimization control method considering unit dynamics, applied to a new energy power system. The new energy power system includes multiple power generation areas connected by interconnecting lines. Each power generation area includes generator units, a measurement unit, and a distributed market coupling controller. The distributed market coupling controller has a built-in preset unit generation response model and a preset market game model, including: The measurement unit collects the regional frequency deviation and tie-line power of adjacent power generation areas, and calculates the regional control deviation based on the regional frequency deviation and the tie-line power. Obtain the nodal electricity price corresponding to the power generation area; The regional control deviation and the node electricity price are input into the preset market game model to generate a power reference value, so that the power reference value and the regional control deviation are coupled in real time; The preset generator set power generation response model generates control commands corresponding to the generator set based on the power reference value, and then adjusts the frequency of the generator set based on the control commands.

[0008] Secondly, according to an embodiment of this application, a power system frequency control device is provided, which is applied to a new energy power system. The new energy power system includes multiple power generation areas connected by interconnecting lines. Each power generation area includes generator sets, a measurement unit, and a distributed market coupling controller. The distributed market coupling controller has a preset generator set generation response model and a preset market game model, including: A preprocessing module is used to collect the regional frequency deviation and tie-line power of adjacent power generation areas through the measurement unit, and calculate the regional control deviation based on the regional frequency deviation and the tie-line power. The data acquisition module is used to obtain the nodal electricity price corresponding to the power generation area; A market calculation module is used to input the regional control deviation and the node electricity price into the preset market game model to generate a power reference value, so that the power reference value and the regional control deviation are coupled in real time; A physical control module is used to generate control commands corresponding to the generator set based on the power reference value using the preset generator set power generation response model, and then adjust the frequency of the generator set based on the control commands.

[0009] Thirdly, an electronic device provided according to an embodiment of this application includes: At least one processor; At least one memory for storing at least one program; When at least one of the programs is executed by at least one of the processors, the power system frequency distributed game optimization control method taking into account unit dynamics, as described in any of the first aspects, is implemented.

[0010] Fourthly, according to the embodiments of the application, a computer-readable storage medium is provided, storing computer-executable instructions, which are used to execute the power system frequency distributed game optimization control method considering unit dynamics as described in any of the first aspects.

[0011] In summary, the power system frequency distributed game optimization control method considering unit dynamics in the above embodiments of this application includes: collecting regional frequency deviation and tie-line power of adjacent power generation areas through a measurement unit, and calculating regional control deviation based on regional frequency deviation and tie-line power; obtaining the node price corresponding to the power generation area; inputting the regional control deviation and node price into a preset market game model to generate a power reference value, so that the power reference value and the regional control deviation are coupled in real time; generating control commands corresponding to the generator set based on the power reference value through a preset unit generation response model, and then adjusting the frequency of the generator set based on the control commands. The preset market game model in this application embodiment can use regional control deviation as a penalty driver, so that regional control deviation can be directly fed back to market decision-making. This means that frequency deviation can be fed back to market decision-making in real time at the second level. Economic optimization and physical frequency regulation evolve synchronously on the same time scale, thereby enabling the generated power reference value and regional control deviation to be coupled in real time. Based on the real-time coupling of power reference value and regional control deviation, the preset unit generation response model describes the dynamic response of the unit to the control command of the power reference value. Then, based on the control command, the frequency of the generator unit is adjusted, and the generator unit can respond to the control command of economic dispatch in real time. Attached Figure Description

[0012] Figure 1 This is a flowchart of the steps of a power system frequency distributed game optimization control method that takes into account unit dynamics, provided in one embodiment of this application; Figure 2 This application provides a hardware architecture diagram of a new energy power system according to one embodiment; Figure 3 This application provides a flowchart of steps in one embodiment; Figure 4 This is a schematic diagram of a software virtual functional module of a distributed market coupling controller provided in one embodiment of this application; Figure 5 This application provides a schematic diagram of the dynamic response curve of frequency deviation under a step disturbance in one embodiment; Figure 6 This application provides a schematic diagram of the convergence curve of node electricity price and strategy bidding under a step disturbance in one embodiment; Figure 7 This is a hardware schematic diagram of an electronic device provided in one embodiment of this application. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0014] It is understandable that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, or the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0015] The power system is a fundamental infrastructure of modern society, and its safe and stable operation is crucial to the national economy and people's lives. Within the power system, frequency is a vital indicator reflecting the balance between active power supply and demand. When the load suddenly increases, the generating capacity is insufficient to meet the load demand, and the system frequency will decrease; conversely, when the load suddenly decreases, the system frequency will increase. Excessive deviation of the frequency from the rated value can lead to equipment damage, system disconnection, and even large-scale power outages.

[0016] Existing power system frequency control is often achieved through Automatic Generation Control (AGC). Traditional AGC is based on the Area Control Deviation (ACE) signal, which comprehensively considers frequency deviation and tie-line power deviation. It responds to load changes by adjusting the power reference value of generator sets, bringing the ACE to zero, thereby restoring the frequency and maintaining inter-area power exchange at the planned value.

[0017] However, with the large-scale integration of renewable energy into inverter interfaces at a high proportion, the system's rotational inertia has decreased significantly, exacerbating the amplitude and frequency of frequency fluctuations. In the aforementioned technical scenario, simply reducing the ACE (Average Frequency Interval) to zero through frequency control is no longer sufficient to meet the operational needs of the power system.

[0018] Existing power systems not only need to ensure physical safety and stability but also need to maximize economic benefits, i.e., to implement economic dispatch to minimize total generation costs. Current frequency control frameworks separate economic dispatch and AGC (Automatic Generation Control) into different time scales: economic dispatch is calculated offline in minutes and then the base point is issued, while AGC is responsible for real-time frequency control in seconds. In traditional power grids, this separation of time scales is feasible; however, in the aforementioned high-proportion renewable energy power grid technology scenario, continuous source-load imbalance forces generating units to frequently deviate from the economically optimal dispatch point, resulting in severe economic penalties. This leads to frequent fluctuations in the frequency control of the units affected by the power system, preventing them from responding to economic dispatch in real time.

[0019] Based on this, the embodiments of this application provide a power system frequency distributed game optimization control method that takes into account the dynamics of the generating units, so that the generating units can respond to the control commands of economic dispatch in real time.

[0020] This application provides a power system frequency distributed game optimization control method that takes into account unit dynamics. It is applied to a new energy power system, which includes multiple power generation areas connected by tie lines. Each power generation area includes generator sets, measurement units, and a distributed market coupling controller. The distributed market coupling controller has a built-in preset generator generation response model and a preset market game model.

[0021] Understandably, traditional centralized control systems require each region to report its status to the global coordination center in real time, which can easily lead to communication bottlenecks. This application's embodiment designs a distributed control law that relies solely on information exchange between adjacent power generation regions. Node electricity prices, bids, and instruction generation are all completed by the local controller, and the power flow dynamics of the virtual tie line only need to know the node prices of adjacent interconnected regions. This architecture avoids the aggregation of massive amounts of data to the central dispatch center. Furthermore, due to the algorithm's topological connectivity requirements, the failure of any single node in the network will not cause the collapse of the entire control system; the remaining nodes can still reach a new equilibrium through the remaining connectivity topology. This significantly reduces the construction and maintenance costs of the wide-area communication network, eliminates the need for a centralized dispatch center, reduces communication and computational burdens, and improves system reliability.

[0022] For example, refer to Figure 2 As shown, there are four interconnected control power generation zones (in Figure 2The power generation areas are displayed as regions (taking regions 1 to 4 as examples). Each region contains generator sets and governors, measurement units, and a distributed market coupling controller. The distributed market coupling controller has a built-in preset market game model. In the cross-regional topology, the generator sets and governors of adjacent regions are interconnected through physical power transmission lines, forming a high-voltage physical network for transmitting active power. The distributed market coupling controllers of each region are connected through an information communication network built via industrial Ethernet or fiber optics. This system achieves strict decoupling and mapping at the physical entity level and the information interaction level. Under this hardware framework, the underlying measurement units collect the physical operation data of the local generator sets in real time and upload it to the local controller. The local controller interacts bidirectionally with the controllers of adjacent regions through the information communication network to obtain game data such as electricity prices and virtual power flow. After distributed computing, the local controller outputs analog or digital control commands to the governors of the downstream generator sets, thus forming a closed-loop hardware control architecture.

[0023] Reference Figure 1 As shown, the power system frequency distributed game optimization control method considering unit dynamics includes, but is not limited to, the following steps: Step S100: The regional frequency deviation and tie-line power of adjacent power generation areas are collected by the measurement unit, and the regional control deviation is calculated based on the regional frequency deviation and tie-line power.

[0024] For example, the regional control deviation (ACE) is a comprehensive indicator that measures regional frequency deviation and tie-line power deviation. In other words, the regional control deviation represents the real-time imbalance between the power generation capacity and load demand in the power generation area.

[0025] Specifically, the regional control deviation satisfies the following expression: ; in, This is the frequency deviation coefficient. For frequency deviation, This refers to the power deviation of the tie line.

[0026] In this embodiment, the controller of each power generation area only communicates with adjacent power generation areas that are physically connected, and obtains the node electricity price of the adjacent power generation areas.

[0027] Step S110: Obtain the node electricity price corresponding to the power generation area.

[0028] It is understandable that there is a communication network between power generation areas, and the node price corresponding to each power generation area is obtained through the communication network at the current moment. The node price is the price determined at the previous moment.

[0029] Step S120: Input the regional control deviation and nodal electricity price into the preset market game model to generate the power reference value, so that the power reference value and the regional control deviation are coupled in real time.

[0030] Understandably, existing frequency control frameworks separate economic dispatch and AGC (Automatic Generation Control) into different time scales: economic dispatch is calculated offline in minutes and then the base point is issued, while AGC is responsible for real-time frequency control in seconds. However, in the aforementioned high-proportion renewable energy grid technology scenario, continuous source-load imbalance forces generator units to frequently deviate from the economically optimal dispatch point, resulting in severe economic penalties and thus frequent fluctuations in the frequency control of the units considered in the power system. The pre-set market game model in this application embodiment can use regional control deviation as a penalty driver, allowing regional control deviation to be directly fed back to market decisions. This means that frequency deviation can be fed back to market decisions in real time in seconds. Economic optimization and physical frequency regulation evolve synchronously on the same time scale, thereby enabling the generated power reference value and regional control deviation to be coupled in real time, providing a foundation for subsequent real-time response control of generator units based on power reference values.

[0031] It is understandable that in the actual electricity market, each generating unit in a power generation area acts as a selfish, rational economic agent, competing on price (submitting bids) to vie for a share of power generation in order to maximize its own profits. This is known as Bertrand competition, a phenomenon that falls under non-cooperative game theory. Therefore, in the context of the electricity market, the pre-set market game model in this application uses a market game framework to describe the non-cooperative game phenomenon in the actual electricity market. By solving the pre-set market game model, at the Nash equilibrium point, each generating unit in the power generation area adopts its optimal strategy. Under the condition that the Nash equilibrium coincides with the solution for maximizing social welfare, an effective Nash equilibrium is reached, yielding a power reference value. At this point, the total power generation cost in the new energy power system is minimized, while simultaneously satisfying the supply and demand balance constraint.

[0032] Understandably, Nash equilibrium refers to a non-cooperative game in which each participant, given the strategies of other participants, chooses the strategy that is optimal for themselves, and no participant has an incentive to unilaterally change their own strategy.

[0033] Step S130: Based on the power reference value, a control command corresponding to the generator set is generated by the preset generator set power generation response model, and then the frequency of the generator set is adjusted based on the control command.

[0034] It is understandable that, based on the real-time coupling of power reference value and regional control deviation, the embodiments of this application describe the dynamic response of the generator set to the control command of the power reference value through a preset generator set generation response model, and then control the generator set based on the control command, enabling the generator set to respond to the control command in real time. Thus, by coupling the power reference value and regional control deviation, a deep real-time coupling between market clearing and the physical dynamic response of the generator set is achieved.

[0035] In some embodiments, the power generation area further includes a speed governor corresponding to the generator set, and adjusting the frequency of the generator set based on control commands includes: acquiring control commands output by the PLC controller; adjusting the speed governor based on the control commands to control the frequency of the generator set.

[0036] It is understood that the embodiments of this application receive the latest active power reference value, and then convert it into AGC control instructions (such as analog voltage / current signals after D / A conversion) that can be recognized by the underlying hardware through the PLC controller, and drive the underlying speed governor to perform physical mechanical actions, thereby controlling the frequency of the generator set and thus controlling the power of the generator set.

[0037] In some embodiments, the preset market game model is determined through the following steps: constructing an optimization objective function that minimizes the generation cost of the new energy power system, and introducing virtual power flow to construct the constraints corresponding to the optimization objective function, wherein the virtual power flow is driven by the difference in nodal electricity prices between adjacent generation areas; using the Lagrange multiplier method to relax the constraints into the optimization objective function, and introducing regional frequency deviation as a penalty driving term to construct the augmented Lagrange function of the preset market game model.

[0038] It is understood that the embodiments of this application construct an optimization objective function for minimizing the generation cost of the new energy power system based on the strategy quotations and power reference values ​​corresponding to all generation areas of the new energy power system. The optimization objective function satisfies the following expression: ; in, For strategic pricing, This is a power reference value; The virtual power flow in this embodiment is driven by the difference in nodal electricity prices between power generation regions, which is used to characterize the willingness of economic power exchange between power generation regions. The introduction of virtual power flow allows the physical power flow to be non-zero in steady state, thereby allowing economic power exchange between regions. The virtual power flow is introduced to construct the constraints corresponding to the optimization objective function. The constraints include (1) the control deviation needs to be balanced with the virtual power flow, (2) the power reference value cannot be negative, that is, the power generation power cannot be negative, and (3) the virtual power flow cannot exceed the physical capacity of the generator set, that is, the virtual power flow is limited. The constraints satisfy the following expression: ; in, This represents the regional control deviation corresponding to the j-th power generation area. For the virtual trend, This is the reference power corresponding to the j-th power generation area. The physical capacity of the motor unit, This represents the virtual power flow between the i-th power generation region and the j-th power generation region; To transform constrained optimization into unconstrained optimization, Lagrange multipliers and penalty terms are introduced. The Lagrange multiplier method is used to relax the constraints into the objective function, and regional frequency deviation is introduced as a penalty driving term to construct the augmented Lagrange function of the pre-defined market game model.

[0039] In some embodiments, the preset market game model satisfies the following expression: ; in, To augment the Lagrange function, Let $j$ be the nodal price corresponding to the $j$-th power generation region. For the virtual trend, This represents the regional control deviation corresponding to the j-th power generation area. This is the reference power corresponding to the j-th power generation area. This is the strategy bid corresponding to the j-th power generation area.

[0040] Understandably, traditional methods suffer from the drawback of time-scale separation. This invention directly uses system control deviation as the penalty driving term for multiplier updates. When a high proportion of renewable energy access brings high-frequency random disturbances, market prices can quickly guide the flexible adjustment resources within the system to immediately redistribute economic power without waiting for long-cycle scheduling instructions. The embodiments of this application significantly improve the system's dynamic adaptability to high-frequency fluctuations from renewable energy, maintaining strict frequency stability while preventing units from deviating from their economic optimum for extended periods, thus improving overall economic efficiency. By breaking down the hierarchical isolation between minute-level economic scheduling and second-level automatic generation control, market-level price updates, virtual power flow, and physical-level frequency deviations are directly coupled into the same set of differential equations, enabling economic price signals to respond instantaneously to frequency fluctuations on a second-level cycle.

[0041] In some embodiments, inputting regional control deviation and nodal electricity price into a preset market game model to generate power reference value includes: using an augmented Lagrangian function with a primal-dual algorithm to iteratively update the virtual power flow, power reference value, and strategy bid corresponding to the power generation region as primal variables; and simultaneously using regional control deviation as a penalty driver to iteratively update the nodal electricity price as the dual variable until convergence to the Nash equilibrium point to obtain the power reference value.

[0042] For example, the power reference value is determined by the following steps: (I) Solving the augmented Lagrangian function using the primal-dual algorithm yields the following expression: ; in, To Find the partial derivative. For the virtual trend, Let J be the regional control deviation of the j-th power generation area. To Find the partial derivative. For the strategy bid corresponding to the j-th power generation area, Let $j$ be the nodal price corresponding to the $j$-th power generation region. To Find the partial derivative. Let $i$ be the nodal electricity price corresponding to the $i$-th power generation region. As a virtual trend, The area control deviation for the i-th power generation area; (ii) Next, based on the above expressions, the dynamic update expressions for nodal pricing, strategic pricing, virtual power flow, and power reference values ​​are derived; 1) The dynamic update expression for nodal electricity prices is shown below: ; in, It is a time constant. Let be the derivative of the nodal price corresponding to the j-th power generation region. For the virtual trend, For the regional control deviation of the j-th power generation region, the nodal price is used as a Lagrange multiplier. The nodal price increases dually, that is, the nodal price is updated along the direction of constraint violation. Node prices rise when there is a power shortage; when When there is excess power, node prices decrease; 2) The dynamic update expression for the virtual trend is shown below: ); in, It is a time constant. Let be the derivative of the virtual power flow between the i-th power generation region and the j-th power generation region. Let $i$ be the nodal electricity price corresponding to the $i$-th power generation region. Let $j$ be the nodal price corresponding to the $j$-th power generation region. For the virtual trend, Let be the regional control deviation for the i-th power generation area. For the virtual trend, For the j-th power generation region, the regional control deviation is denoted as ; the virtual power flow is driven by the price difference between regions, when > At that time, power tends to flow from the low-price region j to the high-price region i. Increasing the augmented term accelerates convergence, leading to price uniformity in steady state. That is, price uniformity is derived from the dynamics of the virtual current flow, not as an assumption. 3) The dynamic update expression for the power reference value is shown below: ; in, It is a time constant. The derivative of the power reference value corresponding to the j-th power generation area. For the strategy bid corresponding to the j-th power generation area, Let $j$ be the nodal price corresponding to the $j$-th power generation region. As a virtual trend, Let be the regional control deviation for the j-th power generation region; the power reference value is driven by both the price-bid difference and the constraint deviation. When the market price is higher than the bid price, power generators have an incentive to increase power generation, which introduces a secondary penalty term for the regional control deviation. The constraint term provides a direct response to ACE. 4) The dynamic update mechanism for the generator's strategy pricing is based on the principle of maximizing individual profits. The dynamic update expression for the strategy pricing is shown below: ; in, It is a time constant. Let be the derivative of the strategy bid corresponding to the j-th power generation region. Let be the gradient of the conjugate of the cost function. The bid evolves dynamically according to the gradient of the conjugate of the cost function, so that the bid equals the marginal cost in the dynamic steady state, reflecting the game-theoretic property that strategic bids eventually converge to the true cost.

[0043] Traditional methods suffer from the drawback of time-scale separation. This application directly uses the system's regional control deviation as the penalty driver for multiplier updates. That is, the driving force for price updates is the local physical power supply-demand imbalance (ACE), and power command updates are also directly driven by the ACE constraint deviation. When a high proportion of renewable energy access brings high-frequency random disturbances, market prices can quickly guide the system's flexible adjustment resources to immediately redistribute economic power without waiting for long-cycle dispatch commands. Thus, because the regional control deviation, representing physical deviation, is embedded in the market dynamic equations in real time, economic dispatch is no longer calculated offline on a minute-by-minute basis and then issued, but responds synchronously with frequency deviations on a second-by-second basis, thus adapting to the high-frequency fluctuations brought by renewable energy. Therefore, by breaking the hierarchical isolation between minute-by-minute economic dispatch and second-by-second automatic generation control, the market-layer price updates, virtual power flow, and physical-layer frequency deviations are directly coupled into the same differential equation set, enabling economic price signals to respond instantaneously to frequency fluctuations on a second-by-second basis. Thus, the embodiments of this application significantly improve the system's dynamic adaptability to high-frequency fluctuations from renewable energy, maintaining strict frequency stability while preventing units from deviating from the economic optimum for extended periods, thereby improving overall economic efficiency.

[0044] Existing fully cooperative models are difficult to implement in real markets. This application, however, uses a dynamic mathematical mechanism to constrain bidding. If a generator deliberately deviates from its true cost and submits a false bid, the allocated electricity it receives under steady-state conditions will result in a loss of its own profit. This mathematical characteristic forces strategic bidding to naturally converge to the true marginal cost during evolution. By introducing a dynamic equation for generator strategic bidding based on gradient descent and projection operators, this application allows each unit to participate in the game as a rational economic agent pursuing profit maximization, without having to directly report the true private generation cost function. This successfully protects the generator's business secrets, achieves incentive compatibility in the design of the electricity market mechanism, and solves the pain point that existing methods cannot be promoted in deregulated electricity markets.

[0045] In some embodiments, the power generation area also includes a PLC controller corresponding to the generator set. The PLC controller has a built-in preset generator set power generation response model. The preset generator set power generation response model is determined by the following steps: obtaining a preset swing equation and a preset tie line equation; constructing a governor dynamic equation describing the inertial response process of mechanical power to power reference value based on the mechanical power of the generator set rotor, the rotor angular frequency deviation, and the droop coefficient, thereby obtaining the preset generator set power generation response model.

[0046] Understandably, existing technologies often overlook the physical delay of generators, leading to overly idealized control clearing. This application pre-calculates the physical inertia of the unit, avoiding misalignment between control commands and execution capabilities caused by issuing overly aggressive dispatch instructions. The smoothness of the commands ensures that the system does not experience severe power overshoot during regulation. Thus, by fully incorporating the generator dynamic response element, including the governor time constant, and the primary frequency regulation term into the continuous-time domain Lyapunov and game-theoretic clearing mathematical model, the upper-level distributed controller can perceive and consider the physical delay and damping characteristics of the underlying mechanical power in real time. Therefore, the embodiments of this application fundamentally eliminate the severe transient frequency oscillations caused when simplified models are directly applied to actual units, significantly reducing mechanical fatigue wear of power generation equipment and ensuring the safe and stable operation of the power grid.

[0047] It is understood that the embodiments of this application describe the electromechanical dynamics of the generator set in the power generation area through a preset swing equation, and describe the interconnection power between different power generation areas through a preset tie line equation. Based on the mechanical power of the generator set rotor, the rotor angular frequency deviation, and the droop coefficient, a governor dynamic equation describing the inertial response process of mechanical power to the power reference value is constructed. The primary frequency regulation term provides a fast response to the frequency deviation, thereby enabling the control algorithm to fully perceive the physical execution lag effect of the unit, and thus obtain the preset unit power generation response model.

[0048] In some embodiments, the preset unit power generation response model satisfies the following expression: ; in, The inertia coefficient, For angular frequency deviation, For mechanical power, For load power, The damping coefficient is... The net power of the connecting lines flowing out of the power generation area, The time constant of the speed controller, This is a power reference value. This is the adjustment coefficient. Let j be the equivalent phase angle of the j-th power generation region. The angular frequency deviation corresponding to the j-th power generation area is... Let be the synchronization coefficient of the tie line between the i-th power generation area and the j-th power generation area. Let be the equivalent phase angle of the i-th power generation region.

[0049] Understandably, the four continuous-time differential equations of the market layer—node price, strategy bid, virtual power flow, and power reference value—are designed using the primal-dual gradient method. The physical layer's regional control deviation (ACE) is directly embedded as a penalty driver, enabling real-time feedback of frequency deviation to market decisions at the second level. Economic optimization and physical frequency regulation evolve synchronously on the same time scale. Because existing game-theoretic AGC schemes use simplified swing equations, assuming instantaneous command execution, but actual governors exhibit mechanical inertia lag, a time-scale misalignment occurs between commands and execution, leading to frequency oscillations in the units. This application explicitly introduces a complete dynamic equation for power generation, including governor time constants and droop coefficients, into the pre-defined unit generation response model. This allows the upper-level market algorithm to perceive the actual response speed and damping of the underlying units at each iteration, thus avoiding overly aggressive dispatch commands. This application utilizes a dual design of complete physical dynamic modeling and deep real-time coupling between the physical and market layers, ensuring that the energy of the closed-loop system monotonically decreases throughout the dynamic process, enabling smooth convergence from any initial state to the equilibrium point without oscillations.

[0050] Based on the Lassalle invariance principle, a rigorous stability proof of the new energy power system is provided, mathematically ensuring that the virtual power flow dynamics inevitably lead to a unified price across the entire network in steady state, and that the closed-loop system derivative is less than or equal to zero. In other words, the mathematical characteristics of the new energy power system in steady state fully conform to the KKT conditions for the social welfare maximization problem, simultaneously achieving: zero frequency deviation, tie-line power reaching the economically optimal planned value, and equal marginal costs across the entire system. This proves that the convergence point of the new energy power system is not only the optimal solution for total social cost but also an efficient Nash equilibrium point under non-cooperative game theory. Furthermore, based on the constructed complete physical-market dynamic model, the closed-loop system derivative can be rigorously proven by constructing a Lyapunov energy function, theoretically guaranteeing that the system can asymptotically and stably converge to this equilibrium point from any initial state when facing continuous load disturbances, completely eliminating the oscillation risks caused by dynamic incompatibility.

[0051] For example, refer to Figure 3 As shown, this application executes the following steps in a loop during real-time operation: Step 1: Parameter Configuration: Configure physical layer parameters and economic parameters according to the actual parameters of the new energy power system, and set the market layer time constant and augmentation coefficient; Step 2: State initialization: Initialize the physical layer state to the current operating point of the system, and initialize the market layer state to the steady-state value or economic scheduling result of the previous time period; Step 3: Real-time measurement: Measure the frequency deviation and tie-line power of each area, and calculate the area control deviation.

[0052] Step 4: Market Layer Update: Update the node price, virtual power flow, power reference value, and strategy price based on the dynamic update expression of the node price, strategy price, virtual power flow, and power reference value; Step 5: Instruction Issuance: The updated power reference value is issued as an AGC instruction to the speed controllers in each area; Step 6: Execute repeatedly: Return to step 3 and execute periodically.

[0053] This is a schematic diagram of the software virtual functional modules of the distributed market coupling controller in an embodiment of the present invention. The accompanying drawing illustrates... Figure 2 The specific software logic structure running inside the hardware controller.

[0054] Reference Figure 4 As shown, to achieve real-time coupling between physical dynamics and market mechanisms, the controller is internally divided into four virtual functional modules connected in series according to the data flow direction, as shown below: Parameter initialization module: used to configure the physical parameters of the local power system (such as inertia coefficient and governor time constant) and the economic parameters and time constant of the market layer game algorithm when the system starts up or is reset, and to complete the initial value assignment of each state variable; Data acquisition and processing module: It communicates with external hardware measurement units and is responsible for receiving local power grid frequency deviation signals and tie-line power signals in real time, and calculating and generating ACE data for the region. The market game collaborative clearing module is the core computing hub of this controller. It further includes a nodal price calculation unit, a virtual power flow coordination unit, and a strategic bidding unit. This module receives physical deviation data (ACE) sent by the data acquisition and processing module, and combines it with the controller price information of neighboring power generation areas obtained through the communication network, i.e., the nodal price of adjacent areas. It solves the differential equations of nodal price, virtual power flow coordination, power command generation, and strategic bidding based on Nash game in parallel in the continuous time domain. In each control cycle (second level), the controller uses the real-time measurement values ​​(frequency deviation, tie line power) and nodal price at the current moment to synchronously update the four state variables with one-step numerical integration, and then immediately issues new commands. This makes economic optimization no longer an offline process of calculation and execution, but a real-time process integrated with physical frequency regulation. Any frequency change caused by load disturbance can trigger price adjustment and power redistribution simultaneously in the same control cycle, which can cope with high-frequency fluctuations of new energy and realize the economic optimization of the whole network market. Command conversion and distribution module: Receives the latest active power reference value output by the market game collaborative clearing module, converts it into AGC control commands (such as analog voltage / current signals after D / A conversion) that can be recognized by the underlying hardware, and drives the underlying speed controller to perform physical mechanical actions.

[0055] Reference Figure 5 As shown, in a multi-entity interconnected scenario comprising four regions, the frequency deviation response process after a load step disturbance is illustrated. The four sub-graphs correspond to regions 1 to 4, respectively. Initially, transient frequency deviations occur in each region. However, under the action of the distributed coupling controller proposed in this invention, by issuing power commands that take into account physical delays in real time, the frequency deviation curves of all four regions smoothly converge to zero within a finite time without overshoot. This simulation result intuitively verifies the reliability of this invention in ensuring the transient stability of interconnected power grids and achieving error-free frequency regulation.

[0056] Reference Figure 6 As shown, the evolutionary trajectory of the market-level game corresponding to the aforementioned physical adjustment process is illustrated. The four subgraphs correspond to the evolution of market participants within four regions. The solid lines represent the regional node electricity prices updated in real-time by the distributed algorithm, while the dashed lines represent the strategic bids submitted by power generators. In the initial stage of the game, power generators, driven by profit, deviate from the electricity price in their strategic bids. However, with the dynamic evolution based on gradient descent and physical ACE penalties, the dashed lines of strategic bids in each region eventually converge with the solid lines of the node electricity prices, reaching a steady state. More importantly, all four independent subgraphs converge to the same vertical coordinate value (i.e., the unified clearing price across the entire network) in the steady state. This result rigorously demonstrates in engineering terms that the control law of this invention can effectively curb the false bidding behavior of power generators, achieve true incentive compatibility, and ensure that the entire network eventually converges to the globally optimal Nash equilibrium point that satisfies the KKT conditions.

[0057] A power system frequency control device is applied to a new energy power system. The new energy power system includes multiple power generation areas connected by interconnecting lines. Each power generation area includes generator units, a measurement unit, and a distributed market coupling controller. The distributed market coupling controller has a preset generator unit generation response model and a preset market game model, including: The preprocessing module is used to collect the regional frequency deviation and tie-line power of adjacent power generation areas through the measurement unit, and calculate the regional control deviation based on the regional frequency deviation and tie-line power. The data acquisition module is used to obtain the nodal electricity price corresponding to the power generation area; The market calculation module is used to input the regional control deviation and nodal electricity price into the preset market game model to generate the power reference value, so that the power reference value and the regional control deviation are coupled in real time. The physical control module is used to generate control commands for the generator set based on the power reference value using a preset generator set power generation response model, and then adjust the frequency of the generator set based on the control commands.

[0058] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned power system frequency distributed game optimization control method that takes into account unit dynamics. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0059] Please see Figure 7 , Figure 7 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 701 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 702 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 702 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 702 and is called and executed by the processor 701 to execute the power system frequency distributed game optimization control method considering unit dynamics according to the embodiments of this application. The input / output interface 703 is used to implement information input and output; The communication interface 704 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 705 transmits information between various components of the device (e.g., processor 701, memory 702, input / output interface 703, and communication interface 704); The processor 701, memory 702, input / output interface 703, and communication interface 704 are connected to each other within the device via bus 705.

[0060] In some embodiments, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described power system frequency distributed game optimization control method taking into account unit dynamics.

[0061] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0062] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0063] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0064] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0065] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0066] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0067] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0068] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0069] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0070] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0071] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A power system frequency distributed game optimization control method considering unit dynamics, characterized in that, This method is applied to a new energy power system, which includes multiple power generation areas connected by interconnecting lines. Each power generation area includes generator units, measurement units, and a distributed market coupling controller. The distributed market coupling controller has a built-in preset generator unit power generation response model and a preset market game model. The measurement unit collects the regional frequency deviation and tie-line power of adjacent power generation areas, and calculates the regional control deviation based on the regional frequency deviation and the tie-line power. Obtain the nodal electricity price corresponding to the power generation area; The regional control deviation and the node electricity price are input into the preset market game model to generate a power reference value, so that the power reference value and the regional control deviation are coupled in real time; The preset generator set power generation response model generates control commands corresponding to the generator set based on the power reference value, and then adjusts the frequency of the generator set based on the control commands. 2.The method of claim 1, wherein, The pre-defined market game model is determined through the following steps: An optimization objective function for minimizing the power generation cost of the new energy power system is constructed, and a virtual power flow is introduced to construct the constraints corresponding to the optimization objective function, wherein the virtual power flow is driven by the difference in the nodal electricity price between adjacent power generation areas. The constraints are relaxed into the optimization objective function using the Lagrange multiplier method, and the regional frequency deviation is introduced as a penalty driving term to construct the augmented Lagrange function of the preset market game model. 3.The method of claim 2, wherein, The pre-defined market game model satisfies the following expression: ； wherein, is the augmented Lagrange function, is the node price corresponding to the jth generation area, is the virtual power flow, is the area control deviation corresponding to the jth generation area, is the reference power corresponding to the jth generation area, is the strategy offer corresponding to the jth generation area, j is a positive integer.

4. The power system frequency distributed game optimization control method considering unit dynamics according to claim 3, characterized in that, The step of inputting the regional control deviation and the nodal electricity price into the preset market game model to generate the power reference value includes: The augmented Lagrangian function is iteratively updated using the primal-dual algorithm, which serves as the original variable; the virtual power flow, the power reference value, and the strategy quotation corresponding to the power generation area. Simultaneously, using the regional control deviation as a penalty driving term, the nodal electricity price, which is the dual variable, is iteratively updated until it converges to the Nash equilibrium point, thus obtaining the power reference value.

5. The power system frequency distributed game optimization control method considering unit dynamics as described in claim 1, characterized in that, The power generation area also includes a PLC controller corresponding to the generator set. The PLC controller has a built-in preset generator set power generation response model, which is determined through the following steps: Obtain the preset swing equation and the preset connecting line equation; Based on the mechanical power of the generator set's rotor, the rotor's angular frequency deviation, and the droop coefficient, a governor dynamic equation is constructed to describe the inertial response process of the mechanical power to the power reference value, thereby obtaining the preset generator set power generation response model.

6. The power system frequency distributed game optimization control method considering unit dynamics according to claim 5, characterized in that, The preset unit power generation response model satisfies the following expression: ； in, The inertia coefficient, The angular frequency deviation is... The mechanical power, For load power, The damping coefficient is... The net power of the connecting lines flowing out of the power generation area, The time constant of the speed controller is... The power reference value is... The adjustment coefficient is the coefficient mentioned above. Let j be the equivalent phase angle of the j-th power generation region. The angular frequency deviation corresponding to the j-th power generation area is... Let be the synchronization coefficient of the tie line between the i-th power generation area and the j-th power generation area. Let be the equivalent phase angle of the i-th power generation region, where i and j are both positive integers.

7. The power system frequency distributed game optimization control method considering unit dynamics as described in claim 5, characterized in that, The power generation area also includes a speed governor corresponding to the generator set, and the adjustment of the frequency of the generator set based on the control command includes: Obtain the control instructions output by the PLC controller; The frequency of the generator set is controlled by adjusting the speed governor based on the control command.

8. A power system frequency control device, characterized in that, This device is applied to a new energy power system, which includes multiple power generation areas connected by interconnecting lines. Each power generation area includes generator sets, a measurement unit, and a distributed market coupling controller. The distributed market coupling controller has a built-in preset generator set generation response model and a preset market game model. The device includes: A preprocessing module is used to collect the regional frequency deviation and tie-line power of adjacent power generation areas through the measurement unit, and calculate the regional control deviation based on the regional frequency deviation and the tie-line power. The data acquisition module is used to obtain the nodal electricity price corresponding to the power generation area; A market calculation module is used to input the regional control deviation and the node electricity price into the preset market game model to generate a power reference value, so that the power reference value and the regional control deviation are coupled in real time; A physical control module is used to generate control commands corresponding to the generator set based on the power reference value using the preset generator set power generation response model, and then adjust the frequency of the generator set based on the control commands.

9. An electronic device, characterized in that, include: At least one processor; At least one memory for storing at least one program; When at least one of the programs is executed by at least one of the processors, the power system frequency distributed game optimization control method taking into account unit dynamics as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing computer-executable instructions, characterized in that, The computer-executable instructions are used to execute the power system frequency distributed game optimization control method that takes into account unit dynamics as described in any one of claims 1 to 7.

Images