Temperature control load frequency modulation method considering time delay influence

Abstract

The invention belongs to the technical field of intelligent control in which a temperature control load participates in power grid adjustment, and particularly relates to a temperature control load frequency modulation method considering time delay influence. Firstly, a frequency domain equivalent model of a temperature control load participating in power system frequency modulation based on time delay factor influence is established; secondly, comprehensively considering the aggregation operation characteristic of the temperature control load and the dual influence of frequent start and stop on the service life of equipment, and establishing a droop control dynamic response model similar to a conventional generator set to describe a dynamic adjustment relation between load power and frequency; and finally, in order to avoid the influence of frequent start and stop on the failure rate and the service life of the temperature control load equipment, dynamically adjusting a frequency deviation trigger value of the equipment in the cluster based on a temperature control load response frequency updating strategy of the hierarchical heterogeneous control architecture. Not only can the response characteristics of the mass temperature control load cluster be accurately represented, but also the frequency fluctuation suppression of the power grid can be realized.

Description

Technical Field

[0001] This invention belongs to the field of intelligent control technology for temperature-controlled loads participating in power grid regulation, and in particular relates to a frequency regulation method for temperature-controlled loads that takes into account the effects of time delay. Background Technology

[0002] Driven by the "dual-carbon" strategy, my country's power system is undergoing a paradigm shift characterized by "dual high characteristics" (high proportion of new energy sources and high proportion of power electronic equipment). Constrained by my country's energy endowment and load distribution, coupled with complex factors such as wind, solar, and hydropower supply and primary energy supply, the frequency regulation demand of the power grid exhibits characteristics of high frequency, fast response, and multi-timescale coordination. Traditional source-side frequency regulation methods face dual bottlenecks in regulation capacity and response speed. Against this backdrop, temperature-controlled loads, as flexible load resources widely deployed on the end-user side, are gradually evolving into a core carrier for tapping the potential of two-way interaction between sources and loads due to their virtual energy storage characteristics and rapid response capabilities, providing more diversified and multi-dimensional regulation methods for the safe and stable control of the power system. On-off temperature-controlled loads, represented by air conditioning and electric heating equipment, are gradually becoming a cutting-edge research hotspot in the field of dynamic power system regulation, thanks to the differentiated time-shiftability of their electricity demand while meeting end-user comfort constraints.

[0003] However, existing temperature-controlled load regulation methods largely ignore the dynamic operating conditions of temperature-controlled loads and have significant shortcomings in the threshold parameter update mechanism. Furthermore, frequent switching between on and off states can affect the lifespan of temperature-controlled load equipment, and if the temperature-controlled load is locked during scheduling, it will not respond to control signals, leading to response deviations. Therefore, a method for setting and updating thresholds for temperature-controlled loads participating in system frequency regulation is needed.

[0004] Furthermore, the dynamic coupling relationship between latency and frequency directly affects the frequency stability of the power system. Due to the surge in traffic in communication links, massive amounts of load-side data experience communication network congestion during transmission, significantly increasing transmission latency and impacting the precise control of demand-side loads. Currently, although dedicated communication channels enable rapid information transmission, engineering practices generally consider a delay of no more than 300ms from the occurrence of a fault to the completion of control. Moreover, the response speed of temperature-controlled loads varies from milliseconds to seconds. This spatiotemporal asynchrony can lead to control command accumulation or system frequency oscillations during frequency regulation under scenarios of sudden changes in grid frequency.

[0005] Existing research is largely based on idealized assumptions: that the system has no time delay and can achieve real-time control of large-scale distributed loads. Therefore, in order to fully mobilize the enthusiasm of temperature-controlled loads to participate in power grid frequency response control and ensure the effectiveness of user participation in frequency security response, there is an urgent need for an effective method to suppress the impact of time delay and load time-varying characteristics on system frequency stability. Summary of the Invention

[0006] The technical problem to be solved by this invention is to establish a dynamic frequency domain equivalent model of temperature-controlled load integrating communication delay and response delay, and considering the aggregation characteristics of temperature-controlled load groups and the dual impact of their frequent start-stop on equipment lifespan, to provide a temperature-controlled load response frequency update strategy based on a hierarchical heterogeneous control architecture, and to introduce a multi-index quantitative evaluation system, so as to achieve accurate characterization of load group response characteristics and effective suppression of power grid frequency fluctuations while ensuring user comfort.

[0007] A method for frequency regulation of temperature-controlled loads considering the effect of time delay includes the following steps, which are performed sequentially:

[0008] Step 1: Establish a frequency domain equivalent model of temperature-controlled loads participating in power system frequency regulation based on the influence of time delay factors.

[0009] Step 2: Taking into account the combined operating characteristics of temperature-controlled loads and the dual impact of their frequent start-stop cycles on equipment lifespan, establish a droop control dynamic response model similar to that of conventional generator sets to describe the dynamic adjustment relationship between load power and frequency.

[0010] Step 3: To avoid the impact of frequent start-stop cycles on the failure rate and lifespan of temperature-controlled load equipment, a temperature-controlled load response frequency update strategy based on a hierarchical heterogeneous control architecture is used to dynamically adjust the frequency deviation trigger value of equipment within the cluster.

[0011] Step 1 specifically involves establishing a frequency domain equivalent model for temperature-controlled loads participating in power system frequency regulation based on the influence of time delay factors:

[0012] Conventional generating units consist of thermal power units, including a governor G. gov and prime mover G g The two parts have the following transfer function:

[0013]

[0014] In the formula: T g T represents the time constant of a conventional thermal power unit governor. RH F HP and T CH These are the reheater time constant, reheater gain, and high-pressure cylinder steam chamber time constant, respectively.

[0015] Therefore, the transfer function G of the series model of conventional units G (s) can be expressed as:

[0016] G G (s)=G gov (s)G g (s) (3)

[0017] Based on the dynamic equation of the generator rotor and the load frequency characteristics, the transfer function of the generator-load model G(s) can be obtained as follows:

[0018]

[0019] In the formula: H is the moment of inertia; D is the system damping coefficient.

[0020] Communication latency originates from the physical constraints and protocol processing in the data transmission link, and is modeled as a pure latency element with the following transfer function:

[0021]

[0022] In the formula: T c This refers to the communication delay duration.

[0023] Response delay originates from the dynamic characteristics and internal structure of the actuator. The control system contains inertial elements, and the delay generated by the frequency detection element is represented by the response delay. The transfer function of the response delay is expressed as:

[0024]

[0025] In the formula: T r T is the inertial time constant. r =0.18.

[0026] The inertial element in the response delay can be represented by a traditional speed governor model, specifically as follows:

[0027]

[0028] Where: ΔP TCLs_P For the aggregate power of temperature-controlled loads participating in primary frequency regulation, k P This is the active-frequency droop control coefficient for temperature-controlled loads.

[0029] Step 2 specifically involves establishing a dynamic response model for temperature-controlled load cluster droop control, referencing a conventional generator set, as follows:

[0030] At time t, the aggregate power of the temperature-controlled load cluster can be expressed as:

[0031]

[0032] In the formula: N is the size of the temperature-controlled load cluster, P i sin This refers to the power of a single temperature-controlled load.

[0033] At this point, the upward and downward adjustment capacities of the temperature-controlled load cluster can be expressed as:

[0034]

[0035] In the formula: P t u and P t d These represent the up / down capacity adjustment for the cluster at time t.

[0036] Therefore, the range of variation of the aggregate power of the temperature-controlled load cluster at time t can be obtained, which can be specifically expressed as:

[0037]

[0038] In the formula: P t agg_max and P t agg_min P represents the maximum and minimum aggregate power that the cluster can achieve at time t, respectively. t agg _con This represents the actual aggregate power after the cluster is under control.

[0039] To avoid shortening equipment lifespan due to frequent start-stop cycles, a locking constraint mechanism corresponding to the shortest start-up and shutdown times of temperature-controlled loads is introduced. This mechanism uses 1 and 0 state values ​​to determine whether the equipment is in a locked state, and thus whether it is allowed to participate in grid frequency regulation. Specifically, this can be represented as:

[0040]

[0041] In the formula: Let t be the preceding state transition time. and These are the minimum start time and minimum shutdown time, respectively.

[0042] Considering the locking constraints, the capacity increase and decrease of the temperature-controlled load cluster at time t will change to:

[0043]

[0044] In the formula: P t u_L and P t d_L These represent the upward and downward capacity adjustments of the temperature-controlled load cluster at time t, respectively, considering the locking constraints.

[0045] At this point, the range of variation in the aggregated power of the temperature-controlled load cluster is:

[0046]

[0047] In the formula: P t agg_max_L and P tagg_min_L These represent the maximum and minimum aggregate power that the temperature-controlled load cluster can achieve at time t, respectively, considering the locking constraints.

[0048] According to equation (12), the actual adjusted capacity P of the temperature-controlled load cluster u_L And the actual reduction in capacity P d_L A dynamic response model for droop control, analogous to that of a conventional generator set under ideal conditions, is established. Δf represents the grid frequency deviation. and For frequency modulation dead zone limitation, ±0.033Hz is used. Δf min and Δf max The minimum and maximum frequency deviation thresholds that the temperature-controlled load cluster can regulate can be expressed as:

[0049]

[0050] In the formula: R u and R d These are the droop control coefficients for the temperature-controlled load cluster under conditions of increased or decreased grid frequency, obtained from the following formula:

[0051]

[0052] Based on the locking constraint in equation (11), the temperature control load is divided into an unlocked open state group, an unlocked closed state group, and a locked state group, where the unlocked open state group is nlock. on and the closed state group nlock off It can be represented as:

[0053]

[0054] Therefore, the non-locking open state group nlock on and the closed state group nlock off The set of temperature-controlled loads can be defined as:

[0055]

[0056] In the formula: For the i1th device in the sorted unlocked open state group, Let n1 be the i2th device in the sorted non-locked state group, and n2 be the total number of devices in their respective sets.

[0057] When the power grid frequency is disturbed, the temperature-controlled load exceeds the dead zone limit when the frequency deviation value Δf exceeds the limit. Then, adjustments are made sequentially, and the system dynamically matches the response capacity based on the real-time frequency deviation. Specifically, the frequency deviation trigger values ​​for each temperature-controlled load in the unlocked on and off states are as follows:

[0058]

[0059] To quantify the compensation for time delay effects, a pre-offset correction is applied to the original frequency deviation trigger values, forming a dynamic response model for the actual droop control of the temperature-controlled load cluster that considers the impact of time delay. The frequency deviation trigger values ​​for each temperature-controlled load are as follows:

[0060]

[0061] Step 3 is described in detail below:

[0062] To achieve precise and timely frequency regulation, a hierarchical heterogeneous control architecture integrating a central coordination layer and a local execution layer is constructed. Interested temperature-controlled load users can sign demand response agreements with the power grid operator, and a two-way interaction between devices and users is established through the local execution layer to achieve grid frequency deviation response.

[0063] To avoid the impact of frequent start-stop cycles on the failure rate and lifespan of temperature-controlled load equipment

[0064] Within a single control cycle, when the temperature-controlled load device at time t is locked due to the execution of the frequency modulation command and cannot respond in time, the response frequency update strategy will update the frequency deviation trigger value of the device in the controllable response queue in real time to fill the response gap.

[0065] Meanwhile, the frequency deviation trigger values ​​of the remaining controllable devices were all shifted upwards, which not only avoided frequent two-way communication between the central coordination layer and the local execution layer, but also ensured the continuity of the cluster's adjustment capabilities.

[0066] The process of temperature-controlled load clusters participating in power grid frequency regulation adopts a phased periodic control architecture, with each regulation cycle T... cycle Divided into multiple time intervals T inter Closed-loop control is achieved through a three-level linkage mechanism of dynamic response modeling, real-time control execution, and adaptive feedback optimization.

[0067] First, the central coordination layer forms a priority response queue for the temperature control load cluster based on the real-time collected parameters of the temperature control load equipment, and dynamically allocates the frequency deviation trigger value of each device and sends it to the local execution layer.

[0068] Subsequently, during the control cycle, the system continuously monitors the grid frequency deviation and compares it with the trigger values ​​of each temperature-controlled load device to determine whether a switching operation is required, and achieves grid frequency correction by adjusting the power output of the device.

[0069] Based on this, the system implements an adaptive feedback optimization mechanism.

[0070] After the temperature-controlled load response control signal is activated, the controlled and locked devices are immediately removed from the response queue. The frequency deviation trigger values ​​of the remaining controllable devices are updated based on the response gap caused by the grid frequency deviation value at the previous moment. This avoids waste of regulation resources due to frequent locking and maintains the continuity of the cluster's frequency regulation capability.

[0071] After completing the above steps, the system automatically enters the next time interval, cyclically executing real-time control and feedback optimization until the current control cycle ends, ultimately achieving efficient utilization of coordinated control and load response resources throughout the entire cycle.

[0072] Through the above design scheme, the present invention can bring the following beneficial effects:

[0073] This invention establishes a frequency response model for temperature-controlled loads that integrates communication delay and response delay, enabling the system to more accurately predict load response behavior. Considering the characteristics of temperature-controlled load clusters and the dual impact of frequent start-stop cycles on equipment lifespan, a response frequency update strategy for temperature-controlled load clusters based on a hierarchical heterogeneous control architecture is designed. This strategy can fill the response gap caused by locking constraints, significantly reducing temperature-controlled load response deviation and enhancing the reliability of load cluster participation in frequency regulation. By introducing a multi-index quantitative evaluation system, this invention achieves accurate characterization of load group response characteristics and effective suppression of power grid frequency fluctuations while ensuring user comfort. Attached Figure Description

[0074] Figure 1 A schematic diagram of the frequency response models for temperature-controlled load clusters and traditional generator sets;

[0075] Figure 2 This is a schematic diagram illustrating the changes in the aggregated power of a temperature-controlled load cluster.

[0076] Figure 3 A schematic diagram showing the power variation of a temperature-controlled load cluster considering lockout constraints;

[0077] Figure 4 A schematic diagram of the dynamic response model for ideal droop control of a temperature-controlled load cluster;

[0078] Figure 5 A schematic diagram of the dynamic response model for actual droop control of a temperature-controlled load cluster;

[0079] Figure 6 A schematic diagram of the dynamic response model for actual droop control of a temperature-controlled load cluster, taking time delay into account;

[0080] Figure 7 A schematic diagram of a hierarchical heterogeneous control architecture for temperature-controlled load clusters aimed at system frequency stability;

[0081] Figure 8 Schematic diagram of the response gap of temperature-controlled load cluster;

[0082] Figure 9 A schematic diagram of a temperature-controlled load cluster regulation method based on RFU strategy;

[0083] Figure 10 Schematic diagram of the process of temperature-controlled load clusters participating in power grid frequency regulation;

[0084] Figure 11 For simulation data;

[0085] Figure 12 Graph showing the results of power grid frequency regulation with different load sizes;

[0086] Figure 13 A graph showing the results of different load powers participating in power grid frequency regulation;

[0087] Figure 14 The temperature control load regulation capability curve;

[0088] Figure 15 A comparison chart of indoor temperatures before and after temperature-controlled load cluster regulation;

[0089] Figure 16 A comparison chart of power deviations in response to temperature-controlled loads;

[0090] Figure 17 This is a graph showing the system frequency deviation and unit load regulation power.

[0091] Figure 18 The frequency modulation curve of the temperature-controlled load participating in the system, taking into account the effect of time delay;

[0092] Figure 19 Unit load regulation power curve considering time delay Detailed Implementation

[0093] The following uses appendix Figures 1-10 The present invention will be further illustrated by the examples.

[0094] The technical problem to be solved by this invention is to establish a dynamic frequency domain equivalent model of temperature-controlled loads that integrates communication delay and response delay, and considering the aggregation characteristics of temperature-controlled load groups and the dual impact of their frequent start-stop cycles on equipment lifespan, to provide a temperature-controlled load response frequency update strategy based on a hierarchical heterogeneous control architecture, and to introduce a multi-index quantitative evaluation system, so as to achieve accurate characterization of load group response characteristics and effective suppression of grid frequency fluctuations while ensuring user comfort; at the same time, it reveals the sensitivity of delay to the dynamic performance of power system frequency, providing theoretical support for the coordinated control of distributed frequency regulation resources.

[0095] This invention is a method for analyzing the frequency regulation capability of temperature-controlled loads considering the effect of time delay, comprising the following steps, which are performed sequentially:

[0096] Step 1: This invention takes into account the effect of time delay and establishes a frequency response model for temperature-controlled loads and traditional generator sets.

[0097] Conventional generating units consist of thermal power units, including a governor G. gov and prime mover G g The two parts have the following transfer function:

[0098]

[0099] In the formula: T g T represents the time constant of a conventional thermal power unit governor. RH F HP and T CH These are the reheater time constant, reheater gain, and high-pressure cylinder steam chamber time constant, respectively.

[0100] Therefore, the transfer function G of the series model of conventional units G (s) can be expressed as:

[0101] G G (s)=G gov (s)G g (s) (3)

[0102] Fluctuations in power system load can cause dynamic adjustments in generator electromagnetic torque, leading to a dynamic imbalance between rotor mechanical power and electromagnetic power. Based on the generator rotor's dynamic equations and load frequency characteristics, the transfer function of the generator-load model G(s) can be obtained as follows:

[0103]

[0104] In the formula: H is the moment of inertia; D is the system damping coefficient.

[0105] In practical temperature-controlled load control systems, there are generally two types of time delays: communication delay and response delay. Communication delay stems from the physical constraints and protocol processing in the data transmission link and can usually be modeled as a pure time-delay element, with the following transfer function:

[0106]

[0107] In the formula: T c This refers to the communication delay duration.

[0108] The response delay originates from the dynamic characteristics and internal structure of the actuator, i.e., there is an inertial element within the control system. Therefore, the delay generated by the frequency detection element can also be represented by the response delay, and its transfer function can be expressed as:

[0109]

[0110] In the formula: T r The inertial time constant is generally T. r =0.18.

[0111] In power system frequency regulation modeling, the inertial element in the response delay can be equivalent to a traditional speed governor model, which can be specifically represented as:

[0112]

[0113] Where: ΔP TCLs_P For the aggregate power of temperature-controlled loads participating in primary frequency regulation, k P This is the active-frequency droop control coefficient for temperature-controlled loads.

[0114] When the power system experiences frequency deviation due to active power disturbances, generator sets adjust their active power output in real time to restore the frequency. Simultaneously, temperature-controlled load clusters dynamically adjust their power consumption based on the frequency deviation, thereby assisting in correcting the power system frequency deviation. Therefore, considering the time delay effect, an equivalent comparison of the frequency response model and control parameters of temperature-controlled loads and traditional generator sets is shown below. Figure 1 As shown in Table 1.

[0115] Table 1. Equivalent Comparison of Temperature-Controlled Load and Key Parameters of Traditional Generator Sets

[0116]

[0117] Step 2: This invention constructs a dynamic response model for temperature-controlled load cluster droop control. Considering the problems caused by frequent start-stop of a large number of temperature-controlled loads leading to shortened equipment lifespan, a locking constraint mechanism based on the shortest start-stop time of the temperature-controlled load is introduced.

[0118] At time t, the aggregate power of the temperature-controlled load cluster can be expressed as:

[0119]

[0120] In the formula: N is the size of the temperature-controlled load cluster, P i sin This refers to the power of a single temperature-controlled load.

[0121] Meanwhile, the corresponding upward / downward adjustment capacity can be expressed as:

[0122]

[0123] In the formula: P t u and P t d These represent the up / down capacity adjustment for the cluster at time t.

[0124] Therefore, the range of variation of the aggregate power of the temperature-controlled load cluster at time t can be obtained, which can be specifically expressed as:

[0125]

[0126] In the formula: P t agg_max and P t agg_min P represents the maximum and minimum aggregate power that the cluster can achieve at time t, respectively. t agg _con The actual aggregation power after the cluster is under control, and the specific changes are as follows: Figure 2 As shown.

[0127] However, in practice, to avoid problems such as shortened equipment lifespan caused by frequent start-stop cycles, this invention introduces a locking constraint mechanism based on the shortest start-stop time for temperature-controlled loads to ensure the reliability of equipment operation. By using 1 and 0 state values ​​to determine whether the equipment is in a locked state, and thus whether it is allowed to participate in grid frequency regulation, it can be specifically expressed as follows:

[0128]

[0129] In the formula: Let t be the preceding state transition time. and These are the minimum on / off times set.

[0130] Taking lockout constraints into account, the up / down capacity adjustment of the temperature-controlled load cluster at time t will change to:

[0131]

[0132] In the formula: P t u_L and P t d_L These represent the up / down adjustment capacity of the temperature-controlled load cluster at time t, respectively, under the consideration of locking constraints.

[0133] At this point, the range of variation in the aggregated power of the temperature-controlled load cluster is:

[0134]

[0135] In the formula: P t agg_max_L and P t agg_min_L These represent the maximum and minimum aggregate power that the temperature-controlled load cluster can achieve at time t, respectively, considering the locking constraints.

[0136] In summary, a schematic diagram of the aggregated power variation of a temperature-controlled load cluster considering lockout constraints can be obtained, as shown below. Figure 3 As shown.

[0137] Under normal operation, the dynamic response model of droop control exhibits a linear relationship. According to equation (12), the actual adjusted capacity P of the temperature-controlled load cluster... u_L And the actual reduction in capacity P d_L Establish an ideal dynamic response model for droop control analogous to that of a conventional generator set, such as... Figure 4 As shown.

[0138] In the figure, Δf represents the power grid frequency deviation. and To limit the frequency modulation dead zone, it is generally taken as ±0.033Hz. Δf min and Δf max These are the minimum and maximum frequency deviation thresholds that the temperature-controlled load cluster can regulate. When the deviation value is lower than Δf min At that time, reduce capacity -P d_L All inputs are used when the deviation value is higher than Δf. max At that time, the capacity P was increased. u_L All are engaged. Therefore, the frequency response of the temperature-controlled load cluster participating in the power grid regulation can be expressed as:

[0139]

[0140] In the formula: R u and R d These are the droop control coefficients for the temperature-controlled load cluster under conditions of increased or decreased grid frequency, obtained from the following formula:

[0141]

[0142] However, this invention focuses on switch-controlled temperature-controlled loads, typically represented by electric heating loads. These loads are characterized by being controlled solely by a switch, resulting in a binary discrete output characteristic of rated power and zero power. This characteristic directly leads to a lack of smoothness in their power regulation capability; that is, the actual droop control model will exhibit a step-like variation, such as... Figure 5 As shown.

[0143] The selection criteria for temperature-controlled load equipment should be based on its frequency regulation participation capability, determined by the equipment's locked state. According to the locking constraint in equation (11), the temperature-controlled load is divided into an unlocked open state group, an unlocked closed state group, and a locked state group, where the unlocked open state group is nlock. on and the closed state group nlock off It can be represented as:

[0144]

[0145] The temperature-controlled load equipment was screened by calculating the operable time in both the unlocked open and closed state groups separately and sorting them in ascending order. Equipment with shorter operable times has a longer controllable time span and therefore participates in grid frequency regulation with higher priority. Based on this, the unlocked open state group (nlock)... on and the closed state group nlock off The set of temperature-controlled loads can be defined as:

[0146]

[0147] In the formula: For the i1th device in the sorted unlocked open state group, Let n1 be the i2th device in the sorted non-locked state group, and n2 be the total number of devices in their respective sets.

[0148] When the power grid frequency is disturbed, the temperature-controlled load exceeds the dead zone limit when the frequency deviation value Δf exceeds the limit. Then, adjustments are made sequentially, and the system dynamically matches the response capacity based on the real-time frequency deviation. Taking a frequency decrease as an example, when the grid frequency deviation is lower than the frequency deviation trigger value of the temperature-controlled load in the non-locked-on state group, the device will automatically participate in grid frequency regulation; the same applies when the frequency increases. Specifically, the frequency deviation trigger values ​​for each temperature-controlled load in the non-locked-on state group and the off state group are as follows:

[0149]

[0150] However, in practical engineering, when the grid frequency deviation reaches the frequency deviation trigger value of a certain temperature-controlled load, the actual response lag of the equipment is not negligible due to factors such as time delay, causing a deviation between the theoretical and actual control capacity, which in turn negatively affects the grid frequency regulation. To quantify and compensate for the time delay effect, a pre-offset correction can be applied to the original frequency deviation trigger value, forming a result such as... Figure 6 The model shown is a dynamic response model for the actual droop control of a temperature-controlled load cluster, considering the effect of time delay. The frequency deviation trigger values ​​for each temperature-controlled load are as follows:

[0151]

[0152] Step 3: To avoid the impact of frequent start-stop cycles on the failure rate and lifespan of temperature-controlled load equipment, this invention proposes a temperature-controlled load response frequency update strategy based on a hierarchical heterogeneous control architecture. By dynamically adjusting the frequency deviation trigger value of devices within the cluster, the reliability of frequency modulation response is ensured while maintaining the existing communication architecture.

[0153] To achieve both precision and timeliness in frequency regulation, this invention constructs a hierarchical heterogeneous control architecture that integrates a central coordination layer and a local execution layer, such as... Figure 7As shown. In addition, willing temperature-controlled load users can sign demand response agreements with the power grid operator and establish a two-way interaction between devices and users through the local execution layer to achieve power grid frequency deviation response. Specifically, the central coordination layer collects and processes temperature-controlled load data using a minute-level control cycle, calculates the upward and downward adjustment capacity that the temperature-controlled load cluster can provide in each cycle, as well as the frequency deviation trigger value of each device, and then constructs a droop control dynamic response model, awaiting response instructions from the local execution layer. The local execution layer collects frequency deviation signals at second-level execution intervals, triggering a switch state switch only when the detected power grid frequency deviation reaches the trigger value of a certain device, enabling temperature-controlled loads to quickly participate in power grid frequency regulation.

[0154] To avoid the impact of frequent start-stop cycles on the failure rate and lifespan of temperature-controlled load equipment, a lockout constraint needs to be considered. This means that after equipment starts or stops, it must maintain its current state for the minimum required duration before switching. However, this control method results in the frequency deviation trigger value of each temperature-controlled load within the same cycle remaining constant. Furthermore, the lockout constraint prevents equipment already participating in frequency regulation from making a secondary response for a short period, creating a response gap and significantly reducing the system's dynamic adjustment margin. Figure 8 As shown.

[0155] At time t, the power grid frequency deviation Δf t Less than the lower limit of the frequency dead zone Temperature-controlled load clusters begin to participate in power grid frequency regulation, when Δf t Reaching the device frequency deviation trigger value hour, The corresponding devices begin to respond, meaning devices on1, on2, and on3 in the unlocked open state group shut down and locked. Due to the locking constraint, from time t until the shortest shutdown time ends, devices on1, on2, and on3 are all in an uncontrollable state. During this time, their corresponding frequency deviation trigger values ​​will be missing, meaning that when the frequency deviation value... At that time, no corresponding equipment in the temperature-controlled load cluster responded. However, at time t+1, the grid frequency deviation value Δf t+1 Less than the lower limit of the frequency dead zone However, the maximum trigger frequency of devices in the unlocked open state group is... because Since the response conditions are not met, the temperature-controlled load cluster cannot participate in grid frequency regulation. Therefore, after each response, it is necessary to further update the frequency deviation trigger value of each device to ensure that there are always controllable devices within the corresponding grid frequency deviation range to respond to the grid's frequency regulation needs.

[0156] To address the aforementioned problems caused by locking constraints, this invention proposes a temperature-controlled load coordination control method based on a response frequency update (RFU) strategy. Specifically, by dynamically adjusting the frequency deviation trigger values ​​of devices within the cluster, the reliability of the frequency modulation response is ensured while maintaining the existing communication architecture. Figure 9 As shown, within a single control cycle, when temperature-controlled load devices on1, on2, and on3 at time t are locked due to the execution of the frequency modulation command and cannot respond immediately, the RFU strategy will update the frequency deviation trigger values ​​of the devices in the controllable response queue in real time: for example, adaptive correction will be made to the frequency deviation trigger values ​​of devices on4, on5, and on6 (adjusted to...). and To fill the response gap, the frequency deviation trigger values ​​of the remaining controllable devices were all shifted upwards, which not only avoided frequent two-way communication between the central coordination layer and the local execution layer, but also ensured the continuity of the cluster adjustment capability.

[0157] The process of temperature-controlled load clusters participating in power grid frequency regulation adopts a phased, periodic control architecture. For example... Figure 10 As shown, each control cycle T cycle Divided into multiple time intervals T inter Closed-loop control is achieved through a three-tiered linkage mechanism of dynamic response modeling, real-time control execution, and adaptive feedback optimization. First, the central coordination layer, based on real-time collected parameters of the temperature-controlled load equipment, forms a priority response queue for the temperature-controlled load cluster and dynamically allocates frequency deviation trigger values ​​for each device, sending them to the local execution layer. Then, within the control cycle, the system continuously monitors the grid frequency deviation and compares it with the trigger values ​​of each temperature-controlled load device to determine whether switching operations are necessary, and corrects the grid frequency by adjusting the device power output. Based on this, the system executes an adaptive feedback optimization mechanism. After the temperature-controlled load responds to the control signal, the controlled and locked devices are immediately removed from the response queue, and the frequency deviation trigger values ​​of the remaining controllable devices are updated based on the response gap caused by the grid frequency deviation value at the previous moment. This avoids wasting control resources due to frequent locking and maintains the continuity of the cluster's frequency regulation capability. After completing the above steps, the system automatically enters the next time interval, cyclically executing real-time control and feedback optimization until the end of the current control cycle, ultimately achieving coordinated control and efficient utilization of load response resources throughout the entire cycle.

[0158] Specific example: The frequency regulation capability analysis of temperature-controlled load considering the effect of time delay provided by this invention includes the following:

[0159] (1) Case Background

[0160] This invention sets up two calculation examples for analysis: the first is a normal fluctuation scenario, that is, the system frequency deviation fluctuates between [-0.1Hz, 0.1Hz], and conducts a frequency regulation performance evaluation of a single temperature-controlled load cluster; the second is a complex disturbance scenario, focusing on the frequency regulation gain of the auxiliary power generation side of the temperature-controlled load cluster, and analyzing the impact of time delay on the dynamic response characteristics of the system.

[0161] (2) Parameter settings

[0162] This invention uses typical daily meteorological temperature data of a certain region (e.g.) Figure 11 (a) Constructing a simulation scenario. Given that the frequency regulation dead zone of a generator set is typically ±0.05Hz, to analyze the independent frequency regulation capability of a temperature-controlled load cluster and its compensation effect on the dead zone of traditional generator sets, white noise simulation is used to generate microgrid frequency deviation data with an initial deviation within ±0.05Hz, such as... Figure 11 As shown in (b). Meanwhile, the initial temperature of the house is set to be uniformly distributed within the range of [19℃, 23℃], and combined with the first-order ETP model of temperature control load (whose equivalent heat capacity and equivalent thermal resistance both follow a Gaussian distribution with standard deviation), the specific simulation parameters are shown in Table 2.

[0163] Table 2 Simulation parameters for temperature control load

[0164]

[0165] 1. The impact of temperature-controlled load scale on power grid frequency regulation:

[0166] For a community microgrid containing 2000 temperature-controlled load devices, the effects of 500, 1000, and 2000 devices participating in grid frequency regulation were tested. The impact of cluster size on grid frequency regulation performance was analyzed during typical time periods. (Specific details are as follows...) Figure 12 As shown.

[0167] Figure 12 (a) shows the results of frequency regulation with 500 temperature-controlled loads. It can be seen that due to the small scale, limited self-regulation capability, and the influence of lockout constraints, the response capacity is insufficient to meet the frequency regulation requirements. However, as the scale expands to 1000 and 2000 units, the response capacity and frequency regulation effect significantly improve, such as... Figure 12 As shown in (c) and (e), to clearly compare the overall frequency quality of the system under different load scales, the relevant indicators in Table 3 are derived from the simulation results. Comparative analysis shows that as the scale of the temperature-controlled load cluster increases, the number of frequency deviation violations decreases significantly. For example, after the load scale increases to 2000 units, the violation rate decreases from 14.3218% before adjustment to 0.1157%; the average rate of change of frequency deviation decreases from 0.7044% to 0.6205%, verifying the positive effect of scale expansion on suppressing frequency fluctuations and enhancing grid stability.

[0168] Table 3 Comparison of Frequency Quality under Different Load Scales (Overall)

[0169]

[0170] Further analysis was conducted using 15-minute local results, incorporating relevant indicators such as the maximum and minimum frequency deviations and root mean square error, as shown in Table 4. Overall, system frequency fluctuations decreased with increasing load size. Notably, when the load size reached 500 units, the minimum frequency was -0.0777Hz, lower than the pre-adjustment level of -0.0712Hz, indicating increased frequency fluctuations. Combined with the analysis in 12(b), it can be seen that when the cluster's upscaling capacity is insufficient, due to the influence of locking factors, the large-scale activation of temperature-controlled load devices triggered by frequencies below the rated value in the early stages will further exacerbate frequency fluctuations and produce negative effects during the dynamic response phase.

[0171] Table 4. Comparison of frequency quality under different load scales (partial)

[0172]

[0173] 2. The impact of temperature-controlled load power on power grid frequency regulation:

[0174] Given the impact of temperature-controlled load clusters of different sizes on power grid frequency regulation, this experiment sets a single variable and, under the condition of ensuring sufficient responsive capacity, integrates temperature-controlled loads with rated power of 1000W, 1600W, and 2000W from a community-level microgrid system to participate in power grid frequency regulation. A selection of typical segments are chosen to demonstrate in detail the impact of temperature-controlled loads of different power levels on the power grid frequency regulation results. Specifically, as shown below... Figure 13 As shown.

[0175] Figure 13 (a) illustrates the results of frequency regulation with a single 1000W temperature-controlled load. Due to the low rated power of a single unit and the lockout constraint, the curve step is small when constructing the dynamic response model for the droop control of the temperature-controlled load cluster. Under the premise of sufficient responsive capacity, the regulation effect of the temperature-controlled load can be more precisely utilized. However, as the load power increases to 1600W and 2000W, with the number of units and the grid frequency difference remaining unchanged, the actual number of temperature-controlled load response devices participating in frequency regulation decreases, resulting in a decline in the frequency regulation effect (see...). Figure 13 (c) and (e)). Furthermore, a comparison of the simulation indicators in Table 5 shows that as the power of the temperature-controlled load increases, the number of frequency deviation violations significantly decreases. For example, with a single load power of 1000W, the violation rate decreased from 14.3218% before adjustment to 0.3564%; simultaneously, the average rate of change of frequency deviation decreased from 0.7044% to 0.6255%, indicating a further reduction in frequency fluctuation amplitude and a significant improvement in grid frequency stability.

[0176] Table 5 Comparison of frequency quality under different load powers (overall)

[0177]

[0178] Analysis of 15-minute local data (as shown in Table 6) reveals that system frequency fluctuations intensify with increasing single-unit load power. At 1000W, the minimum frequency deviation improved from -0.0789Hz to -0.0452Hz. Figure 13 (b) Analysis shows that when the response capacity is sufficient, reducing the power of a single unit can improve the regulation accuracy of the temperature control cluster and effectively suppress grid fluctuations.

[0179] Table 6. Comparison of frequency quality under different load powers (partial)

[0180]

[0181] Meanwhile, the increased power of individual units reduces the number of temperature-controlled loads participating in frequency regulation simultaneously, thus slowing down the fluctuation of the total power of the cluster and reducing the instantaneous impact on the power grid caused by frequent start-stop cycles. Affected by ambient temperature, the total power of the temperature-controlled load cluster is negatively correlated with outdoor temperature. During the high-temperature period from 9:00 to 16:00, the upward adjustment capacity of the load cluster decreases due to reduced heating demand, while the downward adjustment capacity increases due to increased heat dissipation demand. This results in better regulation performance when the frequency exceeds the upper limit during this period, while the regulation capability is relatively weaker when it exceeds the lower limit.

[0182] In addition, from Figure 13 As shown in (a), (c), and (e), the down-regulation capability of temperature-controlled load clusters to the grid frequency is slightly higher than their up-regulation capability. This is related to the inherent operating characteristics of temperature-controlled loads. Since their average duty cycle is typically below 0.5, the on-time of the same load is shorter than its off-time, thus affecting the regulation capability of the load cluster. Taking 2000 temperature-controlled loads participating in frequency regulation as an example (see...),... Figure 14 According to calculations, the average duty cycle of the load cluster is 0.3821, meaning the average shutdown time is 1.6171 times the on time. This characteristic gives the cluster a stronger ability to adjust for frequency reduction.

[0183] When adjusting the temperature-controlled load, user comfort must also be considered. On one hand, a locking constraint mechanism is used to avoid frequent operation of the same device within a short period. On the other hand, while suppressing power grid frequency fluctuations, the room temperature is controlled within the user-defined threshold range to meet user needs. A comparison of indoor temperatures before and after adjustment is provided below. Figure 15 As shown.

[0184] To further verify the effectiveness of the temperature-controlled load coordination control method based on the RFU strategy, this example compares the response performance differences between the two scenarios: no RFU strategy and the scenario with the RFU strategy. For a detailed comparison of the response power deviations, please refer to [link to relevant documentation]. Figure 16 Without the RFU strategy, the lockout constraint causes temperature control equipment to be unable to respond in the short term after being controlled, and the frequency deviation trigger value of each device is fixed within the same period, resulting in a continuous response gap and equipment idleness, forming a significant load response power deviation. The RFU strategy adopted in this example, however, promptly removes the locked load from the controllable queue and updates the frequency deviation trigger value of the responsive load in real time, filling the response gap caused by the lockout constraint and enhancing the reliability of the load cluster participating in frequency regulation.

[0185] (3) Analysis of the effect of temperature control load regulation considering the time delay

[0186] The invention is constructed as follows Figure 1 The power system simulation model shown includes a reheat turbine unit, temperature-controlled load, and conventional load. The system capacity is 800MW, the rated frequency is 50Hz, the primary frequency regulation droop coefficient R is 0.05, and the governor's inertial time constant T... g The voltage output power ratio is 0.2s, and the high voltage output power ratio is F. HP The time constant T of the high-pressure cylinder steam chamber is 0.3. CH The reheater time constant T is 0.3s. RH The time is 7 seconds. The inertia constant H of the power system is 10, and the system damping coefficient D is 1. The total system regulation capacity is 80MW, of which the upper limit of the regulation capacity of the temperature-controlled load cluster is 30MW, and the upper and lower limits of the frequency regulation dead zone are... and The frequency modulation range is ±0.033Hz, with upper and lower limits Δf. max and Δf min It is ±0.2Hz.

[0187] The invention sets up three scenarios for comparative analysis: (1) the temperature-controlled load does not participate in the system frequency regulation, and the required regulation capacity is provided by the generator set; (2) the temperature-controlled load participates in the primary frequency regulation of the system and provides 15MW of regulation capacity; (3) the temperature-controlled load participates in the primary frequency regulation of the system and provides 30MW of regulation capacity. The simulation length is set to 120s, and a load disturbance ΔP is added to the system at the 10th second. L The system frequency drops to 20MW, causing a frequency deviation. The dynamic response curves of the system frequency deviation and the unit load regulation power curves under different conditions are shown below. Figure 17 As shown.

[0188] By introducing three key indicators—maximum frequency deviation, maximum regulating power of the temperature-controlled load, and frequency recovery time—the frequency regulation effect was compared and analyzed, as shown in Table 7. It should be noted that in this example, the frequency recovery time is defined as the time span from the occurrence of the disturbance until the system frequency deviation re-enters the ±0.033Hz allowable range. The results show that when the temperature-controlled load participates in the primary frequency regulation of the system, it can undertake part of the regulating power demand. Furthermore, as the regulating capacity quota increases, the corresponding regulating capacity provided by the generator decreases, significantly improving the system's frequency dynamic response characteristics.

[0189] Table 7 Comparison of Simulation Results under Load Step Disturbance

[0190]

[0191] Further analysis is conducted on the impact of communication delay during the participation of temperature-controlled load clusters in system frequency modulation on the frequency modulation effect. A load disturbance ΔP is set. L The system has a capacity of 30MW, with temperature-controlled loads participating in frequency regulation and providing 20MW of regulation capacity. Then, the system is analyzed and compared for no communication delay and T... c The simulation results are shown at 1s, 2s, 3s, and 4s, as follows: Figure 18 As shown.

[0192] In the absence of communication delay, the power system frequency deviation is minimal, at only -0.1125Hz. As the communication delay gradually increases, the system's dynamic frequency response characteristics show a significant deterioration trend: reaching -0.1782Hz with a 4s delay. The results indicate that the presence of communication delay not only directly exacerbates the system frequency deviation but also induces oscillations and instability in the system frequency. When the delay is in the 1s and 2s range, the system exhibits only weak oscillations with low amplitude; however, when the delay extends to 3 seconds or more, the amplitude of the system frequency oscillations increases significantly, the duration of the oscillations lengthens, and the frequency recovery process is delayed. Specific simulation results are shown in Table 8.

[0193] Table 8. System frequency simulation results considering time delay.

[0194]

[0195] Taking a 3-second delay as an example, let's compare the power regulation characteristics of a traditional generator set and a temperature-controlled load under no-delay conditions, such as... Figure 19 As shown. Among them, Figure 19 (a) indicates that as the system gradually returns to steady state, the temperature control load regulation power shows a gradual downward trend and eventually exits the frequency regulation process, while the generator set regulation power continues to climb until it fully undertakes the regulation capacity required by the system. Figure 19(b) indicates that the delay in frequency regulation command transmission causes the generator set and temperature-controlled load output to exhibit significant amplitude reduction oscillation characteristics. This unplanned power fluctuation not only exacerbates the mechanical stress loss of rotating equipment and shortens the service life of the equipment, but also the frequent start-up and shutdown of temperature-controlled load equipment directly affects the temperature control experience of end users.

[0196] The specific embodiments used in this invention have provided a detailed description of the invention, but are not limited to these embodiments. Any obvious modifications made by those skilled in the art based on the teachings of this invention are within the scope of protection of this invention.

Claims

1. A method for frequency regulation of temperature-controlled loads considering the effect of time delay, comprising the following steps, which are performed sequentially: Step 1: Establish a frequency domain equivalent model of temperature-controlled loads participating in power system frequency regulation based on the influence of time delay factors; Step 2: Taking into account the aggregated operation characteristics of temperature-controlled loads and the dual impact of their frequent start-stop on equipment lifespan, a dynamic response model for temperature-controlled load cluster droop control is established to describe the dynamic adjustment relationship between load power and frequency. A locking constraint mechanism based on the shortest start-stop time of temperature-controlled loads is introduced. Step 3: To avoid the impact of frequent start-stop on the failure rate and lifespan of temperature-controlled load equipment, a temperature-controlled load response frequency update strategy based on a hierarchical heterogeneous control architecture is adopted to dynamically adjust the frequency deviation trigger value of equipment within the cluster.

2. The method for frequency regulation of temperature-controlled load considering the effect of time delay according to claim 1, characterized in that, Step 1 specifically involves establishing a frequency domain equivalent model for temperature-controlled loads participating in power system frequency regulation based on the influence of time delay factors: Conventional generating units consist of thermal power units, including a governor G. gov and prime mover G g Two parts, whose transfer function is, In the formula, T g T represents the time constant of a conventional thermal power unit governor. RH F HP and T CH These are the reheater time constant, reheater gain, and high-pressure cylinder steam chamber time constant, respectively. Therefore, the transfer function G of the series model of conventional units G (s) can be expressed as, G G (s)=G gov (s)G g (s) (3) Based on the dynamic equation of the generator rotor and the load frequency characteristics, the transfer function of the generator-load model G(s) can be obtained as follows: In the formula, H is the moment of inertia; D is the system damping coefficient; Communication latency originates from the physical constraints and protocol processing in the data transmission link. It is modeled as a pure latency element, and its transfer function is: In the formula, T c For communication delay duration; Response delay originates from the dynamic characteristics and internal structure of the actuator. The control system contains inertial elements, and the delay generated by the frequency detection element is represented by the response delay. The transfer function of the response delay is expressed as follows: In the formula, T r T is the inertial time constant. r =0.18; The inertial element in the response delay can be represented by a traditional speed governor model, specifically as follows: In the formula, ΔP TCLs_P For the aggregate power of temperature-controlled loads participating in primary frequency regulation, k P This is the active-frequency droop control coefficient for temperature-controlled loads.

3. The method for frequency regulation of temperature-controlled load considering the effect of time delay according to claim 2, characterized in that, Step 2 specifically involves establishing a dynamic response model for temperature-controlled load cluster droop control, referencing a conventional generator set, as follows: At time t, the aggregate power of the temperature-controlled load cluster can be expressed as: In the formula, N is the total number of units in the temperature-controlled load cluster, and P i sin This refers to the power of a single temperature-controlled load unit. At this point, the upward and downward adjustment capacities of the temperature-controlled load cluster can be expressed as follows: In the formula, P t u and P t d These represent the upward and downward adjustments of the temperature-controlled load cluster at time t, respectively. Therefore, the range of variation of the aggregated power of the temperature-controlled load cluster at time t can be obtained, which can be specifically expressed as follows: In the formula, P t agg_max and P t agg_min P represents the maximum and minimum aggregate power that the cluster can achieve at time t, respectively. t agg_con This represents the actual aggregate power after the cluster is under control. To avoid shortening equipment lifespan due to frequent on / off cycles, a locking constraint mechanism corresponding to the shortest on / off time of the temperature-controlled load is introduced. This mechanism uses 1 and 0 state values ​​to determine whether the equipment is in a locked state, and thus whether it is allowed to participate in grid frequency regulation. Specifically, this can be represented as follows: In the formula, Let t be the preceding state transition time. and These are the minimum start time and minimum shutdown time, respectively. Taking lockout constraints into account, the capacity increase and decrease of the temperature-controlled load cluster at time t will change to: In the formula, P t u_L and P t d_L These represent the upward and downward capacity adjustments of the temperature-controlled load cluster at time t, respectively, considering the locking constraints. At this point, the range of variation in the aggregated power of the temperature-controlled load cluster is: In the formula, P t agg_max_L and P t agg_min_L These represent the maximum and minimum aggregate power that the temperature-controlled load cluster can achieve at time t, respectively, considering the locking constraints; According to equation (12), the actual adjusted capacity P of the temperature-controlled load cluster u_L And the actual reduction in capacity P d_L A dynamic response model for droop control, analogous to that of a conventional generator set under ideal conditions, is established; Δf represents the grid frequency deviation. and For frequency modulation dead zone limitation, ±0.033Hz is used; Δf min and Δf max Let the minimum and maximum frequency deviation thresholds that the temperature-controlled load cluster can regulate be represented as the power grid frequency regulation response quantity of the temperature-controlled load cluster. In the formula, R u and R d These are the droop control coefficients for the temperature-controlled load cluster under conditions of increased or decreased grid frequency, which can be obtained from the following formula. Based on the locking constraint in equation (11), the temperature control load is divided into an unlocked open state group, an unlocked closed state group, and a locked state group, where the unlocked open state group is nlock. on and the closed state group nlock off It can be represented as, Therefore, the non-locking open state group nlock on and the closed state group nlock off The set of temperature-controlled loads can be defined as follows: In the formula, on i1 For the i1th device in the sorted unlocked open state group, off i2 Let n1 be the i2th device in the sorted non-locked off state group, and n2 be the total number of devices in their respective sets. When the power grid frequency is disturbed, the temperature-controlled load exceeds the dead zone limit when the frequency deviation value Δf exceeds the limit. Then, adjustments are made sequentially, and the system dynamically matches the response capacity based on the real-time frequency deviation. Specifically, the frequency deviation trigger values ​​for each temperature-controlled load in the unlocked open and closed states are as follows: To quantify the compensation time delay effect, the original frequency deviation trigger value is pre-offset corrected to form a dynamic response model for the actual droop control of the temperature-controlled load cluster that considers the time delay effect. The frequency deviation trigger values ​​for each temperature-controlled load are as follows:

4. A method for frequency regulation of temperature-controlled load considering the effect of time delay according to claim 3, characterized in that, Step 3 is described in detail below: To achieve precise and timely frequency regulation, a hierarchical heterogeneous control architecture integrating a central coordination layer and a local execution layer is constructed. Interested temperature-controlled load users can sign demand response agreements with the power grid operator and build a two-way interaction between equipment and users through the local execution layer to achieve power grid frequency deviation response. Within a single control cycle, when the temperature-controlled load device at time t is locked due to the execution of the frequency modulation command and cannot respond in time, the response frequency update strategy will update the frequency deviation trigger value of the device in the controllable response queue in real time to fill the response gap. Meanwhile, the frequency deviation trigger values ​​of the remaining controllable devices were all shifted upwards, which not only avoided frequent two-way communication between the central coordination layer and the local execution layer, but also ensured the continuity of the cluster's adjustment capabilities. The process of temperature-controlled load clusters participating in power grid frequency regulation adopts a phased periodic control architecture, with each regulation cycle T... cycle Divided into multiple time intervals T inter Closed-loop control is achieved through a three-level linkage mechanism of dynamic response modeling, real-time control execution, and adaptive feedback optimization. First, the central coordination layer forms a priority response queue for the temperature control load cluster based on the real-time collected parameters of the temperature control load equipment, and dynamically allocates the frequency deviation trigger value of each device and sends it to the local execution layer. Subsequently, during the control cycle, the system continuously monitors the power grid frequency. The deviation is compared with the trigger values ​​of each temperature-controlled load device to determine whether a switching operation is required, and the power grid frequency is corrected by adjusting the power output of the device. Based on this, the system implements an adaptive feedback optimization mechanism; After the temperature-controlled load response control signal is activated, the controlled and locked devices are immediately removed from the response queue, and the frequency deviation trigger values ​​of the remaining controllable devices are updated based on the response gap caused by the grid frequency deviation value at the previous moment. This avoids waste of regulation resources due to frequent locking, while maintaining the continuity of the cluster's frequency regulation capability. After completing the above steps, the system automatically enters the next time interval, cyclically executing real-time control and feedback optimization until the current control cycle ends, ultimately achieving efficient utilization of coordinated control and load response resources throughout the entire cycle.

Images