A MW-level inertia flexible new energy synchronous power generation system and an inertia adjusting method thereof

Abstract

The embodiment of the specification provides a MW-level inertia flexible new energy synchronous power generation system and an inertia adjusting method thereof, wherein the method comprises: online identifying a short circuit ratio (SCR) of a target grid connection point; when the SCR is lower than a preset threshold, generating a control instruction for increasing an inertia coefficient and reducing a speed time constant; when the SCR is not lower than the preset threshold, generating a control instruction for reducing the inertia coefficient and increasing the speed time constant; in response to the control instruction, increasing or reducing or replacing the detachable inertia disc to change a total rotational inertia of the motor stack; and the inverter adjusts an electromagnetic torque according to the inertia coefficient and the speed time constant to change a release rate of kinetic energy of the motor stack rotor. Through the synergistic effect of the actual change of the mechanical inertia and the adjustment of the electromagnetic torque, the inertia support and voltage support capability of the new energy grid connection are significantly improved.

Description

Technical Field

[0001] The embodiments in this specification relate to the field of new energy power generation and grid connection technology, and in particular to a MW-level inertia flexible new energy synchronous power generation system and its inertia adjustment method. Background Technology

[0002] With the continuous expansion of the installed capacity of new energy power generation, photovoltaic power generation, wind power generation and other new energy sources have become the mainstream grid connection method through power electronic inverters. Traditional new energy grid connection systems adopt a direct grid connection architecture of power generation unit-inverter-grid. The energy only undergoes a single-stage DC / AC conversion. The power generation unit has no rotating moving parts and is a static power generation device, which leads to the inherent characteristics of zero inertia and weak support in the new energy grid connection system.

[0003] To compensate for the insufficient inertia of new energy grid connection, several technical routes have been developed. The first is virtual synchronous machine technology, which uses control algorithms to enable power electronic converters to simulate the inertia response characteristics of synchronous generators. This includes virtual inertia regulation methods based on two-layer adaptive inertia control and hybrid energy storage inertia-damping coordinated control strategies based on virtual shaft coupling. The second is grid-connected inverter technology, which transforms grid-following control into grid-connected control, enabling the inverter to exhibit characteristics outside the voltage source and providing inertia and frequency support to the grid. The third is synchronous condenser technology, which configures synchronous condensers at new energy power plants to provide rotational inertia and reactive power support. The fourth is flywheel energy storage technology, which stores kinetic energy through a flywheel rotor, equivalent to a short-term high-power synchronous generator unit, achieving active support for grid voltage and frequency.

[0004] However, the aforementioned existing technologies still have many shortcomings in practical applications.

[0005] First, existing virtual synchronous machine (VSM) and grid-connected inverter technologies rely on control algorithms to simulate inertia, falling under the category of software-defined power supplies. They lack real physical rotating mass, resulting in delays in inertia response and limited duration of inertia support. Second, while synchronous condensers and flywheel energy storage solutions possess real rotating inertia, synchronous condensers are independent bypass devices requiring additional investment and space. Their inertia carriers are independent of the energy conversion path of renewable energy grid connection. Flywheel energy storage systems suffer from high cost and low energy density. Third, the inertia parameters of existing inertia support solutions are generally difficult to adjust flexibly according to changes in grid operating conditions. The rotational inertia of synchronous generators and flywheel energy storage is determined by rotor mass and remains fixed after manufacturing. Although VSM parameters can be set at the software level, these parameters are fixed once set and cannot dynamically adapt to grid strength. Furthermore, in traditional inverter solutions, active and reactive power share converter capacity, leading to coupling competition and making it difficult to simultaneously meet the differentiated needs of active inertia support and reactive voltage support. Finally, there is no existing technology that applies detachable inertia adjustment methods to grid-connected path components of new energy grid-connected power generation systems, and the inertia level lacks a hardware-level adjustment approach.

[0006] Therefore, there is an urgent need for a technical solution that can achieve flexible inertia adjustment and independent control of active and reactive power in the grid connection of new energy sources. Summary of the Invention

[0007] In view of this, embodiments of this specification provide an inertia adjustment method for a MW-level inertia flexible new energy synchronous power generation system. One or more embodiments of this specification also relate to a MW-level inertia flexible new energy synchronous power generation system, a computing device, a computer-readable storage medium, and a computer program, to address the technical deficiencies existing in the prior art.

[0008] According to a first aspect of the embodiments of this specification, an inertia adjustment method for a MW-level flexible new energy synchronous power generation system is provided. The system includes an inverter, a motor driven by the inverter, a synchronous generator coaxially connected to the motor to form a generator stack, and a detachable inertia disk disposed on the rotating shaft of the generator stack. The method includes:

[0009] The system identifies the short-circuit ratio (SCR) of the target grid connection point online. When the SCR is lower than a preset threshold, a control command is generated to increase the inertia coefficient and decrease the speed regulation time constant. When the SCR is not lower than the preset threshold, a control command is generated to decrease the inertia coefficient and increase the speed regulation time constant.

[0010] In response to the control command, the detachable inertia disk can be added, removed, or replaced to change the total moment of inertia of the motor stack;

[0011] The inverter adjusts the electromagnetic torque according to the inertia coefficient and the speed regulation time constant, thereby changing the release rate of the motor stack rotor kinetic energy.

[0012] In one possible implementation, the step of online identification of the short-circuit ratio (SCR) of the target grid connection point includes:

[0013] A non-power frequency disturbance current of a preset amplitude is injected into the target grid connection point, and the voltage phasor and current phasor of the target grid connection point before and after the disturbance current injection are collected;

[0014] The equivalent complex impedance on the grid side is calculated based on the ratio of the voltage phasor difference to the current phasor difference before and after the disturbance.

[0015] The system short-circuit capacity is calculated based on the rated line voltage of the target grid connection point and the modulus of the equivalent complex impedance on the grid side.

[0016] The short-circuit ratio (SCR) is determined based on the ratio of the system's short-circuit capacity to the system's rated apparent power.

[0017] In one possible implementation, the equivalent complex impedance on the grid side is calculated based on the ratio of the voltage phasor difference to the current phasor difference before and after the disturbance, specifically as follows:

[0018]

[0019] In the formula, The equivalent complex impedance on the power grid side, in units of These are the voltage phasor and current phasor at the grid connection point before the disturbance injection; These are the voltage phasor and current phasor at the grid connection point after disturbance injection;

[0020] The system short-circuit capacity is calculated based on the rated line voltage of the target grid connection point and the modulus of the equivalent complex impedance on the grid side, specifically as follows:

[0021]

[0022] In the formula, The short-circuit capacity of the grid connection system is expressed in MVA. This is the rated line voltage at the grid connection point, in kV. This represents the magnitude of the equivalent complex impedance on the power grid side.

[0023] The short-circuit ratio (SCR) is determined based on the ratio of the system's short-circuit capacity to its rated apparent power, specifically as follows:

[0024]

[0025] In the formula, This represents the rated apparent power of this MW-level inertial flexible new energy synchronous power generation system, expressed in MVA.

[0026] In one possible implementation, the preset threshold is a short-circuit ratio (SCR) of 3. When the SCR is lower than 3, the target grid connection point is determined to be a weak grid. When the SCR is not lower than 3, the target grid connection point is determined to be a strong grid.

[0027] When the SCR is less than 3, the detachable inertia disk is added, removed, or replaced to make the equivalent inertia support time of the motor stack greater than or equal to 10 seconds; when the SCR is not less than 3, the detachable inertia disk is added, removed, or replaced to make the equivalent inertia support time less than or equal to 2 seconds.

[0028] In one possible implementation, generating the control commands includes:

[0029] Obtain the real-time operating status set of the system, which includes the grid connection point power angle deviation, frequency deviation, the short-circuit ratio (SCR), and the motor stack speed.

[0030] The real-time running state set is input into the Actor network of a pre-trained Deep Deterministic Policy Gradient (DDPG) model, and the Actor network outputs the inertia coefficient and the speed regulation time constant; wherein, the DDPG model is generated based on a multi-objective weighted reward function that includes frequency deviation penalty, voltage deviation penalty, shaft torque safety penalty and regulation smoothness penalty.

[0031] In one possible implementation, the Actor network includes an input layer, two fully connected hidden layers, and an output layer. The two hidden layers are configured with 256 and 128 neurons respectively and use the ReLU activation function. The output layer uses the Sigmoid activation function to output the inertia coefficient and the speed regulation time constant.

[0032] The Critic network of the DDPG model is based on the Bellman optimal equation to fit the state-action value function, and the complete mathematical expression is:

[0033]

[0034] In the formula, The system state at time t includes the power angle at the grid connection point, frequency deviation, grid short-circuit ratio (SCR), and shaft speed. The control action at time t is the inertia coefficient K and the speed regulation time constant T to be optimized. This is the set of parameters to be trained for the Critic fully connected network, including the weight matrices and bias vectors of each layer; For the target Critic network parameters, a soft update method is used to synchronize the current network parameters to ensure training stability; The immediate reward value is constructed with constraints of minimum frequency deviation, safe shaft torque, and optimal oscillation suppression. This is a discount factor with a value of 0.95, used to weigh the weights of immediate rewards and long-term rewards. For the deterministic policy function of the Actor network, the system state is taken as input and the optimal control action is output.

[0035] The Critic network updates its weights along the negative gradient direction of the temporal difference error during training, and the Actor network updates its weights along the direction of long-term expected reward improvement based on the policy gradient theorem. Furthermore, both the target Actor network and the target Critic network of the DDPG model synchronize the main network parameters through a soft update mechanism.

[0036] In one possible implementation, the multi-objective weighted reward function is constructed by weighting four factors: frequency support effect, voltage stability level, shaft system safety constraint, and adjustment smoothness. Frequency deviation and voltage deviation are negatively correlated with the reward value. A penalty value is deducted when the shaft end torque exceeds three times the rated protection limit. A smoothness penalty value is deducted when the single-step adjustment amplitude of the inertia coefficient is greater than the amplitude threshold.

[0037] One possible implementation also includes a shaft overload protection step:

[0038] Real-time detection of the instantaneous value of the shaft end torque of the motor stack and the speed difference between the motor and the synchronous generator;

[0039] When the shaft end torque exceeds the preset first-level torque threshold, the inverter is controlled to adjust the electromagnetic torque command output to the motor in order to actively suppress shaft torque oscillation.

[0040] When the shaft end torque exceeds the preset secondary torque threshold or the speed difference reaches 5% of the rated speed of the motor stack, the grid connection switch is disconnected to isolate the motor stack.

[0041] One possible implementation also includes multi-time-scale reactive voltage control steps:

[0042] During the sub-transient phase of grid connection disturbance, the electromagnetic response of the damping winding of the synchronous generator is used to provide instantaneous reactive voltage support.

[0043] During the transient phase of grid connection disturbance, the magnetic flux conservation characteristics of the excitation winding of the synchronous generator are utilized to provide rapid reactive power support.

[0044] During the steady-state phase when grid-connected disturbances occur, the excitation parameters are optimized by a neural network based on a self-calibrating attention mechanism, and the excitation system is controlled to adjust the reactive power output of the synchronous generator according to the optimized excitation parameters.

[0045] According to a second aspect of the embodiments of this specification, a MW-level inertial flexible new energy synchronous power generation system is provided, comprising:

[0046] An inverter, the DC side of which is connected to a new energy power generation unit, is configured to convert the DC power output by the new energy power generation unit into AC power.

[0047] An electric motor, the input of which is connected to the AC output of the inverter, is configured to be driven to rotate by the inverter;

[0048] A synchronous generator is coaxially and directly connected to the motor to form a motor stack, which is located on the grid-connected energy transfer path from the inverter to the power grid;

[0049] An inertia adjustment unit includes at least one detachable inertia disk disposed on the motor stack shaft, configured to change the total rotational inertia of the motor stack by adding, removing or replacing the detachable inertia disk;

[0050] The excitation system, independently connected to the synchronous generator, is configured to independently regulate the reactive power output of the synchronous generator;

[0051] The controller, connected to the inverter, the excitation system and the grid connection point measurement point, is configured to identify the short-circuit ratio (SCR) of the target grid connection point online. When the SCR is lower than a preset threshold, it generates a control command to increase the inertia coefficient and decrease the speed regulation time constant. When the SCR is not lower than the preset threshold, it generates a control command to decrease the inertia coefficient and increase the speed regulation time constant.

[0052] The inverter is also configured to adjust the electromagnetic torque according to the inertia coefficient and the speed regulation time constant, thereby changing the release rate of the kinetic energy of the motor stack rotor.

[0053] According to a third aspect of the embodiments of this specification, a computing device is provided, comprising:

[0054] Memory and processor;

[0055] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the above-described inertia adjustment method for the MW-level inertia flexible new energy synchronous power generation system.

[0056] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided that stores computer-executable instructions, which, when executed by a processor, implement the steps of the inertia adjustment method for the MW-level inertia flexible new energy synchronous power generation system described above.

[0057] According to a fifth aspect of the embodiments of this specification, a computer program is provided, wherein when the computer program is executed in a computer, the computer is instructed to perform the steps of the inertia adjustment method for the MW-level inertia flexible new energy synchronous power generation system described above.

[0058] One embodiment of this specification achieves the following beneficial effects: By identifying the short-circuit ratio of the target grid connection point online, control commands for the inertia coefficient and speed regulation time constant are generated based on the relationship between the short-circuit ratio and a preset threshold. These commands then control the addition, removal, or replacement of the detachable inertia disk and the inverter's adjustment of the motor's electromagnetic torque. Since the control commands simultaneously act on the mechanical adjustment of the detachable inertia disk and the electromagnetic torque adjustment of the inverter, the release rate of the motor stack's rotor kinetic energy is affected by both the change in mechanical inertia and the rate of change in electromagnetic torque. These two factors collectively determine the inertia support characteristics exhibited by the motor stack. When the grid strength is weak, increasing the rotational inertia and decreasing the speed regulation time constant allows the rotor kinetic energy to be released at a slower rate, thereby extending the duration of inertia support. When the grid strength is strong, decreasing the rotational inertia and increasing the speed regulation time constant allows the motor stack to recover to a stable operating state more quickly, thus avoiding unnecessary rotor kinetic energy loss. Compared to schemes that rely solely on control algorithms to adjust virtual inertia parameters, this application achieves the adjustment of inertia support capacity through the synergistic effect of actual changes in mechanical inertia and electromagnetic torque regulation, thus overcoming the unilateral constraints of the control algorithm parameter setting range. Attached Figure Description

[0059] Figure 1 This is a flowchart illustrating an embodiment of an inertia-based flexible new energy synchronous power generation system with MW-level inertia, as provided in this specification.

[0060] Figure 2 This is a schematic diagram of a specific MW-level inertia flexible new energy synchronous power generation system provided in one embodiment of this specification;

[0061] Figure 3 This is a schematic diagram of the structure of a specific inertia adjustment unit provided in one embodiment of this specification;

[0062] Figure 4 This is a flowchart illustrating a specific inertia adjustment method for a MW-level flexible new energy synchronous power generation system provided in one embodiment of this specification;

[0063] Figure 5This is a flowchart illustrating a specific DDPG control method provided in one embodiment of this specification;

[0064] Figure 6 This is a structural block diagram of a MW-level inertia flexible new energy synchronous power generation system provided in one embodiment of this specification;

[0065] Figure 7 This is a structural block diagram of a computing device provided in one embodiment of this specification. Detailed Implementation

[0066] Many specific details are set forth in the following description to provide a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar extensions without departing from the spirit of this specification. Therefore, this specification is not limited to the specific implementations disclosed below.

[0067] The terminology used in one or more embodiments of this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of this specification. The singular forms “a” and “the” as used in one or more embodiments of this specification and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0068] It should be understood that although the terms first, second, etc., may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first may also be referred to as second without departing from the scope of one or more embodiments of this specification, and similarly, second may also be referred to as first. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0069] With the continuous increase in the proportion of new energy power generation, the equivalent inertia level of the power system is showing a downward trend, and the problem of frequency stability is becoming increasingly prominent. Although technologies such as virtual synchronous machines, synchronous condensers, and flywheel energy storage have attempted to solve the problem of insufficient inertia from different perspectives, each has its own inherent limitations that are difficult to overcome.

[0070] The root cause lies in the fact that the inertia response of existing virtual synchronous machine technology relies entirely on the simulated output of the control algorithm. Essentially, it involves the power electronic converter dynamically adjusting its active power output based on frequency deviation to simulate the inertia response characteristics of a synchronous generator. This purely software-defined inertia simulation method suffers from response speed limitations imposed by control loop delays and digital sampling periods, making it difficult to achieve the zero-delay response level of real physical inertia. Furthermore, due to the lack of a real rotating mass as an energy carrier, the duration of inertia support is constrained by converter capacity and energy storage capacity, hindering long-term inertia support. The fundamental reason for this deficiency is that the energy source for software-simulated inertia is the rapid power regulation of the power electronic converter, rather than the kinetic energy reserve of the rotating mass; the two are fundamentally different in physical nature.

[0071] Furthermore, while synchronous condensers and flywheel energy storage solutions possess real rotational inertia, the synchronous condenser, as an independent bypass device connected in parallel to the grid, does not participate in the energy conversion process of the new energy power generation system. The inertia carrier and the grid-connected energy path are independent, resulting in redundant system structure and high investment costs. Similarly, flywheel energy storage systems are connected to the grid as independent energy storage units. Their inertia regulation acts on the energy storage circuit rather than the main energy path of new energy power generation, failing to fundamentally change the zero-inertia grid connection characteristic of new energy power generation units. Moreover, the inertia parameters of the above solutions are determined by the rotor's physical structure and remain fixed after manufacturing, making it difficult to flexibly adjust according to dynamic changes in grid operating conditions. This leads to insufficient inertia support under weak grid conditions and the risk of over-adjustment under strong grid conditions.

[0072] On the other hand, traditional grid-connected inverters for new energy sources adopt an architecture where active and reactive power share the inverter capacity, resulting in a coupled and competitive relationship between active and reactive power regulation. When the system needs to simultaneously provide active inertia support and reactive voltage support, their respective occupancy of inverter capacity restricts each other, making it difficult to achieve optimal support simultaneously. The root cause of this problem is that a single inverter needs to undertake both active and reactive power regulation tasks simultaneously, making it impossible to allocate capacity resources independently.

[0073] Therefore, the technical problems faced by existing technologies include, but are not limited to: how to embed real physical rotating inertia into the energy transfer path of new energy grid-connected power generation systems so that new energy power generation has inertia response characteristics equivalent to synchronous generator sets; how to achieve flexible adjustment of inertia level with dynamic changes in grid intensity to adapt to the differentiated needs of different operating scenarios; and how to achieve hardware-level decoupling of active power regulation and reactive power regulation so that the two can be optimized independently and do not interfere with each other.

[0074] To address the aforementioned technical problems, this specification provides an inertia regulation method for a MW-level flexible renewable energy synchronous power generation system. This method achieves flexible inertia regulation by setting up a coaxially rotating motor stack with mechanically adjustable rotational inertia in the renewable energy grid-connected energy transmission path; dynamically and adaptively matching the inertia support strength by sensing grid strength and coordinating the mechanical inertia configuration with electrical speed control parameters; and achieving hardware-level decoupling control of active and reactive power by isolating the reactive power regulation channel from the main power conversion link. This specification also relates to a MW-level flexible renewable energy synchronous power generation system, a computing device, and a computer-readable storage medium, which will be described in detail in the following embodiments.

[0075] See Figure 1 , Figure 1 A flowchart is shown of an inertia adjustment method for a MW-level flexible new energy synchronous power generation system according to an embodiment of this specification. The system includes an inverter, a motor driven by the inverter, a synchronous generator coaxially connected to the motor to form a motor stack, and a detachable inertia disk disposed on the rotating shaft of the motor stack. The method specifically includes the following steps.

[0076] In this paper, the term "motor stack" refers to an integrated rotating unit consisting of a motor and a synchronous generator directly connected coaxially via a shaft. The rotational mechanical energy of the motor is directly transferred to the synchronous generator through the coaxial connection, and both share the same axis of rotation and moment of inertia. The term "grid-connected energy transfer path" is defined by the form of energy flow, namely, the complete energy transfer channel starting from the DC power output of the new energy power generation unit, which is converted into AC power by the inverter, the motor converts electrical energy into mechanical energy, the synchronous generator converts mechanical energy into AC power, and finally feeds it into the power grid. The rotational inertia of the rotating components located on this path directly affects the power generation output process. The term "removable inertia disk" refers to a rotating mass body installed on the motor stack shaft that can be added, removed, or replaced through on-site disassembly and assembly operations. Its function is to change the total moment of inertia of the motor stack. The difference between a removable inertia disk and a flywheel energy storage device is that a removable inertia disk is a component of the motor stack shaft and does not have independent drive and power generation functions, while flywheel energy storage is an energy storage unit independent of the power generation path. The short-circuit ratio (SCR) adopts a well-known definition in the power system field, namely the ratio of the short-circuit capacity of the grid-connected system to the rated apparent power of the renewable energy generation system. It is used to characterize the grid's capacity to support renewable energy generation or the grid strength. The short-circuit ratio can be detected and calculated using methods known in the field, such as the small disturbance injection method and the short-circuit capacity measurement method. The inertia coefficient refers to the proportional gain output by the controller used to adjust the electromagnetic torque of the inverter. This coefficient determines the change in the electromagnetic torque output by the inverter per unit speed deviation and is a key parameter affecting the inertia response strength. The speed regulation time constant refers to the time parameter output by the controller used to adjust the rotor kinetic energy release rate. This parameter determines the speed of the inertia response; a smaller time constant results in faster kinetic energy release and a quicker response, while a larger time constant results in a smoother kinetic energy release and a longer duration.

[0077] Step 101: Identify the short-circuit ratio (SCR) of the target grid connection point online; when the SCR is lower than a preset threshold, generate a control command to increase the inertia coefficient and decrease the speed regulation time constant; when the SCR is not lower than the preset threshold, generate a control command to decrease the inertia coefficient and increase the speed regulation time constant.

[0078] The short-circuit ratio (SCR) of the target grid connection point is identified online to obtain the current grid strength information. The SCR is configured as a quantitative indicator characterizing the grid strength at the target grid connection point, reflecting the grid's capacity to support renewable energy generation in real time. By identifying the SCR online in real time, the latest grid strength information can be obtained promptly when the grid operating state changes, providing accurate predictive basis for subsequent inertia adjustment. Compared to the event-triggered response method based on frequency deviation in traditional virtual synchronous machine technology, this embodiment uses feedforward sensing of the SCR to predict grid strength and adjust inertia configuration in advance before frequency deviation occurs, making the inertia adjustment response more timely and avoiding the response lag caused by passive adjustment after frequency deviation occurs.

[0079] The control commands are configured to match the optimal combination of inertia adjustment parameters according to the grid strength level, ensuring that the inertia support strength adapts to grid demand. The generation of control commands can employ intelligent algorithm-based optimization methods, such as deep reinforcement learning algorithms. These algorithms take grid operating state parameters as input and inertia coefficient and speed regulation time constant as output, obtaining the optimal control strategy through online or offline training. For example, a deep deterministic strategy gradient algorithm can be used, taking grid connection point power angle deviation, frequency deviation, short-circuit ratio (SCR), and generator stack speed as input states. After iterative optimization by a neural network, the inertia coefficient and speed regulation time constant are dynamically output. By dynamically adjusting the inertia coefficient and speed regulation time constant based on the SCR, the inertia adjustment parameters can be adaptively matched to the current grid strength. Under conditions of low SCR (weak grid), increasing the inertia coefficient and decreasing the speed regulation time constant provides stronger inertia support and faster response speed; under conditions of high SCR (strong grid), decreasing the inertia coefficient and increasing the speed regulation time constant reduces inertia configuration to avoid the risk of oscillations caused by over-adjustment. Compared with existing solutions that use fixed inertia parameters, this embodiment achieves dynamic adaptive adjustment of parameters, enabling the system to operate in optimal condition under different power grid intensities.

[0080] In one or more embodiments, the step of identifying the short-circuit ratio (SCR) of the target grid connection point online includes:

[0081] A non-power frequency disturbance current of a preset amplitude is injected into the target grid connection point, and the voltage phasor and current phasor of the target grid connection point before and after the disturbance current injection are collected;

[0082] The equivalent complex impedance on the grid side is calculated based on the ratio of the voltage phasor difference to the current phasor difference before and after the disturbance.

[0083] The system short-circuit capacity is calculated based on the rated line voltage of the target grid connection point and the modulus of the equivalent complex impedance on the grid side.

[0084] The short-circuit ratio (SCR) is determined based on the ratio of the system's short-circuit capacity to the system's rated apparent power.

[0085] The small disturbance injection method injects a non-power frequency disturbance signal and uses the changes in voltage and current before and after the disturbance to calculate the equivalent impedance of the power grid, thereby obtaining the short-circuit capacity and short-circuit ratio. It has high identification accuracy and fast response speed.

[0086] In one or more embodiments, the equivalent complex impedance on the grid side is calculated based on the ratio of the voltage phasor difference to the current phasor difference before and after the disturbance, specifically as follows:

[0087]

[0088] In the formula, The equivalent complex impedance on the power grid side, in units of These are the voltage phasor and current phasor at the grid connection point before the disturbance injection; These are the voltage phasor and current phasor at the grid connection point after disturbance injection;

[0089] The system short-circuit capacity is calculated based on the rated line voltage of the target grid connection point and the modulus of the equivalent complex impedance on the grid side, specifically as follows:

[0090]

[0091] In the formula, The short-circuit capacity of the grid connection system is expressed in MVA. This is the rated line voltage at the grid connection point, in kV. This represents the magnitude of the equivalent complex impedance on the power grid side.

[0092] The short-circuit ratio (SCR) is determined based on the ratio of the system's short-circuit capacity to its rated apparent power, specifically as follows:

[0093]

[0094] In the formula, This represents the rated apparent power of this MW-level inertial flexible new energy synchronous power generation system, expressed in MVA.

[0095] In one or more embodiments, the preset threshold is a short-circuit ratio (SCR) of 3. When the SCR is lower than 3, the target grid connection point is determined to be a weak grid; when the SCR is not lower than 3, the target grid connection point is determined to be a strong grid. When the SCR is lower than 3, the detachable inertia disks are added, removed, or replaced to make the equivalent inertia support time of the motor stack greater than or equal to 10 seconds. When the SCR is not lower than 3, the detachable inertia disks are added, removed, or replaced to make the equivalent inertia support time less than or equal to 2 seconds. Under weak grid conditions, increasing the number of inertia disks increases the equivalent inertia to more than 10 seconds, providing a sufficient inertia support time window for grid frequency recovery. Under strong grid conditions, reducing the number of inertia disks lowers the equivalent inertia to less than 2 seconds, avoiding oscillations caused by inertia over-adjustment. The setting of an equivalent inertia support time greater than or equal to 10 seconds significantly improves the support duration compared to the inertia support time of existing virtual synchronous machine technology (which is usually less than 2 seconds).

[0096] In one or more embodiments, the equivalent inertia support time under weak grid conditions is greater than or equal to 10 seconds. If the equivalent inertia support time is less than 10 seconds, the inertia support time is insufficient in weak grid scenarios, and a secondary frequency drop may occur during frequency recovery. Under strong grid conditions, the equivalent inertia support time is less than or equal to 2 seconds. If it is greater than 2 seconds, the inertia is too strong in strong grid scenarios, which may cause power oscillations. In some embodiments, the equivalent inertia support time under weak grid conditions can be selected as 10-15 seconds, in which case the inertia support time window is sufficient and not over-configured; the equivalent inertia support time under strong grid conditions can be selected as 1-2 seconds, in which case the inertia configuration is lightweight and does not affect system stability. In at least one embodiment, the equivalent inertia support time is 12 seconds under weak grid conditions and 1.5 seconds under strong grid conditions.

[0097] In one or more embodiments, generating the control commands includes:

[0098] Obtain the real-time operating status set of the system, which includes the grid connection point power angle deviation, frequency deviation, the short-circuit ratio (SCR), and the motor stack speed.

[0099] The real-time running state set is input into the Actor network of a pre-trained Deep Deterministic Policy Gradient (DDPG) model, and the Actor network outputs the inertia coefficient and the speed regulation time constant; wherein, the DDPG model is generated based on a multi-objective weighted reward function that includes frequency deviation penalty, voltage deviation penalty, shaft torque safety penalty and regulation smoothness penalty.

[0100] In one or more embodiments, the Actor network includes an input layer, two fully connected hidden layers, and an output layer. The two hidden layers are configured with 256 and 128 neurons respectively and use the ReLU activation function. The output layer uses the Sigmoid activation function to output the inertia coefficient and the speed regulation time constant.

[0101] The Critic network of the DDPG model is based on the Bellman optimal equation to fit the state-action value function, and the complete mathematical expression is:

[0102]

[0103] In the formula, The system state at time t includes the power angle at the grid connection point, frequency deviation, grid short-circuit ratio (SCR), and shaft speed. The control action at time t is the inertia coefficient K and the speed regulation time constant T to be optimized. This is the set of parameters to be trained for the Critic fully connected network, including the weight matrices and bias vectors of each layer; For the target Critic network parameters, a soft update method is used to synchronize the current network parameters to ensure training stability; The immediate reward value is constructed with constraints of minimum frequency deviation, safe shaft torque, and optimal oscillation suppression. This is a discount factor with a value of 0.95, used to weigh the weights of immediate rewards and long-term rewards. For the deterministic policy function of the Actor network, the system state is taken as input and the optimal control action is output.

[0104] The Critic network updates its weights along the negative gradient direction of the temporal difference error during training, and the Actor network updates its weights along the direction of long-term expected reward improvement based on the policy gradient theorem. Furthermore, both the target Actor network and the target Critic network of the DDPG model synchronize the main network parameters through a soft update mechanism.

[0105] In one or more embodiments, the multi-objective weighted reward function is constructed by weighting four factors: frequency support effect, voltage stability level, shaft system safety constraints, and adjustment action smoothness. Frequency deviation and voltage deviation are negatively correlated with the reward value. A penalty value is deducted when the shaft end torque exceeds three times the rated protection limit. A smoothness penalty value is deducted when the single-step adjustment amplitude of the inertia coefficient exceeds the amplitude threshold. The frequency support effect objective aims to minimize frequency deviation, guiding the algorithm to prioritize grid frequency stability. The voltage stability level objective aims to minimize voltage deviation, ensuring voltage quality at the grid connection point. The shaft system safety constraints objective aims to ensure the shaft end torque does not exceed the safety limit, guaranteeing the operational safety of the mechanical system. The adjustment action smoothness objective aims to ensure a moderate single-step adjustment amplitude of the inertia coefficient, avoiding drastic fluctuations during the adjustment process. The comprehensive reward function constructed by these four weighted factors enables the algorithm to achieve balanced optimization across multiple performance dimensions.

[0106] The DDPG algorithm takes four dimensions of state as input: power angle deviation, frequency deviation, SCR, and generator stack speed. This comprehensively covers key information about the power grid and system operating states, providing a sufficient basis for decision-making. A multi-objective weighted reward function comprehensively evaluates the control effect from four dimensions: frequency support, voltage stability, shaft safety, and regulation smoothness, ensuring that the generated inertia coefficient and speed regulation time constant achieve optimal performance across multiple dimensions. The Actor network employs a four-layer fully connected structure, with two hidden layers containing 256 and 128 neurons respectively. Features are extracted using the ReLU activation function, and the output layer constrains the output amplitude using the Sigmoid activation function, keeping the output inertia coefficient and speed regulation time constant within a reasonable range. The Critic network fits the state-action value function based on the Bellman optimal equation. The Bellman equation is a well-known iterative relationship of value functions in reinforcement learning, allowing the Critic network to accurately evaluate the long-term expected reward of taking an action in the current state. The target network synchronizes the main network parameters through a soft update mechanism, avoiding drastic fluctuations in target values ​​during training and ensuring stable convergence.

[0107] In some embodiments, the electric motor is an asynchronous motor or a permanent magnet synchronous motor. In some embodiments, the synchronous generator is an electrically excited synchronous generator. In some embodiments, the new energy power generation unit is at least one of a photovoltaic array, a wind turbine, or an energy storage battery.

[0108] In some embodiments, the system further includes a shaft-end torque sensor and a speed difference detection unit. The shaft-end torque sensor is mounted on the motor stack shaft and connected to the controller, and is configured to send a shaft-end torque signal to the controller. The speed difference detection unit is connected to the controller and is configured to detect the speed difference between the motor and the synchronous generator and send a speed difference signal to the controller.

[0109] Step 102: In response to the control command, add, remove, or replace the detachable inertia disks to change the total rotational inertia of the motor stack. The detachable inertia disks are configured as hardware carriers providing actual physical rotational mass, and their quantity or specifications directly change the total rotational inertia of the motor stack. In one or more embodiments, the addition, removal, or replacement of the detachable inertia disks can be achieved through on-site disassembly and assembly, for example, via a flange connection structure, using bolts to fix the inertia disks to the motor stack shaft. When adjustment is needed, the corresponding number of inertia disks can be removed or added. For example, in a weak power grid scenario, the number of inertia disks can be increased to improve the total rotational inertia, while in a strong power grid scenario, the number of inertia disks can be reduced to decrease the total rotational inertia. The connection methods between the inertia disks and the shaft include, but are not limited to, flange connections, spline connections, keyway connections, and expansion sleeve connections.

[0110] By adding, removing, or replacing the detachable inertia disk to change the total rotational inertia of the generator stack, a real physical rotating mass can be implanted into the energy transmission path of new energy grid connection, giving the inertia response physical realism and making the response speed not limited by the delay of the control algorithm. Compared with the virtual synchronous machine technology that simulates inertia purely in software, the physical inertia disk in this embodiment provides a real rotating mass base, the inertia response has no delay and the support time is determined by the physical rotational inertia, which is not limited by the converter capacity. Compared with bypass inertia solutions such as flywheel energy storage, the inertia disk in this embodiment is directly set on the generator stack shaft in the grid-connected energy transmission path, and the inertia directly acts on the power generation output process, without the need for additional independent energy storage equipment.

[0111] Flange connections use bolts to secure the inertia disc to the shaft flange, offering high load-bearing capacity and easy assembly / disassembly. Spline connections transmit torque through toothed meshing, providing good centering and uniform load distribution. Keyway connections achieve circumferential positioning using flat or wedge keys, resulting in a simple structure and low cost. Expansion sleeve connections achieve keyless connection through expansion force, eliminating keyway stress concentration and ensuring high coaxiality. All four connection methods are mature and reliable detachable connection structures in the mechanical field, allowing for flexible selection based on torque magnitude, assembly / disassembly frequency, and processing conditions.

[0112] In some embodiments, the detachable inertia disk is made of a high-density alloy material with a density greater than or equal to 17 g / cm³, and the mass of a single detachable inertia disk ranges from 50 to 200 kg. In one or more embodiments, the inertia disk material density is greater than or equal to 17 g / cm³, and the mass of a single inertia disk is 50 to 200 kg. If the density is less than 17 g / cm³, the inertia disk volume is too large and the axial dimension is too long under MW-level inertia requirements, affecting the stability of the shaft system; if the single mass is less than 50 kg, the inertia contribution of a single disk is too small, requiring too many disks to meet the inertia requirements, increasing the workload of disassembly and assembly; if the single mass is greater than 200 kg, the single disk is too heavy, making on-site disassembly and assembly difficult. In some embodiments, the density can be selected as 17 to 19 g / cm³, and the single mass can be selected as 80 to 150 kg, at which point the inertia disk volume is moderate and disassembly and assembly are convenient. In at least one embodiment, the density is 18 g / cm³, and the single mass is 100 kg.

[0113] Step 103: The inverter adjusts the electromagnetic torque according to the inertia coefficient and the speed regulation time constant to change the release rate of the motor stack rotor kinetic energy.

[0114] In one or more embodiments, the inverter's electromagnetic torque regulation can be achieved through vector control, such as rotor field-oriented control, by adjusting the torque component of the stator current to control the magnitude of the electromagnetic torque. The inertia coefficient is used to determine the proportional relationship between the electromagnetic torque and the speed deviation, and the speed regulation time constant is used to determine the response speed of torque regulation. For example, when the grid frequency decreases, the inverter increases the electromagnetic torque according to the inertia coefficient to accelerate the motor, releasing rotor kinetic energy to provide inertial support; the smaller the speed regulation time constant, the faster the torque adjustment response speed and the more rapidly the kinetic energy is released.

[0115] By adjusting the electromagnetic torque based on the inertia coefficient and speed regulation time constant using an inverter, the kinetic energy release rate of the physical inertia can be flexibly shaped in software, allowing the shape and duration of the inertia response curve to be adjusted as needed. Compared to the limitations of pure mechanical inertia schemes where kinetic energy is released at an inherent rate and the response curve remains constant, this embodiment provides the flexibility of adjustable inertia response parameters through electrical speed regulation control, enabling the inertia support effect to be optimized according to grid requirements.

[0116] In one or more embodiments, a shaft overload protection step is also included:

[0117] Real-time detection of the instantaneous value of the shaft end torque of the motor stack and the speed difference between the motor and the synchronous generator;

[0118] When the shaft end torque exceeds the preset first-level torque threshold, the inverter is controlled to adjust the electromagnetic torque command output to the motor in order to actively suppress shaft torque oscillation.

[0119] When the shaft end torque exceeds the preset secondary torque threshold or the speed difference reaches 5% of the rated speed of the motor stack, the grid connection switch is disconnected to isolate the motor stack.

[0120] By adding shaft safety protection to the inertia adjustment method, the system can provide inertia support while ensuring the operational safety of the mechanical system. The design of the two-level threshold hierarchical protection system takes into account both safety and system availability: when the first-level threshold is triggered, the electromagnetic torque is actively adjusted to suppress oscillation, and the system still maintains grid-connected operation; when the second-level threshold or speed difference exceeds the limit, the grid is directly disconnected to ensure the safety of the shaft system under extreme operating conditions.

[0121] Regarding the primary and secondary torque thresholds, in one or more embodiments, the primary torque threshold is twice the rated torque of the motor stack, and the secondary torque threshold is three times the rated torque of the motor stack. If the primary threshold is less than twice the rated torque, protection actions may be frequently triggered during normal inertia adjustment, affecting the normal operation of the system; if the secondary threshold is greater than three times the rated torque, the shaft system may have suffered irreversible damage under extreme operating conditions. In some embodiments, the primary torque threshold can be selected as 1.8 to 2.2 times the rated torque, and the secondary torque threshold can be selected as 2.8 to 3.2 times the rated torque. In at least one embodiment, the primary torque threshold is twice the rated torque, and the secondary torque threshold is three times the rated torque.

[0122] In some embodiments, the primary torque threshold is twice the rated torque of the motor stack. In response to the shaft end torque exceeding the primary torque threshold but not reaching the secondary torque threshold, a reverse electromagnetic torque adjustment command is sent to the inverter and a 100ms delay is initiated. If the torque drops within 100ms, the suppression mode is exited; if the torque continues to exceed the limit after 100ms, the grid connection switch is escalated and disconnected. The secondary torque threshold is three times the rated torque of the motor stack. The shaft overload protection is locked when any of the following conditions are met: the system is in a speed ramp phase during startup or shutdown; the duration of the shaft end torque exceeding the limit is less than 10ms; the target grid connection point is in a high-voltage ride-through or low-voltage ride-through condition; the speed difference is less than 1% of the rated speed and the shaft end torque fluctuation is within the 5% dead zone of the rated torque.

[0123] In one or more embodiments, a multi-time-scale reactive voltage control step is also included:

[0124] During the sub-transient phase of grid connection disturbance, the electromagnetic response of the damping winding of the synchronous generator is used to provide instantaneous reactive voltage support.

[0125] During the transient phase of grid connection disturbance, the magnetic flux conservation characteristics of the excitation winding of the synchronous generator are utilized to provide rapid reactive power support.

[0126] During the steady-state phase when grid-connected disturbances occur, the excitation parameters are optimized by a neural network based on a self-calibrating attention mechanism, and the excitation system is controlled to adjust the reactive power output of the synchronous generator according to the optimized excitation parameters.

[0127] By introducing a three-stage control strategy into the independent reactive power regulation channel, reactive power voltage support across the entire timescale from instantaneous to steady state is achieved. The subtransient stage utilizes the natural electromagnetic response of the damping winding, which has the fastest response speed but shortest duration, providing instantaneous voltage support to prevent voltage collapse. The transient stage leverages the flux linkage conservation characteristics of the excitation winding, utilizing the physical property that rotor flux linkage cannot change abruptly to achieve rapid but somewhat lagging reactive power support. The steady-state stage optimizes excitation parameters through a deep neural network based on a self-calibrating attention mechanism, achieving refined steady-state voltage regulation. Each of the three stages contributes its strengths and seamlessly connects, covering the entire process from disturbance occurrence to steady-state recovery.

[0128] In some embodiments, the method further includes: constructing an active power-frequency droop control characteristic with a droop rate of 4% to 6% based on real-time speed feedback of the synchronous generator, so that the system participates in primary frequency regulation of the power grid during grid frequency disturbances. By adding frequency regulation functionality to the inertia regulation method, the system possesses primary frequency regulation capability equivalent to that of a traditional synchronous generator set. The droop control characteristic is constructed based on real-time speed feedback of the synchronous generator; when the grid frequency increases, the speed increases, and the system reduces active power output; when the grid frequency decreases, the speed decreases, and the system increases active power output, forming a self-regulating closed loop of active power-frequency. The droop rate determines the proportional relationship between frequency deviation and active power regulation.

[0129] Regarding the numerical range of the droop rate, in one or more embodiments, the droop rate is 4% to 6%. If the droop rate is less than 4%, the active power regulation per unit frequency deviation is too large, which may lead to over-regulation and power oscillations; if the droop rate is greater than 6%, the active power regulation per unit frequency deviation is too small, resulting in insufficient frequency regulation contribution and difficulty in meeting the primary frequency regulation requirements of the power grid. In some embodiments, the droop rate can be selected as 4.5% to 5.5%, at which point a good balance is achieved between frequency regulation contribution and regulation stability. In at least one embodiment, the droop rate is 5%.

[0130] In some embodiments, the motor is configured to have its active power regulated by the inverter, and the synchronous generator is configured to have its reactive power independently regulated by the excitation system, so that active power regulation and reactive power regulation are separate physical devices in hardware. Active power regulation is achieved by the inverter through the motor, and reactive power regulation is achieved by the excitation system through the synchronous generator. These two are separate power conversion and regulation channels in hardware, and their regulation processes do not interfere with each other.

[0131] There is a clear synergistic relationship between the steps described above: online short-circuit ratio identification, control command generation, physical inertia adjustment, and kinetic energy release rate adjustment. The grid strength information obtained from the online short-circuit ratio (SCR) identification serves as the input for generating control commands. The control commands simultaneously act on both the physical inertia adjustment and electrical speed control channels, ensuring that the increase or decrease in mechanical inertia is synchronized with the adjustment of speed parameters. The rotating mass base provided by physical inertia adjustment provides kinetic energy reserves for inverter speed control, enabling real energy to be dispatched for speed control.

[0132] The following is in conjunction with the appendix Figures 2 to 5 Taking the application of the inertia adjustment method of a MW-level flexible new energy synchronous power generation system provided in this specification in the grid-connected operation scenario of a large-scale ground photovoltaic power station as an example, the inertia adjustment method of the MW-level flexible new energy synchronous power generation system will be further explained.

[0133] This embodiment uses a large-scale ground-mounted photovoltaic power station as an example to illustrate the specific structural composition and connection relationship of a MW-level inertia flexible new energy synchronous power generation system. The photovoltaic power station has an installed capacity of 10MW and is connected to the local power grid. Due to the relatively weak local power grid (low short-circuit ratio), the grid-connected system needs to have strong inertia support capabilities.

[0134] like Figures 2 to 3 As shown, the MW-level inertia flexible new energy synchronous power generation system of this embodiment includes a photovoltaic array (100), an inverter (200), an asynchronous motor (300), a synchronous generator (400), an inertia regulation unit (500), an excitation system (600), and a controller (700). Among them, the photovoltaic array is a specific example of a new energy power generation unit.

[0135] The photovoltaic array (100) outputs DC power, which is connected to the DC side of the inverter (200) via a DC cable. The inverter (200) converts the DC power into AC power with adjustable frequency, driving the asynchronous motor (300) to rotate.

[0136] The rotor of the asynchronous motor (300) and the rotor of the synchronous generator (400) are directly connected via the same shaft, forming a motor stack. This means that the motor and generator rotors are directly connected via the same shaft, without any intermediate transmission links. A mounting flange is provided on the shaft for installing the inertia adjustment unit (500).

[0137] The stator output of the synchronous generator (400) is connected to the grid-connected switch and then connected to the power grid via a step-up transformer. The excitation system (600) is connected to the excitation winding of the synchronous generator (400) and independently adjusts the excitation current, thereby independently adjusting the reactive power output of the synchronous generator.

[0138] The input terminal of the controller (700) is connected to the voltage transformer and current transformer at the grid connection point for collecting grid status information; the output terminal is connected to the control terminal of the inverter (200) and the control terminal of the excitation system (600) for issuing control commands.

[0139] The following example, using the inertia adjustment process when the power grid strength changes, illustrates the specific execution steps of the inertia adjustment method. (Refer to...) Figures 4 to 5 When a photovoltaic power station is connected to a power grid of different strength or when the power grid operation mode changes, causing a change in the short-circuit ratio, the system automatically adjusts the inertia configuration and speed regulation parameters to adapt to the current power grid conditions.

[0140] Step S101: Identify SCR online and generate control commands

[0141] The system collects grid connection point voltage and frequency in real time and identifies the short-circuit ratio (SCR) of the target grid connection point online. When the SCR is lower than a preset threshold, a control command is generated to increase the inertia coefficient and decrease the speed regulation time constant; when the SCR is not lower than the preset threshold, a control command is generated to decrease the inertia coefficient and increase the speed regulation time constant.

[0142] In this embodiment, the preset threshold is taken as SCR=3, that is, when SCR<3 is detected, it is determined to be a weak power grid, and when SCR≥3 is detected, it is determined to be a strong power grid. The inertia coefficient in the control command is represented by K in this embodiment, and the speed regulation time constant is represented by response time T in this embodiment, and the two have corresponding meanings.

[0143] Specifically, when SCR < 3 (weak grid), the controller generates a control command to increase the inertia coefficient and decrease the speed regulation time constant. In this embodiment, an increase inertia command is generated, and the controller outputs control parameters with inertia coefficient K = 8 and response time T = 5ms using the DDPG algorithm. When SCR ≥ 3 (strong grid), the controller generates a control command to decrease the inertia coefficient and increase the speed regulation time constant. In this embodiment, a decrease inertia command is generated, and the controller outputs control parameters with inertia coefficient K = 2 and response time T = 50ms.

[0144] Step S102: Add, remove, or replace the removable inertia disk

[0145] In response to the control command, the detachable inertia disk can be added, removed, or replaced to change the total moment of inertia of the motor stack.

[0146] In this embodiment, the inertia disks are made of high-density alloy material, each weighing 100 kg, making this a specific example of a removable inertia disk. When SCR < 3 (weak grid), in response to the inertia increase command, high-density alloy inertia disks (3 in total) are added to the rotor flange surface, increasing the total rotational inertia J of the motor stack by approximately 50%, and raising the equivalent inertia to over 10 seconds. When SCR ≥ 3 (strong grid), in response to the inertia decrease command, some inertia disks are removed (retaining 1), reducing the equivalent inertia to 2 seconds.

[0147] Step S103: The inverter adjusts the electromagnetic torque to change the kinetic energy release rate.

[0148] The inverter adjusts the electromagnetic torque according to the inertia coefficient and the speed regulation time constant, thereby changing the release rate of the motor stack rotor kinetic energy.

[0149] In this embodiment, when SCR < 3 (weak grid), the inverter quickly adjusts the motor torque based on the inertia coefficient K = 8 and the response time T = 5ms, enabling the rotor to quickly release kinetic energy and provide strong inertia support for the weak grid. When SCR ≥ 3 (strong grid), the inverter slowly adjusts the motor torque based on the inertia coefficient K = 2 and the response time T = 50ms to avoid over-adjustment.

[0150] Step S104: Shaft overload protection control

[0151] Building upon the foregoing, this paper takes the shaft system protection process during a short-circuit fault in the power grid as an example to illustrate the specific execution method of shaft system overload protection. Under extreme conditions such as a power grid short circuit, the torque at the motor stack shaft end may increase sharply, which could cause shaft system damage if no protective measures are taken.

[0152] In this embodiment, the coaxial shaft system of the motor stack is designed according to the transient torque impact capability, and has the overload capability of 2 times the rated torque for 10 seconds and 3 times the rated torque for 200ms.

[0153] Step S201: Real-time monitoring of shaft end torque and speed difference

[0154] The shaft-end torque sensor monitors the torque value in real time, and the speed difference detection unit monitors the speed difference between the motor and the generator in real time.

[0155] Step S202: Torque over-limit suppression

[0156] When a short-circuit fault in the power grid causes a sharp increase in shaft torque, the controller adjusts the inverter's electromagnetic torque command within 100ms to suppress torque oscillation amplitude. Specifically, in response to the shaft torque exceeding a primary threshold, the controller sends a reverse electromagnetic torque adjustment command to the inverter to actively suppress shaft torque oscillation.

[0157] Step S203: Over-limit isolation protection

[0158] If the abnormal torque continues to exceed the protection setting, for example, three times the rated torque for more than 200 ms, the controller immediately issues a command to disconnect the grid-connected switch, isolating the motor stack from the power grid and preventing the shaft system from being subjected to prolonged overstress. Specifically, in response to the shaft end torque exceeding the secondary threshold or the speed difference exceeding the preset limit, the controller triggers the grid-connected switch to trip, achieving fault isolation.

[0159] Corresponding to the above method embodiments, this specification also provides an embodiment of a MW-level inertia flexible new energy synchronous power generation system. Figure 6 This specification illustrates a schematic diagram of a MW-level inertia-based flexible new energy synchronous power generation system according to one embodiment. Figure 6 As shown, the system includes:

[0160] Inverter 601, the DC side of the inverter is connected to a new energy power generation unit, and is configured to convert the DC power output by the new energy power generation unit into AC power;

[0161] An electric motor 602, the input terminal of which is connected to the AC output terminal of the inverter, is configured to be driven to rotate by the inverter;

[0162] A synchronous generator 603 is coaxially and directly connected to the motor to form a motor stack, which is located on the grid-connected energy transfer path from the inverter to the power grid;

[0163] The inertia adjustment unit 604 includes at least one detachable inertia disk disposed on the motor stack shaft, and is configured to change the total rotational inertia of the motor stack by adding, removing or replacing the detachable inertia disk;

[0164] The excitation system 605 is independently connected to the synchronous generator and is configured to independently regulate the reactive power output of the synchronous generator;

[0165] The controller 606 is connected to the inverter, the excitation system and the grid connection point measurement point. It is configured to identify the short-circuit ratio (SCR) of the target grid connection point online. When the SCR is lower than a preset threshold, it generates a control command to increase the inertia coefficient and decrease the speed regulation time constant. When the SCR is not lower than the preset threshold, it generates a control command to decrease the inertia coefficient and increase the speed regulation time constant.

[0166] The inverter is also configured to adjust the electromagnetic torque according to the inertia coefficient and the speed regulation time constant, thereby changing the release rate of the kinetic energy of the motor stack rotor.

[0167] The above is a schematic scheme of a MW-level inertia flexible new energy synchronous power generation system according to this embodiment. It should be noted that the technical solution of this MW-level inertia flexible new energy synchronous power generation system and the technical solution of the inertia adjustment method of the MW-level inertia flexible new energy synchronous power generation system described above belong to the same concept. For details not described in detail in the technical solution of the MW-level inertia flexible new energy synchronous power generation system, please refer to the description of the technical solution of the inertia adjustment method of the MW-level inertia flexible new energy synchronous power generation system described above.

[0168] Figure 7 A structural block diagram of a computing device 700 according to one embodiment of this specification is shown. The components of the computing device 700 include, but are not limited to, a memory 710 and a processor 720. The processor 720 is connected to the memory 710 via a bus 730, and a database 750 is used to store data.

[0169] The computing device 700 also includes an access device 740, which enables the computing device 700 to communicate via one or more networks 760. Examples of these networks include Public Switched Telephone Network (PSTN), Local Area Network (LAN), Wide Area Network (WAN), Personal Area Network (PAN), or combinations of communication networks such as the Internet. The access device 740 may include one or more of any type of wired or wireless network interface (e.g., a network interface card (NIC)), such as an IEEE 802.11 Wireless Local Area Network (WLAN) wireless interface, a Wi-MAX (Worldwide Interoperability for Microwave Access) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, or a Near Field Communication (NFC) interface.

[0170] In one embodiment of this specification, the above-described components of the computing device 700 and Figure 7 Other components, not shown, can also be connected to each other, for example, via a bus. It should be understood that... Figure 7 The block diagram of the computing device shown is for illustrative purposes only and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0171] The computing device 700 can be any type of stationary or mobile computing device, including mobile computers or mobile computing devices (e.g., tablet computers, personal digital assistants, laptop computers, notebook computers, netbooks, etc.), mobile phones (e.g., smartphones), wearable computing devices (e.g., smartwatches, smart glasses, etc.) or other types of mobile devices, or stationary computing devices such as desktop computers or personal computers (PCs). The computing device 700 can also be a mobile or stationary server.

[0172] The processor 720 executes the following computer-executable instructions, which, when executed by the processor, implement the steps of the inertia adjustment method for the MW-level inertia flexible new energy synchronous power generation system described above. The above is an illustrative scheme of a computing device according to this embodiment. It should be noted that the technical solution of this computing device and the technical solution of the inertia adjustment method for the MW-level inertia flexible new energy synchronous power generation system described above belong to the same concept. Details not described in detail in the technical solution of the computing device can be found in the description of the technical solution of the inertia adjustment method for the MW-level inertia flexible new energy synchronous power generation system described above.

[0173] An embodiment of this specification also provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the above-described inertia adjustment method for a MW-level inertia flexible new energy synchronous power generation system.

[0174] The above is an illustrative scheme of a computer-readable storage medium according to this embodiment. It should be noted that the technical solution of this storage medium belongs to the same concept as the technical solution of the inertia adjustment method of the MW-level inertia flexible new energy synchronous power generation system described above. For details not described in detail in the technical solution of the storage medium, please refer to the description of the technical solution of the inertia adjustment method of the MW-level inertia flexible new energy synchronous power generation system described above.

[0175] An embodiment of this specification also provides a computer program, wherein when the computer program is executed in a computer, it causes the computer to perform the steps of the above-described inertia adjustment method for a MW-level inertia flexible new energy synchronous power generation system.

[0176] The above is an illustrative scheme of a computer program according to this embodiment. It should be noted that the technical solution of this computer program and the technical solution of the inertia adjustment method of the MW-level inertia flexible new energy synchronous power generation system described above belong to the same concept. For details not described in detail in the technical solution of the computer program, please refer to the description of the technical solution of the inertia adjustment method of the MW-level inertia flexible new energy synchronous power generation system described above.

[0177] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0178] The computer instructions include computer program code, which may be in the form of source code, object code, executable file, or certain intermediate forms. The computer-readable medium may include any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium may be appropriately added to or subtracted according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media may not include electrical carrier signals and telecommunication signals.

[0179] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments in this specification are not limited to the described order of actions, because according to the embodiments in this specification, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the embodiments in this specification.

[0180] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0181] The preferred embodiments disclosed above are merely illustrative of this specification. The optional embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the embodiments described herein. These embodiments are selected and specifically described in this specification to better explain the principles and practical applications of the embodiments, thereby enabling those skilled in the art to better understand and utilize this specification. This specification is limited only by the claims and their full scope and equivalents.

Claims

1. A method for inertia adjustment in a MW-level flexible renewable energy synchronous power generation system, characterized in that, The system includes an inverter, a motor driven by the inverter, a synchronous generator coaxially connected to the motor to form a motor stack, and a detachable inertia disk disposed on the rotating shaft of the motor stack. The method includes: The system identifies the short-circuit ratio (SCR) of the target grid connection point online. When the SCR is lower than a preset threshold, a control command is generated to increase the inertia coefficient and decrease the speed regulation time constant. When the SCR is not lower than the preset threshold, a control command is generated to decrease the inertia coefficient and increase the speed regulation time constant. In response to the control command, the detachable inertia disk can be added, removed, or replaced to change the total moment of inertia of the motor stack; The inverter adjusts the electromagnetic torque according to the inertia coefficient and the speed regulation time constant, thereby changing the release rate of the motor stack rotor kinetic energy.

2. The method according to claim 1, characterized in that, The step of identifying the short-circuit ratio (SCR) of the target grid connection point online includes: A non-power frequency disturbance current of a preset amplitude is injected into the target grid connection point, and the voltage phasor and current phasor of the target grid connection point before and after the disturbance current injection are collected; The equivalent complex impedance on the grid side is calculated based on the ratio of the voltage phasor difference to the current phasor difference before and after the disturbance. The system short-circuit capacity is calculated based on the rated line voltage of the target grid connection point and the modulus of the equivalent complex impedance on the grid side. The short-circuit ratio (SCR) is determined based on the ratio of the system's short-circuit capacity to the system's rated apparent power.

3. The method according to claim 2, characterized in that, The equivalent complex impedance on the grid side is calculated based on the ratio of the voltage phasor difference to the current phasor difference before and after the disturbance, specifically: In the formula, The equivalent complex impedance on the power grid side, in units of These are the voltage phasor and current phasor at the grid connection point before the disturbance injection; These are the voltage phasor and current phasor at the grid connection point after disturbance injection; The system short-circuit capacity is calculated based on the rated line voltage of the target grid connection point and the modulus of the equivalent complex impedance on the grid side, specifically as follows: In the formula, The short-circuit capacity of the grid connection system is expressed in MVA. This is the rated line voltage at the grid connection point, in kV. This represents the magnitude of the equivalent complex impedance on the power grid side. The short-circuit ratio (SCR) is determined based on the ratio of the system's short-circuit capacity to its rated apparent power, specifically as follows: In the formula, This represents the rated apparent power of this MW-level inertial flexible new energy synchronous power generation system, expressed in MVA.

4. The method according to claim 1, characterized in that, The preset threshold is a short-circuit ratio (SCR) of 3. When the SCR is lower than 3, the target grid connection point is determined to be a weak grid. When the SCR is not lower than 3, the target grid connection point is determined to be a strong grid. When the SCR is less than 3, the detachable inertia disk is added, removed, or replaced to make the equivalent inertia support time of the motor stack greater than or equal to 10 seconds; when the SCR is not less than 3, the detachable inertia disk is added, removed, or replaced to make the equivalent inertia support time less than or equal to 2 seconds.

5. The method according to claim 1, characterized in that, Generating the control commands includes: Obtain the real-time operating status set of the system, which includes the grid connection point power angle deviation, frequency deviation, the short-circuit ratio (SCR), and the motor stack speed. The real-time running state set is input into the Actor network of a pre-trained Deep Deterministic Policy Gradient (DDPG) model, and the Actor network outputs the inertia coefficient and the speed regulation time constant; wherein, the DDPG model is generated based on a multi-objective weighted reward function that includes frequency deviation penalty, voltage deviation penalty, shaft torque safety penalty and regulation smoothness penalty.

6. The method according to claim 5, characterized in that, The Actor network includes an input layer, two fully connected hidden layers, and an output layer. The two hidden layers are configured with 256 and 128 neurons respectively and use the ReLU activation function. The output layer uses the Sigmoid activation function to output the inertia coefficient and the speed regulation time constant. The Critic network of the DDPG model is based on the Bellman optimal equation to fit the state-action value function, and the complete mathematical expression is: In the formula, The system state at time t includes the power angle at the grid connection point, frequency deviation, grid short-circuit ratio (SCR), and shaft speed. The control action at time t is the inertia coefficient K and the speed regulation time constant T to be optimized. This is the set of parameters to be trained for the Critic fully connected network, including the weight matrices and bias vectors of each layer; For the target Critic network parameters, a soft update method is used to synchronize the current network parameters to ensure training stability; The immediate reward value is constructed with constraints of minimum frequency deviation, safe shaft torque, and optimal oscillation suppression. This is a discount factor with a value of 0.95, used to weigh the immediate rewards against the long-term rewards. For the Actor network, it is a deterministic policy function that takes the system state as input and outputs the optimal control action. The Critic network updates its weights along the negative gradient direction of the temporal difference error during training, and the Actor network updates its weights along the direction of long-term expected reward improvement based on the policy gradient theorem. Furthermore, both the target Actor network and the target Critic network of the DDPG model synchronize the main network parameters through a soft update mechanism.

7. The method according to claim 5, characterized in that, The multi-objective weighted reward function is constructed by weighting four factors: frequency support effect, voltage stability level, shaft safety constraint, and adjustment smoothness. Frequency deviation and voltage deviation are negatively correlated with the reward value. When the shaft end torque exceeds 3 times the rated protection limit, a penalty value is deducted. When the single-step adjustment amplitude of the inertia coefficient is greater than the amplitude threshold, a smoothness penalty value is deducted.

8. The method according to claim 1, characterized in that, It also includes shaft overload protection steps: Real-time detection of the instantaneous value of the shaft end torque of the motor stack and the speed difference between the motor and the synchronous generator; When the shaft end torque exceeds the preset first-level torque threshold, the inverter is controlled to adjust the electromagnetic torque command output to the motor in order to actively suppress shaft torque oscillation. When the shaft end torque exceeds the preset secondary torque threshold or the speed difference reaches 5% of the rated speed of the motor stack, the grid connection switch is disconnected to isolate the motor stack.

9. The method according to claim 1, characterized in that, It also includes multi-timescale reactive voltage control steps: During the sub-transient phase of grid connection disturbance, the electromagnetic response of the damping winding of the synchronous generator is used to provide instantaneous reactive voltage support. During the transient phase of grid connection disturbance, the magnetic flux conservation characteristics of the excitation winding of the synchronous generator are utilized to provide rapid reactive power support. During the steady-state phase when grid-connected disturbances occur, the excitation parameters are optimized by a neural network based on a self-calibrating attention mechanism, and the excitation system is controlled to adjust the reactive power output of the synchronous generator according to the optimized excitation parameters.

10. A MW-level inertia flexible new energy synchronous power generation system, characterized in that, include: An inverter, the DC side of which is connected to a new energy power generation unit, is configured to convert the DC power output by the new energy power generation unit into AC power. An electric motor, the input of which is connected to the AC output of the inverter, is configured to be driven to rotate by the inverter; A synchronous generator is coaxially and directly connected to the motor to form a motor stack, which is located on the grid-connected energy transfer path from the inverter to the power grid; An inertia adjustment unit includes at least one detachable inertia disk disposed on the motor stack shaft, configured to change the total rotational inertia of the motor stack by adding, removing or replacing the detachable inertia disk; The excitation system, independently connected to the synchronous generator, is configured to independently regulate the reactive power output of the synchronous generator; The controller, connected to the inverter, the excitation system and the grid connection point measurement point, is configured to identify the short-circuit ratio (SCR) of the target grid connection point online. When the SCR is lower than a preset threshold, it generates a control command to increase the inertia coefficient and decrease the speed regulation time constant. When the SCR is not lower than the preset threshold, it generates a control command to decrease the inertia coefficient and increase the speed regulation time constant. The inverter is also configured to adjust the electromagnetic torque according to the inertia coefficient and the speed regulation time constant, thereby changing the release rate of the kinetic energy of the motor stack rotor.

Images